ABSTRACT
PURPOSE: We sought to determine the reproducibility of magnetic resonance imaging renography using 3 different mathematical models and 2 different approaches to convert the relative signal intensity into quantitative indices. Furthermore, we wanted to examine the influence of fluid intake on the obtained renal parameters. MATERIALS AND METHODS: A total of 10 healthy volunteers 13 to 16 years old were subjected to magnetic resonance imaging 3 times within 10 weeks, including an examination where fluid intake was increased. At each examination 0.1 mmol/kg gadolinium diethylenetriamine pentaacetic acid was administrated intravenously as a rapid bolus during a fast magnetic resonance renography sequence. Images were acquired in the coronal plan, and 1,200 images were recorded during approximately 7 minutes. Cortical data were analyzed to estimate absolute and differential function of renal parameters by converting signal intensities into quantitative units. RESULTS: Using the simple approach that a change in magnetic resonance imaging signal is linearly related to the change in gadolinium diethylenetriamine pentaacetic acid concentration, we found reproducibility in the range of 1% to 5% of all estimations of the differential renal function. The relative glomerular ultrafiltration (ml per minute per cm(3) kidney cortex) was calculated and a reproducibility of 7% was observed for relative glomerular ultrafiltration (using the model based on deconvolution). Increased hydration caused a significant change in most parameters. CONCLUSIONS: Contrast enhanced magnetic resonance renography is reproducible in the normal human kidney but excessive water intake has a significant influence on these parameters. Further studies are required to elucidate whether similar measurements can be applied to a kidney with impaired function.
