Jon Havelock, MD, FRCSC

Polycystic ovary syndrome

Therapy for this reproductive and metabolic disorder remains focused on managing symptoms, including infertility caused by anovulation, and reducing long-term health risks such as endometrial cancer and type 2 diabetes.

ABSTRACT: The clinical presenta- olycystic ovary syndrome drome was the term used for more tion of polycystic ovary syndrome is (PCOS) is a prevalent repro­ than 50 years for the heterogeneous widely variable, with complaints en- P ductive and metabolic disorder clinical features of the disorder now compassing oligomenorrhea, infer- with variable phenotypes and an under­­ known as polycystic ovary syndrome. tility, obesity, hirsutism, endometrial lying pathophysiology that is still not In 1990 the first international defini­ cancer, and diabetes. Community completely understood. While the tion of PCOS was developed, which physicians caring for reproductive- earliest description of the polycystic has since been revised by various age women will invariably encounter ovary dates back to the 17th century,1 professional bodies. The lack of con­ this reproductive and metabolic dis- the characterization of the present- sensus in the definition of PCOS fur­ order resulting from ovarian hyper- day disorder known as PCOS was first ther highlights the uncertainty about androgenism and insulin resistance. detailed by Irving Stein and Michael the pathophysiology of the disorder. While community physicians should Leventhal in 1935.2 In a seminal pa­ However, for the practising physician be aware of the diagnostic criteria per, the two prominent gynecologists a thorough understanding of symptom for polycystic ovary syndrome, it is described a case series of seven wom­ management and prevention of long- more important to have a thorough en with enlarged ovaries associated term complications is more important understanding of symptom manage- with oligomenorrhea or amenorrhea, than an understanding of the different ment and prevention of long-term sterility, and clinical hyperandrogen­ diagnostic criteria for PCOS. complications. Historically, clo- ism. Histopathologic determination of miphene citrate has been used to the disorder was undertaken by wedge Pathophysiology address infertility by inducing ovu- biopsy of the ovaries. The surgical The abnormal findings in PCOS are a lation, with more recent evidence procedure that led to characterization result of ovarian hyperandrogenism3 supporting the use of letrozole as of the disorder also serendipitously first-line therapy for ovulation induc- led to the first therapeutic intervention Dr Havelock is a co-director of the Pacific tion. These and other mainstay treat- for infertile women with PCOS. Five Centre for Reproductive Medicine and a ments may be needed to address of the seven women subsequently clinical assistant professor in the Division anovulation, obesity, and hirsutism. conceived after normalization of their of Reproductive Endocrinology and Infer- Patients should also be monitored menstrual cycles. One woman who tility at the University of British Columbia. for endometrial cancer and type 2 did not conceive was affected by male He is also the former program director for diabetes. factor infertility and the other woman the subspecialty residency in Gynecologic was lost to follow-up. Reproductive Endocrinology and Infertility This article has been peer reviewed. As a result, Stein-Leventhal syn­ at UBC.

210 bc medical journal vol. 60 no. 4, may 2018 bcmj.org Polycystic ovary syndrome

and insulin resistance.4 Evidence sug­ drogens, as demonstrated in the simi­ exclusion of other known disorders. gests that the ovarian hyperandrogen­ lar ultraosonographic findings and The National Institutes of Health ism in PCOS is a result of primary gene expression profile studies on the (NIH) conference on PCOS in 1990 ovarian dysfunction and is second­ ovaries of women with PCOS and the led to the first internationally accept­ ary to disordered gonadotropin activ­ ovaries of androgen-treated female- ed diagnostic criteria. The two criteria ity. While not included in diagnostic to-male transgender individuals.9 (clinical and/or biochemical evidence criteria for PCOS, the elevated level Evidence for the role of insulin of hyperandrogenism and menstrual of serum luteinizing hormone (LH) resistance in the pathophysiology of dysfunction) were based on expert in affected patients due to inappro­ PCOS and ovarian hyperandrogen­ opinion solicited through a question­ priate secretion has long been recog­ ism is demonstrated indirectly by naire. In 2003 the Rotterdam criteria nized.5 LH is the ligand for the LH the findings of hyperandrogenism in developed by the European Society receptor on the ovarian theca cells female subjects with type A insulin of Human Reproduction and Embry­ responsible for ovarian androgen pro­ resistance syndrome, a disorder char­ ology and the American Society for duction. Genome-wide association acterized by a mutation in the insulin Reproductive Medicine (ESHRE/ studies conducted on hyperandro­ receptor gene.10 Insulin contributes to ASRM) allowed for the inclusion of genic subjects with PCOS revealed the biochemical and clinical hyperan­ polycystic-appearing ovaries on ul­ genome-wide significance for a locus drogenism by directly enhancing the­ trasound. This was defined as 12 or mapping to chr 11p14.1 in the region ca cell ovarian androgen production more follicles measuring 2 to 9 mm of the follicle-stimulating hormone in concert with LH,4 and indirectly in at least one ovary, or an ovarian beta polypeptide (FSHB).6 This sin­ by lowering sex hormone-binding volume greater than 10 mL in the gle-nucleotide polymorphism was globulin, the carrier protein respon­ absence of a dominant follicle. The associated with LH levels that result sible for reducing circulating free ESHRE/ASRM diagnostic guidelines in the elevated LH:FSH ratios often testosterone levels.11 The high preva­ only required meeting two of three seen in PCOS, providing further sup­ lence of impaired glucose tolerance criteria (clinical and/or biochemical port for the hypothesis that dysregu­ and type 2 diabetes in women with hyperandrogenism, oligomenorrhea lated gonadotropin secretion in PCOS PCOS has led researchers to consider and/or anovulation, and polycystic leads to secondary hyperandrogen­ the role of insulin sensitizers in treat­ ovaries).12 Most recently, the experts ism. This gonadotropin imbalance ing PCOS. contributing to the Androgen Excess favors an exaggerated intraovarian Society (AES) diagnostic guidelines androgen environment under the in­ Diagnostic criteria required meeting two criteria (clinical fluence of LH, and impaired follicu­ Three sets of diagnostic criteria for and/or biochemical hyperandrogen­ logenesis resulting in anovulation due polycystic ovary syndrome are used ism and either ovarian dysfunction or to a relative FSH deficiency. commonly ( Table 1 ). All require the polycystic ovaries).13 Evidence also suggests that the ovarian hyperandrogenism seen in PCOS is primary, with abnormal Table 1. Diagnostic criteria for polycystic ovary syndrome. ovarian steroidogenesis through over­ National Institutes of European Society for Human Androgen Excess Society expression of the CYP17 gene being Health criteria (1990) Reproduction and Embryology criteria (2006) responsible for androgen biosynthe­ • Must meet both criteria and American Society for • Must meet both criteria Reproductive Medicine Rotterdam sis, as well as increased expression criteria (2003) of the LH receptor, which would po­ • Must meet two of three criteria tentially render the ovarian theca cells Clinical and/or biochemical Clinical and/or biochemical Clinical and/or biochemical more sensitive to LH stimulation.7,8 evidence of evidence of hyperandrogenism evidence of The ovarian hyperandrogenism ap­ hyperandrogenism hyperandrogenism pears to play a role in the appearance Menstrual dysfunction Oligoovulation or anovulation Ovarian dysfunction or of the polycystic ovary on ultrasound polycystic ovaries and the follicular arrest and anovula­ Polycystic ovaries tion that is prevalent in PCOS. The Exclusion of other known Exclusion of other known disorders Exclusion of other known ovarian phenotype may result from disorders is required as is required as well. disorders is required as well. either endogenous or exogenous an­ well.

bc medical journal vol. 60 no. 4, may 2018 bcmj.org 211 Polycystic ovary syndrome

The Rotterdam criteria are the unit. Hirsutism is most commonly as­ the mid-luteal phase (day 21 to 23) of most widely used for PCOS diagnosis, sessed using the modified Ferriman- the menstrual cycle. If ovulation has and like the more liberal AES criteria Gallwey scale to quantify the amount occurred, the level will be 10 nmol/L they allow for different phenotypes of hair growth on various androgen- or higher. of the disorder. Prevalence estimates dependent body areas. However, A diagnosis of PCOS rarely re­ for PCOS obtained using the Rotter­ race and ethnicity play a significant quires the use of ultrasound to con­ dam and AES criteria (12% to 18%) part in hirsutism.16 Additionally, the firm polycystic-appearing ovaries. As were up to twice that obtained using Ferriman-Gallwey scoring system symptom management is the focus in the NIH criteria (9%).14 NIH-defined can be somewhat impractical in ev­ PCOS, ultrasound adds little clinical PCOS is the most common pheno­ eryday clinical practice, and will be value. However, ultrasound may be type, and women with this phenotype affected by a patient’s recent waxing, warranted for investigating a pelvic are most at risk of developing repro­ shaving, or other depilation. Practi­ mass, infertility, or pelvic pain. It is ductive and metabolic abnormalities, cally, hirsutism remains largely a self- important, when possible, that the ul­ specifically type 2 diabetes. Women reported symptom. In the absence of trasound be performed with the use with the Rotterdam PCOS phenotype hirsutism, acne may be considered a of an endovaginal ultrasound probe. without hyperandrogenism are least clinical marker of hyperandrogenism. Furthermore, obtaining an antral at risk of reproductive and metabolic A strict definition of biochemical follicle count in each ovary (all fol­ abnormalities.15 hyperandrogenism in PCOS does not licles 2 to 9 mm) is important since exist. A free testosterone index and the ultrasound diagnostic criteria for Evaluation a free androgen index are thought to PCOS were established by repro­ As PCOS is ultimately a diagnosis be the most sensitive markers of bio­ ductive endocrinologists rather than of exclusion, other endocrinopathies chemical hypernadrogenemia by the radiologists. Finally, there is a signifi­ that have clinical features similar to authors of the Rotterdam criteria.12 cant overlap between the diagnoses those of PCOS must be considered. However, the assay methods are vari­ of polycystic-appearing ovaries and If ovulatory dysfunction exists, or­ able and have significant limitations. normal ovaries, with 30% to 50% of dering tests to rule out causes such as Total testosterone is not a sensitive women younger than 30 having 12 or thyroid dysfunction and hyperprolac­ marker of androgen excess, but mea­ more follicles per ovary.17 This indi­ tinemia (e.g., thyroid-stimulating hor­ surement may be useful if an andro­ cates that a polycystic-appearing ova­ mone, prolactin assay) is imperative. gen-secreting neoplasm is suspected. ry is not pathognomonic of PCOS. Considering the possibility of other Irregular or absent menstrual cy­ more serious causes of androgen ex­ cles are the most common clinical Management cess, such as nonclassic congenital finding of PCOS and are usually iden­ In addition to infertility caused by an­ adrenal hyperplasia (confirmed with tified during history taking.13 Menstru­ ovulation, women with PCOS are at an elevated 17-hydroxyprogesterone al cycle intervals longer than 35 days risk for obesity, hirsutism, endometrial level) and androgen-producing tumors are often anovulatory. If menstrual cancer, and type 2 diabetes and should (confirmed with total testosterone cycles are absent because of ovarian be managed accordingly ( Box ). levels twofold above upper-normal), insufficiency this will be indicated by is also recommended. In the presence a finding of significantly elevated FSH Anovulation of oligomenorrhea or amenorrhea, levels. In PCOS, which is character­ While anovulation can lead to long- measurement of serum FSH and es­ ized by euestrogenic chronic anovula­ term health consequences such as tradiol may be warranted to rule out tion, menstrual withdrawal bleeding endometrial cancer and hyperplasia, ovarian insufficiency or hypogonado­ can typically be induced by a 5-day most PCOS patients will present in­ tropic hypogonadism (hypogonadism to10-day course of progesterone or itially with infertility. It is reasonable of hypothalamic or pituitary origin). progestin. This provides further sup­ to begin by ruling out male factor The most common clinical feature port for anovulation being secondary infertility with semen analysis and to of hyperandrogenism is hirsutism: the to PCOS and not being the result of complete fallopian tube assessment if growth of excessive hair in a male- ovarian insufficiency or hypogonado­ the patient has risk factors for tubal type pattern caused by the conversion tropic hypogonadism. In women with factor infertility (prior ectopic preg­ of vellus hair to terminal hair under less severe menstrual disturbance, se­ nancy or gynecologic surgery, rup­ androgen effect on the pilosebacous rum progesterone can be measured in tured appendix, history of recurrent or

212 bc medical journal vol. 60 no. 4, may 2018 bcmj.org Polycystic ovary syndrome

severe pelvic inflammatory disease). ovulation rate, but with pregnancy Box. Diagnosing and managing polycystic Ovulation induction is the simplest and live birth rates that were similar. ovary disease syndrome and least expensive infertility therapy. Thus, there appears to be no role for For years, clomiphene, a selec­ metformin as a single agent for ovula­ • A diagnosis of polycystic ovary disease tive estrogen receptor modulator first tion induction in PCOS, and a limited syndrome (PCOS) must exclude other causes and include two of the following: shown to induce ovulation in 1961, role for metformin as adjuvant treat­ has been the mainstay of ovulation ment for ovulation induction. – Oligomenorrhea or amenorrhea. induction for PCOS. More recently, – Clinical or biochemical hyperandrogenism. letrozole, an aromatase inhibitor, has Obesity – Polycystic ovaries (> 12 follicles been used off-label for ovulation in­ Obesity is prevalent in 50% to 80% 2–9 mm or volume > 10 mL). 18 15 duction, as first described in 2001. of women with PCOS. Both the PP­ • In addition to infertility caused by A large, multicentre RCT comparing COS I trial22 and the PPCOS II trial23 anovulation, women with PCOS are at both medications in PCOS patients comparing letrozole and clomiphene risk for obesity, hirsutism, endometrial with anovulatory infertility demon­ demonstrated live birth rates approxi­ cancer, and type 2 diabetes. strated superior live birth rates in the mately twofold higher for women • Off-label use of letrozole for ovulation induction (2.5 mg, 5.0 mg, and 7.5 mg 2 letrozole arm (27.5%) compared with with a BMI less than 30 kg/m than daily, from cycle days 3–7 or 5–9) has the clomiphene arm (19.1%), with for women with a BMI greater than been found to be safe and effective. similar twin pregnancy rates (3.9% 35 kg/m2 (PPCOS I) and 39 kg/m2 • Clomiphene citrate, the drug commonly vs 6.9%).19 While letrozole for ovula­ (PPCOS II). While both studies defin­ used in the past for ovulation induction, may be difficult to obtain because tion induction in the setting of PCOS itively demonstrated that a high BMI the only manufacturer in Canada has still remains off-label use, the recent has an adverse effect on response to stopped production. discontinuation of clomiphene pro­ ovulation induction with oral agents, • Oral contraceptive use remains the duction in Canada, along with the evidence of a positive effect for weight first-line therapy for hirsutism and has demonstrated a reduction in risk of superior clinical outcomes with le­ loss in women with infertility second­ endometrial cancer. trozole, has made this the first-line ary to anovulation was lacking until • Metformin use and lifestyle interventions therapy for women with PCOS and recently. Now a trial of obese women have been found to reduce risks anovulatory infertility. Typically, with anovulatory infertility who were associated with type 2 diabetes. therapy with either agent is initiated randomly assigned to either a life­ on cycle day 3 of a spontaneous or style intervention (exercise and diet progestin-induced menstrual bleed for 6 months) or no lifestyle interven­ spontaneous live birth without fertil­ ( Table 2 ). tion has found a significant improve­ ity therapy = 6).24 Subjects in the life­ As insulin resistance is a common ment in live birth rates for women style intervention group lost 4.4 kg feature of PCOS, the use of insulin in the lifestyle intervention group on average during the 6-month pre­ sensitizing agents, particularly metfor­ (number needed to treat with lifestyle fertility treatment intervention.25 Diet min, for treatment of anovulatory in­ intervention to result in 1 additional and exercise as first-line therapy for fertility is physiologically reasonable. Early studies demonstrated ovulation Table 2. Recommendations for inducing ovulation with letrozole or clomiphene. rates of up to 90% in women treated with metformin and clomiphene,20 and Letrozole Clomiphene ovulation rates of 75% in women who 2.5 mg daily on cycle day 3−7 50 mg daily on cycle day 3−7 Initial regimen remained anovulatory on clomiphene (5 days) (5 days) single-agent therapy.21 However, the Pregnancy in Polycystic Ovary Syn­ Indication for increase Absence of ovulation Absence of ovulation 22 drome (PPCOS I) trial comparing How much to increase 2.5 mg daily increment 50 mg daily increment metformin, clomiphene, and metfor­ min plus clomiphene found metfor­ Maximum daily dose 7.5 mg daily 150 mg daily min alone was inferior to clomiphene Treatment duration 6 ovulatory cycles 6 ovulatory cycles alone and to metformin plus clomi­ Serum progesterone > 10 nmol/L Serum progesterone > 10 nmol/L phene. Metformin plus clomiphene Confirmation of ovulation outperformed clomiphene alone in at cycle day 21−23 at cycle day 21−23

bc medical journal vol. 60 no. 4, may 2018 bcmj.org 213 Polycystic ovary syndrome

anovulatory infertility in obese wo­ capacity, and reducing circulating has a theoretical role in management men with PCOS is supported by both free testosterone levels. Second, the of hirsutism, but clinical trial results common sense and well-designed progestin component suppresses pi­ have been inconsistent.28 When OCP clinical research. tuitary LH production, reducing the use alone is ineffective, it is prudent stimulation of ovarian theca cell an­ to use anti-androgen therapies in con­ Hirsutism drogen synthesis under LH stimula­ junction with OCP due to the poten­ Hirsutism is the result of elevated cir­ tion. Certain OCP progestins such as tial teratogenic effects of these agents culating free testosterone acting on the drosperinone and cyproterone acetate in the case of inadvertent pregnancy. pilosebaceous unit to convert vellus function as androgen receptor antago­ hair to terminal hair. Removal of un­ nists, and have a theoretical advantage Endometrial cancer wanted hairs by electrolysis or mech­ over other progestins. OCP use of­ Endometrial cancer prevalence is anical depilation will be a temporary fers the additional benefit of reducing documented to be significantly higher (by 2.7-fold) in women with PCOS.29 However, it is difficult to determine whether PCOS is an independent risk factor for endometrial cancer because many of the common presenting fea­ tures of PCOS (obesity, infertility, As PCOS is ultimately a diagnosis of diabetes, unopposed estrogen/irregu­ lar menstrual cycles) are independent exclusion, other endocrinopathies risk factors for endometrial cancer. that have clinical features similar to Regardless, heightened awareness and monitoring for endometrial can­ those of PCOS must be considered. cer in women with PCOS is war­ ranted, and risk-reduction strategies should be undertaken. While weight loss and exercise are recommended for managing PCOS in obese women, evidence for these as effective ther­ apy for protection against endomet­ solution if the underlying endocrine acne, if present, and provides protec­ rial cancer is lacking. disorder is not treated. The androgen tion against endometrial cancer and Oral contraceptive use has consis­ effect responsible for hirsutism can menstrual cycle irregularity. tently been found to reduce risk of en­ potentially be reduced by decreas­ Women with hirsutism who do not dometrial cancer. OCP use appears to ing androgen production, increasing respond adequately to OCP treatment provide a risk reduction of approxi­ androgen-binding capacity to reduce may benefit from other anti-androgen mately 50%, and the protective effect circulating levels, or reducing andro­ therapies such as spironolactone or seems to last up to 20 years after stop­ gen action at the androgen receptor. finasteride. Spironolactone is a min­ ping OCP use.30 Additionally, rela­ However, individuals with hirsutism eralocorticoid antagonist that also tive reduction of endometrial cancer must be counseled to be patient, as functions as a weak androgen recep­ risk seems to be approximately two­ response to endocrine therapy takes tor antagonist. As well, spironolac­ fold lower in women who have used at least 3 to 6 months in concordance tone reduces the activity of 5-alpha the OCP for 12 years compared with with the hair growth cycle. reductase (the enzyme responsible for women using it for 4 years.31 The oral contraceptive pill (OCP) converting testosterone to the more Some women with PCOS and ir­ remains the first-line therapy for hir­ potent dihydrotestosterone), and re­ regular menstrual cycles and anovula­ sutism because of its effect on andro­ duces testosterone biosynthesis. tion resulting in unopposed estrogen gen production.26 First, the estrogen Daily doses of spironolactone (100 may not tolerate OCP therapy, or component of the OCP increases sex mg) for at least 6 months have been OCP use may be contraindicated. Cy­ hormone-binding globulin levels, re­ shown to reduce hirsutism.27 Finas­ clic progesterone therapy (e.g., 200 sulting in greater androgen-binding teride, a 5-alpha reductase inhibitor, mg Prometrium PO daily for 10 to

214 bc medical journal vol. 60 no. 4, may 2018 bcmj.org Polycystic ovary syndrome

14 days per month, 5 to 10 mg me­ 9. Jansen E, Laven JS, Dommerholt HB, et droxyprogesterone PO daily for 10 to Competing interests al. Abnormal gene expression profiles in 14 days per month) may be a reason­ Dr Havelock has received honoraria and human ovaries from polycystic ovary syn- able option for inducing cyclic menses speaking fees from EMD Serono and drome patients. Mol Endocrinol 2004; and providing progestational effect Merck, companies engaged in producing 18:3050-3063. against unopposed estrogen. Alterna­ fertility treatments and technology. He has 10. Musso C, Cochran E, Moran SA, et al. tively, a progestin-releasing intrauter­ also received a research grant from Ferring Clinical course of genetic diseases of the ine contraceptive device may provide Pharmaceuticals, a company engaged in insulin receptor (type A and Rabson- similar, noncontraceptive benefit.32 developing and marketing products for re- Mendenhall syndromes): a 30-year pro- While all of these therapies represent productive health. spective. Medicine (Baltimore) 2004;83: off-label use, they are generally ac­ 209-222. cepted as appropriate therapy in the References 11. Nestler JE, Powers LP, Matt DW, et al. A circumstances. 1. Azziz R, Adashi EY. Stein and Leventhal: direct effect of hyperinsulinemia on se- 80 years on. Am J Obstet Gynecol 2016; rum sex hormone-binding globulin levels Type 2 diabetes 214:247-256. in obese women with the polycystic ovary When controlling for other risk fac­ 2. Stein IF, Leventhal ML. Amenorrhea as- syndrome. J Clin Endocrinol Metab 1991; tors, PCOS remains an independent sociated with bilateral polycystic ovaries. 72:83-89. risk factor for developing impaired Am J Obstet Gynecol1935;29:181-191. 12. Rotterdam ESHRE/ASRM-Sponsored glucose tolerance (RR 2.5) and type 2 3. Rosenfield RL, Ehrmann DA. The patho- PCOS consensus workshop group. Re- diabetes (RR 4.0).33 Although manage­ genesis of polycystic ovary syndrome vised 2003 consensus on diagnostic crite- ment of diabetes is beyond the scope of (PCOS): The hypothesis of PCOS as func- ria and long-term health risks related to this article, it is possible to recommend tional ovarian hyperandrogenism revisit- polycystic ovary syndrome (PCOS). Hum exercise and weight loss to reduce the ed. Endocr Rev 2016;37:467-520. Reprod 2004;19:41-47. risk of progression from impaired glu­ 4. Diamanti-Kandarakis E, Dunaif A. Insulin 13. Azziz R, Carmina E, Dewailly D, et al. cose tolerance to diabetes.34 Metformin resistance and the polycystic ovary syn- Positions statement: Criteria for defining use may also be considered, given that drome revisited: An update on mecha- polycystic ovary syndrome as a predomi- it is known to have a modest effect de­ nisms and implications. Endocr Rev nantly hyperandrogenic syndrome: An spite being less effective than lifestyle 2012;33:981-1030. Androgen Excess Society guideline. J Clin intervention in reducing diabetes risk. 5. Yen SS, Vela P, Rankin J. Inappropriate Endocrinol Metab 2006;91:4237-4245. Patients with PCOS should be mon­ secretion of follicle-stimulating hormone 14. March WA, Moore VM, Willson KJ, et al. itored for diabetes if they have a BMI and luteinizing hormone in polycystic ovar- The prevalence of polycystic ovary syn- greater than 30 kg/m2 (or greater than ian disease. J Clin Endocrinol Metab drome in a community sample assessed 25 kg/m2 for Asian patients) or have a 1970;30:435-442. under contrasting diagnostic criteria. Hum family history of diabetes, acanthosis 6. Hayes MG, Urbanek M, Ehrmann DA, et Reprod 2010;25:544-551. nigricans, or hyperandrogenism with al. Genome-wide association of polycystic 15. Dumesic DA, Oberfield SE, Stener-Victo- anovulation.29 ovary syndrome implicates alterations in rin E, et al. Scientific statement on the gonadotropin secretion in European an- diagnostic criteria, epidemiology, patho- Summary cestry populations. Nat Commun 2015; physiology, and molecular genetics of Polycystic ovary syndrome remains a 6:7502. polycystic ovary syndrome. Endocr Rev prevalent reproductive and metabolic 7. Comim FV, Teerds K, Hardy K, Franks S. 2015;36:487-525. disorder with variable phenotypes and Increased protein expression of LHCG re- 16. Escobar-Morreale HF, Carmina E, Dewail- an underlying pathophysiology that ceptor and 17α-hydroxylase/17-20-lyase in ly D, et al. Epidemiology, diagnosis and is not completely understood. Mak­ human polycystic ovaries. Hum Reprod management of hirsutism: A consensus ing a diagnosis of PCOS is beneficial 2013;28:3086-3092. statement by the Androgen Excess and but not essential. Therapy remains 8. Wood JR, Nelson VL, Ho C, et al. The mo- Polycystic Ovary Syndrome Society. Hum focused on managing symptoms (in­ lecular phenotype of polycystic ovary syn- Reprod Update 2012;18:146-170. fertility caused by anovulation, obes­ drome (PCOS) theca cells and new candi- 17. Dewailly D, Lujan ME, Carmina E, et al. ity, hirsutism) and reducing long-term date PCOS genes defined by microarray Definition and significance of polycystic health risks (endometrial cancer, type analysis. J Biol Chem 2003;278:26380- ovarian morphology: A task force report 2 diabetes). 26390. from the Androgen Excess and Polycystic

bc medical journal vol. 60 no. 4, may 2018 bcmj.org 215 Polycystic ovary syndrome

Ovary Syndrome Society. Hum Reprod 24. van Oers AM, Groen H, Mutsaerts MA, et Amsterdam ESHRE/ASRM-Sponsored Update 2014;20:334-352. al. Effectiveness of lifestyle intervention 3rd PCOS Consensus Workshop Group. 18. Mitwally MF, Casper RF. Use of an aro- in subgroups of obese infertile women: A Fertil Steril 2012;97:28-38.e25. matase inhibitor for induction of ovulation subgroup analysis of a RCT. Hum Reprod 30. Cibula D, Gompel A, Mueck AO, et al. Hor- in patients with an inadequate response 2016;31:2704-2713. monal contraception and risk of cancer. to clomiphene citrate. Fertil Steril 2001; 25. Mutsaerts MA, van Oers AM, Groen H, et Hum Reprod Update 2010;16:631-650. 75:305-309. al. Randomized trial of a lifestyle program 31. Schlesselman JJ. Risk of endometrial 19. Legro RS, Brzyski RG, Diamond MP, et al. in obese infertile women. N Engl J Med cancer in relation to use of combined oral Letrozole versus clomiphene for infertility 2016;374:1942-1953. contraceptives. A practitioner’s guide to in the polycystic ovary syndrome. N Engl 26. Goodman NF, Cobin RH, Futterweit W, et meta-analysis. Hum Reprod 1997;12: J Med 2014;371:119-129. al. American Association of Clinical Endo- 1851-1863. 20. Nestler JE, Jakubowicz DJ, Evans WS, crinologists, American College of Endocri- 32. Bahamondes L, Valeria Bahamondes M, Pasquali R. Effects of metformin on spon- nology, and Androgen Excess and PCOS Shulman LP. Non-contraceptive benefits taneous and clomiphene-induced ovula- Society disease state clinical review: of hormonal and intrauterine reversible tion in the polycystic ovary syndrome. N Guide to the best practices in the evalua- contraceptive methods. Hum Reprod Up- Engl J Med 1998;338:1876-1880. tion and treatment of polycystic ovary syn- date 2015;21:640-651. 21. Vandermolen DT, Ratts VS, Evans WS, et drome—Part 1. Endocr Pract 2015;21: 33. Moran LJ, Misso ML, Wild RA, Norman al. Metformin increases the ovulatory rate 291-300. RJ. Impaired glucose tolerance, type 2 and pregnancy rate from clomiphene ci- 27. Brown J, Farquhar C, Lee O, et al. Spirono- diabetes and metabolic syndrome in poly- trate in patients with polycystic ovary syn- lactone versus placebo or in combination cystic ovary syndrome: A systematic re- drome who are resistant to clomiphene with steroids for hirsutism and/or acne. view and meta-analysis. Hum Reprod Up- citrate alone. Fertil Steril 2001;75:310- Cochrane Database Syst Rev 2009;(2): date 2010;16:347-363. 315. CD000194. 34. Knowler WC, Barrett-Connor E, Fowler 22. Legro RS, Barnhart HX, Schlaff WD, et al. 28. van Zuuren EJ, Fedorowicz Z, Carter B, SE, et al. Reduction in the incidence of Clomiphene, metformin, or both for infer- Pandis N. Interventions for hirsutism (ex- type 2 diabetes with lifestyle intervention tility in the polycystic ovary syndrome. N cluding laser and photoepilation therapy or metformin. N Engl J Med 2002;346: Engl J Med 2007;356:551-566. alone). Cochrane Database Syst Rev 393-403. 23. Legro RS, Brzyski RG, Diamond MP, et al. 2015;2(4):CD010334. Letrozole versus clomiphene for infertility 29. Fauser BC, Tarlatzis BC, Rebar RW, et al. in the polycystic ovary syndrome. N Engl Consensus on women’s health aspects of J Med 2014;371:119-129. polycystic ovary syndrome (PCOS): The

216 bc medical journal vol. 60 no. 4, may 2018 bcmj.org