PURPOSE: To evaluate the accuracy and feasibility of magnetic resonance voiding cystourethrography (MRVCUG), with or without contrast medium, in detecting vesicoureteral reflux (VUR) by comparison with conventional voiding cystourethrography (VCUG). MATERIALS AND METHODS: Seventy-five patients underwent a series of 55 indirect MRVCUG (I-MRVCUG) and 61 direct MRVCUG (D-MRVCUG) between 2003 and 2009. We retrospectively compared the results from I-MRVCUG and D-MRVCUG with those from VCUG on 150 kidney-ureter units. Ratios of successful completion of the two types of MRVCUG were analyzed in 116 examinations according to sex, age, and among the two groups, with or without sedation. RESULTS: D-MRVCUG was superior to I-MRVCUG in detecting VUR (sensitivity: 96.8% vs. 76.9%; specificity: 96.3% vs. 88.7%; agreement: 96.6% vs. 83.7%; kappa: 0.83 [95% confidence interval (CI): 0.72, 0.94] vs. 0.55 [95% CI: 0.41, 0.69]). The overall ratio of successful completion of the two types of MRVCUG was 76.7% (89/116): there was no significant difference between I-MRVCUG and D-MRVCUG. The successful completion rate was significantly lower in MRVCUG in toddlers compared with preschoolers, infants, schoolchildren, and adults (P < 0.001). CONCLUSION: The two types of MRVCUG (I-MRVCUG and D-MRVCUG) are promising tests without radiation exposure. Both I-MRVCUG and D-MRVCUG are feasible for children except for toddlers.
PURPOSE: To evaluate the accuracy and feasibility of magnetic resonance voiding cystourethrography (MRVCUG), with or without contrast medium, in detecting vesicoureteral reflux (VUR) by comparison with conventional voiding cystourethrography (VCUG). MATERIALS AND METHODS: Seventy-five patients underwent a series of 55 indirect MRVCUG (I-MRVCUG) and 61 direct MRVCUG (D-MRVCUG) between 2003 and 2009. We retrospectively compared the results from I-MRVCUG and D-MRVCUG with those from VCUG on 150 kidney-ureter units. Ratios of successful completion of the two types of MRVCUG were analyzed in 116 examinations according to sex, age, and among the two groups, with or without sedation. RESULTS: D-MRVCUG was superior to I-MRVCUG in detecting VUR (sensitivity: 96.8% vs. 76.9%; specificity: 96.3% vs. 88.7%; agreement: 96.6% vs. 83.7%; kappa: 0.83 [95% confidence interval (CI): 0.72, 0.94] vs. 0.55 [95% CI: 0.41, 0.69]). The overall ratio of successful completion of the two types of MRVCUG was 76.7% (89/116): there was no significant difference between I-MRVCUG and D-MRVCUG. The successful completion rate was significantly lower in MRVCUG in toddlers compared with preschoolers, infants, schoolchildren, and adults (P < 0.001). CONCLUSION: The two types of MRVCUG (I-MRVCUG and D-MRVCUG) are promising tests without radiation exposure. Both I-MRVCUG and D-MRVCUG are feasible for children except for toddlers.