High physical fitness and physical activity are associated with favourable lipid levels, especially a high level of high density lipoprotein cholesterol (HDL-C). A person's skeletal muscle properties, metabolism and percentage of different muscle fibres (ST-%), which may modify coronary heart disease (CHD) risk factors, such as serum insulin, obesity and serum sex hormones may also influence his fitness level and leisure-time physical activity. We studied the associations of physical fitness, physical activity and ST-% with serum lipids and lipoproteins in 72 healthy men. Their parameters were compared with those of 20 men with defined CHD. Significant interrelationships between ST-%, fitness and leisure-time physical activity index (LTPAI) were observed. Multiple regression analysis showed that ST-%, fitness and leisure-time physical activity explained about 32% of the variation in HDL-C in the healthy men. In healthy men ST-% correlated positively with fitness (r(s) = 0.62, P < 0.001) and with LTPAI (r(s) = 0.62, P < 0.001). Fitness level also correlated significantly with LTPAI (r(s) = 0.81, P < 0.001). Serum insulin showed negative associations with ST-% (r(s) = -0.63, P < 0.001) and fitness (r(s) = -0.54, P < 0.001) and LTPAI (r(s) = -0.62, P < 0.001). Free fraction of testosterone correlated negatively with serum HDL-C level (r(s) = -0.34, P < 0.01), with fitness (r(s) = -0.41, P < 0.001) and with LTPAI (r(s) = -0.54, P < 0.001). In sedentary men with the lowest fitness and physical activity the mean of ST-% (45%) was similar to that in CHD patients (44%). However, ST-% in men in the highest tertile of physical activity and fitness (68%) was significantly higher than in CHD patients and in men in the lowest tertile of physical activity and fitness. Skeletal muscle enzyme activity in lipid metabolism was significantly lower in both CHD patients and in sedentary and low-fit men than that in fitter and physically active men. The present data imply that skeletal muscle properties are important determinants of risk profiles, such as physical activity, fitness and serum lipid and lipoprotein patterns. Although fitness is a graded, independent predictor of mortality from CHD, a relatively high fitness level is not enough. This was clearly observed in the clustering analysis, in which the healthy men, according to their ST-%, fitness, leisure-time physical activity and serum sex hormone binding globulin (SHBG), fell into three natural groups: (i) Inactive men with lowest ST-% (mean 42%), lowest fitness (10.7 METs) and lowest HDL-C (1.36 mm/l); (ii) Fit men with high ST-% (66%), high fitness (14.5 METs) and moderately high HDL-C (1.54 mol/l); (iii) Active men with high ST-% (66%), highest fitness (14.9 METs) and highest serum HDL (1.83 mmol/l). The results support the idea that both fitness and physical activity give further protection against CHD by modifying risk factors. Our findings also suggest that skeletal muscle properties should be considered in the studies which assess CHD risk factors and their modifications especially in the field of health-related fitness.