Validation of a clinical prediction rule for pediatric abusive head trauma

Area of Science:

• Pediatric Traumatology
• Clinical Prediction Modeling
• Child Abuse and Neglect Research

Background:

• Pediatric abusive head trauma (AHT) is a significant cause of morbidity and mortality in young children.
• Previous research developed a 4-variable clinical prediction rule (CPR) with high sensitivity for AHT detection.
• Validation of this CPR in a new patient cohort is crucial for its clinical utility.

Purpose of the Study:

• To validate the screening performance of the previously derived 4-variable AHT clinical prediction rule (CPR).
• To assess the sensitivity and specificity of the AHT CPR in a new, equivalent patient population.
• To confirm the rule's effectiveness in identifying AHT in acutely head-injured children.

Main Methods:

• A prospective, multicenter, observational, cross-sectional study design was employed.
• Data from 291 acutely head-injured children under 3 years old admitted to PICUs were collected.
• The AHT CPR was applied using the same inclusion and definitional criteria as the derivation study.

Main Results:

• The 4-variable AHT CPR achieved a sensitivity of .96 in the validation cohort.
• The rule demonstrated a specificity of .46, positive predictive value of .55, and negative predictive value of .93.
• A secondary analysis indicated that the AHT CPR identified 98% of patients ultimately diagnosed with AHT.

Conclusions:

• The 4-variable AHT CPR effectively identifies AHT with high sensitivity in young, head-injured children.
• Key variables include respiratory compromise, specific bruising patterns, subdural hemorrhages, and skull fractures.
• This validated rule can aid clinicians in reducing missed diagnoses of pediatric abusive head trauma.