The Validity of Estimating Morphological Changes in Skeletal Muscle Using MRI in Resistance Trained Men

Resistance training (RT) is a powerful stimulus for increasing skeletal muscle volume (MV). The gold-standard method for assessing morphological changes in skeletal muscle is nuclear magnetic resonance imaging (MRI) using many serial axial images and a small slice gap between images. However, MRI can be cost prohibitive and the analysis can be highly time consuming. Scientists have attempted to use a single slice or a reduced number of slices to estimate MV. However, there is evidence that these methods may not be appropriate for estimating MV. Therefore, the purpose of this investigation was twofold. The first phase of the study was conducted to determine if a single slice MRI image is capable of estimating MV in RT men and to determine the level of error associated with using a reduced number of MRI slices when estimating MV. The second phase was conducted using a case-study analysis to determine if MV changes uniformly along the length of the muscle and if RT load has a differential effect on these results. A total of 16 previously RT men were recruited for this investigation (Age: 23.6 ± 3.4years; RT Experience: 4.9 ± 1.9 years; Height: 178.9 ± 7.4 cm; Body Mass: 86.9 ± 13.7 kg). Continuous transverse MRI images were taken with a 3 Tesla scanner and analyzed with a 5mm slice thickness and no gap. The biceps femoris short head (BFSH), biceps femoris long head (BFLH), semimembranosus (SM), and semitendinosus (ST), vastus lateralis (VL), vastus intermedius (VI), vastus medialis (VM), and the recuts femoris (RF) were analyzed. A single slice at mid-thigh for each muscle was identified and compared against whole MV with all slices using regression analysis. Bland and Altman analysis and one sample t-tests were used to determine agreement between using fewer slices to calculate MV compared to using all available slices. For the second phase, 4 subjects from the larger group were match-paired based on initial strength level and allotted to either low-load RT (LL) or high-load RT (HL) for 10 weeks of RT followed by a second MRI scan. Results showed that using a single slice MRI image at mid-thigh is a poor surrogate for predicting MV in RT men. Although, in all but one muscle, a single slice MRI image was significantly correlated with whole MV (P ≤ .05), standard errors were all very high (15.69-38.26%). Bland and Altman analysis revealed that for all muscles except for the VM and SM, using a slice gap thickness of 2 cm resulted in non-significant differences between methods (P > .05). The limits of agreement were quite small, ranging from 1% to 2.3% for a 2 cm gap. An acceptable error of 1.7% of MV was achieved by using a 3 cm gap in the VM, RF, and ST, whereas the VL, SM, and BFLH needed a smaller 2 cm gap. The VI and BFSH require a 1 cm gap to achieve less than 1.7% error. In the case study analysis, subjects in the HL group achieved greater levels of MV increases compared to the LL. No discernable pattern was found in increases of MV between muscles but there was evidence of region-specific gains in MV, but no pattern was found between groups. The results agree well with previous investigations and show that a single mid-thigh MRI slice should not be used to represent MV. Likewise, we may use larger slice gaps to estimate MV with a low level of error, and this gap varies between individual muscles of the thigh.