Postnatal imaging of prenatally detected hydronephrosis-when is voiding cystourethrogram necessary?

Abstract

To evaluate whether grade 4-5 vesicoureteral reflux (VUR) can be predicted from renal ultrasound (RUS) findings and perform voiding cystourethrograms (VCUGs) only on high-risk patients. The RUS and VCUG images of infants with prenatally detected hydronephrosis admitted to our institution between 2003 and 2013 were re-evaluated. The UTI episodes were collected retrospectively from patient journals. Patients with complex urinary tract anomalies were excluded. One hundred eighty, 44 female and 136 male, patients (352 renal units (RU)), 23 (30 RU) of them having grade 4-5 VUR, were included. The median age of the patients at the time of the RUS was 1.3 (0.1-3.0) months and the median follow-up time was 2.0 (0.1-11.2) years. In multivariate analysis, a visible ureter (OR 12.72; CI 5.33-32.04, p < 0.001) and shorter renal length (OR 2.67; CR 1.50-4.86, p < 0.001) in RUS predicted grade 4-5 VUR while a visible ureter predicted UTIs (OR 5.75; CI 2.59-12.66, p < 0.001). A three-grade risk score for high-grade VUR was developed based on the RUS findings and the patients were categorized into low-, intermediate-, and high-risk groups. The incidence of grade 4-5 VUR was 2.9% in the low-risk, 12.2% in the intermediate-risk, and 52.2% in the high-risk group. The sensitivity and specificity for detecting grade 4-5 VUR were 79 and 82%, respectively. In patients with antenatally detected hydronephrosis, a visible ureter and reduced renal length in RUS are significant risk factors for high-grade VUR. A RUS-based risk scoring would probably reduce the proportion of unnecessary VCUGs.