Purpose:

The non-invasive measurement of post-exercise phosphocreatine (PCr) recovery kinetics using 31-phosphorus magnetic resonance spectroscopy (³¹P-MRS) is a highly prevalent method for in vivo, non-invasive assessment of skeletal muscle energetics. However, ³¹P-MRS methodology is notoriously laboratory-specific, leading to uncertainty about the normal range of PCr recovery kinetics among healthy individuals, as well correlations with disease and demographic factors. Therefore, our aim was to systematically characterize the normal range of PCr recovery measurements from ³¹P-MRS in human skeletal muscles, as well as correlations with age and end-of-exercise pH, among healthy adults and those with muscle-related diseases.

Methods:

PubMed, Web of Science, Cochrane, and Google Scholar databases were searched for articles. Cross sectional, prospective, case report and experimental studies that used isometric, isotonic, and isokinetic muscle flexion to measure the forearm, upper leg, and lower leg muscles, and were published in English, were included. Inclusion required an exercise challenge paradigm targeting skeletal muscle phosphocreatine recovery kinetics among healthy or diseased individuals across the lifespan. Studies must have measured PCr recovery kinetics using ³¹P-MRS.
The search was conducted in March 2023. In total, 2287 relevant studies were initially identified, and 1567 remained after screening for duplicates and off-topic titles. Abstract screening resulted in 185 potentially relevant studies, and screening the full text resulted in a final set of 128 studies that qualified for meta-analysis. These studies were categorized according to the three different muscle groups from which PCr kinetics were measured: forearm, upper leg, and lower leg.
Statistical analysis of mean PCr recovery rates from each study was performed using Stata 17 software. Meta-analysis was conducted using forest plots to determine the pooled summary PCr recovery rate within each muscle group. Random effect analysis was used to weigh each study to control for potential heterogeneity [1]. The mean difference statistic was used for effect size calculations. Single group mean analysis was performed for studies exclusively involving healthy participants whereas two group mean analysis was used for comparative studies involving disease groups and control groups. Inter-study heterogeneity was assessed using the Cochran Q-statistic, and the extent of this heterogeneity was quantified using the I² statistic [2]. In cases where the Q-statistic showed significance, a Galbraith plot was utilized to identify studies responsible for the most significant heterogeneity. Publication bias was examined using Egger’s test and funnel plots. Results of statistical tests were considered significant if P ≤ 0.05.

Results and Discussion:

A PRISMA flowchart [3] of the study selection process is provided in Figure 1. Eighty-six studies included healthy individuals only, while forty-two included individuals with muscle-related diseases. Forest plots showed significant heterogeneity across mean PCr recovery time estimates initially, and outlier studies were identified using Galbraith and funnel plots (see, e.g., Figure 2 for forearm muscle studies among healthy individuals). After removing outlier studies, the pooled estimate of PCr recovery time among healthy adults was 34.41 ± 8.74 s, 35.66 ± 2.17 s, and 33.75 ± 2.82 s, in the forearm, upper leg, and lower leg muscles, and the studies were homogeneous (Q = 10 to 45, I² = 0 to 2.6%, tau² = 0 to 2.3 and P > 0.05). In comparative studies, the overall mean difference in PCr recovery time between disease and control groups was 25.18 ± 11 s, 5.38 ± 2.14 s, and 16.74 ± 4.91 s (Figure 3), in forearm, upper leg and lower leg muscles, and heterogeneity among studies was low to moderate (Q = 7.5 to 19, I² = 0 to 45.08 % and tau² = 0 to 27.18, 0.04 < P < 1.0) after outlier studies were removed. Greater age was associated with longer PCr recovery in upper leg muscles among both healthy (0.387, P < 0.05) and diseased (0.733, P < 0.05) individuals (Table 1). Additionally, longer PCr recovery time was significantly correlated with more acidic end-of-exercise pH in all three muscle groups among healthy individuals.

Conclusions:

Mitochondrial oxidative capacity as indexed by 31P-MRS based PCr recovery time is similar across three different skeletal muscle groups among healthy people measured at a variety of different MRI centers. Common diseases significantly prolong PCr recovery times. Although these values are numerically higher among those with common muscle-related diseases, their effect sizes were not significant in the overall meta-analysis until outlier studies were removed from analysis. Greater age and more acidic pH increase PCr recovery time.

Acknowledgements

No acknowledgement found.