Vesicoureteral reflux by contrast ultrasound, comparison with voiding and retrograde urethrocystography: A prospective accuracy study

Abstract

Diagnosis of vesicoureteral reflux (VUR) plays an important role in the management and follow-up of children presenting with urinary tract infection. This study compared voiding uronosography (VUS) as a diagnostic method for the diagnosis of VUR with the "gold standard", voiding or micturating cystourethrography (MCU). The objective of the study is not only to compare the effectiveness of both methods, but also to assess the feasibility of implementing VUS technically and economically in a large tertiary radiology center in South America. This was a prospective accuracy study that include 41 patients (a total of 85 ureteral units) aged between 1 month and 16 years. VUS was performed in an alternative form as regular MCU, by injecting microbubble contrast solution (SonoVue©) into the bladder through an intermittent catheter and evaluating the presence of reflux with ultrasound. MCU was always performed immediately after VUS. One pediatric radiologist evaluated all VUS exams, while a second independent pediatric radiologist evaluated all MCU exams, with both blinded to each other's assessments. Comparing VUS with MCU in the diagnosis of VUR, we observed a sensitivity of 92.3% (95% CI 74.9-99.1%), specificity of 98.3% (95% CI 90.9-100%), positive predictive value of 96.0% (95% CI 79.6-99.9%), negative predictive value of 96.7% (95% CI 88.5-99.6%) and a Kappa of 0.916 (0.822-1.000). When comparing the VUR grading between VUS and MCU the agreement between the two methods was high with a Kendall's Tau-b > 0.9. VUS failed to diagnose two grade 1 VUR reflux cases in teenagers while diagnosing one grade 4 VUR in a very dilated kidney-ureter unit, which was missed on MCU. VUS and MCU had similar costs when considering the price of the ultrasound contrast agent. VUS is a reliable alternative method to MCU for the diagnosis of VUR with the advantages of not exposing the child to ionizing radiation and potentially being more accessible due to the greater availability of ultrasound equipment. Limitations to this study include the usage of a single ultrasound equipment, which limits its generalizability and not evaluating the interobserver variability. Consideration should be given to implementing VUS for the diagnosis of VUR, particularly in centers without fluoroscopy equipment available, since the exam's performance and cost are comparable to those of the MCU.