Diagnostic imaging of neonatal anorectal malformations: Emphasis on discordant cases and implications for surgical...

Purpose: Accurate preoperative classification of anorectal malformations (ARM) guides surgical planning, yet cross-table radiography may misclassify anatomy. We compared the diagnostic performance of transperineal ultrasound (TPUS) and radiography, with emphasis on cases where these modalities disagreed.

Retrospective review of neonates with ARM who underwent both TPUS and radiography before intervention (2018-2020). Patients in whom physical examination alone provided definitive classification were excluded. Classifications were validated against intraoperative findings. Primary outcome was diagnostic accuracy; discordance patterns were analyzed to identify which modality performed better when clinical uncertainty existed.

Among 61 neonates (17 with anatomy amenable to primary perineal repair, 26 rectourethral bulbar, 18 higher lesions by intraoperative findings), TPUS showed higher sensitivity for identifying anatomy amenable to primary repair compared with radiography (100% vs 17.6%, p < 0.001). Overall accuracy was 90.2% for TPUS versus 73.8% for radiography (p = 0.041). In 21 discordant cases, TPUS was correct in 15 (71.4%, 95% CI: 47.8-88.7%) and radiography in 5 (23.8%, 95% CI: 8.2-47.2%); one case was misclassified by both modalities. Among the 15 cases where TPUS was correct but radiography incorrect, 14 were lesions amenable to primary repair that radiography had systematically misclassified as requiring multi-stage approaches.

When TPUS and radiography disagree, TPUS shows higher accuracy, particularly for identifying anatomy amenable to primary perineal repair. Radiography tended to overestimate complexity in this diagnostically challenging subset. Incorporating TPUS into the diagnostic evaluation may support more informed surgical decision-making.

III (Retrospective diagnostic study).