Can we rescue the Quick Sepsis-related Organ Failure Assessment?

BACKGROUND. Early warning scores (EWSs) or scales for risk assessment have been shown to have good predictive value for identifying patients at high risk of a poor clinical course. Which EWS is best for assessing risk for sepsis, however, is much disputed.
OBJECTIVE. To compare the predictive value of 8 EWSs to identify which patients with infections are likely to have a poor clinical outcome, defined as death or admission to an intensive care unit (ICU) within 30 days.
METHODS. Prospective observation study of patients treated in our emergency department in 2020 in whom infection was suspected. We assessed each patient’s risk for a poor clinical outcome with each EWS and calculated each EWS’s ability to predict a composite outcome of death or ICU admission within 30 days.
RESULTS. One hundred four patients were enrolled; 53 patients (51%) were women and the median age was 77 years (interquartile range, 28.7) years. The composite 30-day outcome occurred in 18 patients (17.3%). The National EWS 2 (NEWS2) had the highest sensitivity (88.9%) and second-best area under the receiver operating characteristic curve (AUC) of 0.772. However, its specificity was low (48.9%). The Quick Sepsis-related Organ Failure Assessment (qSOFA), on the other hand, had low sensitivity (50%) but the highest specificity (87.2%) and AUC (0.796). The qSOFA fulfilled its screening function, however, if the score was 0 (no patient with this score died or entered the ICU) or 2 or more (the composite outcome occurred in 45% of such patients). The lactate-enhanced qSOFA-L increased the tool’s sensitivity to 77.8% and moderately decreased specificity (to 69.9%) and the AUC (to 0.753). All 3 scores had high negative predictive values.
CONCLUSIONS. The EWS most appropriate for predicting complications in patients suspected of having sepsis was the NEWS2. However, the qSOFA-L gave similar results and might become a useful screening tool for managing sepsis appropriately.