Development of a New Screening Tool for Pediatric Septic Shock.

Abstract

Existing screening tools for sepsis in children are limited by suboptimal sensitivity. Our objective was to develop a new, more sensitive screening tool for pediatric septic shock by enhancing 2 aspects of the 4-point Liverpool quick Sequential Organ Failure Assessment (LqSOFA) tool. We performed a secondary analysis of a cohort of children (1 month to 18 years) who presented to a pediatric emergency department (ED) with suspected infection over a 10-year period. Septic shock was defined as intravenous vasoactive infusion within 24-hours of arrival for children with suspected infection and intravenous antibiotics administered. We developed the 4-point quick Pediatric Septic Shock Screening Score (qPS4) by making 2 changes to the 4-point LqSOFA: (1) the pulse rate parameter was replaced with the recently derived Temperature and Age-adjusted Mean Shock Index and (2) standard respiratory rate cutoffs for tachypnea were replaced by cutoffs derived empirically from the study cohort. The other 2 LqSOFA criteria were unchanged (abnormal mentation and capillary refill ≥3 seconds). We defined a positive qPS4 as ≥2 criteria (consistent with LqSOFA). We used the training cohort from the parent study to derive cutoffs for respiratory rate and the validation cohort to compare the qPS4 with LqSOFA and qSOFA. Among the 47,231 encounters in the validation cohort from the parent study, with median age of 4.5 years, qPS4 had an area under the receiver operating characteristic curve for septic shock of 0.94 (95% confidence interval [CI] 0.92 to 0.96). qPS4 ≥2 had a sensitivity of 89.7% (95% CI 84.9% to 94.5%), and a specificity of 92.2% (95% CI 92.0% to 92.5%) for septic shock. In comparison, the LqSOFA achieved an area under the receiver operating characteristic curve of 0.86 (95% CI 0.82 to 0.89), a sensitivity of 56.1% (95% CI 48.3% to 63.9%), and a specificity of 96.8% (95% CI 96.6% to 96.9%). The median time from first positive qPS4 to initiation of an intravenous vasoactive infusion was 2.5 hours (IQR 0.9 to 6.1) compared to 0.7 hours (IQR 0.0 to 4.5) for LqSOFA. The qPS4, with 2 enhancements to the LqSOFA, demonstrated overall improved sensitivity and specificity for pediatric septic shock.