Beyond Vital Signs: Pediatric Sepsis Screening That Includes Organ Failure Assessment Detects Patients With Worse Outcomes

Objective: Pediatric sepsis screening is recommended. The 2005 Goldstein criteria, the basis of our institutional sepsis screening tool (ISST), correlates poorly with clinically diagnosed sepsis. The study objective was to retrospectively evaluate the ISST sensitivity compared to the Pediatric Sequential Organ Failure Assessment (pSOFA). Methods: Single-center retrospective cohort study. The primary outcome was pSOFA score and ISST sensitivity for severe sepsis. Secondary outcomes included clinical outcome measures. Results: In this severe sepsis cohort (N=491), pSOFA and ISST sensitivity were 57.6% and 61.1%, respectively. In regression analysis for a positive pSOFA, positive blood culture (OR 2.2, 95% CI 1.1-4.3, p=0.025), older age (OR 1.006, 95% CI 1.003-1.009, p<0.001), and pulmonary infectious source (OR 3.3, 95% CI 1.6-6.5, p=0.001) demonstrated independent association. In regression analysis for a positive ISST, older age (OR 1.003, 95% CI 1-1.006, p=0.031) and intra-abdominal infectious source (OR 0.3, 95% CI 0.1-0.8, p=0.014) demonstrated independent association. A negative ISST was associated with higher ICU admission prevalence (p=0.01) and fewer ICU free days (p=0.018). A positive pSOFA score was associated with higher ICU admission prevalence, vasopressor requirement, and vasopressor days as well as fewer ICU, hospital, and mechanical ventilation free days (all p<0.001). Exploratory analysis combining the ISST and pSOFA into a hybrid screen demonstrated superior sensitivity (84.3%) and outcome discrimination. Conclusions: The pSOFA demonstrated non-inferior sensitivity to a Goldstein-based institutional sepsis screening model. Further, pSOFA was a better discriminator of poor clinical outcomes. An exploratory hybrid screening model shows superior performance, but will require prospective study.