Magnetic resonance voiding cystography in the diagnosis of vesicoureteral reflux: Comparative study with voiding cystourethrography

Purpose: To evaluate the feasibility of magnetic resonance voiding cystography (MRVC) compared with voiding cystourethrography (VCUG) for detecting and grading vesicoureteral reflux (VUR). Materials and Methods: MRVC was performed upon 20 children referred for investigation of reflux. Either coronal T1-weighted spin-echo (SE) or gradient-echo (GE) (fast multi-planar spoiled gradient-echo (FMPSPGR) or turbo fast low-angle-shot (FLASH)) images were obtained before and after transurethral administration of gadolinium solution, and immediately after voiding. The findings of MRVC were compared with those of VCUG and technetium-99m (Tc-99m) dimercaptosuccinic acid (DMSA) single-photon emission computed tomography (SPECT) performed within 6 months of MRVC. Results: VUR was detected in 23 ureterorenal units (16 VURs by both methods, 5 VURs by VCUG, and 2 VURs by MRVC). With VCUG as the standard of reference, the sensitivity of MRVC was 76.2%; the specificity, 90.0%; the positive predictive value, 88.9%; and the negative predictive value, 78.3%. There was concordance between two methods regarding the grade of reflux in all 16 ureterorenal units with VUR detected by both methods. Of 40 kidneys, MRVC detected findings of renal damage or reflux nephropathy in 13 kidneys, and Tc-99m DMSA renal SPECT detected findings of reflux nephropathy in 17 kidneys. Conclusion: Although MRVC is shown to have less sensitivity for VUR than VCUG, MRVC may represent a method of choice offering a safer nonradiation test that can additionally evaluate the kidneys for changes related to reflux nephropathy.