Determination of skeletal muscle perfusion using arterial spin labeling NMRI: Validation by comparison with venous occlusion plethysmography

T(1)-based determination of perfusion was performed with the high temporal and spatial resolution that monitoring of exercise physiology requires. As no data were available on the validation of this approach in human muscles, T(1)-based NMRI of perfusion was compared to standard strain-gauge venous occlusion plethysmography performed simultaneously within a 4 T magnet. Two different situations were investigated in 21 healthy young volunteers: 1) a 5-min ischemia of the leg, or 2) a 2-3 min ischemic exercise consisting of a plantar flexion on an amagnetic ergometer. Leg perfusion was monitored over 5-15 min of the recovery phase, after the air-cuff arterial occlusion had been released. The interesting features of the sequence were the use of a saturation-recovery module for the introduction of a T(1) modulation and of single-shot spin echo for imaging. Spatial resolution was 1.7 x 2.0 mm and temporal resolution was 2 s. For data analysis, ROIs were traced on different muscles and perfusion was calculated from the differences in muscle signal intensity in successive images. To allow comparison with the global measurement of perfusion by plethysmography, the T(1)-based NMR measurements in exercising muscles were rescaled to the leg cross-section. The perfusion measurements obtained by plethysmography and NMRI were in close agreement with a correlation coefficient between 0.87 and 0.92. This indicates that pulsed arterial techniques provide determination of muscle perfusion not only with superior spatial and temporal resolution but also with exactitude.