Aerobic exercise training-induced reductions in abdominal fat and glucose-stimulated insulin responses in middle-aged and older men.

OBJECTIVE: To test the effects of aerobic exercise training on glucose-stimulated insulin responses in middle-aged and older individuals.

DESIGN: A 9-month moderate-intensity aerobic exercise training trial in 17 men.

SETTING: An academic medical center.

INTERVENTION: Subjects walked, jogged, or cycled at 50 to 60% heart rate reserve (HRR) three times per week for 30 to 45 minutes and progressed over 6 to 9 months until subjects were training at 80 to 85% of HRR for 45 to 60 minutes three to four times per week. Training intensity was stabilized for 2 weeks before retesting. Diets were stabilized on American Heart Association Step I diets before training, and calories increased to prevent weight loss.

MEASUREMENTS: At baseline and after training, subjects underwent measurement of body fat (hydrodensitometry), regional fat distribution (waist-to-hip ratio (WHR)), VO2 max (maximal treadmill testing), diet intake (7-day food records), and glucose and insulin responses during 75 g, 2-hour oral glucose tolerance tests (OGTT) and 2-hour hyperglycemic (+/-7.9 mmol/L) glucose clamps.

RESULTS: Aerobic exercise training increased VO2 max by 15% and decreased body fat from 22.8 +/- 1.6 to 20.8 +/- 1.5% (P < .0001), waist circumference by 2% (P = .038), and WHR by 1% (P = .035). Fasting glucose and insulin levels, and glucose responses during the OGTT did not change, but insulin responses during the OGTT decreased 16% (P = .027) after training. Training reduced early (0-10 minutes) and late (20-120 minutes) phase insulin responses by 14% (P = .017 and .042, respectively), but did not significantly change glucose disposal (+8%, P = .398). Multiple regression analyses showed that changes in waist circumference (r2 = 0.68, P < .0001) and percent body fat (r2 = 0.08, P = .049) were independent predictors of the reductions in the late phase insulin responses with exercise training, however, changes in VO2 max were not (P = .199).

CONCLUSIONS: The decrease in glucose-stimulated insulin secretion with aerobic exercise training in middle-aged and older men appears to be mediated, at least in part, by reductions in the amount of abdominal fat. Regular physical exercise may prevent or ameliorate conditions associated with hyperinsulinemia including dyslipidemia, hypertension, and atherosclerosis in this group.