Figure 4 Mean workload and standard deviation in milliseconds (ms), thru the Mixed Effect Model, for each exercise. Vastus Medialis (VM), Vastus Lateralis (VL) and Gluteus Medius (GM). Table 1 Mean (SD) ��T2 in milliseconds (ms) during each intervention and selleck inhibitor mean difference (95% CI) between exercises for each muscle Discussion The use of MRI to quantify muscle function has been extensively investigated in the last three decades, but the exact mechanism for changes in T2 is not known (Saab et al., 2000). In principle, many factors could contribute, including increase in intracellular and extracellular water content, accumulation of diamagnetic ions (eg. lactate, phosphate, sodium) and decrease in pH (Jenner et al., 1994; Meyer and Prior, 2000).

Some authors had demonstrated a direct relationship of the increase in T2 with the muscle perfusion and volume, but Prior et al. (2001) had documented the increase of T2 in rat muscle even with the occlusion of the vein and electric stimulation (Ploutz-Snyder et al., 1997). Although Cheng et al. (1995) observed that the recovery of muscle pH after exercise is faster than the recovery of T2. The Relaxometry method had good applicability, as demonstrated by Yue et al. (1994), only five repetitions of arm curl with a load of 25% of 1 Repetition Maximum were enough to increase the T2 of the short head of biceps brachii, thus proving the sensitivity of the method. In the same way Adams et al. (1992) could find differences in the increase of the T2 of the concentric and eccentric muscle contraction.

Previous studies provided evidence of a linear association between the ��T2 and the exercise intensity suggesting an index of muscle workload (Yue et al., 1994; Meyer and Prior, 2000). Based in those reports and in our data, it is possible to suggest the type of squat exercise that optimally workload the muscles responsible for stabilizing the patellofemoral joint and the knee dynamic Q angle. The squat exercise presented a significant higher workload of vastus medialis in relation to vastus lateralis, suggesting that the squat exercise presented less risk of patellofemoral maltracking among the 3 squats used in this work. The hypothesis that the contraction of the hip adductor muscles would favor the vastus medialis in the squat associated with isometric hip adduction exercise was not confirmed by our data.

There was no workload difference between vastus medialis and vastus lateralis, suggesting that squat associated with isometric hip adduction does not selectively recruit the vastus medialis muscle. But our hypothesis that the gluteus medius workload decrease with the association of isometric hip adduction was confirmed. Squat associated with isometric hip abduction was the only exercise that presented a balance of vastus medialis, vastus lateralis and gluteus Cilengitide medius muscles, despite a gluteus medius workload similar to squat exercise.