Introduction
Sepsis is a significant cause of morbidity and mortality with an estimated 44,000 patient deaths per year in the UK(1). In a heterogeneous condition like sepsis, affecting multiple organ systems and different age groups, a single definition has proven operationally challenging(1). Following the Third International Consensus meeting on Sepsis and the publication of new guidelines in February 2016, an updated definition of Sepsis was suggested(2). Sepsis is now defined as “life-threatening organ failure due to dysregulated host response to infection”(2). It was clarified that organ dysfunction could be identified by a change in Sequential Organ Failure Assessment (SOFA) score of greater than or equal to 2(3). The SOFA score, however, depended on multiple observations including clinical interventions and laboratory investigations for computation of the score, thus making it time consuming. Using a multivariate regression model the authors suggested a new diagnostic tool called the Quick Sequential Organ Failure Assessment or qSOFA score(4). This score incorporates 3 clinical criteria- Hypotension, Tachypnoea and Altered mental status, and advises that a change in 2 or more of these in a patient with infection would suggest poor outcomes(4).
The qSOFA score is easy to calculate as it is based only on clinical observations and it is reproducible. The authors did not test for inter-user reliability as all the criteria already have known reliability(4).
Case history
A 44 year old lady is brought into Emergency Department (ED) with symptoms of increasing shortness of breath. She complains of having been feeling unwell with fever and coryzal symptoms for the past few days. She was seen immediately upon arrival in the resuscitation department. On initial assessment, she appeared unwell. She was tachypnoeic with a rate of 20/min, had a low blood pressure of 110/50 mm Hg and had a qSOFA score of ‘1’.
When reviewing this patient in the ED, I wondered if there was a predictive tool that I could have used to ensure she received early intensive care input that may have changed her outcome. Currently using the NEWS to make this judgement takes some computation, as observations have to be entered in the Electronic Record to have this score. In such an unwell patient, a bedside scoring system may have enabled me to escalate her care earlier.
In this case, the patient’s qSOFA score was only 1 at presentation. However she was noted to have a National Early Warning Score (NEWS) of 10 and a SIRS of 2.
She subsequently was taken to ITU but did not recover and died less than 48hrs later.
Three part question :
This review was conducted to try and answer the following questions:
In [adult patients presenting to the ED with infection] can [the qSOFA score(4)] be used to [predict the severity of sepsis] and thus escalation of care?
Literature search
Search of Medline through the NHS National Library for Health
# Medline Search terms Results
1 (qsofa).ti,ab 91
2 (“quick sequential organ failure assessment”).ti,ab 33
3 (“quick sepsis related organ failure assessment”).ti,ab 13
4 (1 OR 2 OR 3) 101
5 *INFECTION/ OR *”SYSTEMIC INFLAMMATORY RESPONSE SYNDROME”/ OR *SEPSIS/ 59,263
6 (4 AND 5) 30
7 exp “EMERGENCY MEDICAL SERVICES”/ OR exp “EMERGENCY SERVICE, HOSPITAL”/ OR exp “EMERGENCY RESPONDERS”/ 127,421
8 (“A&E” OR “accident and emergency”).ti,ab 11,493
9 (7 OR 8) 136,337
10 (6 AND 9) 13
Inclusion criteria
-Studies looking at the use of the qSOFA score for prognosis of sepsis in the Emergency department
Exclusion criteria
– Studies combining qSOFA or other scoring systems with additional laboratory investigations such as lactate or bicarbonate.
– Studies that reviewed patients with specific conditions such as pancreatitis, pneumonia etc.
– Studies that looked at prehospital scores were excluded
– Studies that used qSOFA for diagnosis of sepsis
– Epidemiological studies
The results of these searches were crosschecked with EMBASE, Pubmed, CINAHL and HMIC. A total of 27 papers were found using the above search criteria. 13 using MEDLINE, and 14 using EMBASE. 3 papers were excluded due to duplication. The consort diagram in appendix 1 outlines the search. The titles and abstracts of the remaining 24 papers were reviewed and using the above inclusion and exclusion criteria 9 papers were selected for further appraisal.
Further searches using Google scholar and Best BETS were conducted. No additional studies were found. A search of the Cochrane database was also performed, which found no additional papers.
# EMBASE Search terms Results
1 (qsofa).ti,ab 203
2 (“quick sequential organ failure assessment”).ti,ab 57
3 (“quick sepsis related organ failure assessment”).ti,ab 32
4 (1 OR 2 OR 3) 213
5 *INFECTION/ OR *”SYSTEMIC INFLAMMATORY RESPONSE SYNDROME”/ OR *SEPSIS/ 122,134
6 (4 AND 5) 126
7 exp “EMERGENCY MEDICAL SERVICES”/ OR exp “EMERGENCY SERVICE, HOSPITAL”/ OR exp “EMERGENCY RESPONDERS”/ 91,354
8 (“A&E” OR “accident and emergency”).ti,ab 37,634
9 (7 OR 8) 127,298
10 (6 AND 9) 14
Appraisal of literature
Table 1: Study characteristics
Author, date and Journal
Patient number
and study group
Intervention and comparator
Study design
Outcomes
Primary (1o)
Secondary (2o)
Freund. Y et al January 2017, Journal of American Medical Assoc.(5)
879 adult patients
qSOFA versus SIRS
Prospective observational study
1o – in hospital mortality
Finkelstein E.J
Mar 2017-07-06 Critical Care(6)
152 adult patients admitted to ITU
qSOFA versus
SIRS
Low versus high qSOFA score
Retrospective observational analysis
1o – in hospital mortality
2o – ITU and organ-dysfunction free days
Williams J.M et al
Mar 2017
Chest(7)
4172 patients with infection
qSOFA versus SIRS
Retrospective observational analysis
1o – Mortality
and organ dysfunction
Churpek M.M et al
April 2017
American Journal or Respiratory and Critical Care Medicine(8)
30,677 pts
qSOFA versus SIRS versus MEWS versus NEWS
Retrospective observational analysis
1o – Mortality
2o – ITU admission
April M.D et al
May 2017-07-06
Journal of Emergency Medicine(9)
214 adult patient admitted to ITU
qSOFA versus SIRS
Retrospective observational analysis
1o – Mortality
Hwang et al(10)
June 2017
Annals of Emergency Medicine
1395 adult patients
Positive versus negative qSOFA
Retrospective analysis
1o – 28 day mortality
Park. Hyung et al(11)
June 2017
Journal of Critical Care
1009 patients with suspected infection
qSOFA versus SIRS
Retrospective analysis
1o – organ failure
Henning et al(12)
October 2017 (awaiting publication) Annals of Emergency Medicine
7637 adult patients
Sepsis 3 criteria versus Sepsis 1 criteria
Retrospective analysis of prospective observational cohort
1o – in hospital all cause mortality
Singer, A et al(13)
April 2017
Annals of Emergency Medicine
22,530 patients with suspected infection
qSOFA versus MEWS
Retrospective study
1o – Morality
2o – admission and length of stay
Table 2:Key findings and strengths/ limitations of studies
Author, Date and Journal Key findings Study strengths Study limitations Comments
Freund. Y et al(5) January 2017, Journal of American Medical Assoc.
qSOFA has low sensitivity when compared to SIRS
Prospective, multiple EDs involved
Primary outcome was mortality. Authors did not look for any other outcomes
Authors used 1992 Sepsis definitions. Worst observations taken for calculations of scores
Finkelstein E.J(6)
Mar 2017
Critical Care
Greater qSOFA score showed higher risk of primary outcome. qSOFA > 2 was equally sensitive as SIRS > 2.
Patients chosen from wards and Emergency department
Retrospective,
Single centre
Authors were unable to prove that qSOFA was better than SIRS for assessment of organ dysfunction free days
Williams J.M et al(7)
Mar 2017
Chest
qSOFA was specific but not sensitive for organ dysfunction. Similar mortality prediction between qSOFA and SIRS
Prospective cohort
Retrospective observational study
Churpek M.M et al(10)
April 2017
American Journal or Respiratory and Critical Care Medicine
MEWS and NEWS were better predictors of mortality and ITU stay
Large number of patients, comparison to MEWS
Retrospective observational study
April M.D et al(9)
May 2017-07-06
Journal of Emergency Medicine
qSOFA had superior specificity and Negative likelihood ratio than SIRS
End point was mortality
Small sample size
Wide confidence- intervals, Single centre study
Omitted death as an endpoint
Park. Hyung et al(11)
June 2017
Journal of Critical Care
qSOFA has low sensitivity for predicting mortality
Large patient cohort
Retrospective analysis
Hwang et al(12)
June 2017
Annals of Emergency Medicine
qSOFA was more specific but less sensitive than SIRS for mortality prediction
Large patient cohort
Tertiary academic centre (external validity decreased)
Retrospective
Henning et al(14)
October 2017 (awaiting publication) Annals of Emergency Medicine
qSOFA had higher specificity but low sensitivity for prediction of mortality
Large
patient cohort, Prospectively collected patient date
Retrospective analysis
Singer, A et al(15)
April 2017
Annals of Emergency Medicine
qSOFA found to predict mortality, ICU admission and prolonged length of stay
Large number of patients had no vital signs documented
Retrospective study, Single centre
Freund et al(5) conducted the first prospective study looking at use of qSOFA in the Emergency Department. This was a prospective, multicentre, international observational study. Patients were recruited from various academic and non-academic centres in France, Belgium, Spain and Switzerland. All adult patients presenting to ED and thought to have an infection were included. A total of 1088 patients were enrolled. Outcomes reviewed were primarily mortality and secondarily length of stay and ITU admission. The strengths of this study are that it was prospective and the authors chose a mix of hospitals around Europe. Experts reviewed the results were blinded to the outcome. The authors concluded that qSOFA had a sensitivity of 70% and specificity of 79% for in-patient mortality, whereas SIRS had sensitivity of 93% and specificity of 27% for the same outcome.
In appraising this study, I noted that the calculation of the qSOFA score was not made on the patient’s arrival to the ED. The observations were taken at the worst level during their admission in ED. The physicians collecting the data were not blinded to the data collection, so reporting bias could have been an issue, although the experts reviewing the data were blinded. The altered mental status was determined by the physicians, which would have variability based on the physicians experience.
Finkelsztein et al (8) conducted a retrospective analysis of patients listed on the Weill Cornell Medicine Registry. The patients were added to the list prospectively. All adult patients admitted to the Medical ITU and with a suspicion of infection were added to the study. 152 patients were added to the study after exclusion criteria were considered. The primary outcome was all cause in-hospital mortality and secondary outcomes were ITU-free, ventilator free or organ dysfunction free days. The authors noted that for primary outcome of mortality, qSOFA > 2 had a sensitivity of 90% and specificity of 42% compared to SIRS > 2, which showed a sensitivity of 93% but a specificity of only 12%. However, secondary outcome of organ dysfunction-free days had sensitivity and specificity with qSOFA > 2 of 68% and 41%, whereas with SIRS > 2 of 89% and 12%.
The authors concluded that qSOFA was a better predictor of mortality than SIRS, however was not better at predicting morbidity- i.e. ITU-free days, Organ dysfunction-free days.
Several other retrospective studies have also compared the use of qSOFA against SIRS. Williams et al (9) looked at 4,176 patients with infection and looked for primary outcome of mortality when comparing qSOFA to SIRS. Churpek M.M(10) et al reviewed assessments for 30,677 patients to compare the same. Churpek M.M(10) et al also compared the predictive capabilities of MEWS and NEWS and noted that both these performed better than either qSOFA or SIRS.
April M.D et al (11) also conducted a retrospective analysis of patients admitted to ITU and found that qSOFA had greater specificity and negative likelihood ratios than SIRS, but similar sensitivity and positive likelihood ratios.
During the course of this review, several conference abstracts were noted with a similar theme. However, they were not included as they lacked much of the details of the studies, including inclusion and exclusion criteria, randomization and blinding. The overwhelming theme even in these papers was that qSOFA lacks sensitivity and early warning scores such as NEWS have a far higher likelihood of predicting sepsis and mortality.
Limitations of this review
All but one of the studies was conducted on retrospective data. The studies were conducted in hospitals outside the U.K, hence external validity may not apply. The health care system in the United States and Europe where many of the studies were conducted are different to the National Health Service.
Most of the studies conducted a retrospective analysis of data. Several of the study authors comment that this lead to multiple patients being excluded due to incomplete charting of vital signs to enable scores to be accurate. The only prospective study conducted (Freund et al), based the score on the worst observation available and not on those taken when patient first arrived in the ED. This would skew the results towards qSOFA being a useful predictor in the ED.
Personal work
I conducted an audit to assess usefulness of scoring systems in septic patients within my current hospital. A retrospective review of 25 patients admitted to ITU at Homerton University Hospital was conducted. The notes for these patients were reviewed using Electronic patient records. 11 patients were found to have been admitted via the Emergency Department. Vital signs were reviewed, again using
the electronic record. The current electronic record calculates the Modified Early Warning Score automatically based on vital signs input into the system. The worst score was taken and the qSOFA score at the corresponding time was calculated.
Of the 11 patients whose records were reviewed, 1 patient was noted to have minor infection with no markers of sepsis. 5 patients (50%) triggered with qSOFA, as compared to 8 (80%) with SIRS and 9 (90%) with NEWS. The column chart is illustrated in Appendix 1 and data gathered in Appendix 2. NEWS was noted to be more sensitive than qSOFA in diagnosing sepsis. Despite calculating the score using the worst observations, NEWS was still found to be more sensitive to qSOFA in predicting morbidity and mortality.
In addition to this I formulated a survey(Appendix 6). This aimed to assess the practice of colleagues in Emergency Medicine in assessment of sepsis. 81% (28 of 32 respondents) were familiar with qSOFA. Only 6% (2 of 32) currently used the qSOFA score for assessment of patients in the ED. 94% (30 of 32) currently used an Early Warning score in their department, either the Modified Early Warning Score or National Early Warning Score. 62% (20 of 32) felt that the current systems in were not specific enough and generated an increased number of false positives.
My current workplace uses electronic patient data to calculate Early Warning scores. Following discussion with the one of the Authors of the Sepsis-3 Guideline, Prof. Mervyn Singer, I intend to do a further assessment of the retrospective data available at my trust.
Discussion
Although qSOFA appears to be a promising scoring system, the current review shows that it does not have sensitivity to be used for assessment of unwell patients in the Emergency Department, especially when there is already an early warning score available. Although the study by Samir et al(14), was excluded from this review due to it
Conclusion
Currently the use of qSOFA would not be of use in the Emergency department. The above studies do show that qSOFA, especially when greater than 2, has a place in predicting mortality. It does not show significant use in identifying sick patients early, or of predicting organ dysfunction. Although lacking specificity SIRS has shows to be an earlier predictor of sepsis in patients with infection. Currently in the UK, the National Early Warning Score has been validated for the same purpose. As noted in the audit I conducted, qSOFA provided a delayed prediction of severity of illness. The survey I conducted (Appendix 6) also showed that awareness of NEWS is better than that of qSOFA. The difficulties of initiation a new scoring system would outweigh the benefits of re-enforcing one that is actively used at present.
Appendices
Appendix- 1
Flow chart
Appendix – 2
Appendix-3
Appendix -4
Appendix 5
qSOFA- Quick Sepsis-related Organ Failure Assessment(4)
Hypotension with Systolic BP>= 100mm Hg
Respiratory rate > 22/min
Altered mental status
SIRS- Systemic inflammatory response syndrome(14)
Respiratory rate more than 20 breaths/min
Temperature more than 38° C or less than 36° C
Heart rate more than 90 beats/min and
White blood cell count more than 12,000/mm3 or less than 4,000/mm3
Appendix 6
– Have you heard of qSOFA? Yes/No
– What are the elements of qSOFA? Hypotension/ Tachycardia/ Tachypnoea/ Fever/ Confusion
– Does your department currently use qSOFA to determine escalation of care to intensive care for septic patients? Yes/ No
– What score does your department use a trigger mechanism? MEWS/ NEWS/ SIRS/ qSOFA/ SOFA/ Gestalt
– What is your opinion of the current scoring system you use? Too many false positives/ too many false negative/ no opinion
– If shown that qSOFA would be useful for prediction of worsening sepsis, would you consider implementing this in your ED? Yes/ No
Bibliography
1. The Sepsis [Internet]. Vol. 39, Viszeralchirurgie. 2004. 142-149 p. Available from: http://www.thieme-connect.de/DOI/DOI?10.1055/s-2004-818761
2. M. S-HS, C.S. D, C. S, M. S-HS, D. A, M. B, et al. The third international consensus definitions for sepsis and septic shock (sepsis-3). JAMA – J Am Med Assoc [Internet]. 2016;315(8):801–10. Available from: http://linker2.worldcat.org/?jHome=http%3A%2F%2Fwww.hlisd.org%2FAtoZBrowse.aspx%3Fbrowsetype%3Dlibrary&linktype=best
3. Vincent JL, Moreno R, Takala J, Willatts S, De Mendonça A, Bruining H, et al. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. Intensive Care Med. 1996;22(7):707–10.
4. Seymour CW, Liu VX, Iwashyna TJ, Brunkhorst FM, Rea TD, Scherag A, et al. Assessment of Clinical Criteria for Sepsis: For the Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). Jama. 2016;
5. Freund Y, Lemachatti N, Krastinova E, Van Laer M, Claessens Y-E, Avondo A, et al. Prognostic Accuracy of Sepsis-3 Criteria for In-Hospital Mortality Among Patients With Suspected Infection Presenting to the Emergency Department. JAMA [Internet]. 2017;317(3):301–8. Available from: http://linker2.worldcat.org/?jHome=http%3A%2F%2Fwww.hlisd.org%2FAtoZBrowse.aspx%3Fbrowsetype%3Dlibrary&linktype=best
6. E. F, D. J, K. M, M. P, E. S, D. B, et al. Validation of the recently proposed qSOFA score in the weill cornell medicine registry of critical illness. Chest [Internet]. 2016;150(4). Available from: http://linker2.worldcat.org/?rft.institution_id=129754&pkgName=sciDirect&issn=1931-3543&linkclass=to_article&jKey=00123692&issue=4&provider=elsevier&date=2016-10&aulast=Finkelsztein+E.%3B+Jones+D.%3B+Ma+K.%3B+Pabon+M.%3B+Schenck+E.%3B+Berlin+D.%3B+Siempos
7. Greenslade JH, McKenzie J V, Chu K, Brown AFT, Williams JM, Lipman J, et al. Systemic Inflammatory Response Syndrome, Quick Sequential Organ Function Assessment, and Organ Dysfunction: Insights From a Prospective Database of ED Patients With Infection. Chest. 2017;151(3):586–96.
8. Churpek MM, Snyder A, Han X, Sokol S, Pettit N, Howell MD, et al. Quick Sepsis-related Organ Failure Assessment, Systemic Inflammatory Response Syndrome, and Early Warning Scores for Detecting Clinical Deterioration in Infected Patients outside the Intensive Care Unit. Am J Respir Crit Care Med [Internet]. 2017;195(7):906–11. Available from: http://linker2.worldcat.org/?rft.institution_id=129754&spage=906&pkgName=nhshospital&PQUEST.WAYFlessID=48433&issn=1073-449X&linkclass=to_article&jKey=40575&issue=7&provider=PQUEST&date=2017-04&aulast=Churpek%2C+Matthew+M%3B+Snyder%2C+Ashley%3B+Han%2C+Xuan
9. April MD, Aguirre J, Tannenbaum LI, Moore T, Pingree A, Thaxton RE, et al. Sepsis Clinical Criteria in Emergency Department Patients Admitted to an Intensive Care Unit: An External Validation Study of Quick Sequential Organ Failure Assessment. J Emerg Med [Internet]. 2016; Available from: http://linker2.worldcat.org/?rft.institution_id=129754&pkgName=ckjournalathens&issn=0736-4679&linkclass=to_article&jKey=07364679&provider=elsevier&date=2016-11&aulast=April%2C+Michael+D%3B+Aguirre%2C+Jose%3B+Tannenbaum%2C+Lloyd+I%3B+Moore%2C+Tyler%3B+Pingree%2C+Alexander%3B+Thaxton%2C+Robert+E%3B+Sessions%2C+Daniel+J%3B+Lantry%2C+James+H&atitle=Sepsis+Clinical+Criteria+in+Emergency+Department+Patients+Admitted+to+an+Intensive+Care+Unit%3A+An+External+Validation+Study+of+Quick+Sequential+Organ+Failure+Assessment.&title=Journal+of+Emergency+Medicine&rft.content=fulltext%2Cprint&linkScheme=ckeyathens&jHome=https%3A%2F%2Fauth.elsevier.com%2FShibAuth%2FinstitutionLogin%3FentityID%3Dhttps%253A%252F%252Fidp.eduserv.org.uk%252Fopenathens%26appReturnURL%3Dhttps%253a%252f%252fwww.clinicalkey.com%252fdura%252fbrowse%252fjournalIssue%252f07364679&rft.id=info%3Apmid%2F&rft.order_by=preference&linktype=best
10. Hwang SY, Jo IJ, Lee SU, Lee TR, Yoon H, Cha WC, et al. Low Accuracy of Positive qSOFA Criteria for Predicting 28-Day Mortality in Critically Ill Septic Patients During the Early Period After Emergency Department Presentation. Ann Emerg Med [Internet]. 2017; Available from: http://linker2.worldcat.org/?rft.institution_id=129754&pkgName=ckjournalathens&issn=0196-0644&linkclass=to_article&jKey=01960644&provider=elsevier&date=2017-06&aulast=Hwang%2C+Sung+Yeon%3B+Jo%2C+Ik+Joon%3B+Lee%2C+Se+Uk%3B+Lee%2C+Tae+Rim%3B+Yoon%2C+Hee%3B+
11. Hyun Kyung Park, Won Young Kim, Myung Chun Kim WJ and BSK, Park HK, Kim WY, Kim MC, Jung W, Ko BS. Quick sequential organ failure assessment compared to systemic inflammatory response syndrome for predicting sepsis in emergency department. J Crit Care [Internet]. 2017 [cited 2017 Jul 6];42:12–7. Available from: http://linker2.worldcat.org/?rft.institution_id=129754&spage=12&pkgName=ckjournalathens&issn=0883-9441&linkclass=to_article&jKey=08839441&provider=elsevier&date=2017-06&aulast=Park%2C+Hyun+Kyung%3B+Kim%2C+Won+Young%3B+Kim%2C+Myung+Chun%3B+Jung%2C+Woong%3B
12. Henning DJ, Puskarich MA, Self WH, Howell MD, Donnino MW, Yealy DM, et al. An Emergency Department Validation of the SEP-3 Sepsis and Septic Shock Definitions and Comparison With 1992 Consensus Definitions. Ann Emerg Med [Internet]. 2017; Available from: http://linker2.worldcat.org/?rft.institution_id=129754&pkgName=ckjournalathens&issn=0196-0644&linkclass=to_article&jKey=01960644&provider=elsevier&date=2017-10&aulast=Henning+D.J.%3B+Puskarich+M.A.%3B+Jones+A.E.%3B+Self+W.H.%3B+Howell+M.D.%3B+Donnino+M.W.
13. Singer AJ, Ng J, Thode HCJ, Spiegel R, Weingart S. Quick SOFA Scores Predict Mortality in Adult Emergency Department Patients With and Without Suspected Infection. Ann Emerg Med. 2017;69(4):475–9.
14. Haydar S, Spanier M, Weems P, Wood S, Strout T. Comparison of QSOFA score and SIRS criteria as screening mechanisms for emergency department sepsis. Am J Emerg Med [Internet]. 2017;35(11):1730–3. Available from: https://auth.elsevier.com/ShibAuth/institutionLogin?entityID=https://idp.eng.nhs.uk/openathens&appReturnURL=https%3A%2F%2Fwww.clinicalkey.com%2Fcontent%2FplayBy%2Fdoi%2F%3Fv%3D10.1016%2Fj.ajem.2017.07.001
15. Levy MM, Fink MP, Marshall JC, Abraham E, Angus D, Cook D, et al. 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Crit Care Med [Internet]. 2003;31(4):1250–6. Available from: http://content.wkhealth.com/linkback/openurl?sid=WKPTLP:landingpage&an=00003246-200304000-00038