The RIVUR voiding cystourethrogram pilot study: experience with radiologic reading concordance

Two reference radiologists independently review voiding cystourethrograms for the National Institutes of Health sponsored RIVUR (Randomized Intervention for Children with Vesicoureteral Reflux) trial for children with vesicoureteral reflux. A pilot study was required from all clinical centers before enrolling patients.

Digital images were reviewed. Responses were compared and discrepancies adjudicated by teleconference to a final assessment.

A total of 75 studies from 19 sites were reviewed. Discrepancies in vesicoureteral reflux grade level were noted on the left and right side in 11 (15%, kappa 0.85) and 12 (16%, kappa 0.83) ureters, respectively. Other areas of disagreement were the presence of paraureteral diverticulum (left 11%, kappa 0.31; right 9%, kappa 0.34), urethral anatomy (15%, kappa 0.33), whether the child voided (8%, kappa 0.21), the presence of ureteral duplication (left 7%, kappa 0.64; right 3%, kappa 0.78) and the presence of bladder trabeculation (5%, kappa 0.32). Of 83 ureters in which reflux was seen there was grade disagreement about 23 (28%). Of 61 ureters initially assessed as grade II or III reflux by both readers, there was disagreement on 9 (15%). Of these 9 discrepancies 7 (78%) were adjudicated to the higher grade (grade III).

Discrepancies in the assessment of intermediate grade vesicoureteral reflux were noteworthy. Recommendations for patients with grade II or III reflux advanced by studies which rely on a single reading, which categorize only grade III or higher reflux as significant, may not be valid.