Association of lipoprotein(a), insulin resistance, and reproductive hormones in a multiethnic cohort of pre- and perimenopausal women (The SWAN Study)

UMMS Affiliation

Department of Medicine, Division of Preventive and Behavioral Medicine

Publication Date


Document Type



Adult; *African Continental Ancestry Group; *Asian Continental Ancestry Group; Body Mass Index; Cardiovascular Diseases; China; Climacteric; Cross-Sectional Studies; Estradiol; *European Continental Ancestry Group; Female; Follicle Stimulating Hormone; Hispanic Americans; Humans; *Insulin Resistance; Japan; Lipoprotein(a); Middle Aged; Premenopause; Prospective Studies; Sex Hormone-Binding Globulin; Testosterone; Thyrotropin; United States; Women's Health


Life Sciences | Medicine and Health Sciences | Women's Studies


Because evidence suggests lipoprotein (Lp(a)) may be an important cardiovascular risk factor in women, we evaluated whether reproductive hormones may influence Lp(a) concentrations and insulin resistance in a large multicenter study of women transitioning to postmenopause. Data were taken from the Study of Women's Health Across the Nation (SWAN), a prospective multiethnic study of menopausal transition (1,368 Caucasians, 808 African-Americans, 220 Chinese, 216 Hispanic, and 251 Japanese). Blood was assayed for Lp(a), follicle stimulating hormone, estradiol (E2), testosterone, and sex hormone binding globulin (free estradiol index and free androgen index), thyroid stimulating hormone, and glucose and insulin (to calculate insulin resistance). African-American women had twofold higher mean Lp(a) values than women of the other 4 race/ethnic groups, adjusted for body mass index (BMI). Lp(a) was modestly correlated with E2 (r = 0.10) in women without self-reported diagnosed heart disease but not in women with self-reported heart disease. Lp(a) was positively associated with log insulin resistance in women without self-reported heart disease but not in women with self-reported heart disease, an association that was not significant after adjusting for ethnicity, BMI, smoking, age, and site. Race/ethnicity, particularly being African-American, accounted for most of the explained Lp(a) variations. Lp(a) was very modestly associated (r = -0.04) with insulin resistance after adjusting for ethnicity and BMI, and this association was not modified by reproductive hormones, including androgens.

DOI of Published Version



Am J Cardiol. 2003 Sep 1;92(5):533-7.

Journal/Book/Conference Title

The American journal of cardiology

Related Resources

Link to article in PubMed

PubMed ID