Comparison views for subtle physeal injury in the pediatric ankle

This study addresses whether physicians should obtain routine comparison views to detect subtle fractures. Twenty-two cases of Salter-Harris I fractures of the fibula and 16 cases of nonfracture ankle injury were viewed by 26 radiology and orthopedic physicians. For each case, participants made a diagnosis and indicated whether they were highly certain, moderately certain, or uncertain in that diagnosis for the injured side alone and then for the injured side by comparison with the uninjured side (comparison view). Overall film reading accuracy was 43% without comparison views vs 65% with comparison views; accuracy for nonphyseal injuries was 65% vs 88%, accuracy for physeal injuries was 27% vs 49% (all p < 0.0001, X² test). Comparison views increased certainty scores by 0.26 on a 2-point scale (p = 0.0002, t-test). Accuracy and certainty in the diagnosis of ankle injuries improved significantly with comparison views, supporting the opinion that comparison views are valuable for the accurate detection of subtle physeal injury.