MRI-based Measurement of Effects of Strength Training on Intramuscular Fat in People with and without Spinal Cord Injury

Abstract

Introduction: The accurate quantification of the proportion of fat in human muscles could help monitor disease status and test effectiveness of interventions in people with neurological conditions whose skeletal muscles are frequently infiltrated with fat.

Methods: We compared two commonly used magnetic resonance imaging methods to quantify fat in muscles. Measurements were obtained before and after 6 or 8 wk of strength training in a total of 116 muscles spanning the range of intramuscular fat proportions observed in able-bodied young adults and people with spinal cord injury.

Results: We successfully measured fat proportions in all muscles using the mDixon method but were unable to obtain plausible measurements with the T1-weighted method from muscles of able-bodied individuals or from the leaner 23% of muscles of people with spinal cord injury (muscles with less than approximately 8% fat). In muscles with more fat, measurements obtained with the two methods agreed well (intraclass correlation coefficient, 0.88; mean absolute difference, 5%). We also found that, compared with the T1-weighted method, the mDixon method provides a more detailed characterization of fat infiltration in muscle and a less variable measurement of the effect of training on the proportion of fat. The mDixon method showed that 6 or 8 wk of strength training did not appreciably change the proportion of intramuscular fat in either people with spinal cord injury or able-bodied people.

Conclusion: On the basis of these findings, we recommend the use of mDixon methods in preference to T1-weighted methods to determine the effectiveness of interventions aimed at reducing intramuscular fat.