Elsevier

Journal of Critical Care

Volume 46, August 2018, Pages 29-36
Journal of Critical Care

Sepsis/Infection
The systemic inflammatory response syndrome criteria and their differential association with mortality

https://doi.org/10.1016/j.jcrc.2018.04.005Get rights and content

Highlights

  • Individual or combined SIRS criteria may not be equivalent or interchangeable.

  • In >130,00 patients with infection and organ failure, different individual and combinations of SIRS criteria were associated with marked differences in mortality.

  • With any two SIRS criteria, mortality varied from 12% to 31%.

  • The differences in mortality remained unchanged over time.

Abstract

Purpose

Despite the recent Sepsis-3 consensus, the Systemic Inflammatory Response Syndrome (SIRS) criteria continue to be assessed and recommended. Such use implies equivalence and interchangeability of criteria. Thus, we aimed to test whether such criteria are indeed equivalent and interchangeable.

Materials and methods

From 2000 to 2015, we identified patients with infection, organ failure, and at least one SIRS criterion in 179 Intensive Care Units in Australia and New.

Zealand. We studied the association of different SIRS criteria with hospital mortality.

Results

Among 131,016 patients with infection and organ failure, mortality increased from 10.6% for the respiratory rate criterion to 15.8% for the heart rate criterion (P < 0.01); from 10.1% for the high leukocyte count criterion to 20.0% for a low count and from 10.1% for a high temperature to 14.4% for a low temperature criterion. With any two SIRS criteria, hospital mortality varied from 11.5% to 30.8% depending on the combination of criteria. This difference remained unchanged after adjustments and was consistent over time.

Conclusions

Different individual and combinations of SIRS criteria were associated with marked differences in hospital mortality. These differences remained unchanged after adjustment and over time and imply that individual SIRS criteria are not equivalent or interchangeable.

Introduction

Severe sepsis is a major cause of intensive care unit admission and mortality [1,2]. Until recently, sepsis, severe sepsis and septic shock were defined by the consensus statement of the American College of Chest Physicians and the American Society of Critical Care Medicine in 1992 [3]. Since then, randomized controlled trials in sepsis have used such criteria for patient enrolment [[4], [5], [6], [7], [8], [9], [10], [11]], including the need for at least two Systemic Inflammatory Response Syndrome (SIRS) criteria. The sensitivity of this approach in identifying intensive care unit patients with severe sepsis, however, was recently challenged by a large observational study, which found that one in eight septic patients did not have two or more SIRS criteria [12]. This observation and well documented concerns about specificity [13,14] have recently led the joint American Society of Critical Care Medicine (SCCM) and European Society of Intensive Care Medicine (ESICM) Task Force on Sepsis to remove the SIRS criteria from the definition of sepsis [15]. Despite such decision, however, strong support remains for the continued use of SIRS criteria [16,17], and their assessment continues to be reported as part of the diagnosis of sepsis [18], thus fueling a continuing controversy about the validity and clinical usefulness of such criteria.

Such strong support for the continued application of the SIRS criteria as originally described logically implies that each SIRS criterion must be considered equivalent and interchangeable with another. Thus, for example, patients with an elevated respiratory rate and fever are, by definition, taken to be broadly equivalent in terms of pathophysiological state and mortality risk to patients with a low white cell count and tachycardia (both groups would have 2 SIRS criteria). Whether this assumption is correct has significant bearing on the scientific, pathophysiological and clinical validity of the SIRS criteria and their continued use, for example, to assist with patient enrolment into sepsis trials. However, the robustness of this assumption for both different individual criteria and different combinations of criteria has never been tested.

We hypothesized that individual SIRS criteria have markedly different associations with.

hospital mortality in intensive care unit patients with infection and organ failure and that various combinations of two or three SIRS criteria identify septic intensive care unit patients with markedly different mortality risks. Moreover, we hypothesized that the two SIRS criteria (white cell count and temperature) that carry a high or a low value SIRS inclusion trigger have markedly different associations with mortality depending on their fulfilment on the basis of a high vs. low value. Finally, we hypothesized that such differential associations have held steady over more than a decade despite major changes in hospital mortality overall and in each subgroup.

Section snippets

Study design

We conducted a retrospective study of intensive care unit (ICU) patients from January 1, 2000, to December 31, 2015, using data from the Australia and New Zealand Intensive Care Society (ANZICS) Adult Patient Database (APD), run by the ANZICS Centre for Outcome and Resource Evaluation [19]. Data were gathered by means of clinical registry surveillance by data collectors for quality-assurance and benchmarking as previously described [12,20] and capture data on >80% of all intensive care unit

Study population

A total of 1,557,844 patients above 16 years old were treated during the assessment period in the 179 study ICUs. The exclusion process of patients is presented in Table E1. The final study population constituted of 131,016 patients with infection, organ failure and with at least one SIRS criterion. The baseline characteristics of study patients stratified by survivorship are presented in Table 1. Overall, septic shock was present in two thirds of patients, mechanical ventilation was applied to

Key findings

We performed a multicenter observational study of ICU patients with sepsis to assess the prognostic equivalence and interchangeability of SIRS criteria in terms of their individual, combination-related, and high vs. low value association with hospital mortality. We found that, in a cohort of >130,000 sepsis patients, separate individual SIRS criteria had different associations with hospital mortality. Moreover, we found that separate combinations of SIRS criteria and separate low vs. high

Conclusions

In patients with sepsis, we aimed to assess the equivalence and interchangeability of individual, combinations, and dual-fulfilment (high vs. low) SIRS criteria in terms of hospital hospital mortality. We found that, in a cohort of >130,000 patients, separate individual SIRS criteria, separate combinations of SIRS criteria and separate low vs. high values for the white cell count and temperature criteria all had markedly different associations with hospital mortality. We also found that such

Acknowledgements

No external funding.

Contributionship

KMK, MB, DP, DJC, RB: Substantial contributions to the conception or design of the work.

MB, DP: Acquisition of the data.

MB: analysis of the data.

KMK, MB, DP, DJC, RB: interpretation of data for the work.

KMK, MB, DP, DJC, RB: Drafting the work or revising it critically for important intellectual content.

KMK, MB, DP, DJC, RB: Final approval of the version submitted for publication.

KMK, MB, DP, DJC, RB: Accountability for all aspects of the work in ensuring that questions related to the accuracy

Declaration of interest

KMK: none.

MB: none.

DP: none.

DC: none.

RB: none.

Funding

No external funding was granted for this project.

References (25)

  • S.M. Opal et al.

    Effect of eritoran, an antagonist of MD2-TLR4, on mortality in patients with severe sepsis: the ACCESS randomized trial

    JAMA

    (2013)
  • V.M. Ranieri et al.

    Drotrecogin alfa (activated) in adults with septic shock

    N Engl J Med

    (2012)
  • Cited by (0)

    View full text