The Association of Left Ventricular Mass with Cardiovascular Risk Factors in African American Women KATHERINE SHERIF, MD; MICHAEL BARRETT, MD; HARVEY KUSHNER, PHD; BONITA FALKNER, MD ABSTRACT: Background: African American women have disproportionately high rates of myocardial infarc- tion and stroke. Left ventricular hypertrophy is an inde- pendent risk factor for cardiovascular disease. Increases in left ventricular mass (LVM) may precede the expres- sion of hypertension. The purpose of this study was to determine whether LVM is related to cardiovascular risk variables in healthy, premenopausal African American women. Methods: Normotensive or borderline hyper- tensive nondiabetic African American women (N = 52; mean age, 31 years) underwent anthropometric and blood pressure measurements, oral glucose tolerance test, euglycemic clamp, fasting lipid profile, and two- dimensional echocardiography. LVM was calculated by the cube root formula and adjusted for height [LVM index (LVMI)]. Results: LVMI correlated with body mass C ardiovascular disease is a major cause of mor- bidity and mortality in women. Compared with white women, Mrican American women have dis- proportionately higher rates of myocardial infarc- tion and stroke.1,2 Increasing left ventricular mass (LVM) is an independent risk factor for cardiovas- cular disease. 3 Left ventricular hypertrophy (LVH) in Mrican American women confers a higher rela- tive risk for death than for Mrican American men. 4 In women with clinically defined hypertension, sys- tolic blood pressure is a direct correlate of L VM5 as well as an important predictor of LVH in both hy- pertensive and borderline hypertensive women. 6 ,7 Although LVH may be a sequela of chronic hyper- tension, the contribution of other risk factors to an From the Department of Medicine, MCP-Hahnemann School of Medicine (KS, MB, BF) and the Biomedical Computer Research Institute (HK), Philadelphia, Pennsylvania. Submitted March 8, 1999; accepted in revised form December 2, 1999. This work was supported in part by Grants NL51547 and DI46107 from the National Institutes of Health. Correspondence: Katherine Sherif, M.D., Department of Medicine, Room 1281C, MCP Hahnemann University, 3300 Henry Avenue, Phil- adelphia, PA 19129 (E-mail: katherine.sheri!@drexel.edu). THE AMERICAN JOURNAL OF THE MEDICAL SCIENCES index (r = .36, P = 0.009), systolic blood pressure (r = .44, P = 0.001), diastolic blood pressure (r = .43, P = 0.002), and central body fat (r = .42, P = 0.002). LVMI also directly correlated with lipoprotein (a) (r = .34, P = 0.02). Significant independent relationships of other metabolic variables with LVMI were not detected. Dis- cussion: These data show that increased LVMI is asso- ciated with body mass index and central obesity, but not with lipids, insulin resistance, or insulin sensitivity. LVMI is also associated with blood pressure before the expression of severe hypertension in healthy, premeno- pausal African American women. KEY INDEXING TERMS: Blood pressure; Left ventricular mass; African American women; Obesity [Am J Med Sci 2000;320(1): 13-7.] increase in LVM in women is not resolved. The purpose of this study was to determine whether LVM in premenopausal Mrican American women is related to risk factors for cardiovascular disease, including level of blood pressure, body size, and metabolic parameters before clinically significant hypertension. Methods Study Population. The study was conducted on 54 clinically well, nondiabetic African American women. Each participant was drawn from a cohort that has been under study in investigations of blood pressure and cardiovascular risk factors. Subjects in the cohort consisted of African American men and women aged 30 to 35 years who have participated in prospective studies since ado- lescence. s Participants enrolled in this study included normoten- sive patients, subjects with high normal blood pressure, and untreated stage I hypertensives9 based on repeated measure- ments of blood pressure. Exclusion criteria included a known diagnosis of diabetes, history of irregular menses, and use of hormonal contraceptive. All studies were conducted during the follicular phase of the menstrual cycle. The metabolic procedures were conducted on all cohort subjects. A sample of women from the cohort was randomly selected to undergo two-dimensional echocardiography. Written informed consent was obtained from each subject on an institutionally approved protocol. All labora- tory visits and procedures of each patient were completed within 6 to 8 weeks of enrollment. 13