Prognostic Accuracy of qSOFA and SIRS for Mortality in the Emergency Department: A Meta-Analysis and Systematic Review of Prospective Studies

Abstract

Objective. This meta-analysis aimed to determine the prognostic performance of quick sequential organ failure assessment (qSOFA) score in comparison to systemic inflammatory response syndrome (SIRS) in predicting in-hospital mortality in the emergency department (ED) patients. Methods. Eligible studies comparing the performance of qSOFA and SIRS in predicting in-hospital death of ED patients were identified from searching PubMed, Embase, and Cochrane. Raw data were collected, and the pooled sensitivity and specificity were calculated for qSOFA and SIRS. The summary receiver operating curve was also plotted to calculate the area under the curve. Results. A total of 16 prospective studies with 35,756 patients and 2,285 deaths were included. The pooled sensitivity was 0.43 (95% CI: 0.32–0.54) and 0.8 (95% CI: 0.73–0.86) for qSOFA and SIRS, respectively. The pooled specificity was 0.89 (95% CI: 0.84–0.93) and 0.39 (95% CI: 0.3–0.5) for qSOFA and SIRS, respectively. The area under the summary receiver operating curve was 0.76 (95% CI: 0.72–0.8) and 0.67 (95% CI: 0.62–0.72) for qSOFA and SIRS, respectively. A significant heterogeneity was observed for both qSOFA and SIRS studies. Conclusion. The present meta-analysis suggested that qSOFA had a higher specificity but a lower sensitivity as compared with SIRS in predicting in-hospital mortality in the ED patients. qSOFA appeared to be a more concise and simple way to recognize patients at high risk for death. However, the use of SIRS in the ED cannot be completely replaced since the sensitivity of qSOFA was relatively lower.

Soluble urokinase plasminogen activator receptor and procalcitonin for risk stratification in patients with a suspected infection in the emergency department: a prospective cohort study

“…The performance of qSOFA has been controversial with varying results in different studies. In our study, we found an AUC of 0.61 of qSOFA on 30-day mortality, which is considered too low to reliably predict mortality in the ED [ 7 , 19 , 20 ].…”

Modification of qSOFA Increases Mortality Prediction Accuracy in Sepsis Patients

“…Many studies showed that qSOFA has a low sensitivity but high specificity when it is used in ED for predicting mortality. The low sensitivity made the use of qSOFA alone could not satisfy clinicians, especially when it is used in ED for predicting mortality (16,18). A study in patients with secondary peritonitis showed that qSOFA's predictive performance for inhospital mortality has been found to be inferior to SIRS criteria (12).…”

Response to Re: Modified Sequential Organ Failure Assessment score for predicting mortality in emergency department patients with sepsis

“…The net reclassification index, integrated discrimination improvement and decision curve analysis are the previous three statistical methods that are currently frequently utilised to evaluate prediction models. 4 However, the AUC analysis is the only method used in this paper.…”

“…A recent meta-analysis of 16 prospective studies found qSOFA to have a pooled AUC of 0.76 (0.72-0.8) for in-hospital mortality. 4 A systematic review and meta-analysis of 24 studies (23 using qSOFA) found the pooled AUC for qSOFA for mortality to be 0.77 (0.72-0.82). 5 Individual studies with an AUC for qSOFA for predicting 28-day mortality higher than 0.8 include that by Quinten et al, which found qSOFA to have an AUC of 0.85 (0.73-0.96), 6 and Boillat-Blanco et al, with an AUC of 0.8 (0.73-0.87).…”