Share this article on:

Fifteen-year changes in exercise, aerobic power, abdominal fat, and serum lipids in runners and controls

Medicine & Science in Sports & Exercise: January 1991


To describe effects of past as well as current exercise, aerobic power, and subcutaneous fat on the serum lipid profile, two groups of former elite athletes (N = 27 runners, N = 9 bobsledders) and a control group of normal men (N = 23) were investigated. Analysis of variance indicated a significant effect of the type of sports activity on HDL cholesterol, apolipoprotein A-I, and triglyceride levels and on the LDL/HDL cholesterol and apolipoprotein B/A-I ratios, with the most favorable values seen in runners and the least favorable values seen in controls. Of the 27 former elite runners, one third (N = 9) had given up or strongly reduced training. This subgroup showed the steepest 15-yr decrease (from 1973 to 1988) in maximum aerobic power and the largest 15-yr increase in subcutaneous fat, and the lipid profile (measured in 1988) corresponded more to the one of bobsledders and controls than to the one of runners who had remained active. Separate correlational analyses of all runners (N = 27) and nonrunners (N = 32) showed that, in both cohorts, i) the 1988 measurements of exercise, aerobic power, and subcutaneous fat were more predictive for the lipid profile in 1988 than the corresponding 1973 values, ii) anthropometric characteristics, especially abdominal fat, had a stronger relation with serum lipid concentrations than exercise and aerobic power, and iii) 15-yr changes in anthropometric characteristics were, but 15-yr changes in exercise and aerobic power were not, associated with triglyceride, lipoprotein, and apolipoprotein levels in 1988. For example, the 15-yr change in abdominal skinfolds was related, with r = 0.45 (P < 0.01), to the LDL/HDL cholesterol ratio in both runners and nonrunners. The 15-yr changes in exercise and subcutaneous fat themselves were negatively interrelated in runners (r = −0.53; P < 0.01) but unrelated in nonrunners. These results suggest that, in both the presence and the absence of an athletic predisposition, behavioral factors tend to influence the atherogenic risk of the serum lipid profile, predominantly via alterations in body composition.

©1991The American College of Sports Medicine