S P E C I A L A p p r o a c h

F E A T U R E t o

t h e

P a t i e n t

Approach to the Patient: Contraception in Women With Polycystic Ovary Syndrome Bulent O. Yildiz Division of Endocrinology and Metabolism, Department of Internal Medicine, Hacettepe University School of Medicine, 06100 Ankara, Turkey

Polycystic ovary syndrome (PCOS) is a common reproductive and metabolic disorder. Patients with PCOS present with clinical signs of androgen excess (ie, hirsutism and acne), menstrual irregularities, and infertility. Combined oral contraceptive (OC) pills are the first-line medical therapy for the long-term management of PCOS. Containing a combination of estrogen and progestin, OCs restore regular menses, improve androgen excess, and provide effective contraception and protection from endometrial cancer. The benefits of hormonal contraception outweigh the risks in the vast majority of women with PCOS. However, concerns have been raised about potential adverse cardiovascular and metabolic effects of OCs. Currently available evidence indicates an increased relative risk of venous thrombosis associated with OCs varying among different formulations. Arterial thrombosis risk attributable to OCs does not appear to be significantly increased in young nonsmoking women. OC use might be associated with increased risk of diabetes in morbidly obese women with PCOS with severe insulin resistance. A tailored clinical approach to oral contraception in women with PCOS requires individualized risk stratification and management by determination of each PCOS patient’s personal cardiometabolic risk profile at baseline and during follow-up. Before prescribing an OC, clinicians should document individual risk factors including age, smoking, obesity, any degree of glucose intolerance including prediabetes and diabetes, hypertension, dyslipidemia, thrombophilia, and personal or family history of a venous thromboembolic event. (J Clin Endocrinol Metab 100: 794 – 802, 2015)

Case 20-year-old woman presented with a complaint of having “too much facial and body hair” that she found bothersome and embarrassing. She noted hair growth around puberty that had gradually progressed. She had been shaving and waxing regularly for her excess hair. She had always had irregular menses three to four times a year. She had never been pregnant and had not yet been sexually active. She denied galactorrhea. She was not taking any prescription medication or over-the-counter product. She did not pay much attention to her diet. She stated that her mother had excessive body hair and that no family member had type 2 diabetes or cardiovascular disease (CVD). Personal and family history were negative for any venous thrombotic event.

On physical examination, her blood pressure was 115/ 74, and her pulse was 76 beats per minute. Her height was 1.64 m, and weight was 80 kg. Her body mass index (BMI) was 29.8 kg/m2, and waist circumference was 85 cm. She had prominent coarse hair over her upper lip, chin, lower abdomen, lower back, and upper legs (modified FerrimanGallwey score, 14). Her thyroid was normal and palpable. She did not have acanthosis nigricans or any features of virilization, Cushing’s syndrome, or acromegaly. Laboratory tests revealed normal TSH, prolactin, and 17-hydroxyprogesterone, negative human chorionic gonadotropin, and bilateral polycystic ovaries on ultrasound (right ovary, 13 mL, containing 15 follicles each measuring 2–9 mm; left ovary, 11 mL, containing 14 follicles each measuring 2–9 mm). Her total T was 3.1 nmol/L (upper limit of normal, 2.1 nmol/L), and calculated free androgen

ISSN Print 0021-972X ISSN Online 1945-7197 Printed in U.S.A. Copyright © 2015 by the Endocrine Society Received August 12, 2014. Accepted December 31, 2014.

Abbreviations: BMI, body mass index; CI, confidence interval; CVD, cardiovascular disease; EE, ethinyl estradiol; HDL, high-density lipoprotein; LDL, low-density lipoprotein; OC, oral contraceptive; OGTT, oral glucose tolerance test; PCOS, polycystic ovary syndrome; RR, relative risk; VTE, venous thromboembolism; WHR, waist-to-hip ratio.

A

794

jcem.endojournals.org

J Clin Endocrinol Metab, March 2015, 100(3):794 – 802

doi: 10.1210/jc.2014-3196

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 January 2017. at 03:50 For personal use only. No other uses without permission. . All rights reserved.

doi: 10.1210/jc.2014-3196

index (100 ⫻ total T/SHBG) was 7.9 (upper limit of normal, 4.9). The patient was diagnosed as having polycystic ovary syndrome (PCOS), according to all three sets of available diagnostic criteria (1–3). The specific phenotype (clinical/ biochemical androgen excess and ovulatory dysfunction and polycystic ovary morphology) is documented in the medical records as recommended by the recent National Institutes of Health Evidence-based Methodology Workshop on PCOS (4). Metabolic investigations in a fasting state revealed normal total and low-density lipoprotein (LDL)-cholesterol levels, a low high-density lipoprotein (HDL)-cholesterol level of 39 mg/dL (1 mmol/L), and high triglycerides at a level of 198 mg/dL (2.24 mmol/L). Fasting and 2-hour glucose values on a 75-g standard oral glucose tolerance test (OGTT) were 78 mg/dL (4.3 mmol/L) and 124 mg/dL (6.9 mmol/L), respectively. The patient was informed that PCOS is a chronic disorder that requires long-term follow-up and treatment with a combined oral contraceptive (OC) pill that would regulate her menses and make an observable difference in hirsutism after a period of at least 6 months. She was also informed that lifestyle change intervention will be an essential component of the long-term management of PCOS.

Background PCOS is a common and complex disorder characterized by androgen excess, ovulatory dysfunction, and polycystic ovaries. Women with PCOS typically present in the early reproductive years with clinical evidence of hyperandrogenism (ie, hirsutism, acne), menstrual irregularity, or infertility. Available evidence suggests that PCOS is a lifelong disorder associated with long-term cardiometabolic and reproductive implications above and beyond initial clinical presentation and short-term concerns. A personalized and integrative approach according to age and individual needs of each patient throughout her life is required to optimize treatment and prevent long-term health consequences. OCs, often referred to as birth control pills, have traditionally been the mainstay of chronic treatment in PCOS patients not seeking pregnancy. They ameliorate hyperandrogenism and regulate menstrual cycles (5). Current available evidence suggests that the benefits of OCs outweigh the risks in most patients with PCOS. Nevertheless, potential adverse cardiometabolic effects of OCs represent a concern, given that women with PCOS use these drugs for several years (6).

Diagnostic and Therapeutic Strategies Oral contraceptives OCs contain low doses of estrogens and progestins (Table 1). These hormones inhibit the synthesis and se-

jcem.endojournals.org

795

Table 1. Currently Available Low-Dose (ⱕ35 ␮g of EE) Combined OC Pills According to Type of Progestin and EE Dose Progestin Type First generation Norethindrone Norethindrone Norethindrone Norethindrone Ethynodiol diacetate Second generation Norgestrel Norgestrel Levonorgestrel Levonorgestrel Levonorgestrel Third generation Norgestimate Gestodene Desogestrel Antiandrogenic Drospirenone Cyproterone acetate

Progestin Dose, mg

EE Dose, ␮g

1 1.5 0.4/0.5/1 1 1

20 30 35 50 35/50

0.3 0.5 0.1 0.15/0.5 0.125

30 50 20 30 50

0.25 0.75 0.15

35 30 20/30

3 2

20/30 35

cretion of GnRH at the level of the hypothalamus. Estrogens inhibit the selection and development of a dominant follicle by suppression of FSH. Progestins inhibit ovulation via suppression of LH surge. Progestins also make the cervix hostile to sperm penetration by increasing the viscosity of cervical mucus and prevent implantation through an alteration of endometrial lining (7). Most of the currently available OCs contain ethinyl estradiol (EE) as the synthetic estrogenic compound. Pills containing less than 50 ␮g of EE are called “low-dose” OCs. Virtually all low-dose OCs contain ⱕ35 ␮g EE, and the dose of synthetic progestin ranges between 0.1 and 3 mg. The initial OCs in the 1960s contained mestranol in doses as high as 150 ␮g. Starting from the late 1960s, the amount of EE was significantly reduced to the current dose of 20 –35 ␮g to increase efficacy, safety, and tolerability (7). EE has stronger effects than natural estradiol on hepatic metabolism, including synthesis of SHBG, lipoproteins, angiotensinogen, and some estrogen-dependent clotting factors (8). To overcome these metabolic effects, more physiological forms resembling endogenous estrogen have recently been developed. These include 17␤-estradiol, estradiol valerate, and estetrol. OCs containing these natural estrogens induce less metabolic changes than OCs containing EE (8). The OCs are traditionally classified according to generation, which refers to the timing of the introduction of the molecule. Older progestins, called first generation, were approved before the 1970s and include norethindrone (known as norethisterone in Europe) and ethynodiol diacetate. Norgestrel and levonorgestrel are second-

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 January 2017. at 03:50 For personal use only. No other uses without permission. . All rights reserved.

796

Yildiz

Oral Contraceptives in PCOS

generation OCs, whereas norgestimate, gestodene, and desogestrel are third generation (7, 9). Synthetic progestins used in first- and second-generation OCs are chemically related to T (19-nortestosterone derivatives). These progestins vary in their chemical structures, potency, and pharmacokinetics. They bind the androgen receptor with different affinities and show different degrees of androgenic side effects, such as oily skin, acne, and hirsutism. Alternatively, third-generation progestins desogestrel, norgestimate, and gestodene show higher affinity for progesterone receptors and therefore less androgenicity (8). To reduce androgenic side effects, several new progestins derived from progesterone or spironolactone have been developed in the last few decades. These progestins, such as drospirenone, trimegestone, nestorone, cyproterone acetate, and nomegestrol acetate are designed to bind specifically to the progesterone receptor and to have no androgenic, estrogenic or glucocorticoid actions (8, 9). Dienogest, another new progestin, is also structurally related to T, but it is antiandrogenic and has no estrogenic or glucocorticoid activity (9). Among progestins with antiandrogenic effects, cyproterone acetate has the highest antiandrogenic effect. Antiandrogenic potencies of dienogest and drospirenone are approximately 40 and 30% of the potency of cyproterone acetate, respectively (9). Metabolic effects of estrogen in OCs are modulated by the type of the progestin included (8). The more androgenic progestins are able to counteract the stimulatory effects of estrogen on liver proteins and coagulation factors, but less androgenic or antiandrogenic progestins have limited counteraction on the effects of estrogen. Thus, OCs containing third-generation progestins as well as drospirenone and cyproterone acetate have reduced metabolic side effects compared to OCs containing more androgenic progestins (8). Noncontraceptive benefits and side effects of OCs Noncontraceptive benefits of OCs include decreased dysmenorrhea, menorrhagia, and anemia, improvements in acne and hirsutism, and decreased risk of osteoporosis and ectopic pregnancy (10). Long-term OC use is also associated with decreased risk of ovarian and endometrial cancer (10). The most common side effects of OCs that result in poor compliance or discontinuation include abnormal menstrual bleeding, nausea, breast tenderness, headache, and mood changes. Most of these side effects lessen significantly after the first few months of use (8). Many women report that they experience some weight gain during OC use. Available data indicate that OC use might increase adiposity in adolescents (11) and might be asso-

J Clin Endocrinol Metab, March 2015, 100(3):794 – 802

ciated with central redistribution of body fat in young women with PCOS without a recognizable difference in clinical anthropometric measurements, including BMI and waist circumference (12). However, controlled clinical trials have failed to show any association between lowdose OCs and weight gain. In fact, a recent Cochrane review of the 49 available randomized trials did not find any evidence supporting a causal association between OCs and weight gain (13). Current contraindications to the use of OCs The World Health Organization (WHO) and the U.S. Centers for Disease Control and Prevention (CDC) have developed evidence-based guidelines for the use of OCs (14, 15). These documents, based on systematic reviews of available clinical and epidemiological research, are updated regularly. The most recent versions of the WHO and CDC guidelines were published in 2009 and 2010, respectively (14, 15). Absolute and relative contraindications to the use of low-dose OCs according to the WHO guideline are shown in Table 2. The risks and benefits of combined hormonal contraceptives are grouped in the WHO guideline, including both OCs and contraceptive patch and contraceptive ring. Rationale for the use of OCs in the treatment of PCOS OCs are the first-line treatment for many women with PCOS when fertility is not desired (5). They improve clinical manifestations of androgen excess and regulate menstrual cycles. The progestin component of OCs suppresses the secretion of LH and decreases ovarian androgen production. The estrogenic fraction of the pill increases the levels of SHBG, thus resulting in a decrease in circulating free T levels and its bioavailability. The progestin component can compete for 5␣-reductase and the androgen receptor, resulting in a decrease in androgen action. OCs also slightly reduce adrenal androgen production (5). In general, at least 6 months of treatment with OCs is necessary to detect differences in hirsutism or acne of women with PCOS (16). Both ESHRE/ASRM-sponsored PCOS Consensus Group recommendations (5) and The Endocrine Society’s clinical practice guidelines (17) suggest OCs as first-line management for amelioration of clinical and biochemical androgen excess and menstrual irregularity. They also emphasize that there is no definitive evidence for any difference in efficacy of various OCs (5, 17). In accordance, OCs are the most commonly prescribed medications for the long-term management of the syndrome. However, concerns remain regarding safety profile and potential risk of thrombosis and metabolic disease (6).

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 January 2017. at 03:50 For personal use only. No other uses without permission. . All rights reserved.

doi: 10.1210/jc.2014-3196

Table 2. Contraindications to the Use of Low-Dose (ⱕ35 ␮g EE) Combined OC Pills Absolute contraindications (ie, unacceptable health risk) ⬍6 weeks postpartum if breastfeeding Smoker over the age of 35 y (ⱖ15 cigarettes per day) Hypertension (systolic ⱖ160 mm Hg or diastolic ⱖ100 mm Hg) History of deep venous thrombosis/pulmonary embolism Current deep venous thrombosis/pulmonary embolism Major surgery with prolonged immobilization Known thrombogenic mutations (eg, Factor V Leiden, prothrombin mutation, protein S, protein C, and antithrombin deficiencies)a Current case and history of ischemic heart disease Stroke (history of cerebrovascular accident) Complicated valvular heart disease Systemic lupus erythematosus with positive antiphospholipid antibodies Migraine headache with focal neurological symptoms Current breast cancer Diabetes with nephropathy/retinopathy/neuropathy Other vascular disease or diabetes of ⬎20-y duration Active viral hepatitis Severe cirrhosis Liver tumors Selected relative contraindications (risks generally outweigh advantages) Smoker over the age of 35 y (⬍15 cigarettes per day) Adequately controlled hypertension Hypertension (systolic 140 –159 mm Hg, diastolic 90 –99 mm Hg) Migraine headache over the age of 35 Current gallbladder disease Past OC-related history of cholestasis Mild (compensated) cirrhosis Use of drugs that affect liver enzymes Anticonvulsant therapy Antiretroviral therapy Adapted from Ref. 14. a

Routine screening is not recommended because of the rarity of the conditions and the high cost of the screening.

OC use and the risk of venous thrombosis in PCOS An association between use of OCs and venous thromboembolism (VTE) has been reported consistently. The VTE rates in women of reproductive age are 0.5–1, 6 –10, and 50 per 10 000 women-years in the general population, in pregnancy, and in the puerperal period, respectively (18). The risk of VTE is increased 2- to 6-fold in OC users compared to nonusers (19). The risk is highest during the first 3 months of use and returns to that of nonusers within weeks of discontinuation (19). VTE risk depends on the dose of EE and the type of progestin (20). A lowering of estrogen dose from 100 to 50 ␮g has been associated with a decreased risk of venous thrombosis. Whether lowering of the estrogen dose to 30 or 20 ␮g would lead to a further decrease in the risk of venous thrombosis is not clear. In a population-based case-control study including 1524 patients with venous

jcem.endojournals.org

797

thrombosis and 1760 controls taking an estrogen dose of 30 ␮g as the reference, the thrombotic risk was 0.8 (95% confidence interval [CI], 0.5–1.2) for an estrogen dose of 20 ␮g and 1.9 (95% CI, 1.1–3.4) for a dose of 50 ␮g (20). Interestingly, the risk of thrombosis associated with 20 vs 30 ␮g EE in OCs containing gestodene was 0.3 (95% CI, 0.2– 0.7), suggesting that the type of progestin might also influence the thrombosis risk associated with estrogen dose or estrogenicity (20). Newer generation OCs have up to about a 2-fold increased risk of VTE compared with second-generation OCs containing levonorgestrel (21, 22). In recent metaanalyses, the risk of VTE was similar for norgestimate and levonorgestrel, whereas the pooled relative risk (RR) of VTE compared with levonorgestrel was 1.3–1.5 for gestodene, 1.8 –1.9 for desogestrel, 1.6 –1.7 for drospirenone, and 1.6 –1.8 for cyproterone acetate (23–26). PCOS appears to be associated with a prothrombotic state, reflected by a decreased global fibrinolytic capacity and increased activity of plasminogen activator inhibitor type 1, a potent inhibitor of fibrinolysis (27). Recently, two studies reviewing commercially available clinical databases looked into the risk of VTE and its potential association with the use of OCs in women with PCOS (28, 29). In the first study, the prevalence rates of VTE per 100 000 were 374.2 and 193.8 among women with and without PCOS, respectively, regardless of age and OC use (28). Interestingly, a protective effect of OC use for VTE among the patients with PCOS was observed (odds ratio, 0.8; 95% CI, 0.73– 0.97) in this study (28). In the second study, among women not taking OCs, the incidence of VTE was 6.3 per 10 000 person-years for women with PCOS and 4.1 per 10 000 person-years for matched controls, with a RR of 1.55 (95% CI, 1.10 –2.19) (29). The incidence of VTE among women taking OCs was 23.7 per 10 000 person-years for women with PCOS and 10.9 per 10 000 person-years for other OC users, with a RR of 2.12 (95% CI, 1.40 –3.21) (29). Overall, available data indicate an increased RR of VTE with the use of OCs. Clinicians should compare the risk of VTE per OC preparation and evaluate the additional acquired or genetic risk factors for VTE, such as obesity, smoking, advanced age, immobility, and hereditary thrombophilia (30). It should also be kept in mind that the RR of VTE in young healthy women is low, and the absolute risk is even smaller than the risk associated with pregnancy (18). Contribution of OCs to the risk of VTE becomes more important as a woman’s baseline risk increases with age, particularly after the age of 35 years. Subject to inherent limitations and of retrospective cohort studies such as bias and lack of adjustment for confounders, available data suggest a 1.5- to 2-fold increased risk of

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 January 2017. at 03:50 For personal use only. No other uses without permission. . All rights reserved.

798

Yildiz

Oral Contraceptives in PCOS

VTE in PCOS. More research is needed in women with PCOS regarding the risk of VTE and association of this risk with different preparations of OCs. OC use and the risk of arterial thrombosis in PCOS Epidemiological data in healthy women suggest that current use of low-dose OCs might increase the risk of arterial thrombosis, although that risk does not continue once OCs are stopped (31). A large historical cohort study of women between 15 and 49 years old reported a 1.5- to 2-fold increased RR of myocardial infarction and thrombotic stroke among current users of OCs irrespective of the progestin type (32). The same study reported that the risk of arterial thrombosis was increased 1.3- to 2.3-fold in those women taking EE at a dose of 30 – 40 ␮g, whereas thrombosis risk for women who used 20-␮g pills was between 0.9 and 1.7 (32). Considering the very low absolute risk of CVD in women of reproductive age, the number of additional arterial thrombotic events attributable to OC use was one or two per 10 000 women per year (32). The CVD risk is associated with increased age, smoking, and hypertension. Accordingly, young nonsmoking women do not appear to possess any CVD risk due to OC use. Intriguing data from a healthy population suggest that OC use during reproductive years might be protective against CVD later in life (33). In the Women’s Ischemia Syndrome Evaluation study, which evaluated the association between past OC use and angiographic coronary artery disease in postmenopausal women, past OC use was found to be a significant independent negative predictor of coronary artery disease severity after adjustments for several coronary risk factors including age, diabetes mellitus, triglycerides, LDL-cholesterol, smoking, aspirin use, and lipid-lowering medications (33). The current literature supports the fact that many women with PCOS have increased cardiovascular risk factors, suggesting that the risk of CVD is increased in the disorder (5). Obesity, insulin resistance, dyslipidemia, and dysfibrinolysis are generally found to be more common in patients with PCOS compared to healthy women (5). Nevertheless, despite the existence of several CVD risk factors in women with PCOS, there is no direct evidence of increased or premature morbidity or mortality from CVD in the syndrome (5). Unfortunately, there are no data available in the literature assessing potential association of OC use and CVD outcome in PCOS. Because patients with PCOS do not appear to have increased or premature morbidity or mortality from CVD, despite having increased CVD risk factors, it would be interesting to test whether the long-term use of OCs during reproductive years is protective against CVD morbidity and mortality in PCOS later in life.

J Clin Endocrinol Metab, March 2015, 100(3):794 – 802

OC use and the risk of diabetes in PCOS Available data in a healthy population do not support a significant influence of OCs on glucose and insulin homeostasis (34). Cross-sectional data from the Third National Health and Nutrition Examination Survey showed that fasting glucose, insulin, C-peptide, and hemoglobin A1c levels were not different between the current users, past users, and never users of OCs (34). In a prospective cohort study, 98 590 healthy women aged 25 to 42 years were followed for 4 years, and it was found that the incidence of diabetes was not increased in current or past users of OCs (35). Furthermore, a recent meta-analysis has reported that there are no major differences between different OCs regarding their effects on carbohydrate metabolism (36). In PCOS, the results of the prospective studies regarding the effects of different OCs on insulin sensitivity are inconsistent and contradictory in that decreased, unchanged, and increased insulin sensitivity measurements after 3–12 months of use have been reported (6). More importantly, in all but two studies glucose tolerance status did not change (6). Of note, both of these studies included morbidly obese PCOS patients with average BMI of 36.8 and 37.2 kg/m2, respectively (37, 38). In a recent meta-analysis of 35 observational studies and cohorts from randomized controlled trials investigating the association between metabolic changes and use of OCs including different types of progestins, OC use was not associated with any significant change of fasting glucose, fasting insulin, homeostasis model assessment of insulin resistance, or euglycemic hyperinsulinemic clampglucose disposal rate in women with PCOS (39). Included studies showed significant heterogeneity with several limitations. OC use was significantly associated with an increase in HDL-cholesterol and TG levels (P ⫽ .004 for both) in this report (39). Overall, low-dose OC use for up to 1 year does not have a significant adverse impact on insulin sensitivity or glucose tolerance in most of the patients with PCOS (6). However, deterioration of glucose tolerance status might be observed, particularly in morbidly obese women with PCOS (37, 38). It is highly likely that, similar to healthy individuals, the risk of diabetes development depends on individual patient characteristics such as BMI, age, ethnicity, and family history of diabetes. Finally, it remains to be determined prospectively how these variable effects of OCs on insulin sensitivity and glucose tolerance status within 1 year translate into the longer term use of these medications in PCOS.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 January 2017. at 03:50 For personal use only. No other uses without permission. . All rights reserved.

doi: 10.1210/jc.2014-3196

Long-term use of OCs and cardiometabolic risk in PCOS Only a few long-term observational studies evaluating the metabolic effects of OCs in PCOS are available in the literature. In a prospective open-label study, lipid profiles and glucose homeostasis were evaluated in 72 women with PCOS treated with EE/cyproterone acetate for 3 years in comparison with 39 healthy women (40). At baseline, women with PCOS had higher levels of total cholesterol and triglycerides and lower levels of HDL-cholesterol. The investigators observed an increase in triglycerides and HDLcholesterol and a decrease in LDL/HDL ratio in women with PCOS after the treatment. More importantly, insulin and glucose plasma concentrations did not change (40). Another observational study was conducted on 37 PCOS patients with an average follow-up of 10 years (range, 12–180 mo) to assess the long-term effects of OCs on cardiometabolic risk factors in PCOS patients (41). The mean ages at the beginning and the end of the follow-up were 18 and 29 years for the OC users and 21 and 31 years for the non-OC users. Sixteen patients were on OC treatment (EE combined with cyproterone acetate or gestodene or desogestrel), whereas 21 patients had never used OCs. None of the anthropometric measurements changed in non-OC users during the follow-up, including body weight, BMI, waist and hip circumferences, and waistto-hip ratio (WHR). Alternatively, waist circumference and WHR were significantly reduced in OC users. The area under the curve for glucose during OGTT decreased in OC users and was unchanged in non-OC users, whereas the area under the curve for insulin was unchanged in OC users but increased in non-OC users. Finally, HDL-cholesterol and SHBG levels increased significantly only in the OC users, whereas there was no change in non-OC users (41). A recent cross-sectional evaluation of a retrospective cohort reported the metabolic influence of various OCs on 1297 women with PCOS diagnosed according to the Rotterdam criteria between 1983 and 2011 (42). After adjustment for age and BMI, total cholesterol, triglycerides, HDL- and LDL-cholesterol did not show a significant change among current users (1– 8 y duration), ever users, and never users (42). OC use was found not to negatively influence BMI, WHR, and homeostasis model assessment of insulin resistance, and there was even a significant decrease of WHR among ever users (42). On the other hand, studies evaluating the effects of OCs in comparison with low-dose insulin sensitizer and antiandrogen combination in adolescent girls with PCOS suggest that sustained OC use might result in an increase in high-sensitivity C-reactive protein levels and carotid intima media thickness (43). We have also reported an in-

jcem.endojournals.org

799

crease of high-sensitivity C-reactive protein with OC use in young lean women with PCOS (44). Taken together, findings of these studies suggest that insulin resistance worsens during the natural course of the syndrome, whereas long-term OC use does not change or improve the cardiometabolic risk parameters including insulin resistance and lipoprotein profile, at least in adult patients with PCOS. However, available data in adolescents and young women suggest that OC use might be associated with low-grade chronic inflammation and subclinical atherosclerosis that require further investigation.

Controversies and Areas of Uncertainty Definition of PCOS by different diagnostic criteria brings significant heterogeneity to the clinical phenotypes, with potentially varying degrees of cardiometabolic risk starting from the diagnosis. All treatment strategies including the use of OCs in PCOS remain mainly symptomatic in the face of multiple pathophysiological processes and the lack of a full understanding of the development of the syndrome. Head-to-head blinded trials comparing different OCs are lacking. Longitudinal follow-up data on benefits and risks of OCs are not available. It should be emphasized that most of the risk estimates of venous and arterial thrombosis associated with OC use in the general population are derived from case-control and cohort studies providing a RR increase in users compared with nonusers. It is not feasible to conduct randomized controlled trials in PCOS that are large enough to detect differences between various OCs in terms of rare adverse events such as venous and arterial thrombosis. Alternatively, well-designed observational studies with sufficient long-term follow-up deserve more attention. There is a paucity of data regarding the use of hormonal contraceptives other than OCs in PCOS. Extrapolating from the general population, one could use a contraceptive patch (weekly application) or contraceptive ring (monthly insertion) in women with PCOS who are unable to take OCs. PCOS patients with a contraindication to estrogen or who do not have androgen excess and are in need of contraception might be candidates for progestin-only contraceptives including long-acting injectables (depot medroxyprogesterone), etonogestrel-containing implant, or levonorgestrel-containing intrauterine system (45). These methods provide effective protection against endometrial hyperplasia. Depot medroxyprogesterone and etonogestrel are associated with weight gain, insulin resistance, and a worsening metabolic profile. Decreased bone mineral density is another adverse effect with the use of depot medroxyprogesterone (45)

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 January 2017. at 03:50 For personal use only. No other uses without permission. . All rights reserved.

800

Yildiz

Oral Contraceptives in PCOS

Back to Our Patient Our patient with PCOS was young and overweight. Treatment with an OC was the best available option because the patient presented with the main complaints of hirsutism and oligomenorrhea and was not considering pregnancy. OC use in this patient is also important for protection from endometrial carcinoma. As a nonsmoker below 35 years of age with no personal or family history of VTE, diabetes, or hypertension, the patient had no risk factors or contraindications that would prevent OC use. An individualized approach in this patient might favor prescription of a third or newer generation OC with neutral or antiandrogenic properties over a second-generation OC, considering dyslipidemia of the patient characterized by low HDL and increased triglycerides. How to start an OC in a patient with PCOS Before prescribing any kind of OC, a careful history of past and present medical conditions, any drug use, family history, focused physical examination, and laboratory assessments are required (Table 3). Specifically, information regarding migraine and CVD risk factors (smoking, hypertension, obesity, glucose intolerance, dyslipidemia, thrombophilia, previous VTE) is important. Occasionally, a patient with PCOS and multiple risk factors may need to avoid OC use. Blood pressure measurement, BMI, and a pregnancy test are required before a first prescription of OCs, whereas breast examination, pelvic and genital examination, and cervical cytology screening are not routinely recommended because they do not contribute substantially to the safety of OCs. In patients with PCOS, cardiometabolic risk assessment needs to be performed, including a 75-g standard 2-hour OGTT and lipid profile at baseline and during follow-up with regular intervals.

Table 3. Comprehensive Assessment Before Prescription of an OC in PCOS Medical history Age Past and present medical conditions Any drug use Migraine CVD risk factors (smoking, hypertension, obesity, diabetes, dyslipidemia) Thrombophilia (any known disorder of thrombophilia, personal or family history of previous VTE) Physical exam Blood pressure measurement BMI Waist circumference Laboratory tests 75-g 2-h standard OGTT Lipid profile

J Clin Endocrinol Metab, March 2015, 100(3):794 – 802

User preference and individual concerns should be addressed. The patient needs to be informed that OC use does not seem to be associated with weight gain in most women; however, some PCOS patients might gain weight, and clinical assessment of adiposity at regular intervals would be required during follow-up. At the first prescription, all patients should be informed that long-term OC use is safe for the majority but can be associated with minor side effects and with rare but serious harms. Patients should also be informed that there is a small increase in the risk of blood clots with OC use and that there are symptoms that would prompt immediate medical attention, such as warning signs of VTE (leg swelling or pain), visual disturbances, sensory or motor impairment, chest pain, and new headache. A thrombophilia screen is not recommended routinely before prescribing an OC, and a negative screen may not exclude all types of thrombophilia. Women with a family history of VTE in a first-degree relative ⬍45 years of age may indicate an increased likelihood of hereditary thrombophilia. In PCOS patients with a personal history of thrombophilia, progesterone-only pills would be an option. Contraceptive pills containing progestin only (referred to as the mini-pill) do not have a significant impact on coagulation or fibrinolysis and do not significantly alter carbohydrate or lipid metabolism (46). OCs are ideally started on the first day of the cycle and up to fifth day. They can be started at any other time if it is certain that the patient is not pregnant. To follow the schedule written on the packaging correctly, an alternative strategy might be to begin on the first Sunday after bleeding. After 3 months, a revisit needs to be scheduled for assessment of blood pressure and any other problems. OCs should be continued if the patient is comfortable with the drug. An annual visit is required to control for further compliance, side effects, and evaluation of glucose tolerance and lipids.

Conclusions OCs are a key component of the chronic treatment of PCOS, addressing many of the goals of the reproductiveaged women with PCOS, such as amelioration of hyperandrogenic skin manifestations, correction of menstrual irregularities, effective contraception, and protection of endometrium against effects of unopposed estrogen. Although guidelines do not suggest one OC formulation over another in terms of effectiveness, low-dose OCs containing neutral or antiandrogenic progestins may be the choice in the treatment of PCOS regarding the androgen excess

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 January 2017. at 03:50 For personal use only. No other uses without permission. . All rights reserved.

doi: 10.1210/jc.2014-3196

and the metabolic disturbances associated with the disorder. Potential adverse cardiovascular and metabolic effects of OCs have raised some concerns about the safety of long-term use of these drugs in PCOS. However, current evidence suggests that the benefits of oral contraception outweigh the risks in the vast majority of women with PCOS, as in the case presented here. OC use is associated with an increased RR of venous thrombosis, whereas absolute risk is very small. Young, nonsmoking patients with PCOS have no risk of arterial thrombosis attributable to OCs. Nevertheless, OC use might increase the risk of diabetes, particularly in obese patients with severe insulin resistance. Future studies evaluating the long-term effects and safety of OCs in the treatment of PCOS are needed. These studies should adequately consider clinical heterogeneity of the syndrome and variation in the efficacy and safety of different combinations. Meanwhile, the WHO guidelines for the contraindications to OC use should be exercised in women with PCOS, and the precise individualized treatment targets and risk stratification depending on patient characteristics should be determined.

Acknowledgments Address all correspondence and requests for reprints to: Bulent O. Yildiz, MD, Hacettepe University School of Medicine, Department of Internal Medicine, Division of Endocrinology and Metabolism, Hacettepe, 06100 Ankara, Turkey. E-mail: [email protected]. Disclosure Summary: The author has nothing to disclose.

References 1. Zawadzki JK, Dunaif A. Diagnostic criteria for polycystic ovary syndrome. In: Dunaif A, Givens JR, Haseltine F, Merriam GR, eds. Polycystic Ovary Syndrome. Boston, MA: Blackwell Scientific Publications; 1992:377–384. 2. 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; 81:19 –25. 3. Azziz R, Carmina E, Dewailly D, et al. Positions statement: criteria for defining polycystic ovary syndrome as a predominantly hyperandrogenic syndrome: an Androgen Excess Society Guideline. J Clin Endocrinol Metab. 2006;91:4237– 4245. 4. National Institutes of Health. Final Report of Evidence-based Methodology Workshop on Polycystic Ovary Syndrome, December 3–5, 2012. https://prevention.nih.gov/docs/programs/pcos/FinalReport. pdf. 5. Fauser BC, Tarlatzis BC, Rebar RW, et al. Consensus on women’s health aspects of polycystic ovary syndrome (PCOS): the Amsterdam ESHRE/ASRM-Sponsored 3rd PCOS Consensus Workshop Group. Fertil Steril. 2012;97:28 –38.e25. 6. Yildiz BO. Oral contraceptives in polycystic ovary syndrome: riskbenefit assessment. Semin Reprod Med. 2008;26:111–120.

jcem.endojournals.org

801

7. Christin-Maitre S. History of oral contraceptive drugs and their use worldwide. Best Pract Res Clin Endocrinol Metab. 2013;27:3–12. 8. Sitruk-Ware R, Nath A. Characteristics and metabolic effects of estrogen and progestins contained in oral contraceptive pills. Best Pract Res Clin Endocrinol Metab. 2013;27:13–24. 9. Sitruk-Ware R. New progestagens for contraceptive use. Hum Reprod Update. 2006;12:169 –178. 10. ESHRE Capri Workshop Group. Noncontraceptive health benefits of combined oral contraception. Hum Reprod Update. 2005;11: 513–525. 11. Ibáñez L, de Zegher F. Ethinylestradiol-drospirenone, flutamidemetformin, or both for adolescents and women with hyperinsulinemic hyperandrogenism: opposite effects on adipocytokines and body adiposity. J Clin Endocrinol Metab. 2004;89:1592–1597. 12. Aydin K, Cinar N, Aksoy DY, Bozdag G, Yildiz BO. Body composition in lean women with polycystic ovary syndrome: effect of ethinyl estradiol and drospirenone combination. Contraception. 2013; 87:358 –362. 13. Gallo MF, Lopez LM, Grimes DA, Schulz KF, Helmerhorst FM. Combination contraceptives: effects on weight. Cochrane Database Syst Rev. 2014;1:CD003987. 14. World Health Organization. Medical eligibility criteria for contraceptive use. 2009. http://whqlibdoc.who.int/publications/2010/ 9789241563888_eng.pdf. 15. Centers for Disease Control and Prevention. US medical eligibility criteria for contraceptive use, 2010. Morbidity and Mortality Weekly Report. Recommendations and Reports. 2010;59(RR04): 1– 6. 16. Yildiz BO. Assessment, diagnosis and treatment of a patient with hirsutism. Nat Clin Pract Endocrinol Metab. 2008;4:294 –300. 17. Legro RS, Arslanian SA, Ehrmann DA, et al. Diagnosis and treatment of polycystic ovary syndrome: an Endocrine Society Clinical Practice Guideline. J Clin Endocrinol Metab. 2013;98:4565– 4592. 18. Pomp ER, Lenselink AM, Rosendaal FR, Doggen CJ. Pregnancy, the postpartum period and prothrombotic defects: risk of venous thrombosis in the MEGA study. J Thromb Haemost. 2008;6:632– 637. 19. Martinelli I. Counselling women about hormonal therapy. Thromb Res. 2009;123(suppl 2):S74 –S78. 20. van Hylckama Vlieg A, Helmerhorst FM, Vandenbroucke JP, Doggen CJ, Rosendaal FR. The venous thrombotic risk of oral contraceptives, effects of oestrogen dose and progestogen type: results of the MEGA case-control study. BMJ. 2009;339:b2921. 21. Crook D. Do different brands of oral contraceptives differ in their effects on cardiovascular disease? Br J Obstet Gynaecol. 1997;104: 516 –520. 22. Venous thromboembolic disease and combined oral contraceptives: results of international multicentre case-control study. World Health Organization Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception. Lancet. 1995;346:1575– 1582. 23. Martínez F, Ramírez I, Pérez-Campos E, Latorre K, Lete I. Venous and pulmonary thromboembolism and combined hormonal contraceptives. Systematic review and meta-analysis. Eur J Contracept Reprod Health Care. 2012;17:7–29. 24. Stegeman BH, de Bastos M, Rosendaal FR, et al. Different combined oral contraceptives and the risk of venous thrombosis: systematic review and network meta-analysis. BMJ. 2013;347:f5298. 25. de Bastos M, Stegeman BH, Rosendaal FR, et al. Combined oral contraceptives: venous thrombosis. Cochrane Database Syst Rev. 2014;3:CD010813. 26. Plu-Bureau G, Maitrot-Mantelet L, Hugon-Rodin J, Canonico M. Hormonal contraceptives and venous thromboembolism: an epidemiological update. Best Pract Res Clin Endocrinol Metab. 2013; 27:25–34. 27. Yildiz BO, Haznedaroglu IC, Kirazli S, Bayraktar M. Global fibrinolytic capacity is decreased in polycystic ovary syndrome, sug-

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 January 2017. at 03:50 For personal use only. No other uses without permission. . All rights reserved.

802

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

Yildiz

Oral Contraceptives in PCOS

gesting a prothrombotic state. J Clin Endocrinol Metab. 2002;87: 3871–3875. Okoroh EM, Hooper WC, Atrash HK, Yusuf HR, Boulet SL. Is polycystic ovary syndrome another risk factor for venous thromboembolism? United States, 2003–2008. Am J Obstet Gynecol. 2012;207:377.e1– e8. Bird ST, Hartzema AG, Brophy JM, Etminan M, Delaney JA. Risk of venous thromboembolism in women with polycystic ovary syndrome: a population-based matched cohort analysis. CMAJ. 2013; 185:E115–E120. ESHRE Capri Workshop Group. Venous thromboembolism in women: a specific reproductive health risk. Hum Reprod Update. 2013;19:471– 482. Khader YS, Rice J, John L, Abueita O. Oral contraceptives use and the risk of myocardial infarction: a meta-analysis. Contraception. 2003;68:11–17. Lidegaard Ø, Løkkegaard E, Jensen A, Skovlund CW, Keiding N. Thrombotic stroke and myocardial infarction with hormonal contraception. N Engl J Med. 2012;366:2257–2266. Merz CN, Johnson BD, Berga S, Braunstein G, Reis SE, Bittner V. Past oral contraceptive use and angiographic coronary artery disease in postmenopausal women: data from the National Heart, Lung, and Blood Institute-sponsored Women’s Ischemia Syndrome Evaluation. Fertil Steril. 2006;85:1425–1431. Troisi RJ, Cowie CC, Harris MI. Oral contraceptive use and glucose metabolism in a national sample of women in the United States. Am J Obstet Gynecol. 2000;183:389 –395. Chasan-Taber L, Willett WC, Stampfer MJ, et al. A prospective study of oral contraceptives and NIDDM among U.S. women. Diabetes Care. 1997;20:330 –335. Lopez LM, Grimes DA, Schulz KF. Steroidal contraceptives: effect on carbohydrate metabolism in women without diabetes mellitus. Cochrane Database Syst Rev. 2014;4:CD006133. Nader S, Riad-Gabriel MG, Saad MF. The effect of a desogestrelcontaining oral contraceptive on glucose tolerance and leptin concentrations in hyperandrogenic women. J Clin Endocrinol Metab. 1997;82:3074 –3077.

J Clin Endocrinol Metab, March 2015, 100(3):794 – 802

38. Morin-Papunen LC, Vauhkonen I, Koivunen RM, Ruokonen A, Tapanainen JS. Insulin sensitivity, insulin secretion, and metabolic and hormonal parameters in healthy women and women with polycystic ovarian syndrome. Hum Reprod. 2000;15:1266 –1274. 39. Halperin IJ, Kumar SS, Stroup DF, Laredo SE. The association between the combined oral contraceptive pill and insulin resistance, dysglycemia and dyslipidemia in women with polycystic ovary syndrome: a systematic review and meta-analysis of observational studies. Hum Reprod. 2011;26:191–201. 40. Falsetti L, Pasinetti E. Effects of long-term administration of an oral contraceptive containing ethinylestradiol and cyproterone acetate on lipid metabolism in women with polycystic ovary syndrome. Acta Obstet Gynecol Scand. 1995;74:56 – 60. 41. Pasquali R, Gambineri A, Anconetani B, et al. The natural history of the metabolic syndrome in young women with the polycystic ovary syndrome and the effect of long-term oestrogen-progestagen treatment. Clin Endocrinol (Oxf). 1999;50:517–527. 42. Mes-Krowinkel MG, Louwers YV, Mulders AG, de Jong FH, Fauser BC, Laven JS. Influence of oral contraceptives on anthropomorphometric, endocrine, and metabolic profiles of anovulatory polycystic ovary syndrome patients. Fertil Steril. 2014;101:1757–1765.e1. 43. Ibáñez L, Díaz M, Sebastiani G, Marcos MV, López-Bermejo A, de Zegher F. Oral contraception vs insulin sensitization for 18 months in nonobese adolescents with androgen excess: posttreatment differences in C-reactive protein, intima-media thickness, visceral adiposity, insulin sensitivity, and menstrual regularity. J Clin Endocrinol Metab. 2013;98:E902–E907. 44. Harmanci A, Cinar N, Bayraktar M, Yildiz BO. Oral contraceptive plus antiandrogen therapy and cardiometabolic risk in polycystic ovary syndrome. Clin Endocrinol (Oxf). 2013;78:120 –125. 45. Verhaeghe J. Hormonal contraception in women with the metabolic syndrome: a narrative review. Eur J Contracept Reprod Health Care. 2010;15:305–313. 46. Heinemann LA, Assmann A, DoMinh T, Garbe E. Oral progestogen-only contraceptives and cardiovascular risk: results from the Transnational Study on Oral Contraceptives and the Health of Young Women. Eur J Contracept Reprod Health Care. 1999; 4:67–73.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 January 2017. at 03:50 For personal use only. No other uses without permission. . All rights reserved.