Association between Polycystic Ovaries and Extent of Coronary Artery Disease in Women Having Cardiac Catheterization

Coronary artery disease is the most common cause of death among women in developed countries. Established risk factors for coronary disease in women include smoking [1], high serum cholesterol levels [2], low high-density lipoprotein (HDL) cholesterol levels [3], diabetes, and hypertension [2]. Premenopausal women have a lower risk for heart disease than do postmenopausal women [4]. Polycystic ovaries are seen in 22% of women [5, 6] and are associated with hirsutism, infertility, and menstrual disturbances [7]. Insulin resistance [8], higher serum triglyceride levels, and lower HDL cholesterol levels [9] commonly occur in women with polycystic ovaries; these factors may be associated with a higher incidence of coronary artery disease [10, 11]. The metabolic syndrome of hypertension, diabetes, and central obesity (the Stein-Leventhal syndrome) is also associated with polycystic ovaries [12]. We sought to determine whether women with more extensive coronary artery disease (as seen on coronary angiography) are more likely to have polycystic ovaries appearing on ultrasonography than are women with less extensive coronary disease. Methods All women 60 years of age or younger who had had coronary angiography in Auckland, New Zealand, in the preceding 2 years were invited to participate in this study. Women who had had bilateral oophorectomy were excluded. The local ethics committee approved the study. Women who were menstruating were studied between days 5 and 9 of the menstrual cycle. Data were collected on infertility (failure to conceive within 1 year), menstrual cycle irregularity (>a 4-day variation), hypercholesterolemia (requiring medication), hypertension (requiring medication), and a family history of heart disease in a first-degree female relative younger than 60 years of age or in a first-degree male relative younger than 55 years of age. Hirsutism was assessed on the Ferriman-Gallwey scale [13]. Ultrasonography was done prospectively, without knowledge of the coronary angiography results. In consenting patients (71.3%), a transvaginal ultrasound was also done. Polycystic ovaries were defined by modified Adams criteria [14] as eight or more cysts that were between 2 and 8 mm in diameter and were associated with an increase in ovarian stroma. Chest pain (96% of women) or valvular disease (4% of women) was assessed by angiography after administration of nitroglycerin spray. Quantitative angiography was done using a computerized automatic analysis system (Cardiovascular Measurement Systems, Neunen, the Netherlands). The extent of coronary artery disease was evaluated by adding the number of segments containing stenoses of more than 50% diameter according to the Coronary Artery Surgery Study scoring system [15]. A secondary end point was severity of coronary artery disease. Patients were given a score on the basis of luminal diameter narrowing according to the following scale: 25% luminal diameter narrowing, 1 point; 26% to 50%, 2 points; 51% to 75%, 4 points; 76% to 90%, 8 points; 91% to 99%, 16 points; and 100%, 32 points. Each stenosis was multiplied by the appropriate score. Functional significance was evaluated by the Coronary Artery Surgery Study [15] weighting system, which accounts for the severity and location of stenoses. Continuous variables were analyzed using one-way analysis of variance. Discrete variables were analyzed using chi-square tests. Multiple logistic regression analysis was done using the presence or absence of polycystic ovaries as the outcome variable. Explanatory variables include those listed in Table 1 and Table 2 and the extent of coronary artery disease. Explanatory variables were entered into the model in a stepwise fashion if they had a P value of 0.15. These variables included age, extent of coronary artery disease, total cholesterol level, HDL cholesterol level, presence of diabetes, family history of heart disease, and history of smoking. Table 1. Characteristics of Women with Normal Ovaries and Women with Polycystic Ovaries* Table 2. Laboratory Findings in Women with Normal Ovaries and Women with Polycystic Ovaries* Results Of 207 women, 143 (69%) agreed to participate. One or more ovaries were visualized in 99.3% of the study women either by transabdominal (77.6%) or transvaginal (94%) ultrasonography. Neither ovary was visualized in one woman, and her results were excluded. Clinical characteristics are shown in Table 1. Polycystic ovaries were diagnosed by ultrasonography in 42.3% of women. There was no difference in the prevalence of polycystic ovaries between premenopausal women and postmenopausal women (25 of 48 women and 35 of 94 women, respectively; P = 0.13). The differences in ovaries seen by ultrasonography were present in both premenopausal (follicle-stimulating hormone level < 20 IU/L) and postmenopausal (follicle-stimulating hormone level > 20 IU/L) women [16]. One woman had previously received a diagnosis of the polycystic ovary (Stein-Leventhal) syndrome. Compared with women with normal ovaries, women with polycystic ovaries had higher triglyceride levels (P < 0.01) and lower HDL cholesterol levels (P < 0.05), although more women with polycystic ovaries were receiving lipid-lowering medications (P = 0.01). Compared with women who had normal ovaries, women with polycystic ovaries had higher free testosterone levels (P < 0.001) and a trend for increased C-peptide levels (P = 0.06) (Table 2). These differences all persisted when the data were stratified according to whether the women were premenopausal or postmenopausal [16]. Women with polycystic ovaries had more coronary artery segments with greater than 50% diameter loss (1.7 segments [95% CI, 1.1 to 2.3 segments]) than did women with normal ovaries (0.82 segments [CI, 0.54 to 1.1 segments]) (P < 0.01). Women with polycystic ovaries had a higher functional significance score (1767 [CI, 1123 to 2412]) than did women with normal ovaries (1029 [CI, 709 to 1349]) (P = 0.05) and had a trend toward a higher severity score (39.5 [CI, 29.3 to 49.7] and 28.7 [CI, 22.2 to 35.2], respectively; P = 0.06). Forty-nine women with normal ovaries (59.8%) and 32 women with polycystic ovaries (53.3%) did not have coronary artery stenoses of 50% or more (P > 0.2). By multivariate logistic regression analysis, the extent of coronary artery disease was found to be independently associated with the presence of polycystic ovaries (P = 0.032), as was family history of heart disease (P = 0.022). If the extent score increased by 1, the odds ratio increased by approximately 26%. Discussion Our study has three important findings. First, in women having coronary angiography, those with more extensive coronary artery disease were more likely to have polycystic ovaries appearing on sonography than were women with less extensive coronary disease. Second, polycystic ovaries were identified in postmenopausal women. Third, the diagnosis of polycystic ovaries seen on ultrasonography was associated with specific metabolic and endocrine abnormalities. Women with polycystic ovaries have been shown to have risk factors for coronary disease [8-11]. However, a direct association between polycystic ovaries and angiographic coronary artery disease has not been previously reported. Diagnosis of polycystic ovaries in postmenopausal women by ultrasonography has only recently been described [16]. Our finding of a high incidence of polycystic ovaries relates to the patients studied, all of whom had had coronary angiography. The women with polycystic ovaries had high testosterone and triglyceride levels, low HDL cholesterol levels, and some insulin resistance. Women with polycystic ovaries seen on ultrasonography often do not have the clinical features of the polycystic ovarian syndrome. The disorder probably exists on a continuum ranging from no symptoms to the fully developed syndrome. Why should polycystic ovaries appear on ultrasonography more frequently in women with more extensive coronary artery disease than in women with less extensive disease? Polycystic ovaries are the most common cause of anovulation with reduced production of estradiol. Decreased estradiol levels may be associated with coronary artery disease, shown by the reduced incidence of heart disease in postmenopausal women receiving estrogen replacement therapy [17]. Although estrone levels are elevated in women with polycystic ovaries, estrone is less potent than estradiol and may not be cardioprotective. Insulin resistance could be the link between polycystic ovaries and coronary artery disease. We found that C-peptide levels, a stable indicator of insulin production, were higher in women with polycystic ovaries than in women with normal ovaries. Abnormal lipid levels may also account for the increase in significant coronary artery disease. In addition, women with increased testosterone production may have a risk for adverse cardiac events similar to that of men. Hirsutism has also been reported as a risk factor for coronary artery disease in women [18]. What are the implications of our findings? Young women with polycystic ovaries often present with hirsutism, acne, infertility, or menstrual irregularity. Once the diagnosis of polycystic ovaries has been made, lifestyle modifications may be indicated to decrease cardiovascular risk. Further research must be done to investigate the natural history and to determine whether any of the current treatments for polycystic ovaries, such as the oral contraceptive pill, antiandrogens, ovulatory agents, or laparoscopic ovarian surgery, are beneficial. From National Women's Hospital, University of Auckland, and Green Lane Hospital, Auckland, New Zealand. Dr. Farquhar: National Women's Hospital, Epsom, Auckland 1003, New Zealand. Dr. White: Cardiology Department, Green Lane Hospital, Epsom, Auckland 1003, New Zealand.