Académique Documents
Professionnel Documents
Culture Documents
Implications of the new sepsis definition on research and practice
Brian C. Peach
PII: S0883-9441(16)30878-4
DOI: doi:10.1016/j.jcrc.2016.11.032
Reference: YJCRC 52348
Please cite this article as: Peach Brian C., Implications of the new sepsis denition on
research and practice, Journal of Critical Care (2016), doi:10.1016/j.jcrc.2016.11.032
This is a PDF le of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its nal form. Please note that during the production process
errors may be discovered which could aect the content, and all legal disclaimers that
apply to the journal pertain.
ACCEPTED MANUSCRIPT
PT
RI
Implications of the New Sepsis Definition on Research and Practice
SC
Brian C. Peach 1
University of Florida
NU
1
1225 Center DR
PO Box 100197
Gainesville, FL 32610-0197
MA
Corresponding author: Brian C. Peach, bpeach01@ufl.edu
D
P TE
CE
AC
ACCEPTED MANUSCRIPT
Abstract
Introduction: The Society of Critical-Care Medicine and the European Society of Intensive Care Medicine
recently announced a marked change in the sepsis definition. A task force of 19 sepsis clinicians and
researchers made the change based on advances in the understanding of the septic process.
PT
Summary of Change: The task force determined there were numerous justifications for a revision of the
sepsis definition. Reasons for making a change included that there was an overwhelming focus on
inflammation, the sepsis continuum model was misleading, the systemic inflammatory response
RI
syndrome (SIRS) criteria lacked adequate sensitivity and specificity, the multiple definitions for sepsis,
septic shock, and organ dysfunction led to discrepancies in reported incidence and mortality, and they
SC
felt the term severe sepsis was redundant (Singer et al., 2016)1. The systemic inflammatory response
criteria have been replaced by the sequential organ failure assessment (SOFA) score and the qSOFA
score in the newly operationalized definition (Singer et al., 2016)1.
NU
Implications on Research & Practice: There are multiple implications on research and practice, including
potential fiscal effects on hospitals and universities as they adopt the new definition. The change in
MA
definition may have a negative impact on sepsis research in the short-term, but may limit systematic
bias and improve the quality of sepsis research in the long-term.
Conclusions: While the new operationalized Sepsis-3 definition appears on the surface to be an
improvement over the previous iterations, it remains to be seen if clinical outcomes will improve and if
D
research studies will be more robust using the new criteria. Future research on the impact of this change
TE
on research and practice will be essential, to determine if the definition and associated clinical criteria
need further revision.
P
Key Words: sepsis, severe sepsis, septic shock, systemic inflammatory response syndrome (SIRS) criteria,
CE
Sequential [Sepsis-related] Organ Failure Assessment (SOFA) score, Logistic Organ Dysfunction System
(LODS) criteria
AC
Introduction
The Society of Critical-Care Medicine (SCCM) and the European Society of Intensive Care
Medicine (ESICM) recently announced a marked change in the sepsis definition (Singer et al., 2016)1. The
original definition (Sepsis-1) created in 1991 (Bone et al., 1992)2 was revised in 2001 (Sepsis-2) with
the addition of clinical criteria for inadequate tissue perfusion (Levy et al., 2003)3. Since the 2001
revision, there have been numerous advances in understanding the pathology, underlying biological
mechanisms, management, and epidemiology of sepsis, which prompted a review of the Sepsis- 2
definition (Singer et al., 2016)1. A task force of 19 sepsis clinicians and researchers convened in 2014 to
review the current evidence, and to revise the definition through meetings, Delphi processes, analysis of
ACCEPTED MANUSCRIPT
electronic health record databases, and subsequent voting on criteria (Singer et al., 2016)1. After
meeting, the task force circulated their Sepsis-3 definition to 31 international professional societies,
and requested peer review and endorsement of the new definition (Singer et al., 2016)1. The new
PT
definition was announced in a February 23rd, 2016 Journal of the American Medical Association article
(Singer et al., 2016)1, and the implications of this significant change of research and practice will be
RI
discussed later in this article.
SC
Summary of Change
NU
Leading up to the SCCM/ESICM task force creation, several of its members advocated for a
MA
change to the Sepsis-2 definition based on their own teams research findings (Deutschman & Tracey,
2014; Vincent, Opal, Marshall, & Tracey, 2013) 4, 5. The task force determined there were numerous
D
justifications for a revision of the sepsis definition. Reasons for making a change included that there was
TE
an overwhelming focus on inflammation, the sepsis continuum model was misleading, the systemic
inflammatory response syndrome (SIRS) criteria lacked adequate sensitivity and specificity, the multiple
P
CE
definitions for sepsis, septic shock, and organ dysfunction led to discrepancies in reported incidence and
mortality, and they felt the term severe sepsis was redundant (Singer et al., 2016, p.801)1.
AC
Sepsis according to the Sepsis-1 definition was not a disease, but an unbridled cytokine
inflammatory response to an infectious trigger (Bone et al., 1992)2. At the core of the 1991 and 2001
sepsis, severe sepsis, and septic shock definitions is the SIRS criteria. The SIRS criteria included four
clinical signs which when altered were thought to represent an inflammatory response: temperature,
heart rate, respiratory rate, and white blood cell count (Bone et al., 1992)2. In order to be diagnosed
with sepsis, an individual must have at least 2 SIRS criteria and a confirmed or suspected infection (Bone
et al., 1992)2. More recent research however, has shown that sepsis involves activation of both pro-
inflammatory and anti-inflammatory responses (Hotchkiss, Monneret, & Payen, 2013)6. In addition,
ACCEPTED MANUSCRIPT
there are also other endogenous factors that rapidly change the cardiovascular, neurological, endocrine,
and hematological systems in response to infection and the inflammatory response (Deutschman et al.,
2014; Singer, De Santis, Vitale, & Jeffcoate, 2004)4,7. Other factors like the source of infections,
PT
comorbidities, and iatrogenic interventions also have an impact on body response (Deutschman et al.,
2014) 4.
RI
SC
Under the Sepsis-1 and Sepsis-2 definitions, sepsis, severe sepsis, and septic shock were
seen as steps in a continuum. As a patients condition declined, they advanced from one stage to the
NU
next with unresolved septic shock (the final stage) resulting in death. Many patients however lacked
sufficient SIRS criteria under the previous definitions to receive a sepsis diagnosis, but had complicated
MA
courses similar to those with diagnoses resulting in mortality. In a study of 109,663 patients admitted
between 2000 and 2013 to 172 Australian or New Zealand intensive care units with infection and organ
D
failure, 13,278 patients (12.1%) had SIRS-negative sepsis (Kaukonen, Bailey, Pilcher, Cooper, & Bellomo,
TE
2015)8. These results raise the question of whether the SIRS criteria possess adequate construct validity.
P
CE
While individuals with at least 2 aberrations in the SIRS criteria are often septic, many are not.
Other conditions that could be mistaken for sepsis using these criteria are primary hypothermia,
AC
endocrine disorders (e.g. hyperthyroidism / hypothyroidism), autoimmune disorders (e.g. lupus, cystic
fibrosis), cardiac and pulmonary disorders (e.g. CHF, COPD), and uncomplicated infections without a
prospective cohort study of 3,147 patients in 198 ICUs in 24 countries, 87% of patients on admission had
2 SIRS criteria and 93% had two during the admission, with out-of-normal-range respiratory and heart
rates seen in over 70% of patients (Sprung et al., 2006)9. There is a clear direct association between the
number of SIRS criteria and risk for organ failure and mortality (Sprung et al., 2006)9, but the presence of
2 criteria as part of a sepsis diagnosis appears to be overly sensitive and may lead to inflated sepsis
incidence data.
ACCEPTED MANUSCRIPT
Another prospective cohort study of 2085 patients admitted to an Australian tertiary hospital
found that patients with >/=2 SIRS criteria had a sensitivity of 70.6% and a specificity of 37.5%, with a
positive predictive value of 63.7% and a negative predictive value of 45.1% in predicting
PT
microbiologically confirmed infections (Lai & Kruger, 2011)10. Even when weighting the SIRS criteria in
favor of the temperature and white blood cell count scores, patients with three or greater SIRS criteria
RI
had a marginally better positive predictive value of 64.5% and a negative predictive value of 44.6% (Lai
SC
et al., 2011)10. The evidence suggests that using unweighted or weighted SIRS criteria can lead to a high
NU
number of false positives and false negatives.
Another issue addressed in the Singer et al. (2016) paper1, is the variability in use of the terms
MA
sepsis, severe sepsis, and septic shock. At present, there is no diagnostic test for sepsis, nor one that
differentiates the different degrees of illness. There is also no current process to operationalize the
D
definitions, which according to Singer et al. (2016) leads to significant variations in the reported
TE
incidence and mortality rates 1. With the new definition and operationalized clinical criteria, the task
P
force aimed to increase specificity using criteria that identified infection, host response, and organ
CE
infection (Singer et al., 2016)1. Improved discriminative abilities, could lead to a better epidemiological
AC
understanding of the sepsis disorder. The task force eliminated the term severe sepsis, which they
labeled redundant, as nearly all patients with a sepsis diagnosis end up in an intensive care unit
(Singer et al., 2016, p.801) 1. Under the Sepsis-3 definition, there are two diagnostic categories, sepsis or
Recognizing that there are other body processes at play during sepsis besides the exaggerated
inflammatory response, the SCCM and ESICM task force decided to define sepsis as Life-threatening
organ dysfunction caused by a dysregulated host response to infection. with the following clinical
[Sepsis-related] Organ Failure Assessment (SOFA) score of 2 points or more. (Singer et al., 2016, p. 801)
ACCEPTED MANUSCRIPT
1
. Septic shock was defined as a subset of sepsis in which particularly profound circulatory, cellular, and
metabolic abnormalities are associated with a greater risk of mortality than with sepsis alone (Singer et
al., 2016, p. 801)1. Septic shock was operationalized by a vasopressor requirement to maintain a mean
PT
arterial pressure of 65mmHg or greater and serum lactate level greater than 2mmol/L (>18mg/dL) in the
RI
SC
The SOFA scoring tool is not new, but is not a standard tool in many intensive care units. It is a
screening tool for organ dysfunction, with an increase in score of 2 points from baseline being
NU
associated with >10% increased risk of mortality (Singer et al., 2016)1. When accounting for a patients
baseline risk level, a SOFA score of 2 points or greater is associated with a two to twenty-five fold
MA
increased risk of mortality when compared to patients with a SOFA score less than 2 (Seymour et al.,
2016)11 . Variables needed to compute a SOFA score include patients pulmonary-function ratio
D
(PaO2/FiO2 mmHg), platelet counts (x103/), bilirubin (mg/dL), MAP score or vasopressor requirement,
TE
GCS score, creatinine (mg/dL), and urine output (ml/day) (Singer et al., 2016)1. Each variable is graded
P
on a 0-4 scale with predetermined ranges or medication dosages for each level (Singer et al., 2016)1.
CE
In addition to the SOFA score, Singer et al. (2016) encouraged usage of the qSOFA, a newly
AC
created simplified SOFA score that has been validated for use in non-ICU settings (e.g. in ambulances,
emergency rooms, and medical-surgical floors) (Singer et al., 2016)1. The qSOFA includes only 3
variables: systolic blood pressure (100 mm Hg or <), respirations (22/min or >), and a Glasgow Coma
Scale score (13 or <), but any combination of two variables demonstrated predictive validity similar to
that of the SOFA score when used in non-ICU settings (qSOFA AUROC= 0.81; 95% CI, 0.80-0.82 v. SOFA
in non-ICU settings: AUROC= 0.79, 95% CI, 0.78-0.80) (Singer et al., 2016, p.805)1.
emergency departments, intensive care units, wards, step-down units, or post-anesthesia care units,
ACCEPTED MANUSCRIPT
derived from 12 hospitals between 2010 and 2012, evaluated the validity of the SOFA, qSOFA, SIRS, and
Logistic Organ Dysfunction System (LODS) criteria (Seymour et al., 2016)11. In ICU encounters with
suspected infection, there was no statistically significant difference between the SOFA and the more
PT
variable-heavy LODS score in terms of predictive validity (Seymour et al., 2016)11. Both tools however
demonstrated stronger predictive validity when compared to the SIRS criteria and qSOFA score
RI
(Seymour et al., 2016) 11. In encounters with suspected infection in non-ICU settings, the predictive
SC
validity of qSOFA was superior to the SOFA and SIRS criteria (Seymour et al., 2016) 11.
NU
Implications on Research and Practice
MA
As with changes in healthcare policy, changes in definitions and clinical criteria carry financial
implications for schools, hospitals, and other healthcare institutions. For over two decades, medical,
D
nursing, allied health programs, and healthcare institutions have trained clinicians to recognize sepsis
TE
using the previous definition based on the SIRS criteria. A change in the definition will necessitate higher
education institutions to change their critical care content and textbooks, and healthcare facilities to
P
CE
spend the money to re-train their clinicians with no guarantee of better outcomes. The SOFA and qSOFA
scores like screening tools based on the SIRS criteria will be sensitive to iatrogenic interventions such as
AC
medications (e.g. sedation may affect blood pressure, heart rate, respirations, and the Glasgow Coma
Scale score; blood pressure medications can affect the blood pressure; and mechanical ventilation may
affect the respiratory rate). Given that there is no guarantee of better outcomes by using the SOFA and
qSOFA scores, as opposed to other adopted early warning tools (e.g. the Modified Early Warning System
There are however potential financial benefits for hospitals that invest in re-training employees
earlier. Early identification and treatment of sepsis with antibiotics and appropriate fluid resuscitation
has been shown to improve mortality, and is the hallmark of the 2012 Surviving Sepsis Campaign
ACCEPTED MANUSCRIPT
guidelines (Dellinger et al., 2013)12. For every hour that antibiotics are delayed, mortality goes up by
7.6% (Kumar et al., 2006) 13. The use of the sequential organ failure assessment (SOFA) and qSOFA
scores, may result in earlier identification of septic patients, less resource utilization, and better
PT
outcomes. Improved outcomes spells better reimbursement amounts for hospitals from the Center for
Medicare and Medicaid Services (CMS) and insurance providers. In order to compute a SOFA score
RI
however, additional labs (i.e. PaO2, platelet count, creatinine level, and bilirubin level) must be sent,
SC
meaning that a new operationalized definition will bring with it associated costs for hospitals, insurance
NU
companies, and patients that may negate improvements in reimbursement. The benefits and
disadvantages of using the SOFA and qSOFA need to be investigated, and future research will be
MA
necessary to study the impact of this change on hospital budgets and reimbursement income.
sepsis research. Under the previous definition, there were considerable issues with miscoding related to
TE
clinicians misclassification of the level of sepsis. For example, providers would often misclassify a
P
patient as having sepsis when they had severe sepsis, or severe sepsis when they met criteria for
CE
septic shock. By eliminating the severe sepsis category and operationalizing the sepsis and septic
AC
shock definitions, there will likely be less misclassification bias related to provider error, which will
result in more robust epidemiological research studies. There may however continue to be
misclassification issues as inadequately resuscitated patients are started on vasopressors, and coded
according to definition as having septic shock. How this issue will be addressed is yet to be
determined, but should be listed as a possible limitation in published manuscripts pertaining to sepsis.
A major limitation of this change in the short-term is the scarcity of available data for analysis
under the new definition. Routine monitoring of serum bilirubin levels and calculated pulmonary-
function ratios (pf-ratios), 2 data points required to compute a SOFA score, is not currently standard
practice in many intensive care units. Until collecting this data becomes standard, researchers will find it
ACCEPTED MANUSCRIPT
difficult to study sepsis using the new definition. The utility of studies in progress that collect data
A benefit of using the SOFA score in the future will be the additional data available for
PT
researchers to mine. The SOFA score factors in lab values that were not requisite to determine patients
RI
sepsis stage. With this additional data, researchers may be able to perform secondary data analysis
SC
studies that were previously impossible because of insufficient data. In addition, the new CMS sepsis
bundle that must be adopted by hospitals for reimbursement will ensure that more detailed records are
NU
kept in clinical settings.
MA
Conclusion
While the new Sepsis-3 definition appears on the surface to be an improvement over the
D
previous iterations, it remains to be seen if clinical outcomes will improve and if research studies will be
TE
more robust using the new criteria. The Seymour et al. (2016) study demonstrated that the SOFA score
P
is more discriminating that the SIRS criteria in ICU settings, and is less expensive than the LODS score,
CE
which requires additional lab work to be collected for calculation. Additionally the qSOFA score
demonstrated superior predictive validity when compared to the more encompassing SOFA score and
AC
Hospitals and institutions of higher learning will need time to re-train their students to identify
sepsis and septic shock using SOFA and qSOFA. With that said, it may be some time before hospitals
transition away from SIRS, as budget constraints will make it challenging to change their monitoring
tools and re-train their staff. Failure to adopt these tools in a timely manner however may have
important financial implications, and more importantly result in poorer outcomes. Future research on
the impact of this change on research and practice will be essential, to determine if the definition and
Funding source: This research did not receive any specific grant from funding agencies in the public,
PT
Acknowledgements: I wish to thank Dr. Jeannie Cimiotti, PhD, RN, FAAN for her assistance in
RI
proofreading this manuscript.
SC
NU
MA
D
P TE
CE
AC
ACCEPTED MANUSCRIPT
References
[1] Singer, M., Deutschman, C.S., Seymour, C.W., Shanker-Hari, M., Annane, D., Bauer, M., Angus, D.C.
The third international consensus definitions for sepsis and septic shock (Sepsis-3). JAMA 2016; 315(8),
801-10. DOI: 10.1001/jama.2016.0287
PT
[2] Bone, R.C., Balk, R.A., Cerra, F.B., Dellinger, R.P., Fein, A.M., Knaus, W.A., Sibbald, W.J. Definitions
for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. The ACCP/SCCM
Consensus Conference Committee. American College of Chest Physicians/Society of Critical Care
RI
Medicine. Chest 1992; 101(6), 1644-55. PMID: 1303622
[3] Levy, M.M., Fink, M.P., Marshall, J.C., Abraham, E., Angus, D., Cook, D., Ramsay, G. 2001
SC
SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Intensive Care Med 2003;
29(4), 530-8. DOI: 10.1007/s00134-003-1662-x
NU
[4] Deutschman, C.S. & Tracey, K.J. Sepsis: current dogma and new perspectives. Immunity 2014; 40,
463-475. DOI: 10.1016/j.immuni.2014.04.001
[5] Vincent, J.L., Opal, S.M., Marshall, J.C., & Tracey, K.J. Sepsis definitions: time for a change. Lancet
2013; 381(9868), 774-5. DOI: 10.1016/S0140-6736(12)61815-7
MA
[6] Hotchkiss, R.S., Monneret, G., & Payen, D. Sepsis-induced immunosuppression: from cellular
dysfunctions to immunotherapy. Nat Rev Immunol 2013; 13(12), 862-874. DOI: 10.1038/nri3552
D
[7] Singer, M., De Santis, V., Vitale, D., & Jeffcoate, W. Multiorgan failure is an adaptive, endocrine-
mediated, metabolic response to overwhelming systemic inflammation, Lancet 2004; 364(9433), 545-8.
TE
DOI: 10.1016/S0140-6736(04)16815-3
[8] Kaukonen, K-M., Bailey,M., Pilcher, D., Cooper, D.J., & Bellomo,R. Systemic Inflammatory Response
P
Syndrome criteria in defining severe sepsis. N Engl J Med 2015; 372(17), 1629-1638. DOI:
10.1056/NEJMoa1415236
CE
[9] Sprung, C.L., Sakr, Y., Vincent, J-L., Le Gall, J-R., Reinhart, K., Raneiri, V.M.Payen, D. An evaluation of
systemic inflammatory response syndrome signs in the Sepsis Occurrence in Acutely ill Patients (SOAP)
AC
[10] Lai, N.A. & Kruger, P. The predictive ability of a weighted systemic inflammatory response syndrome
score for microbiologically confirmed infection in hospitalised patients with suspected sepsis. Crit Care
Resusc 2011; 13(3), 146-150. PMID: 21880000
[11] Seymour, C.W., Liu, V.X., Iwashyna, T.J., Brunkhorst, F.M., Rea, T.D., Scherag, A., Angus, D.C.
Assessment of clinical criteria for sepsis. For the third international consensus definitions for sepsis and
septic shock (Sepsis-3). JAMA 2016; 315(8), 762-774. DOI: 10.1001/jama.2016.0288
[12] Dellinger, R.P., Levy, M.M., Rhodes, A., Annane, D., Gerlach, H., Opal, S.M., and the Surviving
Sepsis Campaign Guidelines Committee including the Pediatric Subgroup. Surviving Sepsis Campaign:
International guidelines for management of severe sepsis and septic shock: 2012. Crit Care Med 2013;
41(2), 580-637. DOI: doi: 10.1007/s00134-012-2769-8
[13] Kumar, A., Roberts, D., Wood, K.E., Light, B., Parrillo, J.E., Sharma, S., Suppes, R., Feinstein, D.,
Zanotti, S., Taiberg, L., Gurka, D., Kumar, A., & Cheang, M. Duration of hypotension before initiation of
ACCEPTED MANUSCRIPT
effective antimicrobial therapy is the critical determinant of survival in human septic shock. Crit Care
Med 2006; 34(6), 1589-1596. DOI: 10.1097/01.CCM.0000217961.75225.E9
PT
RI
SC
NU
MA
D
P TE
CE
AC