Relation of caffeine intake to blood lipids in elderly women.

The relation of the consumption of caffeine from coffee, tea, and all caffeine-containing soft drinks (hereafter referred to as "cola") to blood lipid levels was studied in 1,035 white women ranging in age from 65 to 90 years (mean, 71.2 years) from October 1986 through October 1988. All study subjects were participants in the Pittsburgh, Pennsylvania, clinic of the Study of Osteoporotic Fractures. Fasting blood samples were tested for total cholesterol, triglycerides, apolipoproteins A-I and B-100, total high-density lipoprotein (HDL) cholesterol, and the HDL2 and HDL3 subfractions. Low-density lipoprotein (LDL) cholesterol was calculated using the Friedewald equation. Current consumption of coffee, tea, and cola was assessed by means of a self-administered personal habits questionnaire. Potential confounders, such as body mass index (weight (kg)/height (m)2), waist/hip ratio, smoking status, and alcohol intake were also measured for each woman. Analyses of variance and tests for mean differences revealed an inverse relation between the consumption of tea and apolipoprotein B-100 and a positive association between the consumption of cola and apolipoprotein B-100. Apolipoprotein A-I levels were found to be positively related to coffee consumption and negatively related to tea consumption. There was no consistent relation between caffeine consumption and total cholesterol, LDL cholesterol, total HDL cholesterol and its subfractions, or apolipoprotein B-100. Adjustment for potential confounders yielded no remaining significant associations between caffeine from any of the major beverage sources and any of the lipid fractions. In conclusion, the inconsistent relations between caffeine from various sources and blood lipids do not support a significant association of blood lipid levels with caffeine consumption in elderly women.