The relationship of physical activity to high density lipoprotein cholesterol in postmenopausal women.

The relationships of physical activity to total HDL, HDL-2 and HDL-3 cholesterol were examined in 255 white postmenopausal women, mean age, 57.6 years. Physical activity was measured by the Paffenbarger Index (a composite index of walking, stair climbing and sports in the past week); by a modified Paffenbarger which included only sport activities; and by an objective activity monitor (LSI). Depending on which activity measure was used, different univariate relationships between physical activity and the HDL-cholesterol subfractions were noted. Physical activity, as measured by sport kcal/week or by objective monitors was related to HDL-2 cholesterol [r = 0.22 (p less than 0.01)] and [r = 0.19 (p less than 0.01)], respectively, but not to HDL-3 cholesterol. Physical activity, as measured by the Paffenbarger kcal/week was only related to HDL-3 [r = 0.15 (p less than 0.05)]. Multiple regression analyses revealed that sport activity was significantly related to total HDL and HDL-2 cholesterol. Activity as measured by objective monitors was not independently related to either HDL-total or HDL-2. Activity as measured by the Paffenbarger kcal/week was an independent determinant of HDL-3, after controlling for body fatness. Insulin, caloric intake and liver function were analyzed as possible underlying mechanisms, but we could not demonstrate any mediating or interacting effect on any of these mechanisms for the association of HDL-cholesterol to physical activity.