In a recent meta-analysis of controlled clinical trials (11 randomized and three nonrandomized) on the effects of exercise training on glycemic control in individuals with type 2 diabetes, Boule and colleagues39 reported a significant (P < 0.001) beneficial exercise effect on glycosylated hemoglobin (HbA1c) (-0.66 percent) compared to the controls. The authors concluded that exercise training reduces HbA1c by an amount that should decrease the risk of diabetic complications. The difference found in this meta-analysis was close to the difference (-0.9 percent) between conventional and intensive glucose-lowering therapy reported in the United Kingdom Prospective Diabetes Study (UKPDS)40,41; an amount that was associated with significant improvement in clinical outcomes (development of microvascular and macrovascular complications of diabetes, including cardiovascular disease). The authors went on to speculate that exercise might result in a greater reduction in cardiovascular complications than with insulin or sulfonylureas since exercise has additional cardioprotective benefits and does not cause weight gain.
We are not aware of a meta-analysis of the effects of exercise training in nondiabetics, most likely because HbA1c, a good, long-term measure of glucose control, is not often measured in nondiabetic subjects. Neither are we aware of a meta-analysis of the effects of exercise on fasting blood glucose in any population. There are a number of individual exercise studies in normal individuals that report fasting glucose1142 and HbA1c43 are significantly improved with exercise training. Unfortunately, none of these studies had a control group. Additionally, there are many exercise-training studies in normal individuals where no decrease in fasting glucose is observed, presumably because these individuals already have normal and healthy fasting glucose levels44 (unpublished observations).
In a comprehensive review by Ivy et al.,45 several lines of evidence are presented that support a key role of exercise and physical activity in prevention and treatment of diabetes. As diabetes is defined by fasting glucose, prevention of diabetes is defined by the prevention of increases in fasting serum glucose. Ivy and colleagues reviewed numerous epidemiological studies (cross-sectional, retrospective, and prospective designs) that provide strong support for the beneficial effect of physical activity in the prevention of type 2 diabetes. Studies on the population of Mauritius showed that the relationship between low levels of physical activity and increased prevalence of impaired fasting glucose and type 2 diabetes existed across both sexes and across the major ethnic groups on the island46-48 and that total physical activity was a significant, independent predictor even after controlling for several major confounders. Mayer-Davies et al.49 reported similar findings in a large, culturally and ethnically diverse sample of men and women. Kriska and colleagues reported the same relationship in Pima Indians.50 Manson et al.,5152 in a prospective study, provided additional evidence of this relationship, including evidence of a dose-response effect of increased exercise frequency and reduced risk of type 2 diabetes. Ivy et al. presented additional evidence in their review of bed-rest studies53-56 and exercise-detraining studies,57-59 which together revealed a rapid deterioration in insulin action and glucose tolerance (if the individual/population was glucose intolerant) with even short periods of bed rest or detraining.
But, does exercise training improve glucose tolerance and prevent development of type 2 diabetes? Exercise training reliably improves insulin action in all subjects but does not affect glucose tolerance in normal individuals. However, in glucose-intolerant individuals, exercise does result in improvements in glucose tolerance. In a seven-day exercise-training study in ten men with mild type 2 diabetes or impaired glucose tolerance, Holloszy's group60 found that two of three who had impaired glucose tolerance had a normal oral glucose-tolerance test after training. Of the seven men with type 2 diabetes, three had normal oral glucose-tolerance tests after training, two had only impaired fasting glucose, and two still had diabetic oral glucose-tolerance tests after seven days of training. The two who still had diabetic oral glucose-tolerance tests had relative hypoinsulinemia, whereas the other eight that improved with exercise all had mild to moderate hyperinsulinemia. The data suggest that when sufficient insulin reserve exists, exercise training has the potential to reverse glucose intolerance and even mild type 2 diabetes.
Results from cross-sectional studies of young and old endurance athletes compared to young and old (lean and not lean) untrained individuals show that untrained individuals have decreased insulin action and decreased glucose tolerance compared to young and old athletes.61 The master athletes, compared to young athletes, had essentially identical insulin and glucose responses to an oral glucose-tolerance test, indicating that age need not result in impaired glucose tolerance. The authors suggested that earlier studies that had shown no improvement in insulin action or glucose tolerance generally used an insufficient exercise stimulus, waited too long after the final exercise bout, and used patients with relative insulin deficiency. When this group trained individuals for one year who were mildly diabetic without insulin deficiency and measured oral glucose-tolerance tests within 18 hours of the last bout of exercise, they found significantly improved glucose tolerance and insulin action and that fasting glucose levels were normalized.62 Data from Reitman et al.63 reported 6-10 weeks of intensive exercise lowered fasting glucose and improved glucose tolerance in obese type 2 diabetic individuals. Other studies in type 2 diabetic subjects support these findings.64,65 While the totality of the data are quite compelling, it is clear that a large, controlled, randomized study of the effects of exercise training in mild type 2 diabetic subjects would provide more definitive evidence for the potential reversal of type 2 diabetes with exercise training.
One cannot discuss the effects of exercise training on progression to diabetes in individuals without mention of the results of the Diabetes Prevention Program.66 In this study, the effects of pharmacologic therapy (metformin) and lifestyle interventions (exercise training at the level of ACSM/CDC recommendations of 30 minutes per day most days of the week, diet, and weight loss for a total of 7 percent body weight) were compared with usual care. The results revealed that the lifestyle intervention reduced the risk of progression to diabetes in this population by 58 percent compared with usual care. The effects of metformin, although statistically and clinically significant, were less impressive in reducing rate of progression to diabetes, which it did by 31 percent compared to usual care. Granted, this trial did not study the effects of exercise alone, but it points out the utility of lifestyle interventions in individuals with metabolic syndrome.
Was this article helpful?