Renal volumetry with magnetic resonance imaging

BACKGROUND No gold standard exists for renal volumetry in vivo. PURPOSE To devise and evaluate segmentation methods on magnetic resonance imaging (MRI) datasets. MATERIAL AND METHODS Five combinations of MRI pulse sequences and measuring methods were used to measure the renal volumes of five men aged 54-72 years scanned before autologous renal stem cell transplantation and three, six, and 12 months post transplantation. RESULTS Renal volume did not change after stem cell transplantation. The results varied considerably: the reproducibility (coefficient of variation) was 4.0-6.0% and measurements took 1-13 min per kidney. Manual segmentation of images from the volumetric interpolated breath-hold examination (VIBE) without fat saturation sequence provided best reproducibility but was time-consuming. Use of the ellipsoid formula from half Fourier acquisition single shot turbo spin echo (HASTE) provided the fastest measurement, but resulted in lower reproducibility. CONCLUSION Renal volumetry based on images from the pulse sequence VIBE without fat saturation acquired using an out-of-phase TE may be investigated further, possibly in combination with the quick ellipsoid formula.