Abstract
Background: The present study aimed to compare changes in muscle size when measured by ultrasound (MT, muscle thickness) and arm circumference (AC) using data from young men.
Methods: The investigation involved data from 3 previous studies involving a total 67 young men who performed resistance training (RT) for 10-12 weeks. Before and after the training period, elbow flexor MT was evaluated by ultrasound and AC was measured. We conducted two-stage individual patient data random effects meta-analyses using both Frequentist and Bayesian hypothesis testing. One-sample analyses examined the absence or presence of change in both MT and AC, and paired analyses examined whether these differed from one another or equivalent.
Results: One-sample analysis supported that both AC (+4.9%; t p = 0.0002; BF10 = 6,255,759,515) and MT (+3.9%; p < 0.0001; BF10 = 7,958,241,773) suggested that change in muscle size had occurred. Frequentist paired comparisons suggested that the estimates of change between both AC and MT measures did not significantly differ (p = 0.1092) but where not statistically equivalent. Bayesian paired comparisons however suggested that MT estimates where greater in magnitude than AC estimates for change in muscle size (BF10 = 16.39174).
Conclusion: Both MT and AC are able to detect RT induced changes in muscle size of the upper arm, but that the magnitude of changes may differ. Thus, care should be taken when comparing or combining estimates using either approach. Relevance for patients: The use of AC might be considered as a practical and low-cost alternative to detect changes in muscle size.
Methods: The investigation involved data from 3 previous studies involving a total 67 young men who performed resistance training (RT) for 10-12 weeks. Before and after the training period, elbow flexor MT was evaluated by ultrasound and AC was measured. We conducted two-stage individual patient data random effects meta-analyses using both Frequentist and Bayesian hypothesis testing. One-sample analyses examined the absence or presence of change in both MT and AC, and paired analyses examined whether these differed from one another or equivalent.
Results: One-sample analysis supported that both AC (+4.9%; t p = 0.0002; BF10 = 6,255,759,515) and MT (+3.9%; p < 0.0001; BF10 = 7,958,241,773) suggested that change in muscle size had occurred. Frequentist paired comparisons suggested that the estimates of change between both AC and MT measures did not significantly differ (p = 0.1092) but where not statistically equivalent. Bayesian paired comparisons however suggested that MT estimates where greater in magnitude than AC estimates for change in muscle size (BF10 = 16.39174).
Conclusion: Both MT and AC are able to detect RT induced changes in muscle size of the upper arm, but that the magnitude of changes may differ. Thus, care should be taken when comparing or combining estimates using either approach. Relevance for patients: The use of AC might be considered as a practical and low-cost alternative to detect changes in muscle size.
Original language | English |
---|---|
Article number | 4 |
Journal | Journal of Clinical and Translational Research |
Volume | 6 |
Issue number | 2 |
DOIs | |
Publication status | Published - 8 Aug 2020 |