To determine the interactive effects of hormones, exercise, and diet on plasma lipids and lipoproteins, serum estrogen and progesterone levels, nutrient intake, and plasma lipid, lipoprotein, and apolipoprotein concentrations were measured in 24 hypoestrogenic amenorrheic and 44 eumenorrheic female athletes. When compared to eumenorrheic athletes, amenorrheic athletes had higher levels of plasma cholesterol (5.47 +/- 0.17 vs. 4.84 +/- 0.12 mmol/L, P = 0.003), triglyceride (0.75 +/- 0.06 vs. 0.61 +/- 0.03 mmol/L, P = 0.046), low-density lipoprotein (LDL; 3.16 +/- 0.15 vs. 2.81 +/- 0.09 mmol/L, P = 0.037), high-density lipoprotein (HDL; 1.95 +/- 0.07 vs. 1.73 +/- 0.05 mmol/L, P = 0.007), and HDL2 (0.84 +/- 0.06 vs. 0.68 +/- 0.04 mmol/L, P = 0.02) cholesterol. Plasma LDL/HDL cholesterol ratios, very low-density lipoprotein and HDL3 cholesterol, and apolipoprotein A-I and A-II levels were similar in the two groups. Amenorrheic athletes consumed less fat than eumenorrheic subjects (52 +/- 5 vs. 75 +/- 3 g/day, P = 0.02), but similar amounts of calories, cholesterol, protein, carbohydrate, and ethanol. HDL cholesterol levels in amenorrheic subjects correlated positively with the percent of dietary calories from fat (r = 0.42, n = 23, P = 0.045) but negatively with the percent from protein (r = -0.49, n = 23, P = 0.017). Thus, exercise-induced amenorrhea may adversely affect cardiovascular risk by increasing plasma LDL and total cholesterol. However, cardioprotective elevations in plasma HDL and HDL2 cholesterol may neutralize the risk of cardiovascular disease in amenorrheic athletes.