The sensitivity of qSOFA calculated at triage and during emergency department treatment to rapidly identify sepsis patients

Sarah M Perman,Byron Drumheller,David F Gaieski,Anish K Agarwal,Adit Ginde,Abhishek Bhardwaj,Mark E Mikkelsen,S Cham Sante,Munish Goyal

doi:10.1038/s41598-020-77438-8

Abstract

The quick sequential organ failure assessment (qSOFA) score has been proposed as a means to rapidly identify adult patients with suspected infection, in pre-hospital, Emergency Department (ED), or general hospital ward locations, who are in a high-risk category with increased likelihood of “poor outcomes:” a greater than 10% chance of dying or an increased likelihood of spending 3 or more days in the ICU. This score is intended to replace the use of systemic inflammatory response syndrome (SIRS) criteria as a screening tool; however, its role in ED screening and identification has yet to be fully elucidated. In this retrospective observational study, we explored the performance of triage qSOFA (tqSOFA), maximum qSOFA, and first initial serum lactate (> 3 mmol/L) at predicting in-hospital mortality and compared these results to those for the initial SIRS criteria obtained in triage. A total of 2859 sepsis cases were included and the in-hospital mortality rate was 14.4%. The sensitivity of tqSOFA ≥ 2 and maximum qSOFA ≥ 2 to predict in-hospital mortality were 33% and 69%, respectively. For comparison, the triage SIRS criteria and the initial lactate > 3 mmol/L had sensitivities of 82% and 65%, respectively. These results demonstrate that in a large ED sepsis database the earliest measurement of end organ impairment, tqSOFA, performed poorly at identifying patients at increased risk of mortality and maximum qSOFA did not significantly outperform initial serum lactate levels.

Highlights

The quick sequential organ failure assessment score has been proposed as a means to rapidly identify adult patients with suspected infection, in pre-hospital, Emergency Department (ED), or general hospital ward locations, who are in a high-risk category with increased likelihood of “poor outcomes:” a greater than 10% chance of dying or an increased likelihood of spending 3 or more days in the ICU
TqSOFA was derived from triage vital signs; the first lactate was drawn at a median of 58 min (IQR 27–123) after presentation; and the maximum quick sequential organ failure assessment (qSOFA) was derived from the worst vitals during the ED length of stay, with a mean length of stay of 477 (± 248) min
AUC 0.50 (0.48–0.52) 0.50 (0.49–0.50) 0.63 (0.60–0.65). In this analysis of a large single center highly granular database of patients admitted to the hospital with sepsis, triage qSOFA ≥ 2 had low sensitivity for in-hospital mortality and triage qSOFA negative sepsis patients were at significant risk of suffering in-hospital mortality (11.7%)

Summary

Introduction

The authors concluded that the new definitions should “facilitate earlier recognition and more timely management of patients with sepsis or at risk of developing sepsis1.” Since this assertion in 2016, numerous authors have analyzed the usefulness of qSOFA in retrospective and prospective cohorts at different points in the care continuum from pre-hospital[16,17] to initial t riage[18,19,20] to the period of ED m anagement[20,21] to in-patient wards and the ICU15; have looked at it as a screening tool for all patients presenting to the ED22 or for those with suspected infection[23,24]; have investigated dynamic changes in qSOFA during ED s tay[20,25]; have analyzed its accuracy as a predictor of ICU admission, length of stay, and in-hospital mortality[26]; have tried to improve the performance of qSOFA by add various biomarkers including lactate27,28, procalcitonin[29], monocyte distribution width[30], and CRP combined with mid-regional proadrenomedullin[31] or vital sign measures including heart rate variability32, EtCO233, and shock index[19]; have examined its utility in high and low resource s ettings[27,29,34]; and have compared it to other scoring systems including SIRS, MEWS, NEWS, and conventional S OFA35,36. Specificity, and area under the receive operator characteristic (AUROC) curve of tqSOFA, maximum qSOFA, and initial venous lactate > 3 mmol/L, which was the cut-off for organ dysfunction in the 2 nd International Sepsis D efinitions[14], at identifying sepsis patients at risk for increased in-hospital mortality

Methods

Results

Conclusion