Low Accuracy of Positive qSOFA Criteria for Predicting 28-Day Mortality in Critically Ill Septic Patients During the Early Period After Emergency Department Presentation.

Abstract

We determine the diagnostic performance of positive Quick Sequential Organ Failure Assessment (qSOFA) scores for predicting 28-day mortality among critically ill septic patients during the early period after emergency department (ED) presentation. This was a retrospective cohort study at a tertiary care academic center. We reviewed a registry of adult (≥18 years) patients who received a diagnosis of severe sepsis or septic shock during an ED stay from August 2008 through September 2014. We identified the point at which patients met 2 or more of the 3 qSOFA criteria (indicating a positive qSOFA score) simultaneously during the initial 24 hours. The diagnostic performance of positive qSOFA score for predicting 28-day mortality was assessed (on ED arrival and within 3, 6, and 24 hours after ED presentation). A total of 1,395 patients were included, and the overall 28-day mortality was 15%. For patients with positive qSOFA score, 28-day mortality was 23% (95% confidence interval [CI] 19% to 28%) on ED arrival, 20% (95% CI 17% to 23%) at 3 hours, 20% (95% CI 17% to 22%) at 6 hours, and 17% (95% CI 15% to 20%) at 24 hours. Positive qSOFA score for predicting 28-day mortality had a sensitivity, specificity, and area under the receiver operating curve, respectively, of 39% (95% CI 32% to 46%), 77% (95% CI 75% to 80%), and 0.58 (95% CI 0.55 to 0.62) on ED arrival; 68% (95% CI 62% to 75%), 52% (95% CI 49% to 55%), and 0.60 (95% CI 0.57 to 0.63) within 3 hours; 82% (95% CI 76% to 87%), 41% (95% CI 38% to 44%), and 0.61 (95% CI 0.58 to 0.64) within 6 hours; and 91% (95% CI 86% to 94%), 23% (95% CI 21% to 25%), and 0.57 (95% CI 0.54 to 0.59) within 24 hours. The diagnostic performance of positive qSOFA score for predicting 28-day mortality was low in critically ill septic patients, particularly during the early period after ED presentation. The study requires further prospective validation because of limitations with its retrospective design and use of single-center data.