Development and Validation of a Model to Predict Pediatric Septic Shock Using Data Known 2 Hours After Hospital Arrival.

Abstract

To use electronic health record data from the first 2 hours of care to derive and validate a model to predict hypotensive septic shock in children with infection. Derivation-validation study using an existing registry. Six emergency care sites within a regional pediatric healthcare system. Three datasets of unique visits were designated:1)training set (five sites, April 1, 2013, to December 31, 2016),2)temporal test set (five sites, January 1, 2017, to June 30, 2018), and3)geographic test set (sixth site, April 1, 2013, to June 30, 2018). Patients in whom clinicians were concerned about serious infection from 60 days to 18 years were included; those with septic shock in the first 2 hours were excluded. There were 2,318 included visits; 197 developed septic shock (8.5%). Lasso with 10-fold cross-validation was used for variable selection; logistic regression was then used to construct a model from those variables in the training set. Variables were derived from electronic health record data known in the first 2 hours, including vital signs, medical history, demographics, and laboratory information. Test characteristics at two thresholds were evaluated: 1) optimizing sensitivity and specificity and 2) set to 90% sensitivity. Septic shock was defined as systolic hypotension and vasoactive use or greater than or equal to 30 mL/kg isotonic crystalloid administration in the first 24 hours. A model was created using 20 predictors, with an area under the receiver operating curve in the training set of 0.85 (0.82-0.88); 0.83 (0.78-0.89) in the temporal test set and 0.83 (0.60-1.00) in the geographic test set. Sensitivity and specificity varied based on cutpoint; when sensitivity in the training set was set to 90% (83-94%), specificity was 62% (60-65%). This model predicted risk of septic shock in children with suspected infection 2 hours after arrival, a critical timepoint for emergent treatment and transfer decisions. Varied cutpoints could be used to customize sensitivity to clinical context.