Noninvasive measurement of absolute renal perfusion by contrast medium-enhanced magnetic resonance imaging.

Abstract

The aim of this study was to validate the quantification of absolute renal perfusion (RP) determined by dynamic magnetic resonance imaging (MRI) and contrast media using an experimental model in the rabbit and a transit-timed ultrasound flow probe around the left renal artery as comparison. An MR-compatible ultrasonic time-of-flight flow-probe was placed around the left renal artery in 9 New Zealand white rabbits. Absolute RP in basal state, after mechanical renal artery stenosis, intravenous dopamine, angiotensin II, or colloid infusion was measured using dynamic MRI and intravenous injection of gadoteridol. The results were correlated to the renal artery flow measured inside the magnet with the transit-timed flow-probe. For the signal intensity concentration conversion, we applied different calibrations according to various velocities measured in the aorta by a phase contrast sequence to correct for inflow effect. MRI-derived RP (in mL/min) was calculated by the maximum upslope method, where RP/volume was defined as the ratio of the cortex contrast enhancement slope over the maximum of the arterial input function determined in the aorta. Reproducible arterial and renal transit curve with excellent contrast to noise ratio were obtained. The MRI derived perfusion was systematically underestimated by comparison to the ultrasonic transit-timed flow-probe but was linearly correlated with these measures (r = 0.80, P < 0.001). Using a flow-sensitive calibration, an accurate arterial input function can be measured from the blood MR signal and used in a realistic model to assess the RP. There was a good correlation between the MR-derived RP and the renal artery blood flow measured by the flow-meter. This experimental study validates absolute RP quantification by MRI and contrast media injection and justifies further clinical studies.