Risk Prediction After a Brief Resolved Unexplained Event

Abstract

OBJECTIVES Only 4% of brief resolved unexplained events (BRUE) are caused by a serious underlying illness. The American Academy of Pediatrics (AAP) guidelines do not distinguish patients who would benefit from further investigation and hospitalization. We aimed to derive and validate a clinical decision rule for predicting the risk of a serious underlying diagnosis or event recurrence. METHODS We retrospectively identified infants presenting with a BRUE to 15 children’s hospitals (2015–2020). We used logistic regression in a split-sample to derive and validate a risk prediction model. RESULTS Of 3283 eligible patients, 565 (17.2%) had a serious underlying diagnosis (n = 150) or a recurrent event (n = 469). The AAP’s higher-risk criteria were met in 91.5% (n = 3005) and predicted a serious diagnosis with 95.3% sensitivity, 8.6% specificity, and an area under the curve of 0.52 (95% confidence interval [CI]: 0.47–0.57). A derived model based on age, previous events, and abnormal medical history demonstrated an area under the curve of 0.64 (95%CI: 0.59–0.70). In contrast to the AAP criteria, patients >60 days were more likely to have a serious underlying diagnosis (odds ratio:1.43, 95%CI: 1.03–1.98, P = .03). CONCLUSIONS Most infants presenting with a BRUE do not have a serious underlying pathology requiring prompt diagnosis. We derived 2 models to predict the risk of a serious diagnosis and event recurrence. A decision support tool based on this model may aid clinicians and caregivers in the discussion on the benefit of diagnostic testing and hospitalization (https://www.mdcalc.com/calc/10400/brief-resolved-unexplained-events-2.0-brue-2.0-criteria-infants).