1

Romford, UK
2
London, UK
*
E-mail: james.rees@bhrhospitals.nhs.uk
1 Houseof Commons Committeeof Public Accounts. TheNational Pro-
gramme for IT in the NHS: an update on the delivery of detailed care
recordsystems. Forty-fthReport of Session201012, London, 2011
2 NHS Connecting for Health. SCR (Summary Care Record) information
for NHS staff, 2013. Available from http://www.connectingforhealth
.nhs.uk/systemsandservices/scr (accessed 17 March 2013)
3 NHS Connecting for Health. History of PACS (Picture Archiving
and Communication System), 2013. Available from http://www.
connectingforhealth.nhs.uk/systemsandservices/pacs/learn (accessed
17 March 2013)
doi:10.1093/bja/aet315
General anaesthesia with laryngeal mask
airway may cause recurrence of
pneumocephalus in a patient with
head injury
EditorA 50-yr-old male patient presented after a road trafc
accident with transient loss of consciousness and bleeding
from the right ear. On examination, the patient was drowsy
with a Glasgow coma scale of 12 and bilateral forearm
fractures. X-raysof theskull, face, andbraincomputerizedtom-
ography (CT) revealed fracture of the skull base and bilateral
pneumocephalus in the frontal region (Fig. 1A). The X-rays
of the forearm revealed fracture of both bones bilaterally.
Pneumocephalus subsided spontaneously with conservative
management as conrmed by repeat CT scan after 10 days
(Fig. 1B). On17thdayafter trauma, the patient was undergoing
operation of the fractures of both bones of the right forearm
under supraclavicular brachial plexus block (SCB). Glycopyrro-
late i.v. 0.2 mg and fentanyl i.v. 100 mg were administered as
premedication. Standard monitoring was applied. As the
patient was uncooperative during the block, anaesthesia
was inducedwithi.v. propofol 150mg, 100%oxygen, andsevo-
urane end-tidal (ET) 3%. Alaryngeal mask airway (LMA) size 5
was inserted and bilateral air entry checked with 23 positive
pressure breaths. The patient was maintainedonspontaneous
breathing with 50%:50% air and oxygen with sevourane ET
2%. SCB was given with 15 ml of 2% lidocaine, 15 ml of 0.5%
bupivacaine, and 50 mg of fentanyl after nerve location with
a 21 G, 10 cm stimuplex needle with 0.7 mA current by the
Kulenkampff technique. The procedure lasted for 50 min and
the tourniquet time was 60 min. The respiratory rate was
around 20 bpm and E

CO
2
was around 3.54.0 kPa throughout
the procedure. At the end of the surgery, LMA was removed
smoothly with no forceful respiratory movements after the
patient was awake and responding.
Tenhours later, thenursingstaff reportedprogressiveabdom-
inal distension with tachypnoea and declining consciousness.
A B C
Fig 1 (A) X-ray skull and CT scan performed immediately after head trauma shows bilateral frontal pneumocephalus. (B) X-ray skull and CT scan
performed 10 days later shows resolution of pneumocephalus. (C) X-ray skull and CT scan performed immediately after general anesthesia
shows recurrent pneumocephalus in the right frontal region.
Correspondence BJA
675

b
y

g
u
e
s
t

o
n

S
e
p
t
e
m
b
e
r

1
1
,

2
0
1
4
h
t
t
p
:
/
/
b
j
a
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

On examination, the bowel sounds were absent. Ultrasound
examination was normal. An arterial blood gas showed respira-
tory alkalosis with pH 7.51, PCO
2
3.2 kPa, PO
2
20 kPa, and HCO
3
20.4. A repeat brain CT with bone window showed right frontal
pneumocephalus and a large defect near the cribriform plate
on the right side (Fig. 1C).The pneumocephalus was managed
with supportive care and the patient was discharged on post-
trauma day 27.
Recurrent pneumocephalus has beenreportedwithpositive
pressure ventilation strategies like mask ventilation during an-
aesthetic induction, continuous positive airway pressure, and
also with simple nasal O
2
cannulae.
13
It has been advised
that inpatients withthe base of skull and facial bone fractures,
excessive positive pressure during mask ventilation should not
be used.
4
Rapid induction and intubation with intubating LMA
in a patient withdifcult airway andpneumocephalus has been
described.
5
Toavoidtheabovecomplication, wehadplannedfor
the surgery of the forearm under regional anaesthesia. But as
thepatient was uncooperative, LMAwasusedwiththepreserva-
tionof spontaneousventilation. Unfortunately, positivepressure
ventilation (PPV) with LMA to check its correct position after in-
sertion seems to have reopened the breach in the duramater.
The respiratory embarrassment caused by paralytic ileus may
alsohave aggravatedthe pneumocephalus. Theresultinghypo-
capnia may have decreased the intracranial pressure, thereby
creating subatmospheric pressure and causing indrawing of
air into the cranium. Therefore, causes of hyperventilation
like respiratory embarrassment, poor analgesia, and anxiety
should be treated promptly.
In conclusion, we would like to caution that pneumocepha-
lus can recur during PPV with LMA in patients with recent
fracture of the skull base and face. The technique of rapid in-
duction and intubation with fast-acting neuromuscular block-
ing agents without positive pressure mask ventilation (PPMV)
and denitive airway control with the tracheal tube is probably
the best way to avoid recurrent pneumocephalus. Any cause of
hyperventilation in the postoperative period which may aggra-
vate pneumocephalus should be corrected.
Declaration of interest
None declared.
I. Gurajala*
M. Azharuddin
R. Gopinath
Hyderabad, India
*
E-mail: indiradevraj@yahoo.co.in
1 Dacosta A, Billard JL, Gery P, Vermesch R, Bertrand M, Bertrand JC.
Posttraumatic intracerebral pneumatocele after ventilation with a
mask: case report. J Trauma 1994; 36: 2557
2 Jarjour NN, Wilson P. Pneumocephalus associated with nasal posi-
tive airway pressure in a patient with sleep apnea syndrome. Chest
1989; 96: 14256
3 OBrienBJ, RosenfeldJV, Elder JE. Tensionpneumo-orbitusandpneu-
mocephalus induced by a nasal oxygen cannula: report on two
paediatric cases. J Paediatr Child Health 2000; 36: 5114
4 Moon HS, Lee SK, Chung SH, Chung JH, Chang IB. Recurred pneumo-
cephalus in a head trauma patient following positive pressure mask
ventilation during induction of anesthesiaa case report. Korean J
Anesthesiol 2010; 59(Suppl.): S1836
5 Noguchi T, ShigaY, KogaK. Theintubatinglaryngeal maskairwayina
patient with cerebrospinal uid rhinorrhea and pneumocephalus.
Masui 2003; 52: 1679
doi:10.1093/bja/aet316
Assessment of anaesthetists ability to
predict difculty of bag-mask ventilation
EditorBag-mask ventilation (BMV) is a vital, life-saving skill
for anaesthetists. The importance of BMV is recognized by
several Difcult Airway Societies worldwide, and, reected in
difcult airwayalgorithms.
1 2
Incurrent literature, the incidence
of difcult BMVhas rangedbetween0.08%and15%
3 4
withfew
studies large enough to report the incidence of impossible BMV,
althoughKheterpal andcolleagues
5
havereportedanincidence
of 0.16% in 2006. However, currently, there is no objective pre-
dictivescorefor difcult BMVroutinelyused. Instead, theindivid-
ual anaesthetists subjective preoperative assessment is heavily
relied upon.
We present a pilot prospective study we conducted at a
350-bed Australian teaching hospital aimed at determining
the accuracy of this assessment by the anaesthetist. During
the routine preoperative assessment, anaesthetists were asked
to predict the difculty of BMV for each patient on a predeter-
mined scale from 1 to 5 (modication of Hans Mask Ventilation
Classication and Description Scale) (Table 1). Intraoperatively,
this same scale was used to record the observed difculty of
BMV. The preoperative and intraoperative scores were then
matched to assess the accuracy of anaesthetists predictions.
All scores 3 were considered difcult and scores of ,3 were
considered easy.
We collected data on 231 episodes in 231 patients. Difcult
BMVwas substantiallyunderestimatedwith16(6.9%) predicted
difcult BMV vs 36 (15.6%) actually difcult BMV (P,0.001). Of
those 215 patients predicted to be easy BMV, 26 (12.3%) were
foundtobedifcult. Similarly, six (37.5%) of the 16 patients pre-
dicted to be difcult were actually easy. Anaesthetists predic-
tion of difcult BMV had a sensitivity of 27.8% and a specicity
of 96.9%.
The results of this pilot study highlight the need for the de-
velopment of objective predictors of difcult BMV that may
be used routinely in the preoperative assessment. The use of
objective data rather than the inaccurate current method
couldfeasibly leadtobetter planningandthereforebetter exe-
cution of a patients ventilation intraoperatively. Further study
to determine the objective predictors of difcult BMVwould be
useful. If these predictors were presented in a clinically applic-
able format, similar to the numerous predictive scores for dif-
cult intubation, thismayenhancethepreoperativeassessment
and improve the accuracy of BMV prediction, ultimately im-
proving the safety of anaesthesia for the patient.
BJA Correspondence
676

b
y

g
u
e
s
t

o
n

S
e
p
t
e
m
b
e
r

1
1
,

2
0
1
4
h
t
t
p
:
/
/
b
j
a
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m