Cameron, D.; Clark, A.; Vermeulen, L. J.; Malekzadeh, A.; Vassiliou, V. S.; Hooijmans, M. T.

Objective: Loss of skeletal muscle mass and performance is a hallmark of ageing. Mitochondrial function has been suggested as a critical determinant of skeletal muscle performance. However, mixed results have been reported regarding mitochondrial function in older individuals. Therefore, the primary objective of this systematic review is to determine whether 31P-MRS-derived {tau}PCr, reflecting mitochondrial oxidative capacity, is reduced in ageing skeletal muscle. Methods: A preregistered systematic literature review was performed using the databases MEDLINE, EMBASE, SPORTDiscus, and Cochrane Central Register of Controlled Trials (CENTRAL). Papers were included if they reported {tau}PCr as measured by 31P-MRS; and studied individuals over 65 years of age in combination with a younger control group. Differences between young and older groups were assessed using random effects meta-analysis. Results: We included 20 papers in total, of which 2 measured 2 muscles, 5 focused on the tibialis anterior (TA) muscle, 11 on the calf muscles, 5 on the quadriceps, and 1 on the flexor digitorum longus. No statistically-significant differences were found in {tau}PCr between older and younger adults for all muscles combined (Hedges' g=0.11 (p=0.487). Inter-study heterogeneity was high ({tau}2=0.36, I2=72.49%, H2=3.64). Sub-analyses for the individual muscles showed longer {tau}PCr in the quadriceps (g=0.65, p<0.001) in older adults, but shorter {tau}PCr in the TA muscle (g=-0.64, p<0.001). For the calf muscles, no differences were detected between older and young individuals (g=0.20, p=0.377). Conclusion: No uniform age-related decline was found for {tau}PCr when comparing all studies together. Substantial heterogeneity was observed between the individual muscles, with {tau}PCr being prolonged in the upper leg muscles in older adults, but shortened in the tibialis anterior. This suggests more work using standardised settings and well-defined cohorts is needed.