Vous êtes sur la page 1sur 475

Anesthesia for Spine Surgery

Anesthesia for Spine Surgery


Ehab Farag, M.D., F.R.C.A.
Staf Anesthesiologist, Departments of General Anesthesia and Outcomes Research, Cleveland Clinic, Ohio, USA
c a mb rid g e u ni ve r si t y pres s
Cambridge, New York, Melbourne, Madrid, Cape Town,
Singapore, So Paulo, Delhi, Mexico City
Cambridge University Press
he Edinburgh Building, Cambridge CB2 8RU, UK

Published in the United States of America by Cambridge University Press, New York

www.cambridge.org
Information on this title: www.cambridge.org/9781107005310

Cambridge University Press 2012

his publication is in copyright. Subject to statutory exception


and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.

First published 2012

Printed in the United Kingdom at the University Press, Cambridge

A catalogue record for this publication is available from the British Library

Library of Congress Cataloguing in Publication data


Anesthesia for spine surgery / [edited by] Ehab Farag.
p. ; cm.
Includes bibliographical references and index.
ISBN 978-1-107-00531-0 (hbk.)
I. Farag, Ehab.
[DNLM: 1. Spinesurgery. 2. Anesthesiamethods. 3. Postoperative Care. WE 725]
617.471dc23 2012007357

ISBN 978-1-107-00531-0 Hardback

Cambridge University Press has no responsibility for the persistence or


accuracy of URLs for external or third-party internet websites referred to in
this publication, and does not guarantee that any content on such websites is,
or will remain, accurate or appropriate.

Every efort has been made in preparing this book to provide accurate and up-to-date information
which is in accord with accepted standards and practice at the time of publication. Although case
histories are drawn from actual cases, every efort has been made to disguise the identities of the
individuals involved. Nevertheless, the authors, editors and publishers can make no warranties that
the information contained herein is totally free from error, not least because clinical standards
are constantly changing through research and regulation. he authors, editors, and publishers therefore
disclaim all liability for direct or consequential damages resulting from the use of material contained
in this book. Readers are strongly advised to pay careful attention to information provided by the
manufacturer of any drugs or equipment that they plan to use.
To my wife, my daughters, and my mother for their incessant prayers,
help, and support and in memory of my late father
Contents
List of contributors ix
Foreword by Dr. Edward Benzel xiii
Foreword by Dr. David Brown xv
Preface xvii

Section 1 General considerations 8.4 An overview of minimally invasive spine


surgery 130
1 Preoperative assessment of the R. Douglas Orr
adult patient 1
Elizabeth A. M. Frost 8.5 Posterior lumbar interbody fusion 138
Virgilio Matheus and William Bingaman
2 Fluid management 25
Maria Bauer, Andrea Kurz, and Ehab Farag 8.6 Minimally invasive procedures for
vertebral compression fractures 141
3 Blood conservation 43 Jason E. Pope and Nagy Mekhail
Robert Helfand
8.7 Endoscopic surgery for Chiari
4 Airway management in spine surgery 50 malformation type I 152
Basem Abdelmalak and D. John Doyle Rodolfo Hakim and Xiao Di
5 Spine imaging 61 8.8 Posterior and anterior thoracic
Doksu Moon, Christian Koopman, and surgery 158
Ramez Malaty Matthew Grosso and Michael Steinmetz
6 Evoked potential monitoring 89 8.9 Surgery for intramedullary spinal cord
Chakorn Chansakul and Dileep R. Nair tumors 164
John H. Shin and Edward C. Benzel
7 Pharmacology of adjunct anesthetic
drugs 106 8.10 Avoiding complications: surgeons point
John E. Tetzlaf of view 171
Michael Kelly and Richard Schlenk
9 Anesthesia for cervical spine surgery 178
Section 2 Spine surgery for adult Alaa A. Abd-Elsayed and Ehab Farag
patients 10 Anesthesia for thoracic spine surgery 188
8 Surgical techniques 113 Rai Avitsian
8.1 Anterior cervical surgery 113 11 Lung isolation during thoracic spine
Iain H. Kalfas surgery 204
Gordon Finlayson and Jay B. Brodsky
8.2 Posterior cervical surgery 118
Kalil G. Abdullah, Jefrey 12 Anesthesia for lumbar spine surgery 216
G. Clark, Daniel Lubelski, and homas Mariel R. Manlapaz, Ajit A. Krishnaney, and
E. Mroz Zeyd Ebrahim
8.3 Intraoperative neurophysiologic 13 Anesthetic management of spinal cord
monitoring: surgeons point of view 122 trauma 228
Manuel Saavedra and Robert F. McLain Brian P. Lemkuil and Piyush M. Patel vii
Contents

14 Anesthesia for patients with spinal cord 21 Preoperative evaluation of the pediatric
tumors 247 patient 358
Stacie Deiner and Jefrey Silverstein Sara P. Lozano and Julie Niezgoda
22 Fluid management and monitoring of the
Section 3 Postoperative care of the pediatric patient 372
Tunga Suresh, Patrick M. Callahan, and
adult patient Peter J. Davis
15 Complications 257 23 Surgical techniques in the pediatric
15.1 Postoperative visual loss 257 patient 388
Lorri A. Lee and Raghu Mudumbai 23.1 Scoliosis 388
David P. Gurd
15.2 Other complications 270
Lorri A. Lee and Karen B. Domino 23.2 Tethered cord: surgical release 397
Mark Luciano
16 Postoperative care in the PACU 281
Maged Argalious 24 Spinal surgery for patients with congenital
heart disease and other associated
17 Postoperative care in the neuro-intensive
conditions 405
care unit 292
Patrick M. Callahan, Tunga Suresh, and
James K. C. Liu, Dani S. Bidros, and
Peter J. Davis
Edward M. Manno
25 Postoperative pain control in pediatric
18 Postoperative acute pain 302
patients 420
Juan P. Cata and Sherif Zaky
Rami Karroum, Loran Mounir Soliman, and
19 Postoperative chronic pain management 321 John Seif
Dmitri Souzdalnitski and Jianguo Cheng

Section 4 Spine surgery for pediatric Index 437


patients
20 Pathophysiology of the pediatric patient 347
Stephen J. Kimatian and Kenneth J. Saliba

viii
Contributors

Alaa A. Abd-Elsayed William Bingaman


Resident Physician, Department of Anesthesiology, Vice-Chairman, Neurological Institute, Department
University of Cincinnati, Cincinnati, OH, USA of Neurosurgery, Cleveland Clinic, Cleveland,
OH, USA
Basem Abdelmalak
Staf Anesthesiologist and Director, Anesthesia for Jay B. Brodsky
Bronchoscopic Surgery, Anesthesiology Institute, Professor, Department of Anesthesia,
Cleveland Clinic, Cleveland, OH, USA Stanford University School of Medicine,
Stanford, CA, USA
Kalil G. Abdullah
Cleveland Clinic Lerner College of Medicine, David Brown
Cleveland Clinic, Cleveland, OH, USA Chairman, Anesthesiology Institute, Cleveland
Clinic, Cleveland, OH, USA
Maged Argalious
Medical Director, PACU and Same Day Surgery; Patrick M. Callahan
Assistant Professor, Cleveland Clinic Lerner College Assistant Professor of Anesthesiology, University
of Medicine, Case Western Reserve University, of Pittsburgh School of Medicine, Department of
Cleveland, OH, USA Anesthesiology, Childrens Hospital of Pittsburgh,
Pittsburgh, PA, USA
Rai Avitsian
Section Head, Neurosurgical Anesthesiology; Juan P. Cata
Program Director, Neuroanesthesia Fellowship; Instructor, Department of Anesthesiology
Assistant Professor of Anesthesiology, Cleveland and perioperative Medicine, Division of
Clinic Lerner College of Medicine Anesthesiology Anesthesiology and Critical Care, he University
and Neurological Institutes, Cleveland Clinic, of Texas MD Anderson Cancer Center, Houston,
Cleveland, OH, USA TX, USA

Maria Bauer Chakorn Chansakul


Clinical Research Fellow, Department of Outcomes Clinical Neurophysiology Fellow, Epilepsy Center,
Research, Institute of Anesthesiology, Cleveland Cleveland Clinic Neurological Institute, Cleveland,
Clinic, Cleveland, OH, USA OH, USA

Edward C. Benzel Jianguo Cheng


Chairman, Department of Neurosurgery, Program Director of Pain Medicine Fellowship,
Neurological Institute, Cleveland Clinic, Cleveland, Departments of Pain Management and
OH, USA Neurosciences, Cleveland Clinic, Cleveland,
OH, USA
Dani S. Bidros
Cerebrovascular Center, Department of Jefrey G. Clark
Neurosurgery, Cleveland Clinic, Cleveland, Cleveland Clinic Lerner College of Medicine,
OH, USA Cleveland Clinic, Cleveland, OH, USA
ix
List of contributors

Peter J. Davis Rodolfo Hakim


Professor of Anesthesiology and Pediatrics, Section of Pediatric and Congenital Neurosurgery,
University of Pittsburgh School of Medicine, Department of Neurosurgery, Cleveland Clinic,
Anesthesiologist-in-Chief, Department of Cleveland, OH, USA
Anesthesiology, Childrens Hospital of Pittsburgh,
Pittsburgh, PA, USA Robert Helfand
Staf Anesthesiologist, Anesthesiology Institute,
Stacie Deiner Cleveland Clinic, Cleveland, OH, USA
Assistant Professor of Anesthesiology, Neurosurgery,
Geriatrics, and Palliative Care, Mount Sinai School of Iain H. Kalfas
Medicine, New York, NY, USA Surgeon, Center for Spine Health, Cleveland Clinic,
Cleveland, OH, USA
Xiao Di
Section of Pediatric and Congenital Neurosurgery, Rami Karroum
Department of Neurosurgery, Cleveland Clinic, Associate Staf Member, Department of Pediatric
Cleveland, OH, USA Anesthesiology, Cleveland Clinic, Cleveland,
OH, USA
Karen B. Domino
Professor, Department of Anesthesiology and Michael Kelly
Pain Medicine, University of Washington, Seattle, Department of Neurosurgery, Center for Spine
WA, USA Health, Cleveland Clinic, Cleveland, OH, USA
D. John Doyle Stephen J. Kimatian
Staf Anesthesiologist, Cleveland Clinic, Cleveland, Chair, Pediatric Anesthesiology, Childrens Hospital,
OH, USA Cleveland Clinic Foundation, Cleveland, OH, USA
Zeyd Ebrahim Christian Koopman
Vice-Chair for Operating Afairs, Department of Neuroradiology Fellow, Cleveland Clinic, Cleveland,
General Anesthesiology, Cleveland Clinic, Cleveland, OH, USA
OH, USA
Ajit A. Krishnaney
Ehab Farag Associate Staf Physician, Center for Spine Health;
Staf Anesthesiologist, Departments of General Associate Staf Physician, Neurological Surgery;
Anesthesia and Outcomes Research, Cleveland Associate Staf Physician, Cerebrovascular Center,
Clinic, OH, USA Cleveland Clinic, Cleveland, OH, USA
Gordon Finlayson Andrea Kurz
Clinical Instructor, Department of Anesthesiology, Department of Outcomes Research, Institute of
Pharmacology, and herapeutics, Division of Critical Anesthesiology, Cleveland Clinic, Cleveland, OH,
Care Medicine, University of British Columbia, USA
Vancouver, BC, Canada
Lorri A. Lee
Elizabeth A. M. Frost Assistant Professor, Department of Anesthesiology
Professor of Anesthesia, Mount Sinai Medical Center, and Pain Medicine, University of Washington,
New York, NY, USA Seattle, WA, USA
Matthew Grosso Brian P. Lemkuil
Center for Spine Health, Cleveland Clinic, Cleveland, Assistant Clinical Professor, Department of Anesthesia,
OH, USA University of California San Diego, CA, USA

David P. Gurd James K. C. Liu


x Staf Physician, Department of Orthopedic Surgery, Resident, Department of Neurosurgery, Cleveland
Cleveland Clinic, Cleveland, OH, USA Clinic, Cleveland, OH, USA
List of contributors

Sara P. Lozano University of Washington, Seattle, WA, USA


Staf, Pediatric Anesthesiology, Cleveland Clinic,
Cleveland, OH, USA Thomas E. Mroz
Director, Spine Surgery Fellowship Program,
Daniel Lubelski Neurological Institute, Center for Spine Health,
Cleveland Clinic Lerner College of Medicine, Departments of Orthopedic and Neurological
Cleveland Clinic, Cleveland, OH, USA Surgery, Cleveland Clinic, Cleveland, OH, USA

Mark Luciano Dileep R. Nair


Head, Pediatric and Congenital Neurosurgery, Section Head, Epilepsy Center, Cleveland Clinic
Pediatric Neurosciences, Cleveland Clinic, Cleveland, Neurological Institute, Cleveland, OH, USA
OH, USA
Julie Niezgoda
Ramez Malaty Staf, Pediatric Anesthesiology, Cleveland Clinic,
Neuroradiology Fellow, Cleveland Clinic, Cleveland, Cleveland, OH, USA
OH, USA
R. Douglas Orr
Mariel R. Manlapaz Staf Physician, Center for Spine Health and
Associate Staf, Department of General Department of Orthopedic Surgery, Cleveland Clinic
Anesthesiology, Cleveland Clinic, Cleveland, Foundation, Cleveland, OH, USA
OH, USA
Piyush M. Patel
Edward M. Manno Professor of Anesthesiology, Department of
Cerebrovascular Center, Department of Anesthesia, University of California, San Diego; Staf
Neurosurgery, Cleveland Clinic, Cleveland, Anesthesiologist, Veterans Afairs Medical Center,
OH, USA San Diego, CA, USA

Virgilio Matheus Jason E. Pope


Resident, Department of Neurosurgery, Cleveland Napa Pain Institute, Napa, CA, USA
Clinic, Cleveland, OH, USA
Manuel Saavedra
Robert F. McLain Division of Plastic Surgery and Neurosurgery Section,
Professor of Surgery and Director, Spine Surgery Department of Surgery, University of Puerto Rico,
Fellowship Program, Cleveland Clinic Center for Medical Sciences Campus, San Juan, Puerto Rico
Spine Health, Cleveland, OH, USA
Kenneth J. Saliba
Nagy Mekhail Surgeon, Department of Pediatric Anesthesiology,
Professor of Anesthesiology, Cleveland Clinic Childrens Hospital, Cleveland Clinic Foundation,
Lerner College of Medicine of Case Western Reserve Cleveland, OH, USA
University and Cleveland Clinic Spine Center,
Cleveland, OH, USA Richard Schlenk
Neurosurgery Residency Program Director, Center for
Doksu Moon Spine Health, Cleveland Clinic, Cleveland, OH, USA
Staf Neuroradiologist, Cleveland Clinic, Cleveland,
OH, USA John Seif
Associate Staf, Department of Pediatric
Loran Soliman Mounir Anesthesiology, Cleveland Clinic, Cleveland, OH,
Staf Member, Departments of General Anesthesia USA
and Pediatric Anesthesia, Cleveland Clinic,
Cleveland, OH, USA John H. Shin
Department of Neurosurgery, Center for Spine
Raghu Mudumbai Health, Neurological Institute, Cleveland Clinic,
Associate Professor, Department of Opthalmology, Cleveland, OH, USA xi
List of contributors

Jefrey Silverstein John E. Tetzlaf


Professor of Anesthesiology, Neurosurgery, Professor of Anesthesiology, Cleveland Clinic
Geriatrics, and Palliative Care, Mount Sinai School of Lerner College of Medicine of Case Western Reserve
Medicine, New York, NY, USA University; Staf Anesthesiologist, Department of
General Anesthesia, Cleveland Clinic, Cleveland,
Dmitri Souzdalnitski OH, USA
Clinical Fellow of Pain Medicine, Department of Pain
Management, Cleveland Clinic, Cleveland, OH, USA Sherif Zaky
Staf Anesthesiologist, Department of
Michael Steinmetz Anesthesiology, Cleveland Clinic, Cleveland, OH,
Center for Spine Health, Cleveland Clinic, Cleveland, USA
OH, USA

Tunga Suresh
Assistant Professor of Anesthesiology, University
of Pittsburgh School of MedicineDepartment of
Anesthesiology, Childrens Hospital of Pittsburgh,
Pittsburgh, PA, USA

xii
Foreword by Dr. Edward Benzel

Ehab Farag has assembled an incredible group of Second, it covers the ield. his cannot be under-
authors for this irst of a kind book, Anesthesia for estimated as an attribute. Anesthesia for Spine Surgery
Spine Surgery. his book is the irst comprehensive text is both deep and broad in content. It truly covers the
devoted to the anesthetic management of the spine sur- ield in many regards. It is thorough. It is comprehen-
gery patient. It addresses both the adult and pediatric sive. It is organized both to facilitate a reading from
domains. It is comprehensive in scope and detailed in cover to cover and to function as a reference.
content. hird, it is a great read. his book feels good in
Anesthesia for Spine Surgery addresses all aspects of my hands and I feel at home while reading it. It is,
the subject at hand. It begins with a group of disser- in fact, diicult to put down. his is, in part, related
tations on the variety of surgical procedures and the to its unique position in the spine surgery and anes-
complications that can ensue. Chapters on imaging thesia domain. It is also related to its wonderful
and surgical technique-speciic nuances follow. hese presentation.
are followed by a variety of chapters that address spe- his book covers everything, and I do mean
ciic spine anesthesia techniques and complications. everything, from visual complications to spinal cord
he book closes with a number of discussions regard- injury. No stones are let unturned. It is a complete,
ing pediatrics-related issues. comprehensive, and user-friendly text. It will soon
How good is this book? It is a must read. Why is it become a great addition to the libraries of all spine
a must read? surgeons, neurosurgeons, and orthopedic surgeons,
First, it is a irst of its kind. Hence, there is no com- and of all neuro-anesthesiologists, anesthesiologists,
petition. I am virtually certain that others will follow. and related professionals. It is, indeed, a classic in the
hey always do when pioneers open and develop a ield. making.

xiii
Foreword by Dr. David Brown

It is a real pleasure to be asked to contribute a foreword that is equally intricate. he willingness of surgeons
to Anesthesia for Spine Surgery by Dr. Ehab Farag and and anesthesiologists to share their approaches to
colleagues. his useful text covers the entire spectrum complex and straightforward spinal surgical patients
of surgical and anesthetic procedures required to be at allows readers of this book to apply the most current
the forefront of clinical care for our patients requiring approaches to minimizing morbidity and optimizing
spine care. Ehab has assembled an outstanding list of safety in their patients. he contributors to this book
contributors; all true experts in their area of clinical are made up of individuals with whom I would be fully
coverage. Many are colleagues at the Cleveland Clinic comfortable in having them providing care for myself
and others are recognized national experts. or members of my family at any time. hey truly are
Spinal surgery has developed over the last 20 years experts.
into a specialized ield in which increasingly com- I congratulate all involved in creating this import-
plex and lengthy surgical procedures are being car- ant work and hope that it really adds value to your care
ried out; these clearly demand an anesthetic approach of this important group of patients.

xv
Preface

Spine surgery has evolved in the last two decades into surgeons, neurologists, and other health care providers
a multidisciplinary specialty including surgery, pain caring for spine surgery patients. Included are chap-
management, neuromonitoring, and neuroradiology. ters for perioperative anesthetic management in adult
his book is the irst comprehensive textbook for anes- and pediatric patients, which cover the preoperative
thesia for spine surgery in both adults and pediatric evaluation, airway management, lung isolation, luid
patients. I believe that in order for the anesthesiologist management, and postoperative care both in the pos-
to provide adequate perioperative care, the anesthesi- tanesthesia care unit and in intensive care. he book
ologist should be armed with the proper knowledge of has chapters that cover in full detail the intraoperative
the diferent facets of the specialty for which he or she anesthetic management from cervical to lumbosacral
administers the anesthetic. his book serves that belief, regions. he anesthetic and surgical complications dur-
for it covers the techniques of surgical procedures ing spine surgery are discussed in full detail as well.
described to the anesthesiologist in a simple approach I would like to express my gratitude to my col-
by world-renowned surgeons. he other aspects of the leagues for their hard work and dedication in accom-
spine surgery specialty (pain management for acute plishing this book. I would also like to acknowledge the
and chronic pain ater spine surgery, intraoperative help and support of my editorial assistants at Cleveland
neuromonitoring, and neuroimaging) are fully cov- Clinic and the Cambridge University Press team.
ered by well-versed experts in those ields.
Ehab Farag, M.D., F.R.C.A.
his book is a very useful resource for anesthesi-
ologists involved in anesthesia for spine surgery, spine

xvii
Section 1 General considerations
Chapter
Preoperative assessment
of the adult patient
1 Elizabeth A. M. Frost

responsibility of the anesthesiologist or CRNA prac-


Key points
ticing alone. he process must consider information
Tests should be ordered only as indicated. from multiple sources including, among others, the
History and physical examination should patient, surgical and medical records, nurse evalua-
precede laboratory studies. tions, and laboratory tests and other tests. As appropri-
Variability of spinal anatomy and physiology ate, consultations may be sought and preoperative tests
dictate that cases involving spine surgery ordered as indicated. In some departments of anesthe-
should be considered on an individual basis. sia, informed consent is obtained separately from surgi-
General cardiac clearance is rarely useful. cal consent; in others consent is signed as part of general
Many diseases and comorbidities involve the hospital consent. In either case, the preanesthetic record
spine and must be considered separately. must note that the anesthetic options, risks and beneits
of anesthesia have been explained to the patient, and he/
she has agreed and accepted the plan.
Introduction Obtaining a history and physically examining the
Procedures on the spine vary in complexity from sim- patient should precede the ordering or performance of
ple discectomy to multi level reconstruction and fusion preanesthesia tests. Such a process includes evaluation
with instrumentation. Moreover, the level of surgery of pertinent medical records, patient interview, assess-
from the cervical area to the coccyx further impacts ment of the risk/beneit for diferent anesthetic tech-
planning for anesthetic management. Procedures may niques and a plan for postoperative pain management,
be planned for months or occur emergently as part of which is especially important for the patient scheduled
multiple trauma. hus many factors determine appro- to undergo complex lower spine surgery. he timing
priate preoperative anesthetic assessment. of the evaluation depends on the degree of surgical
invasiveness, where highly invasive procedures should
General guidelines be done prior to the day of surgery (multilevel lamin-
In 2002 the American Society of Anesthesiologists ectomies with instrumentation) and medium or low
developed a practice advisory to assist in decision mak- risk surgery (minimally invasive single-level laminec-
ing regarding appropriate preanesthetic assessment tomy) may be done on the day before or day of surgery.
and care.1 he advisory is the synthesis of opinions Analysis of the Practice Advisory of the ASA provides
from experts, open forums, public sources, and litera- a good indication of minimal standard of care in the
ture review. It is to be applied to all anesthesiologists United States. Airway assessment and documentation
and those who provide care under the direction of an was considered essential by 100% of consultants and
anesthesiologist including residents, certiied regis- 100% of ASA members. Pulmonary and cardiovascu-
tered nurse anesthetists (CRNA) or students. It applies lar examination was cited by 8188% of respondents
to all age groups and all types of anesthesia and deep as required. In addition, the health care system should
sedation for both surgical and nonsurgical situations. provide appropriate assessment of the severity of the
he advisory does not address emergency situations. patients medical condition and the invasiveness of sur-
Preanesthetic evaluation is the process of clinical gery. In other words, the diagnosis and planned sur-
assessment preceding the delivery of anesthesia. It is the gery should be identiied prior to operation.

1
Anesthesia for Spine Surgery, ed. Ehab Farag. Published by Cambridge University Press. Cambridge University Press 2012.
Section 1: General considerations

Routine testing implies tests that are performed Table 1.1 Upper cervical approaches
without clinical indication and include such items as Anterior Posterior
hemoglobin, urinalysis, chest radiograph (CXR), elec-
Transoral Craniocervical fusion/ixation
trocardiogram, coagulation proile and basic meta-
Transpalatal Atlantoaxial fusion
bolic panel. he reasons for ordering these tests as a
shotgun approach are varied and include such argu- Transmandibular C1C2 wiring/plating
ments as: Anterior retropharyngeal C1C2 transarticular
screw ixation
1. he approach represents good screening (although
for what, is not speciied).
2. It may save an annual physical examination Table 1.2 Middle and lower cervical approaches
(although mammography, colonoscopy, and Supine Prone
prostatic screening are usually recommended in
Anterior cervical discectomy Laminectomy, foraminotomy,
annual physical examinations, these results do not laminotomy
impact anesthetic management). Cloward procedure, Wiring: interspinous,
3. Preoperative testing is medicolegally sound includes insertion of sublaminar
autologous or bank bone
(but tests may cause more harm than good, may or methylmethacrylate
produce false positives or may not be reviewed,
Plating, pedicle screws
thus resulting in even more adverse situations).
4. Testing is required (most hospital do not have
mandated tests written into the policies and Table 1.3 Other levels and approaches
procedures). Anterior cervicothoracic
5. Multiple testing provides income for hospitals and Anterior thoracic: thorascopic techniques
laboratories (true). Posterior thoracic
A study done 20 years ago showed that routine test- Anterior lumbar/lumbosacral
ing cost >$60 billion annually and >60% of tests were Posterior lumbar
not indicated. Approximately 0.2% revealed pertin- Combined anterior posterior
ent abnormalities, that is, a inding that might change Minimally invasive and microdiscectomies
the anesthetic or surgical plan.2 here has been little
change in many areas.
is used in cases of cervical radiculopathy due to degen-
erative disease and in cases of cervical canal stenosis or
Site of surgery removal of intraspinal masses such as ependymomas.
Pathology, usually resulting in pain may occur Multiple levels may be involved.
throughout the cervical spine due to trauma, degen- Lumbar fusion is performed to relieve pain due
erative changes, tumors, lytic lesions, and compres- to intervertebral movement. Segmental instability,
sion. Levels for surgery may be upper cervical (C12), spondylolisthesis, and iatrogenic instability are other
middle and lower cervical, thoracic, lumbar and, more indications. Pedicle screw stabilization involves creation
rarely, sacral/coccygeal. he approaches at each level of a rigid three-column spinal ixation. Posterolateral
are shown in Tables 1.11.3. Anterior upper cervical fusions include laminectomy, discectomy, and grated
approaches are used to relieve compression at the bone to decorticated bone. Posterior lumbar interbody
cervicomedullary junction and stabilize odontoid fusion includes bilateral laminectomies with removal
fractures. hey may also be indicated in resection of of the inferior and part of the superior facets, disc-
tumors such as clival chordomas. Posterior approaches ectomies, autologous or banked bone placement or
are used to correct atlantoaxial or occipitalatlanto cages to the disc spaces. Combined anterior posterior
instability, odontoid and spinal fractures, and cervical approaches are indicated for correction of multilevel
instability. collapse, unstable three-column injury, severe kyphosis,
Supine approach to the middle lower cervical spine scoliosis, or infective or neoplastic conditions. Surgery
is indicated in the removal of osteophytes or herniated involves complete circumferential decompression,
2 discs. Fusion and/or instrumentation allow the disc rigid short segment ixation, and maximal correction
space height to be maintained. he posterior approach of deformities.
Chapter 1: Preoperative assessment adult patient

It is clear that preanesthetic assessment varies Approximately 15% of patients with Downs syn-
depending on the approach required, the pathology drome have atlantoaxial instability and the majority
involved, the invasiveness of the procedure, and the of them are asymptomatic.3 However, they are predis-
presence of other comorbidities, which may be part of posed to subluxation and cervical cord compression
the spinal disease. Assessment also varies depending especially during endoscopy and intubation. Speciic
on whether the surgery is elective or emergent history and physical examination are important to seek
out such symptoms as gait abnormalities, clumsiness
Cervical and elective surgery and increased fatigue when walking. Other indings
include abnormal neck motion (very mobile), upper
Upper levels and lower motor neuron signs such as spasticity, hyper-
High cervical cord tumors such as chordomas or relexia, extensor plantar relexes, loss of bowel or blad-
ependymomas may be resected through the mouth der control, and neck posturing (torticollis).4 he Sharp
an approach rarely used today. Many of these patients and Purser test may also be applied. With the patient in
are young and relatively healthy and require little pre- a sitting position, and the neck lexed, backward pres-
operative testing. However, psychological preparation sure is applied against the forehead while the spinous
includes the awareness that a tracheostomy is usually process of the axis is palpated. A gliding motion may
placed prior to surgery. he approach requires bisec- be felt as subluxation is reduced. he test is positive
tion of the tongue and mandible allowing direct access in about 50% of patients with atlantoaxial instability.
to the back of the mouth. Surgery necessitates the Posterior stabilization is recommended for patients
cooperation of several specialties and is generally very with subluxation prior to any other elective surgery.
long, lasting 24 hours or more. Decannulation of the Intubation should be performed with head stabiliza-
trachea can usually be accomplished within 72 hours tion including two-point pin ixation.
as swelling subsides in the upper airway. Blood loss is Approximately 40% of Downs syndrome patients
minimal. have some form of congenital heart disease, the most
Indicated tests: Complete radiographic series to prevalent of which is endocardial cushion defect.
determine the extent of the pathology. Other anomalies include ventricular and atrial septal
Trisomy 21 is one of the most common chro- defects. All three of these lesions may result in pulmon-
mosomal abnormalities in humans, occurring in ary hypertension. Adults appear to have a higher inci-
1:6800 live births in the United States. he anomaly dence of aortic insuiciency and mitral valve prolapse.
afects many organ systems with implications that hose with previously repaired congenital heart dis-
require close preanesthetic assessment (Table 1.4). ease may have conduction defects, usually let anterior
As life expectancy of these patients has improved, the hemiblock and right bundle branch block. Cardiology
number of adult patients with Downs syndrome pre- consultation may be indicated.
senting for surgery is increasing. Muscle hypotonia here is a general predisposition towards hypoxia.
and ligamentous laxity with atlantoaxial instability Respiratory tract infections secondary to airway
occur frequently and patients may present for stabi- anomalies, immunologic deiciencies and institu-
lization with plating, wiring, or some other means of tional living are contributing factors. Hypotonia and
fusion of C12. sleep apnea with both mechanical and central nervous

Table 1.4 Trisomy 21. Perianesthetic considerations


Anatomic Short stature, obesity, small mouth, large tongue, high arched palate, small mandible and maxilla, short neck
Musculoskeletal Hypotonia, lax joints, unstable atlantoaxial joint, temporomandibular joint disease
Cardiac Congenital defects, aortic insuiciency, mitral valve prolapse
Respiratory Frequent infections, pulmonary hypertension, sleep apnea, atelectasis, airway obstruction
Immune system Altered response, infections, hepatitis B, lymphocytic leukemia increased incidence
Neurologic Mental retardation, seizures, early onset presenile dementia, perioperative agitation
Gastrointestinal system Gastroesophageal relux
Hematologic Polycythemia
3
Endocrine Thyroid anomalies, especially in adults; decreased central and peripheral sympathetic activity
Section 1: General considerations

system factors add to hypoventilation and hypoxia. symptoms. If let untreated, symptoms may worsen
Preoperative assessment should include blood gas and require surgery. here are several possible causes.
analyses, pulmonary function testing, and training in he majority of syringomyelia cases are associated
basic respiratory therapeutic maneuvers. with ArnoldChiari malformation, a condition in
Patients of all ages with Downs syndrome tend to which brain tissue protrudes into the spinal canal. he
have lower blood pressures than normal and mentally malformation consists of a downward displacement of
handicapped controls, perhaps related to a decrease in the cerebellar tonsils through the foramen magnum,
central and peripheral sympathetic activity. Resting sometimes causing noncommunicating hydrocephalus
and stress levels of dopamine beta-hydroxylase, which as a result of obstruction of cerebrospinal luid (CSF)
converts dopamine to norepinephrine, are decreased. outlow. he cerebrospinal luid outlow is caused by
Excretion of epinephrine is decreased, which may be phase diferences in outlow and inlux of blood in the
due to decreased adrenal production even though vasculature of the brain. Any obstruction can cause
plasma epinephrine levels are normal. headaches, fatigue, muscle weakness in the head and
Indicated tests: Chest radiography to exclude aspi- face, diiculty swallowing, dizziness, nausea, impaired
ration or atelectasis, complete blood count to assess coordination, and, in severe cases, paralysis. While the
polycythemia and leukocytosis indicating infection, Chiari malformation may be present at birth, symp-
room air oxygen saturation, electrocardiogram to toms are oten delayed until the 2nd4th decade. Oten
assess conduction defects, thyroid function tests, neck symptoms may appear to be triggered by a fall or minor
radiography, blood glucose to exclude diabetes. trauma. Other causes of syringomyelia include spinal
Syringomyelia is the development of a luid-illed cord tumors, spinal cord injuries such as tethered cord
cyst (syrinx) within the spinal cord (Fig. 1.1). Over syndrome, and meningitis. Early signs and symptoms
time, the cyst may enlarge, damaging the spinal cord of syringomyelia may afect the back of the neck, shoul-
and causing pain, weakness, and stifness, among other ders, arms, and hands and include:
Muscle weakness and wasting (atrophy)
Loss of relexes
Loss of sensitivity to pain and temperature

Later signs and symptoms of syringomyelia are:


Stifness in the back, shoulders, arms, and legs
Pain in the neck, arms, and back
Bowel and bladder function problems
Muscle weakness and spasms in the legs
Facial pain or numbness
On neck lexion, a tingling sensation rapidly
spreading down the trunk and into the legs
(Lhermittes sign)

Syringomyelia can become a progressive disorder and


lead to complications such as scoliosis, Horners syn-
drome, and chronic pain. In other cases, there may
be no associated symptoms and no intervention is
necessary. Typically, surgery for syringomyelia, usu-
ally in the upper cervical spine with a posterior and
prone approach (especially for Chiari malformations)
includes one or more of the following:
Treatment of a Chiari malformation. A
suboccipital craniectomy is performed and a dural
grat may be added around C12 to enlarge the
4 Figure 1.1 Syringomyelia. The thin gray curved line indicates opening of the foramen magnum and restore the
extravasated cerebrospinal luid. low of cerebrospinal luid.
Chapter 1: Preoperative assessment adult patient

Draining of the syrinx. A shunt may be inserted Middle and lower levels; elective
from the syrinx to the abdomen or chest
(syringoperitoneal or syringopleural shunt). Less surgery
commonly, the syrinx can be drained at surgery. Anterior cervical discectomy is commonly performed
Removal of the obstruction. If there is blockage in the treatment of nerve root or spinal cord compres-
within the cord such as might be caused by sion. By decompressing the spinal cord and nerve roots
adhesions or a tumor, surgical removal of the of the cervical spine the corresponding vertebrae can
obstruction may restore the normal low and allow be stabilized and pain and paresthesias relieved. his
luid to drain from the syrinx. procedure is used when other nonsurgical treatments
Correction of an abnormality. If a spinal have failed.
abnormality is hindering the normal low of he nucleus pulposus of the herniated disc bulges
cerebrospinal luid, surgery to correct it such out through the annulus and presses on the nerve root
as releasing a tethered spinal cord may be next to it (Fig. 1.2). his nerve root becomes inlamed
corrective. and causes pain. he problem can also be caused by
degenerative disc disease (spondylosis). he disc
Preanesthetic assessment centers on history and physi- consists of about 80% water and with age dries out
cal examination. Duration and amount of symptoms and shrinks, causing small tears in the annulus and
as well as drug ingestion including narcotics and anti- inlammation of the nerve root. At surgery, the disc
depressants should be documented. All nonsteroidal is completely removed as well as any arthritic bone
analgesics should be discontinued 12 weeks before sur- spurs. To prevent the vertebrae from collapsing and
gery if possible. Frequently the diagnosis of syringomye- to increase stability, the open space is illed with bone
lia is delayed and the patient may have received multiple grat, taken from the pelvis or cadaveric bone or
consultations and many therapies. Neurologic examina- methylmethacrylate. Occasionally a titanium plate is
tion should include a review of any preexisting deicits. screwed on the vertebrae to increase stability during
Muscle and nerve assessments and electromyelography fusion, especially when there is more than one disc
may be indicated to gauge the extent of paresthesias or involved.
numbness. Range of motion of the neck may indicate Recently, endoscopic anterior cervical discectomy
a sharp increase in symptoms during lexion. Nausea, under epidurogram guidance has been described.5
vomiting, and diiculty swallowing should be assessed Contrast dye through a cervical discectomy is used to
and prophylactic therapy given as indicated. generate an epidurogram. Using luoroscopy, endo-
Indicated tests: MRI of the syrinx, CXR if a shunt scopic instruments are advanced to the epidural space
is to placed in the thorax, complete blood count, and and both sot and hard discs can be removed.
coagulation proile if the patient has received analgesic As with other surgery on the spine, ingestion of
medications. all nonsteroidal anti-inlammatory medicines (Advil,

Spinous process
Meninges
Foramen
(filled with adipose tissue) Gray matter
White matter
Dorsal root
Inferior articular process
Ventral root
Superior articular process
Spinal nerve

Posterior tubercle of
transverse process

Anterior tubercle of
transverse process Nucleus pulposus
Herniated disc impinging
Foramen transversium Vertebral body Disc annulus on spinal nerve
Figure 1.2 Vertebra and disc: normal (left) and herniated situation (right) (from http://en.wikidedia.org/wiki/Anterior_cervical_disectomy 5
and fusion).
Section 1: General considerations

Motrin, Nuprin, Aleve, etc.) and anticoagulants external branch of the superior laryngeal nerve. he
(Coumadin, clopidogrel, aspirin) should be stopped recurrent laryngeal nerve enters the pharynx, along
1 to 2 weeks before surgery. In some instances, as for with the inferior laryngeal artery, below the inferior
example recent placement of a drug eluding stent, constrictor muscle to innervate the intrinsic muscles of
discontinuing of clopidogrel may be ill advised. he the larynx responsible for controlling the movements
cardiologist and surgeon should be consulted and the of the vocal folds.
preferred therapy determined. Ingestion of herbs that Unilateral damage may cause hoarseness. Although
can interfere with coagulation should also be stopped. the right recurrent laryngeal nerve is more susceptible
For example, garlic decreases platelet aggregation, gin- to damage due to its relatively medial location, sur-
ger inhibits thromboxane synthetase, gingko inhibits gery is generally performed on the right side more for
platelet activating factor, ginseng interacts with all convenience of the right-handed surgeon. Indeed, the
the anticoagulants to increase their efectiveness, and total rate of persisting recurrent laryngeal nerve dam-
feverfew inhibits serotonin release from platelets. age using a right-sided approach approximates 13%, a
Additionally, the patient should be advised to stop number that can be reduced to 6.5% with a let-sided
smoking (pipes and cigarettes), inhaling snuf, and incision.8 he complication can be further reduced by
chewing tobacco at least 1 week before and 2 weeks ater controlling endotracheal tube (ETT) cuf pressure to
surgery (although complete cessation is preferable) <20 mmHg.9 Noting that cuf pressures are oten much
as these activities can cause bleeding problems. Also, higher than realized and may increase intraoperatively,
nicotine interferes with bone metabolism through proper control by manometer reduced ETT related
induced calcitonin resistance and decreased osteoblas- postprocedural respiratory complications, even in pro-
tic formation. Patients who smoked had failed fusions cedures of short duration. Problems such as cough and
in up to 40% of cases, compared with only 8% among sore throat were signiicantly reduced.10 However, the
nonsmokers.6 Smoking also retards wound healing and signiicance of reducing and controlling cuf pressure
increases the risk of infection.7 in reducing the incidence of vocal cord immobility has
Because risk of damage to the recurrent laryngeal been questioned by one study.11 Laryngeal damage may
nerve is a complication of this surgery, its function also relate to excessive pressures of the retractors on the
should be ascertained preoperatively. Many patients esophagus. he patient should be advised preopera-
with spinal disc disease are smokers and may have tively that while a short period of soreness in the throat
other causes of hoarseness. he recurrent laryngeal is not unusual, should it be accompanied by hoarseness
nerve is a branch of the vagus nerve that supplies motor or last longer than 23 days then further consultation
function and sensation to the larynx. It is referred to as should be sought.
recurrent because the branches of the nerve innerv- A preoperative plan for pain control should be
ate the laryngeal muscles in the neck through a rather in place. To this end it is important to know whether
circuitous route: it descends into the thorax before autologous bone grating is intended. Acute post-
rising up between the trachea and esophagus to reach operative pain and nerve injuries ater anterior iliac
the neck. he let laryngeal nerve branches from the crest bone grat can lead to neuropathic chronic pain.
vagus nerve to loop under the arch of the aorta, pos- A small study evaluated the eicacy of preoperative
terior to the ligamentum arteriosum before ascending. placement of transversus abdominis plane (TAP) block
he right branch loops around the right subclavian under ultrasound-guided technology.12 he authors
artery. Both nerves give of several cardiac ilaments to considered TAP to be an appropriate technique for
the deep part of the cardiac plexus. As they ascend in postoperative analgesia ater bone harvest as about
the neck, branches more numerous on the let than 80% of patients had no pain at 18 months. Whatever
on the right side are given of to the mucous mem- the approach, a plan for pain relief must be detailed as
brane and muscular coat of the esophagus; branches the discomfort experience from the hip is signiicantly
also supply the mucous membrane and muscular ib- greater than that felt in the neck. Other techniques
ers of the trachea, and some pharyngeal ilaments go to include local iniltration, patient-controlled analgesia,
the superior pharyngeal constrictor muscle. he nerve and opioid or nonsteroidal injections.
splits into anterior and posterior rami before supplying Blood loss during this procedure is minimal,
muscles in larynx and supplies all laryngeal muscles although there are rare instances of delayed bleeding
6 except for the cricothyroid, which is innervated by the from vessel damage and retropharygeal hematomas.
Chapter 1: Preoperative assessment adult patient

Indicated tests: Coagulation proile to ensure no postoperatively and functional ability improved mark-
adverse efects of medications or herbs; cervical radio- edly.15 Nevertheless, lung function should be assessed
logic studies to assess the extent of the disease; type and preoperatively to allow a program of respiratory ther-
screen to check for antibodies; complete blood count to apy to be instituted.
ensure adequate hemoglobin levels for optimal wound Cardiac assessment is also critical as many of these
healing. patients may have other problems that afect cardiac
function such as Marfans syndrome. Also, severe cur-
Thoracic levels: elective surgery vatures afect heart rhythm by altering the position
Scoliosis (from Greek: skolisis, crooked) is an of the heart within the chest. Echocardiographic and
abnormal lateral curvature of the spine. It is a complex stress tests as far as the patient can manage are indi-
three-dimensional deformity that is typically classiied cated. It is doubtful whether the patient can be made
as either congenital (caused by vertebral anomalies into any better condition based on results of these tests
present at birth), idiopathic (cause unknown, sub- except that pulmonary infection or aspiration might be
classiied as infantile, juvenile, adolescent, or adult identiied and corrected.
according to when onset occurred), or neuromuscular As with iliac crest grating, pain control becomes
(having developed as a secondary symptom of another a major problem postoperatively. horacic epidural
condition, such as spina biida, cerebral palsy, spinal general analgesia has been shown to be efective.14
muscular atrophy, or physical trauma). he condition Typically the surgery, performed in a prone pos-
afects approximately 20 million people in the United ition and with instrumentation, involves many levels
States.13 Although surgery is more oten performed in and lasts some 68 h. Blood loss may be considerable.
children in whom growth retardation is a signiicant Occasionally patients predonate which may make
problem as the disease progresses, a signiicant number them relatively anemic preoperatively. As a comple-
of adults also undergo the operation. Scoliosis may be mentary surgical procedure a thoracoplasty (or costo-
associated with other conditions such as EhlersDanlos plasty) may be performed to reduce the rib hump that
syndrome (hyperlexibility, loppy baby syndrome, afects most scoliosis patients with a thoracic curve.
and other variants including a high incidence of mitral horacoplasty may also be performed to obtain bone
valve prolapse), CharcotMarieTooth, PraderWilli grats from the ribs instead of the pelvis, regardless of
syndrome, kyphosis, cerebral palsy, spinal muscular whether a rib hump is present. horacoplasty can be
atrophy, muscular dystrophy, familial dysautonomia, performed as part of a spinal fusion or as a separate
CHARGE syndrome (coloboma, heart abnormali- surgery and involves the resection of typically four to
ties, atresia of the nasal choanae, growth retardation, six segments of adjacent ribs that protrude. Each seg-
genital and urinary anomalies, ear deformities, and ment is 2.55 cm long, (the ribs grow back, straight).
deafness), Friedreichs ataxia, proteus syndrome, spina he most common complication of thoracoplasty
biida, Marfans syndrome, neuroibromatosis, con- is increased pain in the rib area during recovery.
nective tissue disorders, congenital diaphragmatic her- Another complication is temporarily reduced pul-
nia, hemihypertrophy, and craniospinal axis disorders monary function (1015% is typical) following sur-
(e.g., syringomyelia, ArnoldChiari malformation). gery. his impairment can last anywhere from a few
Depending on the existence of comorbidities that months to 2 years. Hemothorax and pneumothorax
may be present in >50% of patients, preanesthetic may also occur. Because thoracoplasty may lengthen
assessment may be complex.14 Of most importance is the duration of surgery, patients may also lose more
the assessment of ventilatory function, which may be blood or develop complications from the prolonged
considerably compromised if the curvature exceeds anesthesia.
40. Several comorbidities such as Duchenne muscu- Indicated tests: Type and crossmatch blood, which
lar dystrophy and spinal muscular atrophy are oten will almost certainly be required; coagulation proile
associated with poor pulmonary function tests and if the patient has been receiving analgesics; complete
the simultaneous development of scoliosis will further blood count, especially if predonation has taken place;
aggravate the situation. However, at least one study CXR to rule out infection, pulmonary function tests
showed that despite preoperative forced vital capac- as a baseline of ventilatory function; cardiology con-
ity of <30% and further slight decreases postopera- sult for other (oten rare) comorbidities; assessment of
tively, there were no major pulmonary complications Cobb and pelvic angles. 7
Section 1: General considerations

Tumors may metastasize to the thoracic spine. Lung 31%


Other pathologies that arise in the thoracic area and Breast 24%
require surgery include infection such as tuberculosis, GI tract 9%
and hemangiomas. Potts disease (named ater Percivall Prostate 8%
Pott, 17141788, a London surgeon) is a presentation Lymphoma 6%
of extrapulmonary tuberculosis that afects the inter- Melanoma 4%
vertebral joints. It is most commonly localized in the
Unknown 2%
thoracic portion of the spine. Potts disease results from
Kidney 1%
hematogenous spread of tuberculosis from other sites,
oten pulmonary. he infection then spreads from two Others, including multiple myeloma, 13%
adjacent vertebrae into the adjoining intervertebral Preanesthetic assessment depends on the cause.
disc space. If only one vertebra is afected, the disc is Oten surgery for the primary tumor has been under-
normal, but if two are involved the disc, which is avas- taken, but if vertebral collapse and pain have become
cular, cannot receive nutrients and collapses to be bro- the prominent feature then attempts may be under-
ken down by caseation, leading to vertebral narrowing taken to stabilize the thoracic spine. he approach is
and eventually to vertebral collapse and spinal damage. usually prone and may be prolonged with consider-
A sot tissue mass may form. able blood loss. Many patients have already under-
Spinal metastasis is common in patients with can- gone chemotherapy or radiation therapy and may be
cer. he spine is the third most common site for cancer debilitated.
cells to metastasize, following the lung and the liver. Indicated tests: Identiication of primary tumor;
Approximately 6070% of patients with systemic can- complete blood count and metabolic panel for cancer
cer develop spinal metastasis of which only 10% are patients; CXR to rule out infection or metastasis; car-
symptomatic and 9498% have epidural and/or verte- diac evaluation for patient with general debility.
bral involvement.16 Vertebroplasty is a minimally invasive procedure
Spread from primary tumors is mainly by the performed in the thoracic area in which a iller mater-
arterial route. Retrograde spread through the Batson ial (traditionally polymethylmethacrylate, PMMA) is
plexus during Valsalva maneuver is postulated. Direct injected percutaneously into a vertebral body for treat-
invasion through the intervertebral foramina also can ment of vertebral fractures associated with osteopor-
occur. Besides mass efect, an epidural mass can cause osis, malignant conditions, hemangiomas, and acute
cord distortion, resulting in demyelination or axonal trauma.17
destruction. Vascular compromise produces venous Osteoporosis, or porous bone, is a disease charac-
congestion and vasogenic edema of the spinal cord, terized by low bone mass and structural deterioration
resulting in venous infarction and hemorrhage. of bone tissue. he resulting bone fragility increases
About 70% of symptomatic lesions are found in the fractures of all bones, especially the hip, spine, and
thoracic region of the spine, particularly at the level of wrist, (especially the spine). hese fragility fractures
T4T7. Of the remainder, 20% are found in the lumbar cause both acute and chronic pain and are a major
region and 10% are found in the cervical spine. More source of morbidity and mortality. One in every two
than 50% of patients with spinal metastasis have sev- women and one in four men over 50 years of age will
eral levels of involvement. About 1038% of patients have an osteoporosis-related fracture in their lifetime.
have involvement of several noncontiguous segments. he demographics of this population are predomi-
Intramural and intramedullary metastases are not as nantly whites and females of Asian descent, many with
common as those of the vertebral body and the epidural the comorbidities of the geriatric group. Other risk
space. Isolated epidural involvement accounts for less factors include poor nutritional status, a history of a
than 10% of cases; it is particularly common in lymph- primary relative with bone fragility, inactive lifestyle,
oma and renal cell carcinoma. Most of the lesions are early menopause, smoking, steroid use, and alcohol
localized at the anterior portion of the vertebral body use. Diagnosis depends on a bone mineral density test
(60%). Rarely is there disease in both posterior and and a dual-energy X-ray absorptiometry (DEXA) test.
anterior parts of the spine. In vertebroplasty, the PMMA is injected directly
Primary sources for spinal metastatic disease are as into the bone, whereas in kyphoplasty it is injected ater
8 establishment of a cavity by inlation of a balloon tamp.
follows:
Chapter 1: Preoperative assessment adult patient

As the ability of the procedure to dramatically decrease Indicated tests: Cardiac evaluation in an elderly
pain has been repeatedly demonstrated, it has been debilitated patient; assessment of volume status; com-
applied to many more and sicker patients. It is used as plete blood count to exclude anemia; basic metabolic
a palliative treatment for osteoporotic and malignant panel to exclude renal and hepatic disease; CXR to
vertebral lesions, which weaken vertebrae and cause exclude pneumonia and aspiration; assessment of mul-
chronic pain. tiple medications for drug interactions especially nar-
Preanesthetic assessment is critical in these patients cotic patches; urine analysis to exclude urinary tract
as many of them have signiicant comorbidities. Some infection; availability of an interpreter as many patients
of the considerations include: are of Asian descent
1. Cardiac disease
2. Pulmonary compromise Lumbar level: elective surgery
3. Urinary tract infection Many surgical procedures are performed on the lum-
4. Multiple medications and interactions, including bar spine ranging from a simple minimally invasive and
herbal therapies endoscopic discectomy to multilevel reconstructive
5. Metastatic disease procedure with complex instrumentation. Similarly,
6. Poor nutritional status anesthetic management varies from little more than
7. Narcotic dependency local anesthesia with some sedation to general anes-
thesia. Recently there has been resurgence in the use
8. Limited mobility
of regional anesthesia, especially for single-level disc-
9. Communication diiculties
ectomies. Advantages for this approach are decreased
As age increases so do cardiovascular and pulmonary blood loss, better pain management (especially if epi-
comorbidities. he decrease in pulmonary function dural clonidine is added to the technique), less nausea
associated with osteoporotic vertebral fracture may and vomiting, and a decreased incidence of deep ven-
be clinically signiicant in a patient with already ous thrombosis.20 Endoscopy may be used posteriorly
reduced pulmonary and cardiovascular reserve. for discectomy and anteriorly for instrumentation.
Previously pulmonary function in the osteoporotic Spinal stenosis is the single most common diagnosis
patient was described as normal, perhaps because leading to any type of spine surgery, and laminectomy
height at age 25 years and not current height was is a basic part of the surgical treatment. he lamina of
used in pulmonary function test calculations.18 the vertebra, which itself is not damaged, is removed
On adjusting for this change, a statistically signii- to widen the spinal canal and create more space for
cant decrease in vital capacity and FEV1 suggesting the spinal nerves and thecal sac. Surgical treatment
restrictive lung pattern may be identiied. Also, the that includes laminectomy is the most efective rem-
mortality rate from pulmonary disease (not lung edy for severe spinal stenosis; however, most cases of
cancer) is increased with osteoporotic vertebral spinal stenosis are not severe and respond to bed rest,
fractures. Signiicant improvement in pulmonary nonsteroidal anti-inlammatory agents, and steroids.
function has been shown ater vertebroplasty and Should symptoms include numbness, loss of function,
kyphoplasty, improvement that might increase for and neurogenic claudication, laminectomy is generally
up to 3 months.19 Metastatic lesions may cause ver- indicated. If the spinal column is unstable then fusion
tebral fractures. he primary source may or may not with instrumentation is required.
have been identiied. In addition to the efects upon Spondylolisthesis describes the anterior displace-
the pulmonary system, vertebral fractures also afect ment of a vertebra or the vertebral column in relation
the gastrointestinal system. Loss of vertebral height to the vertebrae below (Fig. 1.3). his pathology occurs
decreases abdominal space and compromises gastro- most commonly in the lumbar spine. A hangmans
intestinal function. Prophylactic antacid therapy is fracture is a speciic type of spondylolisthesis where
indicated. Long-term opioid use causes constipation the C1 vertebra is displaced anteriorly relative to the C2
and decreases nutrient absorption. Insomnia and vertebra due to fractures of the C2 vertebras pedicles.
depressive efects of chronic pain adversely afect Patients presenting for major spinal surgery are
psychological well-being and many patients are also more likely to be male with truncal obesity and fre-
maintained with antidepressants, especially such quently have multisystem disease. Some of the more
herbal preparations as St. Johns Wort. typical indings are shown in Table 1.5.
9
Section 1: General considerations

Table 1.5 The patient for major spine surgery often has many Several studies suggest that such an approach is not
comorbidities
always taken. A multiple-choice survey regarding the
Findings and symptoms Anesthetic implications purposes and utility of cardiology consultations was
Hypertension Well controlled? Medications? sent to randomly selected New York metropolitan area
Efects of general anesthesia? anesthesiologists, surgeons, and cardiologists.22 here
Smoking Respiratory function? Wound was disagreement on the importance and purposes of
healing? Postoperative care? a cardiology consultation on topics such as intraoper-
Obesity Obstructive sleep apnea? ative monitoring, clearing the patient for surgery, and
Airway diiculties? Pulmonary advising as to the safest type of anesthesia and avoidance
hypertension?
of hypoxia and hypotension. his advice was regarded
Diabetes mellitus Perioperative glucose control?
as important by most cardiologists and surgeons but
Multiple pain managements Drug interactions? as unimportant by anesthesiologists. Also, the charts
Renal disease Diuretic therapy? Coronary of 55 consecutive patients aged more than 50 years
artery disease?
who received preoperative cardiology consultations
Hematologic anomalies Anemia? polycythemia? were examined to determine the stated purpose of the
consultation, recommendations made, and concord-
ance by surgeons and anesthesiologists with the rec-
ommendations. Of the cardiology consultations, 40%
contained no recommendations other than proceed
with case, cleared for surgery, or continue current
medications. A review of 146 medical consultations
suggested that the majority of such consultations give
little advice that impacted either perioperative man-
agement or outcome of surgery.23 In only 5 consulta-
tions (3.4%) did the consultant identify a new inding;
62 consultations (42.5%) contained no recommenda-
tions. herefore, careful history taking and physical
evaluation by the anesthesiologist is essential as not
only can situations change over a few weeks but factors
that are critical in anesthetic management may appear
of less signiicance to the cardiologist.
he history should seek to identify cardiac condi-
tions that have been shown to impact perioperative
morbidity and mortality such as unstable coronary
syndromes, prior angina, recent or past myocardial
infarction, decompensated heart failure, signiicant
dysrhythmias, and severe valvular disease.21 A prior
history of placement of a pacemaker or implantable
cardioverter deibrillator (ICD) or a history of ortho-
static intolerance is important. Modiiable risk factors
Figure 1.3 Spondylolisthesis. The body of L5 can be seen slipped
over the sacrum. for coronary heart disease (CHD) should be recorded,
along with evidence of associated diseases, such as
peripheral vascular disease, cerebrovascular disease,
Spinal fusion and instrumentation is major sur- diabetes mellitus, renal impairment, and chronic pul-
gery that is generally planned for many months. hus monary disease. In patients with established cardiac
patients have oten undergone extensive evaluation disease, any recent change in symptoms must be ascer-
before they are even seen in a preanesthetic assessment tained. Accurate recordings of current medications
clinic. However, in consultations key questions should used, including herbal and other nutritional supple-
be identiied to ensure that all of the perioperative ments, and dosages are essential. Alcohol and tobacco
10 caregivers are considered when providing a response.21 use and ingestion of over-the-counter and illicit drugs
Chapter 1: Preoperative assessment adult patient

Table 1.6 METS chart for energy cost of activities

Level Self-care Housework Recreation Work


METS 13 Bathe, dress, comb Wash dishes, dust, set table Walk 2 mph, read, TV, play piano Type, desk, occasional lifting up
hair, put on shoes to 10 lbs
METS 34 Shower, climb stairs, Laundry, weeding, vacuuming, Walk 3 mph, bowl, golf with cart, Light repair work, painting small
wash hair, driving make bed ish from boat jobs, occasionally lift to 20 lbs
METS 45 Sexual intercourse Digging, wax loors, move Walk 3.5 mph, golf and carry clubs, Mix cement, occasionally lift to
furniture, wash car doubles tennis, bicycling 50 lbs, painting exteriors
METS 57 Hanging clothes Split wood, climb ladder, put up Walk 45 mph, singles tennis, Heavy farming, occasionally lift
storm windows softball, cross-country skiing 50100 lbs, heavy industry
METS >8 Saw hardwood by hand, push Jogging 5 mph, football, downhill Heavy construction, frequent
and pull hard, move furniture skiing, cross-country running lifting and carrying (>50 lbs)

should be documented. he history should also seek to with medication. Also, the long-term ingestion of anti-
determine the functional capacity (Table 1.6). hypertensive agents, especially ACE inhibitors, may
Assessment of a patients capacity to perform a cause a decrease in blood pressure intraoperatively,
spectrum of common daily tasks correlates well with necessitating early administration of vasopressors.
maximum oxygen uptake by treadmill testing.24 A Other signs that should be sought include carotid pulse
patient classiied as high risk owing to age or known contour and bruits, jugular venous pressure and pulsa-
CAD but who is asymptomatic and runs for 30 min- tions, auscultation of the lungs, precordial palpation
utes daily may need no further evaluation. In contrast, and auscultation, abdominal palpation, and examina-
a sedentary patient, such as one immobilized by severe tion of the extremities for edema and vascular integrity.
back pain, without a history of cardiovascular disease he inding of weak or absent arterial pulses conirms
but with clinical factors that suggest increased peri- the diagnosis of underlying cardiovascular disease.
operative risk, may beneit from a more extensive pre- Although rales and chest radiographic evidence
operative evaluation. of pulmonary congestion correlate well with elevated
Physical examination should include a review of pulmonary venous pressure in acute heart failure, in
general appearance. Cyanosis, pallor, dyspnea during patients with chronic failure these indings may be
conversation or with minimal activity, CheyneStokes absent and an elevated jugular venous pressure or a
respiration, poor nutritional status, obesity, skel- positive hepatojugular relux are more reliable signs.
etal deformities, tremor, and anxiety are indicators of Peripheral edema is not a reliable indicator of chronic
underlying disease and/or coronary artery disease. A failure unless the jugular venous pressure is elevated or
long-standing history of hypertension and hyperlipi- the hepatojugular test is positive. During cardiac aus-
demia is common, usually treated with several medi- cultation, a third heart sound at the apical area suggests
cations including diuretics, angiotensin-converting a failing let ventricle, but its absence is not a reliable
enzyme (ACE) inhibitors, calcium channel and beta indicator of good ventricular function. Presence of a
blockers, and a statin. In assessing blood pressure, sev- cardiac murmur may or may not be signiicant. For
eral values should be taken from both arms. An elevated example, aortic stenosis poses a higher risk for noncar-
blood pressure may indicate white coat syndrome diac surgery.25 Even if aortic regurgitation and mitral
or failure to adhere to the prescribed regimen. Some regurgitation are minimal, they predispose the patient
patients (and other health care workers) believe that to infective endocarditis should bacteremia occur.
an order nothing to eat or drink excludes all medica- he basic clinical evaluation obtained by history,
tions. If the blood pressure is indeed recorded within a physical examination, and review of the ECG usually
normal range, it is important to remember that, as in provides suicient data to estimate cardiac risk. he
a diabetic patient in whom one measurement is nor- resting 12-lead ECG has been examined perioper-
moglycemic, the disease is still present. In other words, atively to evaluate its prognostic value. Lee et al. studied
even though a mean blood pressure of 6070 mmHg 4135 patients aged 50 years or older undergoing major
may be well tolerated by a young man with normal vas- noncardiac surgery (more than-2 day stay).26 he pres-
culature, such a level may well be too low for someone ence of a pathological Q wave on the preoperative ECG
who may have a baseline mean pressure of 110 (140/90) was associated with an increased risk of major cardiac 11
Section 1: General considerations

complications, deined as an infarction, pulmonary Table 1.7 Cardiac conditions that require further evaluation
edema, ventricular ibrillation, primary cardiac arrest, Unstable coronary Decompensated cardiac
or complete heart block. Pathological Q waves were syndrome failure
found in 17% of the patient population. Based on these
Severe angina Worsening or new onset failure
indings, the authors derived a simple index for the
Recent myocardial infarction Symptomatic dysrhythmias
prediction of cardiac risk for stable patients undergo-
High-grade AV block Severe valvular disease
ing nonurgent major noncardiac surgery. Independent
risk factors included: Mobitz II AV block
1. Ischemic heart disease (history of myocardial
infarction, positive treadmill test, use of 2. Inducing a hyperemic response by
nitroglycerin, chest pain, or ECG with abnormal pharmacological vasodilators such as
Q waves) intravenous dipyridamole or adenosine.
2. Congestive heart failure (history of failure,
pulmonary edema, paroxysmal nocturnal he most common examples presently in use are
dyspnea, peripheral edema, bilateral rales, S3, dobutamine stress echocardiography and intraven-
radiography with pulmonary vascular diease ous dipyridamole/adenosine myocardial perfusion.
3. Cerebral vascular disease (transient ischemic Results of these studies have shown that reversible per-
attack or stroke) fusion defects, which relect jeopardized viable myo-
4. High-risk surgery (major vascular or orthopedic cardium, carry the greatest risk of perioperative cardiac
surgery) death or magnetic imaging with both thallium-201 and
technetium-99m.
5. Need for insulin treatment for diabetes mellitus
Dobutamine stress echocardiography (DSE) is
6. Preoperative creatinine greater than 2 mg/dl.
the method of choice for pharmacological stress test-
Increasing numbers of risk factors correlate with ing with ultrasound imaging. Incremental infusion of
increased risk, yet the risk was lower than described supratherapeutic doses of dobutamine increase myo-
in many of the original indices. Improvements in out- cardial contractility and heart rate, thus inducing let
come may relect selection bias in surgery, advances ventricular ischemic regional wall-motion abnormal-
in surgical technique, anesthesia, and perioperative ities within the distribution of stenotic vessels. he
management of coronary artery disease. he Revised dobutamine infusion may be supplemented with intra-
Cardiac Risk Index has become one of the most widely venous atropine to optimize chronotropic response to
used risk indices.26 stress. Intravenous contrast imaging for let ventricu-
In contrast to these indings, Liu and colleagues lar opaciication may enhance the image and improve
studied the predictive value of a preoperative 12-lead diagnostic interpretation
ECG in 513 patients aged 70 years or older undergoing Several reports have documented the accuracy
elective or urgent noncardiac surgery.27 In this cohort, of DSE to identify patients with signiicant angi-
75% of the patients had a baseline ECG abnormality, ographic coronary disease and indicate that DSE can
and 3.7% of the patients died. Electrocardiographic be performed safely with acceptable patient tolerance.
abnormalities were not predictive of outcome. he Positive test results range from 5% to 50%. he predic-
optimal time interval between obtaining a 12-lead tive value of a positive test is 0% to 33% for events such
ECG and elective surgery is unknown, but general con- as myocardial infarction or death. Negative predictive
sensus suggests that an ECG within 30 days of surgery value range from 93% to 100%.21 he presence of a new
is adequate for those with stable disease. wall-motion abnormality appears to be a powerful
Cardiac conditions indicating the need for further determinant of an increased risk for adverse periopera-
evaluation prior to noncardiac surgery are summa- tive events ater multivariable adjustment for diferent
rized in Table 1.7. clinical and echocardiographic variables.28 he value
Two main techniques are used in preoperative of DES in prediction of perioperative events is further
evaluation of patients who cannot exercise: enhanced by integration with other risk factors such
1. Increasing myocardial oxygen demand (by pacing as angina or diabetes. An ischemic response at 60% or
or intravenous dobutamine) more of maximal predicted heart rate was associated
12 with only a 4% event rate if no clinical risk factors were
Chapter 1: Preoperative assessment adult patient

present versus a 22% event rate in patients with more Table 1.8 Estimated times for return of lung function after
stopping smoking
than 2 risk factors.29 hese indings have been shown to
be predictors of long-term and short-term outcomes.30 Elimination of nicotine 12 h
Beattie et al. conducted a meta-analysis (68 studies) Elimination of carboxyhemoglobin 13 days
comparing stress myocardial perfusion imaging versus Return of ciliary function 67 days
stress echocardiography in 10,049 patients at risk for Decrease of sputum production 68 weeks
MI before elective noncardiac surgery and concluded Normalization of immune system >8 weeks
that both myocardial perfusion imaging and stress
echocardiography detected a moderate-to-large defect
in 14% of patients that was predictive of myocardial of carbon dioxide retention through arterial blood gas
infarction and/or death.31 Mondillo et al. compared analysis may be justiied. If there is evidence of infec-
the predictive value and determined that the predic- tion, appropriate antibiotic therapy is critical. Steroids
tors were the severity and extent of ischemia (dipyrida- and bronchodilators may be indicated, although
mole, p <0.01; dobutamine, p <0.005).32 Only reversible the risk of producing dysrhythmias or myocardial
perfusion defects at scintigraphy were signiicantly ischemia by beta-agonists and hyperglycemia must be
related to perioperative events. he strongest predictor considered. Complete abstinence from tobacco intake
of cardiac events was the presence of more than three for several weeks prior to surgery would be ideal to
reversible defects (p <0.05). A meta-analysis of 58 stud- allow regeneration of lung function but is rarely pos-
ies indicated that perioperative cardiac risk appears to sible. Times required for regeneration of various func-
be directly proportional to the amount of myocardium tions are approximated in Table 1.8.
at risk as relected in the extent of reversible defects A carboxyhemoglobin level of 15% can reduce the
found on imaging.33 Because of the overall low posi- availability of oxygen by up to 25% and while this level
tive predictive value of stress nuclear imaging, it is best may not be signiicant in asymptomatic patients, it may
used selectively in patients with a high clinical risk of present a considerable risk for patients with coronary
perioperative cardiac events. artery disease in whom a favorable myocardial balance
Current recommendations regarding continuing is critical. hese patients should be advised to refrain
drug therapies are: from smoking for at least 24 h prior to surgery. While
1. If the patient is medicated with statins and or beta- pulmonary function tests are not usually helpful in pre-
blockers these medications should be continued. dicting postoperative pulmonary events or the need for
2. Beta-blocking therapy should not be started de mechanical ventilation, a low preoperative oxygen room
novo. air saturation, or low partial pressure of arterial oxygen
3. Some clinicians advise discontinuing ACE may identify patients at higher risk. Other important
inhibitors on the day before surgery. factors in determining postoperative pulmonary com-
4. Patients with a bare metal stent should not have plications include the site and duration of the surgical
surgery for a month, and patients with a drug- procedure and the amount of blood lost. Preoperative
eluding stent should continue clopidogrel for a pulmonary therapy might be useful, if only to acquaint
year and perhaps even longer.34 the patient with the several tools that may be used in the
postoperative period to maintain oxygenation. Clinical
studies suggest that smoking is a risk factor in the pro-
Smoking gression of kidney disease, especially diabetic neph-
he presence of either obstructive or restrictive pul- ropathy. Nicotine promotes mesangial cell proliferation
monary disease places the patient at increased risk of and ibronectin production, and smoking may promote
developing perioperative respiratory complications, the progression of diabetic nephropathy by increasing
compounded by placement in the prone position for the expression of proibrotic cytokines such as trans-
several hours during surgery. Hypoxemia, hypercap- forming growth factor and the extracellular matrix pro-
nia, acidosis, and increased work of breathing can all teins ibronectin and collagen IV.35 hus, evaluation of
lead to further deterioration of an already compro- kidney function and glycemic status, especially in dia-
mised cardiopulmonary system. If signiicant pulmo- betic patients, is important in the smoker and limitation
nary disease is suspected, documentation of response of the use or doses of drugs dependent on the kidney for
to bronchodilators and/or evaluation for the presence excretion is indicated. 13
Section 1: General considerations

Examination of the airway is a prerequisite of all replacement surgery at the Mayo Clinic found that 24%
anesthetic encounters. Nicotine is a signiicant risk fac- of OSAS patients vs. 9% of controls had serious post-
tor for the development and progression of periodontal operative complications (dysrhythmias, myocardial
disease.36 he drug probably acts by decreasing gin- ischemia, re-intubations, and unplanned intensive care
gival blood low, increasing cytokine production, and unit admissions) with most occurring within 72 h of
adversely afecting the immune system to cause loosen- surgery, and resulting in increased hospital length of
ing of teeth and actual tooth loss. Chewing tobacco also stay.43 Evidence-based risk-screening tools that facili-
causes tooth decay due to the high sugar content. Oral tate preoperative risk stratiication of the undiagnosed
cancers and leukoplakia may interfere with intubation OSAS patient, thereby allowing appropriate periop-
or oral airway placement due to bleeding or ulceration. erative care, may help reduce the above-mentioned
Additional exposure to ethanol appears to enhance complications. Once the potential for OSAS has been
adverse changes in the buccal mucosa in vitro in more identiied, the patient may have a formal polysomno-
than an additive efect.37 Noting that surgery might graphic sleep study, which deines the patients apnea/
present a teachable moment, several studies have been hypopnea index (AHI), categorizes the severity of
undertaken to assess the beneit of referring patients OSAS as mild, moderate, or severe, and makes recom-
preoperatively to telephone quitlines.38,39 While many mendations for appropriate continuous positive air-
of these reports are on small groups, there is beneit in way pressure CPAP or nasal CPAP (nCPAP). he use
advising patients on the adverse efects of tobacco and of nCPAP for several weeks preoperatively has been
referring them for further help, especially by facilitat- found to be highly efective at preserving airway pat-
ing referral of smokers for counseling and follow-up. ency during sleep and anesthesia as well as diminishing
Indeed, the preanesthetic assessment interview has relex responses to hypoxia and hypercapnia. his efect
been deemed the teachable moment. Smoking has may result from upper airway stabilization, a residual
been identiied as a major factor in failure to fuse and efect of nCPAP that begins to occur within as little
continued pain ater lumbar surgery.4041 as 4 h of continuous use of nCPAP. Chronic nCPAP
preoperatively has been found to abolish mean, systo-
Obesity lic, and diastolic blood pressure luctuations in OSAS
he obese patient presents many perioperative prob- patients.43 As a result, the risk of cardiac ST segment
lems ranging from the initial assurance of a satisfac- depression and recurrent atrial ibrillation is reduced.
tory airway, to position diiculties, to a high incidence It is recommended that nCPAP and oral appliances be
of obstructive sleep apnea and pulmonary hyperten- continued during the postoperative period. It is also
sion. During the preanesthetic assessment, the airway important to note that patients who have had correc-
must be carefully assessed for ease of intubation. Neck tive surgery for OSAS, such as uvulopalatopharyngo-
circumference >40 cm (17 inches) and heavy jowls, plasty, may still harbor the disease despite lessening or
oten combined with small mouth opening and a large absence of current symptoms.43
tongue may compromise the situation. Appropriate he undiagnosed OSAS patient proves to be a
plans should be made, including informing the patient greater diagnostic dilemma during the preoperative
of a possible awake intubation technique. As spinal dis- screening clinic examination because these patients
ease is rarely conined to only the lumbar area, assess- seldom have sleep studies, which makes accurate risk
ment of a range of motion of the neck must also be stratiication (low, moderate, or severe) to guide intra-
documented as must notation be made of the patients and postoperative management diicult. A presumed
ability to lie lat. diagnosis of OSAS can be inferred from a history of
Obstructive sleep apnea syndrome. he American abnormal breathing during sleep (e.g., loud snoring
Society of Anesthesiologists has introduced guidelines and witnessed apnea by a bed partner), frequent arous-
for anesthetic management of an increasing popula- als from sleep to wakefulness (e.g., periodic extremity
tion of patients who snore and are obese but who have twitching, vocalization, turning, and snorting), severe
not been formally diagnosed as sufering from sleep daytime sleepiness, a BMI of 35 kg/m2, increased neck
apnea, a syndrome that can have serious complications circumference (40 cm [17 inches] for males, 38.5 cm
perioperatively.42 Knowledge of these potential prob- [16 inches] for females), and the presence of coexist-
lems is required for all anesthesiologists. A recent ret- ing morbidities (e.g., essential systemic hypertension,
14 rospective, case control study of patients having joint pulmonary hypertension, cardiomegaly).
Chapter 1: Preoperative assessment adult patient

he ASA task force on OSAS (May 2006) recom- In summary, clinical suspicion of OSAS may be the
mended a risk scoring system.42 he risk score was only preoperative tool available to the anesthesiolo-
achieved through expert opinion, literature review, gist as formal and widely used; preoperative validated
and consensus but it remains to be validated. he questionnaires have not been established.
ASA Risk Score considers the following with points
being assigned for each of three categories (a, b, c) and Pulmonary hypertension
then totaled (d): Symptoms of pulmonary hypertension may develop
(a) Severity of sleep apnea: Based on a sleep study (i.e., very gradually and include shortness of breath, fatigue,
AHI) or clinical indicators if a sleep study is not nonproductive cough, angina pectoris, fainting or syn-
available (i.e., presumptive diagnosis). Points: cope, peripheral edema, and rarely hemoptysis, all of
0 = none; 1 = mild OSA; 2 = moderate OSA; 3 = which may be confused by immobility due to back
severe OSA. pain, obesity, and myocardial ischemia. Pulmonary
One point may be subtracted if a patient has been venous hypertension typically presents with shortness
on CPAP or bilevel positive airway pressure (BiPAP) of breath while lying lat or sleeping (orthopnea or par-
prior to surgery and will be using this consistently oxysmal nocturnal dyspnea), while pulmonary arterial
during the postoperative period. One point should hypertension (PAH) usually does not.
be added if a patient with mild or moderate OSA has A history of exposure to drugs such as cocaine,
a resting PaCO2 exceeding 50 mmHg. methamphetamine, alcohol leading to cirrhosis, and
(b) Invasiveness of the surgical procedure and tobacco leading to emphysema is signiicant. Typical
anesthesia: Based on type of surgery/anesthesia. signs of pulmonary hypertension, include a widely split
Points: 0 = supericial surgery under local or S2 or second heart sound, a loud P2 or pulmonic valve
peripheral nerve block, anesthesia without sedation; closure sound (part of the second heart sound), sternal
1 = supericial surgery with moderate sedation heave, possible S3 or third heart sound, jugular venous
or general anesthesia or peripheral surgery with distension, pedal edema, ascites, hepatojugular relux,
spinal or epidural anesthesia (with no more than and clubbing. Tricuspid insuiciency is consistent with
moderate sedation); 2 = peripheral surgery with the presence of pulmonary hypertension.
general anesthesia or airway surgery with moderate Procedures to conirm the presence of pulmonary
sedation; 3 = major surgery under general anesthesia hypertension and exclude other possible diagnoses
or airway surgery under general anesthesia. include pulmonary function tests; blood tests to exclude
(c) Requirement for postoperative opioids: Points: 0 = HIV, autoimmune diseases, and liver disease; electro-
none; 1 = low-dose oral opioids; 3 = high-dose oral cardiography (ECG); arterial blood gas analyses; CXR
opioids or parenteral or neuraxial opioids. (followed by high-resolution CT scanning if interstitial
lung disease is suspected); and ventilation-perfusion
(d) Estimation of perioperative risk: Based on the
or V/Q scanning to exclude chronic thromboembolic
overall score (06) derived from the points
pulmonary hypertension. Biopsy of the lung is usually
assigned to (a) added to the greater of the points
not indicated unless the pulmonary hypertension is
assigned to (b) or (c).
thought to be due to an underlying interstitial lung dis-
Patients with overall score of 4 may be at increased ease. Lung biopsies are fraught with risks of bleeding
perioperative risk from OSA. Patients with a score of due to the high intrapulmonary blood pressure. Brain
5 may be at signiicantly increased perioperative risk natriuretic peptide level is also used to follow the pro-
from OSA. gress of patients with pulmonary hypertension.
Recently, a study was conducted to evaluate a Diagnosis of PAH requires the presence of pul-
new abbreviated version of the ASA questionnaire monary hypertension with two other conditions.
the STOP Questionnaire.44 Four questions are asked Pulmonary artery occlusion pressure (PAOP or
related to Snoring, Tiredness during the day, Observed PCWP) must be less than 15 mmHg (2000 Pa) and
apnea, and high blood Pressure. Ater screening 2467 pulmonary vascular resistance (PVR) must be greater
patients, the authors concluded that this screening tool than 3 Wood units (240 dyn s cm5 or 2.4 mN s cm5).
was reliable and easy to use. Combined with body mass Although pulmonary arterial pressure can be
index, age, neck size, and sex it had a high sensitivity, estimated on the basis of echocardiography, pressure
especially in patients with moderate to severe OSAS measurements with a pulmonary artery provide the 15
Section 1: General considerations

most deinite assessment. PAOP and PVR cannot be be advised to discontinue the use of herbal prepara-
measured directly with echocardiography. herefore tions such as ginseng, garlic, gingko, and ginger, all of
diagnosis of PAH requires right-sided cardiac cath- which may interfere with coagulation. Most patients
eterization. Cardiac output is more important in presenting for laminectomy have received steroids
measuring disease severity than pulmonary arterial either as a 7-day pack to decrease swelling or as part of
pressure. Normal pulmonary arterial pressure in a per- pain management (epidural steroid injections). While
son living at sea level has a mean value of 1216 mmHg perioperative supplementation of steroids is no longer
(16002100 Pa). Pulmonary hypertension is present advocated, ingestion of this class of drugs should be
when mean pulmonary artery pressure exceeds 25 noted to perhaps explain hyperglycemia or cardiovas-
mmHg (3300 Pa) at rest or 30 mmHg (4000 Pa) with cular instability intraoperatively.
exercise. Mean pulmonary artery pressure (mPAP)
should not be confused with systolic pulmonary artery Renal impairment
pressure (sPAP), which is oten recorded on echocar- Renal dysfunction is associated with cardiac disease,
diogram reports. A systolic pressure of 40 mmHg typi- diabetes, and an increased risk of cardiovascular events.
cally implies a mean pressure of more than 25 mmHg. Preexisting renal disease (preoperative serum creatinine
Roughly, mPAP = 0.61sPAP + 2. levels >2 mg/dl or greater or reduced glomerular iltra-
tion rate) has been identiied as a risk factor for post-
Diabetes mellitus operative renal dysfunction and increased long-term
Type 2 or insulin-resistant diabetes is a common ind- morbidity and mortality compared with patients with-
ing in patients presenting for major back surgery and out renal disease.21,48 Coronary artery bypass patients
its presence heightens the suspicion for cardiac disease; who are more than 70 years old with preoperative cre-
Lee et al. identiied insulin therapy as a signiicant risk atinine levels greater than 2.6 mg/dl are at much greater
factor for cardiac morbidity.26 Older patients with dia- risk for chronic dialysis postoperatively than those with
betes mellitus are more likely to develop cardiac failure creatinine levels below 2.6 mg/dl. One large study has
postoperatively than those without diabetes mellitus conirmed that a preoperative creatinine level greater
even ater adjustment for treatment with ACE inhibi- than 2 mg/dl is a signiicant, independent risk factor for
tors.45 Perioperative management of blood glucose cardiac complications ater major noncardiac surgery.21
levels may be diicult as stress and steroid administra- Creatinine clearance, which incorporates serum
tion will cause hyperglycemia. However, many stud- creatinine, age, and weight to provide a more accurate
ies have shown that wound healing is impaired and assessment of renal function than serum creatinine
neurologic damage increased when excess sugar is alone has been used to predict postoperative complica-
metabolized.4647 tions.49 Ater major surgery, mortality increased when
Patients need careful treatment with adjusted doses both serum creatinine increased and creatinine clear-
or infusions of short-acting insulin based on frequent ance decreased, with creatinine clearance providing a
blood sugar determinations. more accurate assessment

Drug interactions Hematologic disorders


Commonly patients presenting for major back surgery Patients are oten advised to predonate their blood prior
have been in severe and chronic pain for years. hey to major back surgery. As a result they may come to
have attended many clinics and have undergone mul- the hospital relatively anemic. Smokers generally have
tiple therapies. hey may have resorted to over-the- a higher hematocrit and thus these patients while pre-
counter medications and herbal remedies. Frequently senting with values of 36% or even higher may normally
they receive multiple narcotic patches and may have have values of >50%. Notation should be made of avail-
become relatively resistant to the efects of opioids able autologous blood with a plan to replace it promptly.
(in fact, they are addicts). Antidepressants including Anemia imposes a stress on the cardiovascular system
monoamine oxidase inhibitors and selective serot- and was also identiied in the American Society of
onin reuptake inhibitors are oten among their drug Anesthesiologists Postoperative Visual Loss Registry as
armamentarium. Identiication and documentation a risk factor.50
of ingested substances is essential for safe selection Hematocrit <28% is associated with an increased
16 of agents intra- and postoperatively. Patients should incidence of perioperative ischemia and postoperative
Chapter 1: Preoperative assessment adult patient

complications. In the VA National Surgical Quality Emergency spine surgery


Improvement Program database, mild degrees of pre-
Emergency spine surgery is generally related to cervical
operative anemia or polycythemia were associated with
injuries. However, among individuals sufering general
an increased risk of 30-day postoperative mortality and
traumatic injury, the cervical spine is involved in 4.3%
cardiac events in older, mostly male veterans undergo-
of cases, the thoracolumbar spine in 6.3% of cases, and
ing major noncardiac surgery.51 he adjusted risk of
the spinal cord in 1.3% of cases.52 In adults, the most
30-day postoperative mortality and cardiac morbidity
susceptible areas of injury include C5C7 and the tho-
begins to rise when hematocrit levels decrease to less
racolumbar junction, T12L1, areas of the vertebral
than 39% or exceed 51%.
column with the greatest mobility. While patients may
Polycythemia, thrombocytosis, and other condi-
present for release of epidural hematomas ater neu-
tions that increase viscosity and hypercoagulability
raxial anesthesia (very rarely) or ater vertebroplasty or
may increase the risk of thromboembolism or hemor-
other retroperitoneal surgeries, these procedures are
rhage. Appropriate steps to reduce these risks should
oten not diagnosed or undertaken for several hours.
be considered and tailored to the individual patients
he incidence of cervical spine injury (CSI) asso-
particular circumstances.
ciated with blunt trauma is about 0.93%.53 Types of
trauma cases include motor vehicle accidents (5075%),
Consent issues falls (610%), and recreational injuries (515%).5254
Several issues require special acknowledgement dur-
About 20% of patients have more than one cervical
ing the preanesthetic assessment. he patient should
spine fracture. Approximately 2075% of cervical
be aware that there will most probably be a need for
spine fractures are considered unstable and 3070%
blood transfusion. Awake iberoptic intubation may be
are associated with neurologic injuries. Prognosis for
indicated and the patient may require continued venti-
recovery from complete cervical cord lesions is poor
latory support postoperatively. In practice guidelines,
and emphasis must be placed on preventing extension
he American Society of Anesthesiologists have rec-
of neurologic injury once trauma has occurred. Factors
ommended the placement of an arterial cannula but
include: level of consciousness, need for respiratory
not of a central venous catheter or pulmonary artery
assistance, level of the lesion, and severity of neurologic
catheter, given the limited amount of useful informa-
injury, among others. he primary causes of CSI-related
tion obtained from these two monitors in determin-
deaths are cardiac and respiratory complications.5254
ing therapy.50 Anesthesiologists are also advised to tell
Perhaps the most convincing case of the importance
patients that there is a low but real risk of postopera-
of early care of CSI is that of actor Christopher Reeves.
tive blindness. he American College of Surgeons has
A sports-related cervical CSI, Reeves injury brought
been silent on this issue to date. Postoperative visual
to national attention the plight of the patient sufering
loss is probably not a single entity with a single cause.
from trauma-related respiratory, cardiovascular, and
Rather multiple factors have been associated including
sustained systemic damage. hanks in no small part to
long surgery in the prone position, blood loss, relative
his acute trauma care, Reeves was able to regain some
hypotension, diabetes, excessive luid administration,
peripheral nervous function before his untimely death
comorbidities including hypertension, and obesity,
in October 2004 due to cardiovascular complications.
among others.
Indicated tests: Type and crossmatch blood for intra-
operative use; complete blood count to establish base- Acute injury
line; coagulation proile to ensure reversal of efects of hirty-nine percent of fractures occur at C6 and C7,
all anticoagulant medications; room air saturation to with the vertebral body being the most common ana-
assess pulmonary function; appropriate cardiac evalu- tomical site of fracture and 24% of injuries occurring
ation depending on the patients status; CXR to ensure at C2.53 Instability occurs when vertebral displacement
there are no infective processes that require preopera- jeopardizes the spinal cord or nerve roots. To maintain
tive treatment; basal metabolic panel to obtain elec- stability, one element of the injured column must be
trolyte and sugar levels; creatinine clearance for renal preserved. he anterior column contributes more to
function; brain natriuretic peptide as further assess- the stability of the spine in extension. he posterior
ment of cardiac function; carotid and deep venous column adds more during lexion. In hyperextension
Doppler scan to assess other vascular disease. injuries, the anterior elements tend to be disrupted, 17
Section 1: General considerations

whereas in hyperlexion injuries the posterior elem- Spinal shock occurs rapidly ater a complete lesion
ents are disrupted. Both columns may be disrupted of the spinal cord as sympathetic tone is interrupted
with extreme lexion or extension or if either compres- below the level of the injury. Symptoms include hypo-
sive or rotational forces are added.52 Primary mech- tension, bradycardia, decreased peripheral resistance,
anical injury is caused by compression, penetration, loss of bowel/bladder function, loss of sensation and
laceration, shear and/or distraction forces, resulting in deep tendon relexes (DTRs), and total paralyses below
immediate neural damage due to avulsion and devital- the level of the lesion with lail limbs.55 Respiratory
ization of tissues. Spinal cord blood low is severely insuiciency and pulmonary dysfunction are common
reduced within the irst 3060 minutes of injury due in injuries to the cervical spinal cord. Marked reduc-
to hypertensive vasogenic edema as a result of initial tions in expected vital capacity, inspiratory capac-
catecholamine release. Loss of autoregulation leads to ity, and relative hypoxemia occur in severely injured
ischemia and tissue hypoxia. Global perfusion com- patients exacerbating cord ischemia.53 A few key land-
promise from systemic hypotension and tissue hypox- marks of spinal trauma aid in localization:
emia from hypoventilation associated with head injury 1. C3,4,5 innervate the diaphragm; apnea indicates
exacerbate existing perfusion deicits.53 injury above this level.
Fracture and dislocation cause cord compres- 2. Diaphragmatic breathing alone suggest injury at
sion and ischemia. Cord injury may also occur from C5T1.
laceration, contusion, or concussion by bony frag- 3. Presence of the biceps relex with absence of
ments.52 Mechanisms responsible for secondary cord triceps function suggests integrity of C6 and loss
injury including vascular compromise lead to reduced of C7 function.
blood low, loss of autoregulation, vasospasm, throm- 4. Injury above T5 may see the conversion of laccid
bosis, and hemorrhage. Electrolyte shits, permeability paralysis to spastic paralysis, positive Babinski
changes, loss of cellular membrane integrity, edema, sign, return of deep tendon relexes, smaller
and loss of energy metabolism all contribute to pro- vesical capacity with accompanying frequency of
gressive injury. Biochemical changes including neu- micturition, and the mass relex which may occur
rotransmitter accumulation, arachidonic acid release, 23 weeks ater injury.
free radical production, prostaglandin production,
and lipid peroxidation cause axonal disruption and Diagnosis/treatment
cell death.52,54 Glutamate release from damaged cells of
Although there is considerable variation in imaging of
the CNS is responsible for the excitotoxic component
the patient at risk for cervical spine injury, most cent-
of secondary injury. Increased protease activity, loss of
ers rely on multiple plain radiographs (at least three
mitochondrial function, and increased oxidative stress
views) of the cervical spine supplemented by CT scan
as a result of overactivation of glutamate receptors begin
of areas that are diicult to visualize or suspicious for
a cascade of events resulting in selective cell death and
injury. Obtundation, coexisting distracting (i.e., pain-
demyelination around the site of injury leading to an
ful) injuries, head injuries, and intoxication make it
increase in lesion size and scar development.52
impossible to clinically assess many patients without
imaging. If there is evidence of neurologic deicit sug-
Clinical features gesting cervical injury despite normal radiographs and
Although victims of multiple trauma including head CT imaging, MRI may be useful.53
injury usually present to the Emergency Room with he goals of treatment of injuries to the spinal cord
a neck collar in place, traumatic injury to the cervical are to protect the cord from secondary damage, main-
spine alone only occurs in about 20%. Systemic inju- tain alignment of bony structures, and to stabilize the
ries cause hypotension and hypoxia and necessitate spinal column to allow for rehabilitation. Considerable
urgent airway management to preserve cerebral func- controversy has existed over the timing of surgical
tion. Intubation and ventilation should be achieved intervention, but current practice is to operate within
as expeditiously as possible. Patients with cord injury 24 h of injury.56 Surgical indications include: decom-
and systemic injury typically show reduced neurologic pression with or without fusion in a patient with neu-
recovery and increased mortality. However, it is uncer- rologic deterioration, reduction and stabilization when
18 tain whether this poorer outcome is due to more severe conservative management has failed, and surgical
primary injury or the progression of secondary injury.54 intervention for other life-threatening conditions.55
Chapter 1: Preoperative assessment adult patient

Methylprednisolone was advocated some years of trauma patients require a deinitive secured airway
ago as treatment for primary and secondary injury immediately including those with apnea, a Glasgow
in acute spinal cord injury.57 However, current opin- coma scale <9 or sustained seizure activity, unsta-
ion has moved away from this therapy.58 A recent sur- ble mid-face trauma, airway injuries, large lail chest
vey reported that 76% of spine surgeons do not use segment(s) or respiratory failure, high aspiration risk,
methylprednisolone for acute spinal cord injury, a or inability to maintain an airway or oxygenation.
reversal from the practice 5 years previously. In fact, For both complete and incomplete lesions of the spi-
one-third of physicians report they administer methyl- nal cord, manipulation can aggravate the injury and
prednisolone only out of fear of litigation.59 Practice cause ascending deterioration. herefore, the goal is
guidelines and the role of methylprednisolone in the to establish endotracheal intubation without caus-
treatment of acute spinal cord injury had been based ing further damage to the spinal cord.54 Perhaps the
on the National Acute Spinal Cord Injuries Studies most important factor in determining the best tech-
(NASCIS I, II, and III). However, evidence-based nique for intubation is the urgency of the situation.
medicine now suggests that although methylpred- he anesthesiologist must evaluate and assess the risk
nisolone results in neurologic improvement in certain of further cord injury taking into consideration head
types of acute spinal cord injury, its role in prevention and neck movement, the degree of cooperation from
of secondary spinal cord injury remains unclear.60 the patient, anatomy of and trauma to the airway, and
Furthermore, in light of the proven harmful side efects his or her expertise with airway techniques (iberoptic
of high-dose steroids including an increased incidence intubation, Glidescope, Eschmann or Shikani stylets,
of wound infection, pulmonary embolism, hypergly- etc). A cricothyrotomy kit should be available. Collars
cemia, and gastrointestinal hemorrhage, further study whether sot or rigid do not efectively eliminate move-
would deine beneits and limitations of steroid use in ment of the neck during intubation. Manual in-line
acute spinal cord injury.61 Yet a recent study once more stabilization is more efective in immobilizing the neck
indicated improvement in spinal cord function when during intubation, but may cause excessive distraction
methylprednisolone was combined with mouse nerve in C1C2 fractures.52 If possible a neurologic assess-
growth factor.62 ment ater intubation may assure that there has been no
A number of treatments that may require anesthetic change in neurologic status. However, no data suggest
involvement on a semi-emergent basis have bridged the that better neurologic outcomes are achieved by this
translational gap and currently either are in the midst means. In fact, failed awake intubation has been identi-
of human CSI trials or are about to begin such clinical ied as a cause of morbidity and mortality according
evaluation. hese include minocycline, Cethrin, anti- to the latest analysis of diicult airways claims by the
Nogo antibodies, systemic hypothermia, Riluzole, American Society of Anesthesiologists Closed Claims
magnesium chloride in polyethylene glycol, intra- Project (www.asaclosedclaims.org,).1
peritoneal octreotide, modulation of growth factors, Direct laryngoscopy ater induction of anesthesia is
intrathecal administration of magnesium sulfate, and considered an acceptable option encouraged in emer-
human embryonic stem cell-derived oligodendrocyte gent and urgent situations.52 he American Society of
progenitors.6365 Anesthesiologists algorithm should be followed (www.
In all cases airway, breathing, and circulation must asahq.org). Use of axial in-line stabilization reduces
be assessed and addressed even if surgery is not imme- cervical movement by 60% and is preferred to leaving
diately planned or to facilitate placement of a patient a hard collar in place. However, direct laryngoscopy
in traction as a temporizing treatment. Resuscitation may cause greater spinal movement than indirect tech-
and stabilization followed by prevention of secondary niques such as iberoptic intubation. Atlantooccipital
damage to the spinal cord by spinal immobilization extension is necessary to bring the vocal cords into
and airway management are priorities. Cervical spine the line of sight of the mouth. herefore, patients with
injury should be suspected in all mechanisms of injury unstable C1C2 injuries might be at more risk with
involving blunt trauma. direct laryngoscopy. While not ruled out for C-spine
Continuous reassessment of the patient with sus- injury, an awake and alert patient without neck pain
pected cervical spine injury that is fully awake, talk- has minimal risk of such injury. In an intoxicated,
ing, and maintaining their own airway is warranted as comatose patient, the risk of C-spine injury must be 19
the status may deteriorate suddenly. Several categories assumed until a complete diagnostic work-up can
Section 1: General considerations

prove otherwise. A patient able to shrug the shoulders cord injury above T6. It is believed that aferent stimuli
and outwardly rotate the arms is deemed to have intact trigger and maintain an increase in blood pressure via a
C5 innervation. If no other injuries present, acute ven- sympathetically mediated vasoconstriction in muscle,
tilatory support may not be needed. Hypotension when skin, and splanchnic vascular beds. Ascending infor-
seen with bradycardia and hypothermia would suggest mation reaches the major splanchnic sympathetic out-
a high injury, whereas hypotension could also be a low (T5T6) and stimulates a sympathetic response.
sequelae of any of the following conditions: myocardial he sympathetic response causes vasoconstriction
injury, pneumothorax, or occult bleeding.55 In the case below the level of the injury, resulting in hypertension.
of a high injury, serial analyses of forced vital capac- his hypertension stimulates the baroreceptors in the
ity (FVC) and negative inspiratory force is helpful in carotid sinuses and aortic arch. he parasympathetic
determining respiratory functioning. During the com- system is unable to counteract these efects through
prehensive evaluation, opioids and benzodiazepines the injured spinal cord. However, instead, through
should be avoided because of possible depressed ven- the brainstem it attempts to maintain homeostasis by
tilation; however, atropine may be necessary to treat stimulation of the vagus nerve causing bradycardia and
bradycardia. vasodilation above the level of the spinal injury. he
As noted, not all cases come to surgery urgently and parasympathetic impulses are unable to descend past
some may be delayed for days or longer. However, in the lesion, and therefore no changes occur below the
all cases successful surgical management can only be level of injury.
addressed ater a comprehensive preoperative evalu- hus should a noxious stimulus inhibit local vaso-
ation. Airway, breathing, and circulation must be irst constriction, the body responds with autonomic hyper-
assessed and addressed as noted above. Both anter- relexia and causes a generalized vasoconstriction. his
ior and posterior surgical approaches may be used. response can be severe enough to require vasoactive
Cervical discectomy and fusion are oten indicated in drugs as treatment. Additionally, autonomic hyper-
the repair of trauma to the vertebral body. he immedi- relexia will not occur during spinal shock. If a history
ate management of every penetrating spinal cord injury of spasticity is present, or if elicitation of sustained
is the main indication for neurosurgical intervention motor responses upon relex testing occurs without
using laminectomy. Objectives for laminectomy are history of spasticity, then the patient is considered at
to relieve spinal cord or spinal root pressure made by risk for autonomic hyperrelexia. Using either gen-
comminuted bone fragments, blood clots, or foreign eral or spinal anesthesia, as opposed to epidural or
bodies. Especially in cervical cord injury, decompres- axillary analgesia, or sedation, can prevent the relex.
sion of the one or two roots above the transecting lesion Autonomic dysrelexia presents as headache, sweat-
and the subsequent restoration of function could mean ing, severe hypertension, and bradycardia. While it has
the diference between quadriplegia and paraplegia. been recognized that suicient general anesthesia is
Patients with transection above T6 may present efective at controlling spasms and autonomic dysre-
later in the course of their injury for stabilizing pro- lexia, the adverse efects of hypotension and respira-
cedures. Autonomic hyperrelexia (AH) may be tory dysfunction necessitate caution. If a patient with
present, characterized by hypertension, bradycardia, a low-level complete CSI presents without a history of
and vasodilation. It is characterized by hypertension autonomic dysrelexia or troublesome spasms, anes-
associated with throbbing headaches, profuse sweat- thesia may not be necessary.
ing, nasal stuiness, and lushing of the skin above It should be noted that the cardiovascular response
the level of the lesion. Bradycardia, apprehension, and to intubation might change over time, especially in
anxiety, sometimes accompanied by cognitive impair- paraplegics where the pressor response may become
ment are common. AH is believed to be triggered by exaggerated but is abolished in quadriplegics.66
aferent stimuli which originate below the level of the
spinal cord lesion. he stimulus is mediated through Systems assessment
the central and peripheral (somatic and autonomic)
nervous systems. As the name implies, the autonomic Respiratory system
nervous system is responsible for the signs and symp- he diaphragm is innervated by C3C5 and contributes
toms of AH. Normally there is a balance between sym- about 65% of ventilation; therefore a spinal cord injury
20 pathetic and parasympathetic systems lost in spinal above C4 causes respiratory failure. In acute cases of
Chapter 1: Preoperative assessment adult patient

spinal cord injury, there is a high risk of pulmonary failure.52 Invasive monitoring is oten required while
edema.52 he edema may be caused by luid overload transesophageal echocardiography can monitor the
and further exacerbated by eforts of resuscitation. changes in chamber size of the heart when adding luid.
Aspiration oten occurs. Coexisting chest injury may But controversy remains: while many prefer a pulmon-
be present. As the patient cannot generate an efective ary artery catheter, others weigh the beneits of luid
cough to clear secretions, atelectasis and pneumonia administration and consequent raising of pulmon-
develop rapidly following acute injury ary wedge pressure over inotropic vasoconstrictor
Treatment with diuretics, antibiotics, bronchoscopy, agents, mindful that luids can precipitate the risk of
or positive end-expiratory pressure is indicated.55 pulmonary edema and extravascular lung water.52
Chronic CSI patients commonly sufer from
decreased respiratory function due to muscle weak- Musculoskeletal system
ness. Consequently, many of these patients depend Musculoskeletal complications involve peripheral
on mechanical ventilation and are vulnerable to cholinergic responses and osteopenia. Acetylcholine
ventilation-associated pneumonia, involving bacter- receptors are upregulated, and patients can sufer from
ial stasis in uncleared pulmonary secretions. A major spasticity. Succinylcholine is contraindicated due to
complication of chronic CSI is hypoxemia second- the risk of hyperkalemia. Moreover, the bone density
ary to decreased functional residual capacity, where of these patients is in a compromised state and leaves
the administration of supplemental O2 is necessary. the patient vulnerable to osteoporosis, hypercalcemia,
Chronic CSI patients are also at an increased risk for heterotropic ossiication, and muscle calciication.55
aspiration due to an impaired airway.52 Decubitus ulcers and infection are common
Genitourinary system
Cardiovascular system
Patients may sufer long-term problems with bladder
If the spinal cord injury is suiciently high, sympathetic
function, leaving them vulnerable to ascending and
innervation is interrupted, leaving parasympathetic recurring urinary tract infections. Bladder obstruc-
input unopposed and leading to an increased vascu-
tion and infections can lead to further complications
lar space and pooling of blood in compliant vessels as such as pyelonephritis, sepsis, and amyloidosis, which
alpha-receptor-mediated vasoconstriction from sym-
can ultimately lead to renal insuiciency and failure.
pathetic input is virtually absent, resulting in hypo- Management of urinary tract infections and sepsis is
thermia, hypotension, and bradycardia. Hypothermia
complicated by an immunocompromised state.55 Using
is exacerbated in the patient unable to vasoconstrict in less invasive procedures can reduce the risk of nosoco-
response to cold. Bradycardia results due to unopposed
mial infection.
M2 muscarinic receptor action in cardiac myocytes,
unopposed by sympathetic input to beta-1 receptors, Gastrointestinal system
which normally increase heart rate. hese changes Gastrointestinal complications occur in up to 11% of
can lead to complications such as circulatory instabil- patients ater CSI consisting of ileus, gastric distension,
ity and hypotension. Atropine is the drug of choice to peptic ulcer disease, hemorrhage, acalculous cholangi-
block bradycardia. tis, and pancreatitis.52 A high index of suspicion must
Lack of adequate perfusion to the spinal cord be maintained for occult acute abdomen, as the usual
secondary to hypotension can cause more insult to signs of fever, tachycardia, and pain may not develop.
an already injured area.55 Adequate cardiac output he risk for aspiration is increased. A nasogastric tube
should be maintained at an arterial blood pressure can prevent regurgitation by decompression.55
>85 mmHg to prevent additional injury to the spinal
cord. Administering luids and vasopressors will of- Hematologic management
set the pooling of the blood that is caused by vasodila- Patients are at risk for anemia, deep vein thrombosis
tion. However, care must be taken not to luid overload. (40100%), or pulmonary embolism (0.54.6%).52
An inotropic agent such as dopamine or dobutamine Anticoagulants are indicated to prevent thrombus
may be the drug of choice but care should be taken in formation.55
the use of potent alpha agonists such as phenyleph- Indicated tests: Depending on acuteness of the
rine, as substantial increases in cardiac aterload may injury: vital signs to assess level; oxygen saturation
impair cardiac output and precipitate let ventricular to gauge ventilatory ability; complete blood count to 21
Section 1: General considerations

assess infection or occult or actual bleeding; urinalysis 11. Audu P, Artz G, Scheid S, et al. Recurrent laryngeal nerve
to determine infection; basic metabolic panel to evalu- palsy ater anterior cervical spine surgery: the impact of
ate electrolyte status; creatinine clearance to assess endotracheal tube cuf delation, reinlation and pressure
adjustment. Anesthesiology 2006; 15(5): 898901.
kidney function; liver proile to assess hepatic and
nutritional status; CXR to determine any additional 12. Chiono J, Bernard N, Bringuier S, et al. he ultrasound-
injuries, type and crossmatch blood if the patient is a guided transversus abdominis plane block for anterior
iliac bone grat postoperative pain relief: a prospective
trauma victim; coagulation proile if head injury; car-
descriptive study. Reg Anesth Pain Med 2010; 35(6):
diogram if head or neck injury. 5204.

Conclusion 13. Good C. he genetic basis of adolescent idiopathic


scoliosis. J Spinal Res Found 2009; 4(1): 1315.
Patients with spinal cord pathology presenting for sur-
gery may have many and varied problems that encom- 14. Sundarathiti P, Pasutharnchat K, Jommaroeng P.
horacic epidural-general analgesia in scoliosis surgery.
pass all body systems. Preanesthetic assessment may J Clin Anesth 2010; 22(6): 41014.
therefore be extremely complex and must be readily
15. Modi HN, Suh SW, Hong JY, et al. Surgical correction of
prepared to adapt quickly.
paralytic neuromuscular scoliosis with poor pulmonary
functions. J Spinal Disord Tech 2011; 24(5): 32533.
References
1. American Society of Anesthesiologists Task Force 16. Tse V. Spinal metastasis and metastatic disease to the
on Preanesthesia Evaluation. Practice Advisory for spine and related structures. emedicine from web MD:
Preanesthesia Evaluation Anesthesiology 2002; 96: http://emedicine.medscape.com/article/1157987-
48596. overview (accessed November 30, 2010).
2. Roizen M. Preoperative patient evaluation. Can J 17. Peh WC, Munk PL, Rashid F, et al. Percutaneous
Anaesth 1989; 36 Suppl 1: S1319. vertebral augmentation: vertebroplasty, kyphoplasty
and skyphoplasty. Radiol Clin North Am 2008; 46(3):
3. Pueschel SM, Scola FH. Atlantoaxial instability in 61135, vii.
individuals with Down Syndrome. Epidemiologic,
radiographic and clinical studies Pediatrics 1987; 80(4): 18. Schlaich C, Minne HW, Wagner G, et al. Reduced
55560. pulmonary function in patients with spinal
osteoporotic fractures. Osteoporos Int. 1998; (8): 2617.
4. Moore RA, McNicholas KW, Warran SP. Atlantoaxial
subluxation with symptomatic cord compression in a 19. Dong R, Chen L, Gu Y, et al. Improvement in
child with Downs syndrome. Anesth Analg 1987; 66(1): respiratory function ater vertebroplasty and
8990. kyphoplasty. Int Orthop 2009; 339(6): 168994.
5. Liu KX, Massoud B. Endoscopic anterior crevical 20. Jellish WS, halji Z, Srevenson K, et al. A prospective
discectomy under epidurogram guidance. Surg Technol randomized study comparing short and intermediate-
Int 2010 Oct; XX: 3738. term perioperative outcome variables ater spinal or
general anesthesia for lumbar disc and laminectomy
6. Jenkins LT, Jones AL, Harms JJ. Prognostic factors in surgery. Anesth Analg 1996; 83: 55964.
lumbar spinal fusion. Contemp Orthop 10994; 29(3):
17380. 21. ACC/AHA Guidelines on Perioperative Cardiovascular
Evaluation and care for noncardiac surgery. J Am Coll
7. Hilibrand AS. Impact of smoking on the outcome of
Cardiol 2007; 50(17): 159222.
anterior cervical arthrodesis with interbody or strut-
grating. J Bone Joint Surg Am 2001; 83-A: 66873. 22. Katz RI, Barnhart JM, Ho G, Hersch D, Dayan
SS, Keehn L. A survey on the intended purposes
8. Jung A, Schramm J. How to reduce recurrent laryngeal
and perceived utility of preoperative cardiology
nerve palsy in anterior cervical spine surgery: a
consultations. Anesth Analg. 1998; 87: 8306.
prospective observational study. Neurosurgery 2010;
67(1): 1015. 23. Katz RI, Cimino L, Vitkun SA. Preoperative medical
consultations: impact on perioperative management
9. Apfelbaum RI, Kriskovich MD, Haller JR. On the
and surgical outcome. Can J Anaesth. 2005; 52: 697702.
incidence, cause, and prevention of recurrent laryngeal
nervepalsies during anterior cervical spine surgery. 24. Fletcher GF, Balady G, Froelicher VF, et al. Exercise
Spine (Phila Pa 1976) 2000; 25(22): 290612. standards: statement for healthcare professionals from the
American Heart Association. Circulation 1992; 86: 3404.
10. Liu J, Zhang X, Ging W, et al. Correlations between
controlled endotracheal tube cuf pressure and 25. Goldman L, Caldera DL, Nussbaum SR, et al.
22 postprocedural complications: a multicenter study. Multifactorial index of cardiac risk in noncardiac
Anesth Analg 2010; 111(5): 11337. surgical procedures. N Engl J Med 1977; 297: 84550.
Chapter 1: Preoperative assessment adult patient

26. Lee TH, Marcantonio ER, Mangione CM, et al. 39. Warner DO, Klesges RC, Dale LC, et al. Telephone
Derivation and prospective validation of a simple index quitlines to help surgical patients quit smoking and
for prediction of cardiac risk of major noncardiac provider attitudes. Am J Prev Med 2008; 35: S48693.
surgery. Circulation 1999; 100: 10439.
40. Lee JC, Kim MS, Shin BJ. An analysis of the prognostic
27. Liu LL, Dzankic S, Leung JM. Preoperative factors afecting the clinical outcomes of conventional
electrocardiogram abnormalities do not predict lumbar open discectomy: clinical and radiological
postoperative cardiac complications in geriatric prognostic factors. Asian Spine J 2010; 4(1): 2331.
surgical patients. J Am Geriatr Soc 2002; 50: 118691.
41. Harrer SW, Carlson WO. Spinal health and smoking. S
28. Poldermans D, Arnese M, Fioretti PM, et al. Improved D Med 2009; 62(8): 309, 31113.
cardiac risk stratiication in major vascular surgery with
42. Gross JB, Bachenberg KL, Benumof JL, et al.;
dobutamine-atropine stress echocardiography. J Am
American Society of Anesthesiologists Task Force on
Coll Cardiol 1995; 26: 64853.
Perioperative Management. Practice guidelines for the
29. Das MK, Pellikka PA, Mahoney DW, et al. Assessment perioperative management of patients with obstructive
of cardiac risk before nonvascular surgery: dobutamine sleep apnea: A report by the by the American Society
stress echocardiography in 530 patients. J Am Coll of Anesthesiologists Task Force on Perioperative
Cardiol 2000; 35: 164753. Management of patients with obstructive sleep apnea.
30. Bigatel DA, Franklin DP, Elmore JR, Nassef LA, Youkey Anesthesiology 2006; 104: 108193. Available online
JR. Dobutamine stress echocardiography prior to aortic at: http://www.asahq.org/publicationsAndServices/
surgery: long-term cardiac outcome. Ann Vasc Surg sleepapnea103105.pdf.
1999; 13: 1722. 43. Kaw R, Michota A, Jafer A, et al. Unrecognized sleep
31. Beattie WS, Abdelnaem E, Wijeysundera DN, Buckley apnea in the surgical patient: implications for the
DN. A meta-analytic comparison of preoperative stress perioperative setting. Chest 2006; 129(1): 198205.
echocardiography and nuclear scintigraphy imaging. 44. Chung F, Yeqneswaran B, Liao P, et al. STOP
Anesth Analg 2006; 102: 816. questionnaire: a tool to screen patients for obstructive
32. Mondillo S, Ballo P, Agricola E, et al. Noninvasive tests sleep apnea. Anesthesiology 2008; 108(5): 81221.
for risk stratiication in major vascular surgery. VASA 45. Charlson ME, MacKenzie CR, Gold JP, Ales KL,
2002; 31: 195201. Topkins M, Shires GT. Risk for postoperative congestive
33. Klocke FJ, Baird MG, Lorell BH, et al. ACC/AHA/ heart failure. Surg Gynecol Obstet 1991; 172: 95104.
ASNC guidelines for the clinical use of cardiac 46. Bilotta F, Rosa G. Glucose management in the
radionuclide imagingexecutive summary: a report of neurosurgical patient: are we yet any closer? Curr Opin
the American College of Cardiology/American Heart Anaesthesiol 2010; 23(5): 53943.
Association Task Force on Practice Guidelines (ACC/
AHA/ASNC Committee to Revise the 1995 Guidelines 47. Rizvi AA, Chillag SA, Chillag KJ. Perioperative
for the Clinical Use of Cardiac Radionuclide Imaging). J management of diabetes and hyperglycemia in patients
Am Coll Cardiol 2003; 42: 131833. undergoing orthopaedic surgery. J Am Acad Orthop
Surg 2010; 18(7): 42635.
34. ACC/AHA Focused Update on Perioperative
Beta Blockade incorporated into the ACC/AHA 48. Brosius FC III, Hostetter TH, Kelepouris E, et al.
2007 Guidelines in Perioperative Cardiovascular Detection of chronic kidney disease in patients with
Evaluation: www.americanheart.org/presenter. or at increased risk of cardiovascular disease: a science
jhtml?identiier=3004542 (accessed December 1, advisory from the American Heart Association Kidney
2010). and Cardiovascular Disease Council; the Councils on
High Blood Pressure Research, Cardiovascular Disease
35. Obert DM, Hua P, Pilkerton ME, et al. Environmental in the Young, and Epidemiology and Prevention;
tobacco smoke furthers progression of diabetic and the Quality of Care and Outcomes Research
nephropathy. Am J Med Sci 2011; 341(2): 12630. Interdisciplinary Working Group: developed in
36. Malhotra R, Kapoor A, Grover V, et al. Nicotine and collaboration with the National Kidney Foundation.
periodontal tissues. J Indian Soc Periodontol 2010; Circulation 2006; 114: 10837.
14(1): 729. 49. Kertai MD, Boersma E, Bax JJ, et al. Comparison
37. Bor-Caymaz C, Bor S, Tobey NA, et al. Efects of between serum creatinine and creatinine clearance for
ethanol and extract of cigarette smoke on rabbit buccal the prediction of postoperative mortality in patients
mucosa. J Oral Pathol Med 2011; 40(1): 2732. undergoing major vascular surgery. Clin Nephrol 2003;
38. Shi Y, Warner DO. Surgery as a teachable moment 59: 1723.
for smoking cessation. Anesthesiology 2010; 112(1): 50. Lee LA, Roth S, Posner KL, et al. he American Society 23
1027. of Anesthesiologists Postoperative Visual Loss Registry:
Section 1: General considerations

analysis of 93 spine surgery cases with postoperative 59. Hurlbert RJ, Hamilton MG. Methylprednisolone for
visual loss. Anesthesiology 2006; 105(4): 6529. acute spinal cord injury: 5-year practice reversal. Can J
51. Wu WC, Schitner TL, Henderson WG, et al. Neurol Sci 2008; 35(1): 415.
Preoperative hematocrit levels and postoperative 60. Rozet, I. Methylprednisolone in acute spinal cord
outcomes in older patients undergoing noncardiac injury: is there any other ethical choice? J Neurosurg
surgery. JAMA 2007; 297: 24818. Anesthesiol 2008; 20(2): 1379.
52. Stier, G.R., Schell, R.M., Cole, D.J. Neurosurgical 61. Miller, S M. Methylprednisolone in acute spinal cord
diseases and trauma of the spine and spinal cord: injury: a tarnished standard. J Neurosurg Anesthesiol
anesthetic considerations. In: Cottrell JE, Young WL, 2008; 20(2): 1402.
eds. Cottrell and Youngs Neuroanesthesia. 5th ed.
62. Zeng Y, Xiong M, Yu H, et al. Clinical efect of
Mosby : St. Louis; 2010: 35579.
methylprednisolone sodium succinate and mouse
53. Crosby, Edward T. Airway management in adults nerve growth factor for injection in treating acute
ater cervical spine trauma. Anesthesiology 2006; 104: spinal cord injury and cauda equina injury.
1293318. Zhongguo Xiu Chong Jian Wai Ke Za Zhi 2010;
54. Pimental L, Diegelmann L. Evaluation and 24(10): 120811.
management of acute cervical spine trauma. Emerg Med 63. Kwon BK, Sekhon LH, Fehlings MG. Emerging repair,
Clin North Am 2010; 28(4): 71938. regeneration, and translational research advances for
55. Hadley MN, Walters BC. Guidelines for the spinal cord injury. Spine (Phila Pa 1976). 2010; 35(21
management of acute cervical spine and spinal cord Suppl): S26370.
injuries. Clin Neurosurg 2003; 50: 23948. 64. Jellish WS, Zhang X, Langen KE, et al. Intrathecal
56. Fehlings MG, Wilson JR. Timing of surgical magnesium sulfate administration at the time of
intervention in spinal trauma: what does the evidence experimental ischemia improves neurological
indicate? Spine (Phila Pa 1976) 2010; 35(21): 15960. functioning by reducing acute and delayed loss of
57. Belegu V, Oudega M, Gary D, McDonald JW. Restoring motor neurons in the spinal cord. Anesthesiology 2008;
function ater spinal cord injury: promoting spontaneous 108(1): 7886.
regeneration with stem cells and activity based therapies. 65. Erol FS, Kaplan M, Titikci M, et al. Comparison of the
Neurosurg Clin North Am 2007; 18(1): 14368. efects of octreotide and melatonin in preventing nerve
58. Sugimoto IY, Tomioka M, Kai N, et al. Does high dose injury in rats with experimental spinal cord injury. J
methylprednisolone sodium succinate really improve Clin Neurosci 2008; 15(7): 78490.
neurological status in patients with acute cervical 66. Yoo, K Y, Jeong CW, Kim SJ, et al. Altered
cord injury?: a prospective study about neurological cardiovascular response to tracheal intubation in
recovery and early complications. Spine (Phila Pa 1976) patients with complete spinal cord injuries. Br J Anaesth
2009; 34(20); 1214. 2010; 105(6): 7539.

24
Section 1 General considerations
Chapter
Fluid management

2 Maria Bauer, Andrea Kurz, and Ehab Farag

abdominal excursion in the prone position, which


Key points
enhances the venous return). Using transesophageal
Cardiac index decreases in prone position due echocardiography (TEE), Toyota and Amaki2 dem-
to reduced venous return and let ventricle onstrated a decrease in let ventricular volume and
compliance. compliance in prone position for lumbar laminectomy.
Crystalloid should be used for maintenance and hese changes can be explained by inferior vena caval
colloid for replacement of blood loss. compression and decreased let ventricular compli-
Albumin seems to be the suitable colloid during ance due to increased intrathoracic pressure in the
spine surgery. prone position. hese results have been conirmed
Glycocalyx is better maintained by avoiding using a thermodilution pulmonary artery catheter.
hyper- and hypovolemia. Cardiac output decreased from 17% to 24% using this
Goal-directed luid therapy is the ideal way to technique.3 It should be mentioned that vena cava pres-
guide luid management during spine surgery sures vary between 0 40 mmH2O in prone position
especially in prone position. with the abdomen hanging free and >300 mm H2O
with abdominal compression in prone position.4
Consequently the increase in venous pressure not only
Introduction will increase bleeding during spine surgery but also can
Fluid management during spine surgery is very import- impair spinal cord perfusion.
ant and diicult at the same time. Most spine surger- On the other hand, the prone position has a more
ies are performed in prone position. Prone position favorable efect on the respiratory system than the
induces a decrease in cardiac index and cardiac output. supine position. he prone position enhances ventila-
Maintaining stable hemodynamics with proper tissue tionperfusion matching by recruiting dorsal airways,
perfusion requires adequate luid management with- resulting in an increase in lung units and consequently
out luid overloading. he best way to ensure normo- an increased functional residual capacity (FRC) with
volemia in prone position is by utilizing goal-directed near normal ventilationperfusion matching and
luid therapy for luid management during spine sur- reduction in shunt volume. Furthermore, prone posi-
gery in prone position. his chapter will cover the tion has a beneicial efect of positive end-expiratory
pathophysiological changes during prone position, the pressure (PEEP) without the risks of barotrauma or
physiology and the importance of the endothelial gly- interference with cardiac functions. It should be noted
cocalyx, the diferent types of luid, and the most recent that respiratory beneits in prone position depend
advances in goal-directed luid therapy. on maintaining a freely moving abdomen during
surgery.5,6
Pathophysiology of prone position
Prone position decreases cardiac index and venous Endothelial glycocalyx
return. Wadsworth et al.1 measured cardiac index (CI) Since the irst description of blood circulation by
in unanesthetized volunteers. CI was reduced mostly William Harvey (15781657), it has been recognized
in the knee-chest position by 20% and decreased by that an intact barrier is an essential prerequisite for a
3% on the Jackson table (the Jackson table allows free healthy circulatory system and proper luid distribution

25
Anesthesia for Spine Surgery, ed. Ehab Farag. Published by Cambridge University Press. Cambridge University Press 2012.
Section 1: General considerations

between the intravascular and extravascular compart- high extravascular COP should lead to luid shits from
ments. In 1940, Danielli described the existence of a the vessel into the interstitial space, resulting in tissue
thin layer of proteinaceous material on the endothelial edema. Also, the iltration rate across the vascular bar-
surface, which could be associated with the regulation rier is independent of COP in the interstitial space. he
of vascular iltration. Today this layer is called the endo- presence of an intact EG is an integral factor in order
thelial glycocalyx (EG). he structure consists of mem- to maintain an intact vascular barrier, proper iltration
brane-bound proteoglycans and glycoproteins building rate, and avoidance of tissue edema despite high COP
up a network in which plasma or endothelial proteins in the extravascular space. EG retains plasma proteins
are retained.7 he main constituents of the glycocalyx and generates the endothelial surface layer with its own
are syndecan, heparan sulfate, and hyaluronan.8 EG plus high oncotic pressure. In a small gap below the EG, the
bound luids and plasma proteins form the endothelial concentration of proteins is lower than in the interstitial
surface layer (ESL) with a thickness of about 1 m. he space, allowing small net luid iltration into the intersti-
noncirculating part of the plasma ixed within the ESL tial space9 (Fig 2.1). he EG structure makes the arteri-
is approximately 7001000 ml in humans. olar and capillary domains relatively impermeable.
According to the Starling principle published in However, venules represent the suitable site for il-
1896, high vascular colloid osmotic pressure (COP) tration through their gaps and pores. Because colloids
in contrast to low extravascular COP is essential for are able to escape through venular pores, there are low
vascular barrier function. However, it was recently oncotic pressure diferences in addition to low hydro-
indicated that extravascular COP is almost equal to static diferences. he result is a low net iltration in the
intravascular COP.8 According to the Starling principle, venular sections.10 he latter property is in accordance

Capillary segments (continuous) Large-pore venular sections

Hardly any egress of colloidal particles, Easy egress of colloidal particles,


small flow of ultrafiltrate back-diffusion is possible
(IIc and IIt irrelevant, IIc and IIg count)

Pt
Pt IIt
IIt

IIg
IIc

Pc IIv
IIc
Pv
ESL IIc
IIg
EC
IS

[PcPt] large, [IIcIIg] large [PvPt] small, [IIvIIt] small


High resistance to flow of water (sound gap)

Very low filtration rate Low filtration rate


Figure 2.1 Low-ltration concept of lymph production. Pt, Pc, and Pv, hydrostatic pressure in tissue, capillary, and venule, respectively;
t, c, v, e, and g, colloid osmotic pressure in tissue, capillary, venule, endothelial surface layer, and beyond the endothelial glycocalyx,
respectively; ESL = endothelial surface layer (glycocalyx + bound colloid), EC, endothelial cell, IS, interstitial space. (Reprinted with permission
26 from Jacob M, et al., The endothelial glycocalyx aords compatibility of Starlings principle and high cardiac interstitial albumin levels.
Cardiovascular Research 2007;73:575586.)
Chapter 2: Fluid management

with the newly appreciated fact that there is no net interstitial space. hus the ability of ANP to increase
reabsorption of luid in the venular segments of the capillary permeability to water, solutes, and macro-
microcirculation.10 molecules might be at least partially explained by its
In summary, a small luid and protein shit out of capacity to disturb the EG structure.15 Recent studies
the blood vessels occurs at all times, but it is disposed have shown that the classical third space does not
of in a timely manner from the interstitial space via exist.16 he average insensible perspiration is only
the lymphatic system under normal physiological about 0.5 ml/kg/h via skin and airways in the awake
conditions.11 adult. During abdominal surgery, insensible luid
losses increase to only 1 ml/kg/h.17 Avoiding hypo-
The important functions of glycocalyx volemia and hypervolemia, which includes a careful
Endothelial glycocalyx plays a very important role indication for perioperative luid management, is an
in maintaining the proper functions of immune and important element to maintain a healthy EG and there-
coagulation systems. Altered functions of EG lead to fore to limit perioperative luid and protein shits into
an increase in the coagulation and the inlammatory the interstitial space.
response during the perioperative period. Release of he type of luid used for perioperative luid man-
tumor necrosis factor (TNF-) or oxidized lipopro- agement is also very crucial to maintain a competent
teins mediates disruption of ESL.12 Furthermore, it has vascular barrier and reduce the degree of tissue edema.
been reported that perioperative ischemia/reperfusion Isotonic crystalloids are usually used to replace insen-
injury induces shedding of the EG depending on the sible perspiration and urinary output. Colloids, by
duration of ischemia.13 Normally, the small vascular contrast are indicated to replace plasma deicits due
adhesion molecules are within the EG. Degradation of to acute blood loss or protein-rich luid shits toward
the EG exposes adhesion molecules for immunocom- the interstitial space.17 Albumin seems to be the ideal
petent cells, which enhances leukocyte and platelet colloid to maintain the integrity of the vascular bar-
adhesion. Ater shedding of the EG, circulating glyco- rier. his intrinsic efect of albumin is most likely based
calyx components like heparan sulfates have a direct on its electrostatic binding properties. he charges
chemotactic efect on leukocytes and increase their exposed by the molecules forming the EG are mainly
presence at the site of inlammation. Consequently, negative (heparan, dermatan, and chondroitin sul-
the destruction of EG can trigger the inlammatory fates, etc.), whereas albumin carries not only negative
cascade. hus, maintaining the integrity of EG might charges (carboxylate groups) but also positive charges
represent a promising therapy for inlammation and (arginine, lysine) at physiologic pH. he presence of
ischemic/reperfusion injury.8 EG has an important positive charges in albumin will enable it to attach to
mechanosensory role by translating intravascular shear the EG and provide intact ESL. Jacob et al.18 have shown
stress into biochemical activation of endothelial cells in an isolated perfused heart model that providing
to release nitric oxide (NO). EG is a crucial component albumin to the endothelium, before and ater ischemia,
for binding and regulating enzymes involved in the maintains vascular integrity during reperfusion and
coagulation cascade. In addition, the most important alleviates development of tissue edema. It should be
inhibitor of thrombin, factor Xa (antithrombin III), is mentioned that luid accumulation may be adequately
irmly attached to the endothelium. It is therefore not tolerated by young, vigorous patients. However, in
surprising that hyperglycemia-induced loss of EG is elderly or frail patients it may be seriously jeopard-
accompanied by activation of coagulation and vascular ized because of impairment of oxygen delivery to the
dysfunction in diabetics14 (Fig. 2.2). lungs, myocardium, and brain. It is quite interesting
to note in the same study the authors conirmed that
Perioperative fluid management and hydroxyethyl starch (HES 130/0.4) infusion was proved
superior to isotonic saline only in the very late stage of
glycocalyx reperfusion.
Perioperative luid management is one of the key fac-
tors in maintaining the integrity of EG. It is well known
that iatrogenic acute hypervolemia can lead to release
Intravenous fluids commonly used in spine
of atrial natriuretic peptide (ANP). ANP induces shed- surgery
ding of EG components, mainly syndecan-1, thereby Fluid replacement remains a mainstay of perioperative 27
increasing shits of luid and macromolecules into the care, especially when altered hemodynamics or large
Section 1: General considerations

A Glycocalyx under physiological condition Figure 2.2 (a) Physiological role of


the glycocalyx. Endothelial glycocalyx
Erythrocyte Plasma proteoglycan regulates nitric oxide synthase activity,
harbors superoxide dismutase, and serves
as a physical barrier for macromolecules,
including plasma proteins and
lipoproteins. In addition, the glycocalyx
attenuates platelet as well as leukocyte
Platelet adhesion. (b) Consequence of glycocalyx
EC-SOD
perturbation. Glycocalyx perturbation
Plasmaproteins Monocyte results in a pro-atherogenic state,
Glycocalyx
characterized by endothelial dysfunction,
increased vascular permeability, as
Endothelium well as the activation of coagulation
VCAM / ICAM Subendothelial
and cellular adhesion/migration. AT,
NO-synthase Antithrombin space antithrombin; EC-SOD, extracellular
vWF TFPI
superoxide dismutase; ICAM, intercellular
cell adhesion molecule; NO, nitric oxide;
Endothelial function Permeability Coagulation Inflammation
TFPI, tissue factor pathway inhibitor;
VCAM, vascular cell adhesion molecule;
Shear induces Prevention of
NO-sythesis, Inhibition of platelet adherence, leukocyte vWF, von Willebrand factor. (Reprinted
superoxide dismutation Sieving barrier coagulation regulatory factors adhesion with permission from Nieuwdorp M, et al.,
The endothelial glycocalyx: a potential
B Perturbed glycocalyx barrier between health and vascular
disease. Current Opinion in Lipidology
Erythrocyte Plasma 2005;16:507511.)
proteoglycans

TFPI
vWF
Antithrombin
EC-SOD
Monocyte
Plasmaproteins

Perturbed
Adhering glycocalyx
platelets Endothelium
VCAM / ICAM
Subendothelial
No-synthase
vWF space

NO availability Leakage of Platelet adherence Leukocyte adhesion



oxidative stress macromolecules thrombin generation and diapedesis

luid shits are anticipated. Adequate hemodynamic and luid requirements, necessitates careful selection
management in patients undergoing major surgery is of intraoperative luids, and an approach for volume
paramount; however, much controversy exists about substitution that is adapted to the patients needs.
the composition and amount of luids replaced periop- Of the possible causes, trauma and surgery bring
eratively. he debate is of long standing; available data about the most severe blood loss in the greatest fre-
do not provide conclusive evidence to establish univer- quency in the medical setting. Spinal procedures
sally accepted guidelines from any consensus group carry a high risk of signiicant surgical bleeding. he
or professional society. Focus has been shiting from degree of blood loss in spine surgery is highly variable
luid types, cost, and availability, and standard/liberal/ and depends on the presence of predisposing factors.
restrictive regimens to individualized (goal-directed) Moller and colleagues reported an average intraop-
luid management strategies; the latter, especially erative blood loss of 861 ml (range 1003100 ml) and
with colloid administration, appearing to potentially 1517 ml (range 3607000 ml) in patients undergoing
decrease perioperative morbidity and mortality by noninstrumented and instrumented spinal fusion,
improving microcirculatory low and tissue oxygena- respectively.20 Advanced age, preexisting coagulation
tion.19 Complex spine surgery in the prone position, abnormalities and other comorbidities, prolonged
28 especially if characterized by substantial blood loss surgery, increased intra-abdominal pressure due
Chapter 2: Fluid management

to the valveless communication between the inferior cross the vascular endothelium. Iso-oncotic colloids are
vena cava and the vertebral veins,21 posterior spinal commonly used plasma substituents usually in com-
procedures, reconstruction of structural abnormali- bination with crystalloids in cases when the risk of
ties of neuromuscular etiology, fusions involving mul- acute blood loss is high, or large amounts of protein-rich
tiple levels, tumor resections, and revision surgeries luid shits are present or anticipated. heir attributed
have been associated with greater surgical blood loss. potential to produce or restore oncotic pressure desig-
Bleeding can be considerable even in routine cases. nates their main area of indication: rapid restitution and
Neither the potentially devastating sequelae of organ maintenance of intravascular volume, hemodynamic
hypoperfusion in prone spine surgery, nor the adverse stabilization, and improvement of microcirculation.25
outcomes associated with perioperative blood trans- Compared with crystalloids, their use is costly and has
fusion should be underestimated. Restoration and been associated with several adverse efects. Serious
maintenance of optimal tissue perfusion is therefore adverse efects of plasma-derived albumin are rare,
imperative, and, ideally, is tailored to the patients indi- but despite its preparation method and the lack of case
vidual luid requirements during surgery. reports of transmission of viral infections following
Intraoperative luid management, for concep- albumin administration, the potential risk of transmit-
tual convenience, can be divided into two separate ting infections cannot be fully eliminated.2628 An asso-
components: administration of maintenance luids ciation has been found between synthetic colloids (such
to meet basic, predictable volume requirements i.e., as starches, dextrans, or gelatins) and hypersensitivity
luid losses from the extracellular space due to insens- reactions,29 coagulation disorders,30 and kidney dys-
ible perspiration and urine production, and admin- function.31 However, these adverse efects may poten-
istration of resuscitation luids to respond to volume tially be provoked by the infusion of any type of colloidal
deicits that exceed maintenance administration i.e., solution. Although its advantages over crystalloids
mainly blood loss occurring during trauma or surgery, remains questioned by prospective randomized trials
accounting for the intravascular deicit. Resuscitation and meta-analyses,3234 colloid administration has been
luids, based on their constituents, fall into the categor- found to carry a lesser risk of hypoalbuminemia and
ies of crystalloids and colloids. pulmonary edema.23 Colloids remain in the intravascu-
he choice of resuscitation luid type largely lar space for longer, are more efective plasma expand-
depends on the type of volume lost, concurrent elec- ers, and in the presence of an intact endothelium, are
trolyte imbalances, and the distribution of the given known to restore intravascular volume with a prompt
luid within the body; and it may have an impact on and prolonged hemodynamic response and due to
the postoperative outcome.22 Crystalloids are safe, non- their, albeit minimal, ability to pass through the inter-
pyrogenic, nonallergenic solutions that contain difer- cellular pores of the endothelial cells a decreased risk
ing concentrations of inorganic and/or water-soluble of tissue deposition. In the presence of a suspected or
organic (dextrose) particles dissolved in water, with or evident disruption of the bloodbrain barrier, although
without bicarbonate or its precursor. hose contain- highly controversial, colloids should be avoided or used
ing nonorganic particles are used intraoperatively to with caution.35,36 Although numerous types are avail-
provide maintenance hydration and electrolytes. he able, the most commonly used colloids are human albu-
kinetics of crystalloids is determined by sodium difu- min and hydroxyethyl starch solutions.
sion; these solutions therefore primarily distribute in
the extracellular space. heir use for aggressive luid Crystalloids
resuscitation elicits a delayed hemodynamic response,
and may carry an increased risk of luid overload, dilu- Normal saline
tional hypoalbuminemia, and subsequent pulmonary Normal (physiological, 0.9%) saline is an unbalanced
and tissue edema.23,24 Crystalloids contain a suicient isotonic solution of the major extracellular cation and
amount of free water to cause reduction in plasma anion, with no efect on plasma osmolarity. Saline is
osmolarity, and may therefore cause or worsen edema- indicated to replenish anticipated or ongoing sodium-
tous states and compromise oxygen delivery and organ containing luid losses. Normal saline distributes in
function. Protein and nonprotein colloids are solutions the extracellular space: about 25% of the adminis-
of large-molecule substances homogeneously dispersed tered volume tends to remain intravascularly, about
in an isotonic or hypertonic vehicle that do not freely 75% distributes extravascularly. Its volume-expanding 29
Section 1: General considerations

potential is low; for every unit of blood volume lost, To account for further ions, the above equation has
a 4-fold replacement of normal saline is required. As been modiied:43
an easily available, inexpensive, isotonic solution, it is
SID (apparent) = (Na+ + K+ + Mg2+ + Ca2+)
a preferred luid in the perioperative care of the neu-
(Cl lactate) (2.1)
rosurgical patient; however, its composition is non-
physiological in the following ways: as an unbalanced Strong ion diference is therefore a function of both
solution, it contains a concentration of chlorine higher the charge and the concentration of electrolytes, and
than physiological. Evidence supports that massive its driving force is considered to be an independent
infusion of normal saline or normal saline-based luids mechanism imposed on the acidbase regulation. he
alone may predictably induce hyperchloremic, non- electrochemical driving forces generated by the strong
anion-gap metabolic acidosis, with observed reduc- ion diference will ultimately cause alterations of the
tions in plasma pH of as much as 0.3 units.37,38 Normal H+ concentration. SID, as well as weak acids and pCO2,
saline is devoid of bicarbonate or a bicarbonate precur- will determine the inal H+ ion concentration: any
sor to bufer acidbase abnormalities, and lacks other decrease in SID will result in acidosis, any increase in
electrolytes and organic particles present in the plasma, SID will result in alkalosis (resulting from an increase
potentially worsening any preexisting abnormalities. or a decrease, respectively, in [H+] to maintain elec-
he phenomenon of developing metabolic acid- trochemical neutrality). he acidbase disorder ater
osis upon the infusion of large amounts of crystalloids administration of unbalanced solutions can thus be
was irst discussed in a randomized trial in 1994 by explained by acidosis resulting from a decrease in the
McFarlane and Lee,39 whose indings have been con- strong ion diference.42
irmed in subsequent studies.37,40 he phenomenon has Since the severe pathophysiological implications of
traditionally been explained by the concept of dilu- this transient hyperchloremic metabolic acidosis are
tional acidosis: dilutional acidosis occurs when exces- not convincingly supported by currently available evi-
sive amounts of resuscitation solutions devoid of bufer dence, translation of its development into clinical sig-
reduce the concentration of plasma bicarbonate.40,41 niicance should be done with appropriate caution. It
he traditional explanation for the mechanism of should be remembered that the presence of hyperchlo-
acidbase derangements was challenged in 1978 by remic metabolic acidosis may potentially aggravate
Peter A. Stewart, who introduced the term strong any preexisting acidosis of diferent etiology. Acidosis,
ion diference.42 Stewarts nontraditional approach is regardless of the underlying mechanism, may impair
based on three physicochemical principles: irst, the end organ perfusion and performance, blunts ino-
law of electroneutrality (i.e., in a solution, all positively tropic responsiveness to catecholamines, causes coagu-
and all negatively charged ions must be equal), second, lopathy, and, with concomitant hypothermia, increases
the principle of mass conservation (i.e., the mass of morbidity and mortality.38,44,45 Hyperchloremia, as
a closed system will remain constant over time), and suggested by animal studies, has the potential to select-
third, the equilibrium constraints on dissociation ively induce renal vasoconstriction by modulation of
reactions. According to his approach, any changes in renin release and increased sensitivity of the aferent
the H+ and HCO3 concentration resulting in acid arteriole to angiotensin II; and it may further decrease
base derangements are secondary to alterations of (1) renal blood low and glomerular iltration rate, and
pCO2, (2) nonvolatile weak acid concentration (acids prolong time to irst urination, compared with bal-
partially ionized at physiological pH, such as albumin anced luids.4648 Recognition of the phenomenon
and inorganic phosphate), and (3) strong ion concen- and the avoidance of further increase in the Cl load
tration (ions that remain dissociated, thus nearly com- are therefore expected. Additional efects of exces-
pletely ionized within the ranges of physiological pH, sive saline administration include increased bleeding,
such as Na+, K+, Ca2+, Mg2+, Cl, lactate, sulfate, and coagulation derangements, and transfusion require-
ketone bodies). In the plasma, however, adding up all ment, when compared to balanced luids.37,49 One pos-
the strong ions does not result in zero. his accounts sible explanation may lie with the lack of calcium an
for the concept of strong ion diference (SID). Stewart important cofactor of the coagulation cascade in the
originally described this equation as physiological saline. Another possible mechanism may
be the reduced levels of von Willebrand factor antigen
30 SID = (Na+ + K+) (Cl lactate) = 4044 mEq/l and impaired platelet function.50
Chapter 2: Fluid management

Lactated Ringers isolated from pooled human plasma, and has been
Lactated Ringer is an inexpensive, widely available considered to be the gold standard solution for luid
physiological solution that equilibrates freely across resuscitation in the critically ill patient population.57
the intravascular and extravascular luid compart- Of note, the indings of the landmark Saline versus
ments, and restores extracellular luid deicit asso- Albumin Fluid Evaluation (SAFE) study, evaluating
ciated with blood loss. Its constituents and their the efect of albumin versus normal saline administra-
concentrations match those of the plasma, account- tion for luid resuscitation on overall 28-day mortality
ing for less luid shits. Although lactated Ringers has in critically ill patients, demonstrated clinical equiva-
a lesser volume-expanding efect than normal saline, lency between albumin and saline.36 Preparations are
a signiicant advantage of lactated Ringers over saline available in the form of hypo-oncotic (4%), iso-oncotic
is that due to its lactate (bicarbonate precursor) con- (5%) and hyperoncotic (20%, 25%) solutions. As a nat-
tent it does not bring about hyperchloremic acidosis as ural plasma derivative, albumin can be administered in
does normal saline,51 even in large volumes: bicarbo- large amounts;27 however, its potential to induce ana-
nate is generated by both oxidation (70%) and gluco- phylaxis exceeds that of starches.58,59 Albumin, possibly
neogenesis (30%), both biochemical processes taking due to its ability to prolong the antiplatelet activity of
place predominantly in the liver. Both mechanisms nitric oxide (NO), has mild antithrombotic and antico-
result in OH production, which, when combined with agulant efects.60. his natural colloid, contrasting with
CO2, generates bicarbonate over a period of 12 h.52 A artiicial preparations, exhibits anti-inlammatory
healthy volunteer study noted a decreased time to irst properties by (1) suppressing the neutrophil oxida-
urine output in the group infused with lactated Ringers tive burst and spreading,61,62 and (2) the reduction of
with little electrolyte imbalance, and signiicantly less inlammatory cytokine release.63 Albumin adminis-
abdominal discomfort, as compared with the normal tration favorably afects the endothelial barrier func-
saline group.53 For luid maintenance, however, with tion, possibly in a concentration-dependent manner
regard to its potential to improve tissue perfusion, lac- and by modulating molecular charge. Also, it reduces
tated Ringers has not been proven to have advantages subendothelial and interstitial permeability by binding
over normal saline: when compared with 0.9% NaCl, to these layers.64,65 Studies evaluating the renal efects
no diference in urine output, serum creatinine change, of albumin versus hydroxyethyl starch have found no
blood loss and transfusion requirements, or coagula- diference in urine output and serum levels of reten-
tion markers was observed.40,54,55 Administration of tion markers;66,67 however, decreased urine output was
larger volumes of lactated Ringers required for ade- observed in patients receiving albumin, as opposed to
quate volume resuscitation carries the risk of luid those who received lactated Ringers, in the early post-
overload, especially in patients with poor cardiac operative course of abdominal aortic surgery.34 he
reserve, as well as decreased colloid osmotic pressure, glomerular iltration rate has been found to decrease
iatrogenic metabolic alkalosis, and dilutional coagula- with both hyperoncotic albumin and 10% hydroxyethyl
tion disorders. Furthermore, consideration should be starch administration; acute tubular necrosis may also
given to the use of alternative crystalloids in certain occur due to the accumulation of small molecules in
patient populations. Patients with severely impaired the renal tubuli.68 Albumin has the capacity to reduce
liver function are at risk of developing metabolic alka- necrotic tissue volume in the ischemic brain, and to
losis resulting from the impaired hepatic metabolism improve cortical perfusion, as demonstrated in animal
of the lactate. Metformin, however, was found to have models.69,70 Furthermore, in a recent meta-analysis, the
no efect on the rate of lactate turnover and oxidation, use of albumin-containing solutions in septic patients
or gluconeogenesis from lactate.56 was associated with lower mortality than were other
luid resuscitation regimens.71
Colloids
Hydroxyethyl starch
Albumin Hydroxyethyl starch (HES) is a hydrolyzed and
Albumin, a natural plasma protein with a molecular hydroxyethylated derivative of the natural corn starch
weight of 69 000 kDa, accounts for the greatest propor- amylopectin, dissolved in normal saline or other bal-
tion of plasma colloid osmotic pressure. For prepara- anced solvents, and has been developed to serve as an
tions commonly used in clinical practice, albumin is alternative colloid to albumin. HES solutions have been 31
Section 1: General considerations

shown to have favorable efects on microcirculation.19 considered the preferred colloid for luid management
hey decrease plasma viscosity,72 reduce inlammatory during spine surgery.
response by decreasing TNF, IL-1, ICAM-1, and
myeloperoxidase activity along with nuclear factor-B Goal-directed uid therapy
activation, decrease pulmonary capillary leakage,73,74 Maintenance of perioperative normovolemia, adequate
and improve postoperative outcome.75 Although tissue perfusion, and tissue oxygenation is paramount
HES solutions are not indicated to treat or reverse in the care of the surgical patient. However, the amount
hypoalbuminemic states, their potential to restore col- and type of resuscitation luids remains controversial.
loid osmotic pressure in the intravascular space is com- he luid of choice largely depends on the type and
parable to that of albumin.76,77 Hydroxyethyl starches amount of luid lost, but evidence suggests that focus-
are characterized by their molecular weight, degree of ing on individualized volume replacement strategies
molar substitution (referring to the average number may have a greater impact on postoperative outcome
of hydroxyethyl residues per glucose subunit) and the improvement than luid types.
C2/C6 ratio (referring to the site of hydroxyethylation Perioperative volume replacement has tradition-
on the glucose constituent). he traditional classii- ally been guided by static circulatory parameters. Static
cation of these solutions is based on the above-men- hemodynamic parameters, such as blood pressure, heart
tioned physicochemical properties of HES. he range rate, or central venous pressure, however, have not been
comprises the high- ( 400 kDa), medium- (200400 found to be reliable predictors of mild hypovolemia.87
kDa), and low-molecular weight (<200 kDa) solutions; Suboptimal intraoperative volume status (volume
highly substituted (hetastarch, 0.60.75), intermedi- overload, as well as hypovolemia) has been associated
ately substituted (pentastarch, 0.5), or low-substituted with adverse postoperative outcomes. Optimization
(tetrastarch, 0.4) solutions, and solutions with a high of intravascular volume by optimizing low-related
(>8) or low (8) C 2/C6 ratio.25,78 Generally, the higher the dynamic variables, such as cardiac preload, has been
molecular weight, the molar substitution, and the C2/C6 hypothesized to improve microcirculation and tissue
ratio, the slower is the degradation of the HES solution. oxygenation, and therefore to improve clinical out-
Slowly degradable, high molecular weight, old gener- comes. Minimally invasive esophageal Doppler-guided
ation starches (for example, HES 200/0.5) entail the risk luid bolusing has the potential to assess luid respon-
of severe kidney dysfunction,31 as well as delayed, dose- siveness in the anesthetized patient, as well as to opti-
dependent, HES-induced pruritus due to tissue depos- mize the amount of luid administered intraoperatively,
ition, predominantly in macrophages.77 Although the to reduce postoperative complications. Clinical studies
etiology of acute tubular necrosis is multifactorial, vol- demonstrated that Doppler-guided luid replacement
ume resuscitation with HES 200/0.62 increased the risk decreased length of hospital stay and reduced unfavora-
of renal failure, and caused a 2.6-fold increase in the ble outcomes in patients undergoing major abdominal,
risk of acute kidney failure in severely septic patients.79 orthopedic, cardiac, and vascular surgery, especially in
Besides their adverse efect on renal function, the elderly patients.88 Intraoperative luid optimization, a
overall side efect proile of hydroxyethyl starches luid replacement strategy adapted to the patients indi-
appears less advantageous, compared with albumin. vidual needs, may therefore be the strategy of choice to
Adverse efects of slowly degradable HES adminis- improve patient outcomes and, as such, may be more
tration include coagulation abnormalities, bleeding, important than the choice of luid type. he following
and allergic reactions. HES 200/0.5 has been shown to subsection will discuss the principles and techniques
decrease the level of circulating von Willebrand factor for diferent methods for goal-directed luid therapy.
and coagulation factor VIII in healthy volunteers, even
when the administered amounts remained below the Static variables of preload and fluid
recommended cumulative daily dose,80 as well as to responsiveness
inhibit platelet aggregation.81,82 hese side efects, how-
ever, are much less pronounced with the administra- Cardiac lling pressures
tion of low-molecular weight, low-molar substitution Numerous studies have shown that cardiac illing
hydroxyethyl starch solutions (HES 130/0.4).25,28,8386 pressures such as central venous pressure (CVP) and
Until the results of ongoing randomized controlled tri- pulmonary artery occlusion pressure (PAOP) are not
32 als using HES 130/0.4 are known, albumin should be suitable to accurately relect the preload. A recent
Chapter 2: Fluid management

Pulmonary artery occlusion pressure


Recent studies have demonstrated that PAOP is a poor
predictor of preload and volume responsiveness.92 he
use of POAP with a rapid thermistor and electrocar-
SV diogram (ECG) electrode allows recognition along the
rewarming phase of the thermodilution curve of a series
of plateaus, which are due to pulsatile ejections of blood
from the right ventricle (RV). he temperature drop
between two successive beats allows computation of the
RV ejection fraction (RVEF). Knowledge of RVEF allows
Positive-pressure the calculation of the right ventricular end-systolic and
Preload
ventilation end-diastolic volumes from the stroke volume.93
Figure 2.3 The cyclic changes in RV and LV stroke volume are he let ventricular enddiastolic area (LVEDA) has
greater when the ventricles operate on the steep rather than the been measured by transesophageal echocardiography
at portion of the FrankStarling curve. (Reprinted with permission
from Marik PE, Techniques for assessment of intravascular (TEE) in patients undergoing mechanical ventilation.
volume in critically ill patients. Journal of Intensive Care Medicine However, the LVEDA has the same limitations as those
2009;24:329337.) reported for invasive cardiac illing pressures. Subcostal
transthoracic echocardiography (TTE) has been used to
systemic review by Marik et al.89 demonstrated a very measure the diameter of the inferior vena cava (IVC) as
poor relationship between CVP and blood volume, as it enters the right atrium. A collapsed IVC is assumed
well as the inability of CVP/change of CVP to predict to be indicative of volume depletion while a distended
the hemodynamic response to a luid challenge. he IVC is relective of high right atrial pressure. However,
authors recommend that CVP should not be used to measurement of IVC diameter is an indirect indicator of
make clinical decisions regarding luid management. the CVP and is associated with all the limitations of CVP
he cardiac illing pressures measured by either CVP measurement.93 It has been clearly stated by Vincent and
and/or PAOP failed to be a reliable predictor of luid Weil.94 that estimates of intravascular volume based on
responsiveness. First, the response of stroke volume to any given level of illing pressure do not reliably predict a
enhanced preload depends on cardiac function. he patients response to luid administration.
luid bolus will enhance the stroke volume if it is given
in the steep part of the FrankStarling curve of normal
Global end-diastolic volume obtained by
heart function, while it will have no efect or even be transpulmonary thermodilution
harmful if it has been given in lat part of the Frank he measurement of global end-diastolic volume
Starling curve or the failing heart (Fig. 2.3). Second, (GEDV) is considered a volumetric static variable of
the illing pressures are highly dependent on let ven- preload using the mathematical analysis of the trans-
tricular compliance, which is frequently altered in pulmonary indicator dilution curve. Transpulmonary
critically ill patients. he relationship between cardiac indicator dilution curves are used in all commercial
illing pressures and end-diastolic volumes is curvi- available monitors. Temperature in the PiCCO system
linear and varies between individuals. Consequently, and lithium in the LidCO system are the most popu-
there are no absolute illing pressure values that would lar ones. he technique for measurement of GEDV
produce speciic end-diastolic volumes; all depend on requires injection of cold solutions via the central
ventricular compliance. During spine surgery in prone venous catheter in the PiCCO system or lithium via
position, there is an increase in intrathoracic pressure, the peripheral venous catheter in the LidCO system,
which is accompanied by an increase in pericardial followed by calculation of mean transit time (MTt) of
pressure and consequently by an increase in illing the thermal indicator (detection of the downstream
pressures, making them an unsuitable tool to guide changes in temperature), which is usually measured
the luid management in prone position. In conclu- at a central artery (femoral, axillary, brachial) in the
sion, static cardiac illing pressures are not appropri- PiCCO system or a peripheral artery in the LidCO
ate to assess intravascular volume status or to predict system. he product of cardiac output (CO) and MTt
the luid responsiveness in spine surgery, especially in is the volume of distribution of the thermal indicator, 33
prone position.90,91 which is considered the intrathoracic thermal volume
Section 1: General considerations

(ITTV), which theoretically represents the sum of FloTrac/Vigileo


intrathoracic blood volume (ITBV) and extravascular
he FloTrac/Vigileo system can be used to continu-
lung water (EVLW). he product of CO and the expo-
ously measure CO from a peripheral arterial cath-
nential down-slope time of the thermodilution curve
eter. he system measures the pulsatility of the arterial
(DStT) is the pulmonary thermal volume (PTV), which
waveform by calculating the standard deviation of the
is composed of pulmonary blood volume and EVLW.
arterial pressure wave over a 20-second period. his is
ITTV = CO MTt (2.2) multiplied by a constant quantifying arterial compli-
ance and vascular resistance based on patient demo-
PTV = CO DStT (2.3)
graphic data (age, sex, height, and weight). he system
PTV = Pulmonary blood volume + EVLW (2.4) is based on the principle that if pressure is measured
directly and the resistance is known, low (CO) can then
GEDV = ITTV PTV (2.5)
be calculated. he system constantly ine-tunes itself on
Global end-diastolic volume is the diference between the basis of the character of the arterial waveform and
ITTV and PTV and is supposed to be the sum of the autocalibrates every minute.98 he disadvantage of the
right and let heart end-diastolic volumes. system is that it uses autocalibration based on experi-
Sakka et al.95 demonstrated that the ITBV index was mental data and is not calibrated for every patient.
closely correlated with the stroke volume index (SVI,
r = 0.66), whereas neither CVP nor PAOP showed sig- Dynamic variables of fluid responsiveness
niicant correlation (r = 0.10 and 0.06, respectively). Positive pressure ventilation induces cyclic changes
Changes in SVI were also more likely to be mirrored in the loading conditions of the let and right ventri-
by changes in ITBV in this study. GEDV was used as a cles. Mechanical ventilation decreases preload and
useful indicator of preload and potentially as a variable increases aterload of the right ventricle (RV) and
to predict luid responsiveness with acceptable sensi- consequently decreases let ventricle (LV) preload in
tivity and speciicity.96 It should be mentioned that response to a reduction in venous return. Observing
GEDV is a static indicator of preload and may prove and analyzing the resulting efects on stroke volume
less useful than dynamic measures of preload respon- (SV), or its surrogates such as pulse pressure (PP) or
siveness such as systolic and pulse pressure variation. systolic pressure (SP), is known as functional hemo-
However, the use of dynamic measures for preload dynamic monitoring.
responsiveness is limited to mechanically controlled
patients. GEDV is a particularly useful tool in spon- Systolic and pulse pressure variation
taneously breathing patients and thus puts GEDV in Mechanical ventilation induces cyclic changes in systo-
a unique position for guidance of luid administra- lic and pulse pressures, referred to as systolic pressure
tion96,97 (Fig. 2.4). variation (SPV) and pulse pressure variation (PPV).
hese parameters are surrogates for stroke volume (SV)
variation (SVV). During the positive pressure breath
GEDV of mechanical ventilation, LV preload is enhanced by
squeezing of blood from the pulmonary capillaries and
veins into the let side of the heart. Simultaneously, there
is a decrease in LV aterload.99 At the same time, inspi-
RAEDV RVEDV PTV LAEDV LVEDV ration raises intrathoracic pressure, causing a reduc-
tion in right ventricular preload and an increase in
aterload. Over the course of a few heartbeats (because
of the long blood pulmonary transit time), the conse-
ITTV
quent reduction in right ventricular SV impacts on LV
Figure 2.4 Schematic diagram of the relevant intrathoracic illing as the RV and LV are in series. he LV preload
uid compartments and their derivation. RAEDV, right atrial end-
diastolic volume; RVEDV, right ventricular end-diastolic volume; PTV, reduction may induce a decrease in LV stoke volume,
pulmonary thermal volume. LAEDV, left atrial end-diastolic volume; which is at its minimum during the expiratory period
LVEDV, left ventricular end-diastolic volume. GEDV, global end- of mechanical ventilation.93 he variation across the
diastolic volume; ITTV, intrathoracic thermal volume. (Reprinted
34 with permission from Renner J, et al. Monitoring uid therapy. Best respiratory cycle is normal and is essentially the oppo-
Practice & Research. Clinical Anaesthesiology 2009;23:159171.) site physiology of pulsus paradoxus; this reduction is
Chapter 2: Fluid management

Figure 2.5 The eect of respiratory


changes on the magnitude of
biventricular preload dependence.
RV preload
Blood
RV stroke LV stroke
pulmonary LV preload
volume volume
transit time
RV afterload
Pleural
pressure

LV afterload
Transpulmonary LV stroke
pressure volume

LV preload

Pulse pressure maximum Pulse pressure mimimum


at the end of inspiration during expiratory period

exaggerated during hypovolemic states due to collaps- hypotension. Pizov et al.101 induced hypotension in
ibility of the venae cavae and therefore reduces the RV ventilated animals either by hypovolemia or sodium
preload. Also, the transmission of pressure through the nitroprusside infusion. PAOP and CVP were similarly
RV in low-volume states reduces right heart illing. he reduced in both groups, whereas SPV and down were
cyclic changes in RV and LV stroke volume are greater signiicantly increased only in the hemorrhagic group,
when the ventricles operate on the steep rather than the distinguishing between preload and vasodilatation
lat portion of the FrankStarling curve. herefore, the as the cause of hypotension. Identifying the cause of
magnitude of the respiratory changes in LV stroke vol- hypotension as either drug induced or hypovolemia is
ume is an indicator of biventricular preload depend- crucial to correctly treating hemodynamic changes in
ence100 (Fig. 2.5). anesthetized patients.
he arterial systolic pressure variation is deined as Stroke volume can be calculated continuously by
the diference between maximal and minimal systo- the PiCCO system by measuring the systolic portion
lic arterial pressure during one mechanical breath. Its of the aortic pressure waveform and dividing the area
down component is calculated as down = (apneic under the curve by the aortic impedance, which is
minimum systolic blood pressure). SVV and its down determined initially by transpulmonary thermodilu-
component have been shown to be sensitive indicators tion.96 SVV measured by PiCCO has been shown to
of hypovolemia. he increase in systolic pressure above correlate well with SPV and to predict luid responsive-
baseline (baseline is measured at apnea) is deined as ness, being superior in this regard to CVP and PAOP.97
up, and the decrease in systolic pressure below base- In neurosurgical patients, Berkenstadt et al.102 found
line is known as the down component. SPV is the sum that a SVV of 9.5% or more predicted a SV increase of
of up and down. SPV is inluenced by transmitted more than 5% in response to standard luid challenge,
changes in pleural pressure, meaning that increas- with a sensitivity of 79% and a speciicity of 93%. It is
ing inspiratory pressures may falsely accentuate SPV. interesting to note that SVV was able to predict luid
PPV is not inluenced in the same way, as inspiratory response equally well in patients with reduced let ven-
pressure changes are transmitted to both systolic and tricular function (ejection fraction 35%) and those
diastolic components of the arterial waveform, and with normal cardiac function, as demonstrated by
pulse pressure remains unchanged. Consequently, Reuter et al.103 (Fig. 2.8 and Table 2.1)
PVV will relect changes in cardiac output more accu-
rately, and may be a better index of preload responsive- Limitations of heartlung interaction as a predictor of
ness. (Figs. 2.6 and 2.7) uid responsiveness
he irst limitation of using SVV, SPV, and PPV is its
Stroke volume variation and pulse contour analysis restriction to mechanically ventilated patients with
SPV and PPV are helpful to diferentiate between no spontaneous breathing activity. he second limita- 35
hypovolemia and vasodilatation as the cause of tion is its tidal volume dependence. Since respiratory
Section 1: General considerations

40 Figure 2.6 Respiratory cycle-induced


cmH2O changes in arterial systolic pressure.
The reference line indicates apnea and
Paw

allows for measurement of the up and


down components of systolic pressure
5 variation. Paw, positive airway pressure;
SPV, systolic pressure variation. (Reprinted
SPV with permission from Renner J, et al.,
Monitoring uid therapy. Best Practice
& Research. Clinical Anaesthesiology
mmHg

2009;23:159171.)

up
down
mmHg

40 Figure 2.7 Respiratory cycle-induced


changes in pulse pressure (PP). Pulse
cmH2O
Paw

pressure variation is calculated between


the maximal (PPmax) and minimal (PPmin)
values of pulse pressure. Paw, positive airway
5
pressure. (Reprinted with permission from
Renner J, et al., Monitoring uid therapy. Best
mmHg

Practice & Research Clinical Anaesthesiology


2009;23:159171.)
PPmax PPmin

40 Figure 2.8 Respiratory cycle-induced


changes in stroke volume variation
cmH2O

is calculated between the maximal


Paw

(SVmax) and minimal (SVmin) values of


stroke volume. Paw, positive airway
5 pressure. (Reprinted with permission
from Renner J, et al., Monitoring uid
therapy. Best Practice & Research. Clinical
mmHg

Anaesthesiology 2009;23:159171.)

SVmax
SVmin

changes in preload are induced in pleural pressure, during each breath is measured from the arterial pres-
lower tidal volumes reduce the magnitude of SVV or sure waveform, and the slope of the line of best it of
PPV. herefore, using tidal volume 8 ml/kg is currently these three points is calculated. he steeper the gradi-
recommended while SVV or PPV are being monitored. ent, the emptier the circulation and better response to
he Respiratory Systolic Variation Test (RSVT) is sug- luid challenge. Combined RSTV with PPV was found
gested as a solution to this problem. RSVT is conducted to be the most sensitive and speciic predictor of a
by delivering three consecutive mechanical breaths of response to a luid challenge as assessed by TEE. he
36 10, 20, and 30 cmH2O causing a progressive swing in third limitation of PPV or SVV is that it cannot be used
the arterial trace. he minimum systolic blood pressure in patients with arrhythmias.
Chapter 2: Fluid management

Table 2.1 Dynamic variables of uid responsiveness

Variable Description Calculation Monitoring


Delta down (down) Dierence between systolic arterial Invasive arterial pressure
Papnoea SAP
Pexp
pressure (SAP) in apnea and at end-expiration recordings and appropriate
(minimal value during one mechanical monitor
ventilatory cycle)
Delta up (up) Dierence between maximal Invasive arterial pressure
SAP
Pinsp SAP
Papnoea
SAP value during mechanical ventilatory recordings and appropriate
cycle and apneic SAP monitor
Systolic pressure Systolic arterial pressure variation Pinsp SAP
Pexp Invasive arterial pressure
variation (SPV) during one mechanical ventilatory cycle: sum recordings and appropriate
of up + up monitor
Pulse pressure Pulse pressure (PP) variation Invasive arterial pressure
PP
Pmax PPPmin
variation (PPV) calculated from the mean values of four 100 recordings and appropriate
minimum and maximum SVs averaged
during the previous 30 s 2
(
1 PP
Pmax PPPmin ) monitor (PiCCO, LIDCO)

Stroke volume Stroke volume (SV) variation calculated PiCCO, LIDCO


SV
Vmax SVVmin
variation (SW) from the mean values of 4 minimum and 100
maximum 4 SVs averaged during the
previous 30 s 2
(
1 SV
Vmax SVVmin )

Variation in Pleth variability index (PVI) calculates PVI, Masimo Radical-7


pulse oxymetric the respiratory variations in the
Plmax Plmin
plethysmographic plethysmography waveform amplitude Plmax
waveform (POP)
Peak aortic ow Peak aortic blood ow velocity Echocardiography
Vpk max Vpk min
velocity variation variation (Vpeak) during one mechanical 100
(Vpeak) ventilatory cycle
2
(
1 Vpk
max Vpk min )
* PiCCO, Pulse Contour Cardiac Output monitoring system.

Dynamic changes in aortic ow velocity/stroke volume SV following a luid challenge105 demonstrated that use
assessed by esophageal Doppler of an esophageal Doppler probe in multiple-trauma
Esophageal Doppler permits rapid, minimally invasive, patients was associated with a decrease in blood lactate
and continuous estimation of cardiac output. Cardiac levels, a lower incidence of infectious complications,
output measurements by esophageal Doppler compare and a reduced duration of intensive care and hospitals
acceptably with the direct Fick method, and are even stays. A meta-analysis by Walsh et al.88 demonstrated
more reliable than the thermodilution method.104 that intraoperative goal-directed luid therapy (GDT)
he esophageal Doppler technique measures blood using an esophageal Doppler probe signiicantly
low velocity in the descending aorta by means of a reduces postoperative complication rates and length of
Doppler transducer (4 MHz continuous wave or 5 MHz hospital stay.
pulsed wave, depending on manufacturer) placed at
the tip of a lexible probe. he probe is introduced into Suitable and practical techniques for goal-
the esophagus of the mechanically ventilated patient directed fluid therapy during spine surgery
and then rotated so that the transducer faces the aorta, Fluid management during spine surgery in the prone
and a characteristic aortic velocity signal is obtained. position represents a challenge due to decreased RV
he cardiac output is calculated based on the diameter preload induced by increased intrathoracic pressure,
of the aorta (measured or estimated) and the measured and consequently SV. he use of GDT is considered the
low velocity of blood in the aorta. he duration of the preferred method to maintain proper intravascular ill-
aortic velocity corrected for heart rate is called low time ing without luid overload. Fluid overload during spine
corrected (FTc) and is considered a static indicator of surgery is a major problem resulting in increased facial
cardiac preload.93 he esophageal Doppler allows the edema, delayed postoperative extubation, and even in 37
optimization of SV based on the FTc, and the change in increased length of hospital stay.
Section 1: General considerations

During spine surgery, GDT can be easily conducted 6. Tobin A, Kelly W. Prone ventilation its time.
using either SVV or PPV. SVV can be measured by Anaesth Intensive Care 1999; 27: 194201.
pulse contour analysis using the FloTrac/Vigileo sys- 7. Nieuwdorp M, Meuwese MC, Vink H, Hoekstra JB,
tem. Fluid boluses of 200 ml over 2 minutes are usually Kastelein JJ, Stroes ES. he endothelial glycocalyx: a
given to the patient when SVV increases by more than potential barrier between health and vascular disease.
Curr Opin Lipidol 2005; 16: 50711.
14% (sensitivity = 94%, speciicity = 80%).106 he other
technique for SVV is measuring dynamic changes of 8. Paptistella M, Chappell D, Hofmann-Kiefer K,
descending aortic blood low and SV by esophageal Kammerer T, Conzen P, Rehm M. he role of the
glycocalyx in transvascular luid shits. Transfus
Doppler probe. Doppler is used to guide the luid Altern Transfus Med 2010; 11: 92101.
boluses to maintain FTc >0.35 s and/or to keep giving
9. Hu X, Weinbaum S. A new view of Starlings
luid boluses as long as SV still increases by more than
hypothesis at the microstructural level. Microvasc Res
10%. In the PPV technique luid boluses are usually 1999; 58: 281304.
given if PPV is >15% (sensitivity = 100%, speciicity
10. Jacob M, Bruegger D, Rehm M, et al. he endothelial
= 80%).106 In our practice, we prefer to use esophageal glycocalyx afords compatibility of Starlings
Doppler probe or PPV to guide luid management dur- principle and high cardiac interstitial albumin levels.
ing spine surgery, especially in the prone position. It Cardiovasc Res 2007; 73: 57586.
should be remembered that tidal volume should be 11. Fleck A, Raines G, Hawker F, et al. Increased vascular
810 ml/kg during the measurement period. If the permeability: a major cause of hypoalbuminaemia in
blood pressure remains lower than the required target disease and injury. Lancet 1985; 1: 7814.
even ater luid supplementation, we administer vaso- 12. Vink H, Constantinescu AA, Spaan JA. Oxidized
pressors or inotropes to reach target blood pressure. lipoproteins degrade the endothelial surface layer:
implications for platelet-endothelial cell adhesion.
Conclusion Circulation 2000; 101: 15002.
Fluid management during spine surgery in the prone 13. Rehm M, Bruegger D, Christ F, et al. Shedding of the
position is a very diicult but important task to ensure endothelial glycocalyx in patients undergoing major
vascular surgery with global and regional ischemia.
patient safety and successful outcome. Fluid manage- Circulation 2007; 116: 1896906.
ment should be informed by a goal-directed approach.
14. Nieuwdorp M, van Haeten TW, Gouverneur MC,
he basal luid maintenance should not exceed 1 ml/kg/h
et al. Loss of endothelial glycocalyx during acute
using crystalloid solution. his can be supplemented hyperglycemia coincides with endothelial dysfunction
by crystalloids or colloids according to hemodynamic and coagulation activation in vivo. Diabetes 2006; 55:
goals. Colloids are indicated to replace plasma deicits 4806.
due to blood loss or as boluses to maintain normo- 15. Bruegger D, Jacob M, Rehm M, et al. Atrial natriuretic
volemia using a goal-directed approach. peptide induces shedding of endothelial glycocalyx
in coronary vascular bed of guinea pig hearts. Am J
References Physiol Heart Circ Physiol 2005; 289: H19939.
1. Wadsworth R, Anderton JM, Vohra A. he efect 16. Jacob M, Chappell D, Rehm M. he third space fact
of four diferent surgical prone positions on or iction? Best Pract Res Clin Anaesthesiol 2009; 23:
cardiovascular parameters in healthy volunteers. 14557.
Anaesthesia 1996; 51: 81922. 17. Chappell D, Jacob M, Hofmann-Kiefer K, Conzen P,
2. Toyota S, Amaki Y. Hemodynamic evaluation of the Rehm M. A rational approach to perioperative luid
prone position by transesophageal echocardiography. management. Anesthesiology 2008; 109: 72340.
J Clin Anesth 1998; 10: 325. 18. Jacob M, Bruegger D, Rehm M, Welsch U, Conzen
3. Kim KA, Wang MY. Anesthetic considerations in P, Becker BF. Contrasting efects of colloid and
the treatment of cervical myelopathy. Spine J 2006; 6: crystalloid resuscitation luids on cardiac vascular
207S11S. permeability. Anesthesiology 2006; 104: 122331.
4. Pearce DJ. he role of posture in laminectomy. Proc R 19. Kimberger O, Arnberger M, Brandt S, et al. Goal-
Soc Med 1957; 50: 10912. directed colloid administration improves the
5. Nyren S, Mure M, Jacobsson H, Larsson SA, Lindahl microcirculation of healthy and perianastomotic
SG. Pulmonary perfusion is more uniform in the colon. Anesthesiology 2009; 110: 496504.
38 prone than in the supine position: scintigraphy in 20. Moller H, Hedlund R. Instrumented and
healthy humans. J Appl Physiol 1999; 86: 113541. noninstrumented posterolateral fusion in adult
Chapter 2: Fluid management

spondylolisthesis a prospective randomized study: 35. Jungner M, Grande PO, Mattiasson G, Bentzer P.
part 2. Spine (Phila Pa 1976) 2000; 25: 171621. Efects on brain edema of crystalloid and albumin
21. Batson OV. he function of the vertebral veins and luid resuscitation ater brain trauma and hemorrhage
their role in the spread of metastases. Ann Surg 1940; in the rat. Anesthesiology 2010; 112: 1194203.
112: 13849. 36. Finfer S, Bellomo R, Boyce N, French J, Myburgh J,
22. Rose B, Post T, eds. Clinical Physiology of AcidBase Norton R. A comparison of albumin and saline for
and Electrolyte Disorders. 5th ed. New York: McGraw- luid resuscitation in the intensive care unit. N Engl J
Hill; 2001. Med 2004; 350: 224756.
23. Rackow EC, Falk JL, Fein IA, et al. Fluid resuscitation 37. Waters JH, Gottlieb A, Schoenwald P, Popovich MJ,
in circulatory shock: a comparison of the Sprung J, Nelson DR. Normal saline versus lactated
cardiorespiratory efects of albumin, hetastarch, and Ringers solution for intraoperative luid management
saline solutions in patients with hypovolemic and in patients undergoing abdominal aortic aneurysm
septic shock. Crit Care Med 1983; 11: 83950. repair: an outcome study. Anesth Analg 2001; 93:
81722.
24. McIlroy DR, Kharasch ED. Acute intravascular
volume expansion with rapidly administered 38. Ho AM, Karmakar MK, Contardi LH, Ng SS, Hewson
crystalloid or colloid in the setting of moderate JR. Excessive use of normal saline in managing
hypovolemia. Anesth Analg 2003; 96: 15727, table of traumatized patients in shock: a preventable
contents. contributor to acidosis. J Trauma 2001; 51: 1737.
25. Kozek-Langenecker SA. Efects of hydroxyethyl starch 39. McFarlane C, Lee A. A comparison of Plasmalyte 148
solutions on hemostasis. Anesthesiology 2005; 103: and 0.9% saline for intra-operative luid replacement.
65460. Anaesthesia 1994; 49: 77981.
26. von Hoegen I, Waller C. Safety of human albumin 40. Scheingraber S, Rehm M, Sehmisch C, Finsterer
based on spontaneously reported serious adverse U. Rapid saline infusion produces hyperchloremic
events. Crit Care Med 2001; 29: 9946. acidosis in patients undergoing gynecologic surgery.
Anesthesiology 1999; 90: 126570.
27. Kozek-Langenecker SA. Inluence of luid therapy on
the haemostatic system of intensive care patients. Best 41. Mathes DD, Morell RC, Rohr MS. Dilutional acidosis:
Pract Res Clin Anaesthesiol 2009; 23: 22536. is it a real clinical entity? Anesthesiology 1997; 86:
5013.
28. Barron ME, Wilkes MM, Navickis RJ. A systematic
review of the comparative safety of colloids. Arch Surg 42. Stewart PA. Independent and dependent variables of
2004; 139: 55263. acid-base control. Respir Physiol 1978; 33: 926.
29. Kannan S, Milligan KR. Moderately severe 43. Figge J, Rossing TH, Fencl V. he role of serum
anaphylactoid reaction to pentastarch (200/0.5) in a proteins in acid-base equilibria. J Lab Clin Med 1991;
patient with acute severe asthma. Intensive Care Med 117: 45367.
1999; 25: 2202. 44. Burch JM, Ortiz VB, Richardson RJ, Martin RR,
30. Herwaldt LA, Swartzendruber SK, Edmond MB, Mattox KL, Jordan GL, Jr. Abbreviated laparotomy
et al. he epidemiology of hemorrhage related and planned reoperation for critically injured
to cardiothoracic operations. Infect Control Hosp patients. Ann Surg 1992; 215: 47683; discussion
Epidemiol 1998; 19: 916. 4834.
31. Cittanova ML, Leblanc I, Legendre C, Mouquet 45. Yudkin J, Cohen RD, Slack B. he haemodynamic
C, Riou B, Coriat P. Efect of hydroxyethylstarch efects of metabolic acidosis in the rat. Clin Sci Mol
in brain-dead kidney donors on renal function in Med 1976; 50: 17784.
kidney-transplant recipients. Lancet 1996; 348: 46. Wilcox CS. Regulation of renal blood low by plasma
16202. chloride. J Clin Invest 1983; 71: 72635.
32. Choi PT, Yip G, Quinonez LG, Cook DJ. Crystalloids 47. Wilkes NJ, Woolf R, Mutch M, et al. he efects
vs. colloids in luid resuscitation: a systematic review. of balanced versus saline-based hetastarch and
Crit Care Med 1999; 27: 20010. crystalloid solutions on acid-base and electrolyte
33. Alderson P, Schierhout G, Roberts I, Bunn F. Colloids status and gastric mucosal perfusion in elderly
versus crystalloids for luid resuscitation in critically surgical patients. Anesth Analg 2001; 93: 81116.
ill patients. Cochrane Database Syst Rev 2000: 48. hompson RC. Physiological 0.9% saline in the luid
CD000567. resuscitation of trauma. J R Army Med Corps 2005;
34. Virgilio RW, Rice CL, Smith DE, et al. Crystalloid vs. 151: 14651.
colloid resuscitation: is one better? A randomized 49. Martin G, Bennett-Guerrero E, Wakeling H, 39
clinical study. Surgery 1979; 85: 12939. et al. A prospective, randomized comparison of
Section 1: General considerations

thromboelastographic coagulation proile in patients 62. Rhee P, Wang D, Ruf P, et al. Human neutrophil
receiving lactated Ringers solution, 6% hetastarch in activation and increased adhesion by various
a balanced-saline vehicle, or 6% hetastarch in saline resuscitation luids. Crit Care Med 2000; 28: 748.
during major surgery. J Cardiothorac Vasc Anesth 63. Horstick G, Lauterbach M, Kempf T, et al. Early
2002; 16: 4416. albumin infusion improves global and local
50. Schramko A, Suojaranta-Ylinen R, Kuitunen A, hemodynamics and reduces inlammatory response
Raivio P, Kukkonen S, Niemi T. Hydroxyethylstarch in hemorrhagic shock. Crit Care Med 2002; 30: 8515.
and gelatin solutions impair blood coagulation ater 64. Qiao R, Silinger-Birnboim A, Lum H, Tiruppathi C,
cardiac surgery: a prospective randomized trial. Br J Malik AB. Albumin and Ricinus communis agglutinin
Anaesth 2010; 104: 6917. decrease endothelial permeability via interactions
51. Phillips CR, Vinecore K, Hagg DS, et al. Resuscitation with matrix. Am J Physiol 1993; 265: C43946.
of haemorrhagic shock with normal saline vs. lactated 65. Haraldsson B, Rippe B. Serum factors other than
Ringers: efects on oxygenation, extravascular lung albumin are needed for the maintenance of normal
water and haemodynamics. Crit Care 2009; 13: R30. capillary permselectivity in rat hindlimb muscle. Acta
52. White SA, Goldhill DR. Is Hartmanns the solution? Physiol Scand 1985; 123: 42736.
Anaesthesia 1997; 52: 4227. 66. Vogt NH, Bothner U, Lerch G, Lindner KH,
53. Williams EL, Hildebrand KL, McCormick SA, Georgief M. Large-dose administration of 6%
Bedel MJ. he efect of intravenous lactated Ringers hydroxyethyl starch 200/0.5 total hip arthroplasty:
solution versus 0.9% sodium chloride solution on plasma homeostasis, hemostasis, and renal function
serum osmolality in human volunteers. Anesth Analg compared to use of 5% human albumin. Anesth Analg
1999; 88: 9991003. 1996; 83: 2628.
54. OMalley CM, Frumento RJ, Hardy MA, et al. A 67. Vogt N, Bothner U, Brinkmann A, de Petriconi R,
randomized, double-blind comparison of lactated Georgief M. Peri-operative tolerance to large-dose
Ringers solution and 0.9% NaCl during renal 6% HES 200/0.5 in major urological procedures
transplantation. Anesth Analg 2005; 100: 151824, compared with 5% human albumin. Anaesthesia 1999;
table of contents. 54: 1217.
55. Boldt J, Haisch G, Suttner S, Kumle B, Schellhase 68. Rozich JD, Paul RV. Acute renal failure precipitated by
F. Are lactated Ringers solution and normal saline elevated colloid osmotic pressure. Am J Med 1989; 87:
solution equal with regard to coagulation? Anesth 35960.
Analg 2002; 94: 37884, table of contents. 69. Belayev L, Busto R, Zhao W, Clemens JA, Ginsberg MD.
56. Cusi K, Consoli A, DeFronzo RA. Metabolic efects Efect of delayed albumin hemodilution on infarction
of metformin on glucose and lactate metabolism volume and brain edema ater transient middle cerebral
in noninsulin-dependent diabetes mellitus. J Clin artery occlusion in rats. J Neurosurg 1997; 87: 595601.
Endocrinol Metab 1996; 81: 405967. 70. Belayev L, Liu Y, Zhao W, Busto R, Ginsberg MD.
57. Miletin MS, Stewart TE, Norton PG. Inluences on Human albumin therapy of acute ischemic stroke:
physicians choices of intravenous colloids. Intensive marked neuroprotective eicacy at moderate doses
Care Med 2002; 28: 91724. and with a broad therapeutic window. Stroke 2001; 32:
58. Laxenaire MC, Charpentier C, Feldman L. 55360.
Anaphylactoid reactions to colloid plasma substitutes: 71. Delaney AP, Dan A, McCafrey J, Finfer S. he role
incidence, risk factors, mechanisms. A French of albumin as a resuscitation luid for patients with
multicenter prospective study. Ann Fr Anesth Reanim sepsis: a systematic review and meta-analysis. Crit
1994; 13: 30110. Care Med 2011; 39: 38691.
59. Dieterich HJ, Krat D, Sirtl C, et al. Hydroxyethyl 72. Boldt J, Zickmann B, Rapin J, Hammermann H,
starch antibodies in humans: incidence and clinical Dapper F, Hempelmann G. Inluence of volume
relevance. Anesth Analg 1998; 86: 11236. replacement with diferent HES-solutions on
60. Simon DI, Stamler JS, Jaraki O, et al. Antiplatelet microcirculatory blood low in cardiac surgery. Acta
properties of protein S-nitrosothiols derived from Anaesthesiol Scand 1994; 38: 4328.
nitric oxide and endothelium-derived relaxing factor. 73. Rittoo D, Gosling P, Simms MH, Smith SR, Vohra
Arterioscler hromb 1993; 13: 7919. RK. he efects of hydroxyethyl starch compared with
61. Nathan C, Xie QW, Halbwachs-Mecarelli L, Jin WW. gelofusine on activated endothelium and the systemic
Albumin inhibits neutrophil spreading and hydrogen inlammatory response following aortic aneurysm
peroxide release by blocking the shedding of CD43 repair. Eur J Vasc Endovasc Surg 2005; 30: 5204.
40 (sialophorin, leukosialin). J Cell Biol 1993; 122: 74. Feng X, Yan W, Wang Z, et al. Hydroxyethyl starch,
24356. but not modiied luid gelatin, afects inlammatory
Chapter 2: Fluid management

response in a rat model of polymicrobial sepsis with abdominal surgery: systematic review and meta-
capillary leakage. Anesth Analg 2007; 104: 62430. analysis. Int J Clin Pract 2008; 62: 46670.
75. Brandstrup B, Tonnesen H, Beier-Holgersen R, 89. Marik PE, Baram M, Vahid B. Does central venous
et al. Efects of intravenous luid restriction on pressure predict luid responsiveness? A systematic
postoperative complications: comparison of two review of the literature and the tale of seven mares.
perioperative luid regimens: a randomized assessor- Chest 2008; 134: 1728.
blinded multicenter trial. Ann Surg 2003; 238: 6418. 90. Hollenberg SM, Ahrens TS, Annane D, et al. Practice
76. Treib J, Baron JF, Grauer MT, Strauss RG. An parameters for hemodynamic support of sepsis in
international view of hydroxyethyl starches. Intensive adult patients: 2004 update. Crit Care Med 2004; 32:
Care Med 1999; 25: 25868. 192848.
77. Bork K. Pruritus precipitated by hydroxyethyl starch: 91. Pinsky MR. Clinical signiicance of pulmonary artery
a review. Br J Dermatol 2005; 152: 312. occlusion pressure. Intensive Care Med 2003; 29:
78. Niemi TT, Miyashita R, Yamakage M. Colloid 1758.
solutions: a clinical update. J Anesth 2010; 24: 91325. 92. Michard F, Teboul JL. Predicting luid responsiveness
79. Schortgen F, Lacherade JC, Bruneel F, et al. Efects of in ICU patients: a critical analysis of the evidence.
hydroxyethylstarch and gelatin on renal function in Chest 2002; 121: 20008.
severe sepsis: a multicentre randomised study. Lancet 93. Marik PE. Techniques for assessment of intravascular
2001; 357: 91116. volume in critically ill patients. J Intensive Care Med
80. de Jonge E, Levi M, Buller HR, Berends F, Kesecioglu 2009; 24: 32937.
J. Decreased circulating levels of von Willebrand 94. Vincent JL, Weil MH. Fluid challenge revisited. Crit
factor ater intravenous administration of a rapidly Care Med 2006; 34: 13337.
degradable hydroxyethyl starch (HES 200/0.5/6) in 95. Sakka SG, Bredle DL, Reinhart K, Meier-Hellmann
healthy human subjects. Intensive Care Med 2001; 27: A. Comparison between intrathoracic blood volume
18259. and cardiac illing pressures in the early phase of
81. Omar MN, Shouk TA, Khaleq MA. Activity of hemodynamic instability of patients with sepsis or
blood coagulation and ibrinolysis during and septic shock. J Crit Care 1999; 14: 7883.
ater hydroxyethyl starch (HES) colloidal volume 96. Renner J, Scholz J, Bein B. Monitoring luid therapy.
replacement. Clin Biochem 1999; 32: 26974. Best Pract Res Clin Anaesthesiol 2009; 23: 15971.
82. Franz A, Braunlich P, Gamsjager T, Felfernig M, 97. Benington S, Ferris P, Nirmalan M. Emerging trends
Gustorf B, Kozek-Langenecker SA. he efects of in minimally invasive haemodynamic monitoring and
hydroxyethyl starches of varying molecular weights optimization of luid therapy. Eur J Anaesthesiol 2009;
on platelet function. Anesth Analg 2001; 92: 14027. 26: 893905.
83. Westphal M, James MF, Kozek-Langenecker S, Stocker 98. Manecke GR, Jr., Auger WR. Cardiac output
R, Guidet B, Van Aken H. Hydroxyethyl starches: determination from the arterial pressure wave: clinical
diferent products diferent efects. Anesthesiology testing of a novel algorithm that does not require
2009; 111: 187202. calibration. J Cardiothorac Vasc Anesth 2007; 21: 37.
84. Wilkes MM, Navickis RJ, Sibbald WJ. Albumin versus 99. Pinsky MR. Cardiovascular issues in respiratory care.
hydroxyethyl starch in cardiopulmonary bypass Chest 2005; 128: 592S-7S.
surgery: a meta-analysis of postoperative bleeding. 100. Michard F, Teboul JL. Using heart-lung interactions
Ann horac Surg 2001; 72: 52733; discussion 534. to assess luid responsiveness during mechanical
85. Porter SS, Goldberg RJ. Intraoperative allergic ventilation. Crit Care 2000; 4: 2829.
reactions to hydroxyethyl starch: a report of two cases. 101. Pizov R, Yaari Y, Perel A. Systolic pressure variation
Can Anaesth Soc J 1986; 33: 3948. is greater during hemorrhage than during sodium
86. Huttner I, Boldt J, Haisch G, Suttner S, Kumle B, nitroprusside-induced hypotension in ventilated
Schulz H. Inluence of diferent colloids on molecular dogs. Anesth Analg 1988; 67: 1704.
markers of haemostasis and platelet function in 102. Berkenstadt H, Margalit N, Hadani M, et al. Stroke
patients undergoing major abdominal surgery. Br J volume variation as a predictor of luid responsiveness
Anaesth 2000; 85: 41723. in patients undergoing brain surgery. Anesth Analg
87. Shippy CR, Appel PL, Shoemaker WC. Reliability 2001; 92: 9849.
of clinical monitoring to assess blood volume in 103. Reuter DA, Kirchner A, Felbinger TW, et al.
critically ill patients. Crit Care Med 1984; 12: 10712. Usefulness of let ventricular stroke volume variation
88. Walsh SR, Tang T, Bass S, Gaunt ME. Doppler-guided to assess luid responsiveness in patients with reduced 41
intra-operative luid management during major cardiac function. Crit Care Med 2003; 31: 1399404.
Section 1: General considerations

104. Espersen K, Jensen EW, Rosenborg D, et al. levels in multiple-trauma patients: a


Comparison of cardiac output measurement randomized controlled trial. Crit Care
techniques: thermodilution, Doppler, CO2- 2007; 11: R24.
rebreathing and the direct Fick method. Acta 106. Biais M, Bernard O, Ha JC, Degryse C, Sztark F.
Anaesthesiol Scand 1995; 39: 24551. Abilities of pulse pressure variations and stroke
105. Chytra I, Pradl R, Bosman R, Pelnar P, Kasal E, volume variations to predict luid responsiveness in
Zidkova A. Esophageal Doppler-guided prone position during scoliosis surgery. Br J Anaesth
luid management decreases blood lactate 2010; 104: 40713.

42
Section 1 General considerations
Chapter
Blood conservation

3 Robert Helfand

Table 3.1 Methods of blood conservation in spine surgery


Key points
Preoperative optimization of hemoglobin levels
All blood conservation techniques work Preoperative autodonation
better with higher preoperative hemoglobin Antiibrinolytic therapy
concentrations.
Acute normovolemic hemodilution
ESA should be used with caution in spine
Intraoperative cell salvage
surgery due to increased risk of DVT.
Point of care testing of coagulation
Antiibrinolytic therapy is an underutilized
technique.
new eforts at mitigation. Immunologic complications
Cell savers are key elements of blood
have now taken a more key role in the complication pro-
conservation programs.
ile of transfusions. Transfusion-associated acute lung
Blood conservation requires multidisciplinary injury (TRALI) is now recognized as a major complica-
involvement. tion of transfusions. Increased duration of blood stor-
age has been suggested as a source of poor outcomes in
Blood management during spine surgery remains cardiac surgery.2 Even autologous blood can sufer from
a challenge. Perioperative blood administration is clerical errors or infectious complications (see Table 3.2).
expensive, and exposes the recipient to a number of he only way to eliminate transfusion-related complica-
risks. A number of methods are available to decrease tions is to lessen a patients exposure to transfusions.
the need to administer blood during the perioperative he remarkable safety of blood administration has
period (Table 3.1). We will discuss the current tech- come at a considerable cost. Each additional screening
niques and controversies with the goal of lessening the test drives up the cost of each unit of blood products.
patients exposure to blood transfusions. Restrictions on who can donate limit our pool of candi-
he spectrum of blood component-related compli- dates. In an era of sicker patients facing more complex
cations has changed signiicantly in the last few years. surgery, we now face an increasingly limited, expen-
Enhanced technology has allowed for better screening sive, risk-laden resource.
of donated blood units with a dramatic decrease in many One of the most common questions is what does
infectious complications. Estimates of viral transmission blood cost?. Unfortunately, the answer is not straight-
in the blood donor system range from 1:205 000 for hepa- forward. Shander et al. looked at this question at an
titis B to 1:2 135 000 for HIV.1 Extremely sensitive nucleo- American and several European sites.3 Cost estimates
tide screening systems have led to remarkable gains in the ranged from $522 to $1183 dollars per unit of blood.
safety of administration of allogeneic blood units. Costs for additional blood products are also considerable
We now recognize that there are other transfusion and diicult to estimate. A successful blood conserva-
risks that are far greater problems than the previously tion program would have to show a decrease in transfu-
feared viral infections. he West Nile Virus demon- sions, better patient outcomes, and overall lower costs.
strated that new, emerging threats remain ongoing Given the complexity of measuring each cost, blood
problems. Bacterial contaminations, particularly with conservation becomes part of the increasingly complex
platelet products, have now attracted new attention and health care accounting and must prove its success.

43
Anesthesia for Spine Surgery, ed. Ehab Farag. Published by Cambridge University Press. Cambridge University Press 2012.
Section 1: General considerations

Table 3.2 Potential transfusion risks by typea

Type Potential transfusion risk


Infectious Viral: hepatitis A, B, C, E, G; HIV, HTLV-I, HTLV-II, cytomegalovirus, EpsteinBarr virus, parvovirus B19
Bacterial
Variant CreutzfeldtJakob diseaseb
Parasitic: malaria, babesiosis, Chagas disease
Risk of viral contamination with viruses not yet screened for
Noninfectious Hemolytic transfusion reactions
Transfusion errors leading to blood type incompatibility reactions
Febrile nonhemolytic transfusion reactions
Anaphylaxis and urticarial allergic reactions
Posttransfusion purpura
Risks of old blood vs. fresh blood (e.g., microcirculatory occlusion, lack of efect)
Transfusion-related acute lung injury
Circulatory overload
Iron overload
Air embolism
Fat embolism (intraoperative administration of salvaged blood)
Hypotensive reactions (with ACE inhibitors negatively charged leukoreduction ilters)
Metabolic disturbances: citrate toxicity, hypocalcemia, hyperkalemia, acidosis, and hyperammonemia
Hypothermia
Immunologic Multiple organ dysfunction syndrome or multiple organ failure attributed to cytokine release
Postoperative infection
Transfusion-associated sepsis
Increased risk of cancer recurrence
Downregulation of macrophage and T-cell function
Alloimmunization: HLA, especially in patients undergoing chronic transfusion
Transfusion-associated graft versus host disease in immunocompromised and
nonimmunocompromised hosts
HIV = human immunodeiciency virus; HTLV = human T cell lymphoma virus; ACE = angiotensin-converting enzyme;
HLA = human leukocyte antigen.
a
This table presents a broad sampling of potential risks from transfusion of blood products; neither is it all-inclusive
nor is it intended to imply that causality has been proved. Because of limitations in the number of references that
can be cited, those associated with this information are available on request to the corresponding author.
b
Classic CreutzfeldtJakob disease is probably not transmissible through blood products.
Reprinted with permission from reference 16: Shander A and Goodnough LT, Update on transfusion medicine.
Pharmacotherapy 2007;27(9 Pt2):57S68S.)

Predicting which patients are appropriate for presenting for major surgery range from 5% to
aggressive blood conservation measures is also crucial 75.8%. In patients presenting for total hip, knee, or
for successful blood management. Just as transfusion hip fracture surgery anemia rates ranged from 24%
has its costs and limitations, not every patient should to 44%.5 Lower preoperative hemoglobin correlates
receive the same level of conservation techniques. with increased transfusion requirements. While there
Current evidence supports several factors that pre- are a great range of patients presenting for spine sur-
dict the need for transfusion in adult spine surgery: gery, undoubtedly a sizable number will be anemic.
advanced age, preoperative anemia, osteotomy, and Correction of preoperative anemia is the irst step in
fusion. hese have been conirmed by Lenoir et al. with blood management.
their predictive model of transfusion in spine surgery.4 Figure 3.1 shows a targeted anemia work-up for
It is these patients and procedures that will be the best preoperative patients. his focused evaluation should
targets for blood-conserving interventions. be done as soon as it is identiied that a patient for a
major blood loss surgery is anemic. Most preoperative
Preoperative measures anemia is a combination of nutritional deiciencies and
All blood conservation methods work better with ongoing chronic inlammation. herefore it is unlikely
44 patients who have higher hemoglobin concentra- that a single treatment will consistently treat the major-
tions. However, the estimates of anemic patients ity of patients.
Chapter 3: Blood conservation

In the absence of a speciically identiied reason there are ive parenteral iron compounds available. Iron
for the anemia, the common treatment approaches dextran is available in both high- and low-molecular
involve supplementation with iron and administration weight compounds. he high-molecular weight version
of erythropoietic stimulating agents (ESAs). Iron sup- has a poor safety record with a high risk of anaphylaxis.
plementation is needed frequently even with seemingly he advantage of this preparation is the ability to admin-
normal iron stores. Chronic inlammatory conditions ister a large dose of iron in one sitting. he other prepara-
limit the bodys ability to mobilize iron from storage tions, low-molecular weight iron dextran, iron sucrose,
sites. his functional iron deiciency can be overcome by ferric gluconate, and ferumoxytol have at least a 10-fold
pharmacologic iron administration. Figure 3.2 shows a lower risk of serious allergic reactions. hese prepara-
targeted preoperative anemia treatment protocol. tions are generally well tolerated but require multiple
Oral iron is poorly tolerated in most patients. dosings due to the more rapid iron release and lower dose
herefore, while it seems simple to administer PO iron, that can be administered in one sitting.6
the limited absorption from the GI tract as well as the ESAs are commonly used to prepare anemic ortho-
common occurrence of signiicant side efects limits the pedic patients for major surgery. More recently it has
efectiveness of this treatment. Intravenous iron supp- been appreciated that these agents increase the periop-
lementation is much more efective, but requires consid- erative risk of deep venous thrombosis (DVT). he FDA
erable additional resources to administer safely. Currently published an advisory warning against the use of these
drugs in surgical patients who cannot receive pharma-
If hemoglobin is less than 10 gm/dI, then obtain a standard anemia
work-up
cologic DVT prophylaxis.7 Most spine patients are not
candidates for pharmacologic DVT prophylaxis due to
Iron and TIBC, Tsat
Ferritin the risk of postoperative epidural hematoma formation.
Reticulocyte count herefore, the use of ESA for the preparation of anemic
Vitamin B12, if borderline lownormal then MMA
Folic acid, if borderline lownormal then RBC folate+/ homocysteine spine patients needs to be viewed with extreme caution.
RBC smear he safest overall preoperative preparatory regimen
Figure 3.1 Preoperative anemia evaluation. for spine surgery is intravenous iron supplementation.

Patient seen at least 4 weeks before noncardiac surgery


Hemoglobin at surgical office
If Hb < 12 g/dl anemia panel requested (iron and TIBC,
ferritin, RBC folate,vitamin B12, recticulocyte count).

Macrocytic anemia (MCV >100) Hb 1012 g/dl Iron deficiency anemia


or
Microcytic anemia (MCV <80)

FeSO4+ 325 mg POTID x 6 months


Outpt: Fe sucrose (Venofer) 200 mg (250 ml NS 0.9%) ~1h qod.
Normocytic anemia Inpatient Fe gluconate(Ferrlecit) 125 mg IV daily over 1 h
(MCV 80-100) Total dose IV 1 g
B12 or folate deficiency Hb electrophoresis
Referral to hematology Consider GI w/ucolonoscopy +/EGD

Patient undergoing TKA or THA


Folic acid 1 mg PO daily Jehovah's Witnesses Exclusion Criteria:
Cyanocobalamin 1000 g Recent blood donors
IM daily x 5 days Hb < 10 g/dl
Recent GI bleed ( 3 months)
Uncontrolled HTN (> 180/100)
EPO (ProcritEpoetin alpha) 600 Seizure disorders
units/kg2 weekly (40,000 units Blood dyscrasias
weekly) on days 21, 14, 7, and 0 Known history of VTE
before surgery along with PO iron Active malignancy not on chemotherapy
Contraindications for pharmacologic VTE
prophylaxis

Obtain labs on day of EPO injection


(Hb and reticulocyte count)
If Hb > 12 stop EPO
45
Figure 3.2 Targeted preoperative anemia treatment protocol.
Section 1: General considerations

In 714 days, irons stores can be replenished such that thrombosis in those receiving the drug.12 herefore
efective erythropoiesis can correct preoperative ane- there is no current indication for this drug unless all
mia. Since most spine procedures are elective in nature, reasonable means of controlling hemorrhage have
many patients are candidates for this treatment. been exhausted, and then this is at the risk of poten-
tially devastating arterial thrombotic events.
Antifibrinolytic treatments
-Aminocaproic acid (Eaca) and tranexamic acid Preoperative autodonation
(TXA) are two lysine analogs that are inexpensive and Preoperative autodonation (PAD) enjoyed great inter-
safe agents to decrease perioperative blood loss in spine est in the late 1980s and early 1990s. Currently, it is a
surgery. Aprotinin is no longer available ater reports technique that has a much more limited beneit. While
of excessive renal failure in cardiac surgery8 and will the emotional appeal of storing blood for the patients
not be further discussed. Antiibrinolytic agents act own use is considerable, many problems exist. First,
to mitigate the increased ibrinolysis frequently seen without supplemental agents, patients will not replace
during surgery. Both agents prevent the conversion of the donated red cells adequately. he result is that the
plasminogen to plasmin and also directly inhibit plas- patient will likely become anemic and therefore would
mins ability to degrade ibrinogen. be more likely to need transfusions. Transfusing back
Both drugs are administered intravenously as a their donated red cells will expose the patient to a risk
continuous infusion during surgery due to their rapid of clerical error. Stored red cells, whether autologous
renal excretion. TXA is 610 times more potent than or not, do not transport oxygen normally. PAD is
Eaca and is administered in a correspondingly lower not free, and can be inconvenient for the patient. he
dose. Typical protocols are:9 platelets and plasma are discarded and usually about
Eaca loading dose 100150 mg/kg, infusion 1015 half the units are discarded. As the safety of allogeneic
mg/kg/h. blood has increased, the utility of PAD has decreased
TXA loading dose 10 mg/kg, infusion 1 mg/kg/h. considerably.

Antiibrinolytic treatment has been studied in both Acute normovolemic hemodilution


pediatric and adult spine surgery.10 Both have dem-
Acute normovolemic hemodilution (ANH) is in prin-
onstrated eicacy in decreasing intraoperative, post-
ciple a very simple and elegant approach to red cell
operative, and transfusion requirements in complex
conservation as well as sequestration of platelets and
spine surgery. Excessive thrombosis has not been a
plasma. Before surgery, blood is drained into blood
major inding in most trials to date. herefore the use
bags containing a citrate anticoagulant while asanguin-
of lysine analogs represents a safe, low-cost, easy means
ous luid is administered to maintain circulating intra-
of lowering perioperative blood loss and transfusion in
vascular volume. As a result, during surgery the patient
spine surgery.
will lose blood during surgery that contains fewer red
cells. When the bleeding has stopped, the stored whole
Recombinant activated factor VII blood is re-administered. his whole blood contains
he of-label use of recombinant activated factor VII to red cells, platelets, and plasma. Since the blood is
control bleeding in a wide variety of settings (trauma, otherwise not processed and is stored in the operat-
intracranial hemorrhage, surgery, bleeding in the set- ing room in immediate proximity to the patient, issues
ting of anticoagulant therapy) is quite common. his of clerical error and storage problems are eliminated.
drug can activate the coagulation system directly, Cost is minimal, although the collection of the blood
bypassing many of the early steps in the clotting cas- needs good vascular access and some time. he key
cade. Anecdotal use and small series describe success- questions are how low to take the hemoglobin and how
ful control of bleeding when other usual methods have much blood can be saved. he more aggressively you
failed. However, its use in spine surgery is largely unex- withdraw blood, the more likely you are to be success-
plored. Only one phase IIa study has been published on ful, but at the risk of exposing a patient to lower red cell
safety in spine surgery.11 his trial reported nonsignii- levels than they may be able to tolerate.13
cant diferences in blood loss and transfusion volume. As with other blood-conservation strategies, ANH
46 A recent analysis of the of-label uses of this agent high- only works well if the patient has a normal starting
lights the small but signiicant increased risk of arterial hemoglobin. Withdrawal of anemic blood results
Chapter 3: Blood conservation

in storage of relatively fewer red cells and leaves the

coagulant
patient with a low hemoglobin level. his will result in

Anti-
the need to transfuse the stored blood relatively soon,
largely negating any beneit of the efort to store the
blood. Since blood loss in spine surgery is rarely in
sudden, large volumes, ANH can be an efective tech-
nique to minimize the spine surgery patients exposure Surgical site

Re-infusion
to allogeneic blood.

solution

bag
Saline
Collection
reservoir
Intraoperative red cell salvage
he mainstay of most blood conservation programs Re-infusion
in the use of intraoperative cell salvage, commonly
referred to as cell savers. A cell saver is a system con-
Waste
sisting of a means to collect shed intraoperative blood, bag
Centri-
store it until enough has been collected to process, and fuge
then concentrate and wash the salvaged material so that
a clean, safe product can be returned to the patient. Figure 3.3 Schematic of a cell saver system. From Noblood.org
Initially the shed blood is removed with special
surgical suction devices. he vacuum tubing has a he eiciency of cell salvage is limited, especially
channel to immediately mix the shed blood with an during spine surgery. If surgical sponges are used
anticoagulant, generally either heparin or citrate. he to dry the ield, these should be soaked in saline and
vacuum level must be regulated to lessen the phys- the blood-contaminated saline suctioned into the cell
ical trauma and resultant hemolysis. his frequently saver reservoir. Also the suctioning should be done to
results in objections from the surgical team about minimize air entrapment. Despite the most meticulous
the efectiveness of the suction systems. Suctioning is eforts at cell salvage, typically only 6070% of the lost
also best done in a manner that removes the liquid blood can be returned. Operations with large amounts
from pools with less entrainment of air, also to lessen of blood loss will ultimately need allogeneic red cells to
the resultant red cell trauma. he recovered blood is prevent anemia, even with meticulous cell salvage.
iltered upon entry to the storage canister to remove Cell savers should not be used when material that
large contaminants. he shed blood is stored in the should not be administered systemically is given in the
reservoirs until suicient material is available to pro- wound. Cell savers should not be used when the wound
cess in the centrifuge. he storage time should be is irrigated with antibiotics that are not supposed to be
monitored subject to reasonable times set forth by systemically administered. During these times, discard
blood banking policies. suction should be used. Also, cells savers should not be
he centrifuge bowls are speciically designed to used when topical hemostatic agents have been used. If
allow the heaviest elements, the red cells to collect in these materials are administered systemically through
the bottom of the bowl. Shed blood is pumped into the the cells saver, disseminated intravascular coagulation
spinning bowl until a suicient volume of packed red can result.
cells forms and is usually detected by a photo eye on the Cell salvage in controversial in several areas, the
device. Once the bowl is full, saline is pumped through most commonly sited being cancer surgeries and
the red cell pack to wash out residual debris and pro- bacterial infections.14 Cell salvage in cancer surgery
teins so that the resultant product contains concen- exposes the patient to the possibility of returning via-
trated red cells suspended in saline. See Fig. 3.3. ble cancer cells to the circulation from the re-admin-
Typically for spine surgery, a small or medium- istration of salvaged blood. here is little proof that
sized bowl is chosen so that the blood can be processed these transfused cancer cells will inevitably produce
at reasonable time intervals. Fill and wash protocols hematogenous spread of the cancer. Furthermore,
are chosen that typically ill the bowl slowly so that there is some suggestion that transfusion of allogeneic
less material is lost due to too rapid illing, and slow blood may worsen the prognosis in cancer patients.
washing with large volumes of wash saline is chosen to herefore some centers have decided that it is reasona- 47
remove as much bone debris and fat as reasonable. ble to administered salvaged intraoperative blood from
Section 1: General considerations

cancer surgeries, particularly if techniques are used to with unclear end points. Laboratory testing such as
decrease the burden of cancer cells. he two techniques PT/INR, PTT, ibrinogen, and platelet counts can be
are the use of leukodepletion ilters to remove a burden helpful but frequently involves considerable delay.
of the cancer cells or irradiation of the salvaged blood Quantitative measurement of intraoperative coagula-
prior to re-infusion. tion can provide a better guide as to the type of coagula-
Similar arguments can be made for the use of cell tion disorder and the required treatment. Techniques
savers in operations with bacterial contamination. of viscoelastic monitoring of coagulation can be done
he physical process of aggressive cell washing as well in the operating room and provide more deined diag-
as the use of leukodepletion ilters may allow for the nosis of the clotting disorder.
re-infusing of all but the most contaminated of shed here are two main technologies for measuring
blood. Most of these patients are receiving systemic intraoperative clotting: the Sonoclot (Sienco, Inc.) and
antibiotics that can also lessen the chance of dissemi- the hromboelastogram TEG (Haemoscope Corp.)
nating the infectious elements. Since allogeneic blood or a similar technique called ROTEM (Rotem, Inc.).
can be immunosuppressive in its own right, lessening Both techniques follow the in vitro rheologic changes
the patients exposure by using cell savers, even in con- in blood during the coagulation process. Characteristic
troversial areas, may improve the patients outcome. changes can be seen that can diagnose clotting factor
A inal use of the cell saver is to acutely fractionate depletion, hypoibrinogenemia, thrombocytopenia,
the patients blood into its elements in the OR. Some cell ibrinolysis, and anticoagulant administration. Speciic
savers have protocols that allow for the collection of red therapy can be administered based on the characteris-
cells, plasma, and platelets. Depending on the anticipated tic indings.
need in the surgery, for example the need for platelets hese methods are all moderately complex point of
and plasma, the patients blood can either be collected care testing systems. his requires an infrastructure of
and centrifuged or directly drawn into the centrifuge qualiied personnel, and comprehensive quality con-
and the desired elements separated. his process is more trol systems for test validation. Most clinical literature
complicated than ANH, but allows the team to collect exists in the cardiac surgery arena.17 here is compara-
the elements most needed for the speciic surgery. tively little literature to support the use of these tech-
niques in spine surgery.18 Viscoelastic measurement
Postoperative cell salvage of coagulation is a potentially useful but is currently a
Some surgeons leave drains in place ater certain sur- poorly validated point of care testing method for spine
geries. It is very tempting to re-administer this wound surgery.
drainage as a means of supplementing the patients
red cells. Unfortunately, there are several issues with Quality management
administration of wound drainage. his material has Blood banks adhere to strict procedures to ensure the
a hematocrit typically less than 15%. herefore, a large best outcomes from transfusions. Many of the blood
volume of wound drainage will need to be adminis- conservation techniques in spine surgery interface
tered before any signiicant amount of blood is given closely with the blood banking system. herefore,
back to the patient. Direct so-called lip and drip type quality assurance mechanisms frequently derived from
systems administer this luid to the patient without blood bank techniques should be used to monitor blood
any processing.15 he wound drainage contains a large conservation methods. Suggestions include monitor-
number of inlammatory mediators that should not ing the quality of cell saver products and watching for
be administered systemically in an ideal world. here complications from other techniques. Transfusions
are commercial systems that can collect and wash this have signiicant risks and, likewise, haphazard use of
material prior to administration. Whether it is cost blood conservation may be harmful to patients.
efective given the low yields needs to be considered.
Summary
Point of care testing Spine surgery patients frequently need transfusions.
he management of intraoperative coagulation disor- Efective blood conservation starts by making sure that
ders during spine surgery can have a large impact on most patients are not anemic when they start surgery.
48 control of bleeding and the quantity of blood products he risk of DVT with preoperative administration
utilized. Frequently empirical methods are employed of ESA needs to be considered. ANH can be used for
Chapter 3: Blood conservation

patients who can tolerate the acute volume shits and Research and Education Foundation. he risk
the forced anemia that is incumbent for appropriate associated with aprotinin in cardiac surgery. N Engl J
use. Antiibrinolytic agents can be used safely for many Med 2006 Jan 26; 354(4): 35365.
patients. Intraoperative cell salvage is a cornerstone for 9. Eubanks JD. Antiibrinolytics in major orthopaedic
most blood conservation programs. surgery. J Am Acad Orthop Surg 2010 Mar; 18(3): 1328.
10. Henry DA, Carless PA, Moxey AJ, et al. Anti-ibrinolytic
References use for minimising perioperative allogeneic blood
1. Centers for Disease Control and Prevention (CDC). transfusion. Cochrane Database Syst Rev 2007 Oct 17;
HIV transmission through transfusion Missouri and (4): CD001886.
Colorado, 2008. MMWR Morb Mortal Wkly Rep 2010 11. Sachs B, Delacy D, Green J, et al. Recombinant activated
Oct 22; 59(41): 13359. factor VII in spinal surgery: a multicenter, randomized,
2. Koch CG, Li L, Sessler DI, et al. Duration of red-cell double-blind, placebo-controlled, dose-escalation trial.
storage and complications ater cardiac surgery. N Engl J Spine (Phila Pa 1976) 2007 Oct 1; 32(21): 228593.
Med 2008 Mar 20; 358(12): 122939. 12. Levi M, Levy JH, Andersen HF, Trulof D. Safety of
3. Shander A, Hofmann A, Ozawa S, et al. Activity-based recombinant activated factor VII in randomized clinical
costs of blood transfusions in surgical patients at four trials. N Engl J Med 2010 Nov 4; 363(19): 1791800.
hospitals. Transfusion 2010 Apr; 50(4): 75365. 13. Shander A, Perelman S. he long and winding road of
4. Lenoir B, Merckx P, Paugam-Burtz C, et al. Individual acute normovolemic hemodilution. Transfusion 2006
probability of allogeneic erythrocyte transfusion Jul; 46(7): 10759.
in elective spine surgery: the predictive model of 14. Waters JH. Indications and contraindications of cell
transfusion in spine surgery. Anesthesiology 2009 May; salvage. Transfusion 2004 Dec; 44(12 Suppl): 40S-4S.
110(5): 105060. 15. Hansen E, Hansen M. Reasons against the retransfusion
5. Spahn DR. Anemia and patient blood management of unwashed wound blood. Transfusion 2004 Dec; 44(12
in hip and knee surgery: a systematic review of the Suppl): 45S53S.
literature. Anesthesiology. 2010 Aug; 113(2): 48295. 16. Shander A, Goodnough, LT. Update on transfusion
6. Auerbach M, Goodnough LT, Picard D, Maniatis A. he medicine. Pharmacotherapy 2007 Sep; 27(9 Pt 2):
role of intravenous iron in anemia management and 57S68S.
transfusion avoidance. Transfusion 2008 May; 48(5): 17. Shore-Lesserson L, Manspeizer HE, DePerio M, et al.
9881000. hromboelastography-guided transfusion algorithm
7. Shander A, Spence RK, Auerbach M. Can intravenous reduces transfusions in complex cardiac surgery. Anesth
iron therapy meet the unmet needs created by the new Analg 1999; 88: 31219.
restrictions on erythropoietic stimulating agents? 18. Horlocker TT, Nuttall GA, Dekutoski MB, Bryant SC.
Transfusion 2010 Mar; 50(3): 71932. he accuracy of coagulation tests during spinal fusion
8. Mangano DT, Tudor IC, Dietzel C; Multicenter Study and instrumentation. Anesth Analg. 2001 Jul; 93(1):
of Perioperative Ischemia Research Group; Ischemia 338.

49
Section 1 General considerations
Chapter
Airway management in spine surgery

4 Basem Abdelmalak and D. John Doyle

discussed here also apply to clinical airway management


Key points
in general, regardless of the surgical procedure. In partic-
Flexible iberoptic bronchoscope is the device ular, the American Society of Anesthesiologists Diicult
of choice for many clinicians for unstable Airway Algorithm, briely outlined in this chapter and
neck or diicult airway management in spine shown in Fig. 4.1 should be an important starting point
surgery, because this technique is associated for all aspects of clinical airway management.
with minimal movement of the cervical spine To a large extent, the airway management tech-
in comparison with other methods. nique employed for anesthesia for spine surgery will
While many clinicians fear untoward depend on clinical circumstances as well as on the air-
consequences from cervical spine movement way management skills of the anesthesiologist. hree
during intubation in the setting of an unstable options are employed: (1) general endotracheal anes-
neck, these movements are typically small, thesia; (2) general anesthesia using a supraglottic air-
and the clinical implications of such a degree way device such as a laryngeal mask airway; and (3)
of cervical spine movement are not well neuraxial anesthesia, for instance, spinal anesthesia.
established. he irst option is undoubtedly the most preferred.
When anesthesiologists in the United States
and Canada were surveyed regarding their Tracheal intubation in patients
preference in airway management of patients undergoing spine surgery
with cervical spine disease, awake lexible Most patients undergoing spine surgery have their
iberoptic intubation was the irst choice. airway managed via tracheal intubation, especially
Caution should be exercised with the use of if they are undergoing surgery in the prone position.
succinylcholine in patients with trauma to Under ordinary circumstances, tracheal intubation is
the spine, especially those with neurologic straightforward; therefore, ordinary laryngoscopic
symptoms from spinal cord injury. techniques work well. However, patients are occa-
Edema and hematoma formation may sionally encountered who are potentially diicult
sometimes make extubation risky following to intubate, and these patients are usually managed
cervical spine surgery. using video laryngoscopy or lexible iberoptic intu-
Every efort should be made to prevent bation. hree key decisions must be made in such
inadvertent extubation during spine surgery, cases. he irst concerns whether intubation should
especially in the prone position. Inserting a be carried out awake or following the induction of
supraglottic airway can be a quick temporizing general anesthesia (Fig. 4.1). he second concerns the
measure in these situations. tools to employ if diiculty is encountered with ven-
tilation or with intubation (Table 4.1). Finally, there
is the question of how to manage a patient with an
Introduction unstable cervical spine a vital issue that will be dis-
his chapter deals with clinical airway management cussed later in the chapter.
in the context of spine surgery. However, although he trachea of a patient in the supine position is
the focus is on spine surgery, many of the principles oten intubated with an ordinary polyvinylchloride
50
Anesthesia for Spine Surgery, ed. Ehab Farag. Published by Cambridge University Press. Cambridge University Press 2012.
AMERICAN SOCIETY
OF ANESTHESIOLOGISTS

DIFFICULT AIRWAY ALGORITHM


1. Assess the likelihood and clinical impact of basic management problems:
A. Difficult Ventilation
B. Difficult Intubation
C. Difficulty with Patient Cooperation or Consent
D. Difficult Tracheostomy
2. Actively pursue opportunities to deliver supplemental oxygen throughout the process of difficult airway management

3. Consider the relative merits and feasibility of basic management choices:


Intubation Attempts After Induction of
A. Awake Intubation vs.
General Anesthesia

B. Non-Invasive Technique for Initial Invasive Technique for Initial


vs.
Approach to Intubation Approach to Intubation

C. Preservation of Spontaneous Ventilation vs. Ablation of Spontaneous Ventilation

4. Develop primary and alternative strategies:


A. B. INTUBATION ATTEMPTS AFTER
AWAKE INTUBATION INDUCTION OF GENERAL ANESTHESIA

Airway Approached by Invasive Initial Intubation Initial Intubation


Non invasive Intubation Airway Access(b)* Attempts Successful* Attempts UNSUCCESSFUL

FROM THIS POINT


Succeed* FAIL ONWARDS CONSIDER:
1. Calling for Help
2. Returning to Spontaneous
Cancel Consider Feasibility Invasive Ventilation
Case of Other Options(a) Airway Access(b)* 3. Awakening the Patient

FACE MASK VENTILATION ADEQUATE FACE MASK VENTILATION NOT ADEQUATE

CONSIDER / ATTEMPT LMA

LMA ADEQUATE* LMA NOT ADEQUATE


OR NOT FEASIBLE
NON-EMERGENCY PATHWAY EMERGENCY PATHWAY
Ventilation Adequate, Intubation Unsuccessful Ventilation not Adequate,
Intubation Unsuccessful
IF BOTH
Alternative Approaches FACE MASK
Call for Help
to Intubation(c) AND LMA
VENTILATION
BECOME Emergency Non invasive Airway Ventilation(e)
INADEQUATE
Successful FAIL after
Intubation* Multiple Attempts Successful Ventilation* FAIL

Emergency
Invasive Consider Feasibility Awaken
Invasive Airway
Airway Access(b)* of Other Options(a) Patient(d)
Access(b)*

* Confirm ventilation, tracheal intubation, or LMA placement with exhaled CO2


a. Other options include (but are not limited to): surgery utilizing face c. Alternative non-invasive approaches to difficult intubation include
mask or LMA anesthesia, local anesthesia infiltration or regional (but are not limited to): use of different laryngoscope blades, LMA
nerve blockade. Pursuit of these options usually implies that mask as an intubation conduit (with or without fiberoptic guidance),
ventilation will not be problematic. Therefore, these options may be fiberoptic intubation, intubating stylet or tube changer, light wand,
of limited value if this step in the algorithm has been reached via retrograde intubation, and blind oral or nasal intubation.
the Emergency Pathway. d. Consider re-preparation of the patient for awake intubation or
canceling surgery.
b. Invasive airway access includes surgical or percutaneous e. Options for emergency non-invasive airway ventilation include (but
tracheostomy or cricothyrotomy. are not limited to): rigid bronchoscope, esophageal-tracheal combitube
ventilation, or transtracheal jet ventilation.
Figure 4.1 Synopsis of the 2003 ASA diicult airway algorithm. It is expected that future editions of the algorithm will place special
emphasis on the use of video laryngoscopy for situations in which direct laryngoscopy produces an unsatisfactory view of the glottic
structures.
Section 1: General considerations

Table 4.1 Techniques for diicult airway management


This table, from the 2003 ASA diicult airway algorithm, lists some commonly cited
techniques. Since the time of publication, the use of video laryngoscopy (e.g., GlideScope,
McGrath video laryngoscope, Storz video laryngoscope, Pentax AWS) has become
commonplace and therefore belongs to the left-hand column. Naturally, the tools used in
any particular situation will depend on the speciic circumstances.

Techniques for diicult intubation Techniques for diicult ventilation


Alternative laryngoscope blades Esophageal tracheal Combitube
Awake intubation Intratracheal jet stylet
Blind intubation (oral or nasal) Invasive airway access
Fiberoptic intubation Laryngeal mask airway (LMA)
Intubating stylet or tube changer Oral and nasopharyngeal airways
Intubation via LMA Rigid ventilating bronchoscope
Invasive airway access Transtracheal jet ventilation
Light wand Two-person mask ventilation
Retrograde intubation

(PVC) endotracheal tube. hree precautions are Prediction of intubation difficulty:


required here: the tube should be irmly secured with
tape or other means; the tube tip should not be too intubation difficulty scale
close to the carina (or worse, positioned endobronchi- Before attempting tracheal intubation it is very helpful
ally); and the tube should be situated so that kinking is for the anesthesiologist to have a means of predicting
unlikely to occur with head movement. Many anesthe- which patients may have airways diicult to intubate
siologists employ wire-reinforced endotracheal tubes with direct laryngoscopy. While a number of studies
for spine cases, especially for patients in the prone and reviews on the topic are available,58 most clini-
position. he principal advantage of such tubes is that cians will be satisied with the 11-point airway assess-
they are unlikely to kink. Finally, it is important that ment tool included in the 2003 ASA Diicult Airway
cuf pressures in the endotracheal tube be maintained Algorithm. his tool is summarized in Table 4.2. In
under 25 cmH2O to avoid damage to the tracheal addition, ater completion of tracheal intubation, it
mucosa. can sometimes be helpful for the anesthesiologist to
describe any diiculties that might have occurred.
Role of the ASA Difficult Airway Here, the Intubation Diiculty Scale (IDS) introduced
by Adnet and colleagues9 can be useful. It is a numer-
Algorithm ical score indicating overall intubation diiculty based
he American Society of Anesthesiologists (ASA) has on seven descriptors associated with diiculty of
issued guidelines for management of the diicult air- intubation: (1) number of supplementary intubation
way. he guidelines began with an algorithm origin- attempts; (2) number of supplementary operators; (3)
ally published in 1993,1 followed by a revision in 2003.2 alternative techniques used; (4) laryngoscopic grade;
hese guidelines, as well as similar guidelines from (5) subjective liting force; (6) the use of external laryn-
other organizations,3,4 ofer considerable advice to the geal manipulation; and (7) the characteristics of the
clinician facing potentially diicult airway challenges. vocal cords.
he advice emphasizes, for example: (1) the import-
ance of performing an airway evaluation prior to
inducing anesthesia; (2) the importance of providing Laryngoscopes
oxygen at every opportunity; (3) the potential value of Most intubations for spine surgery are performed
awake intubation; and (4) the value of supraglottic air- using traditional Macintosh and Miller laryngoscopes.
way devices such as the laryngeal mask airway (LMA) When the view at laryngoscopy is suboptimal, the use
as a possible airway rescue maneuver, if failure should of introducers such as the Eschmann stylet (gum elas-
52 occur. Figure 4.1 summarizes the 2003 ASA Diicult tic bougie) can sometimes be very helpful. It is used as
Airway Algorithm. follows. When a poor laryngoscopic view of the glottic
Chapter 4: Airway management in spine surgery

Table 4.2 Components of the preoperative airway physical examination as recommended in the ASA Diicult Airway Algorithm (2003
edition)
In ordinary clinical practice, special emphasis is usually placed on the visibility of the oropharyngeal structures with the tongue protruded
when the patient is in the sitting position (Mallampati classiication).

Airway examination component Nonreassuring indings


1. Length of upper incisors Relatively long
2. Relation of maxillary and mandibular incisors during Prominent overbite (maxillary incisors anterior to mandibular
normal jaw closure incisors)
3. Relation of maxillary and mandibular incisors during voluntary Patient cannot bring mandibular incisors anterior to (in front of )
protrusion of mandible maxillary incisors
4. Interincisor distance Under 3 cm
5. Visibility of uvula Not visible when tongue is protruded with patient in sitting
position (e.g., Mallampati class greater than II)
6. Shape of palate Highly arched or very narrow
7. Compliance of mandibular space Stif, indurated, occupied by mass, or nonresilient
8. Thyromental distance Less than 3 ordinary inger breadths
9. Length of neck Short
10. Thickness of neck Thick
11. Range of motion of head and neck Patient cannot touch tip of chin to chest or cannot extend neck

Figure 4.2 Photograph showing a tracheal tube with a preloaded


introducer containing a Coud tip. Intended for situations where
the laryngoscopic view is suboptimal, the upturned distal Coud tip
is placed under the epiglottis or, if visible, above the interarytenoid
notch, followed by advancement of the tracheal tube into the
trachea.
Figure 4.3 The GlideScope video laryngoscope utilizes a color
CMOS video camera and LED light source embedded into a plastic
structures is evident, the intubator places the introducer laryngoscope blade. The standard (adult) blade is 14.5 mm at its
maximum width, and bends 60 at the midline. This coniguration
into the patients mouth and gently advances it through provides a view that is frequently superior to that obtained by
the glottic opening (in the case of a grade II view) or direct laryngoscopy. The video image is displayed on a liquid crystal
anteriorly under the epiglottis (in the case of a grade display (LCD) monitor, and can also be recorded electronically.
An anti-fog mechanism helps ensure that a high-quality image is
III view). Clicks resulting from the introducer passing obtained. In addition to the standard blade, a mid-sized (pediatric)
over the tracheal rings help conirm proper placement blade and a neonatal blade are also available.
of the introducer. With the introducer held steady, one
then railroads a tracheal tube over the introducer into the Storz video laryngoscope, and the Pentax-AWS
the glottis.10 Some clinicians preload a tracheal tube have proved to be particularly valuable, especially in
onto the introducer, as shown in Fig. 4.2. patients with an anterior larynx or in patients with
Special devices such as the McCoy laryngoscope and cervical spine immobilization. Among the available
the Bullard laryngoscope are also popular in some cent- video laryngoscopes, the GlideScope (Fig. 4.3) has the
ers. In recent years, however, video laryngoscopes such largest market share. Figure 4.4 shows a typical view of 53
as the GlideScope, the McGrath video laryngoscope, the glottis during intubation with the GlideScope.
Section 1: General considerations

Figure 4.4 Close-up views from the


GlideScope, as the endotracheal tube
(ETT) passes through the vocal cords.
Note that during ETT placement the
tube tip often tends to hit against the
anterior tracheal wall. This problem is
easily solved by pulling back the stylet by
about 3 cm and then advancing the ETT.
Sometimes it also helps to rotate the ETT
180 to direct the ETT tip more posteriorly
(once the stylet has been removed).

Tracheal intubation in patients with safest method of airway management based on move-
ment criteria was the lexible iberoptic technique.11
cervical spine instability In another study the Bullard laryngoscope caused
In patients with conirmed or suspected instability less cervical spine extension than Macintosh and Miller
of the cervical spine, special precautions are needed laryngoscopes and resulted in a better view when stud-
to minimize the likelihood that laryngoscopy and ied in healthy patients.12 he Bullard laryngoscope also
intubation might result in neurologic injury. Airway performed better than the Macintosh laryngoscope,
management in such patients presents an anesthetic resulting in less neck movement when in the setting of
challenge; and, for decades, clinicians have debated the in-line stabilization.13
safest technique and/or device for securing the airways In another study in healthy patients under general
of these patients. hey have usually focused on the anesthesia, investigators compared the Pentax-AWS,
association between a particular airway management the Macintosh laryngoscope, and the McCoy laryngo-
technique and the degree of neck movement at a given scope with respect to movement of the upper cervical
cervical spine location. spine during intubation. he Pentax-AWS produced
In a randomized, controlled, crossover study in less movement of the upper cervical spine than did the
cadavers with a posteriorly destabilized third cervical Macintosh or McCoy laryngoscopes.14
(C3) vertebra, investigators sought to determine the When mask ventilation was compared to use of a
degree of cervical spine motion for six airway manage- Macintosh #3 laryngoscope, GlideScope, and use of a
ment techniques when manual in-line stabilization was lighted stylet with respect to cervical spine movement
applied. hey compared face mask ventilation, direct at the occiputC1 junction, the C12 junction, the C25
laryngoscopic orotracheal intubation, lexible iberop- region, and C5 thoracic region, motion during bag-mask
tic nasal intubation, use of the Combitube (Kendall- ventilation and lighted stylet intubation was 82% and
Sheridan, Neustadt, Germany), use of the intubating 52% less, respectively, at the four regions of interest than
laryngeal mask in conjunction with lexible iberscope- during Macintosh laryngoscopy. In addition, cervical
guided tracheal intubation, and laryngeal mask airway spine motion was reduced 50% at the C25 region with
54
(LMA) insertion. he investigators concluded that the the GlideScope as opposed to direct laryngoscopy but
Chapter 4: Airway management in spine surgery

Figure 4.5 Illustration of in-line


stabilization (not in-line traction!)
during laryngoscopy and intubation in a
patient with a suspected cervical spine
injury. Note also (where appropriate)
the application of cricoid pressure to
reduce the chance of aspiration. (Image
from: http://www.pharmacology2000.
com/822_1/inline.jpg and Stene JD,
Anesthesia for the critically ill trauma
patient. In: Siegel JH, ed. Trauma:
Emergency Surgery and Critical Care. New
York, Churchill Livingstone, 1987.)

was unchanged in the other three regions.15 Interestingly, cord injury, those experiencing muscle weakness, or
even cricothyrotomy resulted in a small, and clinically individuals who have experienced prolonged inactiv-
insigniicant, degree of movement across an unstable ity and/or immobility. In such cases the administration
cervical spine injury in a cadaver model.16 of succinylcholine increases the risk of hyperkalemia,
Finally, it should be noted that while many clinicians potentially severe enough to cause cardiac arrest.22,23
fear untoward consequences from cervical spine move- Patients are susceptible to succinlycholine-induced
ment during intubation in the setting of an unstable neck, hyperkalemia for up to two months following massive
cervical spine movements are typically small and their trauma or until damaged tissues heal.24 he associated
clinical implications have not been well established.17,18 hyperkalemia is believed to stem from the spread of
In cases where neck movement is deliberately lim- extrajunctional receptors across the muscle membrane
ited via the use of a neck immobilization collar (e.g., upon exposure to succinylcholine. hese receptors
Philadelphia collar), airway management becomes enter a state of prolonged depolarization characterized
even more challenging, as such collars constrict the by massive potassium release.23
mouth opening, making direct laryngoscopy much
more diicult.19 In some cases the anterior part of the Flexible fiberoptic intubation and
collar is removed for laryngoscopy (Fig. 4.5). Although awake intubation
many airway devices have been used in such cases, the Whether as a prelude to spine surgery or for other rea-
lexible iberoptic bronchoscope is the device of choice sons, the use of lexible iberoptic intubation for the
for many clinicians. More recently, newer-generation airway management of patients with cervical spine
video laryngoscopes have been used successfully. For pathology is favored because in contrast with other
example, the Pentax-AWS has been used successfully methods, this technique is associated with minimal
for awake nasotracheal intubation in patients using a movement of the cervical spine. In addition, many cli-
neck collar for cervical spine stabilization,20 while the nicians prefer iberoptic intubation for patients who
GlideScope provides a better glottic view than does have limited neck mobility (e.g., ankylosing spondyli-
Macintosh direct laryngoscopy in these patients.21 tis) even when there is no risk of spinal cord injury.25
While lexible iberoptic intubation can usually be
Use of succinylcholine in patients easily performed under complete general anesthesia,
undergoing spine surgery many clinicians prefer topical anesthesia, with the patient
Caution should be exercised with the use of succinyl- being only lightly sedated (awake iberoptic intuba-
choline in patients with trauma to the spine, especially tion). However, the decision to go either way (awake ver- 55
those with neurologic symptoms resulting from spinal sus asleep intubation) depends on the anesthesiologists
Section 1: General considerations

level of skill, the extent of the patients cooperation, and iberoptic intubation, the method appears to be less
the severity of the neck pathology. A key consideration afected by the presence of secretions or blood. (3) All
underlying the choice between awake versus asleep in attendance can see what is going on, while this is the
iberoptic intubation concerns the safety margin that case only with iberoptic intubation carts that carry a
an awake technique would allow. Speciically, if awake video screen option. (4) here are no special restric-
intubation is not successfully accomplished, the patient tions on the type of ETT that can be placed when using
should be able to maintain his or her own airway. the GlideScope, while this is not the case with iberop-
he awake technique maintains muscle tone that tic methods. (5) he GlideScope is much more rugged
is nearly normal, thus resembling the natural way of than a iberoptic bronchoscope, and is far less likely to
splinting the spine and preventing the extension of be damaged with use. (6) While it is well known that
damage to neural structure.26 Also, awake intubation advancing the ETT into the trachea over the iberoptic
allows for a postintubation neurologic check, per- bronchoscope may fail as a result of the ETT imping-
formed either immediately ater intubation, or even ing on the arytenoid cartilages, this is generally not a
ater the patient has been positioned for surgery before problem with the GlideScope.
induction of general anesthesia. Finally, during awake he above advantages notwithstanding, the use of
intubation, airway relexes are generally maintained to awake iberoptic intubation in a patient with cervical
a degree suicient to prevent pulmonary aspiration spine pathology remains steadfastly preferred because
an important consideration in a patient at high risk for it is gentle to the airway, is generally well tolerated, and
aspiration, such as a trauma patient having a full stom- does not require the application of force to obtain glot-
ach. It is not surprising, then, that anesthesiologists in tic exposure.
the United States27 and Canada28 who were surveyed
regarding their preference in airway management of Airway edema in spine cases
patients with cervical spine disease, favored awake Airway edema frequently accompanies prolonged
iberoptic intubation as their irst choice. surgery of any kind, especially when vigorous luid
It should be emphasized that awake intubation is resuscitation has been carried out, or when a patient
not synonymous with lexible iberoptic intubation, is undergoing surgery in the prone position. his
for awake intubation can be safely accomplished with edema may sometimes make extubation risky. Sagi
airway devices other than iberoptic. hese devices and colleagues identiied hematoma formation and
include, but are not limited to, direct laryngoscopy with pharyngeal edema as the main reasons for airway
Macintosh and Miller laryngoscopes, blind nasal intu- complications (2.4%) following anterior cervical spine
bation, the GlideScope, and the lighted stylet. Typically, surgery.36 Edema is considered an especial contribu-
the airway is anesthetized with gargled and atomized 4% tor to airway complications.37 More edema is observed
lidocaine. Superior laryngeal and transtracheal blocks to result from upper cervical spine surgery C24 than
are also occasionally employed. In addition, judicious from lower cervical spine surgery C5636,38 and to
sedation is usually administered. Midazolam, fentanyl, occur more oten in female than in male patients.36,39
remifentanil, ketamine, propofol, and clonidine have Other risk factors include operative time, adminis-
all been used in this setting. Recently, the use of dexme- tered crystalloid volume, large blood loss, and the
detomidine,29 a selective 2 agonist with sedative, anal- need for blood transfusion.36,40 Massive tongue swell-
gesic, amnestic,30 and antisialagogue properties,31 has ing has been reported as a contributor to airway com-
been reported. One key advantage of this agent is that promise.41 Perioperative steroids were reported to be
it maintains spontaneous respiration with minimal res- helpful in reducing tongue swelling. However, in a
piratory depression. Another advantage is that patients randomized trial, perioperative intravenous steroids
under dexmedetomidine sedation are generally easy were not found to decrease the risk of airway edema,
to arouse,32 a property that has been exploited during and they delayed postoperative extubation in anterior
awake iberoptic-assisted intubation.33 However, these cervical spine surgery.39 Many clinicians have used the
advantages may not hold under very large doses.34 leak test.40 (he presence of a leak around the endotra-
Doyle described the successful use of the GlideScope cheal tube when the cuf is delated suggests that the
in four cases of awake intubation for nonspinal sur- airway is not overly edematous.) However, the test is
gery.35 he following advantages are noteworthy: not without limitations, as the presence of a leak does
56 not guarantee smooth extubation and vice versa.
(1) he view is generally excellent. (2) In contrast to
Chapter 4: Airway management in spine surgery

Spine surgery requiring the use If both ventilation and intubation are diicult, rescue
ventilation via a supraglottic airway is warranted. his
of double-lumen tubes would be followed by reevaluation of the situation and
Many thoracic spine procedures require collapsing a deciding whether the procedure could be completed
lung to facilitate surgical exposure. Use of a double- with the use of such an airway, or whether re-intuba-
lumen tube (DLT) is oten preferred in such a setting. tion is still needed.42 In desperate cases, transtracheal
While this scenario usually does not mandate any spe- jet ventilation or a surgical airway might be necessary
cial considerations beyond the usual procedure for to regain airway access. In addition, the administration
placing a double-lumen tube, in patients with a poten- of intravenous anesthesia may be needed to ensure that
tially diicult airway, an awake intubation technique the patient remains unconscious.
is sometimes preferable. Although it is theoretically If the cervical spine is unstable, in-line stabilization
possible to insert a DLT in an awake patient, most cli- of the spine may help reduce the likelihood of spinal
nicians prefer to initially insert a single-lumen tube uti- cord injury. In this case, iberoptic intubation is usually
lizing an appropriate awake technique. his is followed recommended, although other techniques may also be
by induction of general anesthesia and exchange of acceptable.
the single-lumen tube with a DLT by means of a tube- In the prone position, the challenge is greater. his
exchange catheter. However, before attempting the is a more serious situation because ventilation and
latter step, one should determine the largest diameter intubation in the prone position are much more dii-
of exchange catheter that will it into the tube in place, cult than in the supine position. Possibly exacerbating
and the intended DLT, as a smaller-diameter exchange such diiculty is the edema that may have developed
catheter is prone to kink, creating diiculties when an in the airway.
attempt is made to railroad an endotracheal tube over he traditional teaching about the management of
it. It is helpful to presoak the DLT in warm water to patients who become extubated in the prone position
soten it. Many clinicians perform the exchange with is that they should be promptly lipped into the supine
the aid of direct laryngoscopy or video laryngoscopy. position and then reintubated in that position. For
(Further details are discussed in Chapter 11.) this reason, a gurney should be kept near any patient
Ater the procedure is completed, it is oten desir- undergoing anesthesia in the prone position so that the
able to reverse the above steps and exchange the DLT patient may be lipped if necessary. Moreover, an essen-
with a single-lumen ETT. his is also frequently done tial precaution is that the surgical wound must be fully
by means of an exchange catheter. However, the best covered in a sterile manner before the lip is executed.
time to perform such an exchange might not be imme- An alternate approach oten advocated is to irst
diately postoperatively as it might prove hazardous to establish a temporary airway through the use of a sup-
reintubate the trachea in a patient with a very swol- raglottic airway device such as the intubating laryngeal
len airway following extensive prone positioning. It mask airway. his provides time to choose between
is therefore generally wise to delay this until airway lipping the patient onto a nearby gurney as described
edema has subsided. In the meantime, the endobron- above, iberoptically intubating through the supraglot-
chial blue cuf of the DLT should be delated and the tic airway device in the prone position, or even complet-
DLT withdrawn approximately 2.5 cm. ing the case using only a supraglottic airway device.
Of course, the importance of preventing inadvert-
Management of accidental ent extubation in the irst place cannot be overempha-
extubation sized. Special care must be taken to irmly secure the
Inadvertent extubation during spine surgery can occur tube and to ensure that inadvertent pulling on adjacent
either in the prone or supine position. In the supine tubes such as oral gastric tubes or esophageal tempera-
position, in which stability of the cervical spine is not ture probes will not lead to extubation as well.
a concern, re-intubation is usually straightforward
unless the airway has previously been diicult. Mask Spine surgery under
ventilation with 100% oxygen will be the irst step, usu-
ally followed by immediate re-intubation with the same spinal anesthesia
device that proved helpful earlier. If ventilation proves Spinal anesthesia for spine surgery ofers several advan-
tages. It allows the patient to self-position in prone
57
diicult, intubation should be attempted immediately.
Section 1: General considerations

cases, which will likely reduce the chance of positioning shit the patient to the prone position. Since position-
injuries that occasionally occur when general anesthe- ing an anesthetized patient from the supine position to
sia is used. Moreover, it reduces intraoperative surgi- the prone position is fraught with potential problems
cal blood loss; improves perioperative hemodynamic (such as accidental extubation or injury to peripheral
stability ; reduces pain in the immediate postoperative nerves), some clinicians have advocated utilizing a
period, and therefore the need for analgesics; lowers the supraglottic airway placed in the prone position ater
incidence of postoperative nausea and vomiting; and the patient has self-positioned and general anesthesia
lowers the incidence of lower extremity thromboem- has been induced.
bolic complications. he cumulative efect of these Brimacombe and colleagues46 conducted a ret-
advantages is that patient satisfaction is enhanced and rospective audit of 245 patients, among whom the
discharge from hospital is expedited.43 ProSeal laryngeal mask airway (PLMA) was used in
At Cleveland Clinic Spine Institute, Tetzlaf and prone patients. he authors acknowledged that the use
colleagues44 studied 611 cases of elective lumbar of a classic laryngeal mask airway in prone patients is
spine procedures performed under spinal anesthesia. controversial but that the PLMA may be less problem-
hey found that among perioperative complications, atic, since it forms a better seal and provides access to
nausea and deep venous thrombosis occurred sig- the stomach. heir technique involves the following
niicantly more oten in patients who had undergone steps: (1) he patient adopts the prone position with
general anesthesia than in those who had undergone the head to the side and the table tilted laterally; (2) pre-
spinal anesthesia. hey also found that the use of plain oxygenation to end-tidal oxygen is >90%; (3) anesthe-
bupivacaine was associated with the lowest incidence sia is induced with midazolam/alfentanil/propofol;
of supplemental local anesthetic use intraoperatively (4) facemask ventilation is used; (5) a single attempt at
compared with hyperbaric bupivacaine or hyperbaric digital insertion is made, and if unsuccessful a single
tetracaine. he authors concluded that for lumbar attempt is made at laryngoscope-guided, gum elastic
spine surgery, spinal anesthesia is an efective alterna- bougie-guided insertion; (6) a gastric tube is inserted;
tive to general anesthesia and has a lower rate of minor (7) anesthesia is maintained with sevolurane-N20; (8)
complications. volume-controlled ventilation is maintained at 812
In a subsequent study at the same institution, a ml/kg; (9) emergence from anesthesia occurs in the
case-controlled analysis of 400 patients compared spi- supine position; and (10) PLMA is removed when the
nal and general anesthesia in lumbar laminectomy. patient is awake.
he authors found that anesthetic and operative times In their audit, correctable partial airway obstruc-
were longer for patients receiving a general anesthetic, tion occurred in three patients, but there was no hyp-
and also that these patients experienced more nausea oxia, hypercapnea, displacement, regurgitation, gastric
and a greater need for antiemetics and pain medica- insulation, or airway relex activation. he authors
tion.45 he authors also found that complication rates concluded that when used competently, PLMA is feas-
(for instance, the rate of urinary retention) were sig- ible for inducing and maintaining anesthesia when the
niicantly lower in spinal anesthesia patients. However, patient is in the prone position.
neither group experienced neural injuries, and in Sharma and colleagues47 conducted a prospective
patients receiving a spinal anesthetic, the incidence of audit on the use of the Laryngeal Mask Airway Supreme
spinal headache was lower. hus for patients undergo- (SLMA) in 205 consecutive patients undergoing ortho-
ing lumbar laminectomy, spinal anesthesia is not only pedic surgery in the prone position. Patients positioned
as safe and efective as general anesthesia but it also themselves in the prone position; aterwards, anesthe-
ofers several advantages. sia was induced and the SLMA was inserted. he vast
majority of patients received positive pressure ventila-
The utility of supraglottic airway tion (PPV). No failures of SLMA insertion or of main-
devices in spine surgery tenance of PPV occurred. he authors concluded that
Traditional airway management for patients under- the SLMA is useful for airway management in patients
going surgery under general anesthesia, and in the anesthetized in the prone position, and for subsequent
prone position, is to induce anesthesia in the supine airway management with PPV, with or without neu-
position, secure the airway with a cufed endotracheal romuscular block. Other studies support the use of
58
tube, securely tape the tube in place, and then gently supraglottic airway devices in the prone position.4851
Chapter 4: Airway management in spine surgery

We emphasize that all the above studies and case 11. Brimacombe J, Keller C, Kunzel KH, et al. Cervical
reports are nonrandomized studies; they either spine motion during airway management: a
assessed feasibility or described a technique, or both. cineluoroscopic study of the posteriorly destabilized
third cervical vertebrae in human cadavers. Anesth
For these reasons, they are subject to patients selection
Analg 2000; 91: 12748.
bias and do not prove superiority over other anesthetic
management strategies such as spinal anesthesia and/ 12. Hastings RH, Vigil AC, Hanna R, Yang BY, Sartoris DJ.
Cervical spine movement during laryngoscopy with
or general endotracheal anesthesia. hese reports sim- the Bullard, Macintosh, and Miller laryngoscopes.
ply state that supraglottic airways in prone spine sur- Anesthesiology 1995; 82: 85969.
gery can be done, which is useful information that can
13. Watts AD, Gelb AW, Bach DB, Pelz DM. Comparison
be used in case of emergency loss of the airway in the of the Bullard and Macintosh laryngoscopes for
prone position and/or if the anesthesiologist elects to endotracheal intubation of patients with a potential
use such a technique as the primary choice. Moreover, cervical spine injury. Anesthesiology 1997; 87: 133542.
these reports apply to procedures performed by users 14. Maruyama K, Yamada T, Kawakami R, et al. Upper
who are expert in LMA; the indings may therefore not cervical spine movement during intubation:
apply to infrequent LMA users. luoroscopic comparison of the AirWay Scope, McCoy
laryngoscope, and Macintosh laryngoscope. Br J
References Anaesth 2008; 100: 1204.
1. Practice guidelines for management of the diicult 15. Turkstra TP, Craen RA, Pelz DM, Gelb AW. Cervical
airway. A report by the American Society of spine motion: a luoroscopic comparison during
Anesthesiologists Task Force on Management of the intubation with lighted stylet, GlideScope, and
Diicult Airway. Anesthesiology 1993; 78: 597602. Macintosh laryngoscope. Anesth Analg 2005; 101:
2. Practice guidelines for management of the diicult 91015.
airway : An updated report by the American Society of 16. Gerling MC, Davis DP, Hamilton RS, et al. Efect of
Anesthesiologists Task Force on Management of the surgical cricothyrotomy on the unstable cervical spine in
Diicult Airway. Anesthesiology 2003; 98: 126977. a cadaver model of intubation. J Emerg Med 2001; 20: 15.
3. Frova G. he diicult intubation and the problem 17. Crosby ET. Considerations for airway management for
of monitoring the adult airway. Italian Society of cervical spine surgery in adults. Anesthesiol Clin 2007;
Anesthesia, Resuscitation, and Intensive herapy 25: 51133, ix.
(SIAARTI). Minerva Anestesiol 1998; 64: 36171.
18. McLeod AD, Calder I. Spinal cord injury and direct
4. Cook TM. Diicult Airway Society guidelines. laryngoscopy the legend lives on. Br J Anaesth 2000;
Anaesthesia 2004; 59: 12434. 84: 7059.
5. Randell T. Prediction of diicult intubation. Acta 19. Goutcher CM, Lochhead V. Reduction in mouth
Anaesthesiol Scand 1996; 40: 101623. opening with semi-rigid cervical collars. Br J Anaesth
6. Iohom G, Ronayne M, Cunningham AJ. Prediction of 2005; 95: 3448.
diicult tracheal intubation. Eur J Anaesthesiol 2003; 20: 20. Asai T. Pentax-AWS videolaryngoscope for awake nasal
316. intubation in patients with unstable necks. Br J Anaesth
7. Honarmand A, Safavi MR. Prediction of diicult 2010; 104: 10811.
laryngoscopy in obstetric patients scheduled for 21. Agro F, Barzoi G, Montecchia F. Tracheal intubation
Caesarean delivery. Eur J Anaesthesiol 2008; 25: 71420. using a Macintosh laryngoscope or a GlideScope in
8. Khan ZH, Mohammadi M, Rasouli MR, Farrokhnia 15 patients with cervical spine immobilization. Br J
F, Khan RH. he diagnostic value of the upper lip bite Anaesth 2003; 90: 7056.
test combined with sternomental distance, thyromental
22. Gronert GA, heye RA. Pathophysiology of
distance, and interincisor distance for prediction of
hyperkalemia induced by succinylcholine.
easy laryngoscopy and intubation: a prospective study.
Anesthesiology 1975; 43: 8999.
Anesth Analg 2009; 109: 8224.
23. Gronert GA. Succinylcholine-induced hyperkalemia
9. Adnet F, Borron SW, Racine SX, et al. he intubation
and beyond. 1975. Anesthesiology 2009; 111: 13727.
diiculty scale (IDS): proposal and evaluation of a new
score characterizing the complexity of endotracheal 24. Savarese J, Caldwell J, Lien C, Miller R. Pharmacology
intubation. Anesthesiology 1997; 87: 12907. of Muscle Relaxants and heir Antagonists. 5th ed.
Philadelphia, PA: Churchill Livingstone; 2000.
10. Nolan JP, Wilson ME. Orotracheal intubation in patients
with potential cervical spine injuries. An indication for 25. Langford RA, Leslie K. Awake ibreoptic intubation in
neurosurgery. J Clin Neurosci 2009; 16: 36672. 59
the gum elastic bougie. Anaesthesia 1993; 48: 6303.
Section 1: General considerations

26. Manninen PH, Jose GB, Lukitto K, Venkatraghavan L, corpectomy patients: a prospective, randomized,
El Beheiry H. Management of the airway in patients double-blind study. Spine (Phila Pa 1976) 2009; 34:
undergoing cervical spine surgery. J Neurosurg 22932.
Anesthesiol 2007; 19: 1904. 40. Kwon B, Yoo JU, Furey CG, Rowbottom J, Emery SE.
27. Ezri T, Szmuk P, Warters RD, Katz J, Hagberg CA. Risk factors for delayed extubation ater single-stage,
Diicult airway management practice patterns among multi-level anterior cervical decompression and
anesthesiologists practicing in the United States: have posterior fusion. J Spinal Disord Tech 2006; 19: 38993.
we made any progress? J Clin Anesth 2003; 15: 41822. 41. Miura Y, Mimatsu K, Iwata H. Massive tongue swelling
28. Jenkins K, Wong DT, Correa R. Management choices as a complication ater spinal surgery. J Spinal Disord
for the diicult airway by anesthesiologists in Canada. 1996; 9: 33941.
Can J Anaesth 2002; 49: 8506. 42. Avitsian R, Doyle DJ, Helfand R, Zura A, Farag E.
29. Avitsian R, Lin J, Lotto M, Ebrahim Z. Successful reintubation ater cervical spine exposure
Dexmedetomidine and awake iberoptic intubation for using an Aintree intubation catheter and a Laryngeal
possible cervical spine myelopathy: a clinical series. Mask Airway. J Clin Anesth 2006; 18: 2245.
J Neurosurg Anesthesiol 2005; 17: 979. 43. Jellish WS, Shea JF. Spinal anaesthesia for spinal surgery.
30. Ebert TJ, Hall JE, Barney JA, Uhrich TD, Colinco MD. Best Pract Res Clin Anaesthesiol 2003; 17: 32334.
he efects of increasing plasma concentrations of 44. Tetzlaf JE, Dilger JA, Kodsy M, al-Bataineh J, Yoon HJ,
dexmedetomidine in humans. Anesthesiology 2000; 93: Bell GR. Spinal anesthesia for elective lumbar spine
38294. surgery. J Clin Anesth 1998; 10: 6669.
31. Scher CS, Gitlin MC. Dexmedetomidine and low- 45. McLain RF, Kalfas I, Bell GR, et al. Comparison of
dose ketamine provide adequate sedation for awake spinal and general anesthesia in lumbar laminectomy
ibreoptic intubation. Can J Anaesth 2003; 50: 60710. surgery: a case-controlled analysis of 400 patients.
32. M M. Pharmacology and Use of Alpha-2 Agonists in J Neurosurg Spine 2005; 2: 1722.
Anesthesia, European Society of Anesthesiologists 46. Brimacombe JR, Wenzel V, Keller C. he proseal
Refresher Course, 2003; pp. 3743. laryngeal mask airway in prone patients: a retrospective
33. Abdelmalak B, Makary L, Hoban J, Doyle DJ. audit of 245 patients. Anaesth Intensive Care 2007; 35:
Dexmedetomidine as sole sedative for awake intubation 2225.
in management of the critical airway. J Clin Anesth 47. Sharma V, Verghese C, McKenna PJ. Prospective
2007; 19: 3703. audit on the use of the LMA-Supreme for airway
management of adult patients undergoing elective
34. Ebert T, Maze M. Dexmedetomidine: another arrow for
orthopaedic surgery in prone position. Br J Anaesth
the clinicians quiver. Anesthesiology 2004; 101: 56870.
2010; 105: 22832.
35. Doyle DJ. Awake intubation using the GlideScope video
48. Ng A, Raitt DG, Smith G. Induction of anesthesia
laryngoscope: initial experience in four cases. Can J
and insertion of a laryngeal mask airway in the prone
Anaesth 2004; 51: 5201.
position for minor surgery. Anesth Analg 2002; 94:
36. Sagi HC, Beutler W, Carroll E, Connolly PJ. Airway 11948.
complications associated with surgery on the anterior 49. Dingeman RS, Goumnerova LC, Goobie SM. he
cervical spine. Spine (Phila Pa 1976) 2002; 27: 94953. use of a laryngeal mask airway for emergent airway
37. Emery SE, Smith MD, Bohlman HH. Upper-airway management in a prone child. Anesth Analg 2005; 100:
obstruction ater multilevel cervical corpectomy for 6701.
myelopathy. J Bone Joint Surg Am 1991; 73: 54451. 50. Weksler N, Klein M, Rozentsveig V, et al. Laryngeal
38. Andrew SA, Sidhu KS. Airway changes ater anterior mask in prone position: pure exhibitionism or a valid
cervical discectomy and fusion. J Spinal Disord Tech technique. Minerva Anestesiol 2007; 73: 337.
2007; 20: 57781. 51. Lopez AM, Valero R, Brimacombe J. Insertion and use
39. Emery SE, Akhavan S, Miller P, et al. Steroids and risk of the LMA Supreme in the prone position. Anaesthesia;
factors for airway compromise in multilevel cervical 65: 1547.

60
Section 1 General considerations
Chapter
Spine imaging

5 Doksu Moon, Christian Koopman, and Ramez Malaty

he detailed anatomy of the vertebrae and fusion


Key points
variants is outside the scope of this chapter. However,
Understand the diference between disc disease a brief overview will be provided in the accompanying
and facet disease and how they afect the spinal diagrams. he spinal vertebra can be divided into the
canal and formina. vertebral body and posterior elements. he posterior
Understand the imaging features of infection, elements consist of the neural ring, transverse process,
especially discitis versus osteomyelitis. superior facet joints, inferior facet joints, and spin-
Understand the key indings in spine trauma ous process. he neural ring consists of the posterior
and patterns of injury. aspect of the vertebral bodies, pedicles, and laminae.
Learn an imaging approach for neoplasms of the here are multiple excellent textbooks and chapters for
spine, including intramedullary lesions versus study2,3 (Figs. 5.1 and 5.2).
intradural extramedullary neoplasms.
Marrow
Evaluation of marrow signal abnormalities can be
Introduction diicult even for the experienced neuroradiologist.
Imaging is integral to the diagnosis of spine disease. Marrow signal on T1 images is typically hyperintense
his chapter will provide an overview of imaging of (bright) in the elderly and middle-aged and decreases in
spine disease. It will give a brief overview of normal younger patients. he reason is that older patients have
imaging anatomy and illustrate highlights of patho- increased lipid in their marrow.4 In the very young the
logic processes in the spine. Ater reviewing this T1 marrow signal may be very dark due to the preva-
chapter the reader should be able to describe how lence of red marrow. However, no matter how much
to determine normal marrow signal in the spine, red marrow is present, normal marrow is hyperintense
describe and identify degenerative disc and facet dis- to the intervertebral disc signal (Fig. 5.3).
ease, discuss the appearance of pyogenic discitis on he three main imaging sequences available are
MRI, discuss the four types of spinal arteriovenous T1-weighted (T1W) imaging sequences, T2-weighted
malformations, and summarize common neoplasms (T2W) imaging sequences, and inversion recovery
of the spine. (STIR or IR) sequences. T1 signal is the most useful for
evaluating marrow signal abnormalities. Areas of abnor-
Normal spine mal marrow, whether from a marrow replacement proc-
here are typically 7 cervical vertebrae, 12 rib-bearing ess (e.g., myelodysplasia, sick cell disease, metastasis)
thoracic vertebrae, 5 lumbar type vertebrae, and 5 or edema (e.g., acute or subacute compression fracture)
fused sacral segments. However, many variants includ- are of decreased signal compared with normal mar-
ing fusion of the C1 through C3 vertebrae, fusion of C1 row. Inversion recovery sequences are also very useful.
with the skull base, cervical ribs, 11 or 13 rib-bearing hose sequences typically saturate the fat signal so that
ribs, and transitional lumbosacral vertebra all combine fatty structures such as subcutaneous fat and marrow
to add complexity in the counting process. Careful are rendered as decreased signal. T2W and postcon-
delineation of the spinal levels on imaging is necessary trast sequences are not as useful for evaluating normal
to avoid wrong-level surgery.1 marrow. On T2W sequences, for the most part, normal

61
Anesthesia for Spine Surgery, ed. Ehab Farag. Published by Cambridge University Press. Cambridge University Press 2012.
Section 1: General considerations

Figure 5.1 Normal cervical spine MRI.


(A) Sagittal T1 TR600/TE11. (B) Sagittal
T2 TR4000/TE60. (C) and (D) Axial
gradient echo TR30/TE15: anterior arch
of atlas (AA), posterior arch of atlas (AP),
foramen of Magendie (FM), vertebral
body (V), spinous process (Sp), medulla
(M), cervical spinal cord (C), nucleus
C pulposus (NP), pedicle (P), inferior facet
(IF), superior facet (SF), neuroforamen
(NF), lamina (L).

A B D

Figure 5.2 Normal lumbar spine MRI:


(A) Sagittal T1 TR550/TE11. (B) Sagittal
T2 TR4040/TE60. (C) and (D) Axial T2
TR4040/TE60: thoracic cord (TC), spinous
process (SP), cauda equina (CE), nucleus
pulposus (NP), vertebral body (V), conus
medullaris (CM), annulus ibrosus (AF),
neuroforamen (NF), inferior facet (IF),
superior facet (SF), pedicle (P), transverse
C process (TP).

A B D

Figure 5.3 Myelodysplasia. (A) T1


sagittal T spine TR500/TE11. (B) T2 sagittal
T spine TR2910/TE104. (C) STIR sagittal
T spine TR4200/TE60. Note that marrow
signal in (A) is difusely decreased
compared with the signal in the disc on
T1W images.

62
Chapter 5: Spine imaging

marrow signal is hyperintense and pathologic marrow be degenerative disease present in up to one-third of
signal also tends to be hyperintense due to the edema asymptomatic adults, although more severe degenera-
associated with it. Postcontrast images are also limited tive changes such as extrusions and sequestration are
in evaluating marrow signal since pathologic marrow found rarely in the asymptomatic population.6,7 We can
signal enhances as well as normal marrow signal. divide degenerative spine disease anatomically into
disc disease and facet disease.
Degenerative disease
Degenerative disease is the most common reason for Degenerative disc disease
spine imaging and surgery in the United States. here he primary problem is a defect in the annulus ibrosis,
is evidence that genetic factors as well as environmen- which surrounds the nucleus pulposus. Disc disease
tal factors contribute to disc disease.5 Degenerative can produce spinal canal stenosis and neuroforaminal
disease is diagnosed by imaging. he major pitfall stenosis via disc bulges, disc protrusions, disc extru-
in imaging of degenerative disease is that there may sions, and sequestrations.
he weakest point in the annulus ibrosis tends to be
dorsal, so the irst area afected is the spinal canal and
thecal sac.8 Degenerative disc disease can result in both
spinal canal and neuroforaminal stenosis. he classii-
cation is complex and has evolved over the years, but
there is a common classiication system used in North
America (Nomenclature and Classiication of Lumbar
Disc Pathology) which has been recommended by both
the North American Spine Society and the American
Society of Neuroradiology. According to this system a
disc bulge is greater than 50% (180) of the circumfer-
ence of the disc; a broad-based disc protrusion is hernia-
tion of the disc material greater than 25% and less than
50% of the circumference; a protrusion is displacement
of the disc that is less than 25% of the circumference
of the disc; and an extrusion is displacement or hernia-
tion of the disc in which in at least one plane the edge of
the disc material is greater than the distance of the disc
material at the base9 (Figs. 5.45.6).
Figure 5.4 Neuroforaminal stenosis. Sagittal T2 shows moderate Degenerative end plate changes oten accompany
left L34 stenosis due to loss of disc height and lateral extension of
disc bulge. degenerative disc disease. Type I end plate changes

Figure 5.5 (A) Sagittal T1 TR705/


TE12. (B) Axial gradient echo TR30/TE14
showing multiple disc extrusions in
the cervical spine. The most prominent
ones are at C34 and C45 causing cord
compression.

63
Section 1: General considerations

Figure 5.6 T2 images TR4200/TE117


show a disc sequestration (free fragment)
in the lumbar spine dorsal to L3.

A B

(edematous) show decreased T1 signal and increased signs are hypertrophy of the facet joints (most read-
T2 signal. Type II (fatty) end plate changes show ily visible in the lower lumbar spine). With suicient
increased T1 and T2 signal in the adjacent end plates. degenerative changes, facet hypertrophy alone can
Type III changes (sclerotic) show decreased T1 and result in moderate or several spinal canal stenosis. Facet
T2 end plate signal changes. hese types of end plate disease also afects the neuroforamina, causing narrow-
changes can transition between each other.10 ing due to degenerative changes. Facet efusions are vis-
Even when there appears to be CSF ventral to the ible on T2W images as areas of increased signal. If facet
spinal cord and no impingement or compression of the efusion becomes large enough it can become a synovial
spinal cord by visual inspection, there can be compres- cyst which can cause (asymmetric) spinal canal steno-
sion of the nerves due to compression on the individ- sis. Despite the cystic nature of these lesions they may
ual nerve rootlets that exit from the ventral (anterior) show decreased signal on T2W images depending on
aspects of the spinal cord. Even with current imaging the amount of associated calciication.11
the individual nerve rootlets are not usually visualized
in a typical MRI scan. OPLL
In the most severe cases there is complete efacement Ossiication of the posterior longitudinal ligament
of the CSF surrounding the cord or nerve roots. his (OPLL) occurs most oten in patients of East Asian
can be appreciated on both T1 and T2 axial images. descent, but also in the rest of the population.12 he
Although many herniations indent the thecal sac, exact etiology remains unclear but the process results
some are more laterally placed and compress the nerve in spinal canal stenosis and possible cord compression.
root at the neuroforamen. hese are termed lateral or here is also an association with DISH.13
foraminal protrusions or extrusions.
DISH
Facet disease Difuse idiopathic skeletal hypertrophy (DISH) hap-
Degenerative facet disease can be a cause of back pain pens sporadically. It is identiied by four or more levels
both by itself and also because of direct efects on the of contiguously fused vertebral bodies. here is a cor-
spinal canal and neuroforamina. Degenerative facet relation with increasing age and it is common in the
changes can indirectly cause narrowing of the spinal population over the age of 50 years (males 25%, females
canal and foramina through spondylolisthesis. 15%)14 (Fig. 5.7).
64 he degenerative facet joints can be a source of pain Baastrups disease is the close contact of the spinous
that is unrelated to radicular pain. he most obvious processes (kissing spine). It results in inlammatory
Chapter 5: Spine imaging

changes between the spinous processes including


edema, cyst formation, luid accumulation, and scle-
rotic changes. hese areas of edema are seen on T2 and
inversion recovery as areas of hyperintensity.15
Schmorls nodes. Not signiicant by themselves in
most cases, but in severe cases these are associated with
compression fractures. hese are internal disc her-
niations, with the nucleus pulposus extending through
the cartilaginous end plates rather than the annulus
ibrosus. his situation is physiologic since in normal
patients the annulus is more resistant to axial loads
than the end plates.
Scheuermanns disease. Osteochondrosis of the thor-
acic spine resulting in a kyphosis. he basic process is
interruption of the blood supply and osteonecrosis of
the vertebrae resulting in mild kyphosis.

Sarcoidosis
Sarcoidosis is an idiopathic systemic disease. In the
head and spine, it presents as thickening and enhance-
ment of the leptomeninges. Spinal sarcoidosis can
involve intramedullary, intradural extramedullary,
extradural, vertebral, and disc space lesions.
Intramedullary sarcoid is uncommon. Lesions are
high in signal intensity on T2 and low on T1 and patchy
enhancement ater contrast administration.
Leptomeningeal and dural lesions are more com-
mon than intramedullary lesions and can be seen on
postcontrast T1 weighted images as thin, linear lep-
tomeningeal enhancement or small nodules (Fig. 5.8)
Clinical manifestations do not correlate well with MRI
indings. Clinical manifestations result from local
Figure 5.7 Sagittal reconstruction from a thoracic spine CT nerve or spinal cord compression. Surgical resec-
myelogram showing lowing anterior syndesmophytes in the lower
thoracic spine extending to four levels, relecting DISH (difuse tion is usually necessary to relieve the compression,
idiopathic skeletal hypertrophy). followed by steroids. Diferential diagnosis includes
Figure 5.8 Sarcoidosis. Pre- (A) and
post- (B) contrast sagittal T1 TR500/TE13
MRI shows nodular enhancement in the
conus medullaris.

65
Section 1: General considerations

Figure 5.9 Ankylosing spondylitis.


Sagittal T2 images TR2780/TE115 show
fusion of the vertebral bodies and
facet joints. Note the lack of articular
cartilage between the facet joints on
the parasagittal image (B), which shows
that the facet joints are fused. On the
midsagittal image (A) the intervertebral
discs in the mid- and upper cervical
spine do not extend to the margins of
the vertebral bodies. The C25 vertebrae
show body bridging anterior and
posterior to the intervertebral discs.

A B

lymphoma and carcinomatous metastasis among from the presence of inlammatory pannus surround-
other etiologies.16 ing the dens.
Vertebral involvement is rare with multiple, well-
deined lytic lesions with sclerotic margins. Mimicking Infection
of metastasis by such lesions is also rare.16
Pyogenic discitis and osteomyelitis
Ankylosing spondylitis his most oten involves vertebral bodies (osteomyel-
Ankylosing spondylitis is a relatively common rheu- itis). However, it may also involve posterior elements,
matologic condition in adults with incidence of 1.4% in discs (discitis), epidural space, and paraspinous sot tis-
the general population. his progressive seronegative sues.19 Etiology is most commonly bacterial and usually
spondyloarthropathy results in fusion of the vertebral Staphylococcus aureus, although fungi or parasites may
bodies and sacroiliac (SI) joints. It has an ascending be involved. Infection may be due to hematogenous
course of progress, with the SI joints and lumbar spine spread, contiguous spread, or direct inoculation in the
the most commonly and irst afected.17 his disease setting of trauma or surgery. Infections are most com-
may result in fractures and dislocations, and spinal ste- monly hematogenous and usually via the skin, urin-
nosis with neurologic compromise18 (Fig. 5.9). ary tract, or pulmonary sources.19 In adults, infection
starts in the subchondral portion of the vertebral body
Rheumatoid arthritis and spreads to the disc space and then further along
Rheumatoid arthritis is oten associated with its the vertebral body in a subligamentous fashion.20 In
efects on the peripheral joints, but perhaps its most children, however, disc space infection may be the pri-
devastating efect is in the cervical spine where inlam- mary site with vertebral body infection secondary.19,21
matory changes can result in craniocervical subluxa- Infection most commonly involves the lumbar spine.21
tion. Otentimes it is asymptomatic but it can result Symptoms vary, with pain and malaise being com-
in myelopathy. he two most common presentations mon, and patients may be afebrile. Neurologic deicit
are atlantoaxial subluxation (widening of the anterior and cord compression may occur if infection spreads
atlanto-odontoid distance) and basilar invagination. into the epidural space.21 Plain radiographs are of low
Atlantoaxial subluxation can also result in C12 insta- sensitivity for detecting this infection. Radionuclide
bility. his can be diagnostic with lexionextension bone scans are sensitive but nonspeciic for this infec-
66 plain ilms of the lateral cervical spine (Fig. 5.10). tion. CT scans are also less sensitive.22 Findings on
Myelopathy and cord compression can also result MRI are characteristic, with T1W sequences showing
Chapter 5: Spine imaging

Figure 5.10 Rheumatoid arthritis. (A) Extension view. (B) Neutral view. (C) Flexion view. Note the widening of the anterior atlanto-odontoid
distance (AAOD) on the lexion view.

Figure 5.11 Discitis. Pre- (A) and


postcontrast (B) axial T1 images
through the level of the disc showing
circumferential enhancing soft tissue
surrounding the vertebral body at
the level of discitis. Note the epidural
involvement.

A B

a narrowed disc space and decreased signal in adja- disease, and in rare circumstances, metastasis25 (Figs.
cent vertebral bodies.19,22 Subligamentous or epidural 5.11 and 5.12).
sot tissue luid collections and cortical bone erosion
are common. Postcontrast examination demonstrates Granulomatous spondylitis
enhancement of the infected disc and infected bone. his is most commonly due to Mycobacterium tuber-
Paraspinous abscess, epidural abscess, and meningeal culosis but can be seen in bacterial, viral, parasitic, and
inlammation can all occur and may be detected on fungal etiologies among others, as well as tumors and
MRI imaging.23,24 autoimmune diseases.19 Other implicated organisms
he diferential diagnosis includes granulomatous are bacilli of the Brucella genus.26 Tuberculous spon- 67
spondylitis, calcium pyrophosphate crystal deposition dylitis is most prevalent in middle-aged adults with
Section 1: General considerations

A B C D
Figure 5.12 Discitis/osteomyelitis. (A) Sagittal STIR TR4350/TE810. (B) Sagittal T2 TR4060/TE120. (C) Sagittal T1 TR500/TE12. (D) Sagittal T1
postcontrast TR500/TE12. Typical appearance showing increased marrow signal on T2 and STIR. Decreased signal on T1 and difuse end plate
enhancement on postcontrast images.

predisposing factors including debilitation, immuno- or osteomyelitis seen in 80% of cases.30 Fever and local-
suppression, alcoholism, and drug addiction.19 he lum- ized tenderness are early symptoms but symptoms are
bar spine is also the most common level of involvement, oten nonspeciic. Predisposing conditions include
with nearly 90% of cases having at least two afected diabetes, intravenous drug abuse, multiple medical
vertebral bodies with skip lesions commonly occur- illnesses, and trauma. Epidural abscess may result in
ring.19,27,28 Paraspinous abscesses occur in more than severe neurologic deicit or even death in untreated
50% of cases.27 Tuberculosis can afect only part of the cases.31,32 Myelography and CT myelography demon-
vertebral body, with transverse processes and posterior strate extradural sot tissue mass with blockage of nor-
elements involved only some of the time.19 Tuberculous mal CSF low.32 MRI scans show extradural sot tissue
spondylitis is typically more indolent than pyogenic mass that is isointense to hypointense compared with
osteomyelitis, with insidious onset and symptoms last- spinal cord on T1W images and hyperintense on proton
ing months to years and with untreated patients devel- density and T2W images31 Coexisting signal changes
oping progressive vertebral body collapse and gibbus in adjacent vertebral bodies are oten seen. hree pat-
formation.27,29 CT imaging demonstrates extensive terns are observed ater contrast administration. One
bony destruction and large paraspinous abscesses that is of difuse homogeneous enhancement seen in 70% of
are disproportionate to the amount of bone destruction. the cases, likely representing a phlegmonous stage. he
Epidural extension and subligamentous spread are also second most common inding is a thick or thin enhan-
frequently present.19,27 MRI shows loss of cortical def- cing rim surrounding a liqueied, low-signal pus col-
inition of vertebral bodies involved. Infection spreads lection, representing frank necrotic abscess. Finally, a
beneath longitudinal ligaments to involve adjacent ver- combination of both patterns can be observed.31,32
tebral bodies, with discs sometimes relatively spared.
Posterior elements are commonly involved.27 Subdural abscess
Subdural abscesses are rare and the rarity is due to
Epidural abscess absence of venous sinuses in the spine, wide epidural
Again infection is typically hematogenous from skin, space acting as a ilter, and centripetal direction of spi-
pulmonary, or urinary tract sources, with S. aureus nal blood low.34 Clinical presentation is nonspeciic,
being by far the most common organism.20,30 Two basic with symptoms mimicking acute transverse myelitis
stages are observed. he irst stage demonstrates thick- and spinal epidural abscess among other pathologies.
ened and inlamed tissue with granulomatous mater- Imaging studies reveal an intraspinal space-occupying
ial and embedded microabscesses that represent a mass with features that would localize the lesion to the
phlegmonous stage.31 he second stage demonstrates a subdural compartment.
collection of liquid pus material with frank abscess for-
mation.30 Epidural abscesses are uncommon, with inci- Meningitis
dence on the rise and mean age being 5055 years3133 he cause of meningitis is fungal, parasitic, or viral,
68 hey are commonly extensive and can extend along with pyogenic leptomeningitis being the most com-
multiple vertebral body levels with concomitant discitis mon bacterial infection. he majority of cases occur
Chapter 5: Spine imaging

A B C D

Figure 5.13 Syphilis. Postcontrast T1 (TR500/TE12) images through the thoracic and lumbar spine demonstrate difuse leptomeningeal
enhancement throughout the spinal cord (A) and clumped enhancing nerve roots (C and D) relecting adhesive arachnoiditis. Note the
narrowing of the lateral diameter of the thoracic cord (B) on the axial images.

as manifestation of cerebral meningitis.19 Infection is sources.36 Infectious agents can cause myelitis. Viral
seen as contrast-enhancing tissue that surrounds the infections typically afect the gray matter with her-
spinal cord and nerve roots. hree patterns are seen: pes, coxsackie, and HIV being the more common
one pattern is delicate, smooth, linear enhancement infections.36 Epidural abscess and chronic meningeal
outlining the cord and nerve roots; another is discrete infection such as tuberculosis and fungal meningitis
nodular foci on the surface; a third pattern is of dif- can also cause a secondary myelitis.21,35 Imaging ind-
fusely thickened sot tissues appearing as an intradural ings are nonspeciic, with focal or difuse increased
illing defect. here is no correlation between the pat- intramedullary signal on T2W images with or without
tern of enhancement and etiology of the organism or mass efect and with or without enhancement follow-
disease severity35 (Fig. 5.13). ing contrast administration.35

Paraspinous abscess Intramedullary abscess


Instrumented patients can have seeding of the hard- Frank pyogenic spinal cord abscesses are extremely
ware by hematogenous spread of organisms (Fig. 5.14). rare compared with brain abscesses. here are only a
Hematogenously spread pyogenic organisms can oten few reported cases.19
seed the paravertebral muscles unless the scan is care-
fully examined (Fig. 5.15). Demyelinating diseases
Myelitis Multiple sclerosis
he term myelitis is restricted to inlammatory diseases Multiple sclerosis is a disease of the central nervous
of the spinal cord, with myelopathy being a general term system. here are plaques that occur in the brain and 69
applied to cord dysfunction from noninlammatory spinal cord. Spinal cord plaques are almost universal
Section 1: General considerations

Figure 5.14 Postcontrast T1 images


TR427/TE11 through the cervical thoracic
junction show a prevertebral abscess
(asterisk) anterior to the area of fusion.
There is also epidural enhancement at
C3 through C7 causing narrowing of
the spinal canal without evidence for an
epidural abscess.

A B

Figure 5.15 Psoas abscess. Pre- (A)


and postcontrast (B) axial T1 TR682/TE12
images showing a rounded peripherally
enhancing abscess which was not
apparent on precontrast images in the
left iliopsoas muscle.

A B

at autopsy indings. Plaques occur preferentially in most commonly reveals one or more elongated, poorly
the dorsal lateral cord and do not respect boundaries marginated hyperintense intramedullary lesions on
between speciic tracts or between gray and white mat- T2W images. Focal or generalized cord atrophy can
ter. Multiple sclerosis demonstrates disease onset typi- be seen on T1W images. Acute demyelinating lesions
cally between 15 and 50 years of age, with the peak in may have mass efect and enhance following contrast
the third and fourth decades and with distinct female administration.38,39
predominance. In early disease, there is distinct cervi-
cal spinal cord predilection.37 Initial imaging evaluation Acute transverse myelopathy
in suspected multiple sclerosis is by brain MRI. Spinal Sometimes termed acute transverse myelitis, this is char-
MRI is not required for conirmation when deini- acterized by an acutely developing, rapidly progressing
tive diagnosis of multiple sclerosis is made on clinical lesion afecting both sides of the cord. his is not a true
grounds. In patients suspected to have multiple scle- disease but a clinical syndrome with many causes.40
70 rosis, if brain MRI is normal, MRI examination of the Causes include active infection, postinfectious demye-
spinal cord is appropriate. Imaging of the spinal cord linating disorder (acute disseminated encephalomyelitis
Chapter 5: Spine imaging

or ADEM), immune disorders such as systemic lupus Compressive myelopathy


erythematosus, multiple sclerosis, vascular occlusion
Intramedullary high signal intensity foci on proton
with resultant cord infarction; it may occur following
density or T2W MRI scans in cases of moderate to
vaccination and as a complication of systemic malig-
marked spinal stenosis have been observed.45 his can
nancy similar to limbic encephalitis, and oten the
be due to degenerative disc disease or spondyloarth-
etiology is unknown. Annual incidence is 1 case per mil-
ropathy and is probably related to focal cord ischemia.
lion without age or sex predilection. In the typical case,
Some cases resolve following decompressive surgery.45
there is no prior history of neurologic abnormality, with
Other causes of compressive myelopathy can be due to
time from symptom onset to maximum deicit ranging
mass efect from primary or secondary spine tumors or
from 1 hour to 17 days. Prognosis is poor in most cases,
other epidural lesions such as epidural abscess.36
with severe residual neurologic deicit being common.
Imaging is to exclude treatable conditions that can mimic Degenerative and toxic myelopathies
acute transverse myelitis. hese may include acute disc
Inherited and acquired degenerative disorder such as
herniation, hematoma, epidural abscess, or compres-
Friedrich ataxia and other spinocerebellar degenera-
sion myelopathy. Imaging indings are nonspeciic, with
tions, amyotrophic lateral sclerosis, toxic diseases such
focal cord enlargement on T1W and poorly delineated
as chronic alcoholism, and metabolic disorders such as
hyperintensities on T2W images, and enhancement
vitamin B12 deiciency are miscellaneous causes of spi-
occurring in some cases.36,40,41
nal cord dysfunction.36
Miscellaneous myelopathies and Superficial siderosis
conditions his is due to previous episodes of hemorrhage in the
subarachnoid space. he pathognomonic appearance
Radiation myelopathy is difuse decreased signal surrounding the cord and
his is a rare complication of therapeutic irradiation. brainstem. Hemosiderin is toxin to the neuronal tissue
hree criteria exist to establish the diagnosis. hese causing volume loss of the CNS. Patients most com-
include that the spinal cord must have been included in monly present with hearing loss, ataxia, and myelop-
the radiation ield, that the neurologic deicit must cor- athy46 (Fig. 5.16).
respond to the cord segment that was irradiated, and
that metastasis or other primary spinal cord lesions must Postoperative arachnoiditis
be ruled out.42 Four distinct clinical syndromes exist in Postoperative arachnoiditis is a chronic inlammatory
irradiated spines, with chronic progressive radiation condition causing formation of scar tissue. his results
myelopathy (CPRM) being the most common form in clumping of the nerve roots. he nerve roots may
identiied on imaging studies.4143 Most cases are seen adhere to the dura,which results in a false empty sac;
following radiotherapy of nasopharyngeal carcinoma, they may clump together to form what appears to be
with the area most commonly afected accordingly a few but very thick nerve roots; or they may adhere
being cervical spinal cord. he latent period between to each other in an irregular fashion. Enhancement of
termination of irradiation and onset of symptoms varies the nerve roots may also occur in postoperative arach-
from 3 to 40 months, with most cases occurring between noiditis47 (Fig. 5.17).
9 and 20 months.43 Imaging indings vary. If MRI scans
are observed more than 3 years ater symptom onset, Vascular diseases
cord atrophy without abnormal signal is seen. Scans
performed within 8 months of symptom onset typi- Aneurysms
cally demonstrate long segment hyperintense lesions on Spinal aneurysms are localized saccular dilatations of
T2W images with or without associated cord swelling spine or spinal cord arteries that are frequently asso-
and enhancement following contrast administration.44 ciated with intramedullary spinal cord arteriovenous
malformations. hese are extremely rare. Most com-
AIDS-related myelopathy monly they are seen in the cervical and thoracic spinal
his is probably related to direct injury of neurons by cord. hey are almost always located on one of the main
HIV, although secondary demyelination of posterior high-low vessels feeding the arteriovenous malfor- 71
and lateral columns also occurs.36 mation, with nearly 70% found on the anterior spinal
Section 1: General considerations

Figure 5.16 Supericial siderosis.


(A) Sagittal STIR TR4000/TE60. (B)
Axial gradient echo TR30/TE15. There
is difuse circumference-decreased
signal surrounding the cervical cord
and brainstem due to the susceptibility
artifact from the hemosiderin deposition.
Also note the difuse volume loss in the
cord, which is also a feature of siderosis.

Figure 5.17 Pre- (A) and postcontrast


(B) axial T1 TR580/TE11 images through
the lumbar spine showing enhancement
of the nerve roots relecting
arachnoiditis.

artery. In contrast to intracranial aneurysms, spinal arteries and draining veins that are both enlarged. In
aneurysms do not usually occur at bifurcation points. contrast, arteriovenous istulas drain directly into an
Symptoms are due to subarachnoid hemorrhage in the enlarged venous outlow tract.47,48
majority of cases. Angiography is the deinitive imag- Spinal AVMs are subdivided into four general cat-
ing study.36 egories. Type I is a dural arteriovenous istula that is pri-
marily found in the dorsal aspect of the lower thoracic
Vascular malformations cord and conus medullaris. Most consist of a single
hese are uncommon lesions. Most are arteriovenous transdural arterial feeder that drains into an intradural
malformations or arteriovenous istulas. Cavernous arterialized vein. he draining vein oten extends over
angiomas and capillary telangiectasias are less com- multiple segments. Nearly 60% are spontaneous and
mon, with venous angiomas being rarely seen. approximately 40% are posttraumatic. Progressive
neurologic deterioration likely due to chronic venous
Arteriovenous malformations and hypertension is typical4951 (Fig. 5.18). Type II AVMs
called glomus malformations are intramedullary AVMs
arteriovenous fistulas in which a localized compact vascular plexus is supplied
72 Arteriovenous malformations (AVMs) have a true by multiple feeders from anterior or posterior spinal
nidus of pathologic vessels interposed between feeding arteries. Type II AVMs drain into a tortuous arterialized
Chapter 5: Spine imaging

AVMs are the most common spinal vascular anom-


aly. he most common AVM is type I, with type III
being least common. he thoracolumbar area is the
most common location overall.48,52 Paresis, sensory
changes, bowel and bladder dysfunction, and impo-
tence are common symptoms. Hemorrhage is seen
frequently. Venous hypertension may be important in
the development of cord symptoms.48 Imaging indings
include illing defects of enlarged vessels seen on mye-
lography. Cord atrophy is common. MRI may show
low voids with enlarged vessels. High signal intensity
is oten seen on T2W images, with the cord sometimes
being atrophic. Hemorrhagic byproducts may be pre-
sent. Spinal angiography is the deinitive diagnostic
procedure for evaluation of spinal AVMs.51,52

Cavernous angiomas
hese are similar to intracranial cavernous angiomas,
with imaging indings demonstrating well-circum-
scribed masses. Microscopically they consist of blood-
illed, endothelium-lined spaces. here are localized
hemorrhages of diferent ages. Calciication is rare.
hese are, however, extremely rare lesions. Spinal
angiography is typically normal since these are slow-
low vascular lesions. Findings on MRI scans dem-
onstrate blood products of subacute and chronic ages
with mixed high- and low-signal components. Typical
appearance is a small, high-signal focus on both T1W
and T2W images and typical imaging characteristics
on gradient-refocused scans. If a typical spinal cord
Figure 5.18 Sagittal T2 image (TR7660/TE79) shows increased lesion is identiied on MRI scan, the brain should be
signal in the conus medullaris as well as low voids associated with studied using gradient-refocused sequences to screen
the cauda equina in a patient with a dural AV istula.
for asymptomatic intracranial lesions.5355

venous plexus that surrounds spinal cord. hese AVMs Cord infarction
are located dorsally in the cervicomedullary region, Arterial infarction: Blood supply to the cord depends
with most occurring in younger patients with acute on three longitudinal arterial trunks: a single anterior
onset of neurologic symptoms due to intramedullary spinal artery and paired posterior spinal arteries with
hemorrhage.49 Type III AVMs, called juvenile type, are collateral low comparatively limited. he anterior spi-
large, complex vascular masses that involve the cord and nal artery gains its supply in most individuals from the
oten have extramedullary or even extraspinal exten- artery of Adamkiewicz, which usually originates from
sion. Multiple arterial feeders from several diferent the let-sided spinal arteries directly of the aorta at the
vertebral levels are common. Type IV AVMs are intra- T1012 level. It sports a characteristic hairpin curve.
dural extramedullary arteriovenous istulas that are fed Spontaneous anterior spinal cord infarction primarily
by the anterior spinal artery and lie completely outside afects individuals with severe atherosclerotic disease or
the spinal cord and pia matter. here is no intervening aortic dissection, with other reported etiologies includ-
small-vessel network and the istula drains directly into ing vasculitis and hypertension.56, 5859 Arterial infarction
an enlarged venous outlow tract. Most are anterior to is extremely rare and is most oten seen ater aortic sur-
the spinal cord and fed by the anterior spinal artery and gery. Most cord infarctions occur at the upper thoracic
most occur near the conus medullaris. Progressive neu- or thoracolumbar junction, with extensive involvement 73
rologic deicits are typical.49 ranging from a single segment to multiple levels.56,57
Section 1: General considerations

Clinical symptoms vary with classic presentation includ- symptom onset. Follow-up scans may show cord atro-
ing sudden onset of laccid paraparesis or quadriparesis. phy focally and residual high signal intensity on T2W
Associated sensory loss with preserved touch, vibration, images.58,60 In terms of imaging, chronic cord infarct is
and position sense is common.58 Imaging demonstrates indistinguishable from myelomalacia of other causes.
enlargement of the cord on T1W images with central In acute cases, difusion imaging may be performed,
or anterior intramedullary high signal on T2W images. which shows increased signal on the difusion-weighted
Enhancement following contrast administration may be images and decreased signal on the apparent difusion
initially absent but occurs a few days to a few weeks ater coeicient maps. (Figs. 5.19, 5.20).

Figure 5.19 Infarct. Axial gradient


echo images. Note the increased signal
along the ventral aspect of the cord
in (A) relecting infarction of the cord.
(B) Shows the normal appearance of
the cord above the level of spinal cord
infarction. Also note the descending
aortic dissection.

Figure 5.20 Spinal cord infarct. (A) Sagittal T1 TR500/TE12. (B) Sagittal T2 TR4620/TE117. (C) Sagittal STIR TR3830/TE82. Note that on the
74 sagittal T1 image the spinal cord looks normal but on the sagittal T2 and STIR there is a band of hyperintensity in the ventral aspect of the
cord relecting the area of infarction in the cord.
Chapter 5: Spine imaging

Venous infarction, also known as subacute necrotic vertebral body and disc. he middle column is composed
myelitis, is a less well known entity. Pathologic studies of the posterior longitudinal ligament and posterior
demonstrate enlarged, thick-walled oten thrombosed 1st/3rd vertebral body and disc. he posterior column is
veins with necrosis involving both gray and white mat- composed of the posterior arch, articular processes, and
ter. MRI suggests vascular malformation with serpentine posterior ligamentous complex (supraspinous ligament,
illing defects, thrombosed veins, and cord edema.43,61 interspinous ligament, ligamentum lava, articular pil-
lar capsular ligaments). he basic premise is that injuries
Spine trauma isolated to the middle column and injuries involving two
A standardized approach to sensitive and eicient assess- or more columns are considered unstable. his classii-
ment of the spine is essential, with nearly 11 000 spinal cation system was designed to be applied to the thoracic
cord injuries occurring in the United States annually, and lumbar spine, but allowing for certain modiica-
and total lifetime treatment and rehabilitation costs in tions, the model may be applied to the cervical spine.
the range of $200 000 to $800 000 per patient.62,63 In the Classiication schemes based on injury mechanism
rare absence of CT, initial plain ilm evaluation may be are also proposed as an aid to characterization of not only
warranted, with inclusion of no fewer than three stand- fracture patterns but anticipated ligamentous disruption
ard views: anteroposterior, lateral, and open mouth and subsequent instability. hese mechanisms include
odontoid. With missed fracture rates between 23% and axial loading, lexion, lateral compression, lexion
57% and delays in diagnosis ranging from 5% to 23%, rotation, lexion distraction, extension, and shear.
there has been a global transition to CT evaluation of
the spine in those patients clinically deemed high-risk Cervical spine injuries
as the preferred initial modality of choice.6466 While an all-inclusive review of cervical spine injuries
he isotropic nature of data acquisition allows for is beyond the scope of this chapter, several classic pat-
accurate sagittal and coronal reformatted sequences, terns of injury are discussed.
avoiding the axial limitations of decreased sensitiv- he Jeferson fracture typically occurs in conjunc-
ity to subluxation, increased intervertebral distance, tion with axial loading, such as with a blow to the top of
angulations, and horizontally oriented fractures. he the head, resulting in disruption of the osseous ring of
strength of CT remains in detection of the full spectrum C1 with lateral displacement of the lateral masses, best
of fracture patterns, facet injuries, and vertebral body evaluated in coronal reformatted sequences. he clas-
subluxations and malalignment.67 Addition of intra- sic form is considered stable.70,71
venous contrast in combination with CT angiography Atlantoaxial dissociation is a generic term applied
(CTA) protocols is also commonly used to evaluate for to injuries in which the C1 ring is displaced by rotation
acute vascular injury when suggested by the pattern about the odontoid peg or articular mass. Concomitant
of osseous injury. Typical MDCT (multidetector CT) fractures, transverse atlantal ligamentous injury, and
technique consists of 0.75 mm axial acquisitions with vertebral artery injury may result. Diagnosis is sug-
subsequent axial, sagittal, and coronal reformats utiliz- gested by asymmetry of the atlantodental space and
ing both bone and sot tissue algorithms. anterior or posterior displacement of lateral mass(es);
Magnetic resonance imaging, subsequently dis- stability is variable. hree-dimensional CT reformats
cussed, is the only modality capable of identifying cord oten aid in conirmation of the diagnosis.72
edema, cord hemorrhage, disc injury, and canal com- Odontoid fractures are commonly classiied as
promise in the presence or absence of osseous injury in types I, II, or III. Type I injury is an oblique fracture
the setting of focal neurologic deicit. through the tip; type II, or high odontoid fracture, is
a horizontal fracture through the base; and type III, or
Spinal injury patterns: stability and low odontoid fracture, is generally an oblique fracture
mechanism through the base with extension into the body of C2.
he most widely accepted spinal stability assessment Type I fractures generally involve avulsion of portions
model is a three-column classiication system proposed of the alar ligament as may result from atlantooccipital
by Denis.68,69 he basis of the model is the sagittal des- dislocation and thus should be treated as unstable.73
ignation of anterior, middle, and posterior osteoliga- Traumatic spondylolisthesis as a result of hyperex-
mentous columns. he anterior column is composed of tension distraction in the upper cervical spine results 75
the anterior longitudinal ligament and anterior 2nd/3rd in a hangmans fracture. Classically there is fracture
Section 1: General considerations

Figure 5.21 Two axial images from


a cervical spine ilm in a patient post
fall demonstrates a fracture through
the right pars interarticularis (*) and
lamina (**).

A B

Figure 5.22 (A) Sagittal T1 TR687/TE10. (B) Sagittal T2 TR 2693/TE110. (C) Parasagittal T1 TR6887/TE10. There is a fracture dislocation of the
cervical spine due to a suboptimally treated pedicle fracture. Despite the grade III anterolisthesis there is just mild cord compression. Also
note the dislocation of C67 due to a pars fracture and a jumped facet joint.

of the C2 pars interarticularis bilaterally, with variable vertebral body with small anterior avulsion fragment;
degrees of extension into the posterior elements and retropulsion of the posterior vertebral body fragment;
anterior displacement of the C2 vertebral body relative narrowing of the intervertebral disc space; and wid-
to the C3 vertebral body; it is unstable74,75 ening of the interlaminar and interspinous distances
Flexion-teardrop fractures are severe injuries that compatible with extensive posterior ligamentous dis-
76 are a product of hyperlexion. A constellation of ind- ruption; it is unstable. Resultant cord contusion and
ings is demonstrated: compression of the anterior hemorrhage are common.7678
Chapter 5: Spine imaging

Hyperlexion with rotation may result in a unilat-


erally locked facet as a result of apophyseal joint liga-
mentous rupture and facet joint dislocation, generally
considered stable. Bilateral locked facets occur in the set-
ting of extreme lexion and distraction forces. he result
is anterior dislocation of the bilateral inferior facets in
relation to subjacent superior facets with ligamentous
and capsular rupture; it is unstable.79 (Figs. 5.21, 5.22)

Thoracolumbar spine injuries


he most commonly encountered thoracic and lumbar
spine fracture is the compression or wedge fracture.
Making up nearly 50% of all thoracolumbar fractures,
these injuries are considered stable.80
Seat belt injuries are the result of forceful lexion
about a fulcrum, i.e., seat belt, with associated verte-
bral body compression fracture and variable degrees of
posterior element. Type I (Chance fracture) involves
posterior osseous elements; type II (Smith fracture)
involves posterior ligaments; and type III involves rup-
ture of the annulus ibrosis.80
Spondylolysis is a special scenario, encountered in
the lower lumbar spine in which there is a defect in the
pars interarticularis. When bilateral, instability may
result in anterior displacement or spondylolisthesis of
the vertebral body. he speciic etiology remains a sub-
ject of debate, with speculation about congenital origin
versus sequela of recurrent trauma.
Pedicle fractures are among the least common
injury to the spine. hey can be the result of chronic
or acute trauma and they can be unilateral or bilateral. Figure 5.23 Sagittal T2 image showing a chronic pedicle fracture
Depending on their age they may show edema or fatty at L5 with a small amount of edema in the adjacent bone.
marrow change. Like spondylolisthesis they result in
instability81 (Fig. 5.23) echo) for identiication of cord hemorrhage, fractures,
Sacral insuiciency fractures can be diicult to and disc protrusions; axial T2*GRE for evaluation of
diagnose because they can present as hip pain or lower graywhite diferentiation. STIR (short tau inversion
back pain. MRI is very sensitive on T1W sequences recovery) may also be used to increase the conspicuity
(Fig. 5.24), where the dark marrow edema is outlined of edema.
by the normal bright fatty marrow. Extensive investigation has been performed into
spinal cord injury (SCI) patterns presenting neuro-
MRI and spine trauma logic deicit on MRI and clinical outcomes. Kulkarni
Speciic imaging protocols vary to some degree et al. irst proposed an approach to predicting neuro-
between institutions. In general, commonly used pulse logic outcomes based on three patterns of acute SCI on
sequences include: sagittal SE T1 (spin echo) for dein- MRI including spinal cord hemorrhage (type I), spinal
ition of anatomy and identiication of disc herniations, cord edema (type II), and mixed edema and hemor-
epidural luid collections, subluxations, vertebral body rhage (type III).82 Subsequently, Schaefer et al. added
fractures, cord swelling, and cord compression; sagittal the assessment of injury segment size.83 Additional fea-
FSE/TSE T2 (fast-spin echo/turbo-spin echo) with fat tures of cord compression, cord swelling, and persistent
saturation allowing for characterization of spinal cord cord signal abnormality in follow-up evaluation were
edema, cord hemorrhage, ligamentous injury, disc her- found to be of prognostic signiicance by Yamashita 77
niation, and epidural luid; sagittal T2*GRE (gradient et al.84,85 In general, the presence and extent of cord
Section 1: General considerations

Figure 5.24 Sacral insuiciency. (A) Axial T1 TR400/TE13. (B) Coronal T1 TR425/TE13. (C) Sagittal T1 TR529/TE11. (D) Sagittal STIR TR4820/
TE90. The T1 weighted sequences show the edema from the acute bilateral sacral insuiciency fracture as areas of decreased signal in this
osteoporotic 79-year-old patient. The parasagittal STIR shows the edema as an area of hyperintensity.

hemorrhage is a good indicator of prognosis. Less pre- Careful evaluation of the anterior and posterior
dictable is the secondary cord injury that results from a longitudinal ligaments, supraspinous, nuchal, and
cycle of edema and ischemia.86 interspinous ligaments, as well as the ligamentum
Spinal cord edema as a result of contusion is char- lavum, should be performed on all trauma patients.
acterized by T1 isointentsity to hypointensity with con- Ligamentous tear is depicted as discontinuity of the
comitant T2 hyperintensity. Spinal cord hemorrhage, anticipated normal linear T1 signal hypointensity with
most commonly localizing to the central gray matter at associated sot tissue changes. Adjacent T2-hyperi-
the level of injury, is initially T1 isointense with T2 and ntense edema and hemorrhage should prompt liga-
GRE relative hypointensity to the surrounding cord mentous scrutiny. Partial and intrasubstance tears may
signal. Subsequently, intramedullary hemorrhage of appear as linear T2 signal hyperintensity.
the spinal cord undergoes a typical pattern of degrad- Traumatic disc extrusions are variable in appear-
ation similar to that of intracranial hemorrhage, but ance based on the presence or absence of associated
over a greater period of time due to diferences in cord hemorrhage. If no associated hemorrhage is present,
78 perfusion and hypoxia. Cord transection may have the fragment will have signal intensity similar to the
intercalated hemorrhage. native disc, whereas, hemorrhage may result in T2
Chapter 5: Spine imaging

Figure 5.25 Cord hemorrhage. (A)


Sagittal STIR TR4040/TE95. (B) Axial
gradient echo TR30TE15. Note the
punctate susceptibility artifact (dark)
in the anterior aspect of the cord with
surrounding edema on both sagittal
and axial images in this anticoagulated
patient presenting with paraplegia.

signal hyperintensity of the fragment in comparison neurulation (weeks 34), and secondary neurulation
with the native disc. Identiication is important as open (weeks 56). hese anomalies were classically char-
reduction may be necessary in the setting of cervical acterized speciically by developmental embryologic
facet dislocations to prevent increased neurologic origin.his is a complex strategy with limited clin-
compromise. ical application. he current widely accepted model,
Epidural hematoma presents as a hypointense developed and introduced by Tortori-Donati and
T2*GRE focus adjacent to the cord with isointense T1 Rossi, is a clinical-neuroradiological classiication
signal. It may complicate cord injury or lead directly to strategy relying upon only a few fundamental key fea-
cord ischemia as a result of mass efect. tures: (1) clinical classiication of the dysraphism as
Cord hemorrhage can be best detected using a gra- open (exposed to environment through osseous and
dient echo sequence or T2W image in the acute stage, skin defect) or closed (covered with skin); (2) closed
where the hemorrhage appears as a dark area. In suba- dysraphisms are subsequently divided into those
cute stage the hemorrhage is bright on T1 and T2 (Fig. entities with an associated subcutaneous mass ver-
5.25). sus those without a mass; and (3) inally, those closed
Insensitivity of MRI to acute fracture makes it defects without associated mass are further divided
inappropriate as the initial imaging modality in the set- into simple and complex.87
ting of suspected osseous injury. A solitary exception is
the utility of MRI in the characterization of vertebral
body fractures as acute or chronic. Hemorrhagic blood Open spinal dysraphisms (OSD)
products and marrow edema are relatively conspicu- Open dysraphic defects account for 85% all cases and
ous in the trabecular bone of the vertebral body. are commonly diagnosed by ultrasound (US) or fetal
Carotid dissection and vertebral artery thrombo- MRI following identiication of elevated maternal
sis occur to some degree in the setting of signiicant alpha fetoprotein. Nearly all are associated with Chiari
trauma and should be considered. Cervical vascular II malformations and all exhibit neural placodes.
injury evaluation with CTA, MRA, Doppler sonogra- Neural placodes are a splayed-out segment of embryo-
phy, or catheter angiography may be considered. nal neural tissue which fails to transform through a
process of bending and folding (neurulation) into the
Congenital anomalies of the spine neural tube. Myelomeningoceles make up the over-
whelming majority of open defects at 98%, character-
and spinal cord ized by protrusion portions of spinal cord and nerve
Congenital anomalies of the spine and spinal cord are roots (neural placode) beyond the skin. Myeloceles
generically classiied as spinal dysraphisms. he dys- are second most common, exhibiting a placode that
raphisms, or defective fusion of parts that normally is lush with the skin. Hemimyeloceles and hemimy-
unite, comprise a group of anomalies that result from elomeningoceles are rare entities with features similar
speciic aberrations of the embryologic stages of spi- to those above but associated with diastematomyelia 79
nal development: gastrulation (weeks 23), primary or cord splitting.88
Section 1: General considerations

Closed spinal dysraphisms (CSD) as congenital scoliosis. Segmental spinal dysgenesis is a


less common entity in which an isolated level, usually
with a mass lumbar or thoracic, is afected by the failure of osse-
Deined as a closed dysraphic defect with associated ous, central cord, and nerve root development. Caudal
subcutaneous mass. he majority of masses are local- regression represents a similar total or partial failure
ized to the lower lumbar spine above the intergluteal of spinal column development, oten with associated
fold. Lipomyeloceles and lipomyelomeningoceles developmental anomalies of multiple systems.88
exhibit a placode-lipoma morphology. he result of
invasion the neural tube defect by mesenchymal tissue
that is subsequently induced to form adipose tissue.
Downs syndrome and spinal
he lipomyelocele placodelipoma interface resides abnormalities
within the spinal canal, whereas the lipomyelomenin- Downs syndrome is associated with several spinal
gocele placodelipoma interface is outside of the canal. abnormalities. hese include atlanto-occiptal insta-
Meningoceles are CSF-illed herniated meningeal sacs bility, scoliosis, and cervical spondylosis.89 he most
without cord protrusion. Terminal myelocystoceles are prevalent is atlanto-occipital instability. his can be
meningoceles that contain a herniated terminal syrinx. identiied radiographically in children as widening
Myelocystoceles (nonterminal) are meningoceles that of the anterior atlanto-odontoid distance (AAOD)
contain a dilated herniated central canal and occur greater than 5 mm. his is felt to be caused by pos-
more commonly in the cervicothoracic region. terior transverse ligament laxity.90 his widening of
the AAOD can cause compression of the brainstem by
Closed spinal dysraphisms (CSD) the dens. he frequency is reported as high as 30% in
patients with Downs syndrome.91 However, the per-
without a mass: simple centage of patients who are clinically symptomatic is
Intradural lipoma is simply a lipoma dorsally located much lower. Given the large percentage of patients
within the dural sac. heir presence should prompt who have radiographic instability and yet the paucity
close inspection for tethered cord. Tethered cord is of Downs syndrome patients who are clinically symp-
deined as low-lying conus medullaris below L23 with tomatic, there has been controversy whether there is
associated shortened and thickened ilum terminale >2 a need to screen patients who might be engaged in
mm. he persistent terminal ventricle is a cystic cavity sports or the Special Olympics.92,93 Cervical plain ilms
immediately above the ilum terminale that does not are currently a recommended part of the childhood
enhance and is generally incidental, with no clinical screening by the American National Down Syndrome
signiicance. he dermal sinus is an epithelium-lined Society (1999), but are deemed to be too unreliable
tract connecting neural tissues/meninges with skin for screening purposes by the UK and Ireland Downs
surfaces.88 Syndrome Medical Interest Group.94,95

Closed spinal dysraphisms (CSD) Scoliosis


without a mass: complex Scoliosis is curvature of the spine in the coronal plane.
A dorsal enteric istula is the result of abnormal per- Scoliosis is a three-dimensional phenomenon and has
sistent communication between skin and bowel. he a rotatory component associated with it that can be
neurenteric cyst occurs anterior to the cervicothoracic determined by observing the placement of the pedicles
spinal cord and is a mucin-secreting lined cyst similar in relation to the vertebra. It can be subdivided by cause
to gastrointestinal tract elements. Disorders of noto- (idiopathic, congenital, developmental, neuromuscu-
chordal formation include segmental spinal dysgenesis lar, and tumor associated) as well as age. Idiopathic
and caudal agenesis. Misallocation of notochord cells scoliosis accounts for 80% of the cases of scoliosis.96
results in varying degrees of segmental deicits and Scoliosis is described by the apex of the curvature
less commonly dysgenesis. Wedge vertebra, bony bars/ and the degree of curvature. he apex is given as the
block vertebra, butterly vertebra, and hemivertebra level (vertebral body) and the side to which it is pointed
represent a few of the osseous abnormalities encoun- (let or right). Cobb angle is the measurement used to
80 tered. When associated with pathologic curvature and characterize scoliosis. A Cobb angle of 10 or greater is
alignment of the spine, the phenomena may be classiied typically used as the cut-of for determining scoliosis.
Chapter 5: Spine imaging

Features common to nearly all spinal cord neo-


plasms include: a tendency to enlarge the cord focally
or difusely, T2/PD signal hyperintensity, and enhance-
ment. Once a mass is identiied, subsequent classi-
ication using location allows for greater accuracy in
generating a practical diferential diagnosis: specif-
ically, intramedullary (within the spinal cord), intra-
dural/extramedullary, and extradural.

Intramedullary neoplasms
Intramedullary neoplasms directly involve/origi-
nate from the spinal cord and account for nearly 25%
of all tumors of the spine. he majority (9095%) of
intramedullary neoplasms are malignant.97 Adults
most commonly develop ependymomas and children
astrocytomas. hese two cell types alone make up more
than 70% of intramedullary malignancies.

Ependymoma
Figure 5.26 AP scoliosis ilm measuring the Cobb angle. This
Ependymomas account for 60% of all glial-based
patient has a levoscoliosis with the apex at L1. Note the lines intramedullary/ilum tumors.97 Multiple histologic
overlying the end plates of L4 and T11. These are drawn on the end subtypes exist; most common overall is the cellular
vertebrae, which are the vertebrae demonstrating maximum tilt
in relation to the horizontal plane. The angle formed by the lines
type, arising from ependymal cells lining the central
perpendicular to the end vertebrae is the Cobb angle. canal as well as nests of ependymal cells at the ilum
and sacral regions and occurring with greatest fre-
he Cobb angle is obtained from an AP ilm by quency in the cervical spine. he ilum terminale is
drawing lines parallel to the end plates of the end ver- afected by the myxopapillary subtype, which is
tebral bodies which are at the inlection points of the complicated more commonly by hemorrhage, and by
superior and inferior curves (the end plates which deinition intradural extramedullary. Ependymomas
produce the greatest tilt in relation to horizontal.) are glial in origin, tend to be central in position, and
hese end vertebrae also show the least rotation. If exhibit sharp margins. Cystic changes and hemorrhage
the end plates are diicult to see, a line can be drawn are also observed with moderate frequency. Generally
through the pedicles and that can be used to deter- T1 iso/hypointense, patchy heterogeneous T2 hyperin-
mine what are the end vertebrae. he angle formed tense, T2 hypointense; cap sign is occasionally noted,
from the lines drawn along the superior and inferior the result of hemosiderin staining and development of
end vertebra (or lines perpendicular to them) is the a pseudocapsule on gross specimen. Both homogenous
Cobb angle (Fig. 5.26) and heterogeneous contrast enhancement occur com-
monly, sometimes in the typical cyst with mural nod-
ule coniguration (Fig. 5.27).
Neoplastic disease of the spine
High-contrast sensitivity and multiplanar capability Astrocytoma
make MRI the study of choice when evaluating neo- Astrocytomas are the second most common adult
plastic disease of the spinal cord. Standard T1W pre- intramedullary tumor and the most common
contrast, T1W postcontrast with fat saturation, and intramedullary tumor in children.98 Generally low
T2W pulse sequences are acquired. Addition of GRE grade and iniltrative, astrocytomas demonstrate
increases the conspicuity of cord hemorrhage and more ill-deined margins and eccentric orientation in
STIR allows for sensitive assessment of sites of edema. the cord as opposed to the sharply delineated central
Osseous spinal abnormalities continue to be best morphology of ependymomas. Intratumoral cysts are
evaluated by CT due to the inherent high sensitivity to not uncommon and are sometimes associated with 81
alterations of bone mineralization. proximal or distal syrinx in the pilocytic type. horacic
Section 1: General considerations

Figure 5.27 Ependymoma.


Postcontrast sagittal (A) and axial (B)
T1 TR773/TE10 MRI showing difuse
enhancement and compression of the
conus medullaris which is displaced
anterior on the axial images.

A B

Figure 5.28 Astrocytoma. (A) Sagittal T1 TR681/TE10. (B) Sagittal T2 TR3770/TE113. (C) Sagittal T1 postcontrast TR681/TE10. Nonenhancing
expansile mass on postcontrast image (C) relects a low-grade thoracic spine astrocytoma. Note the increased signal on T2 weighted images.

spine is involved only slightly more frequently than isointensity, T2 hyperintensity, avid enhancement,
cervical spine. here is a minimal male predominance. low voids 99 (Fig. 5.30)
MRI examination reveals mild fusiform widening of
the cord, T1 iso/hypointensity, avid enhancement, and Metastatic disease
T2 signal hyperintensity (Figs. 5.28, 5.29). Lung, breast, colon, lymphoma, and kidney carcino-
mas may all infrequently metastasize to the intramed-
Hemangioblastoma ullary spinal cord but are found in only 2% of cancer
Although predominantly intramedullary and involv- patients at autopsy. MRI appearance: T1 hypointense,
ing the dorsal spinal cord and posterior fossa, T2 hyperintense, avidly enhancing.97 hey can appear
hemangioblastomas may also infrequently involve as enhancing intramedullary lesions, intradural
extramedullary structures. Overall, they account for extramedullary masses. or as leptomeningeal disease
less than 17.2% of cord tumors. he thoracic spine (Figs. 5.31, 5.32)
is afected slightly more frequently than the cervical
spine. he vast majority of lesions are solitary, as many Ganglioglioma
82 as 80%. Approximately 1/3 of cases are associated with Gangliogliomas represent neoplastic ganglion and glial
von HippelLindau syndrome. MRI appearance: T1 cell proliferation resulting in long segment expansion
Chapter 5: Spine imaging

Figure 5.29 Two axial gradient echo


(TR94TE34) images from astrocytoma
above the level of the tumor (A) and at
the level of the tumor in the thoracic
spine (B). Note the marked diference in
the caliber of the cord.

Figure 5.30 Hemangioblastoma. (A)


Sagittal T1 TR700/TE9 postcontrast. (B)
Sagittal T2 TR3000/TE87. Cystic mass
at the craniocervical junction with
enhancing mural nodule.

Figure 5.31 Postcontrast T1 images


TR769/TE12 through the thoracolumbar
junction showing difuse enhancement
in the subarachnoid space relecting
leptomeningeal metastasis. The axial
image (A) shows difuse enhancement
outlining the conus medullaris.

of the cord with associated cystic components. he and scoliosis. MRI features demonstrated are typic-
intramedullary tumor predominantly afects children ally mixed isohypointensity T1 signal, heterogeneous
and young adults. Commonly afecting the cervical and isohyperintensity T2, patchy enhancement, and cystic
upper thoracic cord, the tumor exhibits slow, expan- changes. Calciications, when present help in distin- 83
sive growth resulting in osseous erosions/remodeling guishing from astrocytoma.9799
Section 1: General considerations

Intradural extramedullary neoplasms meningiomas that occur below the level of C7 are most
commonly posterior to the cord. An association with
Meningioma neuroibromatosis II should prompt evaluation for
Meningioma is the second most common primary other abnormalities (Fig. 5.33).
intraspinal tumor.97 It most commonly afects the
thoracic spine followed by the cervical spine. MRI
Schwannoma
demonstrates T1 isointensity, T2 isointensity, and Schwannoma is the most common intraspinal
intense homogeneous enhancement. Dural tail and/ tumor.97 It is an eccentric, exophytic, nerve sheath tumor
or adjacent dural thickening are classic features. hose of Schwann cell origin. Generally afecting the cervi-
cal region, schwannomas oten have both extradural
and intradural components in a so-called dumbbell
shape. MRI indings include: T1 isohypointensity, T2
isohyperintensity, moderate to marked homogenous
enhancement, occasionally target enhancement with
lower central intensity (Fig. 5.34).

Neurofibroma
Neuroibromas are an iniltrative mixture of Schwann
cells and ibroblasts and are associated neuroibroma-
tosis I. MRI indings include: T1 isohypointensity, T2
hyperintensity, and marked homogenous enhance-
ment.76 Neuroibromas may be solitary or multiple
nodular masses.

Myxopapillary ependymoma
Myxopapillary ependymomas account for 83% of
ilum tumors.99 hey are generally slow growing, lead-
Figure 5.32 Noncontrast sagittal T1 TR500/TE11 image showing ing to vertebral scalloping and canal enlargement. MRP
pathologic compression fracture from endometrial carcinoma
metastatic disease. Note that there is mild dorsal displacement of the
indings include: isointense T1W and hyperintense T2W
posterior wall and a small amount of epidural tumor dorsal to L2. with avid enhancement. heir highly vascular nature

Figure 5.33 Sagittal (A) and axial (B)


postcontrast T1 MRI images through the
cervical spine showing an enhancing
dural-based meningioma compressing
the cord at the level of the craniocervical
junction.

84
Chapter 5: Spine imaging

Figure 5.34 Postcontrast T1 images


TR700/TE9 in the lower cervical spine
showing enhancing paraspinous masses
following the nerve roots consistent with
difuse schwannoma.

A B

commonly leads to hemorrhage. Classic cap type T2 2. Harnsberger HR, Osborn A, Macdonald AJ. Diagnostic
hypointensity from hemosiderin and formation of a and Surgical Imaging Anatomy Brain, Head & Neck,
pseudocapsule are occasionally demonstrated. Spine. Salt Lake City, UT: Amirsys; 2006.
3. Modic MT, Masaryk TJ, Ross JS. Magnetic Resonance
Intrathecal metastases: drop mets Imaging of the Spine. 2nd ed. St. Louis, MO: Mosby Year
Book; 1994.
Drop metastases represent seeding of the subarach-
noid space, classically as the result of migration of 4. Fox MW, Ootrio BM, Kilgore JE. Neurologic
complications of ankylosing spondylitis. J Neurosurg
exfoliated tumor cells from primary CNS tumors such 1993; 78: 8718.
as astrocytoma, medulloblastoma, ependymoma, and
5. Matsui H, Kanamori M, Ishihara H, et al. Familial
pineal neoplasms. Non-CNS primary malignancy such
predisposition for lumbar degenerative disc disease: a
as breast, lung, leukemia, and lymphoma also metasta- casecontrol study. Epidemiology 1998; 23: 102934.
size to the leptomeninges. MRI will demonstrate either
6. Powell M, Szypryt P, Wilson M, et al. Prevalence of
multiple enhancing nodules or difuse thickening and
lumbar disc degeneration observed by magnetic
enhancement of the meninges. resonance in symptomless women. Lancet 1986; 328:
13667.
Extradural neoplasms 7. Weishaupt D, Zanetti M, Hodler J, et al. MR imaging
he majority of extradural neoplasms are metastatic in of the lumbar spine: prevalence of intervertebral disk
etiology with less than 5% being primary bone tumors.87 extrusion and sequestration, nerve root compression, end
While discussion of primary bone tumors is beyond the plate abnormalities, and osteoarthritis of the facet joints
scope of this chapter, brief discussion of metastatic mar- in asymptomatic volunteers. Radiology 1998; 209: 6616.
row replacement is in order. As the marrow of the ver- 8. Kramer J. Intervertebral disc disease, causes, diagnosis,
tebral bodies is exposed to a large volume of blood low, treatment and prophylaxis. 2nd ed. New York: George
implantation of malignant cells is common. Growth and hieme Verlag; 1990: 1447.
proliferation of the metastatic foci ultimately results in 9. Nomenclature and Classiication of Lumbar Disc
replacement of the normal T1 bright fatty marrow sig- Pathology. http://www.asnr.org/spine_nomenclature/
nal with T1 signal hypointensity. he accompanying (accesssed December 1, 2011).
T2 signal is generally hyperintense but may be variable. 10. Modic MT, Steinberg PM, Ross JS, et al. Degenerative
Postcontrast imaging is of minimal beneit, but addition disk disease: assessment of changes in vertebral body
of inversion recovery sequences has great utility in exag- marrow with MR imaging. Radiology 1988; 166: 1939.
gerating the conspicuity of pathologic marrow signal. 11. Liu SS, Williams KD, Drayer BP, Spetzler RF, Sonntag
VK: Synovial cysts of the lumbosacral spine: diagnosis
References by MR imaging. AJR 1990; 154: 1639.
1. Mody MG, Nourbakhsh A, Stahl DL, et al. he 12. Hiramatsn Y, Nobechi T. Calciication of posterior
prevalence of wrong level surgery among spine longitudinal ligament of spine among Japanese. 85
surgeons. Spine 2008; 33(2): 1948. Radiology 1971; 100: 30712.
Section 1: General considerations

13. Resnick D, Guerra Jr J, Robinson CA, Vint VC. 30. Numaguchi Y, Rigamonti D, Rothman MI, et al. Spinal
Association of difuse idiopathic skeletal hyperostosis epidural abscess: evaluation with gadolinium-enhanced
(DISH) and calciication and ossiication of the MR imaging. Radiographics 1993; 13: 54559.
posterior longitudinal ligament. AJR 1978: 131(6): 31. Sandhu FS, Dillon WP. spinal epidural abscess:
104953. evaluation with contrast-enhanced MR imaging. AJNR
14. Weinfeld RM, Olson PN, Maki DD, Griiths HJ. he 1991; 12: 108793.
prevalence of difuse idiopathic skeletal hyperostosis 32. Nussbaum ES, Rigamonti D, Standiford H, et al. Spinal
(DISH) in two large American Midwest metropolitan epidural abscess: a report of 40 cases and review Surg
hospital populations. Skeletal Radiol 1997; 26(4): 2225. Neurol 1992; 38: 22531.
15. Resnick D. Degenerative diseases of the vertebral 33. Kricun R, Shoemaker EI, Chovanes GI, Stephens HW.
column. Radiology 1985; 156: 314. Epidural abscess of the cervical spine: MR indings in
16. Smith JK, et al. Imaging manifestations of ive cases. AJR 1992; 158: 11459.
neurosarcoidosis. AJR Am J Roentgenol 2004; 182(2): 34. Bartels RH, deJong TR, Grotenhuis JA. Spinal subdural
28995. abscess. J Neurosurg 1992; 76: 30711.
17. Braun J, Bollow M, Sieper J. Radiologic diagnosis and 35. Gero B, Sze G, Sharif H. MR imaging of intradural
pathology of the spondyloarthropathies. Rheum Dis inlammatory diseases of the spine. AJNR 1991; 12:
Clin North Am 1998; 24: 697703. 100919.
18. Fox MW, Ootrio BM, Kilgore JE, Neurologic 36. Scotti G, Righi C, Campi A. Myelitis and myelopathies.
complications of ankylosing spondylitis, J Neurosurg Riv di Neuroradiol 1992; 5(suppl 2): 4952.
1993; 78: 8718.
37. DeLaPaz R. Demyelinating disease of the spinal cord.
19. Sharif HS. Role of MR imaging in the management of In: Enzmann D, DeLaPaz R, RubinJ, eds. Magnetic
spinal infections. AJR 1992; 158: 133345. Resonance of the Spine. St. Louis: Mosby ; 1990; 42336.
20. Mark AS. MRI of infections and inlammatory 38. Osborn AG, Harnsberger HR, Smoker WRK, Boyer R.
diseases of the spine. MRI Decisions 1991 March/ Multiple sclerosis in adolescents: CT and MR indings,
April: 1226. AJNR 1990; 12: 5214.
21. Brant-Zawadzki M. Infections. In: Newton TH, Potts 39. Larsson E-M, Holtas S, Nilsson O. Gd-DTPA-enhanced
DG, eds. Modern Neuroradiology, vol. 1: Computed MR of suspected spinal multiple sclerosis. AJNR 1989;
Tomography of the Spine and Spinal Cord. San Anselmo, 10: 10716.
CA: Clavedel Press; 1983: 20529.
40. Holtas S, Basibuyuk N, Frederiksson K. MRI in acute
22. hrush A, Enzmann D. MR imaging of infectious transverse myelopathy, Neuroradiol 1993; 35: 2216.
spondylitis. AJNR 1990; 11: 117180.
41. Austin SG, Zee C-S, Walters C. he role of magnetic
23. Sklar EML, Post MJD, Lebwohl NH. Imaging of Infection resonance imaging in acute transverse myelitis. Can
of the lumbosacral spine. Neuroimaging 1993; 3: 57790. J Neurol Sci 1992; 19: 50811.
24. Post MJD, Sze G, Quencer RM, et al. Gadolinium- 42. Zweig G, Russell EJ. Radiation myelopathy of the
enhanced MR in spinal infection. J Comput Assist cervical spinal cord: MR indings. AJNR 1990; 11:
Tomogr 1990; 14: 7219. 118890.
25. Resnick D. Inlammatory diseases of the vertebral 43. Michikawa M, Wada Y, Sano M, et al. Radiation
column. Categorical course on Spine and Cord imaging. myelopathy: signiicance of gadolinium-DTPA
American Society of Neuroradiology; 1988: 5149. enhancement in the diagnosis. Neuroradiol 1991; 33:
26. Sharif HS, Aideyan OA, Clark DC, et al.. Brucellar and 2869.
tuberculosis spondylitis: comparative imaging features. 44. Wang P-Y, Shen W-C, Jan J-S. MR imaging in radiation
Radiology 1989; 171: 41925. myelopathy. AJNR 1992; 13: 104955.
27. Smith AS, Weinstein MA, Mizushima A, et al. MR 45. Takahashi M, Yamashita Y, Sakamoto Y, Kojima R.
imaging of characteristics of tuberculosis spondylitis vs. Chronic cervical cord compression: clinical signiicance
vertebral osteomyelitis. AJNR 1989; 10: 61925. of increased signal intensity on MR images. Radiol
1989; 173: 219224.
28. Arabi KM, Al Sebai MW, Al Chakaki M. Evaluation of
radiological investigations in spinal tuberculosis. Int 46. Levy M, Turtzo C, Llinas RH. Supericial siderosis: a
Orthop (SICOT) 1992; 16: 1657. case report and review of the literature. Nat Clin Pract
Neurol 2007; 3: 548.
29. Boxer DI, Pratt C, Hine AL, McNicol M. Radiological
86 features during and following treatment of spinal 47. Ross JS, Masaryk TJ, Modic MT, et al. MR imaging of
tuberculosis. Br J Radiol 1992; 65: 4769. lumbar arachnoiditis. AJR 1987; 149: 102532.
Chapter 5: Spine imaging

48. Rodesch G, Lasjaunias P, Berenstein A. Embolization 64. Blackmore CC, Ramsey SD, Mann FA, et al. Cervical
of spinal cord arteriovenous malformations. Riv di spine screening with CT in trauma patients: a cost
Neuroradiol 1992; 5(suppl 2): 6792. efectiveness analysis. Radiology 1999; 212: 11725.
49. Anson JA, Spetzler RF. Classiication of spinal 65. Nunez DB, Quencer RM. he role of helical CT in the
arteriovenous malformations and implications for assessment of cervical spine injuries. Am J Roentgenol
treatment. BNI Quarterly 1992; 8: 28. 1998; 171: 9517.
50. Nichols DA, Rufenacht DA, Jack CR Jr, Forbes GA. 66. Nunez DB, Zuluaga A, Fuentes Bernardo DA, et al.
Embolization of spinal dural arteriovenous istula with Cervical spine trauma; how much do we learn by
polyvinyl alcohol particles: experience in 14 patients. routinely using helical CT? Radiographics 1996; 16:
AJNR 1992; 13: 93340. 130718.
51. Beaujeux RL, Reizine DC, Casasco A, et al. 67. Woodring JH, Lee C. he role and limitations of
Endovascular treatment of vertebral arteriovenous computed tomography scanning in the evaluation of
istula. Radiology 1992; 183: 3617. cervical trauma. J Trauma 1992; 33(5): 698708.
52. Naidich TP, McLone DG, Harwood-Nash DC. Vascular 68. Denis F. he three column spine and its signiicance in
malformations. In: Newton TH, Potts DG, eds. Modern the classiication of acute throacolumbar spinal injuries.
Neuroradiology, vol 1: Computed Tomography of the Spine 1983; 8: 817.
Spine and Spinal Cord. San Anselmo, CA: Clavedel
69. Denis F. Spinal instability as deined by the three
Press; 1983: 397400.
column spine concept in acute spinal trauma. Clin
53. Ogilvy CS, Louis DN, Ojemannn RG. Intramedullary Orthop 1984; 189: 65.
cavernous angiomas of the spinal cord: clinical
presentation, pathologic features, and surgical 70. Han SY, Witten DM, Mussleman JP. Jeferson fracture of
management. Neurosurgery 1992; 31: 21930. the atlas. Report of six cases. J Neurosurg 1976; 44: 368.

54. Anson JA, Spetzler RF. Surgical resection of 71. Keterson L, Benzel E, Orrison W, Coleman J. Evaluation
intramedullary spinal cord cavernous malformations. and treatment of atlas burst fractures (Jeferson
J Neurosurg 1993; 78: 44651. fractures). J Neurosurg 1991; 75: 213.

55. Bourgouin PM, Tampieri D, Johnston W, et al. Multiple 72. Moore KR, Frank EH. Traumatic atlantoaxial rotatory
occult vascular malformations of the brain and spinal subluxation and dislocation. Spine 1995; 20: 1928.
cord: MRI diagnosis. Neuroradiology 1992; 34: 11011. 73. Anderson LD, DAlonzo RT. Fractures of the odontoid
56. Friedman DP, Flanders AE. Enhancement of gray process of the axis. J Bone Joint Surg Am 1974; 56: 1663.
matter in anterior spinal infarction. AJNR 1992; 13: 74. Fielding JW, Francis WR Jr, Hawkins RJ, et al.
9835. Traumatic spondylolisthesis of the axis. Clin Orthop
57. Yuh WTC, March EE, Wang AK, et al. MR Imaging of 1989; 239; 47.
spinal cord and vertebral body infarction. AJNR 1992; 75. Mirvis SE, Young JW, Lim C, Greenberg J. Hangmans
13: 14554. fracture: radiologic assessment in 27 cases. Radiology
58. Takahashi S, Yamada T, Ishii K, et al. MRI of anterior 1987; 163: 713.
spinal artery syndrome of the cervical spinal cord. 76. Scher AT. Tear-drop fractures of the cervical spine
Neuroradiology 1992; 35: 259. radiological features. S Afr Med J 1982; 61: 355.
59. Mikulis DJ, Ogilvy CS, McKee A, et al. Spinal cord 77. Schneider R, Kahn E. Chronic neurologic sequelae of
infarction and ibrocartilagenous emboli. AJNR 1992; acute trauma to the spine and spinal cord. Part 1: the
13: 15560. signiicance of the acute lexion or tear-drop fracture-
60. Hirono H, Yamadori A, Komiyama M, et al. MRI of dislocation of the cervical spine. J Bone Joint Surg Am
spontaneous spinal cord infarction: serial changes in 1956; 38: 985.
gadolinium-DTPA enhancement. Neuroradiol 1992; 34: 78. Kim KS, Chen HH, Russell EJ, Rogers LF. Flexion
957. teardrop fracture of the cervical spine: radiographic
61. Enzmann DR. Vascular diseases. In: Enzmann DR, characteristics. Am J Roentgenol 1989; 152: 319.
DeLaPaz R, Rubin J, eds. Magnetic Resonance Imaging of 79. Shanmuganathan K, Mirvis SE, Levine AM. Rotational
the Spine. St. Louis, MO: Mosby ; 1990: 51037. injury of cervical facets: CT analysis of fracture patterns
62. [No authors listed]. Spinal cord injury: facts and igures with implications for management and neurologic
at a glance. J Spinal Cord Med 2000; 23(1): 513. outcomes. Am J Roentgenol 1994; 163: 1165.
63. Pope A, Tarlov AR. Disability in America: Toward a 80. Rogers LF. he roentgenographic appearances of
National Agenda for Prevention. Washington, DC: transverse or chance fractures of the spine: the seat belt 87
National Academy Press; 1991. fracture. Am J Roentgenol 1971; 111: 8449.
Section 1: General considerations

81. Guillodo Y, Botton E, Saraux A, et al. Contralateral Orthopaedic Society Survey: the Israeli sample.
spondylolysis and fracture of the lumbar pedicle in an J Pediatr Orthop B. 2000; 9(3): 18592.
elite female gymnast: a case report. Spine 2000; 25(19): 91. Alvarez N, Rubin L. Atlantoaxial instability in adults
25413. with Down syndrome: a clinical and radiological
82. Kulkarni MV, McArdle CB, Kopanicky D, et al. Acute survey. Appl Res Ment Retard 1986; 7(1): 6778.
spinal cord injury: MR imaging at 1.5 T. Radiology 1987; 92. American Academy of Pediatrics. Committee on Sports
164: 83743. Medicine. Atlantoaxial instability in Down syndrome.
83. Schaefer DM, et al. Magnetic resonance imaging of Pediatrics 1984; 74(1): 1524.
acute cervical spine trauma. Correlation with severity 93. Pueschel SM, Scola FH, Perry CD, Pezzullo JC. Atlanto-
of neurologic injury. Spine 1989; 14(10): 10905. axial instability in children with Down syndrome.
84. Yamashita Y, et al. Acute spinal cord injury: magnetic Pediatr Radiol 1981; 10(3): 12932.
resonance imaging correlated with myelopathy. Br J 94. Down Syndrom Health Care Guidelines.
Radiol 1991; 64(759): 2019. http://www.dsacc.org/downloads/healthcare/
85. Yamashita Y, et al. Chronic injuries of the spinal cord: dshealthcareguidelines.pdf. New York: National
assessment with MR imaging. Radiology 1990; 175(3): Down Syndrome Society (accessed
84954. December 1, 2011).
86. Leypold BG, Flanders AE, Burns AS. he early evolution 95. Basic Medical Surveillance Essentials for People with
of spinal cord lesions on MR imaging following Downs Syndrome. http://www.dsmig.org.uk/library/
traumatic spinal cord injury. AJNR 2008; 29: 101216. articles/guideline-cervsp-4.pdf. Nottingham, UK:
87. Tortori-Donati P, Rossi A, Cama A. Spinal dysraphism: Downs Syndrome Medical Information Services
a review of neuroradiological features with (accessed December 1, 2011).
embryological correlations and proposal for a new 96. Kim H, Kim HS, Moon ES, et al. Scoliosis imaging: what
classiication. Neuroradiology 2000; 42: 47191. radiologists should know. Radiographics 2010; 30(7):
88. Rufener SL, Ibrahim M, Raybaud CA, et al. Congenital 182342.
spine and spinal cord malformations pictorial review. 97. Bloomer CW, Ackerman A, Bhatia R. Imaging for spine
AJR 2010; 194: S26S37. tumors and new applications. Top Magn Reson Imaging
89. Bosma GP, van Buchem MA, Voormolen JH, van 2006; 17: 6987.
Biezen FC, Brouwer OF. Cervical spondylarthrotic 98. Karagianis A, Klufas R, Schwartz R. MRI of cervical
myelopathy with early onset in Downs syndrome: ive spine neoplasms. Appl Radiol 2003; 12: 2638.
cases and a review of the literature. J Intellect Disabil Res 99. Koeller K, Rosenblum R, Morrison AL, et al. Neoplasms
1999; 43(4): 2838. of the spinal cord and ilum terminale: radiologic-
90. Merrick J, Ezra E, Josef B, et al. Musculoskeletal pathologic correlation. Radiographics 2000; 20(6):
problems in Down Syndrome European Paediatric 172149.

88
Section 1 General considerations
Chapter
Evoked potential monitoring

6 Chakorn Chansakul and Dileep R. Nair

during surgery. he purpose of NIOM is to minimize


Key points
the risks of iatrogenic neurological deicits when
Neurophysiologic intraoperative monitoring is clinical examination is not possible by detecting early
the use of electrophysiological techniques changes in the neural structures prior to irreversible
to assess functional integrity of the nervous damage.
system during surgery, to prevent iatrogenic Various electrophysiological methods can be used
injuries to the neural structures. intraoperatively during spine surgery. Somatosensory
Somatosensory evoked potentials are used evoked potential monitoring and motor evoked potential
primarily for spinal cord monitoring. his monitoring are used primarily to monitor the functional
technique mainly assesses the function of integrity of the spinal cord, although both techniques
the proprioceptive sensory pathway in the also assess the entire neuraxis from the peripheral nerve
dorsolateral funiculus of the spinal cord; hence, level to the cerebral hemisphere. Continuous electro-
injury to the motor pathway may at times be myography (EMG) monitoring is efective in detecting
missed. injury to the peripheral nervous system. Triggered EMG
Motor evoked potential monitoring is a very (tEMG) monitoring or pedicle screw stimulation can
eicacious modality to evaluate the function of prevent neurological impairments related to misplaced
the motor pathways during surgery. hardware in surgery for spine deformities.
Continuous electromyography monitoring is
a sensitive technique that can be used during Somatosensory evoked potential
operations to detect potential damages to the monitoring
nerve roots and peripheral nerves.
Triggered EMG monitoring can be beneicial in Background
assessing the accuracy of pedicle screw placement. Somatosensory evoked potentials (SSEPs) are responses
recorded over the limbs, spine, and scalp following
stimulation of peripheral nerves. hese potentials are
Introduction believed to represent activity in the proprioceptive sen-
Spine surgery risks injury to spinal cord, nerve roots, sory pathway when low-intensity electrical stimulation
plexuses, as well as peripheral nerves. Although the over- is used. Stimuli are conducted peripherally by large-
all incidence of neurological complications ater spine diameter, heavily-myelinated, fast-conducting nerve
surgery is low, the sequelae of such injuries can be debili- ibers, and conducted centrally by the dorsal column
tating and create tremendous burdens for individuals medial lemniscal system.
and their families. he potential number of patients who here are several synapses in the proprioceptive
are at risk for such complications has increased signii- sensory pathway. he irst-order neurons in the dor-
cantly over the past few decades as the number and com- sal root ganglion receive sensory input from peripheral
plexity of spine operations has increased dramatically. nerve ibers and extend their central processes into the
Neurophysiologic intraoperative monitoring spinal cord. he central axons of the dorsal root gan-
(NIOM) is the use of electrophysiological techniques glion travel within the ipsilateral fasciculus gracilis and
to monitor the functional integrity of neural structures fasciculus cuneatus to the caudal medulla, where they
89
Anesthesia for Spine Surgery, ed. Ehab Farag. Published by Cambridge University Press. Cambridge University Press 2012.
Section 1: General considerations

synapse on the second-order neurons in perspective have high amplitude and are reliably reproducible.
nucleus gracilis and nucleus cuneatus. he second-order Stimulation can be performed on any major peripheral
axons decussate as internal arcuate ibers and ascend as nerves and can be either unilateral or bilateral. However,
the medial lemniscus to the third-order neurons in the it is good practice to monitor all four extremities since
ventral posterolateral (VPL) nucleus of the thalamus. the responses in other limbs can be used as controls
he third-order axons project into the primary somato- to diferentiate focal injury from systemic factors, such
sensory cortex in the contralateral postcentral gyrus. as anesthetic efects, hypothermia, or hypotension.
he traveling volley of the action potentials propagat- Commonly stimulated nerves include the median and
ing along these pathways, or the responses generated at ulnar nerves at the wrist, common peroneal nerve at
the sites of the synapses or within the sensory pathway the knee, and posterior tibial nerve at the ankle. he
nuclei, can be recorded at diferent sites in the limb, selection of the peripheral nerve to be stimulated is
spine, and scalp as somatosensory evoked potentials. determined by the segmental level of the spine surgery.
SSEP monitoring has been used widely for NIOM Upper limb SSEPs are generally required for cervical
during a wide variety of surgical procedures. It has spinal cord monitoring. Median nerve stimulation
been the primary spinal cord monitoring modality for is commonly used for upper limb SSEPs. When the
several decades. Several animal models have demon- surgery involves the lower cervical segments (C78),
strated that the duration and severity of damage to the ulnar nerve SSEP monitoring is suggested instead as
spinal cord correlate with the degree of changes in the the damage can be missed by median SSEPs. Posterior
SSEP indings.1 One large multicenter survey showed tibial or common peroneal nerve SSEPs are necessary
that the incidence of postoperative neurologic deicits for thoracic spinal cord monitoring.
ater scoliosis surgery was 0.46% with SSEP monitor- Several types of electrodes can be used for stimu-
ing and 1.04% without.2 lation. Transcutaneous stimulation can be performed
SSEPs are appropriate for NIOM for several rea- through the use of standard metal disc electroenceph-
sons. First, SSEP waveforms have a deinable latency alography (EEG) electrodes, or bar electrodes. hese
and amplitude that can be quantiied for compari- electrodes should be applied irmly to the skin over
son throughout the surgical procedure. Second, the the selected nerve with collodion and sealed with plas-
responses are reproducible with reasonable stability, so tic tape or sheet to prevent drying and contamination
changes related to injury can be identiied with coni- with blood or other luids and to ensure stable SSEP
dence. hird, stimulation can be performed at almost responses. Adhesive surface electrodes can also be
every nerve that contains the sensory ibers. Fourth, the applied, but carry the risk of dislodgement. Contact
entire somatosensory pathway can be assessed along impedance of less than 5 k is recommended for trans-
the course of surgery. Finally, the neural generators for cutaneous stimulation. Subdermal needle electrodes
each waveform are known within practical precision, can be useful in the operating room setting, in certain
so localization of the damage to the nervous system conditions producing excessive adipose tissue, edema,
is possible and this allows the appropriate corrective or unusually thick skin, and when neuropathy is pre-
measures to be taken.3 However, an important caveat sent. he electrodes need to be secured in a similar
is that SSEP monitoring alone does not directly assess fashion as disc and bar electrodes. he operating room
the motor function carried by the pyramidal pathway. personnel should be informed of the needle electrode
Postoperative motor deicits can occur, although infre- locations, so that necessary care can be observed to
quently, with no associated changes in SSEP intraop- avoid needle sticks.
eratively.4 In addition, although SSEPs are good basic he electrical stimulus should be monophasic
indicators of the functional integrity of the spinal cord, rectangular pulses delivered using either a constant-
limited information is provided regarding the function voltage or a constant-current stimulator. he pulse
of nerve roots, because SSEPs are a composite of multi- width should be between 100 and 300 s (optimally
ple action potentials that enter the spinal cord through 200300 s). he stimulus intensity should be sui-
several segments.5 cient to produce a small visible twitch of the muscle,
typically 3040 mA.
Methodology SSEPs can be recorded with standard surface elec-
SSEPs are typically obtained with electrical stimula- trodes on the scalp, or subdermal needle electrodes.
90 tion of a peripheral nerve because responses generally Contact impedance should be maintained at less than
Chapter 6: Evoked potential monitoring

Table 6.1 Electrode designation for recording SSEPs smaller the noise, the fewer trials will be required. he
Electrode number of averaging trials should be just enough to
designation Location obtain reliable recordings in order to give the surgeon
feedback as quickly as possible. In general, 200600
Cc, Ci Contralateral and ipsilateral central
(C3 and C4 of the international 10/20 system) stimuli are usually required to display high-quality,
CPc, CPi Contralateral and ipsilateral centroparietal
well-deined, reproducible SSEP waveforms of 110
(half way between C3 or C4 and P3 or P4) V. he baseline responses must be obtained prior to
Fpz Frontopolar, midline critical stages and then followed continuously when
critical structures are at risk.
CPz Midway between Cz and Pz
C2S 2nd cervical spine
C5S 5th cervical spine Significant findings
T10S 10th thoracic spine Following the stimulation of a peripheral nerve, SSEP
responses can be recorded along the proprioceptive
T12S 12th thoracic spine
somatosensory pathway. he nomenclature of SSEP
L2S 2nd lumbar spine
waveforms is according to the direction of peak delec-
Epc, Epi Contralateral and ipsilateral Erbs point tion (N = negative, P = positive) and the latencies of the
(2 cm superior to the midpoint of the clavicle)
peak response in milliseconds. he number following
AC Anterior cervical above thyroid cartilage
the N and P derives from the average latency at which
Pf Popliteal fossa the particular waveform is recorded in normal healthy
REF Noncephalic reference controls. For example, an N20 response is an SSEP
waveform with upward delection recorded at approxi-
Table 6.2 Recommended montages for SSEP recording mately 20 ms ater the median nerve is stimulated at
the wrist. he nomenclature of SSEP responses has not
Median or ulnar Posterior tibial
been standardized, so the peak nomenclature may vary
Generator nerve nerve
slightly between diferent institutions.
Peripheral nerve Epi-Epc Popliteal fossa SSEPs following the median or ulnar nerve stimu-
Spinal cord C5S-Fpz T12S-REF lation include the following waveforms (Fig. 6.1):
Subcortical Cpi-REF Fpz-C5S Erbs point potential (EP). EP is the response
Cortical CPc-CPi CPz-FPz recorded with an electrode placed over Erbs
point referenced to an electrode in the same
5 k. Electrodes are placed at standard sites over the location contralaterally. EP represents the
peripheral nerve, lumbar or cervical spine, and scalp. volley of action potentials in sensory ibers
he standard designation and the recommended mon- traveling through the brachial plexus.
tage are listed in Tables 6.1 and 6.2, respectively. he N13. N13 is the waveform recorded with an
recommended bandpass is typically 30100 Hz and electrode at the ith cervical spine referred to
20003000 Hz for the low- and high-frequency ilter, Fpz. N13 probably represents a dorsal horn
respectively. Sensitivity of 510 V/cm and sweep of postsynaptic potential generated by collaterals
210 ms/cm are usually adequate to identify the SSEP of the primary aferent ibers in the lower
waveforms.6 cervical cord. Some investigators suggest that
One of the major technical limitations to record- N13 can be a far-ield potential that originates
ing SSEPs is that the sensory response amplitude is low in the ipsilateral dorsal column pathway at the
compared with the noise, such as muscle and motion cervicomedullary junction.7
artifacts and electrocardiograph, electroencephalo- P14. P14 is a subcortically generated, far-ield
graph, or electromagnetic activity in the environment. potential, best obtained referentially from
he use of averaging improves signal-to-noise ratio by scalp electrodes. Its neural generators remain
summing the activity that is time-locked to the stimu- controversial. Some believe that this waveform
lus trigger, and iteratively subtracting random back- originates from the thalamus, while some
ground noises. here is no speciic number of stimuli evidence suggests that it probably relects
required to be averaged. he larger the signal and the activity in the caudal medial lemniscus. 91
Section 1: General considerations

Left median nerve stimulation Left posterior tibial nerve stimulation

N20

P37
N18

P14
N18 P37
N34

P14 P31

PF
Erbs point

5 V 2V

5 ms 10 ms

Figure 6.1 Normal somatosensory evoked potentials. The igure shows two graphs. The one on the left of the igure shows normal
somatosensory evoked potential waveforms obtained in the operating room after left median nerve stimulation. The graph on the right
shows a normal somatosensory evoked potential following left posterior tibial nerve stimulation. The traces colored in red are obtained at the
beginning of monitoring (also called baseline recordings). The green traces are the most current obtained waveforms at the time of surgery.

N18. N18 is a subcortically generated, far-ield (T10S, T12S) or upper lumbar spines (L2S).
potential, recorded referentially from scalp It mainly represents postsynaptic activity
electrodes ipsilateral to the stimulated nerve. It generated in the dorsal horn of the lumbar
may occur as early as 16 ms ater stimulation of spinal cord, analogous to the N13 response of
the peripheral nerve, and can persist for 6 ms or the median nerve SSEP.
more. N18 likely represents postsynaptic activity N34. N34 is a subcortically generated, far-ield
from multiple generator sources in the brainstem potential recorded referentially from the Fpz
and perhaps the thalamus. he precise location of electrode. It is analogous to the N18 response
the neural generators remains to be determined. following median nerve stimulation. he neural
N20. N20 is a near-ield potential, recorded generator of this potential is controversial, and
from a scalp electrode contralateral to the it may represent postsynaptic activity from
stimulus. It represents the response arising many generator sources in the brainstem, and
from the primary somatosensory cortex in the perhaps, the thalamus.
postcentral gyrus. his is the most important P37. P37 is the response which is generally
waveform of the upper limb SSEP monitoring. recorded by scalp electrodes placed
somewhere between the midline and the
he SSEP responses ater the stimulation of the poster- centroparietal locations. his potential
ior tibial nerve consist of the following (Fig. 6.2): represents the activation of the primary
Popliteal fossa potential (PF). PF is the waveform cortical somatosensory area of the leg,
recorded with an electrode placed over the contralateral to the stimulus. As the cortical
popliteal fossa. his response relects the volley representation of the leg is in the midline,
of action potentials in sensory ibers of the the orientation of the P37 dipole can be
posterior tibial nerve. variable, and at times may cause paradoxical
92 Lumbar potential (LP). LP is the response recorded lateralization, i.e., maximal P37 activity is
referentially over the dorsal lower thoracic recorded over the ipsilateral scalp. As a result,
Chapter 6: Evoked potential monitoring

it is crucial to record from both midline and Table 6.3 Efects of commonly used anesthetic agents on SSEP
responses
ipsilateral scalp locations to be certain that the
P37 response is absent. Agents Latency Amplitude
he SSEP responses are recorded as a baseline prior to Inhalational anesthetic agents
the operation, and continuously monitored through- Nitrous oxide Increase Decrease
out the surgery especially during critical stages. he Halothane Increase Decrease
most commonly used warning criteria by most Enlurane Increase Decrease
centers for identifying a signiicant change of the
Isolurane Increase Decrease
SSEPs is a 50% decrement in amplitude from the base-
Deslurane Increase Decrease
line and/or a latency prolongation of 510% over the
baseline value. hese cutofs are generally guidelines Sevolurane Increase Decrease
for high likelihood of damage and are supported by Intravenous anesthetic agents
several animal and clinical studies as warning criteria Barbiturates
that predict neurological deicits if uncorrected, but Low dose No change No change
are also reversible if appropriate actions are taken. High dose Increase Decrease
It is also important to note trends during monitor- Benzodiazepines Increase Decrease
ing so that early warning can be given to the surgeon
Opioids Increase Decrease
in order to prevent any impending damage to neural
Etomidate Increase Increase
structures.
However, several factors beside injury to the ner- Ketamine Increase Increase
vous system can afect SSEP waveforms. hese factors Propofol Increase Decrease
include the following:
1. Anesthesia cortical SSEPs begin to change with a drop in
Anesthetic agents, especially inhalational agents, amplitudes and a prolongation in latency. A loss
generally result in a reduction in amplitude and a of cortical SSEPs occurs if cerebral blood low
prolongation of the interpeak latency of the SSEP drops below 15 ml/100 g/min. his rate of low
responses, particularly the cortical potentials (N20 is not adequate to maintain cortical electrical
and P37). Subcortical and spinal responses tend activity but is just above the critical threshold
to be less afected by anesthesia; thus, this can be for permanent neuronal injury. herefore, a loss
a helpful clue to guide that the changes in SSEP of cortical SSEP responses is an early warning
waveforms may result from systemic factors and sign for decreased cerebral perfusion and, if not
permit continued monitoring. he reduction in corrected, may result in permanent neurological
SSEP amplitude directly correlates with the depth deicits. Subcortical and spinal SSEP responses
of anesthesia, so the level of anesthesia should tend to be more resistant to ischemia, and
be kept as light as possible. However, this may measurable electrical signals may be recorded
not always be possible, and conlicts between the even ater blood low to the generator sites has
needs of the surgeon, the anesthesiologist, and the ceased for several minutes.
neurophysiologist are sometimes unavoidable. Generally, SSEP waveforms are minimally
Hence, efective communication is necessary so afected if mean arterial pressure is kept more than
that compromises can be reached. Generally SSEP 70 mmHg due to cerebral autoregulation. However,
monitoring can be successful with almost every the degree of SSEP degradation with decreases
anesthetic technique. in blood pressure varies between individuals,
he efects of commonly used anesthetic agents especially among patients with underlying
are listed in Table 6.3. cardiovascular and cerebrovascular diseases.
2. Blood pressure 3. Temperature
A certain amount of perfusion to neural tissue is Diminished body temperature results in a decrease
required to meet the metabolic demands. If the in neural conduction velocity and can cause a
demands are not met, electrical activity of neural prolongation of SSEP peak latencies. For every 1C
tissue will start to shut down. When cortical decrease in body temperature, the latency of the 93
blood low is reduced below 18 ml/100 g/min, N20 response increases every 0.751.0 ms. Cortical
Section 1: General considerations

Table 6.4 Localization of neural dysfunction based on the involved were questioned about monitoring techniques
pattern of changes in SSEP responses
and the warning criteria for signiicant SSEP changes.
Pattern of SSEP Responses were obtained from 153 surgeons and 90
Locus of neural injury degradation neurophysiologists. SSEP monitoring was performed in
Peripheral nerve Unilateral loss of Erbs point, 51 263 cases of spine surgery (53% of 97 586 total cases).
e.g., limb malpositioning subcortical, and cortical signals he majority (60%) of the cases were scoliosis surgery.
Spinal cord dysfunction Intact Erbs point potential Compared with the historical data, the rate of neurologi-
Prolongation or loss of cal deicits reduced from 0.72% to 0.55%. he incidence
subcortical and cortical signals of persistent neurological deicits decreased from 0.46%
Cerebral dysfunction Intact Erbs point potential and to 0.31%. he rate of major neurological deicits (e.g.,
e.g., cerebral ischemia subcortical signal
Unilateral loss of cortical
paraplegia) dropped signiicantly from 0.61% to 0.24%.
responses All these changes are statistically signiicant (p < 0.001).
Systemic factors Global cortical loss he false-negative rate was 0.127%, while the false-
e.g., anesthetic efect, Intact subcortical signals positive rate was 1.51%. he negative predictive value
hypotension was 99.93%, signifying that the monitoring is likely to
be accurate if SSEPs remain stable throughout the sur-
SSEP signal disappears when the temperature is gery. he positive predictive value was 42%, suggesting a
less than 22C. Subcortical and spinal responses are tendency to false alarm. However, as surgeons may have
generally more resistant to hypothermia and can prevented the neurological deicits based on the intra-
provide a clue to systemic etiology of SSEP changes. operative monitoring information, the false positive
In addition, alteration in temperature afects the may not be truly false. SSEP monitoring is estimated
metabolism of anesthetic agents, which can also to prevent one neurological deicit for every 200 cases
contribute to the changes in SSEP responses. monitored. he cost to prevent one complication is esti-
When there is a signiicant change in SSEP responses mated to be around $120 000, which is still less than the
consistent with the warning criteria, it must be inter- lifetime cost of medical care for a young paraplegic.
preted within its clinical context and several sys- According to a systematic review by Fehlings et al.,8
temic factors mentioned above must be considered. eight studies evaluating the diagnostic characteristics
Localization of the neural insult based on the pattern of SSEP monitoring were identiied. he sensitivity
of SSEP degradation allows appropriate measures to and speciicity of unimodal SSEP monitoring ranged
be taken, and may help guide the surgeon to the most from 0% to 100% and 27% to 100%, respectively. he
appropriate course of action to correct the dysfunction positive predictive value was 15100%, and the nega-
related to surgical complications. Some important pat- tive predictive value was 95100%. he investigators
terns of SSEP changes and the localization of the neural concluded that the overall strength of the evidence for
insult are demonstrated in Table 6.4. unimodal SSEP monitoring was very low, signifying
uncertain estimates of efect. he quality of studies was
Outcomes poor. he quantity was high, and the consistency
In addition to merely predicting the neurological dei- was poor. However, when SSEP monitoring is used
cits, in order to be clinically useful NIOM must be able in combination with other neurophysiologic modali-
to cost-efectively alter operative morbidity. Ethical ties, such as motor evoked potential monitoring, the
considerations prevent the use of prospective, rand- overall strength of evidence with respect to sensitivity
omized controlled trials on surgical outcome of intra- and speciicity is high. his systematic review recom-
operative SSEP monitoring. As a result, the evidence mends the use of SSEP monitoring together with other
on the eicacy of SSEP monitoring mostly is from case neurophysiologic modalities for spine surgery where
series and historical controls. the spinal cord or nerve roots are deemed to be at risk.
he largest multicenter study to assess the eicacy
of SSEP monitoring during spine surgery to date is that Motor evoked potential monitoring
of Nuwer et al.2 In this study, questionnaires were sent
to the US members of the Scoliosis Research Society Background
94 (SRS). Surgeons were questioned about the morbidity Although SSEP monitoring has been proven to be use-
and the use of SSEP monitoring. he neurophysiologists ful for neurophysiologic intraoperative monitoring and
Chapter 6: Evoked potential monitoring

had been a primary modality for monitoring the func- MEPs are the responses recorded ater stimulation
tional integrity of the spinal cord over the past several of the motor pathways of the central nervous system
decades, it is not always accurate at predicting damage (cerebral hemispheres or spinal cord). MEPs assess the
to the motor pathways. In the past, neurophysiologists functional integrity of the descending motor pathway
assumed that a signiicant injury to the nervous sys- at various levels. he purpose of MEPs in NIOM is to
tem that results in motor deicits would be suicient stimulate rostral to the structure at risk and record the
to cause changes in the SSEP responses. However, it is responses at a distal site.
well documented that a damage sparing the posterior here are several types of MEPs depending on
columns can cause debilitating motor deicits but may where the stimulation occurs and where the responses
not signiicantly alter the SSEP recording.4 his is due are recorded. Transcranial MEPs (TcMEPs) can be
to the fact that the main motor pathways of the spinal attained by stimulation through the skull, activating the
cord are located in the dorsolateral funiculus, sepa- primary motor cortex which is located at the precentral
rated from the somatosensory pathways which are in gyrus. he signal recording is possible at the level of
the dorsal columns. Furthermore, the anterior spinal muscle (compound muscle action potential, or myo-
artery supplies the motor pathway and a signiicant genic MEPs), nerve (neurogenic MEPs), or spinal cord
portion (approximately 75%) of the arterial supply to (D wave recording). Stimulation can also be performed
the spinal cord, while the dorsal columns are mainly directly in the spinal cord during operations and sig-
supplied by the posterior spinal arteries. herefore, nals are recorded distally from peripheral nerves or
SSEP responses do not accurately represent the vascu- muscles. he spinal stimulation has drawbacks includ-
lar state of the spinal cord. In addition, neurons in the ing the potential for antidromic stimulation of sensory
motor gray matter of the spinal cord are more vulner- pathways leading to a motor evoked response.10 At each
able to ischemia than axons in the dorsal column sen- location, either electrical or magnetic stimulation can
sory white matter. hus, assessment of the pyramidal be performed.
pathway would be more sensitive to vascular events of Stimulation of the primary cortex produces acti-
the spinal cord than SSEP monitoring. vation of contralateral muscles. A propagated action
Because prevention of motor deicits is always the potential travels down the corticospinal tract to the
main objective of NIOM, monitoring the integrity spinal cord where it activates the anterior horn cells.
of the corticospinal tract has always been an issue of he signal then travels along the peripheral nerve,
interest. Prior to the widespread use of motor evoked traverses the neuromuscular junction, and stimulates
potential monitoring, the only way to evaluate the the muscle ibers leading to muscular contraction.
function of the motor pathway during surgery was the
Stagnara wake-up test.9 Patients are awakened dur- Methodology
ing the operations and asked to move their feet. his MEPs can be elicited with either electrical or magnetic
technique possesses several drawbacks. First, it causes stimulation. In awake subjects, magnetic stimulation is
signiicant delay to the surgery. Furthermore, moni- generally the preferred technique because a magnetic
toring certain patient populations, such as those with pulse is painless while an electric shock is painful. In
cognitive or hearing impairments, would be techni- anesthetized patients, however, magnetic stimulation
cally challenging. Additionally, this test cannot assess has no advantage over electric stimulation, and indeed
the motor pathway continuously, and, thus, when posi- possesses several major drawbacks. For example, equip-
tive, it is possible that a substantial period of time has ment used for magnetic stimulation is more expensive
elapsed from the onset of injury. and more cumbersome in the operating room setting,
Motor evoked potential (MEP) monitoring was and magnetic MEPs are more sensitive than electric
developed in 1980, and has since emerged as an MEPs to anesthetic agents. herefore, intraoperative
extremely efective way to assess the functional integ- MEP monitoring is most commonly performed by
rity of the corticospinal tract. Changes in MEPs are electrical stimulation.
sensitive in detection of postoperative neurological Transcranial stimulation of the motor cortex is
deicits. MEP monitoring has become the gold standard minimally invasive and allows monitoring of the
for neuromonitoring of the motor pathways, especially entire motor neuraxis. he techniques of transcranial
since the advent of multipulse technique for transcra- electrical stimulation vary among institutions. he 95
nial electrical stimulation of the motor pathways. single-pulse stimulation technique is highly sensitive
Section 1: General considerations

Left MEP Right MEP

Left biceps Right biceps

Left FDI Right FDI

Left adductors Right adductors

Left ant tib Right ant tib

Left AHL Right AHL

Figure 6.2 Myogenic motor evoked potentials after transcranial electrical stimulation. The igure shows myogenic motor evoked potentials
obtained in the operating room following multipulse transcranial electrical stimulation. The traces on the left are from the muscle groups
listed from the left side of the body, whereas the traces on the right are from the muscle groups on the right. The red traces are the baseline
recordings while the green are the most recently obtained motor evoked potentials.

to anesthetic efects, so multipulse stimulation is gen- electrodes placed in the interlaminar space). In add-
erally performed. Anodal stimulation with a short ition, electrical stimulation can be performed in the
duration (0.05 ms), rapid rise time pulses to subcuta- operative ield with interspinous or epidural electrodes
neously placed EEG electrodes at C3 and C4 is com- using a distant anode in the subcutaneous tissue, but
monly used. he output current range is 01000 mA, this is more technically challenging. Although this
from a source voltage as high as 200800 V. he pulse technique is less sensitive to anesthetic agents, MEP
width range can be varied from 50 to 500 s, and the responses obtained via direct spinal cord stimulation
interstimulus interval is typically between 1 and 4 are less likely to represent motor function because, in
ms. hese parameters are adjusted until reproducible addition to the anterograde activation of the motor
responses can be observed in all the muscles exam- pathways, spinal cord stimulation produces antidromic
ined. Transcranial electrical stimulation is relatively sensory responses via retrograde activation of the dor-
contraindicated in patients with epilepsy, due to the sal column, similar to traditional SSEP recordings.10
theoretical possibility of seizure kindling efect. Other Stimulation of the motor system at either the cor-
relative contraindicationa are the presence of a pace- tical or the spinal cord level results in activation of mul-
maker, infusion pumps, cochlear implants, aneurysm tiple descending tracts and causes contraction of several
clips and other retained metal fragments, a history of muscles. herefore, theoretically, recording electrodes
skull fracture, or other intracranial diseases. he risk can be placed anywhere along the descending pathway.
beneit ratio for use of this modality in these scenarios MEPs are generally recorded from muscles (myo-
should be assessed prior to its use. genic MEPs) in most institutions with surface, subder-
Spinal cord stimulation of the motor system can be mal, or intramuscular electrodes as a compound muscle
achieved through the use of a nasopharyngeal/esopha- action potential (CMAP). he potentials can be obtained
96
geal active and laminar needle electrode (needle from most limb muscles relatively easily, and recording
Chapter 6: Evoked potential monitoring

techniques are not invasive. he selection of the mus- responses are exceptionally variable in size and con-
cles for recording is determined by the clinical question. iguration. A complete loss of the MEP waveform is
Typically, the ankle extensor and lexor muscles and the considered signiicant and indicative of potential dam-
quadriceps muscles are used in the lower limbs, while the age to the motor pathways. However, given the all-or-
intrinsic hand muscles, forearm extensors, and arm lex- none nature of this approach, there is a concern that
ors are monitored in the upper extremities. Unilateral this method is not sensitive enough in detecting sub-
dysfunction of the motor pathway can be identiied with tle injury to the corticospinal tract, and postoperative
the CMAP recording. MEP responses have high signal- motor deicits are possible.11
to-noise ratio, so no averaging is required, and feedback A modiication of the all-or-none criteria includes
can be instantaneous. Disadvantages of recording from measuring the CMAP amplitude preoperatively as
muscles include variability in the CMAP responses with a baseline value, then measuring relative changes in
the level of neuromuscular blockade. Cortically evoked amplitude during surgery to determine whether a sig-
CMAPs are more variable and more sensitive to anesthe- niicant change has occurred. Due to signiicant vari-
sia than spinally evoked CMAPs. ability in CMAP signals, an 80% reduction in amplitude
MEPs can be recorded from the spinal cord with from baseline in at least one out of six recording sites
transcranial stimulation (D wave recording). his has is required for a signiicant change. When this criter-
the advantage of being relatively insensitive toward ion was used in a study of 142 patients who underwent
anesthetic agents and not afected by full neuromus- corrective surgery for spinal deformity, a sensitivity of
cular blockade. Moreover, SSEPs can be recorded reli- 100% and a speciicity of 91% were achieved.12
ably if the cerebral cortex and peripheral nerve are he threshold criterion is another approach that has
stimulated at the same time. he disadvantages of the been proposed in an attempt to improve the sensitiv-
spinal cord recordings are that they require insertion ity in detecting damage to the motor tracts. According
of epidural leads, which usually requires a posterior to the study by Calancie and Molano involving 903
approach to the spinal cord, and the recordings are patients who underwent spinal surgery at either cer-
limited to the level above T11. Although rare, there is vical, thoracic, or lumbar levels, the increase in the
a risk of hematoma from placement of the recording threshold required for obtaining CMAP responses of
electrodes. he recordings from the lumbar cord are 100 V or more that persists for more than one hour and
generally not as reliable. Importantly, the side respon- is not due to systemic factors is highly correlated with
sible for deterioration in the recorded volleys cannot be postoperative motor deicits.11
readily identiied as the epidural electrode may move Finally, the morphology criterion has also been
during surgery, such as during spinal distraction. proposed. his approach detects impairment of motor
MEP responses can also be obtained from the peri- conduction in the corticospinal tract by tracking
pheral nerves (neurogenic MEPs). Neurogenic MEPs changes in the morphology of MEP signals, such as
are relatively immune to the efects of muscle relaxants reduction in complexity of the MEP waveforms. In par-
compared with myogenic MEPs. However, neurogenic ticular, Quiones-Hinojosa and colleagues observed
MEPs are technically challenging. Only a few nerves changes in the CMAP responses from a polyphasic to a
supplying upper and lower extremities are suitable for biphasic waveform, or from polyphasic to biphasic and
recording. he responses have low signal-to-noise ratio, ultimately to loss of signal, in 28 patients who under-
so averaging of more than 100 responses is necessary. went intramedullary spinal cord tumor resection. he
alterations in morphology persisted despite signiicant
Significant findings raise in the threshold voltage. he reduction in the
When recording MEP responses from the muscle (myo- complexity and/or loss of the MEP waveforms signii-
genic MEPs), the compound muscle action potentials cantly correlated with postoperative motor deicits.13
will be obtained (Fig. 6.2). TcMEP responses can at times luctuate extremely
here are four routinely used warning criteria for strongly. Abrupt loss or marked attenuation of the
interpretation of TcMEP responses: (1) the all-or-none CMAP amplitude can occur immediately ater record-
criterion, (2) the amplitude criterion, (3) the threshold ing a robust response. his may result from spontane-
criterion, and (4) the morphology criterion. ous luctuation of excitability of the anterior horn cells.
he all-or-none criterion is the method that is used herefore, signiicant changes of CMAP response must
widely in most institutions, because TcMEP CMAP be veriied before deciding that the changes are real. 97
Section 1: General considerations

Table 6.5 Efects of commonly used anesthetic agents on CMAP when the patient movement is acceptable, and the
responses
surgeon should be notiied before the stimulus is
Agents Latency Amplitude applied.
Inhalational anesthetic agents 3. Temperature
Nitrous oxide Increase Decrease MEP responses are generally less afected by
Halothane Increase Decrease
changes in temperature than are SSEP responses.
Hypothermia may result in a gradual increase in
Enlurane Increase Decrease
stimulation threshold.
Isolurane Increase Decrease
Deslurane Increase Decrease When the MEP signals are recorded from the spinal cord,
two types of waveforms are obtained: D (direct) and I
Sevolurane Increase Decrease
(indirect) waves. Cortical stimulation generates a ser-
Intravenous anesthetic agents
ies of descending action potentials in the corticospinal
Barbiturates tracts. he D wave is generated by depolarization of the
Low dose Increase Decrease axon hillocks of the large motor neurons. It is followed by
High dose Increase Decrease a series of I waves that result from synaptic depolariza-
Benzodiazepines Increase Decrease tion of interneurons within the cortical gray matter.
Opioids No change No change Compared with CMAP recordings, the D wave is
Etomidate No change No change relatively resistant to anesthesia because there are no
synapses involved between the stimulating and the
Ketamine No change No change
recording sites. he I waves, in contrast, are severely
Propofol Increase Decrease
afected by inhalational anesthetic agents. he D wave
is also immune to neuromuscular blockade as the
Similarly to the SSEP monitoring, several systemic recording does not involve the activation of muscle
factors can afect CMAP signals. hese factors include ibers. he latency of the D wave may be temporarily
the following:3 prolonged if the exposed spinal cord is cooled, either
1. Anesthesia by cold saline irrigation or by a low operating room
MEPs can be signiicantly afected by temperature.
anesthetic agents. hese agents can abolish the he warning criteria used by most institutions for
response at multiple sites that involve synaptic identifying a signiicant change of the D wave are a
transmission, particularly at the level of the 50% reduction in amplitude from the baseline and/or
cerebral cortex and anterior horn cells and less a latency prolongation of 10% over the baseline value.
so at the neuromuscular junction. Halogenated D wave monitoring has been shown to be correlated
inhalational anesthetic agents can easily reduce with the postoperative motor functions. A complete
the CMAP responses by blockade at the cortex and loss of TcMEP CMAP responses with at least 50%
anterior horn cells. If these agents are necessary, preservation of the D wave amplitude generally cor-
the concentration must remain very low (0.5%). responds to transient paraplegia ater the operation.
he efects of commonly used anesthetic agents Patients with complete loss of the D wave during sur-
are listed in Table 6.5. gery are likely to have permanent postoperative motor
2. Paralytics deicits.14 Caveats for the D wave monitoring are that
Since CMAPs are motor responses, the use of epidural D waves can produce false positive results
neuromuscular blocking agents will suppress during a scoliosis surgery. A decrease of up to 75%
or eliminate these potentials. However, partial or an increase of D wave amplitude can be found in a
neuromuscular blockade is sometimes beneicial number of patients despite unchanged muscle CMAP
for the surgery and the MEP monitoring since responses and neurologic outcomes.15
it improves surgical retraction and can lead to
substantial reduction in patient movements Outcomes
following the stimulation, thus lessening the risk As discussed above, it had been assumed in the past that
of injuries. A neuromuscular blockade up to 50% the motor tracts should be spared if the SSEP responses
98 may be used during MEP monitoring with CMAP. are stable during surgery. Although this is true in
MEP stimulation must be performed at times the majority of cases, motor deicits in the setting of
Chapter 6: Evoked potential monitoring

normal SSEP can occur. he development of reproduc- peripheral nerves are at risk, such as during spine surgery,
ible MEP recording techniques has enabled the moni- EMG monitoring can be utilized to minimize the chance
toring of motor pathways during spine surgery. of injury to these structures, by detecting impending
Several studies have evaluated the utility of MEP damages to the peripheral nerve caused by manipula-
monitoring during surgery. In these reports, there tion, traction, compression, or vascular events.
were no patients with intraoperative preservation Each muscle iber is innervated by an alpha motor
of MEPs who developed new motor deicits postop- neuron in the spinal cord. Conversely, an axon of a
eratively. Overall, there is a good correlation between single motor neuron may innervate from a few muscle
changes in the MEP and postoperative motor func- ibers (as in ocular muscles) to more than 500 ibers
tion. All patients with new postoperative deicits had (as in the gastrocnemius). A motor neuron plus all cor-
at least a 50% reduction in the amplitude of the MEP responding muscle ibers that it innervates comprise
responses.16,17 he false positive rate of MEP moni- a motor unit. When the mechanical irritation of the
toring has varied among the studies because diferent nerve roots or peripheral nerves during the surgery
warning criteria were used. When relatively strict crite- is suicient, there will be axonal depolarization that
ria, such as an all-or-none response of the CMAPs, are results in the activation of the corresponding muscula-
employed, there is a reduction in the number of false- ture, which can be recorded by EMG. he consequent
positive cases. In addition, the false-positive cases may EMG indings provide immediate feedback to the sur-
not be truly false because surgeons may be able to geon regarding the efects of his or her actions.
prevent postoperative neurological deicits based on EMG can be recorded from any muscle accessible
the intraoperative monitoring information. to needle, wire, or surface electrodes. he selection of
According to a systematic review by Fehlings et al., the muscles for recording is determined by the struc-
two studies evaluating the diagnostic characteristics tures at most risk. Commonly used spinal nerve-inner-
of MEP monitoring were determined to be adequate vated muscles are listed in Table 6.6. Each spinal nerve
to undergo further evaluation. he sensitivity and root innervates a group of muscles, which is termed
speciicity of MEP monitoring varied due to diferent the myotome for that nerve root. On the other hand,
warning criteria being used. In summary, both the sen- most muscles are innervated by several spinal nerve
sitivity and the speciicity of MEP monitoring varied roots.
from 81% to 100%. he positive predictive value and
the negative predictive value ranged from 17% to 96%
and 97% to 100% respectively. he overall strength of
Methodology
A variety of electrodes can be used to record EMG
evidence for unimodal MEP monitoring is very low,
activity intraoperatively. Surface and subcutaneous
so any estimates of efect are very uncertain. he quality
electrodes can capture some muscle activities of inter-
of studies was poor. he quantity was poor, and the
est, but generally these electrodes are inadequate since
consistency was high. When MEP monitoring is used
they cannot reliably record activity deep in a muscle
in combination with other neurophysiologic modali-
nor can they precisely identify the speciic respon-
ties, the overall strength with respect to sensitivity and
sible muscle. Standard concentric and monopolar
speciicity is high. MEP monitoring together with
needle electrodes can record EMG activity with excel-
other neurophysiologic modalities is recommended in
lent quality, but have limitations related to their being
spine surgery where the spinal cord or nerve roots are
bulky and diicult to keep in place and out of the way of
deemed to be at risk.8
the surgeon and anesthesiologist. Subcutaneous EEG
needle electrodes are commonly used; when a record
Continuous electromyography from deeper muscles is required, ine Nichrome wires
can be inserted with a hollow-bore needle.
monitoring (free-running EMG monitoring is generally recorded with stand-
electromyography) ard gains of 100500 V, and a sweep speed of 10200
ms/division. he commonly used bandpass is 2030
Background Hz and 20 000 Hz for the low- and high-frequency il-
Electromyography (EMG) is a neurophysiologic tech- ter, respectively. EMG recording is typically presented
nique for evaluating and recording the electrical poten- on a monitor as well as over a speaker, and the activity 99
tials generated by muscle ibers. When nerve roots or of interest can be printed or stored for later review.
Section 1: General considerations

Table 6.6 Commonly used muscles in EMG monitoring


Significant findings
Root Muscles he potentials of major interest for intraoperative
Cervical myotomes EMG monitoring are neurotonic discharges (Fig. 6.3).
C1 None Neurotonic discharges are distinctive, high-frequency
C2 Sternocleidomastoid bursts of motor unit potentials recorded from a mus-
C3 Sternocleidomastoid, trapezius
cle when the nerve is mechanically or metabolically
stimulated.
C4 Trapezius, rhomboids, levator scapulae
Neurotonic discharges may appear as rapid, irregu-
C5 Deltoids, biceps brachii
lar bursts lasting several milliseconds or prolonged
C6 Biceps brachii, brachioradialis, pronator teres, lexor trains lasting up to one minute. An EMG burst repre-
carpi radialis
sents near-simultaneous activation of multiple axons,
C7 Triceps brachii, forearm extensors
while an EMG train is repetitive iring of one or more
C8 Forearm lexors, pronator quadratus motor units.
Thoracic myotomes When neurotonic discharges are present, the cause
T1 Intrinsic hand muscles of such EMG activity needs to be identiied immedi-
T212 Intercostal and paraspinal muscles ately. he most important etiology to be considered is
T68 Upper rectus abdominis trauma to the nerve roots or peripheral nerves, particu-
T810 Middle rectus abdominis
larly blunt mechanical irritation/injury. Additionally,
nonmechanical irritation including temperature
T1012 Lower rectus abdominis
(such as cold saline or heat from electrocautery) and
Lumbosacral myotomes
osmotic irritation can produce intense EMG activ-
L1 Quadratus lumborum, paraspinal muscles ity. Recognition of these nonmechanical etiologies is
L2 Iliopsoas important, because they generally do not have clinical
L3 Quadriceps femoris, adductor longus, adductor implications. Mechanical irritation, however, is associ-
magnus ated with a risk of damage to the corresponding nerve
L4 Quadriceps femoris, adductor longus, adductor either immediately or with repetitive trauma.
magnus
Neurotonic discharges must be distinguished from
L5 Tibialis anterior, gluteus medius several electromyographic activities including semi-
S1 Gastrocnemius, biceps femoris, gluteus rhythmic voluntary motor unit action potentials if the
maximus
patient is not deeply anesthetized, ibrillation poten-
S25 Anal sphincter, urethral sphincter tials if the muscle has been partially denervated, end
plate potentials, complex repetitive discharges, and

Left EMG Right EMG Figure 6.3 Neurotonic discharges on


EMG. The igure shows a neurotonic
discharge recorded from free running
EMG arising from the right quadriceps
muscle. This would implicate potential
Adductors irritation of the right L24 nerve root.

Quadriceps
Neurotonic discharge
Ant tibialis

Med gastroc

AHL

100
Chapter 6: Evoked potential monitoring

myokymic discharges. hese activities can be difer- and should be avoided as much as possible. Inhalational
entiated by typical iring patterns and action potential anesthetic agents or narcotic anesthesia are usually pre-
characteristics. ferred when EMG monitoring is necessary, although
he most important feature suggesting clinical short-acting, nondepolarizing neuromuscular blocking
signiicance of neurotonic discharges is a relation to agents titrated to produce a 50% reduction of the baseline
surgical events. he onset of EMG discharges with a motor action potentials still allow neurotonic discharges
surgical activity suggests a causative role. he activity to be recorded. While such a level of muscle relaxation
that does not correspond with surgical actions tends increases the likelihood of undesired movements during
to be of benign etiology, such as return of muscle tone the surgery, movements of the patient can be prevented
and voluntary muscle contraction. Some muscles, such with adequate levels of narcotics or inhalational anesthe-
as the anal sphincter, are more likely to demonstrate sia. Additional agents such as fentanyl or midazolam
return of tone than others, and this can be seen during may be necessary to lessen background muscle contrac-
low levels of neuromuscular blockade and low anes- tions and associated motor unit potentials. A continu-
thetic depth. Voluntary muscle contraction can rarely ous monitor of the degree of blockage is recommended
occur in the case of low anesthetic depth, and electri- if partial neuromuscular blockade is used.
cal activity typically precedes gross clinical movement.
Moreover, EMG activity can be observed prior to inci- Outcomes
sion in some patients. his usually relates to the reason Although EMG was one of the irst neurophysiologic
for surgery (e.g., radiculopathy) and in such cases only modalities to be used for intraoperative monitoring,
increased EMG activity over the preincision baseline there are still limited studies to determine the outcome
would relect further nerve irritation. of patients undergoing EMG monitoring during spine
he intensity of EMG activity correlates roughly with surgery.
the degree of irritation. Since each motor unit potential In a case series reported by Beatty and colleagues,
represents a separate depolarizing axon, an increase in EMG monitoring was performed in a total of 150
the number of axons afected by any irritative sources patients who underwent spinal surgery for radicu-
would result in a larger number of distinct motor unit lopathy (120 had lumbar surgery, and 30 underwent
potentials, and thus a greater intensity of recorded EMG cervical operations). All surgeries were performed to
activity. EMG activity that persists ater the end of the relieve symptoms due to disc herniation, spondylosis,
ofending surgical action indicates a relatively intense or both. During the operations, continuous intraoper-
initial irritation, and is oten described as signiicant. ative EMG recordings were obtained from the muscle
Its presence oten suggests some level of ongoing insult corresponding to the involved nerve root. In baseline
or injury. EMG activity that appears as short bursts or recordings acquired in the operating room 10 minutes
low-frequency trains is relatively more benign, and prior to the lumbar surgery, electrical discharge or ir-
commonly is a physiological phenomenon resulting ing was detected from the corresponding muscle in 22
from the excitation of mechanoreceptors of the axon. Its of 120 patients (18% of the cases). Once the nerve was
presence, at worst, poses a low risk for persisting injury. decompressed, electrical activity diminished. hese
EMG monitoring can produce a false negative result electrical discharges were produced with regularity on
for sharp nerve transection.18 Furthermore, in patients nerve root retraction. he authors concluded that con-
with underlying abnormal motor nerves, EMG monitor- tinuous EMG monitoring can be easily accomplished
ing may fail to detect additional intraoperative damage and yields valuable information that indicates when
even though such injury is blunt mechanical trauma.19 the nerve root is adequately decompressed or when
EMG monitoring is more likely to yield false-negative undue retraction is exerted on the root.20
indings in patients with some underlying neurological Jimenez et al. reported that the incidence of post-
diseases, particularly in those with disorders of the neu- operative C5 palsies reduced from 7.3% to 0.9% as a
romuscular junction or the muscle. Classic examples of consequence of continuous intraoperative EMG moni-
these conditions include myasthenia gravis, botulinum toring. he authors also noted that no patient sufered a
toxin treatment for dystonia, and muscular dystrophy. postoperative C5 palsy when there was no intraopera-
Because EMG monitoring detects the activity in the tive evidence of root irritation.21
skeletal muscles, neuromuscular blocking agents will Continuous intraoperative EMG plus SSEP moni-
signiicantly attenuate the EMG and CMAP activity toring has been used commonly in spinal surgery to 101
Section 1: General considerations

prevent postoperative neurological deicits. However, surrounding bone serves as an insulator to electrical
only limited data are available on the sensitivity, spe- conduction. hus, a higher amount of electrical current
ciicity, and predictive values of multimodality tech- is required to activate the surrounding nerve roots. On
niques. Gunnarsson et al. retrospectively analyzed the other hand, if the screw perforates the pedicle wall,
a prospectively accumulated series of patients who a low-impedance pathway will be created between the
underwent intraoperative monitoring with SSEP hole and nearby exiting nerve roots, and the stimula-
and EMG during thoracolumbar spine surgery. he tion threshold will be signiicantly reduced. As a result,
analysis focused on the correlation of intraoperative evaluation of the integrity of the pedicles can be based
electrophysiological indings with the development on the minimum level of electric current required to
of postoperative neurological deicits. here were 213 activate nearby nerve roots.
patients who underwent surgery on a total of 378 levels;
32.4% underwent an instrumented fusion. Signiicant Methodology
EMG activity was noted in 77.5% of the patients and Triggered EMG monitoring can be performed by using
signiicant SSEP changes were seen in 6.6%. Fourteen an electrode to directly stimulate the pedicle screw at
patients (6.6%) developed new postoperative neuro- increasing current intensities. Monopolar electrodes
logical symptoms. Of those patients, all had signiicant such as nasopharyngeal electrodes are commonly used,
EMG activation while only four had signiicant SSEP and are placed within the pedicle screw holes and/or on
changes. Intraoperative EMG activation had a sensitiv- hardware. Responses are recorded with needle electrodes
ity of 100% and a speciicity of 23.7% for the detection that are typically placed in the appropriate muscle group
of a new postoperative neurological deicit, whereas corresponding to the level of spinal surgery (Table 6.6).
SSEP had a sensitivity of 28.6% and speciicity of 94.7%. Direct nerve root stimulation using <2 mA can be
he authors concluded that intraoperative EMG moni- tried prior to pedicle screw stimulation to ensure that
toring has a high sensitivity for the detection of a new the stimulus current is accurately delivered. CMAP
postoperative neurological deicit but a low speciicity. responses that are time locked to the stimulation can
On the other hand, SSEP has a low sensitivity but a high be recorded from the monitored muscle innervated by
speciicity. he authors also noted that combination of that nerve.
intraoperative EMG and SSEP monitoring is helpful Several stimulation parameters and techniques of
for predicting and possibly preventing neurological pedicle screw stimulation have been described in the
injury during thoracolumbar spine surgery.22 literature. he rates of pulsatile stimulation range from
1 to 5 Hz with pulse durations between 50 and 300 ms.
Triggered EMG monitoring (pedicle he intensity of the stimulation is gradually increased
screw stimulation) from zero until a reliable and repeatable EMG response
is obtained from corresponding muscle groups or a pre-
Background determined maximum stimulus intensity is achieved.
Spinal instrumentation and fusion has become the It is generally recommended that the maximum stimu-
standard procedure for spinal stabilization over the lation intensity is 50 mA.
past few decades because it provides a permanent solu- he EMG activity is typically recorded with stand-
tion to spinal instability. Spinal instrumentations are ard gains of 100500 V, and a sweep speed of 10200
usually anchored to the vertebral column with screws ms/division. he bandpass is generally set at 2030 Hz
placed in the pedicles to provide support and allow and 20 000 Hz for the low- and high-frequency ilter,
bony fusion. Malpositioned screws that have breached respectively. he CMAP responses are presented on a
the medial or inferior pedicle wall may impinge on monitor for review (Fig. 6.4).
exiting nerve roots, causing radiculopathy of the cor-
responding level. he incidence of such complications Significant findings
has been estimated to be between 2% and 10%.23 here are several warning thresholds described in the
Triggered EMG monitoring or pedicle screw literature and each institution may use slightly diferent
stimulation is a technique that can be used to deter- cut-of values. In general, pedicle screw placements that
mine whether screws have breached the pedicle wall are associated with stimulation thresholds greater than
102 and thus pose a risk of injury to the exiting nerve root. 10 mA suggest that the pedicle wall is intact, and are
When a screw is accurately placed in the pedicle, the unlikely to represent a risk of postoperative neurological
Chapter 6: Evoked potential monitoring

EMG TRIG Figure 6.4 Pedicle screw


stimulation triggered EMG.
T9 Myotome The igure shows triggered
EMG obtained following
stimulation of a pedicle screw at
T10 myotome an intensity of 3 mA. This result
would suggest the possibility
of perforation of the medial
T11 myotome pedicle wall.

T12 myotome

L1 myotome

L24 myotome

L34 myotome

100 v
5 ms

deicits. A stimulation threshold between 10 and 20 mA with stimulation threshold greater than 7 mA, while
reasonably implies that no breach of the medial wall has exposed screws near a nerve root tend to have stimula-
occurred, while a threshold response of >15 mA indi- tion thresholds of less than 5 mA.
cates a 98% chance of accurate screw positioning on False-negative results can occur due to various
postoperative CT scan.24 Stimulation thresholds greater factors, such as the use of muscle relaxants, current
than 20 mA signify a strong probability that there is no spread, or preexisting nerve damage. hese factors
breach of the medial pedicle wall. need to be kept in mind to ensure the accuracy of ped-
A perforation of the pedicle wall is possible if pedi- icle screw stimulation. he degree of muscle relaxation
cle screws stimulate nearby exiting nerve roots at less can be assessed by a train-of-four test. Triggered EMG
than 7 mA. In such cases, the surgeon may choose to monitoring is best performed when no paralytics are
remove or adjust the screw at the corresponding site. used and four of four twitches are optimal for reliable
In a situation where redirection is not feasible or the recording. Excessive luid, blood, or sot tissue around
screw is a keystone to the success of spinal fusion, the the head of the screw at the time of stimulation can
surgeon may opt to leave that screw in place ater deter- potentially shunt current away from the screw, and
mining that the pedicle wall is unlikely to be perforated thus result in no activation of the nerve roots. Lastly,
in a clinically signiicant manner. his is typically con- chronically compressed nerve roots have higher stimu-
irmed by probing the pedicle hole and/or obtaining a lation thresholds, with literature reports ranging from
luoroscopy to look for a potential impingement on a 6 to >10 mA, compared with 2 mA for a normal nerve
nerve root. he presence of neurotonic discharges on root.19 As a result, in nerve roots where there is known
spontaneous EMG monitoring corresponding to screw or suspected damage, direct nerve root stimulation
placement or probing strongly suggests that a perfora- threshold is essential to establish a baseline value.
tion with nerve impingement is present.
Because the low-threshold stimulation inten- Outcomes
sities only implies a low-impedance pathway between he data on postoperative outcomes in patients who
the pedicle hole and exiting nerve roots, false-posi- underwent pedicle screw stimulation or triggered EMG
tive results can be seen in patients with osteoporosis monitoring are conlicting in the literature. In a pro-
resulting in thin pedicle walls or cracked pedicles. spective study conducted by Raynor and colleagues, the
Nonetheless, there is a relationship between thresh- sensitivity of rectus abdominis-triggered EMG to assess
old stimulation intensities and the exposure of a ped- the placement of thoracic screw was evaluated. A total of
icle screw. Cracked pedicles, osteoporotic bone, or a 677 thoracic screws were placed into 92 patients. Screws 103
minimally exposed screw are likely to be associated were inserted between T6 and T12 and were stimulated
Section 1: General considerations

with a CMAP recorded from the rectus abdominis monitoring 1000 patients would be approximately $1
muscle. hreshold values were compared both in abso- million. Since the estimated incidence of postoperative
lute values and relative to other locations in the same neurological deicits resulting from the placement of
patient. he stimulation thresholds can be divided into pedicle screws is around 210%, at least 20 patients
three groups: group A (n = 650 screws) had thresholds would develop some neurological complications ater
greater than 6.0 mA with intraosseus placement con- the operations. he average medical costs to correct a
irmed radiographically; group B (n = 21) had thresh- patients postoperative outcome and to rehabilitate that
olds less than 6.0 mA but with an intact medial pedicle patient are usually more than $50 000. hus, the use
border on reexamination and radiographic conirm- of triggered EMG monitoring during spine surgery is
ation; and group C (n = 6) had thresholds less than 6.0 cost-efective.23
mA with medial pedicle wall perforations conirmed by
tactile and/or visual inspection. herefore, the positive
predictive value of triggered EMG monitoring is 22% (6
Conclusion
NIOM can minimize potential neurological deicits
of 27 patients with threshold less than 6.0 mA). here
that may result from spine surgery. SSEPs are the most
were no postoperative neurological deicits. Group B
commonly used modality for monitoring spinal cord
screws averaged a 54% decrease of stimulation thresh-
function and have had the widest application. SSEP
olds, while group C had a 69% reduction in threshold
monitoring can identify the injury early enough to
from baseline (p = 0.016). he authors concluded that
alert the surgeon in most cases. However, anterior spi-
for assessment of thoracic pedicle screw placement,
nal cord injury that may result from a vascular acci-
triggered EMG thresholds of less than 6.0 mA, together
dent or direct trauma can be missed, and addition of
with 6065% reduction of stimulation threshold from
MEP monitoring in some spinal procedures may bet-
the mean of all other thresholds in a given patient,
ter protect the motor pathways. Neurotonic discharges
should alert the surgeon to suspect a medial pedicle
recorded from peripheral muscle during spontaneous
wall breach.25
EMG monitoring are sensitive to irritation of the nerve
However, the above data were contradictory to the
root. Triggered EMG monitoring is helpful to assess
study by Reidy et al., which was a prospective study
the accuracy of pedicle screw placement and, thus, can
that determined the use of intercostal EMG moni-
alert the surgeon when the damage may be occurring.
toring as an index of the accuracy of the placement
A thorough familiarity with these monitoring tech-
of pedicle screws in the thoracic spine. A total of 95
niques, as well as knowledge of the beneits and limi-
thoracic pedicle screws were placed into 17 patients.
tations of each modality, enhance the value of NIOM
A CMAP was recorded from the corresponding inter-
and the neurological and functional outcomes during
costal or abdominal muscles. Postoperative CT was
spinal procedures.
performed to determine the position of the screw. he
stimulus thresholds were correlated with the position
of the screw on the CT scan. Using 7.0 mA as a warning References
threshold, the sensitivity of EMG was 50% in detect- 1. Daube JR, Rubin DI, eds. Clinical Neurophysiology.
ing a breached pedicle and the speciicity was 83%. Oxford: Oxford University Press; 2009.
here were eight unrecognized breaches of the pedicle 2. Nuwer MR, Dawson EG, Carlson LG, et al.
based on triggered EMG results alone. horacic ped- Somatosensory evoked potential spinal cord
icle screws were placed accurately in more than 90% of monitoring reduces neurologic deicits ater scoliosis
surgery: results of a large multicenter survey.
patients. hese authors concluded that triggered EMG
Electroencephalogr Clin Neurophysiol 1995; 96: 611.
monitoring did not signiicantly improve the reliability
of pedicle screw placement.26 3. Husain AM, ed. A Practical Approach to
Neurophysiologic Intraoperative Monitoring. New York:
Regarding the cost-efectiveness of pedicle screw Demos Medical Publishing; 2008.
stimulation, the experience of Toleikis over 1000 cases
4. Lesser RP, Raudzens P, Lders H, et al. Postoperative
suggests that triggered EMG monitoring is worth-
neurological deicits may occur despite unchanged
while. he cost of monitoring for a typical instru- intraoperative somatosensory evoked potentials. Ann
mented fusion involving pedicle screw placement with Neurol 1986; 19: 225.
continuous and triggered myogenic techniques was
104 5. Tsai TM, Tsai CL, Lin TS, et al. Value of dermatomal
estimated to be $1000 or less. herefore, the cost for somatosensory evoked potentials in detecting acute
Chapter 6: Evoked potential monitoring

nerve root injury: an experimental study with special 16. Levy WJ, York DH, McCafrey M, Tanzer F. Motor
emphasis on stimulus intensity. Spine 2005; 30: evoked potentials from transcranial stimulation of
E5406. the motor cortex in humans. Neurosurgery 1984; 15:
6. American Clinical Neurophysiology Society. Guideline 21427.
11B: Recommended standards for intraoperative 17. Kitagawa H, Itoh T, Takano H, et al. Motor evoked
monitoring of somatosensory evoked potentials. potential monitoring during upper cervical spine
Bloomield, CT: American Clinical Neurophysiology surgery. Spine 1989; 14: 107883.
Society ; October, 2009. 18. Nelson KR, Vasconez HC. Nerve transaction
7. Lueders H, Lesser R, Hahn J, et al. Subcortical without neurotonic discharges during intraoperative
somatosensory evoked potentials to median nerve electromyographic monitoring. Muscle Nerve 1995; 18:
stimulation. Brain 1983; 106: 34172. 2368.
8. Fehlings MG, Brodke DS, Norvell DC, et al. he 19. Holland NR. Intraoperative electromyography. J Clin
evidence for intraoperative neurophysiological Neurophysiol 2002; 19: 44453.
monitoring in spine surgery: does it make a diference? 20. Beatty RM, McGuire P, Moroney JM, et al. Continuous
Spine 2010; 35: S3746. intraoperative electromyographic recording during
9. Vauzelle C, Stagnara P, Jouvinroux P. Functional spinal surgery. J Neurosurg 1995; 82: 4015.
monitoring of spinal cord activity during spinal surgery. 21. Jimenez JC, Sani S, Braverman B, et al. Palsies of the
Clin Orthop Relat Res 1973; 93: 1738. ith cervical nerve root ater cervical decompression:
10. Toleikis JR, Skelly JP, Carlvin AO, Burkus JK. Spinally prevention using continuous intraoperative
elicited peripheral nerve responses are sensory rather electromyography monitoring. J Neurosurg Spine 2005;
than motor. Clin Neurophysiol 2000; 111(4): 3: 927.
73642. 22. Gunnarsson T, Krassioukov AV, Sarjeant R, et al. Real-
11. Calancie B, Molano MR. Alarm criteria for motor- time continuous intraoperative electromyographic and
evoked potentials: whats wrong with the presence-or- somatosensory evoked potential recordings in spinal
absence approach? Spine 2008; 33: 40614. surgery: Correlation of clinical and electrophysiologic
12. Langeloo DD, Lelivelt A, Louis Journe H, et al. indings in a prospective, consecutive series of 213
Transcranial electrical motor-evoked potential cases. Spine 2004; 29: 67784.
monitoring during surgery for spinal deformity: a study 23. Toleikis RJ. Neurophysiological monitoring during
of 145 patients. Spine 2003; 28: 104350. pedicle screw placement. In: Deletis V, Shils JL, eds.
13. Quiones-Hinojosa A, Lyon R, Zada G, et al. Changes Neurophysiology in Neurosurgery. New York: Academic
in transcranial motor evoked potentials during Press; 2002: 23164.
intramedullary spinal cord tumor resection correlate 24. Shi YB, Binette M, Martin WH, et al. Electrical
with postoperative motor function. Neurosurgery 2005; stimulation for intraoperative evaluation of
56: 98293. thoracic pedicle screw placement. Spine 2003; 28:
14. Morota N, Deletis V, Constantini S, et al. he role 595601.
of motor evoked potentials during surgery for 25. Raynor BL, Lenke LG, Kim Y, et al. Can triggered
intramedullary spinal cord tumors. Neurosurgery 1997; electromyograph thresholds predict safe thoracic screw
41: 132736. placement? Spine 2002; 27: 20305.
15. Ulkatan S, Neuwirth M, Bitan F, et al. Monitoring 26. Reidy DP, Houlden D, Nolan PC, et al. Evaluation of
of scoliosis surgery with epidurally recorded motor electromyographic monitoring during insertion of
evoked potentials (D wave) revealed false results. Clin thoracic pedicle screws. J Bone Joint Surg Br 2001; 83:
Neurophysiol 2006; 117: 2093101. 100914.

105
Section 1 General considerations
Chapter
Pharmacology of adjunct anesthetic drugs

7 John E. Tetzlaf

one pharmacologic option for the elements of the


Key points
anesthetic for major spine surgery. One unique elem-
Pharmacologic choices for major spine surgery ent of major spine surgery that dictates the choice of
are dictated by the severity of the surgery, various adjunct drugs is central nervous system (CNS)
patient comorbidity, and the techniques of monitoring. he pharmacologic options for spine sur-
spinal cord monitoring. gery will be presented in the context of requirements
he approaches to spinal cord monitoring imposed by this monitoring.
require anesthetic choices compatible with the
speciic monitoring technique. Advanced neurophysiologic
When the wake-up test is planned, monitoring
intraoperative emergence and focused
With the continued evolution of spine surgery and the
neurologic examination require agents with
development of better instrumentation options, more
predictable recovery.
and more aggressive surgical procedures are being per-
When evoked potentials are selected, agents formed from the sacrum to the foramen magnum. A
with predictable impact on the amplitude and common element for all of these procedures is risk of
latency of the evoked signal are required. injury to the spinal cord and/or nerve roots, and com-
Continuous infusion of intravenous agents has plex spinal cord monitoring has become a routine part
an important role in the anesthetic technique of prevention of injury to the CNS. With the applica-
when evoked potential monitoring is part of the tion of these monitoring techniques to major surgery
surgical plan. comes the requirement to adjust the anesthetic drug
selection to allow the early detection of potential risk
to the spinal cord. Two broad categories for strategy are
Introduction the wake-up test and neurophysiologic monitoring
Surgery of the spine can be elective in the case of of the spinal cord. he approach to the pharmacologic
spinal stenosis, nerve root entrapment, or disc herni- choices is diferent for these two options.
ation; urgent as in the case of metastatic disease with
developing neurologic deicit; or an acute emergency
as in the case of fracture or fracture-dislocation. he
Anesthetic approach for the
anesthetic techniques are dictated by the degree of wake-up test
urgency involved, the indication for the surgery, and he wake-up test1 was irst introduced for scoliosis
the level of the spine to be operated on. Spine surgery surgery to prevent catastrophic spinal cord injury and
has a wide range of acuity from relatively minor surgi- involves awakening the patient ater completion of key
cal procedures to some of the most invasive procedures phases of instrumentation and performing a focused
performed. he anesthetic drugs selected for major neurologic examination prior to completing the sur-
spine surgery are dictated by the comorbidities in the gery. Any motor or sensory deicit not present prior to
patients health history, the severity of the surgery, and the start of surgery is taken as a sign that spinal cord
the preference of the anesthesiologist. As with almost function is impaired (distraction, kinking, vascu-
everything in anesthesia, there is always more than lar compromise) as a result of instrumentation, and

106
Anesthesia for Spine Surgery, ed. Ehab Farag. Published by Cambridge University Press. Cambridge University Press 2012.
Chapter 7: Pharmacology of adjunct anesthetic drugs

removal of some or all of the rods, screws, and/or wires anesthesia to ensure that any changes in evoked poten-
to decrease distraction on the spinal cord is indicated tials can be attributed to the surgical procedure and
to prevent irreversible injury. Ater the adjustment not to changes in the depth of anesthesia.4,5 When sig-
of instrumentation, the wake-up test is repeated to niicant changes in the evoked potentials are detected,
determine whether the correction has restored nor- adjustment of the instrumentation has been shown to
mal neurologic function. Clinical experience conirms preserve neurologic function.6 he basal anesthetic can
that adjustment of instrumentation ater an abnormal include a volatile agent, but not to exceed one-half MAC,
wake-up test can result in reversal of the neurologic because the volatile agents at higher doses suppress the
deicit.2 evoked potential signs in a dose-related manner. he
he anesthetic technique for major spine surgery choice of which evoked potential monitoring approach
using a wake-up test is focused around agents that allow (SSEP, MEP, or both) depends on the speciic spine sur-
for a reasonably rapid return of consciousness. Volatile gery procedure and the preferences of the surgical team.
agents have a role in this anesthetic technique, although he choice selected has direct impact on the anesthetic
when used as the primary agents they can result in slow drug selection. For SSEP alone, the anesthetic choice
emergence. As a result, a volatile agent is oten used with can include one-half MAC volatile agents, a nondepo-
one or more other classes of agents to reduce the time to larizing muscle relaxant to prevent gross movement
responsiveness. he options include injection or infu- and motion artifact on the SSEP recording, and add-
sion of members of the fentanyl family, propofol, and/ itional anesthetic agents to deepen the anesthesia and
or dexmedetomidine. Neuromuscular blockade must ensure amnesia. he options to deepen the anesthesia
either be avoided or maintained at a level where reversal include intermittent injection or infusion from the fen-
is possible. If the neurologic examination has changed tanyl family, infusion of propofol or infusion of dexme-
when there is an abnormal wake-up test, the anesthetic is detomidine. When MEP monitoring is selected, there
deepened using the same strategy, the instrumentation are similar requirements to SSEP monitoring, except
is adjusted by the surgeon, and the patient is awakened nondepolarizing muscle relaxants must be avoided to
again. If the neurologic examination reverts to baseline, allow measurement of motor evoked activity. his, in
the risk of spinal cord injury is greatly reduced. he turn, imposes a greater burden on the additional agents
limitations of this technique are the time involved, the because they must deepen the anesthetic suiciently to
preparation of the patient to ensure they are aware that ensure that sudden movement does not disrupt the sur-
they will be awakened intraoperatively, and the poten- gery during intervals of critical instrumentation in the
tial to lose the airway or lines during the emergence if absence of neuromuscular blockade. he same options
agitation occurs. In extreme cases, patient movement include infusions of the fentanyl family, propofol, and/
could potentially cause injury to the spinal cord or or dexmedetomidine, with combinations of two or more
nerve roots related to excessive movement. Although being more common because of the need to withhold
both options are efective, the combination of deslu- neuromuscular blockade. Because the volatile agent
rane with remifentanil provided more rapid emergence dose is ixed, the depth of anesthesia can be adjusted
for a wake-up test than propofol and remifentanil.3 to the surgical stimulus using these intravenous agents
with minimal impact on the evoked potential signals.
Anesthetic options for
neurophysiologic monitoring Pharmacologic implications
An alternative to the wake-up test is neurophysiologic
monitoring of spinal cord function. A small number of of anesthetic agents used during
centers used neurophysiologic monitoring of spinal cord major spine surgery
function in addition to using the wake-up test. he two
options are somatosensory evoked potentials (SSEPs) Inhaled anesthetic agents
and motor evoked potentials (MEPs). Each imposes he volatile anesthetic agents and nitrous oxide have
diferent conditions on the anesthetic drug selection. In been used successfully in virtually every kind of major
contrast to the wake-up test approach, there is no need spine surgery. All of the volatile agents cause a dose-
to plan for intraoperative emergence, and the strategy dependent decrease in signal amplitude of evoked
focuses on selecting an anesthetic technique compatible potentials with an increase in signal latency,7 although 107
with evoked potential monitoring, and a steady level of there are diferences in this impact among the volatile
Section 1: General considerations

Table 7.1 Intravenous agents used for spine surgery

Agent Elimination Dose range Issues with use Contraindications


Thiopental Hepatic 35 mg/kg Hypotension Porphyria
Methohexital Hepatic 0.51.0 mg/kg Hypotension Porphyria
Etomidate Hepatic 0.10.2 mg/kg Nausea, myoclonus, adrenal ? Prolonged infusion
or 510 g/kg/ suppression for sedation (adrenal
min suppression)
Propofol Hepatic 12 mg/kg (induction) Pain with injection, mild
50100 g/kg/min (sedation) antiemetic
100200 g/g/min (primary
anesthetic)
Dexmedetomidine Hepatic 1 g/kg loading over 20 min Hypotension with rapid
0.20.7 g/kg/h maintenance infusion of loading dose
Hypertension with high doses
Ketamine Hepatic 0.52.0 mg/kg induction Increased oral secretion, Preexisting dysphoria
25100 g/kg/min maintenance sympathomimetic properties,
infusion bronchodilator
Midazolam Hepatic 0.1 mg/kg for anxiolysis, less when Reduced dose with advanced
used as adjunct age or renal impairment

agents that have been studied.8 Isolurane and enlurane patients with poryphyria, as they may induce fatal
have less impact on evoked potentials than halothane at attacks of porphyria.
MAC equivalent doses.9 Although these diferences can hiopental causes a dose-related decrease in amp-
be measured reproducibly, there is no clinically signii- litude and increase in signal latency of short duration
cant diference. he newer inhaled agents, sevolurane with modest doses.7 Evoked potential monitoring is
and deslurane have a similar impact to isolurane.1012 possible shortly ater administration of thiopental,18
When brainstem integrity must be monitored, high although bolus administration during critical surgi-
doses of volatile agents probably should be avoided.13 cal instrumentation could create changes that would
Nitrous oxide causes a dose-related reduction in the appear the same as spinal cord ischemia. Continuous
amplitude of evoked potentials which must be consid- infusion of thiopental allows the measurement of som-
ered if nitrous oxide is part of the anesthetic plan.14 At atosensory evoked potentials and detection of spinal
70%, nitrous oxide reduced the amplitude of SSEPs by cord pathology.19
50% in adults15 and children.16 Deslurane, combined
with remifentanil, may provide appropriate conditions Etomidate
for rapid emergence if a wake-up test for critical neuro- Etomidate as a single-dose induction agent has a role
logic examination is required intraoperatively.17 in major spine surgery when hemodynamic stability
Table 7.1 shows intravenous agents used for spine during induction is important, and when used in this
surgery, which are discussed separately below. manner it will have little impact on evoked potential
recording. Etomidate can be used as an induction agent
Barbiturates at 0.20.4 mg/kg and can be used for maintenance of
he best-studied class of drugs for impact on evoked short cases at 10 g/kg/min, although this has limited
potentials is the barbiturate induction drugs. he most application to spine surgery. Termination of action is
commonly used barbiturate during spine surgery is by redistribution, and the short redistribution half life
thiopental. he most common use would be for induc- (25 minutes) limits the duration of action of a single
tion. Enthusiasm for continuous infusion is limited dose to about 5 minutes. Injection can be painful, and
by the hemodynamic consequences (aterload reduc- myoclonus is frequently observed shortly ater injec-
tion causing hypotension) and the cumulative efects. tion. Etomidate inhibits adrenal enzymes that produce
hiopental (35 mg/kg) produces unconsciousness cortisol, and even single doses have been reported to
within 1 minute with a duration, if not dosed, of 58 produce adrenal suppression. Sustained infusion of
108 minutes. Methohexital (0.5 mg/kg) is an infrequently etomidate for sedation in the intensive care setting has
chosen alternative. Neither should be selected in been highly associated with adrenal suppression
Chapter 7: Pharmacology of adjunct anesthetic drugs

In contrast with the barbiturates, etomidate causes Table 7.2 Elimination half-life (EHL) and context-
sensitive half-time (CSHT) for opioids
an acute increase in the amplitude of evoked potentials
with minimal increase of latency.7,20 Under certain cir- Agent EHL (min) CSHT (min)
cumstances, etomidate infusion has been reported to Fentanyl 475 >100 a
improve intraoperative SSEP monitoring21 and detec- Sufentanil 562 26 a
tion of potential spinal cord ischemia during instru-
Alfentanil 111 51 a
mentation. Conversely, by comparison with infusion
Remifentanil 48 <5 b
of fentanyl, the accentuation of scalp electrode signals
from etomidate could make detection of acute changes a
After 200 minutes or longer infusion.
b
more diicult.22 After any infusion.

30 minutes, or as an infusion at 12 g/kg/h. At lower


Propofol doses, fentanyl infusion can be used to reduce the need
Propofol can be used in spine surgery for induction or for either propofol or inhalation agent. Sufentanil can
maintenance of anesthesia. When administered as a use used in a similar manner with loading doses of 0.1
bolus at 1.02.0 mg/kg, propofol induces unconscious- 0.3 g/kg and maintenance infusion at 0.52.0 g/kg/h.
ness within 1 minute, with a duration of 78 minutes. It he slightly longer duration of action requires termi-
can be used as an adjunct to ensure continuous uncon- nation of sufentanil earlier in the procedure (4560
sciousness and amnesia with an infusion rate of 50100 minutes prior to emergence) to avoid delayed awaken-
g/kg/min or as a sole anesthetic at 100200 g/kg/min. ing. Alfentanil is generally used as an adjunct to other
Injection is generally preceded by lidocaine due to the infused anesthetics, achieving analgesia with a short
pain with injection of propofol. he high clearance rate half-life at 0.52.0 g/kg/min. Remifentanil is unique
ensures rapid emergence even ater prolonged infu- within the fentanyl family because of the ester linkage
sions of propofol. he respiratory depressant proper- in its molecular structure, rendering it susceptible to
ties of propofol during sedation should be considered rapid ester hydrolysis. his results in rapid termination
and airway management equipment and personnel of action. It is not used as a sole anesthetic, and when
should be immediately available. An attractive part of used as an adjunct for balanced anesthesia during spine
the proile of propofol is its mild antiemetic properties. surgery (0.11.0 g/kg/min) it provides analgesia and
When propofol is used as a sole agent, vivid dreaming sparing of other agents that is rapidly resolved when
has been reported. the infusion is stopped. he elimination half-life (EHL)
he impact of propofol on evoked potential signals and the context-sensitive half-time (CSHT, deined as
is clinically insigniicant as long as boluses are avoided the time for the central compartment to decrease by
during critical instrumentation.2325 Continuous infu- 50% from termination of infusion) can be used to deter-
sion of propofol mixed with opioids allows for satis- mine when to end an infusion to allow for emergence
factory monitoring of spinal cord function during (Table 7.2), especially if the CSHT is considered as the
scoliosis and vertebral fracture.15 Targeted controlled time for termination of brain action of the opioid.
infusion of propofol for a wake-up test provides a Continuous infusion of fentanyl is consistent with
smoother, more rapid intraoperative emergence than deepening anesthesia while allowing THE recording
manual adjustment based on dose and observation.26 of somatosensory evoked potentials.30 Infusions of
At high doses, propofol has been reported to decrease fentanyl or morphine were equivalent in the impact
the amplitude of transcranial electrical motor evoked on evoked potential recording, allowing for clinical
potentials.27 Interestingly, peripheral blood low was spinal cord monitoring, although more impact was
increased and blood loss decreased28 due to selective noted with bolus injection compared with infusion.31
vasodilation from propofol29) compared with sevolu- Remifentanil is an excellent addition to either deslu-
rane during major spine surgery. rane or propofol to facilitate intraoperative emergence
for a wake-up test.3 Remifentanil combined with less
The fentanyl family than half-MAC of isolurane allowed for emergence
Fentanyl can be used during spine surgery as a com- during surgery to facilitate intentional fracture of
ponent of induction (25 g/kg) combined with the cervical spine in two patients with chin-on-chest
another induction agent, or as an element of balanced consequences of ankylosing spondylitis.32 Among the 109
anesthesia with intermittent boluses at 2550 g every rapid-emergence opioid combinations, intraoperative
Section 1: General considerations

emergence for wake-up testing was more rapid with of anesthesia if there is hypovolemia or bronchospasm,
remifental than with alfentanil.33 and as an induction agent only would have minimal
impact on spinal cord monitoring.
Dexmedetomidine Low-dose ketamine infusion (1 g/kg/min) has
Dexmedetomidine can be used to achieve sedation and been used as an additive during major spine surgery
analgesia with alpha adrenergic agonist action when to achieve hemodynamic stability and to reduce the
administered as an infusion. he loading dose is 1 g/kg amount of other agents needed to maintain anesthe-
and is administered slowly over 20 minutes. he sed- sia.40 Motor evoked potential monitoring is not signii-
ation and analgesia achieved by the loading dose can be cantly altered by the use of ketamine.41,42
maintained with an infusion at 0.20.7 g/kg/h. Rapid
administration of the loading dose can activate vaso- Benzodiazepines
constriction and cause signiicant hypertension. he most common benzodiazepine used for major
Dexmedetomidine can be used an additive to vir- spine surgery is midazolam. Midazolam is selected as
tually any basal anesthetic, and has demonstrated the a short-acting sedative as a premedicant, or as part of a
ability to allow lower doses of the primary agents and balanced anesthetic regimen to ensure amnesia. When
rapid emergence either during or at the conclusion injected intravenously, anxiolysis peaks at 1.52.0 min-
of surgery. he total propofol was reduced and emer- utes. Midazolam is arguable the best sedative agent to
gence was more rapid when dexmedetomidine was achieve amnesia.
added to remifentanil plus propofol than with propo- Benzodiazepines have a role during major spine
fol and remifentanil alone.34 When dexmedetomidine surgery to ensure amnesia during times when anes-
infusion was introduced during deslurane/remifen- thetic doses must be kept low. However, there is
tanil anesthesia, SSEP and MEP recording was not impact on amplitude and latency with diazepam43
disturbed.35 his was also reported during total intra- and with amplitude ater midazolam with no impact
venous anesthesia.36,37 Although not an issue at lower on latency.20 his would make bolus injection of either
doses, high doses of dexmedetomidine decreased benzodiazepine unwise during intervals of critical sur-
the amplitude of transcranial motor evoked poten- gical instrumentation.
tials compared with propofol/remifentanil anesthe-
sia in children.38 his attenuation did not occur when References
the dexmedetomidine was let at the lower level and 1. Vauzella C, Stagnara P, Jouvinroux P. Functional
the propofol concentration was increased. Clinical monitoring of spinal cord activity during spinal surgery.
experience with higher doses of dexmedetomidine Clin Orthop 1973; 93: 738.
has been associated with compromise of intraopera- 2. Hall JE, Levine CR, Sudhir HG. Intraoperative
tive spinal cord monitoring using transcranial motor awakening to monitor spinal cord function during
evoked potentials.39 Harrington rod instrumentation and spinal fusion.
J Bone Joint Surg 1978; 60A: 5336.
Ketamine 3. Grottke O, Dietrich PJ, Wiegels S, Wappler F.
Ketamine has its most common application in spine Intraoperative wake up test and postoperative
emergence in patients undergoing spinal surgery: a
surgery as an induction agent. An induction dose of
comparison of intravenous and inhaled anesthetic
0.51.5 mg/kg produces amnesia, analgesia, and uncon- techniques using short acting anesthetics. Anesth Analg
sciousness. At lower doses, amnesia and analgesia can 2004; 99: 15217.
be achieved with minimal respiratory depression. he
4. McTaggart Cowan RA. Somatosensory evoked
unconsciousness is diferent from other induction potentials during spinal surgery. Can J Anaesth 1998;
agents, being more like catalepsy, and eye opening or 45: 38792.
involuntary motor activity may occur if other agents 5. Grundy BL. Intraoperative monitoring of
or nondepolarizing muscle relaxants are not coadmin- somatosensory evoked potentials. Anesthesiology 1983;
istered. An unattractive aspect of ketamine is the dys- 58: 7287.
phoria and psychomimetic properties reported ater 6. Lyon R, Lieberman JA, Grabovac ME, Hu S. Strategies
ketamine use as a sole agent. hese events are much for managing decreased motor evoked signals while
110 less likely when benzodiazepines or propofol are also distracting the spine during correction of scoliosis.
administered. Ketamine can have a role in the induction J Neurosurg Anesthesiol 2004; 16: 16770.
Chapter 7: Pharmacology of adjunct anesthetic drugs

7. Banoub M, Tetzlaf JE, Schubert A. Pharmacologic of fentanyl and nitrous oxide. Anesth Analg 1988; 67:
and physiologic inluences afecting sensory evoked 5829.
potentials. Anesthesiology 2003; 99: 71623. 21. Sloan TB, Ronai AK, Toleikis JR. Improvement of
8. Pathak KS, Ammadio BS, Scoles PV. Efects of somatosensory evoked potentials by etomidate. Anesth
halothane, enlurane, and isolurane in nitrous oxide Analg 1988; 67: 5829.
on multi-level somatosensory evoked potentials. 22. McPherson RW, Sell B, Traystman RJ. Efects of
Anesthesiology 1989; 70: 20714. thiopental, fentanyl, and etomidate on upper extremity
9. Pathak KS, Ammadio M, Kalamchi A, et al. Efects of somatosensory evoked potentials in humans.
halothane, enlurane, and isolurane on somatosensory Anesthesiology 1986; 65: 5848.
evoked potentials during nitrous oxide anesthesia.
23. Nathan N, Tabaraud F, Lacroix F. Inluence of propofol
Anesthesiology 1987; 66: 7537.
concentrations on multiple transcranial motor evoked
10. Scholz J, Bischof P, Szafarczyk W. Comparison of potentials. Br J Anaesth 2003; 91: 4939.
sevolurane and isolurane in ambulatory surgery; Results
24. Pechstein U, Nadstawek J, Zentner J. Isolurane
of a multicenter study. Anesthetist 1996; 45: 5806.
plus nitrous oxide versus propofol for recording
11. Haghighi SS, Sirintrapun SJ, Johnson JC. Suppression of of motor evoked potentials ater high frequency
spinal and cortical somatosensory evoked potentials by repetitive electrical stimulation. Electrocephalogr Clin
deslurane anesthesia. J Neurosurg Anesthesiol 1996; 8: Neurophysiol 1998; 108: 17583.
14853.
25. Kajiyama S, Sanuki M, Kinoshita H. Efect of bolus
12. Vaugha DJ, hornton C, Wright DR. Efects of propofol administration on muscle evoked potential
diferent concentrations of sevolurane and deslurane (MsEP) during spine surgery. Masui 2001; 50: 86773.
on subcortical somatosensory evoked potentials in
anesthetized, non-stimulated patients. Br J Anaesth 26. Russell D, Wilkes MP, Hunter SC, et al. Manual
2001; 86: 5967. compared to targeted infusion of propofol. Br J Anaesth
1995; 75: 5626.
13. Samra SK, Vanderzant CW, Domer PA, Sackellares
JC. Diferential efects of isolurane on human 27. Nathan N, Tabaraud F, Lacroix F, et al. Inluence of
median nerve somatosensory evoked potentials. propofol concentrations on multipulse transcranial
Anesthesiology 1987; 66: 2935. motor evoked potentials. Br J Anaesth 2003; 91:
P34937.
14. Sebel PS, Flynn PJ, Ingram DA. Efect of nitrous
oxide on visual, auditory and somatosensory evoked 28. Albertin A, LaColla L, Gandoli A, et al. Greater
potentials. Br J Anaesth 1984; 56: 14037. peripheral blood low but less bleeding with propofol
versus sevolurane during spine surgery. Spine 2008; 33:
15. Kalkman CJ, Traast H, Zuurmond WA, Bovill JG. 201722.
Diferential efects of propofol and nitrous oxide on
posterior tibial nerve somatosensory cortical evoked 29. Holzman A, Schmidt H, Gebhardt MM, et al. Propofol-
potentials during alfentanil anaesthesia. Br J Anaesth induced alterations in the microcirculation of hamster
1991; 66: 4839. striated muscle. Br J Anaesth 1995; 75: 4526.
16. Schaney CR, Sanders J, Kuhn P, LaJohn S, Heard C. 30. Schubert A, Drummond JC, Peterson DO, Saidman
Nitrous oxide with propofol reduces somatosensory LJ. he efect of high-dose fentanyl on human median
evoked potential amplitude in children and adolescents. somatosensory evoked responses. Can J Anaesth 1987;
Spine 2005; 30: 68993. 34: 3540.
17. Rodola F, DAvolio R, Chierichini, et al. Wake-up 31. Pathak KS, Brown RH, Cascorbi HF, Nash CL.
test during major spinal surgery under remifentanil Efects of fentanyl and morphine on intraoperative
balanced anaesthesia. Eur Rev Med Pharmacol Sci 2000; somatosensory cortical-evoked potentials. Anesth
4: 6770. Analg 1984; 63: 8337.
18. Shimoji K, Kano T, Nakashima H. he efects of 32. Kimball-Jones PL, Schell RM, Shook JP. he use of
thiamylal sodium on electrical activities of the central remifentanil infusion to allow intraoperative awakening
and peripheral nervous system in man. Anesthesiology for intentional fracturing of the anterior cervical spine.
1974; 40: 2349. Anesth Analg 1999; 89: 105961.
19. Ganes T, Lundar T. he efect of thiopentone on 33. Imani F, Jafarian A, Hassani V, Khan ZH. Propofol-
somatosensory evoked responses and EEGs in comatose alfentanil vs. propofol-remifentanil for posterior spinal
patients. J Neurol Neurosurg Psychiatry 1983; 46: 50914. fusion including wake-up test. Br J Anaesth 2006; 96:
5836.
20. Koht A, Schutz W, Schmidt G. Efects of etomidate,
midazolam, and thiopental on median nerve 34. Ngwenyama NE, Knanyezi E, Anderson J, 111
somatosensory evoked potentials and the additive efect Hoernschmeyer DG, Tobias JD. Efects of
Section 1: General considerations

dexmedetomidine on propofol and remifentanil infusion 39. Mahmoud M, Sadhasivam S, Sestokas AK, Samuels
rates during total intravenous anesthesia for spine P, McAulife J. Loss of transcranial electric motor
surgery in adolescents. Pediatr Anesth 2008; 18: 11905. evoked potentials during pediatric spine surgery with
35. Bala E, Sessler DI, Nair DR, et al. Motor and dexmedetomidine. Anesthesiology 2007; 106: 3936.
somatosensory potentials are well maintained in 40. Hadi BA, Al Ramadani R, Daas R, Naylor I, Zeiko R.
patients given dexmedetomidine during spine surgery. Remifentanil in combination with ketamine versus
Anesthesiology 2008; 109: 41725. remifentanil in spinal fusion surgery- a double blind
36. Anschel DJ, Aherne A, Soto RG. Successful intraoperative study. Int J Clin Pharmacol her 2010; 48: 5428.
spinal cord monitoring during scoliosis surgery using 41. Ubags LH, Kalkman CJ, Been HD, et al. he use of
a total intravenous anesthetic regimen including ketamine or etomidate to supplement sufentanil/
dexmedetomidine. J Clin Neurophysiol 2008; 25: 5661. nitrous oxide anesthesia does not disrupt the
37. Tobias JD, Goble TJ, Bates G, et al. Efects of monitoring of myogenic transcranial motor evoked
dexmedetomidine on intraoperative motor and responses. J Neurosurg Anesthesiol 1997; 9: 22833.
somatosensory evoked potential monitoring during 42. Yang LH, Lin SM, Lee WY, et al. Intraoperative
spinal surgery in adolescents. Pediatr Anesth 2008; 18: transcranial electrical motor evoked potential
108288. monitoring during spinal surgery under intravenous
38. Mahmoud M, Sadhasivam S, Salisbury S, et al. ketamine or etomidate anesthesia. Acta Neurochir 1994;
Susceptibility of transcranial electric motor-evoked 127: 1917.
potentials to varying targeted blood levels of 43. Grundy BL, Brown RH, Greenbergh BA. Diazepam
dexmedetomidine during spine surgery. Anesthesiology alters cortical potentials. Anesthesiology 1979; 51:
2010; 112: 136473. 53843.

112
Section 2 Spine surgery for adult patients
Chapter
Surgical techniques

8
8.1 Anterior cervical surgery
Iain H. Kalfas

Key points does not provide access to pathology lying in front


of the spinal cord where a majority of cervical lesions
Anterior cervical surgery is a frequently occur. Removal of the posterior supporting elements
performed spinal procedure. of the cervical spine also creates the potential for the
Indications for this surgery include development of a postoperative spinal deformity.2
degenerative, traumatic, neoplastic, and Over the past two decades, improvements and
deformity disorders of the cervical spine. developments in preoperative imaging techniques,
Surgical goals: decompression of the neural surgical microscope technology, and spinal instru-
elements followed by reconstruction and mentation devices have all contributed to a rapid rise
stabilization of the involved spinal segments. in the volume of anterior cervical surgery performed.
Anatomic complexity of the anterior cervical Approximately 350 000 anterior cervical surgeries are
region creates the potential for a variety of now performed in the United States each year. It is a far
intraoperative and postoperative complications. more commonly performed procedure than cervical
Close collaboration between the surgical laminectomy.3
and anesthesia teams before and during
anterior cervical surgery helps minimize the Indications for anterior cervical
development of these complications and to
optimize clinical outcome. surgery
he primary indication for anterior cervical surgery
is clinically correlative compression of the ventral epi-
Introduction dural space. his is most commonly due to herniated
Anterior cervical surgery is one of the more common disc material or intervertebral osteophyte (bone spur)
procedures performed by neurosurgeons and ortho- formation (spondylosis) (Fig. 8.1). Other pathologies
pedic spine surgeons. It is a versatile and efective sur- that can cause compression of the neural structures
gery that can address a wide variety of spinal disorders (spinal cord and nerve roots) include cervical trauma,
afecting the cervical spine. deformity, neoplasms, and osteomyelitis/discitis.
he anterior approach to the cervical spine was Anterior cervical surgery is also indicated when a dis-
irst developed in the 1950s as an alternative to the ruption of the cervical bone and supporting ligament-
more commonly performed laminectomy proced- ous structures has produced an unstable spinal column
ure.1 A laminectomy is performed through a posterior that potentially places the spinal cord at risk for new or
approach to the cervical spine. his approach has sev- further injury.
eral factors that limit its efectiveness for many patients Although less frequently performed, laminec-
with cervical pathology. In particular, laminectomy tomy remains an efective surgical option for some

113
Anesthesia for Spine Surgery, ed. Ehab Farag. Published by Cambridge University Press. Cambridge University Press 2012.
Section 2: Spine surgery for adult patients

Figure 8.1 Sagittal MRI demonstrating a herniated disc (arrow)


at the C56 level creating compression of the ventral aspect of the
spinal cord.

Figure 8.2 Lateral radiograph demonstrating anterior cervical


conditions. Its primary indication is for pathology pro- ixation following a discectomy and interbody fusion.
ducing compression of the dorsal epidural space. It is
also indicated for most intradural pathology such as
he most common type of anterior cervical surgery
spinal cord neoplasms and vascular malformations.
is a discectomy which involves the removal of a single or
Rarely, both an anterior and a posterior approach
multiple intervertebral discs. his procedure is typically
may be necessary to optimally manage cervical path-
performed for symptoms resulting from a cervical disc
ology. his combined approach is typically indicated
herniation or from spondylosis (bone spur formation).
when there is a marked deformity of the cervical spine
Less frequently, one or several vertebrae as well as their
in addition to signiicant epidural compression. It
adjacent intervertebral discs may need to be removed
may also be indicated for added stabilization follow-
to adequately decompress the neural structures. his
ing extensive anterior decompression requiring the
procedure, termed corpectomy, is typically used for
removal of several vertebrae.
the management of cervical trauma and neoplasms that
afect the structural integrity of the vertebral bodies.
Surgical technique overview Following the appropriate decompression of the
he anterior cervical surgical approach can be used neural elements through either a discectomy or a cor-
to access the cervical spine from the C23 disc space pectomy procedure, the involved spinal column must
superiorly to the C7T1 disc space inferiorly. he sur- be reconstructed. his involves placing one or several
gical dissection and approach follows a natural tissue structural bone grats into the site of the discectomy or
plane lying between the trachea and esophagus medi- corpectomy. Alternatively, a metal or carbon iber cage
ally and the sternocleidomastoid muscle and carotid can be used instead of a bone grat. A spinal ixation
artery laterally. hese structures can be gently retracted plate and screws helps secure the bone grat or cage in
114 to provide a generous exposure of the anterior surface place and adds additional structural support to the sur-
of the cervical spine. gical site (Fig. 8.2).
Chapter 8.1: Anterior cervical surgery

Postoperatively, patients are typically placed in a period of transient hypotension in these patients may
hard cervical collar for a short period. Patients requir- signiicantly increase the potential for hypoperfusion
ing more extensive spinal reconstruction may require a of a spinal cord that already has its vascular supply
longer period of postoperative bracing. Rarely, patients compromised by the compressive pathology.
with major spinal reconstruction may need to be placed Following induction, the patients head and neck
into a halo immobilization brace for several months. are placed in a neutral to slightly extended position.
A relatively recent development in anterior cer- he endotracheal tubing is positioned so that it extends
vical surgery is the insertion of an artiicial disc.4 his directly over the patients forehead. his allows for opti-
procedure, termed cervical arthroplasty, is indicated mal access to both sides of the anterior cervical region
for the treatment of single-level disc degeneration or by the surgical team. Some surgeons prefer to place
herniation in patients with radiculopathy. It involves their patients into gentle traction with a halter sling
performing a discectomy and decompression of the that its under the chin and the back of the skull.
epidural space followed by the insertion of an articu- he administration of corticosteroids is optional.
lating device into the disc space. he theoretical advan- In selected patients with severe neurologic impairment
tage of arthroplasty is that it preserves segmental spinal due to signiicant neural compression, preoperative
motion and potentially reduces the incidence of degen- administration of corticosteroids may protect the spi-
eration of the adjacent intervertebral discs. his devel- nal cord during surgery, although limited data exist to
opment of adjacent-level changes has been noted to support this approach. Corticosteroids may also help
occur at a rate of up to 3% per year following anterior to minimize sot tissue edema and airway issues result-
cervical fusion surgery.5 ing from surgical dissection and retraction. he disad-
vantage of the use of steroids is that they interfere with
the early stages of bone healing and may compromise
Immediate preoperative management the long-term fusion outcome of the procedure.
Several general principles apply to all patients undergo- he routine use of intraoperative electrophysio-
ing anterior cervical surgery. Although most patients logical monitoring (somatosensory or motor evoked
with anterior cervical pathology can undergo gentle potentials) during anterior cervical surgery is not con-
extension of the head and neck for endotracheal intu- clusive. Advocates of monitoring claim that its use low-
bation, some patients will require that any extension ers the risk for postoperative neurologic deicits. Motor
be limited or not used at all. his is typically the case evoked potentials have been found to be far more help-
in patients with a traumatic cervical injury that has ful than somatosensory evoked potentials.6
destabilized their spinal column. he argument against the routine use of evoked
Another less common contraindication to cervi- potential monitoring is that since the risk of neuro-
cal extension is severe spinal canal stenosis. hese logic deterioration following anterior cervical surgery
patients may exhibit a inding termed Lhermittes phe- is relatively low without monitoring, the additional
nomenon. his neurologic sign occurs when afected time and expense involved with monitoring may not
patients extend their neck and experience a shocklike justify its use for every case. he use of monitoring is
shooting down their spinal column or into their arms best determined by each individual surgeons personal
and legs. It indicates the presence of signiicant spinal preference and the speciic surgical problem being
cord compression with a high degree of sensitivity to addressed in each patient.
motion.
Any patient with a contraindication to cervical
extension should be considered for an awake intubation, Surgical management
preferably with iberoptic assistance. Communication Anterior cervical surgery can be performed through
between the spinal surgeon and the anesthesiologist either a transverse incision extending from the trachea
prior to intubation is advised to help assure the selec- to the sternocleidomastoid muscle or an oblique inci-
tion of the most appropriate intubation approach for sion made along the medial border of the sternocleido-
each patient. mastoid muscle. Although a right-sided approach is
During the induction process it is critical to avoid technically easier for a right-handed surgeon, the let-
any prolonged period of hypotension. his is particu- sided approach to the cervical spine may ofer some
larly important in patients who present with a myelop- minimal reduction to the risk of injury to the recur- 115
athy due to compression of the spinal cord. Even a brief rent laryngeal nerve. he recurrent laryngeal nerve on
Section 2: Spine surgery for adult patients

the right side lies between the trachea and esophagus the neural structures. his is typically performed with
but can take an aberrant course and lie more rostral the aid of a surgical microscope using a high-speed drill
than the nerve on the let side. his anatomic variance and a series of bone rongeurs. Although epidural ven-
may make the nerve more prone to a retraction injury ous bleeding can be encountered it is rarely excessive
with a right-sided approach. Alternatively, a let-sided and relatively easy to control. It typically subsides once
approach risks injury to the thoracic duct. adequate decompression has been achieved. Leakage of
Early in the dissection process, the carotid artery cerebrospinal luid can also occur during decompres-
is identiied and retracted laterally. Injury to this ves- sion but is relatively uncommon and can be controlled
sel is very rare. Any bleeding that occurs during the with the application of a dural sealant material.
approach is typically venous in origin and relatively Following adequate decompression, the structural
easily controlled. he vertebral arteries are typically stability of the spinal column must be reestablished. his
not visualized during the procedure unless the disc- involves placement of a suitable bone grat(s) or syn-
ectomy or corpectomy is extended too widely. thetic cage(s) into the site of the discectomy or corpec-
Following exposure of the spinal column, self- tomy. In the past, bone grats were typically harvested
retaining retractor blades are positioned to retract the from the patients iliac crest or ibula. he morbidity and
trachea and esophagus medially and the carotid artery pain associated with this process as well as the improve-
and jugular vein laterally. he retractor blades are posi- ment in tissue banking practices has made the use of
tioned under the longus colli muscles which run lon- banked allograt bone much more common today.
gitudinally on either side of the cervical spine midline. he appropriately sized grat(s) or cages(s) are
If these retractor blades are allowed to migrate from placed between the vertebral end plates on either side
their submuscular position, they may cause a perfor- of the decompressed site. If halter traction has not been
ating injury to the esophagus or excessive retraction used, placement of a grat or cage into the site of recon-
of the trachea. A rare complication of retractor blade struction may require the anesthesia team to provide a
migration is the development of a unilateral Horners brief period of manual inline cervical traction.
syndrome due to encroachment on the sympathetic Once the bone grat(s) or cages(s) are in pos-
chain lateral to the longus colli muscles. ition, anterior cervical plate ixation is placed to fur-
Rarely, excessive pressure on the carotid artery by ther secure them (Fig. 8.2) hese titanium plates are
the lateral retractor blade may potentially compromise attached to the spine by screws inserted into the verte-
blood low through the vessel. Palpation of the superi- brae on either side of the reconstructed site. In addition
cial temporal arteries may alert the anesthesia team to to enhancing the fusion rate, plate ixation also mini-
a possibility of reduced low necessitating a reposition- mizes the need for postoperative bracing.8,9
ing of the retractor blades by the surgical team. Correct
initial placement of the lateral retractor blade beneath Immediate postoperative
the longus colli muscle minimizes the risk of reduced management
carotid blood low. Postoperatively, most patients can be extubated in the
Retraction of the trachea has been proposed to operating room. Following more extensive multilevel
potentially lead to vocal cord paralysis due to endo- corpectomy procedures in which there may be excessive
tracheal cuf pressure on the recurrent laryngeal nerve sot tissue edema it may be necessary to leave the patient
within the endolarynx. In an efort to reduce the inci- intubated until the edema subsides. Most of these patients
dence of vocal cord paralysis, one study proposed that can be extubated 1218 hours following surgery.
immediately following tracheal retraction, the cuf Although complications from anterior cervical sur-
pressure should be released and then re-inlated to the gery are relatively rare, the anatomic complexity of this
just-sealed pressure of 15 mmHg. he cuf pressure is region allows for the potential development of a variety
monitored throughout the procedure and delated with of diferent problems. he most serious complication
any further increases in pressure. his approach resulted following this surgery is airway compromise and frank
in a reduction in the incidence of vocal cord paralysis obstruction requiring urgent re-intubation or trache-
from 6.4% to 1.7% in a series of 900 consecutive patients ostomy. his may occur in the immediate postopera-
who underwent anterior cervical spine surgery.7 tive period or several days ater surgery. he common
116 Surgical decompression proceeds with removal of causes of airway compromise following anterior cervi-
the disc or bone material producing encroachment on cal surgery are pharyngeal edema, wound hematoma,
Chapter 8.1: Anterior cervical surgery

ixation plate and bone grat. Injury to the spinal cord


or nerve roots is relatively rare but can occur. he most
common long-term complication is failure of the bone
grat to fuse. A review of 4589 cases in the Cervical Spine
Research Society database revealed a total complication
rate of 5.3% following anterior cervical surgery.13
he most common neurologic complication fol-
lowing cervical spine surgery is C5 nerve root palsy. A
meta-analysis report indicated an average incidence of
C5 dermatomal weakness following anterior cervical
surgery of 4.3%.14 Most (92%) C5 palsies are unilateral
and occur within the irst week of surgery. Although the
pathogenesis for selective C5 involvement is unclear, sev-
eral theories have been promoted generally based on the
roots speciic anatomic location and course. Proposed
mechanisms include traction of the root produced by
a shiting of the spinal cord following decompression,
ischemia due to involvement of the C5 radicular artery,
and lack of cross-innervation of the deltoid muscles
which are primarily supplied by the C5 root.14,15 he
prognosis for spontaneous recovery of a C5 motor radic-
ulopathy is relatively good in most cases.
Postoperatively, most patients are placed into a
cervical collar for a short period (12 weeks). hey are
typically discharged from the hospital 12 days post-
operatively. he clinical outcome for patients under-
going anterior cervical surgery for radiculopathy is
Figure 8.3 Lateral radiograph demonstrating dislodgement of an
anterior ixation device and the bone graft following a three-level very good, with over 90% achieving a successful out-
corpectomy procedure. come.1 Patients who undergo anterior cervical surgery
for myelopathy have a clinical improvement rate of
or dislodgement of the bone grat and ixation plate approximately 70%.13
(Fig. 8.3).10,11
Sagi et al. reported a retrospective series of 311 pat- Conclusion
ients who underwent anterior cervical surgery. Nineteen Anterior cervical surgery is a common spinal pro-
patients (6.1%) had an airway complication and six (1.9%) cedure. Its use for a variety of cervical disorders has
required re-intubation. One patient died. All airway expanded with advancements in surgical microscopy,
complications except two were attributable to pharyn- ixation devices and bone grating technologies. he
geal edema. Variables that were found to be statistically anatomic complexity of this region requires a close col-
associated with airway complications following anterior laboration between the surgical and anesthesia teams
cervical surgery were exposure of more than three ver- in order to limit the development of complications and
tebral bodies, blood loss >300 ml, exposure of the upper to optimize the clinical outcome following surgery.
cervical spine (C24) and an operative time >5 hours.
A history of myelopathy, spinal cord injury, pulmonary References
problems, smoking, anesthetic risk factors, and the 1. Cloward RB. he anterior approach for removal of
absence of a wound drain did not correlate with a higher ruptured cervical discs. J Neurosurg 1958; 15: 60217.
incidence of postoperative airway complications.12
2. Raynor RB. Anterior and posterior approaches to the
In addition to airway obstruction other potential cervical spinal cord, discs, and roots: a comparison of
complications include transient sore throat, dysphagia, exposures and decompression. In: he Cervical Spine
hoarseness, dysphonia, vocal cord paralysis, esophageal Research Society, Editorial Committee, ed. he Cervical 117
perforation, wound hematoma, and dislodgement of the Spine. Philadelphia: JB Lippincott; 1989: 65969.
Section 2: Spine surgery for adult patients

3. Wang MC, Kreuter W, Wolla CE, et al. Trends and 9. Kaiser MG, Haid RW, Subach BR, et al. Anterior cervical
variations in cervical spine surgery in the United States. plating enhances arthrodesis ater discectomy and
Medicare beneiciaries 19922005. Spine 2009; 9: 95561. fusion with cortical allograt. Neurosurgery 2002; 50:
4. Goin J, Van Clenbergh F, van Loon J, et al. 22938.
Intermediate follow-up ater treatment of degenerative 10. Emery SE, Smith MD, Bohlman HH. Upper airway
disc disease with the Bryan cervical disc prosthesis: management ater multi-level corpectomy for
single and bi-level. Spine 2003; 28: 26738. myelopathy. J Bone Joint Surg Am 1991; 73: 54450.
5. Hilibrand AS, Carlson GD, Palumbo MA, et al. 11. Bookvar JA, Philips MF, Telfeian AE. Results and risk
Radiculopathy and myelopathy at segments adjacent factors for anterior cervicothoracic junction surgery.
to the site of a previous anterior cervical arthrodesis. J Neurosurg 2001; 94: 1217.
J Bone Joint Surg 1999; 81: 51928. 12. Sagi HC, Beutler W, Carroll E, et al. Airway
6. Hilibrand AS, Schwartz DM, Sethuraman V, et al. complications associated with surgery on the anterior
Comparison of transcranial electric motor and cervical spine. Spine 2002; 9: 94953.
somatosensory evoked potential monitoring during 13. Zeidman SE, Ducker TB, Raycrot J. Trends and
cervical spine surgery. J Bone Joint Surg 2004; 86: 124853. complications in cervical spine surgery: 19891993.
7. Apfelbaum RI, Kriskovich MD, Haller JR. On the J Spinal Disord 1997; 10: 5236.
incidence, cause and prevention of recurrent laryngeal 14. Sakaura H, Hosono N, Mukai Y, et al. C5 palsy ater
nerve palsies during anterior cervical spine surgery. decompression surgery for cervical myelopathy. Spine
Spine 2000; 25: 290612. 2003; 28: 244751.
8. Kalfas IH. he anterior cervical spine locking plate: a 15. Kaneko K, Hashiguchi A, Kato Y, et al. Investigation of
technique for surgical decompression and stabilization. motor dominant C5 paralysis ater laminoplasty from
In: Fessler RG, Haid RW, eds. Techniques in Spinal the results of evoked spinal cord responses. J Spinal
Stabilization. New York: McGraw-Hill; 1996: 2533. Disord Tech 2006; 19: 35861.

8.2 Posterior cervical surgery


Kalil G. Abdullah, Jefrey G. Clark, Daniel Lubelski, and homas E. Mroz

Key points Introduction


Posterior cervical spine surgery is among the most
Posterior surgical spine surgery is a common
common procedures performed by spine surgeons. A
elective procedure that is indicated for
variety of indications require a posterior approach, of
decompression and stabilization of the spinal
which several diferent techniques may be employed
cord and exiting nerve roots.
to decompress and reconstruct the spine follow-
It is essential to maintain vigilance in the ing trauma, degeneration, neoplasm, or infection.
monitoring and positioning of the cervical Posterior spine surgery at any level elicits a set of con-
spine surgery patient, whether elective or siderations for the surgical team, but posterior cervical
traumatic due to the possibility of sudden spine surgery in particular requires vigilance with
neurologic deterioration. regard to positioning, intubation, and intraoperative
Careful positioning is paramount in this patient hemodynamic status. In this chapter, we briely intro-
population to allow for appropriate surgical and duce indications for posterior spinal surgery, patient
airway access. positioning, and speciic considerations necessary for
Induction of hypotension to protect from blood safe anesthesia in these patients.
loss is generally unnecessary and can result in
ischemic compromise to the spinal cord. Indications for posterior spine surgery
During traumatic intervention, mean arterial Posterior surgery is indicated when exposure of
pressure should be carefully monitored and the dorsal elements of the spine or spinal cord is
118 kept above 90 mmHg. required. Among the most common indications for
Chapter 8.2: Posterior cervical surgery

posterior surgery is cervical spondylotic myelopathy surgery must be handled with the understanding that
and radiculopathy. In these cases, the cervical spine catastrophic neurologic deterioration is a possibility
is exposed through a midline incision and the bony if positioning, intubation, and surgical handling are
elements including the vertebral process, lamina, and compromised. Careful collaboration with the surgical
foramina are visualized and then oten modiied (i.e., team prior to and during intubation and positioning is
laminoplasty) or removed (i.e., laminectomy, forami- mandatory.
notomy) to decompress the exiting nerve roots and/
or spinal cord. In general there are two main catego- Patient positioning
ries of posterior surgery: decompression and fusion. Prone positioning is the standard for posterior cer-
Laminoforaminotomy (i.e., removing a small por- vical surgery. Historically, the sitting position was
tion of the unilateral lamina and facet joint) addresses occasionally used but the rates of complications (e.g.,
foraminal stenosis (i.e., nerve root compression) at one air emboli, ischemia, instability) were unacceptably
or more levels and does not require fusion because it high. In the extremely unlikely case of morbid obesity
does not involve removing spinous processes or inter- or ventilatory restriction requiring a sitting position,
spinous ligaments, and hence, is not a destabilizing pro- then Doppler ultrasound is an option for embolism
cedure. Laminectomy is typically reserved for patients monitoring.1
with spinal cord compression at greater than three lev- Most patients are placed in cranial tongs for poster-
els and is typically followed by an instrumented fusion. ior cervical surgery, but some surgeons prefer a foam
Laminectomy without fusion has fallen out of favor pillow. Advantages of the cranial tongs include inten-
due to the known association with postlaminectomy tional intraoperative positional changes, and absence
kyphosis, and possibly late neurologic deterioration as of facial or ocular pressure. he decision for an awake
a consequence of the structural demise. he purpose of iberoptic intubation should be made ater careful con-
the instrumentation is to provide provisional stability sideration of the cervical stability and the preoperative
until biological fusion occurs across the surgical lev- active range of motion. In patients who are deemed
els. Laminoplasty, used to treat myelopathy, is a pro- unstable or unable to neurologically tolerate cervical
cedure that involves making precise cuts on bilateral stenosis (i.e., severe stenosis) by the treating sur-
laminae at multiple levels, hinging the laminae open, geon, an awake iberoptic intubation is indicated. It is
and then ixing them in that position with small plates important to assess preoperative active range of motion
or sutures. his procedure does not involve fusion, in patients with myelopathy regardless of etiology. If
and is considered as eicacious as laminectomy and active cervical extension or lexion result in subjective
fusion for the treatment of cervical myelopathy with or objective neurologic compromise, then an awake
and without radiculopathy. he majority of posterior iberoptic intubation is indicated to avoid injury.
surgery involves the subaxial spine (C3T1); however, To prepare the prone elective patient, intubation is
occipital-cervical surgery is not uncommon. Certainly, irst accomplished and the patient is turned from supine
surgical approaches may vary signiicantly to address to prone using a log roll technique, maintaining the neck
infection or neoplasm, but the common principles of in a neutral position. Typically, a member of the sur-
posterior spine surgery include the need to expose, gical team controls the head during the turn while the
decompress, and possibly, perform an instrumented anesthesiologist manages the airway. To avoid elevated
fusion. thoracic or abdominal pressures, cushions are placed
Trauma to the cervical spine and instability due to along the margins of the upper torso. he surgeon then
infection or neoplasia has speciic ramiications for adjusts and negotiates the Mayield clamp to secure the
the surgical team. Whenever destabilization occurs head and cervical spine in the desired position, and the
in the cervical spine secondary to trauma, a multi- bed may be placed in slight reverse Trendelenburg.2 It
tude of factors must be taken into consideration (the is important that the endotracheal tube is secured and
relevant components are discussed below). In certain guided appropriately so that it is unobstructed and not
situations, it may be necessary to stage surgical inter- in competition with any other apparatus. Unlike in sur-
vention to include anterior and posterior approaches, gical situations where the patient is supine, inherent
but that discussion is beyond the scope of this chapter. diiculty will arise should the tube need to be adjusted
Regardless of future intervention or perioperative care, while the surgeon is operating on the prone patient with
each patient with cervical trauma undergoing posterior a possibly unstable spine. 119
Section 2: Spine surgery for adult patients

For those patients undergoing intervention follow- When the patient is placed in the prone position there
ing trauma, the positioning process is similar. Should is oten a drop in blood pressure. Dharmavaram and
these patients already have undergone halo ixation, it colleagues conducted a study comparing prone posi-
allows for more controlled movement of the patient from tioning systems on hemodynamic and cardiac func-
supine to prone. Patients in the emergent setting without tion. hey found various changes in cardiac function
ixation must be moved with typical stabilizing technique dependent on the manufacturer of the operating room
and can then be placed into traction by the surgeon. table; some tables decreased cardiac index and stroke
volume, whereas others contributed to a decrease in
Elective interventions cardiac preload. Cardiac output was reduced in all of
he most common elective procedures include lamino- them.6 he mechanism of action is thought to be poor
foraminotomy, laminectomy and fusion, and lamino- venous return and a change in ventricular compliance.
plasty for spondylotic myelopathy and radiculopathy. Even moderate hypotension may exacerbate spinal
Spinal instability is rare in cases of spondylotic myel- cord injury that is produced from manipulation or dis-
opathy or radiculopathy involving the subaxial cervical traction of the region during surgery. Understandably,
spine. However, as mentioned, it is important to pre- in the hypotensive state there is limited oxygen sup-
operatively assess the active range of motion to ensure ply and impaired clearance of metabolites. he subse-
that the patient does not neurologically decompensate quent elevated levels of carbon dioxide and lactic acid
in certain neck positions. Patients with rheumatoid lead to lower pH and concomitant damage to the cells.
disease, however, can have subaxial or atlantoaxial hese changes are irreversible, and the most successful
(C12) instability, and it is important to preoperatively avoidance of perioperative spinal cord injury is well-
discuss this particular aspect of the case with the sur- executed preventative methods and communication
geon.3,4 Certain maneuvers can narrow the spinal canal between the surgical and anesthetic teams.
depending on underlying pathology. Patients with It is important to maintain perfusion of the spinal
subaxial spondylotic myelopathy can undergo narrow- cord in both acute injury and elective spine surgery.
ing of the spinal canal during neck extension, while According to a Practice Advisory issued by a task force
those with atlantoaxial subluxation will have narrow- of the American Society of Anesthesiologists, blood
ing of the spinal canal during lexion. In these patients, pressure during induction of deliberate hypotension
a iberoptic intubation is preferred. in healthy, nonhypotensive patients should be main-
he elective patient is unlikely to require additional tained within 24% of baseline mean arterial pressure
consideration during the intubation process. he anes- (MAP) or with a minimum systolic blood pressure of
thetic considerations as to whether to use a iberoptic 84 mmHg (range 50120 mmHg).7 hese guidelines
intubation or routine direct laryngoscopy will mostly should absolutely be paired with clinical judgment.
be due to the patients other comorbidities or underly- Speciically, the deinition of healthy in the above
ing disease process (i.e., rheumatoid arthritis, Downs report should be contextualized. Patients undergo-
syndrome) and not the speciic planned surgical inter- ing elective spine surgery are of a demographic that
vention. It should be noted that there exists a difer- includes many comorbidities, which include obesity,
ential amount of cervical motion that occurs in the peripheral vascular disease, diabetes, and other entities
upper cervical vertebrae when using diferent intuba- that negatively inluence perfusion. As such, we rec-
tion techniques. Sahin et al.5 found, as expected, that ommend that MAP be manipulated at a higher range
iberoptic laryngoscopy resulted in a signiicantly (8090 mmHg MAP) to account for these factors. In a
smaller range of motion between vertebrae when com- systematic review by Ahn and Fehlings, a MAP >8085
pared with direct intubation. mmHg was recommended, and it was emphasized that
Patients undergoing elective procedures may patients with preexisting SCI or at high risk for SCI
require hemodynamic monitoring with a radial artery intraoperatively should be maintained at this MAP,
catheter, and patients with signiicant comorbidities even if use of vasopressors is needed. Minimizing blood
may require more invasive monitoring. he patient loss in nearly all patients undergoing elective surgery of
should be carefully observed to ensure adequate per- the posterior cervical spine should not be an anesthetic
fusion of the spinal cord. States of low blood low are priority. hese surgeries typically involve minimal
oten seen during spinal surgery as a result of an efect blood loss and in many ordinary surgeries requir-
120 of the anesthetic drugs or the positioning of the patient. ing posterior intervention it is far more important to
Chapter 8.2: Posterior cervical surgery

maintain perfusion to vital respiratory centers through ofers additional stability. his is particularly helpful in
an appropriate MAP.8 patients with ankylosing spondylitis-related fractures
It is oten misunderstood that hypotension is the in whom the cervical spines are frequently very unsta-
greatest contributor to postoperative vision loss during ble. It is imperative that the surgeon and anesthesiolo-
prone position spine surgery.9. In the only study that gist have a thorough discussion of the intubation and
has examined this phenomenon in a casecontrol man- surgical plan prior to the intubation.
ner, the authors found no diference in blood pressure Patients with infections and tumors involving the
between the group of patients with or without post- spinal column should be considered to have unstable
operative vision loss. Instead, length of surgery and cervical spines unless stated otherwise by the surgeon.
overall blood loss were statistically signiicant factors.10 here a multitude of diferent surgeries used for these
Nonetheless, vigilance for postoperative vision loss is entities; however, the key tenets of decompressing
an absolutely essential component of spine surgery and the neural elements, resecting pathologic tissue, and
has been reported in many diferent types of patient reconstructing the spine are followed. he manage-
population, and is discussed at length elsewhere in this ment of these patients from an anesthesia standpoint is
book.10 Signs of hemodynamic instability should be similar to patients undergoing elective procedures.
attended to in the usual fashion, with alertness for both Spinal cord injuries require special hemodynamic
arterial hypo- and hypertension, and swings in blood considerations. Traumatic cord injuries involve sec-
pressure should be avoided. ondary mechanisms of injury, including hypoper-
fusion. It is extremely important to maintain proper
Traumatic interventions perfusion through careful control of the blood pres-
he patterns of traumatic cervical spine injuries vary, sure during surgery. In those patients where spinal
but all patients with cervical injuries (i.e., fractures, cord injury is known or assumed, perfusion of the
subluxations, and dislocations) should be managed cord is paramount. As in elective procedures perfusion
the same in terms of intubation and positioning. In the should be emphasized over blood loss. To start, MAP
emergent setting, the patient usually presents to the should be greater than 90 mmHg. he pressure should
operating room ater an appropriate traumatic work- be maintained at this level, which may involve the use
up by the surgical team. By this point, the neuro- of vasopressors such as norepinephrine and dopamine
logic status of the patient should have been assessed. (epinephrine should not be used as it has alpha recep-
However, there is debate as to whether or not intuba- tor ainity, possible decreasing cord perfusion).
tion in the operating room should be performed using Vigilance to the hemodynamic state is very impor-
iberoptic awake intubation. his provides the advan- tant during traumatic posterior cervical spine surgery.
tage of assessing neurologic function ater intubation, Injuries to the cervical spinal cord can result in neu-
and can result in minimal displacement of the neck. rogenic shock due to a loss of sympathetic tone below
It is usually done under light sedation and requires the site of the injury. his decreases or eliminates the
a cooperative patient. he complicating aspects of ability of the cord to autoregulate perfusion. Studies
awake intubation are related to risks of aspiration suggest that the severity of abnormal cardiac control
and a prolonged induction time. Further, this may be correlates well with the severity of cervical spinal cord
an understandably uncomfortable experience for the injury14 but is not related to the location of the injury
conscious patient, who may not be in a fully coopera- in the upper (C15) or lower (C67) cervical spine.15
tive state. Several other methods of intubation have hus, regardless of the region of injury within the cer-
been proposed in case reports and small series,11,12 vical spine, awareness of these phenomena should be
and a detailed algorithm for the appropriate methods maintained. Neurogenic shock should be immediately
of induction following cervical trauma (including the treated with luids to keep central venous pressure
decision as to whether or not to employ rapid-sequence within 46 mmHg. Vasopressor infusions can then be
induction and appropriate anesthetic regimens) can be given to maintain adequate spinal cord perfusion.
found elsewhere.13 Regardless of the type of intubation,
immobilization of the cervical spine is paramount Summary
and must remain an urgent priority during the intu- he patient undergoing posterior spine surgery
bation process. It is advantageous to place the patient requires a special level of vigilance by the surgical team.
in a rigid cervical collar prior to positioning, as this Complications that may arise during intubation or 121
Section 2: Spine surgery for adult patients

preparation of the patient can be due to trauma or under- a report by the American Society of Anesthesiologists
lying pathology present in the cervical spine. hose Task Force on Perioperative Blindness. Anesthesiology
patients presenting with spinal trauma must be assessed 2006; 104(6): 131928.
appropriately preoperatively and the intubation must 8. Ahn H, Fehlings MG. Prevention, identiication, and
proceed with minimal disruption of the spine. treatment of perioperative spinal cord injury. Neurosurg
Focus 2008; 25(5): E15.
References 9. Baig MN, Lubow M, Immesoete P, Bergese SD, Hamdy
1. Mayer HM. Minimally Invasive Spine Surgery. 2nd ed. EA, Mendel E. Vision loss ater spine surgery: review of
New York: Springer; 2006. the literature and recommendations. Neurosurg Focus
2007; 23(5): E15.
2. Denaro L, DAvella D, Denaro V. Pitfalls in Cervical
Spine Surgery. Berlin: Springer; 2010. 10. Myers MA, Hamilton SR, Bogosian AJ, Smith CH,
Wagner TA. Visual loss as a complication of spine
3. Kim KA, Wang MY. Anesthetic considerations in the surgery. A review of 37 cases. Spine (Phila Pa 1976)
treatment of cervical myelopathy. Spine J 2006; 6(6 1997; 22(12): 13259.
Suppl): 207S11S.
11. Avitsian R, Lin J, Lotto M, Ebrahim Z.
4. Wattenmaker I, Concepcion M, Hibberd P, Lipson
Dexmedetomidine and awake iberoptic intubation for
S. Upper-airway obstruction and perioperative
possible cervical spine myelopathy: a clinical series.
management of the airway in patients managed
J Neurosurg Anesthesiol 2005; 17(2): 979.
with posterior operations on the cervical spine for
rheumatoid arthritis. J Bone Joint Surg Am 1994; 76(3): 12. Schuschnig C, Waltl B, Erlacher W, Reddy B, Stoik W,
3605. Kapral S. Intubating laryngeal mask and rapid sequence
induction in patients with cervical spine injury.
5. Sahin A, Salman MA, Erden IA, Aypar U. Upper cervical
Anaesthesia 1999; 54(8): 7937.
vertebrae movement during intubating laryngeal mask,
ibreoptic and direct laryngoscopy: a video-luoroscopic 13. Raw DA, Beattie JK, Hunter JM. Anaesthesia for spinal
study. Eur J Anaesthesiol 2004; 21(10): 81923. surgery in adults. Br J Anaesth 2003; 91(6): 886904.
6. Dharmavaram S, Jellish WS, Nockels RP, et al. Efect 14. Tuli S, Tuli J, Coleman WP, Geisler FH, Krassioukov
of prone positioning systems on hemodynamic and A. Hemodynamic parameters and timing of surgical
cardiac function during lumbar spine surgery: an decompression in acute cervical spinal cord injury.
echocardiographic study. Spine (Phila Pa 1976) 2006; J Spinal Cord Med 2007; 30(5): 48290.
31(12): 138893; discussion 1394. 15. Bilello JF, Davis JW, Cunningham MA, Groom TF,
7. American Society of Anesthesiologists Task Force Lemaster D, Sue LP. Cervical spinal cord injury and the
on Perioperative Blindness. Practice advisory for need for cardiovascular intervention. Arch Surg 2003;
perioperative visual loss associated with spine surgery: 138(10): 11279.

8.3 Intraoperative neurophysiologic monitoring:


surgeons point of view
Manuel Saavedra and Robert F. McLain

Key points Dorsal decompression addresses dorsal


columns most directly and SEPs are oten
Spinal monitoring is provided at the surgeons
adequate. Volar compression is most common
discretion there is no standard of care.
with tumors and fractures, and requires
Motor evoked potentials (MEPs) can be
anterior column decompression. MEPs are
obtained with partial relaxation when the
recommended in these cases.
patient is not paralyzed, meaning that neural
When changes in latencies or amplitude are
responses to pressure, heat, or electroconduction
recognized during spinal cord decompression
are not absent. he surgeon still gets immediate
122 or spinal instrumentation, the irst step is
feedback from surgical maneuvers that may
to check for equipment errors, then quickly
impinge on neural tissues.
Chapter 8.3: Intraoperative neurophysiologic monitoring

monitoring is during spinal cord or nerve root manipu-


assess medication issues, patient temperature,
lation. Patients who already exhibit signs of spinal cord
oxygenation, and perfusion pressure. All of these
distress or compression are particularly at risk during
should be optimized immediately at the same
surgery, and monitoring of these cases is most helpful.
time the surgeon is informed of the change.
In cervical spine surgery, electrophysiologic moni-
In cases where the altered recordings appear real toring is oten requested simply to ensure that rare events
and directly associated with the spinal condition, are not missed. During cervical decompression, EMG
surgical implants need to be assessed for direct monitoring may help avoid stress on the C5 nerve roots,
impingement on the neural structures, spinal one of which oten experiences injury during surgery.
correction assessed for over-distraction of the In cases of cervical trauma, tumor, or infection, SSEPs
cord, and margins of decompression assessed for and MEPs may detect subtle changes in cord pressure
unrecognized bone or sot tissue compression. due to positioning, surgical decompression, or changes
Particular attention needs to paid to SCM signals in blood pressure. SSEP and MEP monitoring are most
that wax and wane directly with changes in blood oten requested in patients with cervical myelopathy,
pressure. Patients with labile systolic pressure however diicult the signals may be to obtain in patients
and borderline cord perfusion are at risk for with chronic disease. Real-time monitoring of cord func-
postoperative cord stroke if their postoperative tion can help prevent permanent spinal cord injury.
blood pressure is allowed to drit downward in In thoracic spine surgery, monitoring is also used
recovery or postoperative care. Intraoperative in the face of spinal cord compression due to tumor,
monitoring can identify this risk and alert the infection, or trauma, but is also useful to avoid injury
anesthesiologist to the need for tight pressure when placing spinal implants such as hooks or pedi-
management. cle screws. In these patient changes in cord perfusion
can result in signiicant cord injury, and maintenance
of blood pressure as well as avoidance of physical pres-
Introduction sure on the cord may result in a much better outcome,
Intraoperative neurophysiologic monitoring has been facilitated by careful neurophysiologic monitoring.
used in spine surgery to minimize injury to critical While lumbar surgery usually avoids the spinal cord
neural structures from operative manipulations. Since itself, the conus medullaris is at risk in upper lumbar sur-
the irst recording of somatosensory evoked potentials gery. Likewise speciic nerve roots may be at risk during
in 1947 by Dawson,1 the evolution of intraoperative decompression or manipulation of the lumbar spine.
monitoring has advanced dramatically. he advent of Placement of pedicle screws and other spinal implants
diferent types of techniques and its combinations has may stress the nerve roots and can potentially damage
made possible monitoring of both motor and sensory them directly. EMG monitoring as well as SEP monitor-
nervous pathways according to the requirements of a ing can give important feedback during lumbar surgery.
variety of surgical procedures. Deformity surgery is particularly challenging in
his chapter provides an overview of the various pediatric and adult patients. Congenital deformities are
neurophysiologic monitoring techniques used intra- oten associated with neurologic abnormalities, such
operatively, including somatosensory evoked poten- as tethered cord or diastematomyelia, and even small
tials (SSEPs), motor evoked potentials (MEPs), and changes in spinal alignment may stress the cord and
electromyography (EMG). Challenges of neurophysi- cause neurologic deicit. Correction of large idiopathic
ologic monitoring include the presence of electromag- curves may involve manipulation of the thoracic spinal
netic interference and the use of anesthetic agents that cord or even the cervical level, and electrophysiologic
can alter recordings. monitoring may identify stress in the spinal cord long
before it can be identiied in any clinically helpful way.
Indications for spinal monitoring he combination of anterior surgery involving liga-
Depending on the surgery planned, the surgeon may tion of small segmental vessels with a large posterior
request SSEP, MEP, EMG monitoring, or a combina- surgery involving correction of multiple segments of
tion of approaches in support. While these studies the spine can put the cord at high risk for injury. Spinal
may be helpful in many diferent kinds of cases, the monitoring during this sort of surgery can be par-
most common reason to request electrophysiologic ticularly important in determining when correction 123
Section 2: Spine surgery for adult patients

is too much, and in making sure that blood pressure ultimately relay sensory information to the primary
and other factors are optimized to maintain cord per- somatosensory cortex (Brodmann areas 3, 1, and 2).3
fusion. By warning the surgeon of the need to reduce SSEPs are recorded with scalp electrodes ater elec-
the correction, or remove speciic implants, electro- tric stimulation of aferent peripheral nerves. Median
physiologic monitoring can help prevent permanent and ulnar nerves are used in the upper limbs while
and severe neurologic injuries. tibial and peroneal nerves are selected for monitoring
the lower limbs. Figure 8.4 shows the usual location of
Somatosensory evoked potentials SSEP electrodes.
(SSEPs) Changes in amplitude greater than 50% and/
Somatosensory evoked potential were initially or decrease in latency greater than 10% are gener-
described for monitoring the spinal cord during ally accepted as a guideline for notifying the surgeon
deformity correction for scoliosis.2 he reinement of of impending spinal cord injury.3,4 Nuwer et al., ater
technology and technique has made SSEPs the main- reviewing a large multicenter survey of more than
stay in spinal cord monitoring. 50 000 procedures using intraoperative SSEPs, reported
SSEPs consist of monitoring neuronal integrity an overall 92% sensitivity and 98.9% speciicity in the
in the dorsal columnmedial lemniscus pathway. ability of SSEPs to detect new postoperative neurologic
Receptors localized in the skin, tendons, and muscles deicits.5
generate information on tactile discrimination, vibra- he amplitude of SSEPs can potentially be altered by
tion and conscious proprioception. hese sensory the use of halogenated agents, nitrous oxide, hypother-
modalities relay signals to irst-order neurons whose mia, hypotension, and electrical interference. Latency
soma are located at dorsal root ganglia. hey project can be afected by temperature changes.3
to the spinal cord via the medial root entry zone, giv- SSEPs are generally used as a surveillance tool dur-
ing rise to the gracilis and cuneatus fasciculi, carrying ing spinal decompression and instrumentation. hey
sensation from the lower and upper extremities. he are generally considered less sensitive to motor inju-
ibers decussate at the medullary level forming the ries, which may occur during anterior decompression
medial lemniscus. hey ascend to the thalamus and and treatment of fractures and tumors.

Figure 8.4 Arrangement for


Recording somatosensory evoked potential
Somatosensory cortex electrodes.

Upper extremity

Lower extremity

Stimulation
Tibialis nerve

124
Chapter 8.3: Intraoperative neurophysiologic monitoring

Motor evoked potentials (MEPs)


Somatosensory evoked potentials have an excellent
ability to assess the dorsal column and lateral sensory
tract function. However, signiicant motor deicits
have been seen in patients undergoing spinal surgery
despite normal SSEPs.6,7 First introduced by Levy and Right hand
York in 1983,8 motor evoked potentials were developed
to assess the function of the anterior spinal columns
and motor pathways.
MEP monitoring is based on transcranial stimula-
tion (electrical) through scalp needles. he stimulus
over the skull elicits a response from the underlying
motor cortex at the lowest threshold. he initial volt-
age used is typically close to 100 V, and a train of stim-
uli are used to record motor evoked response from the
muscles contralateral to the stimulated side. Figure 8.5 Right hamstring
shows normal responses of diferent muscle groups. If
no response is seen, the voltage may be increased by
50 V until MEP responses are seen.
MEP responses can be interpreted with difer-
ent methods. he all-or-nothing criterion is based on
complete loss of signal compared with the baseline
response.9 he amplitude criterion compares baseline Right gastrocnemius
responses in terms of amplitude changes.10 he thresh-
old criterion analyzes changes in voltage requirements
compared with baseline responses.11 Changes in wave-
form morphology can also be analyzed by tracking
changes in the pattern and duration of MEPs.12
he development of multipulse stimulation has
allowed monitoring with less rigorous anesthetic Right foot
parameters. Total intravenous anesthesia is used for
optimal acquisition of signals. Compounds such as
nitrous oxide, volatile agents, and muscle relaxants Figure 8.5 Motor evoked potential normal signal recording.
are excluded, and short-acting agents such as fentanyl
and propofol are relied upon to achieve anesthetic con-
trol. Although this may pose more of a challenge to Electromyography (EMG)
the anesthesia team and possibly the surgeon, total IV Electromyography (EMG) is the recording of electri-
anesthesia ofers clear beneits in obtaining MEPs over cal muscle activity. In 1948, Du Bois-Reymond dem-
inhaled anesthetics.13 onstrated the irst nerve action potentials and muscle
Anterior spinal surgery, which oten involves electrical activity. Changes in EMG recordings are
manipulation of structures that may be putting pres- indirect indicators of function of the innervating nerve.
sure on the spinal cord to start with, and which also Intraoperative use is helpful in localization and assur-
may involve interruption of small blood vessels which ance of peripheral nerve integrity and function.
supply the spinal cord, is oten supported with MEP Multiple EMG needles are placed into muscles
monitoring to determine the earliest event in which of interest. Practically any muscle can be monitored,
increased blood pressure support or steroids may be including face, tongue, and sphincter musculature.
necessary. he MEP feedback may also alert the sur- EMG is recorded continually with a low-noise ampli-
geon to the need for further decompression, or a change ier. Recordings are shown on a monitor and also sent
in the position of a grat or implant. to a speaker to provide auditory feedback. Changes 125
Section 2: Spine surgery for adult patients

in electrical activity can be seen and heard. During responses picked up in the EMG window are cautery
spine surgery, EMG can be used with two techniques. devices, electrocardiography leads, and high-speed
he irst technique captures spontaneous EMG activ- drills.3,15
ity. Spontaneous muscle activity is monitored with
recording electrodes placed in the muscles of interest Multimodality intraoperative
and based on the structures at risk. Even though no
stimulation is performed, surgical manipulation such
monitoring (MIOM)
Multimodality intraoperative monitoring (MIOM),
as stretching, pulling, or nerve compression produces
combines SSEPs, MEPs, and EMG according to the struc-
neurotonic discharges resulting in activity in the cor-
tures at risk in surgery, taking advantage of the individ-
responding innervated muscle or muscle group. At this
ual strength of each modality for accurate monitoring
point the surgeon is notiied in order to reassess his or
Kelleher et al.,16 ater reviewing a prospective ana-
her technique so as to avoid neural injury. he second
lysis of 1055 patients, showed a very low evidence sup-
technique is called triggered EMG. his technique has
porting unimodal SSEPs or MEPs as a valid diagnostic
been used in segmental instrumentation procedures
test for measuring intraoperative neurologic injury.
requiring pedicle screws for ixation. he main goal is to
On the other hand, there was strong evidence suggest-
reduce the risk of breaching walls of the vertebral pedi-
ing that multimodality of neuromonitoring is sensitive
cle in order to avoid the neural structures. Typically, a
and speciic for detecting intraoperative neurologic
monopolar electrode is used to stimulate the top of the
injury during spine surgery.3,17
pedicle screw at increasing current intensities. Needle
Even though there is no Class I evidence in the lit-
electrodes in the appropriate muscle groups will meas-
erature supporting the use of monitoring in spine sur-
ure the muscle action potential during the stimula-
gery,18 it is recommended that the use of MIOM be
tion. A pedicle breach would signiicantly reduce the
considered in complex spine surgery where the spinal
stimulation threshold. A threshold response between
cord or nerve roots are at risk of injury.3,17,19
10 and 20 mA gives a reasonable probability that no
breach of the medial wall has occurred, thresholds
>15 mA indicate a 98% likelihood of accurate screw
Eects of anesthesia on recording
positioning on postoperative CT scan.14 Figure 8.6 of electrical impulses
shows triggered EMG. Intraoperative neurophysiologic monitoring places
Sometimes, the electrodes will pick up interference certain demands on anesthesia. he main concern is to
from various sources that may be mistaken for spiking tailor the anesthetic agents in order to help maximize
or training EMG activity. Potential sources of artifact signal acquisition during the surgical procedure.

Figure 8.6 Triggered EMG.

Triggered EMG

Right vastus hamstring

Right tibialis anterior

126
Chapter 8.3: Intraoperative neurophysiologic monitoring

here are many factors that may afect evoked typically delivered as continuous infusion, with a load-
potential signal recording. Any physical parameter or ing dose of 0.51 g/kg followed by an infusion of 0.2
drug that afects the axonal conduction may change 0.5 g/kg/h.20 Lidocaine depresses the amplitude and
the evoked potential waveform. In general, long neural prolongs the latency of SSEPs; nevertheless, the wave-
tracts with more synapses are more susceptible to anes- forms are preserved and interpretable when used as
thesia. MEPs are usually more sensitive to interference part of narcotic-based anesthesia.27 Lidocaine infusion
by anesthesia than SSEPs. Evoked potential signals are is used with a loading dose of 1.5 mg/kg (induction) and
more diicult to obtain from lower than upper extrem- followed by lidocaine infusion at a rate of 40 g/kg/h.
ities.15 When the anesthetic conditions are optimized he maximum recommended dose is 4 mg/min.20,27
for MEPs, they are usually also acceptable for SSEPs. Ketamine has the beneit of enhancing signals
Another important factor is the neurologic condition acquired from evoked potentials. his can be achieved
of the patient. When a preexisting neurologic deicit is with clinically relevant doses.28 Ketamine can be used
present, signals are more diicult to obtain.17 with a loading dose of 0.51.0 mg/kg as part of the
General anesthesia should be optimized to obtain induction, followed by an infusion of 0.3 mg/kg/h. It
useful monitoring potentials that can help guide the is administered throughout the surgical procedure and
surgical progress. In general, most anesthetic agents is usually terminated at least 30 minutes prior to the
depress the amplitudes and increase the latency of end of the surgery. Ketamine can be coadministered
evoked potentials. Etomidate and ketamine are the with the lidocaine infusion. he standard concentra-
exception to the rule, since they enhance SSEP and tion of lidocaine is an 8% solution in dextrose-based
MEP amplitude. Either way, etomidate is limited due crystalloid. Ketamine (250 mg) is added to a 250 ml IV
to the association with adrenocortical suppression.18 bag of lidocaine. When the infusion pump is set for a
Neuromuscular blockade has a negative impact on lidocaine dose of 40 g/kg/min, the resultant ketamine
the ability of monitoring to attain acceptable record- dose that will be administered is 0.3 mg/kg/h.2
ings.20,21 Neuromuscular blocking agents such as Changes in temperature may alter intraoperative
succinylcholine or nondepolarizing agents such as signal recording. Hypothermia increases the latency
rocuronium may be used in order to facilitate endo- and decreases conduction velocities. Hyperthermia
tracheal intubation only.20 reduces the latency and increases the conduction vel-
All halogenated inhalational agents produce a dose- ocity.20 Hypoxemia can cause evoked potential deteri-
related increase in latency and reduction in the ampli- oration before other clinical parameters have changed.
tude of cortically recorded SSEPs.15,19 Concentrations SSEP response changes with hematocrit are consistent
of 0.5 MAC of volatile anesthetics should allow with this optimum range. A decrease in amplitude has
acceptable acquisition of MEPs.22 Some recommend been noted with mild anemia, followed by an increase
to avoid volatile anesthetic agents and instead rely on in latency at hematocrits of 1015%; further latency
propofol-based anesthesia.20 changes and amplitude reductions were observed at
Nitrous oxide reduces SSEP cortical amplitude and hematocrits less than 10%. hese changes are partially
increases latency when used alone or when combined with restored by an increase in the hematocrit.29
halogenated inhalational agents, opioids, or propofol.23 An alarming change in SSEP or EMG responses
Total intravenous anesthesia (TIVA) gives the opti- can be seen when blood low is reduced to one of the
mal conditions for intraoperative neurophysiologic extremities. his can be expected when a tourniquet is
monitoring.20 Recommendations for this anesthetic used over the lower extremity, as in the case of harvest
approach use propofol, a synthetic narcotic such as suf- of a ibular grat. It may also be seen commonly dur-
entanil, an infusion of the local aesthetic lidocaine, and ing anterior approach surgery for disc replacement or
frequently the NMDA receptor antagonist ketamine. interbody fusion of the lumbar spine, when pressure
Propofol is used for induction and for maintenance is placed on either the aorta or the common iliac ves-
of general anesthesia (75150 g/kg/min). Propofol sels. Responses should return to normal quickly ater
does not afect latency but produces a dose-dependent the pressure is relieved, and ater the tourniquet is
reduction in the amplitude of MEPs.24 Opioids have a removed. If this does not happen, an alarm should be
limited impact on MEPs but there are reports suggest- sounded. his may be the irst sign of a vascular throm-
ing a suppressive efect of alfentanil, fentanyl, remifen- bosis or plaque embolism that can compromise or
tanil, and sufentanil.20,25,26 Sufentanil or remifentanil are result in loss of the lower extremity. 127
Section 2: Spine surgery for adult patients

Table 8.1 Summary of diferent neuromonitoring modalities commonly used in spine surgery

SSEPs MEPs EMG


Monitor Dorsal column medial and lemniscus Function of the motor pathway. Peripheral nerve integrity and
pathway Anterolateral column function
Abnormal Decrease in amplitude (>50%) and or Complete loss of signal Changes in wave Manipulation of peripheral nerve
recording decrease in latency (>10%) amplitude, form, threshold to stimuli response produces neurotonic discharges
Sensitivity/ 92%/98.9% 100%/96% 46100%/23100%
speciicity

Since it was demonstrated that temporary occlu- can be reversed. If a bone grat or cages have been
sion of segmental vessels in the thoracic spine could be placed near the spinal cord these should be removed
relected in SSEP changes in many patients, surgeons and repositioned to ensure that there is no subtle pres-
are more cautious about ligation of blood vessels in sure on the neural elements.
the thoracic region. horacic spine surgeons are much If no evident cause for the changes can be identiied,
more likely to spare segmental vessels during anterior a Stagnara wake-up test may be considered to verify the
approaches these days, but may occasionally place a electrophysiologic observation. In the presence of a doc-
temporary clamp on a large segmental artery to deter- umented deicit, the patient will be given steroids, the
mine whether vascular changes are relected in SSEP or implants will be locked in place without excessive cor-
MEP monitoring. rection, and the wound will be closed carefully over a
drain. he patient will be transferred to an intensive care
Impact on surgical outcomes and environment where oxygenation and blood pressure
Because intraoperative spinal cord injury is an uncom- can be maintained at an optimal level. Serial neurologic
mon event, and diicult to attribute to any single cause, examinations can be carried out, and reimaging of the
very large numbers of patients are required to show spine performed to identify any reversible lesion.
signiicant impact of any intervention. Because of this
there are few Level I data available to conclusively sup- Summary
port the use of intraoperative monitoring in spinal Surgeons will request electrophysiologic monitor-
surgery. Nonetheless, the devastating impact of spinal ing based on their anticipated needs during surgery.
cord injury on functional outcome and survival drives Combined multimodality monitoring using SSEP,
surgeons to seek any opportunity to reduce risk. MEP, and EMG (Table 8.1) modalities tends to over-
Subtle changes in SSEP averaging, or in MEP sig- come the shortcomings of each individual modality
naling, can guide the surgeon to alter the surgical and provide the most comprehensive and sensitive test
approach, and can alert the anesthesiologist to change during surgery. As modalities become more reliable,
the anesthetic environment. Particularly in scoliosis surgeons will regularly depend on electrophysiologic
and deformity surgery, where the neural elements are monitoring when the spinal cord is under pressure or
manipulated but not seen, electrophysiologic monitor- at risk.
ing can alert the surgical team to make changes to save
the neural elements from serious irreparable injury. References
If changes are seen, the surgeon can make an imme- 1. Dawson GD. Cerebral responses to nerve stimulation in
diate reassessment of the implants placed and the man. Br Med Bull 1950; 6(4): 3269.
decompression performed, and look for any evidence 2. Nash CL Jr, Lorig RA, Schatzinger LA, Brown RH.
of pressure caused by surgical instruments, pledgets, Spinal cord monitoring during operative
or bone or tissue fragments. he anesthetist will imme- treatment of the spine. Clin Orthop Relat Res 1977;
diately reassess blood pressure, raising it to an optimal 1977; 126: 1005.
level, and assess the body temperature and oxygena- 3. Gonzalez A, Jeyanandarajan D, Hansen C, Zada
tion. If inhaled anesthetics are being used, they should G, Hsieh PC. Intraoperative neurophysiological
be discontinued and IV barbiturates and opioids used monitoring during spine surgery: a review. Neurosurg
instead. he surgical team may choose to release the Focus 2009; 27(4): E6.
128 instrumentation and ease of correction, and may select 4. Aglio LS, Romero R, Desai S, Ramirez M, Gonzalez AA,
speciic implants to remove to see whether the changes Gugino LD. he use of transcranial magnetic stimulation
Chapter 8.3: Intraoperative neurophysiologic monitoring

for monitoring descending spinal cord motor function. 17. Wilson-Holden TJ, Padberg AM, et al. Eicacy of
Clin Electroencephalogr 2002; 33(1): 3041. intraoperative monitoring for pediatric patients with
5. Nuwer MR, Dawson EG, Carlson LG, Kanim LE, spinal cord pathology undergoing spinal deformity
Sherman JE. Somatosensory evoked potential spinal surgery. Spine 1999; 24(16): 168592.
cord monitoring reduces neurologic deicits ater 18. Wagner RL, White PF, et al. Inhibition of adrenal
scoliosis surgery: results of a large multicenter survey. steroidogenesis by the anesthetic etomidate. N Engl J
Electroencephalogr Clin Neurophysiol 1995; 96: 611. Med 1984; 310(22): 141521.
6. Hilibrand AS, Schwartz DM, Sethuraman V, Vaccaro 19. Zentner J, Albrecht T, et al. Inluence of halothane,
AR, Albert TJ. Comparison of transcranial electric enlurane, and isolurane on motor evoked potentials.
motor and Somatosensory evoked potential monitoring Neurosurgery 1992; 31(2): 298305.
during cervical spine surgery. J Bone Joint Surg Am 20. Pajewski TN, Arlet V, Phillips LH. Current approach
2004; 86-A: 124853. on spinal cord monitoring: the point of view of the
7. Hsu B, Cree AK, Lagopoulos J, Cummine JL. neurologist, the anesthesiologist and the spine surgeon.
Transcranial motor-evoked potentials combined with Eur Spine J 2007; 16 (Suppl 2): S11529.
response recording through compound muscle action 21. Van Dongen EP, ter Beek HT, et al. Within-patient
potential as the sole modality of spinal cord monitoring variability of myogenic motor-evoked potentials to
in spinal deformity surgery. Spine 2008; 33: 11006. multipulse transcranial electrical stimulation during
8. Levy WJ Jr., York DH. Evoked potentials from the two levels of partial neuromuscular blockade in aortic
motor tracts in humans. Neurosurgery 1983; 12: 422. surgery. Anesth Analg 1999; 88(1): 227.
9. Kothbauer KF, Deletis V, Epstein FJ. Motor-evoked 22. Sekimoto K, Nishikawa K, et al. he efects of volatile
potential monitoring for intramedullary spinal anesthetics on intraoperative monitoring of myogenic
cord tumor surgery: correlation of clinical and motor evoked potentials to transcranial electrical
neurophysiological data in a series of 100 consecutive stimulation and on partial neuromuscular blockade
procedures. Neurosurg Focus 1998; 4(5): e1. during propofol/fentanyl/nitrous oxide anesthesia in
humans. J Neurosurg Anesthesiol 2006; 18(2): 10611.
10. Calancie B, Molano MR. Alarm criteria for motor-
evoked potentials: whats wrong with the presence-or- 23. Van Dongen EP, ter Beek HT, Schepens MA, et al..
absence approach? Spine 2008; 33: 40614. Efect of nitrous oxide on myogenic motor potentials
evoked by a six pulse train of transcranial electrical
11. Langeloo DD, Lelivelt A, Louis Journee H, Slappendel stimuli: a possible monitor for aortic surgery. Br J
R, de Kleuver M. Transcranial electrical motor-evoked Anaesth 1999; 82(3): 3238.
potential monitoring during surgery for spinal deformity:
a study of 145 patients. Spine 2003; 28: 104350. 24. Nathan N, Tabaraud F, et al. Inluence of propofol
concentrations on multipulse transcranial motor
12. Quinones-Hinojosa A, Lyon R, Zada G, et al. Changes evoked potentials. Br J Anaesth 2003; 91(4): 4937.
in transcranial motor evoked potentials during
intramedullary spinal cord tumor resection correlate 25. Kalkman CJ, Drummond JC, et al. Efects of propofol,
with postoperative motor function. Neurosurgery 2005; etomidate, midazolam, and fentanyl on motor evoked
56: 98293. responses to transcranial electrical or magnetic
stimulation in humans. Anesthesiology 1992; 76(4): 5029.
13. Pechstein U, Nadstawek J, Zentner J, Schramm
26. Scheuler KM, Zentner J. Total intravenous anesthesia
J. Isolurane plus nitrous oxide versus propofol
for intraoperative monitoring of the motor pathways:
for recording of motor evoked potentials ater
an integral view combining clinical and experimental
high frequency repetitive electrical stimulation.
data. J Neurosurg 2002; 96(3): 5719.
Electroencephalogr Clin Neurophysiol 1998; 108: 17581.
27. Schubert A, Licina MG, et al. Systemic lidocaine and
14. Shi YB, Binette M, Martin WH, Pearson JM, Hart RA. human somatosensory-evoked potentials during
Electrical stimulation for intraoperative evaluation sufentanil-isolurane anaesthesia. Can J Anaesth 1992;
of thoracic pedicle screw placement. Spine 2003; 28: 39(6): 56975.
595601.
28. Erb TO, Ryhult SE, et al. Improvement of motor-evoked
15. Deiner S. Highlights of anesthetic considerations for potentials by ketamine and spatial facilitation during
intraoperative neuromonitoring. Semin Cardiothorac spinal surgery in a young child. Anesth Analg 2005;
Vasc Anesth 2010; 14(1): 513. 100(6): 16346.
16. Kelleher MO, Tan G, Sarjeant R, Fehlings M. Predictive 29. Nagao S, Roccaforte P, et al. he efects of isovolemic
value of intraoperative neurophysiological monitoring hemodilution and reinfusion of packed erythrocytes on
during cervical spine surgery a prospective analysis of somatosensory and visual evoked potentials. J Surg Res
1055 consecutive patients. J Neurosurg Spine 2008; 8: 1978; 25(6): 5307. 129
215-221.
Section 2: Spine surgery for adult patients

8.4 An overview of minimally invasive spine surgery


R. Douglas Orr

Key points endoscope is introduced into the epidural space and


also potentially into the thecal sac to allow visualiza-
he minimally invasive approach allows the tion of pathology. Since its development it has also
surgeon to perform the same operation with been used to treat pathologies. It has been used to
less collateral damage. break down adhesions, potentially resect discs, and
Less invasive does not mean less risk. decompress spinal stenosis. his procedure is gener-
Initially on adoption of these techniques OR ally done under local anesthetic and the portal used
times will be longer but they should eventually is only a few millimeters wide. he scope is inserted
become the same. through the sacral hiatus. It is advanced into the lum-
here should be less blood loss and less bar spine under radiographic guidance. Once there
postoperative pain than in open techniques. it can either through direct mechanical pressure
Each approach has subtle diferences and or the use of lasers break down adhesions or resect
intraoperative requirements and it is compressive tissue, and corticosteroids can be dir-
important to discuss these in advance with ectly deposited into the epidural space.1216 It can also
the surgeon. be performed through an interlaminar approach.17
here is very little blood loss and is generally done as
Minimally invasive spinal surgery is a catch-all term an outpatient procedure. It has relatively limited uses
that has been applied to a number of techniques used and is not widely performed.
to treat spinal pathology. Techniques have been devel- Kambin described an extraforaminal approach to
oped to allow decompression of the nerves and spinal the lumbar disc as a way of dealing with symptomatic
cord, resection of intradural and extradural spinal lumbar disc herniations. Initially he described a two-
tumors, correction of deformity, and stabilization of portal technique with an arthroscope inserted from
instability. he overwhelming philosophy of mini- one side and a working channel to allow introduc-
mally invasive spinal surgery is to minimize the collat- tion of instruments inserted into the disc space from
eral damage that occurs in accessing spinal pathology the opposite side.18 his technique was later evolved
through traditional open approaches. It is felt that by through the development of working channel spinal
doing so better outcomes can be achieved. Many tech- endoscopes.19 he scopes range in size from 5 to 8 mm
niques are involved, each with their own indications, in diameter. hey are usually inserted under local anes-
complications, and outcomes. his chapter will review thetic ater tissue dilation. hrough the working chan-
many of these techniques and the rationale for their nel one can use a variety of graspers, drills, and lasers
use. It will also include anesthetic considerations dur- to remove pathology. his can be used to decompress
ing these surgeries. foraminal stenosis or to resect herniated discs, and
Traditional open approaches to the spine have has been described for the treatment of central sten-
been shown to cause signiicant muscle damage. osis. his technology has been used in the cervical,
Radiographic, histochemical, and clinical studies have lumbar, and thoracic spine.2026 Recent comparative
shown this.18 Minimally invasive approaches have studies have looked at percutaneous discectomy and
been developed to lessen this injury. In general they compared it with open microdiscectomy in the lumbar
use smaller incisions, respecting anatomic planes and spine. In general these studies show that patients have a
separating or dilating muscles rather than cutting or shorter hospital stay and less immediate postoperative
detaching them. It is felt that by doing so recovery will pain but equivalent long-term results.2731 here are no
be quicker and outcomes will be better.9,10 comparative studies of thoracic or cervical uses of this
One of the earliest minimally invasive techniques technique. he endoscopic transforaminal technique
was used predominantly for diagnosis, and this is has not become widely accepted in North America and
130 epiduroscopy.11 In this procedure a small iberoptic is done mostly in specialized outpatient centers.
Chapter 8.4: Minimally invasive spine surgery

he next minimally invasive technique to be


developed and achieve a degree of widespread accept-
ance was endoscopic anterior lumbar interbody
fusion.32 his is performed through a transperitoneal
approach with laparoscopic visualization. Numerous
studies were published with initially good results.3339
As time went on further studies questioned the safety
and eicacy of the technique.4044 It has largely been
abandoned.
he modern era of more widely accepted minim-
ally invasive spine surgery techniques began with the
development of the tubular retractor by Foley. his
technique was originally developed to treat far lateral A
and central lumbar disc herniations.45,46 In this tech-
nique under radiographic guidance a series of dilators
are used to split the muscles to allow placement of the
tubular retractor. Initially the procedure used a small
endoscope but now most surgeons use a microscope.
Further development of the use of tubular retractors
has allowed them to be used to decompress the spine
for lumbar stenosis47 and for fusion.48
Figure 8.7A shows the placement of the tubular
retractor over a series of dilators. Figure 8.7B shows
a 16 mm diameter tubular retractor being used for a
microdiscectomy.
Numerous studies have looked at the use of tubu-
lar retractors for discectomies and decompressions. In
B
general they show that in the short term patients have a
faster recovery with less postoperative pain, less blood Figure 8.7 (A) Intraoperative photograph showing placement of
loss, and less use of narcotics.49,50 In the longer term a 16 mm tubular retractor over a series of dilators. (B) Intraoperative
photograph with a Penield probe in the 16 mm retractor during a
it does not appear as though results are any diferent microdiscectomy.
from open techniques. hese procedures do show a
signiicant learning curve and initially operative times
are longer than for the equivalent open procedure, but began by inserting a Jamshidi needle into the pedicle
with experience they seem to take equivalent or less and then placing a guide wire over which the pedicle
time than the open procedure.5052 is prepared and the screw inserted. Figure 8.8 shows
In order to progress from decompressive proce- intraoperative photographs of percutaneous screw and
dures to fusions it was necessary to develop the ability rod placement. Technologies have been developed to
to place pedicle screw instrumentation through small allow placement of screws over multiple levels.
incisions. his led to the development of a number Percutaneous screws can be used as temporary
of techniques for percutaneous pedicle screws. hese ixation in trauma patients55 and have occasionally
screws are typically used in conjunction with some been used to stabilize tumor patients in the absence of
form of fusion, either an anterior lumbar interbody fusion. Otherwise all percutaneous instrumentation
fusion (ALIF), a minimal access lateral fusion (LLIF), systems need to be combined with some sort of fusion.
or a minimal access transforaminal lumbar inter- In most cases these involve placement of implants and
body fusion (TLIF). he irst description of this used bone grat or bone grat substitute into the intervert-
a suprafascial implant and was not really practical.53 ebral disc space. All of the current minimally inva-
he irst clinical use of current types of implants was sive interbody fusion techniques developed from
reported by Foley.54 hese screws are inserted with more standard open techniques and through the use
either luoroscopy or image guidance. he procedure of specialized retractors have evolved into minimally 131
Section 2: Spine surgery for adult patients

A D

Figure 8.8 (cont.)

invasive approaches. Anterior interbody fusion (ALIF)


is performed through an anterior retroperitoneal mini
open approach. Typically this is done through a trans-
verse incision overlying the rectus muscle. Ater the
rectus sheath is divided, the remainder of the approach
is done with blunt dissection and as a result is relatively
atraumatic. his approach has signiicant intraopera-
tive risks, particularly at the L45 level due to the need
to mobilize the common iliac vein in order to access the
disc space. Vascular injuries are uncommon but can be
a signiicant problem.56,57
B Posterior lumbar interbody fusion (PLIF) was ini-
tially described by Cloward58 as an open technique. It
has been adapted for use with tubular retractors.59 In
this procedure the disc is removed from a posterior
approach working in the spinal canal. Ater removing
the disc a bone grat or implant is placed. his proced-
ure carries a risk of intraoperative nerve injury due to
the necessity of retracting the thecal sac to access the
disc. As a result it is oten performed with intraopera-
tive neurologic monitoring.
he transforaminal lumbar interbody fusion (TLIF)
was initially described as an open technique60 and was
also converted to a minimally invasive approach.61
he entry to the disc space is more lateral through the
C region of the foramen and as a result has a lower risk
of neurologic injury. his is the fusion procedure that
Figure 8.8 (A) Intraoperative photograph during placement of has been most widely adopted as a minimally inva-
a percutaneous screw. On the upper right is a guide wire being
placed through a Jamshidi needle. On the lower left are the dilators sive approach. Figure 8.9 shows intraoperative photo-
being placed over the wire. (B) Lateral Fluoro image showing graphs of a TLIF being performed through a 22 mm
previously placed interbody cages. A tap is being advanced over tubular retractor.
the uppermost wire and one screw has been placed. (C) After
placement of 5 of 6 screws, showing the extension towers used to Comparative studies looking at open versus min-
guide the rod into the screw heads. (D) Advancing of the rod from imally invasive TLIF fusions show that TLIF patients
132 superior though the extension towers. have a shorter postoperative stay, less blood loss, less
Chapter 8.4: Minimally invasive spine surgery

A A

B B

Figure 8.9 (A) Intraoperative photograph showing a Kerrison Figure 8.10 (A) Intraoperative photograph showing MIS LLIF
punch in a 22 mm tubular retractor during an MIS TLIF. On the left retractor. Cables protruding from the retractor are iberoptic light
are the extension towers and rod-passing tools of a percutaneous cords to assist visualization. Unlike the tubes in the previous igures
instrumentation system. (B) Lateral Fluoro image showing the retractor is expandable. (B) AP Fluoro image showing placement
interbody cage in the disc space and rod passage during MIS TLIF. of an implant trial in the disc space. Screws are temporary retaining
pins to hold the retractor in position.

postoperative narcotic use, and a faster recovery. At


1 year and 2 years there is no diference in outcome nerve injury. Ater placement of a probe into the disc
compared with open procedures.6266 Early in the learn- space, serial dilators are used to enlarge the approach
ing experience operative times tend to be longer than until a retractor can be placed.69 Figure 8.10 shows the
seen in open techniques, but with experience the oper- retractor in place and a trial implant in position. his
ation can be done in equal or shorter time.65,67,68 technique can be used over multiple levels and has now
Minimally invasive lateral fusion (LLIF) is done been used in the minimally invasive treatment of scoli-
through a retroperitoneal and trans-psoas approach. osis.70 his procedure does not require mobilization of
In this procedure the patient is placed in a lateral pos- the great vessels and so has a low risk of vascular com-
ition. Under a combination of direct palpation and plication.71 here is a relatively high risk of injury to the
radiographic guidance a probe is placed down onto the lumbosacral plexus and this necessitates EMG moni-
spine through the psoas muscle. Because of the pres- toring.72 In general, high fusion rates can be obtained
ence of the lumbar nervous plexus this approach is done with good clinical success. hese procedures are oten 133
with stimulated EMG monitoring to lessen the risk of backed up with posterior instrumentation.71 hey can
Section 2: Spine surgery for adult patients

be done stand-alone and are sometimes augmented complications of the procedure just as one would for a
with an anterior plate.73 traditional open operation. In the case of any anterior
he trans-sacral lumbar interbody fusion is a tech- surgery there is the risk of catastrophic bleeding if the
nique for fusion of L5S1, and now L45 and L5S1. great vessels are injured. his can occur with all of the
It utilizes the presacral interval as an approach and is interbody techniques and, in the case of vascular injury
done through a paracoccygeal incision.74 It has been during a posterior interbody technique, may not be
reported to have good clinical success and low compli- readily apparent. In patients with sudden loss of blood
cation rates,75 though rectal injury has been reported.76 pressure this should be considered.
If a surgeon is just beginning to use minimally inva-
Minimally invasive treatment of sive techniques, operative times will typically be longer
than with the equivalent open procedure. With time
idiopathic scoliosis and kyphosis and experience this should decrease to similar opera-
horacoscopy has been used extensively for anterior tive times and sometimes even shorter as there is less
release and instrumentation in idiopathic scoliosis and time needed to open and close the incision. For the
kyphosis and is used as an alternative to open thoracot- anesthesiologist it is important to know where the sur-
omy. It is associated with shorter operative times, faster geon is on the learning curve and to plan accordingly.
recovery, and better cosmesis.77 horascopic instru- Long procedure duration early in the experience can
mentation for correction of scoliosis has been done, but increase problems with core body temperature and
many have abandoned it to due to higher complication pressure over bony prominences.
rates particularly late hardware failure.7881 his can be In general minimally invasive surgery (MIS) tech-
done with the patient either in the lateral position or as niques have lower intraoperative blood loss. As a result
a simultaneous prone procedure accompanied by pos- there should be lower demand for luids intraopera-
terior instrumentation.82 In either case double-lumen tively. At the authors institution one of our early MIS
intubation is used and the lung collapses on the upper patients had a postoperative myocardial infarct attrib-
side. Scoliosis can be treated either from the convexity uted to volume overload ater receiving a 2 l luid bolus
or the concavity. It is important for the anesthesiologist before a kyphoplasty that had 15 ml blood losses.
to be aware of which side the approach will be done from Most studies have shown less immediate post-
when it comes time to delate the lung. operative pain and lower postoperative narcotic
demands. In the authors practice most open spinal
Conditions treated with minimally fusion patients require PCA IV narcotics for the irst
2448 h where most MIS patients can be managed with
invasive spine surgery PO pain medication.
Initial reports of minimally invasive surgery in the Depending on the technique used, neuromuscu-
spine looked at common degenerative conditions such lar blockade may be required or contraindicated. In
as herniated disc, lumbar spinal stenosis, degenerative the ALIF approach it is vital to have blockade to allow
disc disease, and spondylolisthesis. As techniques have retraction. Loss of blockade during the procedure
evolved many more conditions are able to be treated can lead to loss of retraction and risk of vessel injury.
through minimally invasive techniques. Trauma,83 In contrast, the need for stimulated EMG monitor-
deformity,8486 and tumors8789 have all been reported ing during LLIF surgery means that neuromuscular
to have been treated to minimally invasively. As these blockade is contraindicated. In some case both pro-
techniques evolve it is likely that there will be a minim- cedures are done and as a result blockade is required
ally invasive option for most spinal surgeries. for part of the procedure and contraindicated later. In
these cases short-acting blockade agents must be used.
Anesthetic considerations in minimally It is important to conirm with the surgeon before the
start of the procedure what the requirement for block-
invasive spine surgery ade is.
In general minimally invasive techniques allow spine
surgeons to do the same procedures with less collat-
eral damage. It is important to remember that smaller Conclusion
134 incision does not mean smaller risk to the patient. As Minimally invasive techniques are rapidly evolving
a result it is important to be prepared for the potential and are changing the way that spine surgery is done. In
Chapter 8.4: Minimally invasive spine surgery

the future it will be a larger part of most practices. It has 10. German JW, Foley KT. Minimal access surgical
been shown to lead to less postoperative pain, lower techniques in the management of the painful lumbar
blood loss, shorter hospital stays, and shorter recov- motion segment. Spine 2005; 30(16 Suppl): S529.
ery. In many cases patients are preferentially seeking 11. Blomberg R. A method for epiduroscopy and
out minimally invasive options as they perceive these spinaloscopy. Presentation of preliminary results. Acta
techniques are superior. Long-term studies have yet to Anaesthesiol Scand 1985; 29(1): 11316.
show beneits in the long term over traditional open 12. Ruetten S, Meyer O, Godolias G. Endoscopic surgery
techniques, but if these techniques have equivalent of the lumbar epidural space (epiduroscopy): results of
therapeutic intervention in 93 patients. Minim Invasive
success and equivalent complication rates then the
Neurosurg 2003; 46(1): 14.
improved short-term outcome would justify the more
widespread adoption of these techniques. hose that do 13. Kitahata LM. Recent advances in epiduroscopy. J Anesth
2002; 16(3): 2228.
not have equivalent outcomes should fall out of favor
over time, as has been seen with laparoscopic ALIF and 14. Ruetten S, Meyer O, Godolias G. Application of
thorascopic instrumentation for scoliosis. holmium: YAG laser in epiduroscopy: extended
practicabilities in the treatment of chronic back pain
hese techniques still carry the same risks as syndrome. J Clin Laser Med Surg 2002; 20(4): 2036.
the open procedures and in some cases the smaller
15. Gillespie G, MacKenzie P. Epiduroscopy a review.
incision makes complications harder to deal with.
Scott Med J 2004; 49(3): 7981.
Anesthesiologists must remember this and be prepared
for the possibility. 16. Igarashi T, et al. Lysis of adhesions and epidural
injection of steroid/local anaesthetic during
epiduroscopy potentially alleviate low back and leg
References pain in elderly patients with lumbar spinal stenosis. Br J
1. Gejo R, et al. Serial changes in trunk muscle Anaesth 2004; 93(2): 1817.
performance ater posterior lumbar surgery. Spine 1999; 17. Avellanal M, Diaz-Reganon G. Interlaminar approach
24(10): 10238. for epiduroscopy in patients with failed back surgery
2. Kawaguchi Y, Matsui H, Tsuji H. Changes in serum syndrome. Br J Anaesth 2008; 101(2): 2449.
creatine phosphokinase MM isoenzyme ater lumbar 18. Kambin P. Diagnostic and therapeutic spinal
spine surgery. Spine 1997; 22(9): 101823. arthroscopy. Neurosurg Clin N Am 1996; 7(1): 6576.
3. Kawaguchi Y, et al. Back muscle injury ater posterior 19. Kambin P, Savitz MH. Arthroscopic microdiscectomy:
lumbar spine surgery. Topographic evaluation of an alternative to open disc surgery. Mt Sinai J Med 2000;
intramuscular pressure and blood low in the porcine 67(4): 2837.
back muscle during surgery. Spine 1996; 21(22):
20. Mathews HH. Transforaminal endoscopic
26838.
microdiscectomy. Neurosurg Clin N Am 1996; 7(1): 5963.
4. Kawaguchi Y, Matsui H, Tsuji H. Back muscle injury 21. Ditsworth DA. Endoscopic transforaminal lumbar
ater posterior lumbar spine surgery. A histologic and discectomy and reconiguration: a postero-lateral
enzymatic analysis. Spine 1996; 21(8): 9414. approach into the spinal canal. Surg Neurol 1998; 49(6):
5. Kawaguchi Y, Matsui H, Tsuji H. Back muscle injury 58897; discussion 5978.
ater posterior lumbar spine surgery. Part 2: Histologic 22. Jho HD. Endoscopic transpedicular thoracic
and histochemical analyses in humans. Spine 1994; discectomy. Neurosurg Focus 2000; 9(4): e4.
19(22): 2598602.
23. Yeung AT, Yeung CA. Advances in endoscopic disc and
6. Styf JR, Willen J. he efects of external compression spine surgery: foraminal approach. Surg Technol Int
by three diferent retractors on pressure in the erector 2003; 11: 25563.
spine muscles during and ater posterior lumbar spine
surgery in humans. Spine 1998; 23(3): 3548. 24. Choi G, et al. Percutaneous endoscopic interlaminar
discectomy for intracanalicular disc herniations at
7. Weber BR, et al. Posterior surgical approach to the L5-S1 using a rigid working channel endoscope.
lumbar spine and its efect on the multiidus muscle. Neurosurgery 2006; 58(1 Suppl): ONS5968; discussion
Spine 1997; 22(15): 176572. ONS5968.
8. Motosuneya T, et al. Severe scoliosis associated with 25. Ruetten S, et al. Use of newly developed instruments
Costello syndrome: a case report. J Orthop Surg 2006; and endoscopes: full-endoscopic resection of lumbar
14(3): 3469. disc herniations via the interlaminar and lateral
9. Fessler R, et al., he development of minimally invasive transforaminal approach. J Neurosurg Spine 2007; 6(6): 135
spine surgery. Neurosurg Clin N Am 2006; 17(4): 4019. 52130.
Section 2: Spine surgery for adult patients

26. Nellensteijn J, et al. Transforaminal endoscopic surgery 41. Zdeblick TA, David SM. A prospective comparison
for symptomatic lumbar disc herniations: a systematic of surgical approach for anterior L4-L5 fusion:
review of the literature. Eur Spine J 2010; 19(2): laparoscopic versus mini anterior lumbar interbody
181204. fusion. Spine 2000; 25(20): 26827.
27. Peng CW, Yeo W, Tan SB. Percutaneous endoscopic 42. Liu JC, et al. Is laparoscopic anterior lumbar interbody
discectomy: clinical results and how it afects the quality fusion a useful minimally invasive procedure?
of life. J Spinal Disord Tech 2010; 23(6): 42530. Neurosurgery 2002; 51(5 Suppl): S1558.
28. Peng ZW, et al. A case-control study comparing 43. Kaiser MG, et al. Comparison of the mini-open versus
percutaneous radiofrequency ablation alone laparoscopic approach for anterior lumbar interbody
or combined with transcatheter arterial fusion: a retrospective review. Neurosurgery 2002; 51(1):
chemoembolization for hepatocellular carcinoma. Eur J 97103; discussion 1035.
Surg Oncol 2010; 36(3): 25763. 44. Button G, et al. hree- to six-year follow-up of stand-
29. Peng CW, YeoW, Tan SB. Percutaneous endoscopic alone BAK cages implanted by a single surgeon. Spine J
lumbar discectomy: clinical and quality of life outcomes 2005; 5(2): 15560.
with a minimum 2 year follow-up. J Orthop Surg Res 45. Perez-Cruet MJ, et al. Microendoscopic lumbar
2009; 4: 20. discectomy: technical note. Neurosurgery 2002; 51(5
30. Lee DY, et al. Comparison of percutaneous endoscopic Suppl): S12936.
lumbar discectomy and open lumbar microdiscectomy 46. Foley KT, Smith MM, Rampersaud YR.
for recurrent disc herniation. J Korean Neurosurg Soc Microendoscopic approach to far-lateral lumbar disc
2009; 46(6): 51521. herniation. Neurosurg Focus 1999; 7(5): e5.
31. Kim MJ, et al. Targeted percutaneous transforaminal 47. Khoo LT, Fessler RG. Microendoscopic decompressive
endoscopic diskectomy in 295 patients: comparison laminotomy for the treatment of lumbar stenosis.
with results of microscopic diskectomy. Surg Neurol Neurosurgery 2002; 51(5 Suppl): S14654.
2007; 68(6): 62331.
48. Foley KT, Holly LT, Schwender JD. Minimally invasive
32. Zucherman JF, et al., Instrumented laparoscopic spinal lumbar fusion. Spine 2003; 28(15 Suppl): S2635.
fusion. Preliminary results. Spine 1995; 20(18): 2029
34; discussion 20345. 49. Rahman M, et al. Comparison of techniques for
decompressive lumbar laminectomy: the minimally
33. Olsen D, McCord D, Law M. Laparoscopic discectomy invasive versus the classic open approach. Minim
with anterior interbody fusion of L5S1. Surg Endosc Invasive Neurosurg 2008; 51(2): 1005.
1996; 10(12): 115863.
50. Yagi M, et al. Postoperative outcome ater modiied
34. Regan JJ, et al. Laparoscopic fusion of the lumbar unilateral-approach microendoscopic midline
spine in a multicenter series of the irst 34 consecutive decompression for degenerative spinal stenosis.
patients. Surg Laparosc Endosc 1996; 6(6): 45968. J Neurosurg Spine 2009; 10(4): 2939.
35. Zdeblick TA. Laparoscopic spinal fusion. Orthop Clin 51. Fourney DR, et al. Does minimal access tubular assisted
North Am 1998; 29(4): 63545. spine surgery increase or decrease complications
36. Burkus JK, et al. Six-year outcomes of anterior lumbar in spinal decompression or fusion? Spine 2010;
interbody arthrodesis with use of interbody fusion 35(9 Suppl): S5765.
cages and recombinant human bone morphogenetic 52. Parikh K, et al. Operative results and learning curve:
protein-2. J Bone Joint Surg Am 2009; 91(5): 11819. microscope-assisted tubular microsurgery for 1- and
37. DeBerard MS, et al., Outcomes of posterolateral versus 2-level discectomies and laminectomies. Neurosurg
BAK titanium cage interbody lumbar fusion in injured Focus 2008; 25(2): E14.
workers: a retrospective cohort study. J South Orthop 53. Lowery GL, Kulkarni SS. Posterior percutaneous spine
Assoc 2002; 11(3): 15766. instrumentation. Eur Spine J 2000; 9 (Suppl 1): S12630.
38. Regan JJ, et al. Laparoscopic approach to L4-L5 for 54. Foley KT, Gupta SK. Percutaneous pedicle screw
interbody fusion using BAK cages: experience in the ixation of the lumbar spine: preliminary clinical
irst 58 cases. Spine 1999; 24(20): 21714. results. J Neurosurg 2002; 97(1 Suppl): 712.
39. Kleeman TJ, et al. Laparoscopic anterior lumbar 55. Beringer W, et al., Percutaneous pedicle screw
interbody fusion at L4-L5: an anatomic evaluation and instrumentation for temporary internal bracing of
approach classiication. Spine 2002; 27(13): 13905. nondisplaced bony Chance fractures. J Spinal Disord
40. Pellise F, et al. Low fusion rate ater L5-S1 laparoscopic Tech 2007; 20(3): 2427.
136 anterior lumbar interbody fusion using twin stand- 56. Brau SA, et al. Vascular injury during anterior lumbar
alone carbon iber cages. Spine 2002; 27(15): 16659. surgery. Spine J 2004; 4(4): 40912.
Chapter 8.4: Minimally invasive spine surgery

57. Brau SA. Mini-open approach to the spine for anterior 72. Tohmeh AG, Rodgers WB, Peterson MD. Dynamically
lumbar interbody fusion: description of the procedure, evoked, discrete-threshold electromyography in the
results and complications. Spine J 2002; 2(3): 21623. extreme lateral interbody fusion approach. J Neurosurg
58. Cloward RB. he treatment of ruptured lumbar Spine, 2011; 14(1): 317.
intervertebral disc by vertebral body fusion. III. Method 73. Sharma AK, et al. Lateral lumbar interbody fusion:
of use of banked bone. Ann Surg 1952; 136(6): 98792. clinical and radiographic outcomes at 1 year: a
59. Khoo LT, et al. Minimally invasive percutaneous preliminary report. J Spinal Disord Tech 2011 Jun; 24(4):
posterior lumbar interbody fusion. Neurosurgery 2002; 24250.
51(5 Suppl): S16671. 74. Marotta N, et al. A novel minimally invasive presacral
60. Blume HG. Unilateral posterior lumbar interbody approach and instrumentation technique for anterior
fusion: simpliied dowel technique. Clin Orthop Relat L5-S1 intervertebral discectomy and fusion: technical
Res 1985; (193): 7584. description and case presentations. Neurosurg Focus
2006; 20(1): E9.
61. Schwender JD, et al. Minimally invasive transforaminal
lumbar interbody fusion (TLIF): technical feasibility 75. Patil SS, et al. Clinical and radiological outcomes of axial
and initial results. J Spinal Disord Tech 2005; 18 (Suppl): lumbar interbody fusion. Orthopedics 2010; 33(12): 883.
S16. 76. Botolin S, et al. High rectal injury during trans-1 axial
62. Adogwa O, et al. Comparative efectiveness of lumbar interbody fusion L5-S1 ixation: a case report.
minimally invasive versus open transforaminal lumbar Spine 2010; 35(4): E1448.
interbody fusion: 2-year assessment of narcotic use, 77. Newton PO, et al. Anterior release and fusion in
return to work, disability, and quality of life. J Spinal pediatric spinal deformity. A comparison of early
Disord Tech 2011 Dec; 24(8): 47984. outcome and cost of thoracoscopic and open
63. Karikari IO, Isaacs RE. Minimally invasive transforaminal thoracotomy approaches. Spine 1997; 22(12): 1398406.
lumbar interbody fusion: a review of techniques and 78. Wong HK, et al. Results of thoracoscopic instrumented
outcomes. Spine 2010; 35(26 Suppl): S294301. fusion versus conventional posterior instrumented
64. Wang J, et al. Minimally invasive or open fusion in adolescent idiopathic scoliosis undergoing
transforaminal lumbar interbody fusion as revision selective thoracic fusion. Spine 2004; 29(18): 20318;
surgery for patients previously treated by open discussion 2039.
discectomy and decompression of the lumbar spine. Eur 79. Newton PO, et al. horacoscopic multilevel anterior
Spine J 2011; 20(4): 6328. instrumented fusion in a goat model. Spine 2003;
65. Villavicencio AT, et al. Minimally invasive versus open 28(14): 161419; discussion 1620.
transforaminal lumbar interbody fusion. Surg Neurol 80. Kim DH, Jaikumar S, Kam AC. Minimally invasive
Int 2010; 1: 12. spine instrumentation. Neurosurgery 2002; 51(5 Suppl):
66. Wu RH, Fraser JF, Hartl R. Minimal access versus open S1525.
transforaminal lumbar interbody fusion: meta-analysis 81. Newton PO, et al. Surgical treatment of main thoracic
of fusion rates. Spine 2010; 35(26): 227381. scoliosis with thoracoscopic anterior instrumentation.
67. Wang J, et al. Comparison of one-level minimally Surgical technique. J Bone Joint Surg Am 2009; 91
invasive and open transforaminal lumbar interbody (2 Suppl): 23348.
fusion in degenerative and isthmic spondylolisthesis 82. Lieberman IH, et al. Prone position endoscopic
grades 1 and 2. Eur Spine J 2010; 19(10): 17804. transthoracic release with simultaneous posterior
68. Peng CW, et al. Clinical and radiological outcomes of instrumentation for spinal deformity: a description of
minimally invasive versus open transforaminal lumbar the technique. Spine 2000; 25(17): 22517.
interbody fusion. Spine 2009; 34(13): 13859. 83. Rampersaud YR, Annand N. Dekutoski MB. Use
69. Ozgur BM, et al. Extreme Lateral Interbody Fusion of minimally invasive surgical techniques in the
(XLIF): a novel surgical technique for anterior lumbar management of thoracolumbar trauma: current concepts.
interbody fusion. Spine J 2006; 6(4): 43543. Spine 2006; 31(11 Suppl): S96102; discussion S104.
70. Anand N, et al. Minimally invasive multilevel 84. Wang MY, Mummaneni PV. Minimally invasive surgery
percutaneous correction and fusion for adult lumbar for thoracolumbar spinal deformity: initial clinical
degenerative scoliosis: a technique and feasibility study. experience with clinical and radiographic outcomes.
J Spinal Disord Tech 2008; 21(7): 45967. Neurosurg Focus 2010; 28(3): E9.
71. Rodgers WB, Gerber EJ, Patterson J. Intraoperative and 85. Dakwar E, et al. Early outcomes and safety of the
early postoperative complications in extreme lateral minimally invasive, lateral retroperitoneal transpsoas
interbody fusion: an analysis of 600 cases. Spine 2011; approach for adult degenerative scoliosis. Neurosurg 137
36(1): 2632. Focus 2010; 28(3): E8.
Section 2: Spine surgery for adult patients

86. Anand N, et al. Mid-term to long-term clinical and 88. Mannion RJ, et al. Safety and eicacy of intradural
functional outcomes of minimally invasive correction extramedullary spinal tumor removal using a
and fusion for adults with scoliosis. Neurosurg Focus minimally invasive approach. Neurosurgery 2011;
2010; 28(3): E6. 68(1 Suppl Operative): 20816; discussion 216.
87. Haji FA, et al. Minimally invasive approach for the 89. Uribe JS, et al., Minimally invasive surgery treatment
resection of spinal neoplasm. Spine 2011; 36(15): for thoracic spine tumor removal: a mini-open, lateral
E101826. approach. Spine 2010; 35(26 Suppl): S34754.

8.5 Posterior lumbar interbody fusion


Virgilio Matheus and William Bingaman

Key points
his chapter will:
Review the indications for performing a
posterior lumbar interbody fusion.
Explain the critical portions of the surgical
procedure from an anesthesia point of view.
Discuss possible complications related to the
procedure.
Review some critical steps the anesthesiologist
can follow to minimize complications
Summarize the postprocedural care.

Introduction
Posterior lumbar interbody fusion (PLIF) is a sur-
gical intervention to restore and stabilize the sagittal
alignment of the spine and distract the neuroforami-
nal space. Cloward irst introduced it in 1945 with the Figure 8.11 Sagittal MRI of the lumbar spine. Note the
intention to treat disc herniations.1 he operation is anterior displacement of L4 on L5 (arrow) representing a grade II
performed via a posterior approach with wide lamin- spondylolisthesis.
ectomy, which allows excellent neural decompression
and access to the anterior vertebral column (disc space) extremity pain as the nerve root exiting at the involved
for distraction and stabilization. level becomes chronically compressed. Other less com-
mon indications for PLIF include recurrent lumbar disc
Indications herniation, chronic medically intractable degenerative
Lumbar spondylolisthesis is the condition most disc disease, and pseudoarthrosis ater attempted lum-
commonly treated by lumbar fusion (Fig. 8.11). bar posterolateral transverse process fusions (Table 8.2).
Spondylolisthesis results most commonly from degen- Relative contraindications include previous interbody
erative disease but also may occur due to trauma, fusion attempts, upper level spinal pathology (places the
infection, or iatrogenic causes following lumbar decom- spinal cord at risk from retraction), and multiple levels
pression surgery. he pathophysiology of lumbar spond- of spinal instability where PLIF may lead to excessive
ylolisthesis involves anterior translation of the spine at the blood loss. General contraindications to any lumbar
involved level due to instability in the supporting spinal fusion procedure include advanced osteoporosis and
138 structures including the disc and facet joints. his leads signiicant preexisting medical conditions that place the
to mechanical back pain and eventually radicular lower patient at risk.
Chapter 8.5: Posterior lumbar interbody fusion

Table 8.2 Common indications for PLIF

Indications Contraindications Complications


Spondylolisthesis Pathology above L2 where retraction Hardware failure or misplacement
may cause cord injury.
Recurrent disc herniation Severe osteoporosis precluding safe General surgical complications
hardware implantation (e.g., infection, bleeding)
Failure of intertransverse Previous lumbar decompression with Radiculopathy from manipulation
process fusion severe epidural scaring of the nerve roots
Degenerative disc disease Arachnoiditis Vascular injury (iliac vessels)

Procedure
Positioning and patient preparation
Given the prone positioning, the potential for long
operative times, and the potential for blood loss requir-
ing transfusion, general anesthesia with endotracheal
intubation is usually the recommended technique. It is
worth mentioning that since most of these patients have
degenerative disease of the spine, the anesthesiologist
and spine surgeon should discuss the safety and feasi-
bility of maneuvering the cervical spine for intubation
purposes. Once the airway is secured, the endotracheal
tube is securely fastened to avoid dislodgement in the
prone position. Traditional intraoperative monitoring
should include intra-arterial monitoring of blood pres- Figure 8.12 Variety of available head-holding systems. On the left
is the rigid head holder allowing clamp ixation of the head; in the
sure, bladder catheterization, and adequate intraven- center is the foam pillow; and on the right are cranial tongs allowing
ous access. Central venous pressure monitoring is also the head to move freely when attached to a pulley system.
sometimes recommended depending on the underlying
medical condition of the patient. Prior to inal posi-
tioning, the surgical team should discuss whether the
patients head will be placed on a face cushion or head
clamp system to decrease the incidence of postoperative
vision loss (Fig. 8.12). Careful attention should be paid
to the patients eyes to avoid corneal abrasions, which
have a much higher incidence in the prone position.
he blood loss expected for posterior lumbar fusion
difers depending on several variables including single
vs. multiple operative levels, abdominal decompres-
sion (positioning and surgical table), surgeons experi-
ence, and open vs. minimally invasive technique (Fig.
8.13). Recent studies have quoted an average blood loss
of 900 ml without iliac crest bone grat harvesting and
1400 ml when bone grat is harvested.2 More recent Figure 8.13 Jackson frame surgical bed. The abdomen is allowed
studies quote lower blood losses ranging between 200 to freely hang through the center opening, relieving pressure and
and 300 ml.3 Careful attention to patient positioning decreasing venous congestion in the epidural plexus.
can help to reduce blood loss by relaxing the abdominal
musculature and decreasing the amount of blood pre- Attention to other intraoperative concerns includes
sent in the epidural venous plexus.4,5 Finally, intraop- the avoidance of hypothermia through the use of body
erative cell salvage can be utilized to minimize the risks warmers and the possible need for electrophysiologic 139
associated with blood transfusion. monitoring to avoid nerve root injury. While not a
Section 2: Spine surgery for adult patients

standard of care, intraoperative nerve root moni- intervertebral spacers and bone grat. his interverte-
toring is sometimes performed and may necessitate a bral bone grat will be compressed when the patient
change in anesthetic management. stands up and will serve as the ultimate strength of
the surgical fusion. Ater the spacers are placed, the
Surgical technique remainder of the disc space is packed with morselized
A midline incision is made at the desired level of the bone taken from the iliac crest or from the bone
lumbar spine. he subcutaneous tissue is dissected removed during the surgical decompression. Ater
using electrocautery down to the level of the dorsal this is completed, the nerve root foramina are checked
fascia. A subperiosteal exposure of the lumbar spine to make sure the roots are adequately decompressed.
is performed. Careful attention to hemostasis during horough inspection of the hardware, decompression
exposure is critical to avoid excessive blood loss over site, and hemostasis is performed one last time, ater
the course of the procedure. Intraoperative radio- which adequate closure by layers is performed. A inal
graphic veriication of the correct level is made. A intraoperative radiograph may be obtained to verify
bilateral laminectomy is performed exposing the the- adequate positioning of the hardware prior to emer-
cal sac. Careful mobilization of the dural sac as well gence from anesthesia (Fig. 8.14). A subfascial drain
as the exiting root is performed medially in order to will be inserted prior to skin closure based on the sur-
expose the vertebral disc. he aim of laminectomy geons preference. In some instances the surgeon will
is to allow for relief of the spinal stenosis and allow place an epidural catheter prior to closure for post-
exposure of the pedicles for safe insertion of the titan- operative pain control.
ium pedicle screws for stabilization. For this part of
the procedure luoroscopy or stereotactic navigation Postprocedural care
may be used to accomplish adequate placement of the Normal postoperative care will include avoiding hyper-
intervertebral hardware.6 Once the pedicle screws tension, adequate analgesia, and safe mobilization of
are placed, the annulus of the afected disc space is the patient in the hospital environment. Antibiotics
divided and the disc space is sequentially opened with are given from just prior to the induction of anesthesia
mechanical disc space distractors. his is made safer and stopped within 24 hours aterwards. Corticosteroid
by removal of the inferior facet joint of the superior therapy may be utilized to help reduce inlammation
vertebral level allowing for wider exposure and eas- and improve patient comfort. Assessment by physical
ier preparation of the interbody space. Once the disc and occupational therapy allows for safe mobiliza-
space height is restored to the desired value, titanium tion and discharge planning. Orthotics is generally
plates or rods are placed across the pedicle screws to consulted to it the patient for a rigid or semi-rigid lum-
hold the disc space in the correct position. At this bar orthosis. Attention to hospital risk management
point, the disc material is removed with a combination includes prevention of deep venous thrombosis through
of disc shavers and curettes to allow for placement of the use of a variety of antithrombotic devices/drugs

Figure 8.14 AP (A) and lateral (B)


weight-bearing radiographs of the
lumbar spine following surgery. Note
the construct composed by the pedicle
screws and ramps interconnecting both
vertebral bodies. Small radiopaque dots
in the disc space represent the interbody
graft in place.

140 (A) (B)


Chapter 8.6: Vertebral compression fractures

and the prevention of urine and wound infection by References


aggressive control of blood sugar and aseptic technique.
1. Cloward RB. he treatment of ruptured intervertebral
discs by vertebral body fusion. I. Indications, operative
Complications techniques, ater care. J Neurosurg 1953; 10: 15468.
Complications can be divided into those speciic to the 2. Yasuhisa A, Masaki T, Hisashi K. Comparative study of
operation and those common to spinal surgery. he iliac bone grat and carbon cage with local bone grat in
general complications include risk of blood loss, infec- posterior lumbar interbody fusion. J Orthop Surg 2002;
tion, deep venous thrombosis, urinary tract infection, 10(1): 17.
ileus, and cardiopulmonary complications related to 3. Freudenberger C, Lindley E, Beard E, et al. Posterior vs.
general anesthesia.79 Complications speciic to the anterior lumbar interbody fusion with anterior tension
procedure are centered on nerve root or thecal sac band plating: retrospective analysis. Orthopedics 2009;
injury due to manipulation/retraction of the neural 32(7): 492.
elements during surgery and those related to implanta- 4. Singh A, Ramappa M, Bhatia C, Krishna M. Less invasive
tion of the hardware. Postoperative neuralgia occurs in posterior lumbar interbody fusion and obesity: clinical
approximately 7% of patients, likely related to surgical outcomes and return to work. Spine 2010; 35(24): 211620.
manipulation of the involved nerve root.10 Hardware 5. Frymoyer J, Wiesel S. horacolumbar spine-Interbody
complications include improper placement and/or fusion. In: he Adult and Pediatric Spine: Principles,
migration. he hardware consists of titanium pedicle Practice and Surgery. 3rd ed. Philadelphia, PA:
screws, rods/plates, and the interbody spacers. As with Lippincott Williams & Wilkins; 2003: 11416.
any implanted device, improper placement can result 6. A. Vaccaro, T. Albert. Posterior lumbar interbody
in neural or vascular injury. A potentially devastating fusion. In: Spine Surgery: Tricks of the Trade. 2nd ed.
complication is violation of the anterior longitudinal New York: hieme; 2008: 11545.
ligament and iliac vessel injury leading to catastrophic 7. Hosono N, Namekata M, Makino T, et al. Perioperative
hemorrhage. Long-term complications include pseu- complications of primary posterior lumbar interbody
doarthrosis (failure of the bony fusion) and adjacent fusion for nonisthmic spondylolisthesis: analysis of risk
factors. J Neurosurg Spine 2008; 9: 4037.
segment failure and may lead to ongoing back pain
requiring further surgery. he overall complication 8. Cho KJ, Park SR, Kim JH, et al. Complications in
rate associated with PLIF varies widely in the literature posterior fusion and instrumentation for degenerative
lumbar scoliosis. Spine 2007; 32: 22327.
with rates ranging from 6.7% to 68%.8,9 It is important
to note that most of these are transient with no result- 9. Krishna M, Pollock RD, Bhatia C. Incidence, etiology,
ing permanent neurologic deicits. classiication, and management of neuralgia ater
posterior lumbar interbody fusion surgery in 226
patients. Spine 2008; 8: 3749.
Conclusion 10. Kasis AG, Marshman LA, Krishna M, et al. Signiicantly
Posterior lumbar interbody fusion is a safe and efect- improved outcomes with a less invasive posterior
ive technique to treat disorders of the intervertebral lumbar interbody fusion incorporating total
disc, especially lumbar spondylolisthesis. facetectomy. Spine 2009; 34: 5727.

8.6 Minimally invasive procedures for vertebral compression


fractures
Jason E. Pope and Nagy Mekhail

Key points he sites of vertebral compression fracture are


commonly in the thoracolumbar junction
Vertebral compression fractures are common
(T12/L1), followed by T7/T8,
in the aged population and carry signiicant
Risk factors for vertebral compression
morbidity and mortality consequences. 141
fractures are most commonly a consequence of
Section 2: Spine surgery for adult patients

junction (T12/L1), followed by (T7/T8), whereas the


osteoporosis, and are less commonly associated
morphology of vertebral compression fractures can
with primary or secondary cancers.
vary through crush, biconcave, and wedge (most com-
Management of vertebral compression fractures
mon).8 A high degree of clinical suspicion for cancer
follows the principles of orthopedic fracture needs to be maintained in fractures outside of these
management locations and in patients without the commonly asso-
Vertebral augmentation procedures are very ciated risks factors (see below).
efective in eliminating pain and disability in Dreadfully, an estimated 75 000100 000 cancer-
the appropriately selected patient. related fractures occur annually. Metastatic disease,
here are multiple laws in the literature including all stages of multiple myeloma, stage III pros-
discounting its beneit for selected patients. tate cancer, and stage IV breast and lung cancer are
commonly implicated in VCFs;913 3070% of those
who die from cancer annually have bone metastasis.14
Introduction Distribution of tumor types causing VCF includes: lung
Vertebral compression fractures (VCFs) are most (20%); breast (6%); myeloma (22%); prostate (32%); and
commonly a consequence of osteoporosis and less fre- others, including bladder, thyroid, and survival of pedi-
quently of primary or metastatic cancers. Osteoporosis atric solid tumors (20%).15 Metastatic bone lesions are
is a systemic skeletal disease that is characterized by either osteoblastic or osteolytic, where the latter predis-
reduced bone strength, caused by a deiciency in peak pose to a higher risk of fracture.16 Osteoblastic lesions
bone mass during growth and development, inad- are characterized by increased bone density, maintained
equate bone formation, and excessive bone resorp- bone strength, but decreased bone integrity (stifness),
tion, resulting in disruption in architecture and mass.1 and are common in prostate cancer. Osteolytic lesions are
Secondary osteoporosis can be a consequence of medi- characterized by decreased bone density and decreased
cations (steroids, chemotherapy, anticonvulsants), strength and integrity, and are common in patients with
endocrinopathies (hyperthyroidism, Cushings syn- multiple myeloma. Not only does cancer cause VCF, so
drome, hyperparathyroidism, and hypogonadism), does its treatment.17 he long-term use of oral glucocor-
toxins (alcohol, tobacco), cancers, and malnutrition. ticoids increases the risk of fracture 2.6-fold.18
he National Osteoporosis Foundation estimates that he consequences of vertebral compression frac-
100 million people worldwide and 44 million people tures are far reaching. Patient-associated consequences
in the United States are at risk for developing osteo- are both physical and psychological. Physical conse-
porosis. Over 2 million fragility fractures in the United quences include bone pain, disability, radiculopathy,
States can be attributed to osteoporosis, of which 26% spinal cord compression, kyphosis, and related reduc-
are vertebral body fractures.2. Moreover, two-thirds tion on lung volume, abdominal content compression,
are let undiagnosed and even fewer are appropriately impaired physical function, impaired gait, sleep disor-
treated.2 Not surprisingly, VCF prevalence increases ders, decreased adult activities of daily living (ADL),
with age, as 1 in 2 women and 1 in 4 men over the age of and hypercalcemia. Patients with VCF are 23 times
50 years will have an osteoporosis-related fracture.2,3 more likely to die of pulmonary causes, i.e., pneumonia
Vertebral compression fractures are the most because of increased work of breathing, decreased func-
common osteoporotic fracture and an estimated 700 tional residual capacity, and the associated kyphosis
000 occur annually.4 Over 150 000 people are hospi- causing reduced lung capacity. Psychological conse-
talized for pain and management of spinal fractures quences include depression, reduced self-esteem, anx-
each year.5 Approximately 260 000 patients are diag- iety, reduced autonomy, and reduction in the quality of
nosed with their irst painful fracture each year,5 and life.19 Van Schoor looked at 334 people aged 65 or older
ater the irst, the risk of a subsequent VCF increases with VCF assessed by radiography and SF-12 (quality
5-fold, 12-fold ater two or more, and 75-fold ater two of life measure) and reported that patients with three
or more and low bone mass below the 33rd percentile.6 or more vertebral compression fractures had a loss of
VCFs commonly occur spontaneously and can result quality of life compared with patients with stroke.19
from minimal or no trauma. Minor activities, such as VCFs increase the risk of mortality.5 Johnell reported
bending, liting, and coughing have been implicated a mortality of near 80% for patients 5 years ater verte-
142 with developing vertebral compression fractures.7 he bral compression fracture (mean age of patients at the
sites of fracture are commonly in the thoracolumbar time of fracture was 78.6 years).20
Chapter 8.6: Vertebral compression fractures

he economic burden is as telling as the consequen- Table 8.3 Indications for vertebral augmentation
ces of VCF for the patient. Annual costs of osteoporotic Symptomatic VCF
fractures continue to rise; estimates were as high as Osteoporotic/osteolytic/malignant/benign fractures
$13.8 billion in 1995 with an increase to $16.7 billion in refractory to conservative medical therapy
2003.21,22 Two minimally invasive options for vertebral Multiple compression fractures where further spinal
body augmentation are vertebroplasty and kyphoplasty. deformity would result in depressed respiratory function
Both entail injecting polymethylmethacrylate into the Painful fracture secondary to osteonecrosis (Kummels
disease)a
fractured vertebrae to ix the fracture and relieve the
pain. A detailed discussion of these techniques follows. Unstable fracture with movement of wedge deformity
Chronic nonunion traumatic fracture
Pain localization and presentation suggestive of VCF
Patient selection (bandlike, localized, axial)
An appropriate history, physical examination, and Ideally acute/subacute fractures (<1 year old) although
radiographic testing are a necessity and cannot be symptomatic chronic fractures suitable candidates
understated. As alluded to previously, historical indi- Older patient age > younger patient age
cators for suspicion of VCF include age greater than a
Vascular necrosis of the vertebral body after a vertebral
50 years, exposure to known medications with known compression fracture.
bone strength and architecture detriment (including
glucocorticoid therapy), endocrinopathies, known
cancers with either primary or metastatic vertebral Table 8.4 Absolute contraindications for vertebral
body sites, a diagnosis of osteoporosis, and female sex. augmentation
Furthermore, relatively minor or no traumatic history Asymptomatic stable fracture
is oten reported, and includes cough, sneeze, bending, Efective conservative medical measurement
and twisting maneuvers. Oten, patients report awak-
Systemic or localized infection
ening with pain. he pain is typically localized and axial
Osteomyelitis of target vertebra
in nature, with or without a bandlike radiation, and can
be elicited longitudinally up to 10 cm (4 inches) (one to Uncorrected coagulopathy
two levels) from the VCF (caudal presentation is more Allergy to bone cement or contrast agents
common than rostral, anecdotally). he pain is typi- Acute traumatic fracture of nonosteoporotic vertebra
cally exacerbated by movement. Spinal percussion may Prophylaxis in osteoporotic patients
be helpful to delineate the symptomatic vertebral com- Fractured pedicles
pression fracture level, but this sign is not sensitive or Burst fracture (traumatic injury resulting in pieces of
speciic.23 Indications, as well as relative and absolute vertebral body shattering into surrounding tissues)a
contraindications are listed in Tables 8.38.5.24 Retropulsed fragment causing signiicant canal
Radiologic correlation is paramount. Modalities compromiseb
employed include plain ilms, magnetic resonance a
Although some authors advocate using vertebroplasty for
imaging, bone scan, and computed tomography. Axial burst fracture management.25
b
Some authors consider retropulsed fragments relative
and sagittal MRI are generally recommended currently, contraindications.
as it provides anatomic detail and suggests fracture age. a,b
Both conditions may require open ixation rather than
Uniformly, acute fractures typically have a low T1 and minimally invasive techniques.
high T2 signal and T2 STIR signal on MRI, secondary
to edema and increased water content.
Computed tomography (CT) with a bone scan is an Table 8.5 Relative contraindications for vertebral compression
excellent imaging modality when MRI cannot be used, fractures
either secondary to inability to tolerate MRI positioning Radicular pain (caused by a compressive syndrome unrelated to
or if the patient has preexisting implanted metal devices vertebral body collapse
(ferromagnetic). Bone scans with increased tracer Younger age
uptake are suggestive of recent fracture. CT scans difer- Tumor extension into the epidural space
entiate fat from other sot tissue and accurately describe Severe vertebral body collapse >70% (vertebra plana)
bony architecture. Importantly, CT exposes the patient 143
to ionizing radiation, whereas MRI does not.
Section 2: Spine surgery for adult patients

ater true AP and lateral images are conirmed, a mark


Management is placed approximately 5 cm (2 inches) lateral and
Management of vertebral compression fractures is
parallel to the superior end plate of the target vertebral
largely medical or surgical. Medical management
body, with the projected trajectory to the upper pos-
includes symptom reduction by conservative man-
terior portion of the pedicle waist (1011:00 on LEFT
agement: analgesics, bracing, and institution of pre-
and 23:00 on the RIGHT). Ater local iniltration with
ventative measures (supplementation of calcium
local anesthetic (50:50 mixture of 0.75% bupivacaine
and vitamin D; calcitonin; antireabsorption agents
and 2% lidocaine with epinephrine 1:400 000), the
(bisphosphonates); chemotherapy/radiation therapy
tip of the trochar is advanced to the pedicle. Precise
for cancer; and hormonal therapy (estrogen, PTH).
correction of trajectory is paramount and should be
here are questions regarding conservative manage-
checked with sequential AP and lateral images. For
ment with analgesics: How long should they be trialed
the extrapedicular approach, the trajectory is similar
while attempting to mitigate the aforementioned VCF
to the transpedicular approach, although entry is out-
sequelae? Innately, orthopedic principles of fracture
side the pedicles, either lateral to the junction of the
management include anatomy restoration, rigid ix-
transverse process and superior articular process or
ation, minimal tissue disruption, and safe and early
between the rib and transverse process at the superior
mobilization.26 In can be argued that conservative
lateral pedicle wall, depending on the level of the tho-
management for symptomatic VCF does not fulill
racic vertebra. An en-face, coaxial approach may aid in
these well-established principles.
tool placement, but is usually not advocated.
here are some authors who advocate venography
Techniques prior to cement injection to evaluate for potential
routes of cement extravasation. he physical proper-
Vertebroplasty ties of cement are diferent from those of the contrast
Deramond and Gilbert introduced vertebroplasty in medium (i.e., viscosity), and injectate radiographic
France in 1984 for treatment of hemangiomas. he spread correlation is not guaranteed.27 Venography is
procedure was then employed in the United States in therefore not routinely performed, exclusive of vascu-
1993. he procedure is typically performed in an out- lar lesions. In patients with osteoporosis or hemangi-
patient setting under monitored anesthesia care and omas, 2.54 cm3 of cement can provide optimal illing
local anesthesia. Routinely, perioperative antibiotics of the vertebra; in tumor smaller volumes of 1.52.5 cm3
are given within 30 minutes prior to the procedure. he are suicient.28 he cement is mixed and the viscosity
procedure itself takes between 30 and 60 minutes, with is checked, as it is recommended that it should have
the majority of the time dedicated to patient position- toothpaste consistency. he injection should be mon-
ing and luoroscopic C-arm alignment, and biplanar itored under live luoroscopy with care in emptying the
luoroscopy is advocated. he patient is self-positioned trocar before removing it to avoid extravasation. he
prone to ensure reduce joint stress and minimize risk of patient is instructed to remain prone until the cement
intraoperative compressive neuropathy. True AP and hardens (815 minutes). Some advocate a postproce-
true lateral luoroscopic views are required for success- dure CT scan, while others do not.29 he patient is to
ful and safe needle placement. In the AP projection, the stay recumbent for 35 h and neurologic evaluation is
pedicle outline is the waist of the pedicle, not the base. performed. Refer to Fig. 8.15.
A true AP projection is where the pedicles are in the
upper half of the vertebral body, the spinous process is
equidistant between the pedicles, and the end plates are Kyphoplasty
parallel. True lateral projection assurance is conirmed Kyphoplasty is a variation of percutaneous vertebro-
by parallel end plates and superimposed pedicles. Two plasty, where the diference lies in mechanical fracture
common approaches are described: extrapedicular or reduction and cavity creation through the percutan-
transpedicular, and both are intended to be bilateral. eous introduction of an inlatable balloon tamp (IBT).
Transpedicular approaches, where appropriate (low Typically, this procedure is performed under general
thoracic and lumbar) have been advocated to reduce the anesthesia. Patient positioning, imaging, and trocar
chance of cement extravasations. It is vitally important placement are the same as for vertebroplasty. Biopsy
144 to consider fracture orientation and trajectory before is usually taken to determine or conirm the cause of
needle positioning. For the transpedicular approach, the vertebral body compression fracture. he IBT has
Chapter 8.6: Vertebral compression fractures

Figure 8.15 Lateral (A) and AP (B) images of vertebroplasty.

bidirectional and unidirectional inlating properties,


and can range from 10 to 20 mm in length. It is cru-
B
cial to understand that the operator controls the volume
of IBT inlation, as it is recommended to increase the Figure 8.16 Lateral (A) and AP (B) luoroscopic views of trocar
volume in 0.25 cm3 increments (180 turn of plunger). placement for kyphoplasty.
he maximum pressure recommended is 400 psi (2.76
Mpa) without decay and the maximum volume is 4 cm3 he more viscous cement is recommended for kypho-
for the 10 and 15 mm IBT and 6 cm3 for the 20 mm IBT. plasty because the injection occurs under lower pressure
he goal is IBT introduction into the anterior one-third and ills an iatrogenic void, in contrast to vertebroplasty
of the vertebral body. he IBT is then removed and the where injection is under high pressure without cavity
bone iller device (BFD) is introduced approximately creation. It is recommended not to add additives to the
35 mm into anterior cortex. Cement mixing is simi- cement, e.g., antibiotics. It is recommended to inject the
lar to that for vertebroplasty, although the consistency cement via the BFD to 5 mm of posterior wall, start-
is usually much more viscous, and the cement appears ing anteriorly, under live luoroscopy to completely ill
doughy; it usually takes 8 minutes, depending on the void created and to reduce occurrence of extravasa- 145
ambient temperature, rigor in mixing, and handling. tion.30 Refer to Figs. 8.16 through 8.18.
Section 2: Spine surgery for adult patients

Figure 8.18 (cont.)

Table 8.6 Technical considerations

Kyphoplasty Vertebroplasty
Figure 8.17 After cement injection in lateral projection. In-patient procedure, Outpatient
increasingly outpatient
General anesthesia, monitored Monitored anesthesia care
anesthesia care
Single or bi trans/extrapedicular Single or bi trans/
extrapedicular
8 G trochar 1113 G trochar
Procedure time slower than Procedure time faster than
vertebroplasty kyphoplasty
Cost $$$$ Cost $
VCF anatomy restoration ++++ ++
Cement viscosity higher Cement viscosity lower
Injection under lower pressure Injection with higher pressure

Central cavity creation


As described briely earlier, kyphoplasty utilizes
curettes to create a space to accommodate the IBT
and cement deposition. A variation of vertebroplasty
involving similar mechanisms with IBT deployment
has also been described.31 Important technical consid-
erations regarding the decision to perform either pro-
cedure are listed in Table 8.6.

Complications
A Complications associated with vertebral body aug-
mentation are procedure speciic and are traditionally
Figure 8.18 Anteroposterior (A) and lateral (B) images of
kyphoplasty.
either procedural or extravasations of cement in nature.
Collectively, the clinically meaningful complication
146 rate of vertebroplasty is reported to be approximately
Chapter 8.6: Vertebral compression fractures

60 Figure 8.19 Percentage of occurrence


Vertebroplasty and distribution of cement leakage
Kyphoplasty by location in vertebroplasty and
kyphoplasty. (From Hulme P, Kebs J,
50
Ferguson S, Berlemann U, Vertebroplasty
and kyphoplasty: a systematic review of 69
clinical studies. Spine 2006;31:19832001.
40 Reproduced with permission.)
Occurrence (%)

30

20

10

0
Epidural Paraspinal Intradiscal Pulmonary Foraminal

0.70 suspected of increasing adjacent vertebral body com-


Pulmonary emboli
pression fractures, with rates of 1252%.35 Severe clin-
0.60 Neurologic complications
ical complications for vertebroplasty and kyphoplasty,
0.50
including pulmonary emboli and neurologic seque-
lae occurred at rates of 0.6% and 0.1%, and 0.6% and
Incidence (%)

0.40 0.03%, respectively.33 Pulmonary emboli likely resulted


from venous uptake via paravertebral veins. Other pul-
0.30 monary reactions to cement, as for long bone fracture
repair, can induce a pulmonary hypertension.
0.20
Extravasation management is dependent on symp-
0.10 tomatology and includes observation or surgical
decompression. Perivertebral leakage of cement has
0.00 been reported to occur in up to 65% of treated osteo-
Vertebroplasty Kyphoplasty porotic VCFs.36 Nevertheless, the majority of patients
Figure 8.20 Incidence of severe clinical complications following are asymptomatic.37,38 Some may experience transient
vertebroplasty and kyphoplasty. (Created from data presented in neurologic indings, from mechanical compression or
Moreland DB, Landi MK, Grand W, Vertebroplasty: Techniques to
avoid complications. Spine J 2001;1:6671.) chemical irritation, but these rarely persist for more
than 4 weeks.

13%, and complications include infection, bleeding,


increased pain, numbness, tingling, extravasation of Outcomes and ecacy
cement, paralysis, and death. he overall complica- Relief observed ater vertebral body augmentation has
tion rate of vertebroplasty, as reported by Murphy and consistently been reported to occur almost immedi-
Deramond in 2000, was 1.3% for osteoporosis, 10% for ately ater the procedure to 12 days postoperatively.39,40
cancer, and 2.5% for hemangiomas.32 In comparison, he hypothesized mechanisms of pain relief from ver-
Hulme reviewed the distribution of cement leakage tebral body augmentation include immobilization of
and percentage of occurrence for kyphoplasty and ver- the fracture, and/or intraosseous neural destruction
tebroplasty and reported, respectively, 11% and 32% from the exothermic PMMA (polymethylmethacryl-
epidural, 48% and 32.5% paraspinal, 38% and 30.5% ate) reaction, and/or a direct neural cytotoxic efect
intradiscal, 1.5% and 1.7% pulmonary, and 1.5% and of the PMMA. However, recent studies suggest that
1.7% foraminal. Refer to Figs. 8.19 and 8.20.33,34 Both mechanical stabilization is the most likely mode of 147
vertebral body restoration procedures have been pain relief.41,42
Section 2: Spine surgery for adult patients

he biomechanics of the spine following vertebral Mean normal stress


body augmentation have been investigated as there Augmented
have been some suggestions of an increased risk of
adjacent vertebral body fracture, and that the risk may
be diferent ater vertebral body augmentation with
cavity creation (kyphoplasty). Under normal physio-
logic conditions, load is shared between the discs, the
vertebral bodies, the ligaments, and the facets. Anterior
Inc : 10
load shit, increased vertebral body stifness, and load Time : 1.000e+ L4
transfer through the intervertebral disc are suggested
0.000e + 00
reasons for potential increased potential for VCF at
adjacent levels. See Fig. 8.21. he response of a FSU 3.7503e 01
(functional stif component) to a load is governed by 7.500e 01
the less stif component, i.e., the intervertebral disc, 1.125e + 00
which is not severely disrupted. In a prospective study 1.500e + 00
of 25 patients with 34 fractures who were treated with 1.875e + 00
vertebroplasty and followed for 2 years, 52% had devel- 2.250e + 00
oped at least one VCF, where the odds ratio of develop- L5
2.265e + 00
ing an adjacent VCF to a treated, augmented vertebra
3.000e + 00
was 2.27 versus 1.44 for a new fracture unrelated to the
treated vertebra and was not statistically signiicant.43
Although static compression tests demonstrate con-
tinued slight redistribution of stress and stifness fol- A
lowing augmentation that may contribute to adjacent
Mean normal stress
vertebral body fracture,44 it is clear that subsequent
Non-augmented
VCF development is multifactorial and may be a con-
sequence of progression of disease rather than slightly
altered biomechanics of a slightly increased load and
geometric misalignment.
Luo et al. demonstrated that vertebral augmentation
could restore normal spine mechanics.45 Numerous
studies have reported incidence rates of VCF ater aug- Inc : 10
mentation with PMMA for kyphoplasty and vertebro- Time : 1.000e+ L4
plasty but are inconsistent in fracture deinitions and 0.000e + 00
time spans, with interpretation confounded by age, 3.7503 01
sex, bone density, and disease progression, to name a 7.500e 01
few factors. Furthermore, when comparing the bio-
1.125e + 00
mechanics of the spine following fusion with the alter-
1.500e + 00
ation following vertebral body augmentation, it is clear
1.875e + 00
that vertebral body augmentation is more physio-
logic. Compared with anatomy restoration by postural 2.250e + 00
L5
reduction (commonly lordotic positioning), kypho- 2.265e + 00
plasty doubled the inal height achieved by the former 3.000e + 00
and was maintained throughout follow-up of 1 year.
he clinical beneits of vertebral body augmenta-
tion are seemingly impressive. As demonstrated by B
Ledlie and Renfro, in 77 patients followed for 2 years Figure 8.21 A comparison of stresses in a motion segment, showing
ater kyphoplasty, 90% had complete pain relief; use of the load shift that results from rigid cement augmentation. (Adapted
analgesic medications decreased 87%; there were no from Hadley C. Abdulreham A, Zoarski GH, Biomechanics of vertebral
148 bone augmentation. Neuroimag Clin N Am 2010 15967.)
procedure-related complications; and vertebral body
Chapter 8.6: Vertebral compression fractures

augmentation was maintained.46 Comparing augmen- lists some of the diferences and similarities between
tation with medical management at 6 months, Komp kyphoplasty and vertebroplasty. Ater factoring in
et al. demonstrated progressive vertebral body height operating room time, general anesthesia, and overnight
loss in 94% of patients and subsequent fracture rate of admission, kyphoplasty costs 1020 times more than
65% with nonsurgical treatment, against 0% vertebral vertebroplasty.49 Ignoring the extrapolated costs, the
body height loss and 37% subsequent fracture rate fol- vertebroplasty kit without cement costs approximately
lowing kyphoplasty.47 Komp et al. also demonstrated $400, whereas the kyphoplasty kit costs $3400.49
that kyphoplasty improved pain and function by VAS Despite the kyphoplasty vs. vertebroplasty debate, a
(visual analogue scale) reduction and Oswestry Back mountain of evidence exists suggesting their appropriate
Disability Index reduction in patients compared with use in the management of vertebral compression frac-
nonsurgical management in fractures with mean age of tures to reduce both morbidity and mortality. In stark
1 month that were followed to 6 months.47 Furthermore, contradiction, the New England Journal of Medicine pub-
in comparing chronic fracture management greater lished two articles regarding the routine use of vertebro-
than 1 year old, kyphoplasty improved pain and func- plasty in fracture management in 2009 and concluded
tion by VAS and EVOS Functional Scale at 6 months.48 that there was no beneicial efect of vertebroplasty as
Controversy persists between kyphoplasty and ver- compared with a sham procedure in patients with pain-
tebroplasty.4952 Regarding anatomy restoration, both ful osteoporotic vertebral fractures up to 3 months of
vertebroplasty46 and kyphoplasty improve height and follow-up, and that there was no statistically signiicant
angular deformity; however, most agree that kypho- diference in pain improvement in the sham procedure
plasty is more efective. Kyphoplasty, on average, vs. vertebroplasty in pain and pain-related disability.56,57
improved angular deformity by 53% in one study and In a later reply, Buchbinder et al. state that58
47% in another.53,54 Hiwatashi et al, however, concluded Vertebroplasty appears to confer no beneit over sham
that the statistical signiicance of height restoration procedure or usual care, and it poses riskit would be
was not signiicant.55 Furthermore, Ledlie and Renfro neither appropriate nor moral to ofer this treatment in
reported maintained height of 88% prefracture height at routine care.
2 years.46 Kumar et al. prospectively analyzed 52 patients Let us look a bit closer. Buchbinder et al. performed
randomized to either kyphoplasty or vertebroplasty with a multicenter, randomized double-blind, placebo-
mean follow-up of approximately 42 weeks, recording controlled trial in which participants with one or two
outcomes pre- and post-procedure using VAS, Oswestry painful osteoporotic vertebral fractures that were of
Disability Index, the EuroQol-5D questionnaire, and less than 12 months duration and unhealed, as con-
the Short-Form 36 Health Survey. Both vertebroplasty irmed by magnetic resonance imaging, were randomly
and kyphoplasty were efective at improving pain on assigned to undergo vertebroplasty or a sham proced-
VAS (8.05.5 cm vs. 7.52.5 cm), functional disability ure. he sham procedure was introduction of the trocar
(57.638.4 vs. 50.728.8), and quality of life. However, to the posterolateral portion of the facet, with the sharp
kyphoplasty provides better results, which are main- stylet being replaced with a blunt one to mimic tapping
tained over long-term follow-up.52 and entry into the vertebral body ater local iniltration
Interestingly, Liu et al. randomly assigned 100 of osteum with local anesthetics.
patients with thoracolumbar compression fractures Participants were stratiied according to treatment
to either kyphoplasty or vertebroplasty using PMMA center, sex, and duration of symptoms (<6 weeks or
and assessed VAS scores and vertebral body height an 6 weeks). Outcomes were assessed at 1 week and at
kyphotic wedge angle from reconstructed CT images 1, 3, and 6 months and included the primary outcome
with follow-up to 6 months and concluded that degree of pain assessment by NRS at 3 months, and second-
of pain reduction, wedge angle improvement, and ary outcomes of quality of life (QUALEFFO), a VCF-
vertebral body height improvement did not difer sig- speciic questionnaire, AQoL (Assessment of Quality
niicantly and, secondarily to expense, suggested verte- of Life) questionnaire, sensitive to the elderly, and the
broplasty for osteoporotic VCFs.52 EQ-5D (European Quality of Life-5 Dimensions scale).
Literature support, as shown, can seemingly sub- he Kallmes study involved the random assignment
stantiate both sides of the debate on kyphoplasty vs of 131 patients who had one to three painful osteopor-
vertebroplasty. Reported advantages of either proce- otic vertebral compression fractures to undergo either
dure need to be weighed against their risks. Table 8.6 vertebroplasty or simulated procedure (as described 149
Section 2: Spine surgery for adult patients

previously) following 4 weeks of medical management, References


and patients were followed at 3, 14, and 90 days with
1. Raisz LG. Pathogenesis of osteoporosis: concepts,
the primary outcome of pain relief with the modi- conlicts, and prospects. J Clin Invest 2005; 115(2):
ied RolandMorris Disability Questionnaire (RDQ) 331825.
and numerical rating scale (NRS), where secondary 2. National Osteoporosis Foundation. http://www.nof.org
measures were EQ-5D and the Study of Osteoporosis (accessed December 1, 2011).
FractureActivities of Daily Living (SOF-ADL).
3. Oneil TW, Felsenberg D, Varlow J, et al. Prevalence of
here were multiple laws. No physical examin- vertebral deformity in European men and women: the
ation was performed and therefore pain could not European vertebral osteoporosis study. J Bone Miner Res
be attributed speciically to the VCF discovered on 1996; 11: 101018.
MRI. Furthermore, other sources of back pain were 4. Riggs BL, Melton LJ. he worldwide problem of
not controlled, as the aged population commonly has osteoporosis: insights aforded by epidemiology. Bone
symptomatic facet arthropathy. his could explain the 1995; 17: 50511.
description of pain relief following the procedure (i.e., 5. Cooper C, Atkinson EJ, OFallon WM, et al.
blocking of the medial branches with local iniltration Incidence of clinically diagnosed vertebral fractures:
in the sham group). Furthermore, because pain gen- a population-based study in Rochester, Minnesota,
erators rarely occur in an isolated presentation, the 19851989. J Bone Miner Res 1992; 7: 2217.
speciicity and sensitivity of the outcomes for VCF and 6. Ross PD, Davis JW, Epstein RS, Wasnich RD. Pre-
vertebroplasty versus sham is uncertain, as no follow- existing fractures and bone mass predict vertebral
up examination was performed. Interestingly, both fracture incidence in woman. Ann Intern Med 1991;
studies only treated only two levels (some patients had 114(11): 91923.
three levels of evidence VCF). One hundred and forty- 7. Aslan S, Karicioglu O, Katirici Y, et al. Speed-bump
one patients declined to participate in the Buchbinder induced spinal cord injury. Am J Emerg Med 2005; 23: 563.
study, while 300 declined to participate in the Kallmes 8. Nevit MC, Ross PD, Palermo L, et al. Association
study without an exclusion explanation. One explan- of prevalent vertebral fractures, bone density, and
ation could be that they feared randomization into the alendronate treatment with incident vertebral fractures:
sham group because of severe pain. efect of number and spinal location of fractures. Bone
1999; 25: 61319.
In the natural course osteoporotic fractures typi-
cally heal at 68 weeks, although MRI evidence of frac- 9. Berenson JR, Lichtenstein A, Porter L, et al. Eicacy of
pamidronate in reducing skeletal event in patients with
ture lingers. Furthermore, in-patients were excluded in
advanced multiple myeloma. NEJM 334(8): 48893.
the study and medical management was required for
4 weeks, removing the subacute VCF patient popula- 10. Brinker H, Westin J, Abildgaard N, et al. Failure of oral
pamidronate to reduce skeletal morbidity in multiple
tion and resulting in intervention on healed VCF. Only
myeloma: a double blind placebo controlled trial. Br J
32% and 44% of subjects enrolled in these studies were Haematol 1998; 101: 2806.
within this time frame, for Buchbinder and Kallmes,
11. Melton LJ, Kyle RA, Achenbach SJ, Oberg AL, Rajkumar
respectively.59 hese articles were overwhelmingly SV. Fracture risk with multiple myeloma: a population-
accepted as unequivocal evidence and incorrectly based study. J Bone Miner Res 2005; 20: 48793.
deride vertebroplasty as inefective.
12. Berruti A, Dogliotti L, Bitossi R, et al. Incidence of
skeletal complications in patients with bone metastatic
Conclusions prostate cancer and hormone refractory disease. J Urol
Comparative efectiveness research and evidence-based 2000; 164: 124853.
medicine are paramount to ensuring appropriate and 13. Diamond TH, Bucci J, Kersley JH, et al. Osteoporosis
accessible patient-centered pain care. Vertebroplasty and spinal fractures in men with prostate cancer: risk
and kyphoplasty are techniques that can reduce morbid- factors and efects of androgen deprivation therapy.
ity and mortality associated with vertebral compression J Urol 2004. 172: 52932.
fracture. Most contend that kyphoplasty, although more 14. Coleman RE. Metastatic bone disease: clinical features,
expensive and oten requiring general anesthesia and pathophysiology and treatment strategies. Cancer Treat
23-hour observation, is better at restoring vertebral body Rev 2001; 27: 16576.
height, with equivocal evidence of reduction in cement 15. Bilariki K, Anagnostou E, Masse V, et al. Low bone
150 extravasations secondary to central cavity creation. mineral density and high incidences of fractures and
Chapter 8.6: Vertebral compression fractures

vitamin d deiciency in 52 pediatric cancer survivors. 31. Vallejo R, Benyamin R, Floyd B, et al. Percutaneous
Horm Res Paediatr 2010; 74(5): 31927. cement injection into a created cavity for the treatment
16. Patel B, DeGroot H. Evaluation of the risk of pathologic of vertebral body fracture: preliminary results of a new
fractures secondary to metastatic bone disease. Orthop J vertebroplasty technique. Clin J Pain 2006; 22: 1829.
2001; 24: 61217. 32. Murphy KJ, Deramond H. Percutaneous vertebroplasty
17. Una Cidn E. Steroids, cancer and vertebral fractures: a in benign and malignant disease. Neuroimaging Clin N
dreaded complication. J Oncol Pharm Pract 2011; 17(3): Am 2000; 10: 53545.
279281. 33. Moreland DB, Landi MK, Grand W. Vertebroplasty:
18. Van Staa TP, Leukens HG, Abenhaim L, Zhang B, techniques to avoid complications. Spine J 2001; 1: 6671.
Cooper C. Use of oral corticosteroids and risk of 34. Hulme P, Kebs J, Ferguson S, Berlemann U.
fractures. J Bone Miner Res 2000; 15(6): 9931000. Vertebroplasty and kyphoplasty: a systematic review of
19. Silverman SL, Minshall ME, Shen W, Harper KD, 69 clinical studies. Spine 2006; 31: 19832001.
Xiw S. he relationship of health-related quality of 35. Fribourg D, Tang C, Sra P, Delamerter R, Bae H.
life to prevalent and incident vertebral fractures in Incidence of subsequent vertebral body fracture ater
postmenopausal women with osteoporosis: results kyphoplasty. Spine 2004; 29(20): 22706.
from the Multiple Outcomes of Raloxifene Evaluation
36. Cortet B, Cotton A, Boutry N. Percutaneous
Study. Arthritis Rheum 2001; 44(11): 261119.
vertebroplasty in the treatment of osteoporotic
20. Johnell O. Mortality ater osteoporotic fractures. vertebral compression fractures: an open prospective
Osteoporosis Int 2004; 15: 3842. study. J Rheumatol 1999; 26: 22228.
21. Truumees E. Osteoporosis [Editorial]. Spine 2001; 26: 37. Schmidt R, Cakir B, Mattes T. Cement leakage during
9302. vertebroplasty: an underestimated problem? Eur Spine J
22. Silverman SL. he clinical consequences of vertebral 2005; 14: 46673.
compression fractures. Bone 1992; 13(suppl 2): 2731. 38. Nassbaum DA, Gailloud P, Murphy K. A review
23. Gaughen JR, Jensen ME, Schweickert PA, et al. Lack of of complications associated with vertebroplasty
preoperative spinous process tenderness does not afect and kyphoplasty as reported to the food and drug
clinical success of percutaneous vertebroplasty. J Vasc administration medical device related website. J Vasc
Interv Radiol 2002; 13: 11358. Interv Radiol 2004; 15: 118592.
24. Soin A, Mekhail N. Minimally invasive procedures 39. Deramond H, Depriester C, Galibert P,
for vertebral compression fractures. Rajs Practical et al. Percutaneous vertebroplasty with
Management of Pain. 4th ed. Chapter 61. New York: polymethylmethacrylate. Technique, indications, and
Elsevier Mosby ; 2008: 111926. results. Radiol Clin North Am 1998; 36: 53346.
25. Shin JJ, Chin DK, Yoon YS. Percutaneous vertebroplasty 40. Truumees E, Hilirand A, Vaccaro AR. Percutaneous
for the treatment of osteoporotic burst fractures. Acta vertebral augmentation. Spine J 2004; 4: 21829.
Neurochir (Wien) 2009 Feb; 151 (2): 1418. 41. Urrutia J, Bono CM, Mery P, Rojas C. Early histologic
26. Helfet DL, Kloen P, Anand N, Rosen H. Open reduction changes following polymethylmethacrylate injection
and internal ixation of delayed unions and nonunions (vertebroplasty) in rabbit lumbar vertebrae. Spine 2008;
of fractures of the distal part of the humerus. J Bone 33(8): 87782.
Joint Surg Am 2003; 85: 3340. 42. Hadley C. Abdulreham A, Zoarski GH. Biomechanics
27. Gaughen JR, Jensen ME, Schweickert PP, et al. of vertebral bone augmentation. Neuroimaging Clin N
Relevance of antecedent venography in percutaneous Am 2010; 20(2): 15967.
vertebroplasty for the treatment of osteoporotic 43. Grados F, Depriester C, Cayrolle G, et al. Long-term
compression fractures. AJNR 2002; 23: 5946. observations of vertebral osteoporotic fractures treated
28. Gangi A, Wong LLS, Guth S, et al. Percutaneous by percutaneous vertebroplasty. Rheumatology 2000;
vertebroplasty: indications, techniques and results. 39(12): 141014.
Semin Intervent Radiol 2002; 19: 26570. 44. Baroud G, Nemes J, Heini P, et al. Load shit of the
29. Venmans A, Klazen CA, Lohle PN, et al. Percutaneous intervertebral disc ater a vertebroplasty: a inite-
vertebroplasty and pulmonary cement embolism: element study. Eur Spine J 2003; 12: 423.
results from VERTOS II. AJNR Am J Neuroradiol 2010 45. Luo J, Adams MA, Dolan P. Vertebroplasty and
Sep; 31(8): 14513. kyphoplasty can restore normal spine mechanics
30. Miller JC. Radiology Grand Rounds Massachusetts following osteoporotic vertebral fracture. J Osteoporos
General Hospital Department of Radiology 2009 July; 7(7). 2010. doi:10.4061/2010/729257. 151
Section 2: Spine surgery for adult patients

46. Ledlie JT, Renfro MB. Kyphoplasty treatment of vertebral compression fracture: a prospective,
vertebral fractures: 2 year outcomes show sustained comparative, and randomized clinical study. Osteoporos
beneits. Spine 2006; 31: 221320. Int 2010 Feb; 21(2): 35964.
47. Komp M, Ruetten S, Godolias G. Minimally-invasive 53. Gaitanis IN, Carandang G, Phillips FM, et al. Restoring
therapy for functionally unstable osteoporotic geometric and loading alignment of the thoracic spine
vertebral fractures by means of kyphoplasty: with a vertebral compression fracture: efects of balloon
prospective comparative study of 19 surgically and 17 (bone tamp) inlation and spinal extension. Spine J
conservatively treated patients. J Miner Stofwechs 2004; 2005; 5: 4554.
11(Suppl 1): 1315. 54. Crandall D, Slaughter D, Hankins PJ, Moore C, Jerman
48. Kasperk C, Hillmeier J, Noldge G, et al. Treatment of J. Acute vs. chronic vertebral compression fractures
painful vertebral fractures by kyphoplasty in patients with kyphoplasty: early results. Spine J 2004; 4(4):
with primary osteoporosis: a prospective nonrandomized 41824.
controlled study. J Bone Miner Res 2005; 20: 60412. 55. Hiwatashi A, Westesson A, Ypshiura T, et al. Kyphoplasty
49. Mathis JM, Ortiz O, Zoarski GH. Vertebroplasty and vertebroplasty produce the same degree of height
versus kyphoplasty: a comparison and contrast. Am J restoration. Am J Neuroradiol Apr 2009; 30: 66973.
Neuroradiol 2004; 25: 8405. 56. Buchbinder R, Osborne RH, Ebeling PR, et al. Randomized
50. Kumar K, Nguyen R, Bishop S. A comparative analysis trial of vertebroplasty for painful osteoporotic vertebral
of the results of vertebroplasty and kyphoplasty fractures. N Engl J Med 2009; 361: 55768.
in osteoporotic vertebral compression fractures. 57. Kallmes DF, Comstock BA, Heagerty PJ, et al. A
Neurosurgery 2010 Sep; 67(3 Suppl Operative): 17188; randomized trial of vertebroplasty for osteoporotic
discussion 188. spinal fractures. N Engl J Med 2009; 361: 56979.
51. Anselmetti GC, Muto M, Guglielmi G, Masala S. 58. Buchbinder R, Osborne R, Staples M. Correspondence
Percutaneous vertebroplasty or kyphoplasty. Radiol Reply. N Engl J Med 361; 21.
Clin North Am. 2010 May; 48(3): 6419. 59. Lotz JC. Trials of vertebroplasty for vertebral fractures.
52. Liu JT, Liao WJ, Tan WC, et al. Balloon kyphoplasty N Engl J Med. 2009; 361(21):2098; author reply
versus vertebroplasty for treatment of osteoporotic 2099100.

8.7 Endoscopic surgery for Chiari


malformation type I
Rodolfo Hakim and Xiao Di

recurrence continues to pose great challenges to sur-


Key points
geons, especially when associated with syringomye-
he symptoms and signs of Chiari I lia.1,2 Although the complete pathogenic mechanisms
malformation patients can be diverse. are yet to be elucidated, it is widely accepted that either
Patients with Chiari I tend to have a crowded an inborn or an acquired descent of the cerebellar ton-
posterior fossa. sils creates a craniospinal pressure dissociation and an
he current surgical techniques proposed for impaired cerebrospinal luid (CSF) low due to crowd-
the treatment of symptomatic Chiari I ing at the level of the craniovertebral junction (CVJ).3,4
malformation are very diverse among Consequently, the irst line of surgical therapy ofered
diferent authors. at many institutions is a craniovertebral decompres-
A purely endoscopic decompressive surgery is a sion, usually including suboccipital craniectomy with
novel technique in a minimally invasive modality. removal of the posterior arch of the atlas. Sometimes
a laminectomy of the axis with or without intradural
exploration, reduction of the tonsils, duraplasty, and
Introduction even a shunt implantation for syringomyelia may be
152 he management of Chiari malformation type I included as part of the surgical treatment. he speciic
(CM I) with less postoperative complications and surgical steps in this operation continue to undergo
Chapter 8.7: Chiari malformation type I

modiications as surgeons attempt to identify the opti- an impaired gag relex, vocal cord paralysis, decreased
mum procedure. In addition, postoperative complica- trapezius muscle strength, dysmetria, and truncal and
tions including pseudomeningocele, meningitis, CSF apendicular ataxia.
leak, intradural adhesions, cerebellar ptosis, and cer- With regard to the spinal cord, the symptoms
vical instability have brought about not only failures of occurring from anterior cord compression, posterior
the operation but also potentially severe consequences cord compression, or a cord syrinx are motor (atrophy,
for some of the patients. spasticity, and muscle weakness) or sensory (anesthe-
In order to simplify and make the surgical decom- sia, hypoalgesia, hyperesthesia, burning dysesthesia,
pression less invasive, a novel and entirely endoscopic decreased temperature sensation, and decreased posi-
procedure has been employed for the decompression tional sense). Also, urinary and/or fecal incontinence
of Chiari malformation type I. as well as impotence are described. On examination,
We describe here a new, minimally invasive tech- scoliosis, muscle weakness, muscle atrophy, decreased
nique for removing the subocciput as well as the pos- ine-motor ability, trophic phenomena, hyper- or
terior elements of the cervical vertebrae under direct hyporelexia, Babinski sign, and dissociated sensory
endoscopic visualization. he aim of this procedure loss may be found.
is to ameliorate as much as possible the clinical symp-
toms. his is done by enlarging the foramen magnum Pathophysiology
and vicinity and thereby facilitating the low of CSF at Although it is diicult to assess, patients with Chiari I
the CVJ. malformation seem to share a common denomina-
tor which is a tight or crowded posterior fossa. It has
Clinical manifestations of Chiari I been suggested that this is a disorder of the para-axial
he symptoms and the clinical presentation of Chiari I mesoderm in which the posterior fossa is underdevel-
malformation are diverse. Although headache is the oped and therefore its normally developed contents are
most common symptom, ocular, otoneurologic, brain- tight or overcrowded.10 With time, the result can be a
stem, lower cranial nerve, cerebellar, and spinal cord downward squeeze of the normally developed hind-
disorders may occur as well. he headache is more brain through the foramen magnum establishing as an
common in the suboccipital region; its description can anatomical cerebellar tonsillar herniation. he clinical
range from a light pressure to intense pounding that manifestations may be due to direct compression of the
can radiate upward to the vertex or all the way to the nearby nervous structures and probably from a caudal
retroorbital region as well as downward to the neck and cephalad restricted CSF low through the foramen
and shoulders. It usually intensiies when lowering the magnum. A genetic component with autosomal domi-
heads position and with Valsalva maneuvers. nant or recessive inheritance has been documented
he ocular symptoms that can occur in Chiari in those families where more than one member has a
I patients are phosphenes, photophobia, blurred Chiari I.
vision, diplopia, visual ield cuts, and visual blackouts. Interestingly, a secondary or acquired form of
Funduscopic examination may reveal bilateral papille- Chiari I is sometimes found in patients with pseudo-
dema as well as decreased or absent venous pulsations. tumor cerebri, posterior fossa lesions with mass efect,
Among the otoneurologic symptoms, one may ind hydrocephalus, central nervous system infections,
dizziness, vertigo, loss of equilibrium, oscillopsia, tin- multiple lumbar punctures, spinal CSF istulas, and/or
nitus, nausea and vomiting, and hypo- or hyperacusia. lumboperitoneal shunts.2729
On examination, rotary, lateral, and/or downbeat nys-
tagmus may be found.2,526 Surgical technique
he symptoms correlated with the brainstem, lower General anesthesia is induced in the patient using
cranial nerves, and the cerebellum that can be described endotracheal intubation. he neck should be kept in
by the patient are sleep apnea, syncope, shortness of neutral position during intubation and positioning
breath, dysphagia, dysarthria, facial pain or numbness, especially in those patients with a tight craniocervi-
throat pain, palpitations, hoarseness, incoordination, cal junction, brainstem or spinal cord compression,
tremor, and decreased ine motor function. On exam- or with a syrinx. Venous and arterial lines are placed.
ination, one may ind facial hypoesthesia, trigeminal It is very important to have a constant arterial pres- 153
and/or glossopharyngeal neuralgia, glossal atrophy, sure as well as heart rate monitoring throughout the
Section 2: Spine surgery for adult patients

head pinning and patient positioning as well as dur- and longitudinally to the bodys axis to secure good
ing surgery; inadvertent pressure, trauma, or ische- expansion of the chest and therefore facilitate adequate
mia over the brainstem and spinal cord can usually be ventilation. Emphasis on making sure that the shoul-
suspected by sudden bradycardia and hypotension. ders are being supported by the chest rolls is important;
Also, it must be kept in mind that besides the neural this provides more stability and therefore decreases
structures involved and exposed during the proced- unnecessary torque and pull on the neck. Also, it is not
ure, the vertebral arteries are near or within the sur- uncommon for patients to complain of shoulder, lower
gical ield and therefore can be inadvertently injured neck, and interscapular postoperative pain when the
as well. Electrophysiological monitoring such as som- unsupported shoulders have been pulling forward by
atosensory and/or motor evoked potentials are utilized gravity. In female patients, the breasts should be posi-
throughout the case, more so in those patients with tioned within the space between the chest rolls so that
any of the following: articular instability/ligamentous they are free from pressure points.
laxity, tight craniocervical junction, the presence of a Although a moderate neck lexion deinitely facili-
syrinx, progressive neurologic deicit, or any signs of tates surgical exposure in most cases, a slightly ante-
brainstem involvement. Broad-spectrum antibiotics lexed or even neutral head position provides us with
are given for 24 hours ater surgery; the irst dose is an approach to the CVJ, especially when the patients
given ideally at least 30 minutes (but not more than 60 condition does not allow the neck adequate lexion. As
minutes) prior to skin incision. Cefazolin is our irst a general rule a minimum distance of about two inger
choice antibiotic although, if the patient is allergic, we widths should exist between the jaw and the sternum
usually use vancomycin as an option. All patients will when lexing the head. One must keep in mind that
have a Foley catheter placed. Diuretics are not used overlexion poses risks for vascular, spinal cord, and
routinely. We try to stay away from the use of steroids brainstem injuries in an already tight craniocervical
to avoid their side efects. Since most of our patients are space, as well as decreased venous jugular blood low
healthy individuals except for their Chiari malforma- and compromise of the airway or endotra