Risk Prediction After a Brief Resolved Unexplained Event.

Area of Science:

• Pediatrics
• Clinical Decision Making
• Emergency Medicine

Background:

• Brief Resolved Unexplained Events (BRUE) are common in infants, but serious underlying pathologies are rare (4%).
• Current American Academy of Pediatrics (AAP) guidelines may lead to unnecessary investigations and hospitalizations as they do not effectively differentiate high-risk patients.
• There is a need for improved risk stratification tools for infants presenting with BRUE.

Purpose of the Study:

• To derive and validate a clinical decision rule for predicting the risk of a serious underlying diagnosis or event recurrence in infants with BRUE.
• To compare the performance of a new risk prediction model against existing AAP guidelines.

Main Methods:

• Retrospective analysis of 3283 infants presenting with BRUE across 15 children's hospitals (2015-2020).
• Logistic regression was used in a split-sample approach to derive and validate a risk prediction model.
• Performance was evaluated using sensitivity, specificity, and area under the curve (AUC).

Main Results:

• 17.2% of infants had a serious underlying diagnosis or recurrent event.
• The AAP higher-risk criteria had high sensitivity (95.3%) but low specificity (8.6%) for serious diagnoses (AUC=0.52).
• A derived model incorporating age, previous events, and medical history showed improved prediction (AUC=0.64). Infants >60 days had a higher risk of serious diagnosis (OR=1.43).

Conclusions:

• The majority of infants with BRUE do not have serious underlying conditions.
• Two novel models were developed to predict the risk of serious diagnosis and recurrence.
• These models can serve as decision support tools to aid clinicians and caregivers in discussions regarding diagnostic testing and hospitalization for BRUE.