Abstract
The association between cardiometabolic disorders and polycystic ovary syndrome (PCOS) is well established. Women with PCOS are at increased risk for obesity, insulin resistance, diabetes, dyslipidemia, hypertension, and cardiovascular disease (CVD). 1 –4 Exploration of optimal methods to screen for the eventual development of CVD in a typically young reproductive-aged population diagnosed with PCOS is encouraged. Coronary artery calcification (CAC), a surrogate marker of subclinical atherosclerosis and an increasingly recognized prognostic marker for CVD risk, may become useful. In their recent systematic review and meta-analysis, Osibogun et al. 5 investigated the existing studies evaluating the association between PCOS and CAC. Their conclusions provide support for an early presence of CAC among women with PCOS. Assessing for CAC has the potential to serve as a powerful early marker of increased risk for CVD in this at-risk population.
In their review, Osibogun et al. performed a multidatabase search for all studies related to PCOS and CAC. The authors reviewed 36 unique articles, eventually including 7 studies (3 prospective cohort and 4 cross-sectional) with data for 2341 participants. Six of the seven studies were then included in a pooled meta-analysis. The primary outcome of interest was CAC presence among patients diagnosed with PCOS, as indicated by a CAC score >0 (CAC >0). The inclusion criteria were peer-reviewed studies on PCOS and CAC in human populations. The methodological quality was assessed using the Newcastle-Ottawa Scale (NOS). Two separate pooled analyses were performed for three prospective cohort and three cross-sectional studies.
In addition, a combined pooled analysis of all six studies was done. One study included in the review but omitted from meta-analysis was excluded because CAC outcome was defined in that study as ≥10, rather than >0. 6 Five of the included studies reported a mean age of participants between 32 and 49.4 years, whereas two reported a median age between 38.5 and 42 years. All included studies were conducted in the United States. The prospective cohort studies were awarded seven stars using NOS, considered strong quality, whereas the cross-sectional studies were awarded six stars, indicating fair quality.
Meta-analysis of the three prospective cohort studies demonstrated a pooled odds ratio (OR) of 2.48 for CAC >0 among women with PCOS (2.11–2.84) with no significant heterogeneity (I 2 = 0.10%, p = 0.97). Meta-analysis of the three cross-sectional studies revealed a pooled OR of 1.88 for CAC >0 among women with PCOS (0.71–3.06) with no significant heterogeneity (I 2 = 13.95%, p = 0.87). In a combined meta-analysis of all six studies, the pooled OR was 2.24 for CAC >0 among PCOS women (1.80–2.68) with no significant heterogeneity (I 2 = 12.18%, p = 0.96). These findings suggest that in their meta-analysis of a small number of observational studies, the authors found a positive correlation between the presence of CAC and PCOS.
Although the authors' findings are certainly intriguing, there are limitations. The sample sizes of included studies are small and likely underpowered. Few negative studies have been published on this topic. Two of the included studies were performed by the same author and likely included overlap in study participants—the Pittsburgh PCOS cohort. 7,8 Finally, at the same time, no significant heterogeneity was reported in the analyses; the diagnostic criteria for PCOS are decidedly different across studies. Similarly, other confounding variables were controlled for in each study: one prospective cohort study adjusted for age and body mass index, another matched PCOS controls by age and race, and another adjusted for age, BMI, race, education, smoking, menopausal status, systolic blood pressure, triglycerides, and Homeostatic Model Assessment for Insulin Resistance (HOMA-IR). 9 There are known significant racial disparities in CVD risk and expression of disease by risk factor severity, and the ability of a new marker to identify disease across different races should be evaluated.
Before universal screening for CAC among PCOS patients becomes a reality, future studies must consider the risks and costs of screening and establish good management algorithms to respond to both positive and negative CAC results. The significance of a CAC score of 0 is not yet well understood, nor is the clinical relevance of minimal CAC scores. CAC reflects the body's attempt to respond to developing atherosclerotic lesions, but its presence may not precisely predict the future risk of an early CVD event. The knowledge of CAC presence in and of itself may prove motivating prompting patients to engage in preventive lifestyle behaviors to reduce CVD risk (e.g., heart-healthy diets, adequate exercise, and smoking cessation). Satisfactory evidence of CAC's validity and reliability as a predictive marker for CVD is still needed.
In addition, the appropriate management of patients with detected CAC in a reproductive-aged population can be challenging. Interventions proposed by the study's authors, including metformin or statin therapy, will need to be individualized. Newer information is emerging that statins can be used in this patient population; however, they should be used in conjunction with low-density lipoprotein and non-high-density lipoprotein cholesterol measurements and target age-appropriate goals.
Before CAC can be established as a standard part of CVD risk assessment among PCOS patients, its relevance and incorporation into existing screening paradigms must be carefully examined. The Endocrine Society and Androgen Excess and PCOS Society 10 currently provide guidelines for the screening and managing of glucose intolerance and dyslipidemia among women with PCOS. All who provide care for patients with PCOS should similarly consult the American Heart Association (AHA) Guidelines for appropriate CVD risk assessment and prevention. 11 Given the heterogeneity of PCOS presentation, it is likely that phenotypes with more severe metabolic disturbances will require more aggressive screening, treatment, and monitoring of CVD risk factors.
Although these and many other questions remain, the recent study by Osibogun et al. introduces the exciting possibility of exploring CAC detection as a screening tool for CVD risk in PCOS. Future studies in this area should focus around validating the tool's accuracy in predicting CVD risk among this population, determine its cost-effectiveness and patient acceptance compared with existing screening methods, and provide algorithms for the appropriate management of CAC detection results.