Abstract
Context:
Women with polycystic ovary syndrome (PCOS) have a high prevalence of cardiovascular disease (CVD) risk factors including dyslipidemia. Lipoproteins are heterogeneous, and measurement of serum lipids provides only the size of the pool and does not predict their function or composition. Recently, high-density lipoprotein cholesterol (HDL-C) function, as determined by cholesterol efflux capacity from macrophages, has been shown to be an independent predictor of subclinical CVD.
Objective:
The aim of the study was to comprehensively evaluate lipoprotein profile including lipid particle size and number and cholesterol efflux capacity in PCOS to better define CVD risk.
Design and Setting:
A case control study was performed at an academic PCOS center.
Patients:
Women with PCOS (n = 124) and geographically matched controls (n = 67) were included in the study.
Main Outcome Measures:
The primary outcome was to measure HDL-C efflux capacity by an ex vivo system involving the incubation of macrophages with apolipoprotein (Apo) B-depleted serum from subjects, and the secondary outcome was to measure lipid particle size and number using nuclear magnetic resonance spectroscopy.
Results:
Women with PCOS had significantly higher body mass index and blood pressure but similar HDL-C and low-density lipoprotein cholesterol levels compared to controls. The mean ApoA1 levels were lower, and the ApoB/ApoA1 ratio was higher in PCOS subjects compared to controls (P < .01). There were no differences in ApoB levels. Women with PCOS had an 7% decrease in normalized cholesterol efflux capacity compared to controls (P < .003). Cholesterol efflux capacity in PCOS correlated with body mass index, ApoA1, HDL-C, and the presence of metabolic syndrome. In a multivariable regression model, PCOS was significantly associated with diminished cholesterol efflux. PCOS was also associated with an atherogenic profile including an increase in large very low-density lipoprotein particles, very low-density lipoprotein (VLDL) size, and small low-density lipoprotein cholesterol particles (P < .01).
Conclusions:
Our novel findings of decreased cholesterol efflux and an atherogenic lipid particle number and size pattern in women with PCOS, independent of obesity, further substantiate the increased risk of CVD in this population.
Polycystic ovary syndrome (PCOS) is one of the most common endocrine disorders in reproductive-age women. In addition to having gynecological and dermatological manifestations, these women are at an increased risk for cardiovascular disease (CVD) (1). The latter is evident from an increased prevalence of traditional CVD risk factors such as dyslipidemia, glucose intolerance, diabetes, obesity, and hypertension. In addition, the prevalence of metabolic syndrome is increased at a young age in PCOS (2). Longitudinal studies in this population demonstrate a persistent increase in risk factors such as diabetes, dyslipidemia, and hypertension over a 20-year period (3). Several studies have also demonstrated an increase in subclinical atherosclerosis measured by coronary artery calcium scores, carotid artery intima media thickness, and endothelial dysfunction in PCOS (4). A few studies have reported that perimenopausal and postmenopausal women with symptoms suggestive of PCOS have a shorter cardiovascular event-free survival as compared to controls (5, 6). Collectively, these data support the notion that PCOS is associated with significant cardiometabolic morbidities starting in early reproductive life.
Insulin resistance has been implicated in the underlying pathophysiology of PCOS. The characteristic dyslipidemic profile (high triglycerides [TGs] and low high-density lipoprotein-cholesterol [HDL-C]) associated with insulin resistance is the most common metabolic abnormality in young women with PCOS. In the Framingham Heart Study, women in the lower quartiles of HDL-C had a significantly increased risk of myocardial infarction over a 12-year follow-up period (7). Most successful trials of lipid-modulating therapy to prevent coronary heart disease (CHD), however, have evaluated therapies to lower low-density lipoprotein cholesterol (LDL-C). It has been questioned whether low serum HDL-C is merely a bystander or is causal in CVD (8). HDL-C has several functions, including inhibition of LDL-C oxidation, transport of cholesterol from peripheral cells to liver, antiapoptotic effects, antithrombotic effects, and antioxidant effects. However, measurement of serum HDL-C provides only the size of the HDL-C pool and does not predict the function or composition of HDL-C. HDL-C and apolipoprotein (Apo) A-1, a major protein constituent of HDL-C, play a key role in reverse cholesterol transport by promoting cholesterol efflux from peripheral cells, such as cholesterol-laden macrophages, and subsequently delivering acquired cholesterol to the liver for excretion (9). Recently, measurement of cholesterol efflux capacity from macrophages, a metric of HDL-C function, has been demonstrated to be a stronger predictor of CVD than HDL-C levels (10), strengthening the argument that measures of HDL-C function may be more useful than HDL-C levels as predictors of CVD risk. Compared to the conventional lipid panel, detailed lipid phenotyping by nuclear magnetic resonance (NMR) may provide assessment of lipid particle number and size, and these measures have been shown to be helpful in better characterizing dyslipidemia.
Although women with PCOS commonly have elevated TGs, when assessed by the conventional lipid panel, total cholesterol, HDL-C, and LDL-C are not consistently altered. We therefore performed comprehensive lipid phenotyping in well-defined women with PCOS compared to geographically matched controls. Our objectives were to assess HDL-C function, by measuring cholesterol efflux capacity, and to obtain detailed lipoprotein profiles by measuring lipid particle size and number by NMR spectroscopy in women with PCOS and controls. We hypothesized that the hyperandrogenic and insulin-resistant states in PCOS would reduce cholesterol efflux and shift the particle composition to a more atherogenic profile.
Subjects and Methods
Subjects
The University of Pennsylvania Institutional Review Board approved this study, and all participants gave written, informed consent. Women with PCOS between the ages of 18 and 50 years were recruited. The diagnosis of PCOS was based on evidence of the Rotterdam criteria, and all subjects had clinical or biochemical hyperandrogenism and oligomenorrhea (11). All PCOS patients had TSH, prolactin, dehydroepiandrosterone sulfate, and 17-α hydroxyprogesterone levels measured to rule out other conditions that might mimic symptoms of PCOS. None of the subjects had existing CVD or current symptoms of CVD such as shortness of breath and chest pain. Control subjects were recruited from a random sample of healthy women seen for an annual examination at the Gynecology Clinic. All controls had regular menstrual cycles and no complaints of hirsutism or acne. Other exclusion criteria for all subjects included pregnancy, lactation, hysterectomy, menopause, and chronic illnesses such as asthma and inflammatory bowel disease.
Methods
The blood samples drawn from women with PCOS and controls were immediately centrifuged, and serum was aliquoted and stored at −70°C until batch analyzed. Total cholesterol, TGs, HDL-C, and glucose were measured using standard enzymatic methods. LDL-C was mathematically derived. ApoA1 and ApoB were measured using an immunoturbidimetric assay (Roche Diagnostics). Insulin, leptin, adiponectin, and T were measured by RIA using human specific kits (EMD Millipore), and human resistin was measured by ELISA (EMD Millipore). The interassay and intra-assay coefficients of variation for all measured biochemical parameters were < 5%. Insulin resistance was calculated by the homeostasis model assessment (HOMA) using the formula: glucose (mg/dL) × insulin (mU/L)/405. Smoking status was defined on the basis of self-reported current or past smoking. Body mass index (BMI) was calculated as kilograms per meter squared. Metabolic syndrome was defined as the presence of three of the following five criteria: TG ≥ 150 mg/dL, HDL-C ≤ 50 mg/dL, blood pressure (BP) ≥ 130/85 mm Hg, BMI ≥ 30 kg/m2, and fasting glucose ≥ 100 mg/dL (12).
HDL-C efflux was measured by a validated ex vivo system involving the incubation of macrophages with ApoB-depleted serum from subjects. A total of 774 cells derived from a murine macrophage cell line were plated and radiolabeled with 74 kBq of 3H-cholesterol per milliliter. ABCA1 was up-regulated by means of a 6-hour incubation with 0.3 mm 8-(4-chlorophenylthio)-cAMP. Subsequently, efflux mediums containing 2.8% ApoB-depleted serum were added for 4 hours. To prepare ApoB-depleted serum, samples were thawed before ApoB precipitation. Briefly, 40 parts polyethylene glycol solution (20% PEG 8000 MW in 200 mm glycine buffer, pH 7.4) was added to 100 parts serum, mixed by pipetting, and then incubated at room temperature for 20 minutes before spinning in a microcentrifuge at 10 000 rpm for 30 minutes at 4°C. ApoB-containing lipoproteins are pelleted by this procedure, and the supernatant, which contains the HDL fraction, is recovered and diluted in 14 mm MEM-HEPES (no bicarbonate) + 0.15 mm cAMP to 2.8% (equivalent to 2% serum). All steps were performed in the presence of the acyl-coenzyme A cholesterol acyltransferase inhibitor CP113,818 (2 μg/mL). Liquid scintillation counting was used to quantify the efflux of radioactive cholesterol from the cells. The quantity of radioactive cholesterol incorporated into cellular lipids was calculated by means of isopropanol extraction of control wells not exposed to patient serum. Percentage efflux was calculated by the following formula: [(becquerels of 3H-cholesterol in mediums containing 2.8% ApoB-depleted serum − becquerels of 3H-cholesterol in serum-free mediums) ÷ becquerel of 3H-cholesterol in cells extracted before the efflux step] × 100. All assays were performed in duplicate. The interassay and intra-assay coefficients of variation for the HDL efflux assay were < 10%. Lipoprotein particle concentration and diameters were measured using an automated NMR (13).
Statistical analysis
Continuous variables were summarized (means ± SD), and graphical methods were employed to characterize distributions and skewness. The Mann-Whitney test was used to compare continuous variables across groups, and Spearman's coefficient was calculated to determine correlations between continuous variables. Associations between categorical variables were tested using χ2 tests. Multivariable linear regression analysis was performed to examine the relationship between HDL efflux and PCOS, adjusting for relevant covariates. Logistic regression was used to examine the relationship between metabolic syndrome, HDL-C efflux, and other lipids while controlling for PCOS and other relevant covariates. A priori power calculations assumed a sample size of 191 subjects, α error of 5%, β error of 20%, and an estimated cholesterol efflux SD equal to twice the difference in mean efflux between groups (PCOS and controls). Based on these assumptions, the study was sufficiently powered to detect a difference of at least 0.08 in efflux between groups. P < .05 was considered statistically significant. All statistical tests were performed using STATA 12 software (StataCorp).
Results
We analyzed samples on a total of 124 subjects with PCOS and 67 controls (Table 1). The PCOS subjects were younger, with significantly higher total T levels and fewer menses compared to controls (P < .01). No significant differences were noted in the two groups for a history of hypertension, diabetes, smoking, and family history of coronary artery disease. Family history of type 2 diabetes was significantly higher in women with PCOS (P < .01). On comparing cardiometabolic risk factors, women with PCOS had an adverse CVD risk profile including significantly higher BMI, higher systolic and diastolic BP, and more use of antihypertensive medications (Table 2). There was no difference in the mean fasting glucose, insulin, or HOMA values in the two groups. The adipokine adiponectin was significantly lower, and leptin levels were significantly higher in PCOS subjects after controlling for BMI and age (P < .01). The prevalence of metabolic syndrome in women with PCOS was significantly higher compared to controls (P < .01), again reflecting an increased risk for diabetes and CVD.
Table 1.
Demographic Features of Subjects With PCOS and Controls
| Demographics | PCOS | Controls |
|---|---|---|
| n | 124 | 67 |
| Age (mean), y | 28.2 ± 6.1 | 31.6 ± 6.7a |
| Age range, y | 18–50 | 19–49 |
| Race, % | ||
| White | 69.8 | 69.4 |
| Black | 20.6 | 20.7 |
| Asian | 6.3 | 2.3 |
| Other | 4.0 | 2.3 |
| No. of menses per year | 3.98 ± 3.4 | 11.69 ± 2.1a |
| Total T, ng/dL | 53.2 ± 43.4 | 31.3 ± 17.1a |
| Family history of coronary artery disease, % | 19.3 | 16.4 |
| Family history of type 2 diabetes, % | 37.1 | 19.4a |
Data are presented as mean ± SD unless specified otherwise.
P < .01.
Table 2.
Cardiometabolic Risk Factors in Subjects With PCOS and Controls
| PCOS | Controls | |
|---|---|---|
| n | 124 | 67 |
| BMI, kg/m2 | 33.1 ± 7.6 | 29.0 ± 7.1a |
| <25 | 17.5% | 38.8% |
| 25–30 | 17.5% | 13.4% |
| >30 | 65.1% | 47.7% |
| Systolic BP, mm Hg | 124.4 ± 12.4 | 108.3 ± 13.6a |
| Diastolic BP, mm Hg | 72.1 ± 7.8 | 67.3 ± 8.3a |
| Current hypertensive therapy | 8/125 (6.4%) | 2/65 (3%) |
| Smoker current (any amount) | 12/125 (9.6%) | 1/65 (1.5%) |
| Glucose, mm | 0.43 ± 0.05 | 0.46 ± 0.04 |
| Insulin, pm | 103.41 ± 108.46 | 100.63 ± 53.44 |
| HOMA | 4.4 ± 6.4 | 4.2 ± 4.8 |
| Adiponectin, nm | 0.81 ± 0.56 | 1.08 ± 0.52a |
| Leptin, pm | 1.94 ± 1.1 | 1.33 ± 1.03a |
| Resistin, pm | 943.92 ± 487.35 | 882.36 ± 528.39 |
| Metabolic syndrome prevalence | 34.6% | 15.7%a |
Data are presented as mean ± SD, unless specified otherwise. To convert values for adiponectin to ng/mL, multiply by 26.385. To convert values for leptin to ng/mL, multiply by 16.67. To convert values for resistin to ng/mL, divide by 51.3. To convert values for glucose to mg/dL, multiply by 200. To convert values for insulin for μIU/mL, multiply by 0.14.
P < .01.
Lipoprotein profiles
Next, we performed detailed lipid phenotyping and compared differences in PCOS and controls (Table 3). Women with PCOS had higher TG levels compared to control subjects (P < .01). However, there were no differences in both HDL-C and LDL-C levels between the groups. The mean ApoA1 levels were significantly lower, and the ApoB/ApoA1 ratio was higher in cases compared to controls (P < .01). ApoA1, ApoB, and ApoB/ApoA1 correlated with BMI categories (overweight BMI, 25–30 kg/m2; obese BMI, ≥30 kg/m2; P < .05). ApoA1 inversely correlated with HOMA, an indicator for insulin resistance, and total T (P < .02), whereas ApoB/ApoA1 positively correlated with HOMA and total T (P < .01). Serum nonesterified fatty acid (NEFA) levels were significantly increased in PCOS subjects compared to controls (P < .01) and correlated with total T levels (P < .01).
Table 3.
Lipid Profile, Apo Levels, and HDL-C Function in PCOS Subjects and Controls
| PCOS | Controls | |
|---|---|---|
| n | 124 | 67 |
| Total cholesterol, mm | 0 ± 0.98 | 4.9 ± 0.89 |
| HDL-C, mm | 1.42 ± 0.42 | 1.49 ± 0.46 |
| LDL-C, mm | 4.32 ± 1.32 | 4.0 ± 1.13 |
| TG mm | 1.66 ± 1.05 | 1.27 ± 0.79a |
| Lipid-lowering therapy | 2/125 (1.6%) | 1/65 (1.5%) |
| ApoA1, g/L | 1.61 ± 0.38 | 1.74 ± 0.35a |
| ApoB, g/L | 0.84 ± 0.23 | 0.79 ± 0.19 |
| ApoB/ApoA1 | 0.55 ± 0.2 | 0.47 ± 0.16a |
| NEFA, mm | 0.6 ± 0.3 | 0.43 ± 0.2a |
| HDL functionb | ||
| Cholesterol efflux capacity | 0.98 ± 0.17 | 1.05 ± 0.17a |
Data are presented as mean ± SD, unless specified otherwise. To convert values for cholesterol to mg/dL, multiply by 38.61. To convert values for HDL-C to ng/dL, multiply by 38.61. To convert values for LDL-C to mg/dL, multiply by 38.61. To convert values for TG to mg/dL, multiply by 88.496. To convert values for ApoA1 to mg/dL, multiply by 100. To convert values for ApoB to mg/dL, multiply by 100.
P < .01.
PCOS, n = 115; controls, n = 56.
HDL-C function measured by cholesterol efflux assays
Because both groups had similar mean HDL-C levels, we examined HDL-C function by performing cholesterol efflux assays. Women with PCOS had a 7% decrease in normalized cholesterol efflux capacity compared to controls (P < .003; Table 3). As expected, ApoA1 and HDL-C were the strongest predictors of increased cholesterol efflux capacity (P < .001), whereas BMI, fasting insulin, HOMA, and total T were all associated with decreased efflux capacity (P < .01). In a multivariable regression model including HDL-C and BMI as covariates, PCOS was associated with significantly diminished cholesterol efflux (β, −0.05; 95% confidence interval [CI], −0.1, −0.009; P < .02). The effect of PCOS on cholesterol efflux was not modified by race or ethnicity. Although adiponectin levels correlated with cholesterol efflux (P < .001), in the multivariable model including HDL-C and BMI, addition of adiponectin did not alter the relationship between PCOS and cholesterol efflux. Cholesterol efflux was lower in all women with metabolic syndrome (median value, 0.93; interquartile range, 0.83–1.02) than in those without metabolic syndrome (median value, 1.01; interquartile range, 0.88–1.14; P < .02).
Lipid particle number and size by NMR spectroscopy
We next used NMR spectroscopy to better phenotype lipid particle concentration and size (Table 4). There was a significant increase in large very low-density lipoprotein (VLDL) in women with PCOS, contributing to the corresponding increase in VLDL particle size. Despite similar calculated mean LDL-C levels, women with PCOS had higher total LDL particle numbers compared to controls (P < .02), and these were primarily composed of small LDL particles. The total HDL-C was increased in PCOS with more small HDL-C particles (P < .01), again resulting in smaller HDL particle size. After adjusting for age and BMI, PCOS was significantly associated with large VLDL particles, VLDL size, and small LDL-C particles (P < .01). To test the hypothesis about whether insulin resistance or androgens are predictors of an atherogenic lipid profile, we analyzed correlations between HOMA and total T and the lipid parameters. Fasting insulin and HOMA correlated with large VLDL (P < .05), whereas T levels negatively correlated with total HDL (P < .02).
Table 4.
Lipoprotein Subclass Particle Concentration and Lipoprotein Size Using NMR Spectroscopy
| PCOS | Controls | P Value | |
|---|---|---|---|
| n | 124 | 67 | |
| Particle concentration | |||
| Total VLDL and chylomicrons, nm | 51.44 ± 24.16 | 45.73 ± 17.56 | .2 |
| Large VLDL and chylomicrons, nm | 4.04 ± 3.7 | 2.37 ± 1.73a | .001 |
| Medium VLDL, nm | 18.36 ± 12.64 | 15.27 ± 7.62 | .39 |
| Small VLDL, nm | 29.55 ± 14.54 | 29.16 ± 14.19 | .79 |
| Total LDL, nm | 1067.76 ± 391.75 | 919.57 ± 300.34b | .02 |
| IDL, nm | 201.01 ± 125.31 | 253.85 ± 181.39 | .18 |
| Large LDL, nm | 200.94 ± 143.27 | 203.07 ± 147.08 | .87 |
| Small LDL, nm | 652.91 ± 367.95 | 434.33 ± 280.17a | <.0001 |
| Total HDL, μm | 39.09 ± 9.14 | 35.64 ± 7.83a | .01 |
| Large HDL, μm | 6.77 ± 4.44 | 7.68 ± 3.77 | .06 |
| Medium HDL, μm | 14.39 ± 7.62 | 12.77 ± 6.4 | .27 |
| Small HDL, μm | 18.32 ± 5.96 | 16.21 ± 5.37b | .02 |
| Particle size, nm | |||
| VLDL | 49.58 ± 6.04 | 46.78 ± 5.28a | .006 |
| LDL | 21.03 ± 5.73 | 20.67 ± 0.65 | .07 |
| HDL | 9.34 ± 0.49 | 9.55 ± 0.44a | .006 |
Data are presented as mean ± SD.
P < .01.
P < .05.
Discussion
Our study comprehensively analyzed HDL-C function and lipid particle number and size in a large cohort of women with PCOS. Supporting previous data, we found that young women with PCOS have a high prevalence of CVD risk factors, including hypertension, obesity, and metabolic syndrome, compared to controls. The conventional lipid profile demonstrated higher TG levels in PCOS, but the mean HDL-C and LDL-C levels were similar in both groups. We also found significantly lower levels of lipoprotein ApoA1, a major constituent of HDL-C, in PCOS. We assessed the function of HDL-C using the ex vivo cholesterol efflux assay and observed a significant reduction in efflux capacity in women with PCOS. Finally, NMR spectroscopy findings demonstrated an atherogenic lipid profile associated with PCOS, comprised of high VLDL particle number and size and increased small dense LDL, independent of obesity. Collectively, these findings demonstrate significant dyslipidemia associated with PCOS, which may provide further links to the increased risk of CVD in this population.
Total HDL-C measured in blood consists of several heterogeneous particles and does not assess HDL-C function. To determine the function of HDL-C, we performed cholesterol efflux assays that measure the capacity of HDL-C to remove cholesterol from macrophages. This is the first study to examine cholesterol efflux capacity in women with PCOS, and despite similar HDL-C levels, we report a significant decrease in cholesterol efflux capacity in cases compared to controls. As observed in other populations, cholesterol efflux capacity in PCOS also correlated with BMI, ApoA1, HDL-C, and the presence of metabolic syndrome (10, 14). In a multivariable regression model, PCOS status was significantly associated with lowered HDL-C efflux capacity. Adiponectin levels also significantly correlated with cholesterol efflux, although inclusion of adiponectin in the model did not alter the association between PCOS and HDL efflux. Recently, a positive correlation has been demonstrated between adiponectin levels and HDL-C efflux in subjects with diabetes, and improvement in efflux capacity was elegantly demonstrated with the addition of adiponectin in vitro (15). In our study, there was a negative correlation between total T and cholesterol efflux capacity. In a study examining the impact of gender on efflux capacity, serum estradiol levels did not correlate with cholesterol efflux capacity (16). Our data in the PCOS model confirm that the static measurement of HDL-C has inherent limitations in assessing HDL-C function. PCOS is a state of chronic low-grade inflammation (17), and atherogenic HDL profiles and impaired cholesterol transport have been associated with inflammatory states (14, 18). Although not assessed in our study, the clinical implications of our observations may be significant because impaired HDL-C efflux capacity has been reported to correlate with subclinical atherosclerosis, coronary artery disease, and ischemic cardiomyopathy (10, 19).
Our series comprehensively demonstrates an atherogenic lipid particle number and size pattern in women with PCOS independent of obesity (high VLDL particles and size and small LDL-C). The lipoprotein changes we noted in the NMR lipoprotein subclass profile for both LDL-C and HDL-C were not fully apparent in the conventional lipid panel. In fact, in our study there was no difference in calculated mean LDL-C or ApoB levels, the latter being a more accurate measurement of the relative number of LDL particles than LDL-C measurement. An association between small LDL-C particles and PCOS has been reported in smaller studies using gel electrophoresis (20, 21). The mechanism by which PCOS exerts its effect on lipoprotein quantity and quality is unknown. It has been proposed that androgens may contribute to dyslipidemia by influencing central body fat distribution and hepatic lipase activity, and exacerbating insulin resistance (22). Although we found a negative association between total T levels and ApoA1, NEFA, cholesterol efflux, and total HDL-C particles, the independent role of hyperandrogenism per se has not been clearly established. The association between androgens and the atherogenic lipid particle profile in other studies is not consistent (23–25). This could, however, be explained by genetic differences in populations, small sample sizes, and variations in T assays.
In other disease models, both insulin resistance and obesity have also been shown to be significant contributors to the presence of atherogenic lipid profiles (26). Our findings on NMR spectroscopy are very similar to those reported in subjects with insulin resistance—namely, an increase in VLDL size, large VLDL particle concentrations, small LDL particles, and number of LDL particles; a decrease in HDL size; and a modest increase in small HDL (26). Serum LDL-C levels do not change, whereas the concentration of small LDL particles and total LDL particles increases progressively as the severity of insulin resistance and the number of metabolic risk factors increases. In our study, HOMA correlated with VLDL particle number. Adipose tissue dysfunction plays a central role in the development of hypertriglyceridemia, which in turn is associated with the above-described atherogenic lipid profile, a decrease in insulin sensitivity, and changes in adipokines (22). Although our study was not designed to determine causality, women with PCOS had elevated TGs and decreased adiponectin levels. In fact, hypertriglyceridemia is a fairly consistent finding in lean and obese women with PCOS (27). A meta-analysis of 27 prospective studies with over 10 000 incident CHD cases demonstrated an adjusted odds ratio for CHD of 1.7 (95% CI, 1.6–1.9) comparing individuals in the top third of serum TG levels with those in the bottom third (28). Taken together, these data suggest that interventions targeted at weight loss measures may help restore the atherogenic dyslipidemia by improving insulin sensitivity and decreasing hypertriglyceridemia.
Our findings could explain some of the increase in CVD risk observed in women with PCOS. As mentioned in the introduction, there is significant evidence for subclinical atherosclerosis in PCOS. A few studies have also demonstrated early impairment of endothelial structure and function in lean women with PCOS (29). Differences in echocardiographic findings, such as a decrease in left ventricular ejection fraction, have been reported between women with PCOS and controls (30). Although there are no prospective longitudinal studies following large cohorts of well-defined women with PCOS, a recent meta-analysis included five studies assessing risk of both nonfatal and fatal CHD and stroke (31). Using the random-effects model, the pooled relative risk was 2.02 (95% CI, 1.47–2.76) for CHD or stroke when evaluating women with PCOS compared to those without PCOS. When limited to the only two studies that adjusted for BMI, the relative risk decreased to 1.55 (95% CI, 1.27–1.89). Given the available data, there is evidence to suggest that women with PCOS are at an increased risk for developing adverse cardiovascular-related outcomes.
In summary, detailed lipid phenotyping revealed several abnormalities in both HDL-C and LDL-C in women with PCOS. An increase in small LDL particles that contribute to increased cholesterol deposition in macrophages in arterial walls, accompanied by impaired HDL-C function, may add to the increased CVD risk in women with PCOS. These data strongly support educating all PCOS patients about the associated risk of dyslipidemia and the need for frequent lipid screening (1). Intervention trials including dietary modifications, exercise therapy, and pharmacotherapy should include lipid phenotyping as an outcome measure to better understand whether these treatments reverse atherogenic changes in lipoproteins and their function.
Acknowledgments
This work was funded by National Institutes of Health Grant DK19525 to the Diabetes Research Center RIA Biomarkers Core at the University of Pennsylvania. A.R. and J.H. were recipients of FOCUS medical student fellowship awards at the University of Pennsylvania.
Disclosure Summary: The authors have no conflicts of interest to declare.
Footnotes
- Apo
- apolipoprotein
- BMI
- body mass index
- BP
- blood pressure
- CHD
- coronary heart disease
- CI
- confidence interval
- CVD
- cardiovascular disease
- HDL-C
- high-density lipoprotein cholesterol
- HOMA
- homeostasis model assessment
- LDL
- low-density lipoprotein
- LDL-C
- LDL cholesterol
- NEFA
- nonesterified fatty acid
- NMR
- nuclear magnetic resonance
- PCOS
- polycystic ovary syndrome
- TG
- triglyceride
- VLDL
- very low-density lipoprotein.
References
- 1. Wild RA, Carmina E, Diamanti-Kandarakis E, et al. Assessment of cardiovascular risk and prevention of cardiovascular disease in women with the polycystic ovary syndrome: a consensus statement by the Androgen Excess and Polycystic Ovary Syndrome (AE-PCOS) Society. J Clin Endocrinol Metab. 2010;95(5):2038–2049 [DOI] [PubMed] [Google Scholar]
- 2. Moran LJ, Misso ML, Wild RA, Norman RJ. Impaired glucose tolerance, type 2 diabetes and metabolic syndrome in polycystic ovary syndrome: a systematic review and meta-analysis. Hum Reprod Update. 2010;16(4):347–363 [DOI] [PubMed] [Google Scholar]
- 3. Wang ET, Calderon-Margalit R, Cedars MI, et al. Polycystic ovary syndrome and risk for long-term diabetes and dyslipidemia. Obstet Gynecol. 2011;117(1):6–13 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Dokras A. Cardiovascular disease risk in women with PCOS. Steroids. 2013;78(8):773–776 [DOI] [PubMed] [Google Scholar]
- 5. Solomon CG, Hu FB, Dunaif A, et al. Menstrual cycle irregularity and risk for future cardiovascular disease. J Clin Endocrinol Metab. 2002;87(5):2013–2017 [DOI] [PubMed] [Google Scholar]
- 6. Shaw LJ, Bairey Merz CN, Azziz R, et al. Postmenopausal women with a history of irregular menses and elevated androgen measurements at high risk for worsening cardiovascular event-free survival: results from the National Institutes of Health–National Heart, Lung, and Blood Institute sponsored Women's Ischemia Syndrome Evaluation. J Clin Endocrinol Metab. 2008;93(4):1276–1284 [DOI] [PMC free article] [PubMed] [Google Scholar] [Retracted]
- 7. Abbott RD, Wilson PW, Kannel WB, Castelli WP. High density lipoprotein cholesterol, total cholesterol screening, and myocardial infarction. The Framingham Study. Arteriosclerosis. 1988;8(3):207–211 [DOI] [PubMed] [Google Scholar]
- 8. Robinson JG. Low high-density lipoprotein cholesterol and chronic disease risk marker or causal? J Am Coll Cardiol. 2010;55:2855–2857 [DOI] [PubMed] [Google Scholar]
- 9. Rader DJ. Regulation of reverse cholesterol transport and clinical implications. Am J Cardiol. 2003;92:42J–49J [DOI] [PubMed] [Google Scholar]
- 10. Khera AV, Cuchel M, de la Llera-Moya M, et al. Cholesterol efflux capacity, high-density lipoprotein function, and atherosclerosis. N Engl J Med. 2011;364:127–135 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Rotterdam ESHRE/ASRM-Sponsored PCOS Consensus Workshop Group. Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome. Fertil Steril. 2004;81(1):19–25 [DOI] [PubMed] [Google Scholar]
- 12. Grundy SM, Cleeman JI, Daniels SR, et al. Diagnosis and management of the metabolic syndrome: an American Heart Association/National Heart, Lung, and Blood Institute Scientific Statement. Circulation. 2005;112:2735–2752 [DOI] [PubMed] [Google Scholar]
- 13. Otvos JD, Mora S, Shalaurova I, Greenland P, Mackey RH, Goff DC., Jr Clinical implications of discordance between low-density lipoprotein cholesterol and particle number. J Clin Lipidol. 2011;5:105–113 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Mehta NN, Li R, Krishnamoorthy P, et al. Abnormal lipoprotein particles and cholesterol efflux capacity in patients with psoriasis. Atherosclerosis. 2012;224(1):218–221 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Wang M, Wang D, Zhang Y, Wang X, Liu Y, Xia M. Adiponectin increases macrophages cholesterol efflux and suppresses foam cell formation in patients with type 2 diabetes mellitus. Atherosclerosis. 2013;229(1):62–70 [DOI] [PubMed] [Google Scholar]
- 16. Badeau RM, Metso J, Kovanen PT, Lee-Rueckert M, Tikkanen MJ, Jauhiainen M. The impact of gender and serum estradiol levels on HDL-mediated reverse cholesterol transport. Eur J Clin Invest. 2013;43(4):317–323 [DOI] [PubMed] [Google Scholar]
- 17. Duleba AJ, Dokras A. Is PCOS an inflammatory process? Fertil Steril. 2012;97(1):7–12 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. de la Llera Moya M, McGillicuddy FC, Hinkle CC, et al. Inflammation modulates human HDL composition and function in vivo. Atherosclerosis. 2012;222(2):390–394 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Patel PJ, Khera AV, Wilensky RL, Rader DJ. Anti-oxidative and cholesterol efflux capacities of high-density lipoprotein are reduced in ischaemic cardiomyopathy. Eur J Heart Fail. 2013;15(11):1215–1219 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Rizzo M, Berneis K, Hersberger M, et al. Milder forms of atherogenic dyslipidemia in ovulatory versus anovulatory polycystic ovary syndrome phenotype. Hum Reprod. 2009;24:2286–2292 [DOI] [PubMed] [Google Scholar]
- 21. Doi SA, Abbas JM, Parkinson L, Chakraborty J, Akanji AO. LDL species heterogeneity in the atherogenic dyslipidemia of polycystic ovary syndrome. Am J Clin Pathol. 2008;129:802–810 [DOI] [PubMed] [Google Scholar]
- 22. van de Woestijne AP, Monajemi H, Kalkhoven E, Visseren FL. Adipose tissue dysfunction and hypertriglyceridemia: mechanisms and management. Obes Rev. 2011;12(10):829–840 [DOI] [PubMed] [Google Scholar]
- 23. Phelan N, O'Connor A, Kyaw-Tun T, et al. Lipoprotein subclass patterns in women with polycystic ovary syndrome (PCOS) compared with equally insulin-resistant women without PCOS. J Clin Endocrinol Metab. 2010;95:3933–3939 [DOI] [PubMed] [Google Scholar]
- 24. Sidhwani S, Scoccia B, Sunghay S, Stephens-Archer CN, Mazzone T, Sam S. PCOS is associated with atherogenic changes in lipoprotein particle number and size independent of body weight [published online February 15, 2011]. Clin Endocrinol (Oxf). doi:10.1111/j.1365–2265.2011.04015.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Dejager S, Pichard C, Giral P, et al. Smaller LDL particle size in women with polycystic ovary syndrome compared to controls. Clin Endocrinol (Oxf). 2001;54(4):455–462 [DOI] [PubMed] [Google Scholar]
- 26. Garvey WT, Kwon S, Zheng D, et al. Effects of insulin resistance and type 2 diabetes on lipoprotein subclass particle size and concentration determined by nuclear magnetic resonance. Diabetes. 2003;52(2):453–462 [DOI] [PubMed] [Google Scholar]
- 27. Wild RA. Dyslipidemia in PCOS. Steroids. 2012;77(4):295–299 [DOI] [PubMed] [Google Scholar]
- 28. Sarwar N, Danesh J, Eiriksdottir G, et al. Triglycerides and the risk of coronary heart disease: 10,158 incident cases among 262,525 participants in 29 Western prospective studies. Circulation. 2007;115(4):450–458 [DOI] [PubMed] [Google Scholar]
- 29. Orio F, Jr, Palomba S, Cascella T, et al. Early impairment of endothelial structure and function in young normal-weight women with polycystic ovary syndrome. J Clin Endocrinol Metab. 2004;89(9):4588–4593 [DOI] [PubMed] [Google Scholar]
- 30. Orio F, Jr, Palomba S, Spinelli L, et al. The cardiovascular risk of young women with polycystic ovary syndrome: an observational, analytical, prospective case-control study. J Clin Endocrinol Metab. 2004;89(8):3696–3701 [DOI] [PubMed] [Google Scholar]
- 31. de Groot PC, Dekkers OM, Romijn JA, Dieben SW, Helmerhorst FM. PCOS, coronary heart disease, stroke and the influence of obesity: a systematic review and meta-analysis. Hum Reprod Update. 2011;17(4):495–500 [DOI] [PubMed] [Google Scholar]