Diagnosis of noncalcareous hydronephrosis: role of magnetic resonance urography and noncontrast computed tomography.

OBJECTIVES: To evaluate the role of magnetic resonance urography (MRU) and noncontrast computed tomography (NCCT) in the diagnosis of noncalcareous hydronephrosis when excretory urography (intravenous urography) is either contraindicated or inconclusive.

METHODS: A total 108 consecutive patients with noncalcareous hydronephrosis were included in this study. In all patients, intravenous urography was either contraindicated or could not determine the diagnosis. In all patients, calculus obstruction was excluded by NCCT and all underwent heavily T2-weighted MRU. The final definitive diagnosis was established by retrograde or antegrade ureterography, endoscopy, or open surgery and was considered the reference standard for the diagnosis of obstruction. Normal kidneys in patients with unilateral obstruction were considered the reference standard for the absence of obstruction. The results of MRU were compared with those of NCCT regarding sensitivity, specificity, and overall accuracy.

RESULTS: Of the 108 patients, 5 had bilateral obstruction and the remaining 103 had unilateral obstruction. Of the latter group, 5 had a solitary kidney; therefore, the total number of renal units was 211 (113 obstructed and 98 normal units). Ureteral strictures were identified by NCCT in 15 (28%) of 54 and by MRU in 45 (83%) of 54 patients. Bladder, ureter, or prostate tumors causing ureteral obstruction could be diagnosed in one half of the 54 patients with such tumors by NCCT (27 of 54) and in all but 2 patients by MRU (52 of 54). Both NCCT and MRU could identify all extraurinary causes of obstruction. Overall, of the 113 kidneys with noncalculus obstruction, the cause could be identified by MRU in 102 (sensitivity of 90%) and by NCCT in 47 (sensitivity of 42%), a difference of statistically significant value in favor of MRU (P <0.001). The specificity of T2-weighted MRU and NCCT was 100% and 99%, respectively (not a statistically significant difference). The overall accuracy of T2-weighted MRU and NCCT was 95% and 68%, respectively (P <0.001).

CONCLUSIONS: In patients with ureteral obstruction in whom intravenous urography is not helpful and after NCCT has excluded stone disease, heavily T2-weighted MRU is a sensitive and specific method in the identification of the cause of obstruction.