Comparing the effects of 12 months aerobic exercise and resistance training on glucose metabolism among prediabetes phenotype: A explorative randomized controlled trial.

AIMS: The pathophysiology of each phenotype of prediabetes is unique that promotes different levels of diabetes and cardiovascular disease risks. Exercise guidelines for individuals with prediabetes including both aerobic and resistance training could improve metabolic control, but its effects on different prediabetes subtypes are unclear. The aim of this explorative randomized controlled trial was to evaluate the effects of aerobic training (AT) or resistance training (RT) on glucose metabolism and lipid profile by different prediabetes subtypes with.

METHODS: A randomized controlled trial in which 128 individuals with isolated impaired fasting glucose (i-IFG; n = 39), isolated impaired glucose tolerance (i-IGT; n = 29), combined glucose tolerance (CGI; n = 27) and isolated elevated HbA1c (n = 33) were randomly assigned to the control group, AT group and RT group, respectively. Supervised exercise training, including AT and RT were completed at moderate intensity for 60 min per day, three non-consecutive days per week for 12 months. The primary outcome was improvement in glucose metabolism. Secondary outcomes included measure of lipid profile and if these effects were moderated by the prediabetes phenotype.

RESULTS: Of the initial 128 participants, 118 finished the study, but all participants were included in the intention-to-treat analyses. The improvement in 2 h postprandial plasma glucose (2 hPG) between group difference (AT vs. RT) at 12 months was 0.87 (95% CI, -1.59 to-0.16; p < 0.05). Compared with RT group, AT significantly decreased the 2hPG in participants with i-IGT at 12 months (-1.66, 95% CI -3.04 to -0.28; p < 0.05).

CONCLUSIONS: AT program conferred benefits in improving 2 h PG and HbA1c compared with RT for prediabetes. These findings may moderate by prediabetes phenotype, and AT appeared more effective in i-IGT. A future trial with large sample size and long time follow up of prediabetes phenotype groups are needed.