Voiding Cystourethrography Revisited: Descriptive Statistics for the Detection of Vesicoureteral Reflux

Abstract

To determine which factors might influence the detection of vesicoureteral reflux (VUR), we retrospectively reviewed factors including fluoroscopy time, number of true radiographic acquisitions, and patient characteristics from a large number of voiding cystourethrography (VCUG) examinations. Nine hundred eighty-seven VCUG examinations performed between March 2006 and March 2009 were randomly selected for review. Data recorded were presence of VUR, patient age and sex, examination indication, follow-up status, presence of a diagnostic radiology trainee, fluoroscopy time, and number of true radiographic acquisitions. For initial examinations, descriptive comparison and logistic analyses were performed. To evaluate which variables related to reflux identification, we analyzed variables by logistic regression after stratifying by patient age (≤ 1 or > 1 year) for both the full sample and cases grouped by VUR severity (grades I and II [mild] or grades III-V [moderate to severe]). Nine hundred eighty-seven VCUG examinations were evaluated for the study (65.5% female; mean age, 3 years 2 months; age range, 2 weeks-16 years), and 761 cases met the inclusion criteria. VUR was detected in 101 of 349 infants (40 mild, 61 moderate to severe) and in 107 of 412 children older than 1 year (52 mild, 55 moderate to severe). A wide range of fluoroscopy times was similar between both positive and negative cases (0.033-4.233 minutes). The number of true radiographic acquisitions differed significantly between the negative and positive cases. Descriptive statistics and logistic regression analyses for a large number of VCUG examinations in a pediatric population are summarized. Our results showed that high fluoroscopy times were not associated with a higher likelihood of VUR. There may be a small benefit to the use of true radiographic acquisitions for detecting VUR.