Introduction: In patients with peripheral artery disease (PAD), chronic ischemia leads to claudication and muscle fatigue as a result of diminished capillary density and mitochondrial dysfunction. Noninvasive measures of perfusion are important for assessing the severity of PAD and may predict response to revascularization. The spatial heterogeneity of muscle perfusion is thought to reflect capillary density abnormality (1-3). Unlike arterial spin-labeling methods, tracer kinetic models applied to DCE MRI can quantify exercise-induced calf-muscle hyperemia, with high image contrast and spatial resolution. We test our novel method in healthy subjects under three workloads.

Methods and materials: Three healthy subjects underwent a validated plantar-flexion exercise protocol using an MRI-compatible ergometer (Fig. 1) while supine in a 3T clinical MRI scanner (TimTrio, Siemens) with two flex body coils (HIPAA compliant, IRB approved). Exercise Protocol: For each subject, the studies over two separate visits to obtain data from the right leg at 3 different loads: 4 lbs, 8 lbs and 16 lbs loads, with plantar flexion at 1 Hz for 3 min, for a work rate of 2.7, 5.4 and 10.8 watts. For each load, baseline images were acquired before exercise. Five seconds before exercise terminated, 0.05mmol/kg gadoteridol (ProHance; Bracco) was injected intravenously at a rate of 4 ml/sec, and immediately after exercise cessation, dynamic imaging started with a temporal resolution of 1.5 sec, for about 4 min. With our ergometer, subjects dorsiflex using the tibialis anterior (TA) or peroneus muscles at lower weights and plantarflex, using primarily the gastrocnemius (GC) muscle at all weights. MRI Protocol: We obtained images at three slice positions: axial slice at knee level, axial slice at calf muscle, and sagittal slice centered at knee level. A saturation-recovery prepared FLASH sequence: inversion time 300 ms, TR 527 ms, TE 1.42 ms, flip angle 15°, pixel bandwidth 1002 Hz, slice thickness 10 mm, number of averages 1, image matrix 128×128, field of view 180×180 mm. Image Analysis: To quantify gadolinium contrast concentration ([Gd]) from the signals using the saturation-recovery formula, we also estimated proton density (M₀) using the above sequence with a long TR of 4 sec, at 4 min after exercise. To quantify muscle perfusion from the dynamic images, [Gd] values in muscle and in popliteal artery were estimated from MR signals (4), and then for each muscle voxel, the temporal course of [Gd] enhancement was analyzed by a tracer kinetic model (5) to estimate muscle perfusion (F) on a voxel-by-voxel basis. Regions of interest (ROI) were manually drawn to get averaged perfusion in each muscle group, including gastrocnemius, soleus and peroneal muscles.

Results: Representative FLASH images from the post-exercise (16 lbs) images at baseline and maximum enhancement (25 sec following contrast injection, Fig 2) show that gastrocnemius and peroneal muscles enhanced visibly, with mild enhancement of the soleus. Fig. 3 shows examples of [Gd] arterial input function sampled from popliteal artery and [Gd] enhancement curves from a voxel of gastrocnemius and of soleus. Fig. 4a shows examples of muscle perfusion maps of a same subject but at different exercise loads. Averaging over the subjects, perfusion of the different muscle groups at different exercise loads are shown in Fig. 4b. As expected, perfusion in gastrocnemius increased almost linearly from 4 lbs to 8 lbs, then to 16 lbs. The spatial maps of perfusion illustrate that at mid-level of work, the gastrocnemius showed heterogeneous enhancement, with the most perfused muscles regions reaching 60-80 ml/min/100g while at higher workload homogeneous perfusion at about 80 ml/min/100g.

Discussion: DCE-MRI of exercise-induced hyperemia, analyzed using a tracer kinetic model, provides high spatial resolution perfusion maps of the calf muscle. With increased load, plantar flexion stimulated increasing perfusion in gastrocnemius of healthy subjects, and the pattern of hyperemia is notable for the heterogeneity of perfusion at low work load which became homogeneous at high work load, as predicted in the literature (1-3). In conclusion, post exercise DCE-MRI offers a very promising technique for quantitatively assessing regional calf-muscle perfusion. As a future study, we will test the value of the technique in evaluating muscle function in PAD patients.