A meta-analysis of the chronic effect of exercise on phase angle in non-athletes

Abstract

Regular exercise elicits positive health and body composition adaptations that can be measured by electrical bioimpedance analysis (bia). The phase angle from bia has been proposed as a biomarker of cellular health; however, there is a lack of consensus about the long-term effects of exercise on phase angle in non-athletic populations. Purpose: to determine the chronic effect of physical exercise on the phase angle of non-athletes. Methods: an aggregate data meta-analysis of randomized controlled trials was performed. The search included articles published in english, spanish, and portuguese using the ebscohost platform, pubmed, redalyc, science direct collection, and scopus databases, cross-referencing, and expert review. Standardized mean difference effect sizes (es) were analyzed using the inverse variance heterogeneity (ivhet) model. Non-overlapping 95% confidence intervals were considered statistically significant. The magnitude of the ess was interpreted as trivial (0-0.19), small (0.20-0.49), moderate (0.50-0.79), and large (≥ 0.80). Heterogeneity was analyzed by cochran’s q test, inconsistency between studies with the i2 test, and small-study effects by visual inspection of doi plots as well as the lfk index. Results: twenty-six effect sizes were computed from 11 studies (n = 15 experimental, n = 11 control). For the overall between-group differences, the es for the phase angle was small and non-significant (es = -0.35, q = 0.52, p = 1.00; i2 = 0%; lfk = -0.43; 95%ci = -1.01, 0.32). The es for exercise groups on the phase angle was statistically significant and considered as small to moderate (es = 0.44; q = 4.93, p = 0.99; i2 = 0%; lfk = 0.46; 95% ci = 0.28, 0.60), while for the non-exercise control groups it was also statistically significant and considered as trivial to small (es = -0.20; q = 5.15, p = 0.88; i2 = 0%; lfk = 0.45; 95%ci = -0.40, -0.01). Conclusion: exercise did not change the phase angle in non-athletes compared to controls. There was a small and significant mean effect of exercise training on the phase angle in exercising individuals from baseline.