Elsevier

Journal of Infection

Volume 76, Issue 3, March 2018, Pages 241-248
Journal of Infection

Early Warning Scores do not accurately predict mortality in sepsis: A meta-analysis and systematic review of the literature

https://doi.org/10.1016/j.jinf.2018.01.002Get rights and content

Highlights

  • Early Warning Scores are commonly used to assess patients in hospital.

  • The available evidence suggests they are inaccurate at predicting mortality.

  • EWS cannot rule in (LR+ 1.79) or rule out (LR- 0.59) mortality.

  • The studies involved are generally at moderate-high risk of bias.

Abstract

Objectives

Early Warning Scores are used to evaluate patients in many hospital settings. It is not clear if these are accurate in predicting mortality in sepsis. We performed a systematic review and meta-analysis of multiple studies in sepsis. Our aim was to estimate the accuracy of EWS for mortality in this setting.

Methods

PubMED, CINAHL, Cochrane, Web of Science and EMBASE were searched to October 2016. Studies of adults with sepsis who had EWS calculated using any appropriate tool (e.g. NEWS, MEWS) were eligible for inclusion. Study quality was assessed using QUADAS-2. Summary estimates were derived using HSROC analysis.

Results

Six studies (4298 participants) were included. Results suggest that EWS cannot be used to predict which patients with sepsis will (positive likelihood ratio 1.79, 95% CI 1.53 to 2.11) or will not die (negative likelihood ratio 0.59, 95% CI 0.45 to 0.78). Two studies were rated as low risk of bias and one as unclear risk of bias on all domains. The other three studies were judged at high risk of bias in one domain.

Conclusion

Early Warning Scores are not sufficiently accurate to rule in or rule out mortality in patients with sepsis, based on the evidence available, which is generally poor quality.

Introduction

Sepsis is a major problem in emergency departments and hospitals causing significant morbidity and mortality.1, 2 A steady increase in the incidence of sepsis has been noted worldwide, with recent UK estimates of around 50 cases of severe sepsis per 100,000 people per year.2 Collecting data on accurate mortality figures is difficult, but recent estimates suggest an estimated case mortality of between 20%–35%, although rates differ widely by definition.3, 4 Timely diagnosis of patients with sepsis can be difficult, because of the differences in clinical presentations, and few uniquely identifying features, especially in the elderly.5

Once patients with sepsis are identified, rapid treatment has been shown to improve mortality, although nearly all evidence comes from observational studies.6, 7, 8 Due to the significant number of patients who present with sepsis, it is critical to identify the patients at risk of deterioration, and patients requiring urgent treatment or critical care input. Alongside this, there is also a need to identify the significant number of patients who are likely to have a good outcome and hence can be managed more conservatively.

Early Warning Scores (EWS) are physiological composite scores comprising pulse rate, blood pressure, temperature, respiratory rate, mental state and oxygen saturation. Each of these physical observations is given a score, where 0 is considered normal. Simple addition of these observations allows a total score to be calculated, usually between 0 and 12. Different versions of EWS often have minor modifications, such as the addition of points if the patient is receiving oxygen therapy, or variations in specific cut-offs.

Recently, the UK National Institute for Health and Care Excellence (NICE) has published guidelines on the recognition, diagnosis and management of sepsis.9 These guidelines recommend considering the use of Early Warning Scores in assessing patients with suspected sepsis in acute hospital settings, and highlight this area as a key research topic. The 2015 National Confidential Enquiry on Patient Outcome and Death report on sepsis also suggested use of EWS as a principal recommendation in all care settings, specifically to assess severity of sepsis and to prioritise urgent care.10

Over the last ten years, EWS have been introduced into nearly all UK hospitals, and are already recommended in NICE guidance for monitoring critically ill patients in hospital and Royal College of Physician guidance for monitoring of all adult patients in acute hospital settings.11, 12 They have recently been introduced into emergency departments, and also into many ambulances services and community settings and primary care.13 Given these scores are often calculated on every patient, there have been concerns about the volume of workload created and the sensitivity of these scores for identifying unwell patients.14

EWS are often used, both informally and formally, to guide treatment decisions such as the best location for care (inpatient, outpatient, ICU), and the level of monitoring or seniority of doctor that should see the patient. In some centres, a certain score (for example, greater than 5), will trigger a pager alert to senior medical staff or critical care outreach services. Although there is some evidence that this method identifies sick patients, the evidence relating to patients with sepsis is limited.15, 16

Section snippets

Methods

A protocol was developed based on recommended standards for conduct of systematic reviews.17, 18 The review was registered with PROSPERO (CRD42016047125), and the PRISMA guidelines were followed when reporting results.

Results

Fig. 1 shows the flow of studies through the review process. The searches identified 620 possible studies, of which 6 were included (4298 participants). Two of the studies were conducted in the UK,21, 22 and one each in Germany,23 Israel,24 Italy25 and Turkey26 (Table 1). The Italian study was available as a conference abstract only.

Four studies were performed in emergency departments21, 22, 23, 26; two were in internal medicine.24, 25 Three studies were relatively small in size (<400

Summary of findings

We identified six relevant studies for inclusion in our review. Results suggest that EWS cannot be used to predict who will (positive likelihood ratio 1.79, 95% CI 1.53, 2.11) or will not (negative likelihood ratio 0.59, 95% CI 0.45, 0.78) die in patients with sepsis. Sensitivity analyses suggested that results were similar when restricted to studies at low risk of bias, studies with low concern regarding applicability, studies with low or unclear concerns regarding applicability, studies

Conclusions

EWS have poor prognostic value in predicting sepsis mortality. Based on the existing data, which is of poor quality, EWS should not be used on their own to guide prognosis in patients with sepsis, and are unlikely to be reliable alone in identifying patients at risk or death. Further work is needed to assess the role of EWS in patients with suspected sepsis.

Role of the funding source

Penny Whiting's time was supported by the National Institute for Health Research (NIHR) Collaboration for Leadership in Applied Health Research and Care (CLAHRC) West at University Hospitals Bristol NHS Foundation Trust. The funders had no role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report; or in the decision to submit the article for publication. All researchers were fully independent from the funders.

Competing interests

All authors have completed the ICMJE uniform disclosure form at www.icmje.org/coi_disclosure.pdf and declare: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years; no other relationships or activities that could appear to have influenced the submitted work.

Authors' contributions

FH and MA conceived the idea for the review. FH and PW drafted the article. FH, DA, AB and MA served as content experts in the field of EWS. PW provided methodological support. FH, DA and AB undertook screening and data extraction. FH and DA performed the risk of bias assessment. All authors contributed to the interpretation of results, commented on draft manuscripts and have given their approval for publication. The views expressed in this article are those of the author(s) and not necessarily

Ethics approval

As this was a systematic review no ethics approval was needed.

Data sharing statement

The full dataset is available in tables included in the review.

Transparency

The lead author affirms that this manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned (and, if relevant, registered) have been explained.

Patient involvement

Patients were not involved in the design or conduct of this study.

Acknowledgements

We would like to thank Alison Richards, NIHR CLAHRC West, for conducting the literature searches.

References (33)

  • NICE

    Sepsis: Recognition, Diagnosis and Early Management | Recommendations | Guidance and Guidelines | NICE

  • National Confidential Enquiry into Patient Outcome and Death

    NCEPOD – Sepsis: Just Say Sepsis! (2015).

  • NICE

    Acutely Ill Adults in Hospital: Recognising and Responding to Deterioration

  • RCP Acute Medicine Task Force

    National Early Warning Score (NEWS)

  • NasMeD

    Future National Clinical Priorities for Ambulance Services in England

  • S. Jarvis et al.

    Aggregate National Early Warning Score (NEWS) values are more important than high scores for a single vital signs parameter for discriminating the risk of adverse outcomes

    Resuscitation

    (2014)
  • Cited by (0)

    View full text