Factors associated with an atherogenic lipid profile in premenopausal women without clinical cardiovascular disease

BACKGROUND: To investigate different factors associated to a non desirable lipid profile in premenopausal women without cardiovascular disease. To determine the independent factors of lipid profile as a whole of the sample, for planning preventive studies. PATIENTS AND METHODS: We study (March 1994 to June 1996) premenopausal women with alcohol consumption less than 14 g/day and normal serum level of glucose. Group I: women with a non desirable lipid profile (total cholesterol [TCH, mg/dl]/high density lipoprotein cholesterol [HDL-C, mg/dl] > or = 5). Group II: with a desirable lipid profile (TCH/HDL-C < 5). The following factors were analyzed: age, body mass index (BMI), waist/hip ratio (W/H), systolic blood pressure (SBP, mmHg), fasting plasma insulin (fpI, microU/ml), cigarette smoke (CS) and presence of parents with history of non insulin dependent diabetes mellitus (NIDDM) or hypertension. Statistical methods: Mann-Whitney and Student statistics. Contingency-table analysis (chi 2 statistic). Pearson correlation and multiple linear regression. RESULTS: We analyzed 126 women (age = 30 +/- 8.2; 95% CI, 29-32; TCH = 197 +/- 36; 95% CI, 190-203 mg/dl), with 20 women (group I) and 106 (group II). Women from group I had higher values of W/H (0.83 +/- 0.04 vs 0.78 +/- 0.06; p < 0.001), BMI (29.9 +/- 9 vs 24.6 +/- 4.9; p < 0.03), fpI (12.9 +/- 10.4 vs 7.8 +/- 3.5; p < 0.05), SBP (125.9 vs 117; p < 0.02), as well as higher percentage of smokers (75 vs 40%; p < 0.01) and parents with NIDDM (60 vs 26%; p < 0.01) or hypertension (60 vs 49%; NS). No differences of age were detected (32 +/- 7.3 vs 30 +/- 8.3; NS). BMI (0.32; p < 0.01), W/H (0.50; p < 0.01), SBP (0.27; p < 0.01) and fpI (0.33; p < 0.01) were positively correlated with TCH/HDL-C ratio (n = 126). In multiple regression analysis (n = 126), W/H (regression coefficient = 6.1; 95% CI, 3.1-9.1), fpI (regression coefficient = 0.045; 95% CI, 0.018-0.072) and CS (regression coefficient = 0.5; 95% CI, 0.336-0.667) were the only independent predictors (p < 0.01) of the TCH/HDL-C ratio, controlling a 46% of the variance (R2 = 0.46). CONCLUSIONS: Our data indicates that central obesity, hyperinsulinemia and cigarette smoke are independently associated to a high risk cardiovascular lipid profile in premenopausal women without cardiovascular disease. This study suggests the importance of these factors in the management of early lipid control in these women.