Comparison of magnetic resonance urography with ultrasound studies in detection of fetal urogenital anomalies

Abstract

Purpose Prenatal ultrasonography detects the vast majority of urogenital anomalies, but in some cases the diagnosis remains in doubt. We assessed the potential of magnetic resonance urography (MRU) in the evaluation of different urogenital anomalies in fetuses when ultrasound study was equivocal. Patients and methods We retrospectively reviewed the medical records of 46 fetuses in whom the presence of urogenital anomalies was suspected at ultrasound studies, but remained inconclusive. Fetal MRU was performed within the same week as ultrasound studies. All patients underwent MRU, comprising overview, fast, thick-slab, heavily T2-weighted sequences, followed by focused, high-resolution T2-weighted sequences obtained in sagittal, axial and coronal planes. T1-weighted sequences were obtained in selected cases for assessment of the gastrointestinal tract. All MRU results were compared with ultrasound findings. Sensitivity of each imaging modality was estimated based on definite diagnoses made after birth or abortion. Results The mean (range) gestational age was 27 (18–36) weeks. The final diagnosis was ureteropelvic junction obstruction in 12, ureteral dilation (due to vesicoureteral junction obstruction) in five, ureterocele in five, posterior urethral valve in 16, multicystic dysplastic kidney in six, mesenteric cyst in one and abdominoscrotal hydrocele in one. Overall diagnostic sensitivity of fetal MRU was 96% compared to sonography with 58% sensitivity ( p < 0.05). Fetal MRU studies provided additional information to sonography in 17 (37%) cases, and were especially more sensitive in evaluation of ureteral anatomy. Conclusions Fetal MRU can accurately diagnose a wide variety of urinary tract disorders and must be regarded as a valuable complementary tool to ultrasound in the assessment of the urinary system, particularly in cases of inconclusive ultrasound findings. The present study had a selection bias, as only fetuses with possible anomalies proposed by sonography were referred for MRU; however, this is the population that probably benefits most from MRU studies.