Sie sind auf Seite 1von 18

PRIMER

Polycystic ovary syndrome


Ricardo Azziz1,2, Enrico Carmina3, ZiJiang Chen4,5, Andrea Dunaif 6, Joop S.E.Laven7,
Richard S.Legro8, DariaLizneva1,9, BarbaraNattersonHorowtiz10, HelenaJ.Teede11
andBulentO.Yildiz12
Abstract | Polycystic ovary syndrome (PCOS) affects 520% of women of reproductive age worldwide.
The condition is characterized by hyperandrogenism, ovulatory dysfunction and polycystic ovarian
morphology (PCOM) with excessive androgen production by the ovaries being a key feature of
PCOS. Metabolic dysfunction characterized by insulin resistance and compensatory
hyperinsulinaemia is evident in the vast majority of affected individuals. PCOS increases the risk for
type2 diabetes mellitus, gestational diabetes and other pregnancy-related complications, venous
thromboembolism, cerebrovascular and cardiovascular events and endometrial cancer. PCOS is a
diagnosis of exclusion, based primarily on the presence of hyperandrogenism, ovulatory dysfunction
and PCOM. Treatment should be tailored to the complaints and needs of the patient and involves
targeting metabolic abnormalities through lifestyle changes, medication and potentially surgery for
the prevention and management of excess weight, androgen suppression and/or blockade,
endometrial protection, reproductive therapy and the detection and treatment of psychological
features. This Primer summarizes the current state of knowledge regarding the epidemiology,
mechanisms and pathophysiology, diagnosis, screening and prevention, management and future
investigational directions of the disorder.

Polycystic ovary syndrome (PCOS) is a common dis Childrenmight present with premature pubarche and
order in women that is characterized by hyperandro adolescents with early signs of androgenization (for
genism (that is, evidence of excess male hormone or example, acne and hirsutism) and menstrual irregularity.
androgen effect; for example, clinically, such as hirsut Postmenopausal women with PCOS carry an increased
ism, and/or biochemically, such as hyperandrogenaemia risk for metabolic and cardiovascular comorbid
or excess levels of androgen), ovulatory dysfunction ities, although hyperandrogenic symptoms ameliorate
(including menstrual dysfunction) and polycystic duringmenopause.
ovarian morphology (PCOM; an excessive number of Women with PCOS have an increased risk for meta
preantral follicles in the ovaries). The clinical presenta bolic abnormalities and T2DM, infertility, obstetrical
tion is heterogeneous and can be categorized in several complications, endometrial cancer and mood disorders.
phenotypes, depending on the presence or absence of These women also probably have an increased risk for
characteristic features (FIG.1). Metabolic abnormalities, cardiovascular and cerebrovascular events, venous
mainly insulin resistance and compensatory hyper thromboembolism and ovarian cancer.
insulinaemia, are evident in a majority of affected
individuals1, especially among those women who also Epidemiology
show hyperandrogenism2. Between 1 in 6 and 1 in 20 Prevalence
women of reproductive age (520%) are affected by the The prevalence of PCOS is remarkably similar world
Correspondence to R.A.
Department of Obstetrics and
disorder worldwide. In 2004, the economic impact of wide. The prevalence of clinically evident PCOS in
Gynecology, Medical College PCOS exceeded US$4billion in the United States alone, women of reproductive age from the United States,
of Georgia, Augusta even without considering the cost of the increased risk Europe, Asia and Australia ranges between 5% and 9%5
University, 1120 15th Street, of obstetrical complications, type2 diabetes mellitus based on the original 1990 US National Institutes of
CB2209, Augusta,
(T2DM) and other disorders3. Health (NIH) diagnostic criteria6 (BOX1; FIG.1). Using the
Georgia30912, USA.
RAZZIZ@augusta.edu Although signs and symptoms are most evident in broader 2003 Rotterdam criteria7,8 (FIG.1), now endorsed
women of reproductive age, the disorder also carries by the NIH and accepted internationally, the prevalence
Article number: 16057
doi:10.1038/nrdp.2016.57 risk and symptoms in prepuberty and postmeno of PCOS ranges from 5.5% to 19.9%5. The larger vari
Published online 11 Aug 2016 pause, which are only now beginning to be identified4. ation using the 2003 Rotterdam criteria might be the

NATURE REVIEWS | DISEASE PRIMERS VOLUME 2 | 2016 | 1



2
0
1
6
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
PRIMER

Author addresses should be considered abnormal in white or black, and


probably even in Mongoloid or Asiatic, women11. Indeed,
1
Department of Obstetrics and Gynecology, Medical in a study of 228 patients with minimal unwanted hair
College of Georgia, Augusta University, 1120 15th Street, growth (with an mFG score of 5), >50% demonstrated
CB2209, Augusta, Georgia30912, USA. an androgen excess disorder 12. Thus, many women
2
Department of Medicine, Medical College of Georgia,
have excessive or unwanted hair growth but may not
Augusta University, Augusta, Georgia, USA.
3
Department of Health Sciences and Mother and Child be afforded appropriate evaluation because they are
Care, University of Palermo, Palermo, Italy. notdeemed to be sufficiently hirsute.
4
Center for Reproductive Medicine, Shandong Provincial
Hospital Affiliated to Shandong University, Jinan, China. Metabolic dysfunction
5
Shanghai Key Laboratory for Assisted Reproduction Many women with PCOS demonstrate basal and glucose-
andReproductive Genetics, Shanghai, China. stimulated hyperinsulinaemia and are insulin resist
6
Department of Medicine, Northwestern University ant, independent of body mass index (BMI). Insulin-
Feinberg School of Medicine, Chicago, Illinois, USA. mediated glucose disposal, quantitated by euglycaemic
7
Division of Reproductive Medicine, Department of OBGYN, clamp, was on average 3540% lower in patients with
Erasmus Medical Centre, Rotterdam, The Netherlands.
PCOS than in matched controls1,13.
8
Department of Obstetrics and Gynecology, College of
Medicine, Pennsylvania State University, Hershey, Data indicate that the incidence of metabolic syn
Pennsylvania, USA. drome, gestational diabetes mellitus, impaired glucose
9
Department of Reproductive Health Protection, Scientific tolerance (IGT) and T2DM is increased in premenopausal
Center of Family Health and Human Reproduction, Irkutsk, women with PCOS compared with age-matched and
Russian Federation. BMI-matched controls14. No less than 2% of women with
10
Division of Cardiology, David Geffen School of Medicine, PCOS progress from baseline normoglycaemia to T2DM
University of California, Los Angeles, California, USA. every year and 16% progress from IGT to T2DM15,16. The
11
Monash Centre for Health Research and Implementation, prevalence of T2DM in PCOS continues to increase dur
School of Public Health and Preventive Medicine, Monash ing the late reproductive years17,18. Data on the prevalence
University in partnership with Monash Health, Clayton,
of T2DM in postmenopausal women with PCOS are lim
Victoria, Australia.
12
Division of Endocrinology and Metabolism, Department ited, but the few reports available do not demonstrate a
of Internal Medicine, Hacettepe University School of further substantial increase in the incidence of IGT or
Medicine, Hacettepe, Ankara, Turkey. T2DM in postmenopausal women with PCOS18,19.
In a systematic review and meta-analysis on meta
bolic syndrome, the prevalence of IGT (odds ratio (OR):
consequence of variation in the sensitivity of the tests 2.4; 95% CI: 1.44.5) and T2DM (OR: 4.43; 95%CI:
used for detecting the disorder. 4.14.8) was far higher in patients with PCOS than in
The characteristic features of PCOS lead to several controls and the risk of metabolic disease in those with
different phenotypes (FIG.1). The phenotypic distrib PCOS was clearly demonstrated14. In this systematic
ution of PCOS in epidemiological studies in unselected review, on subgroup analysis, increased prevalence of
(unbiased) populations is 4045% for phenotype A and IGT and metabolic syndrome was observed when com
phenotypeB combined, ~35% for phenotypeC and ~20% paring lean women with and without PCOS, although
for phenotypeD5. An important point to keep in mind no study actually compared the prevalence of T2DM in
is that only studies in unselected (medically u nbiased) thesepopulations.
populations will enable us to clearly identify the true
phenotype of PCOS. Compared with unselected patients, Obesity
patients with PCOS in the clinical setting are more obese, The effect of obesity on PCOS and PCOS on obesity is
more hirsute, more hyperandrogenaemic and show a complex, and strong evidence of an association is cur
greater proportion of phenotypeA and phenotypeB9, rently lacking. Although PCOS occurs in obese and lean
reflecting substantial referralbias. women, a recent systematic review and meta-analysis
concluded that obesity was more prevalent in women
Dermatological abnormalities with PCOS than in women without PCOS20. However,
Hirsutism, acne and androgenic alopecia are clinical all but two of the studies reviewed recruited their
signs and symptoms of hyperandrogenism. In clin patients from hospitals or clinics. By contrast, studies in
icalstudies, hirsutism affects ~6575% of black and unselected (medically unbiased) populations have sug
white patients with PCOS, which is dramaticallyhigher gested that BMI distribution was more similar between
than otherwise expected (02%)10. Acne affects 1525% patients with PCOS identified in unselected populations
of patients with PCOS and varies with e thnicity; and controls and that BMI was higher than in patients
although it is unclear whether the prevalence of acne with PCOS in referral (clinically biased) settings9,21. These
is significantly increased in these patients over that data indicate that much of the obesity in women with
observed in the generalpopulation10. PCOS may be driven by self-referral, as obesity is one of
Although hirsutism is frequently defined visually by the primary depressors of quality of life(QOL)2224.
a modified FerrimanGallwey (mFG) score of 6, which In addition, evidence that obesity drives the develop
corresponds to the 95th percentile of the population ment or prevalence of PCOS is conflicting. A higher
studied, other studies indicate that an mFG scoreof3 incidence of PCOS was observed among those who were

2 | 2016 | VOLUME 2 www.nature.com/nrdp



2
0
1
6
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
PRIMER

obese based on studies of women seeking bariatric sur scores seem to be significantly correlated with the
gery or dietary interventions25,26 in a retrospective study degree of insulin resistance 40. A meta-analysis of
of a national birth cohort using self-reported symptoms27 28studies including 2,384 patients with PCOS and
and in a community-based longitudinal observational 2,705 controls found that more-severe emotional dis
study 28. However, in an unselected population in the tress was present in women with PCOS than in con
United States, no significant difference in PCOS preva trols38. However, even though the hirsutism, obesity and
lence according to BMI was detected29, although another infertility associated with the syndrome were shown to
similarly designed study in Turkey showed greater rates be in some way linked to severe emotional distress in
of PCOS as mean BMI increased30. The fact that there is women with PCOS, these factors alone did not fully or
no relationship between PCOS prevalence in different consistently account for the high prevalence of anxiety
countries, which differ in mean population BMI, further anddepression38.
supports that obesity does not drive the development
of PCOS5. In addition, no association between genetic Long-term morbidity
variants in genes known to be involved in obesity and In addition to the complications of subfertility, meta
genes involved in PCOS has been found3133. bolic dysfunction and dysglycaemia, and psychological
dysfunction, patients with PCOS are at long-term risks
Subfertility for additional disorders.
PCOS is the primary cause of anovulatory subfertil
ity, with major health and economic costs; however, Obstetrical complications. Patients with PCOS are at
community-based data are limited on prevalence and risk of experiencing complications during pregnancy.
treatment trends in infertility in PCOS. Most infertil Apopulation-based study of singleton births among
ity and PCOS data are based on selected populations 3,787 women with PCOS and >1 million without
managed in hospital or in fertility clinics, and national PCOS registered in the Swedish medical birth regis
funding policies on assisted reproduction vary substan try between 1995 and 2007 indicated that pregnan
tially, making comparisons difficult. In a community- cies in women with PCOS had significantly higher
based study, the self-reported prevalence of PCOS was rates of pre-eclampsia, very preterm birth (defined as
5.8%. Infertility was noted by 72% of women reporting <32weeks of gestation in the study) and gestational dia
PCOS compared with 16% of those not reporting PCOS betes mellitus41. Infants born to women with PCOS had
(P<0.001); infertility was 15-fold higher in women a higher risk of being large for gestational age, having
reporting PCOS, independent of BMI34. One retro meconium aspiration syndrome (in which the stool of
spective study followed a cohort of 786 women with the infant defecated into the amniotic fluid, generally
PCOS who were diagnosed >30years ago from hospital under stress, is inhaled) and having a low (<7) Apgar
records35. In this selected population, 66% of women score (which assesses the condition of the newborn
reported infertility. Overall, PCOS seems to be the most by valuing breathing effort, heart rate, muscle tone,
common cause of anovulatory infertility and further reflexes and skin colour) at 5minutes. The increased
studies are needed on the natural history of this feature risk of adverse pregnancy and birth outcomes could not
in community-basedsamples. be fully explained by the use of assisted reproductive
technologies, which are used more frequently by women
Psychological manifestations with PCOS41. Other studies have confirmed these find
Depression and anxiety are more common and more ings4244, regardless of the criteria used to define PCOS.
severe in women with PCOS than in women without Metformin treatment in pregnant women with PCOS
the disorder 3638, regardless of the phenotype of PCOS does not seem to reduce pregnancy complications in
or the presence of obesity 37,39. Interestingly, depression thedisorder 45.

1990 US NIH criteria


2006 AE-PCOS criteria
2003 Rotterdam criteria

Phenotype A Phenotype B Phenotype C Phenotype D

Hyperandrogenism
Present Present Present Absent
and hirsutism

Ovulatory dysfunction Present Present Absent Present

Polycystic ovarian
Present Absent Present Present
morphology

Figure 1 | Diagnostic criteria and phenotypes of PCOS. Polycystic ovary syndrome (PCOS) NatureisReviews
classified| Disease
into Primers
fourseparate phenotypes (AD), according to the presence or absence of three characteristics: hyperandrogenism
(eitherbiochemical or clinical), ovulatory dysfunction and polycystic ovarian morphology. Only phenotype A requires
allthree features of PCOS to be present. The various diagnostic criteria currently available for PCOS include a greater
orfewer number of PCOS phenotypes. AE-PCOS, Androgen Excess-PCOS Society; NIH, National Institutes of Health.

NATURE REVIEWS | DISEASE PRIMERS VOLUME 2 | 2016 | 3



2
0
1
6
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
PRIMER

Cardiovascular and cerebrovascular complications. 20year retrospective cohort study 58 reported increased
Cardiovascular disease (CVD) markers (for example, risk of myocardial infarction in patients with PCOS as
vascular calcification and the thickness of the vas compared with controls.
cular wall) point to a higher risk of CVD in women In addition, in women with the more-severe pheno
with PCOS than in controls, although an increased type of PCOS (that is, phenotype A and phenotypeB),
number of actual cardiovascular events has been diffi the relative risk for CVD was reported to be 1.3 (REF.59).
cult to demonstrate46. Compared with healthy controls, However, negative studies have also been reported;
significant coronary calcification is more prevalent in Wildetal.60 did not find any difference in cardiovascu
women with PCOS47,48; the thickness of the intimal layer lar mortality in women with PCOS versus controls when
of the carotid wall has been reported to be greater in controlling for BMI. The Mayo Clinic cohort did not
women with PCOS49,50; the incidence of aortic calcifi reveal any increase in cardiovascular morbidity, includ
cation was reported to be higher in one study 48, and ing myocardial infarction61. A 21year longitudinal study
the Dallas heart study showed that arterial stenosis was in Sweden demonstrated that the incidence of cardio
more prevalent in women with PCOS based on coro vascular events was comparable between women with
nary angiography 51. Questions have been raised as to PCOS and controls19. To date, studies investigating the
whether these findings indicate a true increase in actual risk of cardiovascular events in PCOS are derived from
cardiovascularmortality. clinical cohort studies and are, therefore, subject to refer
Data on cardiovascular events (for example, myo ral bias9. In addition, substantial variability in ethnicity,
cardial infarctions) in PCOS are conflicting. The inci PCOS criteria and study design c onfound the findings.
dence of CVD strongly increases after 50years of age in The incidence of cerebrovascular events was slightly
the general population, and a similar increase is expected increased in older women with PCOS compared
to be present after menopause in women with PCOS52. with the general population62,63. The risk of venous
The few followup studies available do not demonstrate thromboembolisms is increased (OR: 1.5) compared
a greater number of cardiovascular events in women with BMI-matched controls64, with the risk of venous
with PCOS who are of late reproductive age than in age- thromboembolisms being twofold higher in women with
matched controls5355, although the number of patients PCOS who take oral contraceptive pills (OCPs) than in
followed is too few to be able to detect small changes the general population56,65.
in incidence. Hospitalization data in PCOS have indi
cated that the number of cardiovascular events might be Risk of malignancies. Not unexpectedly, considering the
increased during late reproductive years, although these concurrence of hyperoestrogenic anovulation and hyper
studies have many biases56. Two small studies52,57 and one insulinaemia, women with PCOS have an increased risk
for endometrial cancer (OR: 2.7). They may also have
an increased risk for ovarian cancer, although the OR
Box 1 | Diagnostic criteria for PCOS* for this malignancy is unclear. However, no associated
1990 US NIH criteria increased risk of breast cancer has been shown66.
Patients are diagnosed with polycystic ovary syndrome
(PCOS) if they have all of the following criteria: Mechanisms/pathophysiology
Oligo-anovulation The pathophysiology and intrinsic mechanisms under
Clinical and/or biochemical signs of androgen excess lying PCOS are complex because aetiologies vary and
the different features are considerably intertwined
2003 Rotterdam criteria
(FIG.2). The interplay between these mechanisms results
Patients are diagnosed with PCOS if they have two of the
following three criteria: in and perpetuates the clinical features of PCOS, includ
ing hyperandrogenism, PCOM and ovulatory dysfunc
Oligo-anovulation
tion, in addition to the associated mood disturbances,
Clinical and/or biochemical signs of androgen excess
psychosexual dysfunction and long-term morbidities.
Polycystic ovarian morphology (PCOM) Inaddition, the development of PCOS has a strong
2006 Androgen Excess-PCOS Society criteria|| geneticcomponent.
Patients are diagnosed with PCOS if they have all of the
following criteria: Genetic factors
Clinical and/or biochemical signs of androgen excess Familial clustering and the results from twin studies
Ovarian dysfunction, including oligo-anovulation strongly support an underlying genetic basis for PCOS.
and/or PCOM For example, having a mother or sister with PCOS con
*All criteria require the exclusion of similar, mimicking veys a 3050% risk of developing PCOS6769. The corre
disorders, such as thyroid dysfunction, hyperprolactinaemia, lation for PCOS between monozygotictwinsisters was
adrenal hyperplasia, androgen-secreting tumours and twice as high as the dizygotictwincorrelation. Genetic
iatrogenic androgen excess, among others. See REF.6.
factors were suggested to explain 66% of the variance

Criteria proposed by the European Society for Human
Reproduction and Embryology and the American Society according to the univariate genetic model2. To date, large
forReproductive Medicine expert conference held in numbers of genetic studies have identified almost 100
Rotterdam7,8. ||Criteria proposed by an expert Task Force of the susceptibility genes related to PCOS. Although the can
Androgen Excess and PCOS Society140. NIH, National Institutes didate gene approach is reasonable to explore the genetic
of Health.
origin of PCOS, it is neither an efficient nor a consistent

4 | 2016 | VOLUME 2 www.nature.com/nrdp



2
0
1
6
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
PRIMER

Progesterone Phenotypes associated with the variants discovered


Hypothalamus may also provide an insight into the pathophysiol
Abnormal ogy conferred by the susceptibility genes. Genotype
GnRH phenotype correlation studies in Han Chinese women
pulsation
Insulin resistance Pituitary with PCOS demonstrated that the THADA and
Insulin resistance gland DENND1A variants were associated with endocrine and
of muscle and liver metabolic disturbances78. In populations of European
Adipocyte dysfunction
ancestry, DENND1A was observed to be a risk allele for
androgen excess and anovulation79, a variant near FSHR
Abnormal
Hyperinsulinaemia gonadotropin levels was associated with lower levels of follicle-stimulating
LH/FSH ratio hormone (FSH)79 and a variant near RAB5B seemed to
be associated with glucose m etabolism dysfunction80.
Pancreas Studies in women with a single PCOS clinical feature
identified a specific genetic association for LHCGR and
Ovary INSR with anovulation, and THADA and DENND1A
with polycystic ovaries. C9orf3 and rs4385527 con
Ovulatory dysfunction ferred a particular risk for all three of the definitive
manifestations of PCOS (that is, hyperandrogenism,
ovulatory dysfunction and PCOM), which suggests
Hyperandrogenism Follicular arrest their fundamental role in the aetiology of the disorder 81.
In silico analysis based on data from GWAS is another
Adrenal androgens SHBG levels in the liver PCOM approach that may assist in deciphering the mech
anisms underlying PCOS. According to pathway analy
Figure 2 | The pathophysiology of PCOS. The pulsatile release of gonadotropin- sis, INS, GNAQ, PLCB3, STXBP1, SMC3, PLCB2 and
Nature Reviews | Disease Primers
releasing hormone (GnRH) from the hypothalamus is often disturbed in polycystic ovary PLCZ1 are significantly associated with oocyte m eiosis
syndrome (PCOS), leading to luteinizing hormone (LH) hypersecretion by the pituitary
and the regulation of insulin secretion82.
gland, which induces ovulary dysfunction and hyperandrogenism. This perturbed
secretion of LH seems to arise early in puberty and is related to disturbed inhibition of In addition, functional studies focusing on the loci
GnRH secretion by progesterone. Although serum follicle-stimulating hormone (FSH) elucidated by GWAS have been performed. A recent
levels are generally normal, follicles seem to be more resistant to FSH in women with study measured DNA methylation and gene expres
PCOS than in controls. This effect might be due to increased levels of intra-ovarian sion of 11 Chinese GWAS risk loci in subcutaneous
anti-Mllerian hormone (AMH). Notably, genetic and epigenetic variants contribute adipose tissue of patients with PCOS. This study found
considerably to susceptibly for most of these alterations. Environmental factors that the genetic variants in LHCGR and INSR might
contribute somewhat less, most by exacerbating insulin resistance and dysregulated have changed the expression level via modification on
gonadotropin secretion. PCOM, polycystic ovarian morphology; SHBG, sex methylation. Hypomethylation of LHCGR was con
hormone-binding globulin. cordant with LHCGR overexpression in non-obese
patients, but not in the obese ones, whereas hyper
method for such a complex polygenetic disease70 (BOX2). methylation of INSR was not associated with different
High-throughput genome-wide association studies gene expression between obese and non-obese women
(GWAS) provide a more comprehensive, unbiased, with PCOS. In this study, no significant difference was
discovery-driven approach to explore the genetic basis found in genes of other GWAS loci, after correction for
of complex disorders (TABLE1). multipletesting 83.
Two GWAS in Han Chinese women with PCOS Despite the vast progress in the identification of
identified 11 susceptibility loci71,72. Some of these genes, PCOS loci, the quantitative traits associated with the
such as INSR, FSHR and C9orf3, have been confirmed disorder and the underling mechanisms are still largely
in subsequent family-based studies73,74. A genome-wide unknown. However, the elucidation of genotype
association study in a population of white women with phenotype associations should be the aim in the
European ancestry identified two novel loci in the region postGWAS era.
of GATA4NEIL2 and FSHBARL14EP 75. The loci near
C9orf3, which was also observed in Han Chinese studies, Gonadotropic derangements
was also confirmed. In a second genome-wide association In normal circumstances, immature oocytes mature
study including white women of European descent, three under the influence of several hormones, most
novel susceptibility genes ERBB4, RAD50 and KRR1 notably FSH, and ovulation as well as final maturation
were observed with genome-wide significance. The occur upon luteinizing hormone (LH) stimulation.
Mendelian randomization analyses suggested a causal Aneuroendocrine abnormality in PCOS may include
role of PCOS risk single-nucleotide polymorphisms for increased gonadotropin-releasing hormone (GnRH)
higher BMI, insulin resistance and lower levels of sex pulse frequency, which increases the frequency and
hormone-binding globulin (SHBG) in PCOS. Other pulse amplitude of LH over FSH production. This
previously reported genes, namely, YAP1, THADA and abnormality results in increased circulating LH/FSH
FSHB, were also replicated76. Notably, the loci identified ratio and is frequently observed in lean, but not obese,
in GWAS so far account for perhaps no more than 10% women with PCOS 84,85. The finding that increased
of the heritability of the disorder 77. LH pulses and enhanced daytime LH pulse secretion

NATURE REVIEWS | DISEASE PRIMERS VOLUME 2 | 2016 | 5



2
0
1
6
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
PRIMER

Box 2 | Selected genetic variants associated with PCOS LH hypersecretion is also detrimental to normal fol
licular growth and might cause premature luteinization
Candidate gene studies70 have identified genes involved in: of granulosa cells (leading to hypertrophy, lipid accu
Androgen biosynthesis: CYP1A1, CYP11A, CYP17A1, CYP19 and HSD17B6 mulation and other changes in the follicle that normally
Androgen action: AR, SHBG, SRD5A1 and SRD5A2 occur after ovulation)87. Follicular growth often resumes
Insulin signalling: INSR, IRS1, IRS2, PPARG and CAPN10 following the replacement of oestradiol and progester
Metabolism: ADIPOQ and FTO one during the luteal phase of the menstrual cycle in
Folliculogenesis: FSHR, LHCGR and AMHR2 women with PCOS characterized by anovulation, which
might be related to the reduction in the LH/FSH ratio
Inflammation: IL1A, IL1B, IL6, IL18, PAI1, FBN3, TNF and MEP1A
and ovarian FSH resistance95.
Similarly, reduction in negative-feedback mech
anisms at the level of the pituitary gland by adminis
are already observed early during puberty in girls trating anti-oestrogens or by reducing the conversionof
with hyperandrogenism indicates that abnormalities androgens to oestrogens through the administration
in the pulsatile release of GnRH might underlie the ofaromatase inhibitors will lead to increased release of
development of PCOS, at least in some patients86. The FSH and subsequent resumption of follicular growth
increased LH/FSH ratio, and the resistance to FSH in anovulatory PCOS87. Overstimulation of LH induces
in the ovaries (see below), further enhances hyper hypersecretion of theca cell-derived androgens, which
secretion of androgens in theca cells in ovarian f ollicles, further impairs follicular maturation by promoting the
which impairs follicular development 86 and reduces initiation of primordial follicle growth and increas
the inhibition of GnRH pulse frequency by proges ing the number of growing small antral follicles 96.
terone, further promoting the development of the Overstimulation of theca cells by LH is further exacer
PCOSphenotype86. bated by the gonadotropic action of insulin on theca
cells, acting either directly through the insulin receptor
Ovarian follicular arrest or indirectly through the IGF1 receptor 97.
The coordination and interaction of LH, FSH, insulin- Intra-ovarian factors that modulate follicular
like growth factor 1 (IGF1), anti-Mllerian hormone recruitment and growth, including members of the
(AMH), enzymes involved in androgen conversion and transforming growth factor family (for example,
possibly other factors are disturbed in PCOS leading AMH, inhibins, activins, bone morphogenetic pro
to oligo-ovulation (irregular ovulation) or anovulation teins and growth differentiation factors (GDFs)), other
(the absence of ovulation)87 (FIG.3). In PCOS, the selec growth factors and cytokines might also contribute to
tion of a dominant follicle (that is, the follicle that the abnormal follicle development and function seen
proceeds to ovulation in each cycle) does not occur in PCOS87. The oocyte and its surrounding granulosa
regularly 88, which is a consequenceof insufficient cells produce many of these factors98. For example,
secretion of FSH and local inhibition ofFSHaction. oocyte-derived GDF9 is crucial for normal follicu
Follicular FSH resistance might be caused by other logenesis and is dysregulated in women with PCOS99.
intra-ovarian regulators of FSH action. One such factor Although inhibins, activins, follistatin and IGF1 all have
is the increased levels of AMH in PCOS, which might a crucial role in folliculogenesis, their possible permis
reduce the FSH sensitivity of individual ovarian folli sive role in the pathophysiology of ovarian dysfunction
cles89 and block the conversion of androgens to oestro in women with PCOS remains to be demonstrated87.
gens via the inhibition of aromatase activity, thereby
further contributing to hyperandrogenism. Finally, Insulin resistance and hyperinsulinaemia
genetic variations from normal in the FSH molecule Under normal circumstances, as insulin sensitivity
itself and in its receptor might be partially responsible decreases, insulin secretion increases to maintain a
for some of the differences in FSH sensitivity between constant hyperbolic relationship, a relationship that is
patients with PCOS and healthycontrols90. expressed by the Disposition Index 100. In women with
Increased circulating levels of AMH arise as a PCOS, basal insulin secretion rates are increased101,
consequence of an increased number of small antral although insulin secretory responses to a glucose load
(maturing) follicles and increased production of AMH are generally inadequate, resulting in a lower Disposition
per follicle91. Although reduced levels of AMH in small Index than age-matched and BMI-matched control
primordial and transitional follicles of women with women101103. Thus, despite the presence of hyperinsulin
anovulatory PCOS can initially promote the recruit aemia, women with PCOS have relative pancreatic cell
ment of additional growing follicles92, hypersecretion dysfunction104. Women with PCOS also demonstrate
of AMH in granulosa cells of more-mature small antral decreased hepatic extraction of insulin105,106, which
follicles could subsequently impair further follicular contributes to their hyperinsulinaemia.
growth by inhibiting FSH and aromatase action89,93. The molecular mechanisms that drive insulin
Thus, in patients with anovulatory PCOS, circulat resistance in PCOS differ from those in other common
ing FSH levels, although at low-to-normal concen insulin-resistant states, such as obesity and T2DM.
trations, generally will not be enough to overcome Inmuscle, serine phosphorylation of the i nsulinrecep
the inhibition of aromatase activity by AMH in the tor and of insulin receptor substrate1 (IRS1) is incre
antralfollicle94. ased107, resulting in impaired insulin signalling 108,109

6 | 2016 | VOLUME 2 www.nature.com/nrdp



2
0
1
6
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
PRIMER

and in constitutive activation of MEK1 and MEK2 strong evidence supports that adipocytes and adipo
(MEK1/2) in PCOS110,111 (FIG.4). The PCOS-associated cyte function are aberrant in PCOS, favouring insulin
insulin resistance is selective, affecting metabolic, but resistance and subclinical inflammation. The periph
not mitogenic, signalling pathways110,112, which might eral insulin resistance observed in PCOS might be the
explain the paradox of the persistent reproductive result, at least in part, of adipocyte dysfunction (FIG.5).
actions of insulin in the face of systemic insulin resist For e xample, inflammatory cytokines (such as tumour
ance. Defects in insulin signalling persist in cultured necrosis factor and IL6) suppress insulin-mediated glu
cells107,113. Two studies have suggested that skeletal cose transport more in adipocytes derived from patients
muscle is no longer insulin resistant in long-term cul with PCOS than in adipocytes derived from matched
ture113,114, whereas one study found persistent defects in controls116. Women with PCOS seem to have larger
insulin responsiveness in cultured PCOS myotubes115. adipocytes117, lower lipoprotein lipase activity 117 and
The reasons for these discrepant findings are unclear, impaired catecholamine-induced lipolysis118 compared
although these observations suggest that both intrin with matchedcontrols.
sic abnormalities and extrinsic factors in the invivo Inflammatory cytokines also suppress adiponectin
environment account for insulin resistance in skeletal secretion to a greater degree in adipocytes derived from
muscle of women with PCOS113,114. Abnormalities in patients with PCOS than in matched controls, favouring
insulin action are also observed in adipose tissue and the development of a more pro-inflammatory, insulin-
adipocytes in women with PCOS, although the nature resistant environment116. Glucose transporter4 (GLUT4;
of the defects differs113,115. also known as SLC2A4) protein expression is decreased
in adipocytes in PCOS119, similar to levels observedin
Adipose tissue dysfunction adipocytes derived from patients with T2DM120122.
Although women with PCOS can show little differ Overall, adipocyte functioning, including the stimula
ence in fat distribution and possibly in overall BMI, tion of glucose transport 109, GLUT4 production119,123,

Table 1 | Current genome-wide association studies in PCOS


Number (cases; controls) Susceptibility Mapped genes SNPs Refs
loci
Han Chinese ethnicity
Discovery set: 744; 895 2p16.3 LHCGR rs13405728 72
First replication set: 2,840; 5,012 2p21 THADA rs12478601 and rs13429458
Second replication set: 498; 780 9q33.3 DENND1A rs2479106 and rs10818854
Discovery set: 2,254; 3,001 2p16.3 FSHR rs2268361 and rs2349415 71
Replication set: 8,226; 7,578 9q22.32 C9orf3 rs3802457 and rs4385527
11q22.1 YAP1 rs1894116
12q13.2 RAB5B and SUOX rs705702
12q14.3 HMGA2 rs2272046
16q12.1 TOX3 rs4784165
19p13.2 INSR rs2059807
20q13.2 SUMO1P1 rs6022786
European ancestry
Discovery set: 984; 2,964 8p32.1 GATA4NEIL2 rs804279 75
Replication set: 1,799; 1,231 11p14.1 KCNA4FSHB rs11031006
9q22.32 C9orf3 rs10993397
Discovery set: 5,184; 82,759 2q34 ERBB4 rs1351592 76
Replication set: 7,229*; 181,645 11q22.1 YAP1 rs11225154
2p21 THADA rs9563201
11p14.1 FSHB rs11031006
5q31.1 RAD50 rs13164856
12q21.2 KRR1 rs1275468
12q13.2 ERBB3 rs7312770
17q12 ERBB2 rs7218361
9q33.3 DENND1A rs10760321
PCOS, polycystic ovary syndrome; SNP, single-nucleotide polymorphism. *Cases were defined by either the 1990 US National
Institutes of Health (NIH) criteria or the 2003 Rotterdam criteria.

NATURE REVIEWS | DISEASE PRIMERS VOLUME 2 | 2016 | 7



2
0
1
6
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
PRIMER

Androgens Oestrogens

Granulosa Thecal
AMH FSH cell
cell

Primordial
follicle Primary
Preantral LH
follicle Early antral Late antral
follicle IGF1 Preovulatory
follicle follicle
follicle Cystic follicle

Figure 3 | Ovarian follicular maturation arrest in PCOS. Normal ovulation is the result of synchronized signalling
between centrally released gonadotropins and factors produced in the developing follicleNature of the Reviews
ovary. Anovulation in
| Disease Primers
women with polycystic ovary syndrome (PCOS) is characterized by arrested follicle growth at the early antral stage.
Hypersecretion of luteinizing hormone (LH) and insulin-like growth factor1 (IGF1) leads to hyperandrogenism, which
results in follicular maturation arrest93. In addition, high levels of anti-Mllerian hormone (AMH) in PCOS block follicle-
stimulating hormone (FSH) action, contribute to hyperandrogenism and inhibit the recruitment of further primordial
follicles. Dashed line indicates androgen to oestrogen conversion.

and insulin-stimulated inhibition of lipolysis124,125, are One of the consequences of hyperandrogenism


defective in PCOS. Epigenetic dysregulation of adipo is hirsutism. Androgens, primarily testosterone and
cyte function has been observed in PCOS, primarily of dihydrotestosterone, through their effect on the andro
microRNA93 (miR93) and miR223, which seem to gen receptor, stimulate, among other factors, ornithine
have a role in suppressing GLUT4 content and alter decarboxylase synthesis in the hair follicle, which in
ing glucose transport 123,126. In contrast to myocytes, turn stimulates polyamine production. Polyamines are
nodefects have been found in the classic insulin signal multifunctional cationic amines that are indispensable
ling pathway in adipocytes in PCOS, including in insulin for cellular proliferation, including hair growth in the
binding and in insulin receptor expression123. hair follicle.

Hyperandrogenism Mood disturbances and psychosexual dysfunction


The increased ovarian androgen production observed Causal factors underpinning the mood disturbance in
in PCOS is mainly due to enhanced androgen synthe PCOS remain unclear. The complex hormonal milieu
sis by follicular theca cells, which show an increased might contribute, although the clinical features of PCOS
expression of several genes encoding steroidogenic seem to also adversely affect mood36,129,130. Further poten
enzymes127. A recent study reported that a candidate tial contributors might include delayed diagnosis, poor
gene for PCOS, DENND1A, was overexpressed in theca diagnostic experience and the chronic and complex
cells obtained from patients with PCOS127, further sup nature of the condition131,132.
porting the notion that, at least in some patients, ovarian
androgen excess is a genetically determined feature of Diagnosis, screening and prevention
PCOS. In addition, the expression of the gene encod Diagnosis
ing the rate-limiting enzyme in androgen biosynthesis There are currently three main diagnostic criteria for
(CYP17A1) is increased in theca cells obtained from defining PCOS (BOX1; FIG.1). The evaluation of PCOS
women with PCOS, which might contribute to a higher entails determining the presence or absence of: hyper
conversion of progestogen precursors to androgens127. androgenism, ovulatory dysfunction and PCOM.
Theca cells isolated from women with PCOS are more Hyperandrogenism is clinically determined based on
responsive in terms of androgen secretion to insulin the presence of hirsutism using a visual scoring system,
and LH than theca cells of healthy controls97. In addi such as the mFG method11, and biochemically measur
tion to directly stimulating ovarian androgen secretion, ing the levels of circulating androgens. The clinical
hyperinsulinaemia contributes to hyperandrogenism in detection of ovulatory dysfunction is generally based on
PCOS by reducing hepatic synthesis of SHBG, leading a history of polymenorrhoea or oligo-amenorrhoea, or
to increased free testosterone fractions104. by assessing ovulatory function using luteal phase pro
Although the ovaries are the main source of hyper gesterone levels in hirsute women who are otherwise
androgenism in PCOS, between 20% and 30% of eumenorrhoeic. Ovarian ultrasonography (FIG.6) is used
patients also show adrenal androgen excess suggesting to identifyPCOM.
adrenocortical hyperfunction128. Adrenocortical dys All current criteria also call for the exclusion of
function in PCOS might be secondary to a generalized related or mimicking disorders. Hyperprolactinaemia
exaggeration in the responsivity, but not sensitivity, to and thyroid dysfunction, which can result in ovulatory
adrenocorticotropic hormone, but genetics might have dysfunction, should be excluded. Non-classic adre
arole. nal hyperplasia, androgen-secreting neoplasias, rare

8 | 2016 | VOLUME 2 www.nature.com/nrdp



2
0
1
6
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
PRIMER

syndromes of insulin resistance, with or without lipo Current recommendations are to screen all women
dystrophy, and the use of anabolic or androgenic drugs with PCOS for glucose intolerance using a 2-hour oral
and idiopathic hirsutism, which can also present with glucose tolerance test. Frequency of testing remains
signs and symptoms of androgen excess, should also be to be determined and ranges from every other year
excluded. Regardless of the criteria used to define PCOS, in all patients with PCOS to recurrent testing only in
greater emphasis should be given to defining the actual those with additional diabetes risk factors, including
phenotypes of PCOS (FIG.1). Different diagnostic criteria obesity 36,138140.
take different phenotypes into account. PhenotypesAC Assessment of general CVD risk factors, including
are considered hyperandrogenic, whereas phenotypeD age, BMI, sleep apnoea, smoking status, family history,
is non-hyperandrogenic. In addition, phenotype A lipid and liver enzyme profiles and blood pressure, is rec
and phenotypeB and, to a somewhat lesser extent, ommended at diagnosis. Subsequent assessment should
phenotypeC, are associated with a higher risk of con be based on individual overall risk, with the frequency
comitant metabolic dysfunction, which is less so for of re-testing still under discussion36,138,139,141. For women
phenotypeD133136. Overall, most investigators today use with a long history of irregular menstrual cycles with
the broader 2003 Rotterdam criteria for PCOS, but stress out endometrial protection, screening for endometrial
that the specific phenotypes included must be clearly hyperplasia might also be relevant, given the high risk
recognized and d ocumented as they differ substantially. of endometrial cancer in women withPCOS56.
Screening for psychological features is important
Screening and prevention of PCOS given their high prevalence. Effective, readily available
As the cause of PCOS is still unclear, and is probably screening questionnaires have been developed to assist
multifactorial, a specific plan for early risk prediction screening for mood disorders in PCOS in clinical prac
of diagnosis and treatment is not yet possible. However, tice36. If depression, anxiety and psychological features
studies in at-risk paediatric populations, principally are detected, treatment also includes conventional treat
first-degree female relatives of women with PCOS, have ment for mood disorders and treatment of factors that
elucidated various features that will facilitate the clini could affect QOL, such as hirsutism and excess weight36.
cian to identify those at risk for developing PCOS (BOX3).
Given that longitudinal studies are few and generally Management
short term, the exact extent to which these factors deter Therapy of PCOS should be tailored to the individ
mine the risk of PCOS, particularly in families without ual patient, and is often multifactorial. Lifestyle inter
other first-degree relatives with PCOS, is not well defined. ventions are first-line treatment for PCOS, and small
To what extent early prediction and treatment of lifestyle changes can improve metabolic dysfunction,
PCOS can ameliorate the disorder is unclear. Early treat ovulation, fertility and mood142. Other management
ment of excess weight gain, hyperandrogenic symptom options are aimed at improving metabolic dysfunc
and menstrual dysfunction will result in clinical improve tion, hyperandrogenism, reproductive therapy, and
ment, although it is unclear whether this will prevent psychological and emotionalstatus.
further progression of the disorder. One study reported
on the possibility that early treatment with metformin
might reduce progression to PCOS in girls with low Insulin
birth weight and precocious pubarche137. Further studies Insulin
ofearly markers of disease and early intervention trialsof receptor Cell membrane
atrisk children and adolescents are cruciallyneeded.
Myocyte
Screening for associated morbidities
Rigorous international evidence-based guidelines clearly
Y
recognize the need to routinely screen for metabolic S P
(glucose tolerance and lipid status) and p sychological Y MEK1/2
(anxiety and depression) features associated with Y
PCOS36,138. Screening of metabolic abnormalities should IRS1 S P
not include testing fasting insulin levels in r outine S P
practice as available assays lack adequate accuracy and
sensitivity 36. Fasting blood glucose levels alone to detect
dysglycaemia is not recommended as they seem to under Metabolic eects Mitogenic eects
diagnose IGT and T2DM in PCOS, in which patients of insulin signalling of insulin signalling
primarily demonstrate skeletal muscle-based and adipo
cyte tissue-based insulin resistance, rather than hepatic Figure 4 | Molecular mechanisms of insulin
Nature Reviews resistance
| Disease Primers
in muscle in PCOS. In myocytes (and fibroblasts),
resistance36,104. Although visually evident acanthosis
constitutive activation of MEK1/2 leads to increased
nigricans (hypertrophy of the basal layers of the epider phosphorylation (P) of serine (S) residues on the insulin
mis of the skin, usually in body folds) seems to be a good receptor and on insulin receptor substrate 1 (IRS1), which
predictor of insulin resistance and hyperinsulinaemia, hampers insulin signalling in the metabolic, but not the
well-controlled prospective studies of the predictive value mitogenic, pathways. PCOS, polycystic ovary syndrome;
of this dermatological sign are lacking. Y,tyrosine.

NATURE REVIEWS | DISEASE PRIMERS VOLUME 2 | 2016 | 9



2
0
1
6
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
PRIMER

Metabolic dysfunction and obesity lifestyle programme should always include psychological
Lifestyle modification. Lifestyle intervention is the support, social support and avoidance of toxic substances
primary treatment of metabolic dysfunction in PCOS (for example, smoking, alcohol and drugs).
and also improves fertility 143. Lifestyle intervention pro
grammes improve ovulation in 4050% of women with Bariatric surgery. For those women who fail to control
PCOS, 3040% of whom are able to achieve a spontane their weight on diet alone, bariatric surgery is an impor
ous pregnancy 144. Reduction of body weight in women tant option, but should be reserved for women with
with PCOS who are overweight or obese improves PCOS with severe obesity (a BMI of >40) or with mod
their metabolic profile and induces ovulatory cycles145. erate obesity (a BMI of >35) who also have additional
A small reduction in body weight (of at least 5%) can health issues. A meta-analysis of 13 primary studies has
improve ovulation, suggesting that the results depend shown that bariatric surgery decreased the incidence
more on energy restriction or changes in fat distribution of PCOS symptoms from 45.6% to 7.1%, with a mean
than on weight loss perse146. weight loss of 57.2%153.
Different hypocaloric diets with various macronutri
ent compositions have been used with no significant dif Medical treatment. Several pharmacological interven
ferences in results147. In some patients, high-protein diets tions can be used when lifestyle modifications fail to
might be better tolerated and lead to more satiety 148,149. manage metabolic dysfunction and dyslipidaemia in
However, long-term results of any diet are poor because women with PCOS. Many of these may also have indirect
of early dropout and low long-termcompliance. beneficial effects on the hyperandrogenism and ovula
Physical exercise can also help to reduce body tory dysfunction of the disorder. Metformin, a biguanide
weight 150, but most studies only show modest or no approved for the treatment of T2DM that suppresses
weight loss in women with PCOS, even with intensified hepatic gluconeogenesis and improves peripheral insulin
exercise programmes151,152. However, physical exercise sensitivity, can be used for the prevention of T2DM and
improved insulin resistance, promoted changes in fat impaired glucose tolerance when lifestyle modification
distribution and reduced cardiovascular risk in women fails. Metformin improves body composition and insu
with PCOS151, but should be performed at least 30min lin levels in women with PCOS who are not obese, but
utes per day for at least 5days per week. An effective has no significant effect on BMI, fasting glucose or lipid
levels154. A recent systematic review and meta-analysis
suggests that the combination of lifestyle modification
Insulin
Inammatory cytokines with metformin reduced BMI in women with PCOS to a
Glucose Insulin greater degree than lifestyle modification alone155.
receptor Thiazolidinediones (peroxisome proliferator-
activated receptor agonists or activators) are more effec
tive than metformin in lowering fasting insulin and in
Altered improving insulin resistance in PCOS156, yet less effec
Insulin-mediated adipokine
glucose uptake levels IRS1 tive in reducing BMI and triglyceride levels157. Owing to
potentially serious adverse effects, thiazolidinediones are
not currently recommended for the routine treatment of
GLUT4 production Insulin signalling insulin resistance in PCOS138,158.
Lipid droplet Inositol isomers (secondary messengers involved
Altered
GLUT4 miRNA Insulin- in several signalling pathways, including the insulin
expression inhibited pathway), in particular, combinations of myo-inositol
lipolysis TG and dchiro-inositol, have been shown to have insulin-
GLUT4 Lipolysis mimetic properties and to lower post-prandial blood
expression miR-223 miR-93 glucose. In PCOS, treatment with inositol isomers has
been shown to significantly improve the regularity of the
Nucleus menstrual cycle, the endocrine and metabolic parameters
and insulin resistance159. However, controversy remains
regarding the extent of the benefit and the exact dosing
Adipocyte FFA and glycerol of these substances160.
A meta-analysis evaluating acarbose (an glucosidase
Figure 5 | Molecular mechanisms of insulin resistance in adipose tissue in PCOS. inhibitor) treatment in PCOS demonstrated significant
Nature Reviews | Disease Primers
Although adipose tissue accounts for only 10% of insulin-stimulated whole-body glucose improvement in lipid profile, but an inconclusive effect
uptake , this tissue is crucial in determining systemic glucose homeostasis. For example,
232
onBMI161.
adipose tissue controls plasma free fatty acid (FFA) levels and secretes adipokines, Statins (3-hydroxy-3-methylglutaryl coenzymeA
among other substances, that substantially modulate insulin action and control systemic
(HMG-CoA) reductase inhibitors, which reduce choles
insulin sensitivity, energy balance and metabolic homeostasis. Adipocyte dysfunction in
polycystic ovary syndrome (PCOS) results in decreased levels of insulin-stimulated terol synthesis) can be used according to standard indi
glucose transport, decreased glucose transporter 4 (GLUT4; also known as SLC2A4) cations in PCOS. Statins are more effective than placebo
production, decreased insulin-stimulated inhibition of lipolysis and altered microRNA in reducing total cholesterol and triglyceride levels in
(miRNA) expression. IRS1, insulin receptor substrate 1; TG, triglyceride. Image courtesy PCOS162. In combination with metformin, they further
of Y.-H. Chen, Augusta University, Augusta, Georgia, USA. improve dyslipidaemia and markers of inflammation,

10 | 2016 | VOLUME 2 www.nature.com/nrdp



2
0
1
6
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
PRIMER

but the combination is less effective in improving


insulin sensitivity than metformin alone163. The risk
of statins in human pregnancy remains controversial
and current guidelines recommend discontinuing use
beforeconception.
Various pharmacological agents have been approved
for use in weight loss164. The use of orlistat (a lipase
inhibitor that reduces the intestinal absorption of fat)
is associated with a reduction in BMI in women with
PCOS, but its effect on insulin sensitivity remains con
troversial165. Several studies have been conducted on
the use of vitaminD in PCOS, but data from a recent
meta-analysis does not support the suggestion that vit
aminD supplementation improves insulin sensitivity in Figure 6 | Typical polycystic ovarian morphology.
Nature
Polycystic ovarian morphology Reviews | Disease
is characterized by Primers
the disorder 166.
enhanced central thecalstromal volume and increased
numbers of preovulatory follicles ringing the ovarian
Hyperandrogenism cortex. Image courtesy of J.S.E.L.
Suppression of ovarian androgen secretion. Combin
ation OCPs and the less commonly used transdermal
combination contraceptives effectively suppress ovarian ovary in patients with PCOS via laparoscopy) has been
androgen excess and are recommended as first-line man suggested as an alternative treatment to ovarian wedge
agement for the treatment of menstrual abnormalities, resection because of the lower risk of complications,
hirsutism and acne in women with PCOS who are not including pelvic adhesions and premature ovarian
seeking fertility 138. OCPs suppress gonadotropin release failure174. However, LOD only modestly suppresses
and consequently inhibit ovarian androgen secretion androgen production in PCOS175.
in women with PCOS. The oestrogen in OCPs also
stimulates hepatic production of SHBG, which in turn Anti-androgens. Drugs that block the action of andro
reduces the free (active) fraction of circulating andro gens are off-label medications used in the management
gens (and oestrogens) (an effect that is not achieved by of PCOS176182 and availability varies internationally. This
transdermal contraceptive preparations). Progestins in group includes androgen receptor blockers (for example,
OCPs can also directly inhibit androgen biosynthesis and spironolactone, flutamide and cyproterone acetate) and
impair androgen receptor binding 167. In addition, OCP 5reductase inhibitors (for example, finasteride)176,183
use considerably reduces the risk of endometrial hyper (TABLE2). Anti-androgens need to be prescribed along
plasia and endometrial cancer 66, while providing effective with secure contraception owing to their teratogenic
contraception when anti-androgen therapy is alsoused. potential (that is, risk of feminization of a male fetus)177.
The most widely prescribed OCPs contain ethinyl Anti-androgens also have variable adverse effect profiles,
estradiol and progestin, although there is no evidence which can be considerable in the case of finasteride and
that any one formulation is superior to any other for the flutamide. A combination therapy of anti-androgens with
treatment of PCOS138. Whether increased risk of cardio OCPs should be considered after failure to achieve the
vascular, venous thromboembolism and metabolic desired outcome with the OCP alone, or as initial ther
comorbidities associated with OCP use should be con apy in more-severe cases of hirsutism. Improvement in
sidered, especially as women with PCOS are already at hirsutism is usually observed after >6months. Most anti-
increased risk, remains controversial168,169. Appropriate androgens can be used in adolescents, but the efficacy and
contraindications should be elucidated and excluded safety of spironolactone and finasteride have not been
before starting OCP use170. well established in these younger patients. In addition to
Other medical avenues to suppress ovarian steroido anti-androgens, a topical solution of eflornithine (also
genesis in patients with PCOS include continuous treat known as difluoromethylornithine) hydrochloride
ment with progestin66 or administration of a long-acting an irreversible inhibitor of follicle ornithine decarboxylase
GnRH analogue171, but these are used much less fre can be used to treat unwanted facial hair growth184.
quently than OCPs. Insulin-sensitizing agents (for
example, metformin and thiazolidinediones) also provide Cosmetic treatments. Hormonal suppression of andro
modest improvement in hyperandrogenism172. gen secretion and peripheral androgen blockade will
In addition to medical therapy, ovarian surgery can ameliorate the effects of androgens on hair follicles,
reduce ovarian steroidogenesis. For example, ovarian minimizing the progression and further development
wedge resection (surgical removal of part of an ovary) of the dermatological symptoms of the disorder, includ
decreases the number of antral follicles, suppresses the ing hirsutism, acne and androgenic alopecia. However,
secretion of androgens, improves the endocrine status thesetherapies will be less effective for the treatment of
of other intra-ovarian factors and enables many patients these features once established. Consequently, hormonal
with PCOS to achieve regular ovulatory cycles173. More therapy should be combined with counselling regarding
recently, laparoscopic ovarian drilling (LOD; in which the use of cosmetic treatments for hirsutism (for e xample,
1015 small holes are burned into the surface of the shaving, depilation, laser epilation and electrology),

NATURE REVIEWS | DISEASE PRIMERS VOLUME 2 | 2016 | 11



2
0
1
6
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
PRIMER

Box 3 | Risk factors for the development of PCOS The goal of fertility treatment in women with PCOS
is to restore monofollicular ovulation and achieve single
Family history is the most important risk factor for ton pregnancy, given the predisposition of women with
developing PCOS6769. Other risk factors, depending PCOS to adverse pregnancy outcomes, including pre-
onage, include: eclampsia, gestational hypertension, gestational diabetes
Before birth mellitus and preterm labour 186. Fertility treatment should
-- Small for gestational age* start with counselling about success rates, discontinu
Childhood and peripuberty ation of harmful habits (especially smoking), screening
-- Early-onset obesity219222 for medical comorbidities and treating excess weight.
-- Increased dehydroepiandrosterone sulfate levels
Pursuing low-tech therapies, such as lifestyle modifi
during the onset of puberty219222
-- Premature pubarche223,224
cation and/or dose escalation of oral medications (for
-- Hyperinsulinaemia219222 example, clomifene or letrozole; see below) to achieve
Adolescence225228
ovulation, often requires patience of both the clin
-- Obesity, overweightness or rapid weight gain225228 ician and the patient. The chance of conceiving is only
-- Irregular menstrual or oligo-amenorrhoea225228 510% per ovulatory cycle in women with PCOS (versus
-- Presence of polycystic ovarian morphology225228 1015%per cycle in women without PCOS)187,188.
-- High androgen levels225228
-- Development of unwanted terminal hair growth on Medical treatment. First-line treatment for infertility is
the face or body229 aimed at restoring ovulation by interfering with inappro
*Although some investigators have suggested that girls born priate oestrogen feedback mechanisms (for example, clo
small for their gestational age are at higher risk for insulin mifene citrate, a selective oestrogen receptor modulator)
resistance and possibly PCOS230, other studies have indicated or oestrogen production in adipose tissue (for example,
that most women with PCOS are actually born of normal or
letrozole, an aromatase inhibitor). Letrozole is s uperior
large size for gestational age27,231.
to clomifene at achieving a live birth by 4050%188,
although the benefit might be greater in obese women
with PCOS. Multiple pregnancy rates with these med
acne(for example, topical antibacterials, and topical or ications are in the range of 38% (that is, 38% of the
oral retinoids) and androgenic alopecia (for example, pregnancies are not singletons).
minoxidil and hair transplantation)185. Metformin is a relatively ineffective infertility agent
in PCOS. When used alone, it has the lowest pregnancy
Endometrial protection rates compared to other oral agents, but also the lowest
Owing to their hyperinsulinaemic hyperoestrogenic multiple pregnancy rates154. Thus, metformin tends to be
anovulatory state, patients with PCOS are at increased used in combination with other medications. For exam
risk for endometrial hyperplasia and/or endometrial car ple, the combination of clomifene and metformin may
cinoma. They are also at increased risk for unpredictable be superior to treatment with clomifene alone187,189. The
abnormal uterine bleeding, which, in addition to being optimal number of ovulatory cycles with oral ovulatory
disruptive and a nuisance, can result in bouts of severe agents without achieving a pregnancy before moving on
dysfunctional uterine bleeding and consequent anae to other therapies is unknown, but is probably no more
mia. As such, protecting the endometrium from exces than five orsix187,188.
sive uncontrolled oestrogenic proliferation is c rucial in Second-line treatment tends to be gonadotropin ther
patients with PCOS. This can be readily achieved by treat apy (either as a combination of LH and FSH, or FSH
ing the patients with a progestogenic agent, most often in only), which is used in a low-dose regimen to minimize
the form of a combination OCP or a cyclic progestogen the risk of multiple pregnancies and ovarian hyperstimu
(such as oral micronized progesterone and medroxy lation syndrome (OHSS). The risk of OHSS is increased
progesterone acetate). Although some agents, such as in PCOS owing to a higher chance of multiple follicu
insulin-sensitizing agents, might seem to improve the lar recruitment. Gonadotropin therapy in expert hands
regularity of vaginal bleeding, this alone is not a guarantee with strict cancellation guidelines if evidence of exces
that the bleeding is resulting from regular progesterone- sive ovarian follicular recruitment is observed can result
induced withdrawals (for example, ovulation) and that in low multiple pregnancy rates (<5%)190,191. Pregnancy
the patients endometrium is protected. rates with gonadotropins are probably higher than with
first-line oral treatment 192, although the greater expense
Reproductive therapy and higher potential risks outweigh their choice as
Women with PCOS should be counselled to seek preg first-lineoption.
nancy earlier rather than later, unless further data
regarding the possibility of a prolonged reproductive Ovarian surgery. Ovarian surgery can also be used to
window owing to improved ovarian reserve emerge. induce ovulation. Although bilateral ovarian wedge
Ineither case, timely intervention will allow optimal pre resection is relatively effective at inducing ovulation in
conception, amelioration of risk factors for pregnancy patients with PCOS who are resistant to clomifene173
complications and adequate time to pursue effective, but and can be performed laparoscopically 193, LOD has
relatively safe, and affordable treatment strategies rather become the preferred surgical alternative because of the
than at a laterstage. lower risk of complications, particularly postoperative

12 | 2016 | VOLUME 2 www.nature.com/nrdp



2
0
1
6
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
PRIMER

adhesion formation174,175. LOD is generally recommended specific diseases, treatments, and short-term and long-
in patients with increased levels of LH, clomifene citrate term disabilities from the perspective of the patient 203.
resistance, an inability or unwillingness to proceed to The QOL of patients with PCOS is significantly reduced
gonadotropin ovulation induction and/or needing to in all domains (including functional ability, physical
undergo laparoscopy for other indications174,194. Long- aspects, general health perception, vitality, social and
term complications of LOD mainly include adhesion emotional aspects and mental health) compared with
formation and diminished ovarian reserve, especially healthy women204. Important negative determinants
when a large amount of drilling is performed195. of the QOL of patients with PCOS include the pres
ence ofobesity, hirsutism, androgenic alopecia, acne,
Invitro fertilization. Third-line therapy for women menstrual dysfunction and infertility, amongothers.
with PCOS is invitro fertilization (IVF). If IVF is used, The PCOS health-related QOL questionnaire
women with PCOS have similar or better pregnancy (PCOSQ) developed by Cronin et al.205 is the only
rates than women with other indications, and long- validated PCOS-specific tool and includes items in five
term followup studies suggest that, over a lifetime, domains, including body hair, emotions, weight, infertil
fecundity of women with PCOS may match population ity and menstrual problems205. A modified version of the
means196. In the United States, IVF is associated with questionnaire, adding four acne-related items, has been
multiple pregnancy rates of ~30%. This rate will prob proposed206. Studies using the PCOSQ or the modified
ably be lower in other countries in which single embryo version of this tool consistently report reduced QOL
transfer is recommended, possibly even lower than the with different contributing factors, including menstrual
multiple pregnancy rates observed using other ovulation disturbances, hirsutism, acne, obesity and infertility 38.
induction methods. Interestingly, although depression scores are signifi
Several modifications to IVF have been proposed cantly correlated with insulin resistance, PCOSQ scores
to prevent multiple births and OHSS in patients with arenot 40.
PCOS, including invitro maturation (IVM) of immature Limited available data indicate that weight loss in
oocytes that are retrieved without gonadotropin stimula overweight or obese women with PCOS achieved by diet
tion197 and elective cryopreservation of all embryos and ary restriction alone or combined with exercise improves
transfer in a subsequent frozen embryo transfer cycle both depressive symptoms and PCOS-specific QOL
after ovarian recovery. Pregnancy rates are lower with scores, except for the scores in the body hair domain142.
IVM than with IVF198,199, although IVF results might be An observational study of metformin treatment in
better overall with elective cryopreservation than with women with PCOS reported improvements in health-
fresh IVF embryo transfer 200. However, randomized related QOL scores and emotional well-being, and these
trials in women with PCOS are lacking for both of these improvements significantly correlated with body weight
IVF modifications. Metformin treatment in patients reduction and menstrual cycle normalization207. However,
with PCOS u ndergoing IVF might also reduce the risk subsequent randomized controlled trials have not sup
ofOHSS201,202. ported this observation, and the addition of metformin
to lifestyle modification did not seem to have an effect on
Quality of life QOL208. Similarly, a randomized trial assessing lifestyle
Health-related QOL is a multidimensional concept that modification programme combined with OCP use in
includes domains related to physical, mental, emotional obese adolescents with PCOS showed an improvement
and social functioning in response to the effects of inPCOSQ scores, whereas the addition of metformin

Table 2 | Anti-androgens available for use in PCOS


Agent Mechanism of action US FDA-approved indication Adverse effects Refs
Spironolactone Competitive inhibitor Treatment of low renin Dyspepsia; dry skin; decreased libido; hypotension; 178,
of AR binding, hypertension, hypokalaemia and polyuria; menstrual irregularity and polymenorrhoea 179
antimineralocorticoid, limits Conn syndrome when not administered with an OCP; teratogenic in
suppression of 5reductase early pregnancy, principally on male fetuses; rare skin
activity and suppresses LH sensitivity to sunlight; and rare hypokalaemia
Cyproterone Competitive inhibitor of AR Palliative treatment of patients with Breast swelling; amenorrhoea; decreased libido; 180
acetate binding, limits suppression advanced prostatic carcinoma teratogenic in early pregnancy, principally on
of 5reductase activity and male fetuses; rare liver toxicity; rare reduced
decreases LHdependent adrenal response to ACTH stimulation; and rare
androgen secretion oestrogen-related osteoporosis
Flutamide Competitive antagonist of Management of locally confined Dry skin; discoloured (green) urine; decreased libido; 176,
AR and reduces synthesis stage B2stageC and stage D2 teratogenic in early pregnancy, principally on male 181
of DHT metastatic carcinoma of the prostate fetuses; and rare liver toxicity (occasionally fulminant)
Finasteride Competitively binds to Treatment of benign prostatic Dry skin; headache; decreased libido; 182
and inhibits steroid typeII hyperplasia andteratogenic in early pregnancy, principally
5reductase onmale fetuses
ACTH, adrenocorticotropic hormone; AR, androgen receptor; DHT, dihydrotestosterone; LH, luteinizing hormone; OCP, oral contraceptive pill; PCOS, polycystic
ovary syndrome.

NATURE REVIEWS | DISEASE PRIMERS VOLUME 2 | 2016 | 13



2
0
1
6
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
PRIMER

to the lifestyle modification programme combined ovulation induction therapies, with a low risk of adverse
withOCPs did not influence PCOSQ scores further 209. events including multiple pregnancies, is also desir
Finally, OCP use improves hirsutism and menstrual dis able. These therapies might involve repurposing other
turbances along with PCOSQ scores, but without any drugs (such as those developed for T2DM, breast c ancer
significant change in depression or anxietysymptoms210. or psychiatric indications) or developing new drugs
Women with PCOS also have a more negative body specifically for PCOS. For example, a neurokinin recep
image than women without PCOS, worsened by the tor antagonist has been shown to suppress sex steroid
presence of hirsutism and weight gain130, which is associ levels through alteration in gonadotropin modulation217.
ated with greater depression and anxiety 211 and a lower Further candidate genes identified by GWAS in women
QOL. Likewise, the prevalence of eating disorders, which with PCOS offer supposition about potential pathways,
are in turn also significantly associated with depression including through phosphoinositide 3kinase and
and anxiety, is increased in PCOS36. Women with PCOS mitogen-activated protein kinase, among other path
also suffer from greater psychosexual dysfunction, which ways127. Such strategies could also potentially be used
correlates with poor QOL, sexual dissatisfaction and for other reproductive disorders
reduced feminine identity 207,212. Overall, mood disorders
seem to be more common and more severe in women Hyperandrogenism
with PCOS, and clinician awareness and corresponding Although the skin complications are often viewed as
screening are important. purely cosmetic aspects of the disorder, these features
are among the most injurious of the traits of PCOS, sub
Outlook stantially affecting self-esteem, psychosocial adaption
Although important advances have been made in our and QOL. Future studies must focus on better under
understanding of PCOS in the past two decades, much standing of the physiology and mechanism underlying
remains to be elucidated. We highlight some, although hair follicle cycling and hair growth, and the impact of
certainly not all, areas in which research isneeded. androgens, other steroid hormones, skin environment,
as well as identifying how this process can be targeted,
Metabolic dysfunction including defining the period and mechanisms by which
Currently, limited data are available on the tissue- this process becomes irreversible. These studies will per
specific aetiologies of the metabolic dysfunction under mit the elucidation of novel approaches to the inhibition
lying PCOS, and further studies are needed to better and treatment of unwanted excess hair terminalization
understand the molecular and genetic aetiologies of (hirsutism) or miniaturization (androgenicalopecia).
these pathologies in muscle, fat and other tissues. The development of novel and improved peripheral
Furthermore, we need improved understanding of how androgen receptor blockers, particularly those that do
metabolic dysfunction relates, in a temporal manner, not have systemic effects, is necessary. Anti-androgens
to the development of the other clinical features of the are needed that can be administrated, even topically,
disorder (for example, ovarian dysfunction), perhaps without the need for concomitant OCPs or other forms
through prospective studies of at-risk children. of contraception. In addition, approval of the use of
Future studies should also focus on the development current and novel anti-androgens from the appropri
of preventive therapies to minimize the development of ate government agencies is urgently needed, as the lack
obesity in peripubertal and adolescent girls. In addition, of such approvals has a negative impact on their use in
greater clarity is needed regarding the effect of metabolic many parts of theworld.
dysfunction on the cardiovascular and cerebrovascu
lar complications associated with PCOS, and whether Evolutionary aspects of PCOS
managing metabolic complications might reduce the Every aspect of human health and disease has emerged
risk of these complications. In addition, well-controlled through evolutionary processes, including natural
prospective studies concerning the predictive value of selection, and many of the current ailments affecting
acanthosis nigricans as a predictor of insulin resistance our society stem directly from evolutionary maladap
and hyperinsulinaemia in PCOS areneeded. tation. PCOS seems to be an evolutionary-conserved
Finally, as it is clear that insulin resistance alone is disorder:PCOSisa relatively common disorder with a
not sufficient to induce the full metabolic dysfunction uniform prevalence worldwide, is heritable with similar
of PCOS, better understanding of what other aetio genetic variants across ethnicities and was described
logical factors are necessary or permissive is also needed. in the medical literature as early as 1,000years ago218.
Furthermore, a male phenotype of PCOS, most notably An evolutionary biology approach to studying PCOS
around defects of metabolic action, is likely to exist, that focuses on factors leading to its development and
based on the absence of sex linkage213216; however, this persistence might help to elucidate its fundamental
potential syndrome remains to be demonstrated, better pathophysiology.
characterized and its risks defined. When considering the evolutionary aspects of PCOS,
it is peculiar that the condition has persisted given its
Reproductive dysfunction effect on reproduction. This paradoxical feature might
A better understanding of the doseresponse relation be explained by a vulnerability to PCOS, which refers
ships between the amount of weight loss and improve to the inherent potential for pathology to be embed
ments in ovulation is needed. Establishing low-cost ded within biological characteristics associated with

14 | 2016 | VOLUME 2 www.nature.com/nrdp



2
0
1
6
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
PRIMER

normal physiological development and function. For Epidemiology of PCOS


example, a vulnerability to PCOS, although detri More precise population studies of the epidemiology,
mental to the fertility of affected individuals, could be phenotype and genetics of PCOS worldwide are crucially
associated with improved fertility or fetal outcomes, or needed, and will also help to elucidate the evolutionary
other fitness-enhancing physiological phenomenon, in path of PCOS as humanity marched across the conti
most of the other related women within a population. nents. These studies should be carried out in various
Ultimately, novel insights into the evolutionary origins racial and ethnic groups in different parts of the world
of PCOS will emerge through a broad consideration of and in similar or related racial and ethnic groups who
the potential adaptive and beneficial aspects of vulner have lived for some time in different geographical
ability to the disorder. For example, the study of iso regions or environs (for example, West Africans com
lated populations of people who have uncontrolled pared with African-Americans). These investigations
fertility, high levels of exposure to natural pathogens will also assist in understanding the role of environ
and low consumption of processed carbohydrates might mental factors on the epidemiology, phenotype and
beinstrumental. possibly genetics and epigenetics ofPCOS.

1. Stepto,N.K. etal. Women with polycystic ovary ofobesity, changes in body composition and 28. Teede,H.J. etal. Longitudinal weight gain in women
syndrome have intrinsic insulin resistance on impairment of glucose tolerance, and that identified with polycystic ovary syndrome: results of
euglycaemichyperinsulaemic clamp. Hum. Reprod. hyperinsulinaemia in PCOS is not the result anobservational study in young women. Obesity
28, 777784 (2013). ofdecreased insulin clearance. (Silver Spring) 21, 15261532 (2013).
2. Vink,J.M., Sadrzadeh,S., Lambalk,C.B. 14. Moran,L.J., Misso,M.L., Wild,R.A. & Norman,R.J. 29. Yildiz,B.O., Knochenhauer,E.S. & Azziz,R. Impact
&Boomsma,D.I. Heritability of polycystic ovary Impaired glucose tolerance, type2 diabetes and ofobesity on the risk for polycystic ovary syndrome.
syndrome in a Dutch twin-family study. J.Clin. metabolic syndrome in polycystic ovary syndrome: J.Clin. Endocrinol. Metab. 93, 162168 (2008).
Endocrinol. Metab. 91, 21002104 (2006). asystematic review and meta-analysis. Hum. Reprod. 30. Yildiz,B.O., Bozdag,G., Yapici,Z., Esinler,I.
This study clearly demonstrates the very high Update 16, 347363 (2010). &Yarali,H. Prevalence, phenotype and
heritability of PCOS through twin studies. 15. Legro,R.S., Gnatuk,C.L., Kunselman,A.R. cardiometabolic risk of polycystic ovary syndrome
3. Azziz,R., Marin,C., Hoq,L., Badamgarav,E. &Dunaif,A. Changes in glucose tolerance over time under different diagnostic criteria. Hum. Reprod. 27,
&Song,P. Health care-related economic burden of inwomen with polycystic ovary syndrome: acontrolled 30673073 (2012).
thepolycystic ovary syndrome during the reproductive study. J.Clin. Endocrinol. Metab. 90, 32363242 31. Ewens,K.G. etal. FTO and MC4R gene variants are
life span. J.Clin. Endocrinol. Metab. 90, 46504658 (2005). associated with obesity in polycystic ovary syndrome.
(2005). 16. Norman,R.J., Masters,L., Milner,C.R., Wang,J.X. PLoS ONE 6, e16390 (2011).
4. DiamantiKandarakis,E., Spritzer,P.M., & Davies,M.J. Relative risk of conversion from 32. Cai,X., Liu,C. & Mou,S. Association between
SirPetermann,T. & Motta,A.B. Insulin resistance normoglycaemia to impaired glucose tolerance or fatmass- and obesity-associated (FTO) gene
andpolycystic ovary syndrome through life. noninsulin dependent diabetes mellitus in polycystic polymorphism and polycystic ovary syndrome:
Curr.Pharm. Des. 18, 55695576 (2012). ovarian syndrome. Hum. Reprod. 16, 19951998 ameta-analysis. PLoS ONE 9, e86972 (2014).
5. Lizneva,D. etal. The criteria, prevalence and (2001). 33. Louwers,Y.V. etal. BMI-associated alleles do not
phenotypes of PCOS. Fertil. Steril. 106, 615 17. Hudecova,M. etal. Diabetes and impaired glucose constitute risk alleles for polycystic ovary syndrome
(2016). tolerance in patients with polycystic ovary syndrome independently of BMI: acasecontrol study.
6. Zawadzki,J. & Duniaf,A. in Polycystic Ovary along term followup. Hum. Reprod. 26, PLoSONE 9, e87335 (2014).
Syndrome (eds Dunaif,A., Givens,J.R., Haseltine,F.P. 14621468 (2011). 34. Joham,A.E., Teede,H.J., Ranasinha,S., Zoungas,S.
& Merriam,G.R.) 377384 (Blackwell Scientific 18. Gambineri,A. etal. Polycystic ovary syndrome is a risk & Boyle,J. Prevalence of infertility and use of fertility
Publications,1992). factor for type2 diabetes: results from a long-term treatment in women with polycystic ovary syndrome:
7. Rotterdam ESHRE/ASRM-Sponsored PCOS prospective study. Diabetes 61, 23692374 (2012). data from a large community-based cohort study.
Consensus Workshop Group. Revised 2003 consensus 19. Schmidt,J., LandinWilhelmsen,K., Brannstrom,M. J.Womens Health (Larchmt) 24, 299307 (2015).
on diagnostic criteria and long-term health risks &Dahlgren,E. Cardiovascular disease and risk factors 35. Wild,S., Pierpoint,T., Jacobs,H. & McKeigue,P.
related to polycystic ovary syndrome (PCOS). in PCOS women of postmenopausal age: a21year Long-term consequences of polycystic ovary
Hum.Reprod. 19, 4147 (2004). controlled followup study. J.Clin. Endocrinol. Metab. syndrome: results of a 31year followup study.
8. Rotterdam ESHRE/ASRM-Sponsored PCOS 96, 37943803 (2011). Hum.Fertil. (Camb.) 3, 101105 (2000).
Consensus Workshop Group. Revised 2003 consensus 20. Lim,S.S., Davies,M.J., Norman,R.J. & Moran,L.J. 36. Teede,H.J. etal. Assessment and management of
on diagnostic criteria and long-term health risks Overweight, obesity and central obesity in women polycystic ovary syndrome: summary of an evidence-
related to polycystic ovary syndrome. Fertil. Steril. 81, with polycystic ovary syndrome: asystematic review based guideline. Med. J.Aust. 195, S65112 (2011).
1925 (2004). and meta-analysis. Hum. Reprod. Update 18, 37. Dokras,A., Clifton,S., Futterweit,W. & Wild,R.
9. Ezeh,U., Yildiz,B.O. & Azziz,R. Referral bias in 618637 (2012). Increased risk for abnormal depression scores in
defining the phenotype and prevalence of obesity in 21. LuqueRamirez,M. etal. Referral bias in female women with polycystic ovary syndrome: asystematic
polycystic ovary syndrome. J.Clin. Endocrinol. Metab. functional hyperandrogenism and polycystic ovary review and meta-analysis. Obstetr. Gynecol. 117,
98, E1088E1096 (2013). syndrome. Eur. J.Endocrinol. 173, 603610 (2015). 145152 (2011).38.
This is the first study to clearly demonstrate that 22. Guyatt,G., Weaver,B., Cronin,L., Dooley,J.A. VeltmanVerhulst,S.M., Boivin,J., Eijkemans,M.J.
the phenotype of PCOS observed in clinical studies &Azziz,R. Health-related quality of life in women &Fauser,B.J. Emotional distress is a common risk in
may be very different from that observed in withpolycystic ovary syndrome, a self-administered women with polycystic ovary syndrome: asystematic
medically unbiased populations, indicating the questionnaire, was validated. J.Clin. Epidemiol. 57, review and meta-analysis of 28 studies. Hum. Reprod.
need to better understand the true phenotype 12791287 (2004). Update 18, 638651 (2012).
ofPCOS. 23. Khomami,M.B., Tehrani,F.R., Hashemi,S., 39. Moran,L.J., Deeks,A.A., GibsonHelm,M.E.
10. Azziz,R. etal. The Androgen Excess and PCOS Farahmand,M. & Azizi,F. Of PCOS symptoms, &Teede,H.J. Psychological parameters in the
Society criteria for the polycystic ovary syndrome: hirsutism has the most significant impact on the reproductive phenotypes of polycystic ovary
thecomplete task force report. Fertil. Steril. 91, quality of life of Iranian women. PLoS ONE 10, syndrome. Hum. Reprod. 27, 20822088 (2012).
456488 (2009). e0123608 (2015). 40. Cinar,N. etal. Depression, anxiety and
11. DeUgarte,C.M., Woods,K.S., Bartolucci,A.A. 24. Dilbaz,B., Cinar,M., Ozkaya,E., Tonyali,N.V. cardiometabolic risk in polycystic ovary syndrome.
&Azziz,R. Degree of facial and body terminal hair &Dilbaz,S. Health related quality of life among Hum. Reprod. 26, 33393345 (2011).
growth in unselected black and white women: different PCOS phenotypes of infertile women. J.Turk. 41. Roos,N. etal. Risk of adverse pregnancy outcomes
towarda populational definition of hirsutism. J.Clin. Ger. Gynecol. Assoc. 13, 247252 (2012). inwomen with polycystic ovary syndrome: population
Endocrinol. Metab. 91, 13451350 (2006). 25. Gosman,G.G. etal. Reproductive health of women based cohort study. BMJ 343, d6309 (2011).
12. Souter,I., Sanchez,L.A., Perez,M., Bartolucci,A.A. electing bariatric surgery. Fertil. Steril. 94, 42. Sterling,L. etal. Pregnancy outcomes in women
& Azziz,R. The prevalence of androgen excess among 14261431 (2010). withpolycystic ovary syndrome undergoing invitro
patients with minimal unwanted hair growth. 26. AlvarezBlasco,F., BotellaCarretero,J.I., fertilization. Fertil. Steril. 105, 791797.e2 (2016).
Am.J.Obstet. Gynecol. 191, 19141920 (2004). SanMillan,J.L. & EscobarMorreale,H.F. Prevalence 43. Pan,M.L., Chen,L.R., Tsao,H.M. & Chen,K.H.
13. Dunaif,A., Segal,K.R., Futterweit,W. and characteristics of the polycystic ovary syndrome Relationship between polycystic ovarian syndrome
&Dobrjansky,A. Profound peripheral insulin inoverweight and obese women. Arch. Intern. Med. and subsequent gestational diabetes mellitus:
resistance, independent of obesity, in polycystic ovary 166, 20812086 (2006). anationwide population-based study. PLoS ONE 10,
syndrome. Diabetes 38, 11651174 (1989). 27. Laitinen,J. etal. Body size from birth to adulthood as e0140544 (2015).
This is one of the first studies to conclusively a predictor of self-reported polycystic ovary syndrome 44. Palomba,S. etal. Pregnancy complications in women
demonstrate that women with PCOS have symptoms. Int. J.Obes. Relat. Metab. Disord. 27, with polycystic ovary syndrome. Hum. Reprod. Update
significant insulin resistance that is independent 710715 (2003). 21, 575592 (2015).

NATURE REVIEWS | DISEASE PRIMERS VOLUME 2 | 2016 | 15



2
0
1
6
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
PRIMER

45. Vanky,E. etal. Metformin versus placebo from first polycystic ovary syndrome: asystematic review and useful practicalguidelines on how to take care of
trimester to delivery in polycystic ovary syndrome: meta-analysis. Int. J.Cardiol. 176, 486487 (2014). these patients in daily practice.
arandomized, controlled multicenter study. J.Clin. 65. Carmina,E., Campagna,A.M. & Lobo,R.A. A20year 88. Fauser,B.C. & VanHeusden,A.M. Manipulation
Endocrinol. Metab. 95, E448E455 (2010). followup of young women with polycystic ovary ofhuman ovarian function: physiological concepts
This is the first study to clearly demonstrate that syndrome. Obstet. Gynecol. 119, 263269 (2012). andclinical consequences. Endocr. Rev. 18, 71106
metformin treatment from the first trimester to 66. Dumesic,D.A. & Lobo,R.A. Cancer risk and PCOS. (1997).
delivery did not reduce pregnancy complications Steroids 78, 782785 (2013). 89. Broekmans,F.J. etal. Anti-Mllerian hormone
inwomen with PCOS. 67. KahsarMiller,M. & Azziz,R. The development of andovarian dysfunction. Trends Endocrinol. Metab.
46. Carmina,E. Polycystic ovary syndrome: metabolic thepolycystic ovary syndrome: family history as a risk 19, 340347 (2008).
consequences and long-term management. Scand. factor. Trends Endocrinol. Metab. 9, 5558 (1998). 90. Valkenburg,O. etal. Genetic polymorphisms of
J.Clin. Lab. Invest. Suppl. 244, 2326; discussion 26 68. Legro,R.S., Driscoll,D., Strauss,J.F. 3rd, Fox,J. GnRHand gonadotrophic hormone receptors affect
(2014). &Dunaif,A. Evidence for a genetic basis for the phenotype of polycystic ovary syndrome.
47. Christian,R.C. etal. Prevalence and predictors of hyperandrogenemia in polycystic ovary syndrome. Hum.Reprod. 24, 20142022 (2009).
coronary artery calcification in women with polycystic Proc. Natl Acad. Sci. USA 95, 1495614960 (1998). 91. Laven,J.S. etal. Anti-Mllerian hormone serum
ovary syndrome. J.Clin. Endocrinol. Metab. 88, 69. Lerchbaum,E., Schwetz,V., Giuliani,A. concentrations in normoovulatory and anovulatory
25622568 (2003). &ObermayerPietsch,B. Influence of a positive family women of reproductive age. J.Clin. Endocrinol. Metab.
48. Talbott,E.O. etal. Evidence for an association history of both type2 diabetes and PCOS on metabolic 89, 318323 (2004).
between metabolic cardiovascular syndrome and and endocrine parameters in a large cohort of PCOS 92. Stubbs,S.A. etal. Anti-Mllerian hormone protein
coronary and aortic calcification among women with women. Eur. J.Endocrinol. 170, 727739 (2014). expression is reduced during the initial stages of
polycystic ovary syndrome. J.Clin. Endocrinol. Metab. 70. Kosova,G. & Urbanek,M. Genetics of the polycystic follicle development in human polycystic ovaries.
89, 54545461 (2004). ovary syndrome. Mol. Cell. Endocrinol. 373, 2938 J.Clin. Endocrinol. Metab. 90, 55365543 (2005).
49. LuqueRamirez,M., MendietaAzcona,C., (2013). 93. Visser,J.A. etal. Increased oocyte degeneration
AlvarezBlasco,F. & EscobarMorreale,H.F. 71. Shi,Y. etal. Genome-wide association study identifies andfollicular atresia during the estrous cycle in
Androgenexcess is associated with the increased eight new risk loci for polycystic ovary syndrome. antiMllerian hormone null mice. Endocrinology 148,
carotid intimamedia thickness observed in young Nat.Genet. 44, 10201025 (2012). 23012308 (2007).
women with polycystic ovary syndrome. Hum. Reprod. 72. Chen,Z.J. etal. Genome-wide association study 94. CatteauJonard,S. & Dewailly,D. Pathophysiology
22, 31973203 (2007). identifies susceptibility loci for polycystic ovary ofpolycystic ovary syndrome: therole of
50. Lass,N., Kleber,M., Winkel,K., Wunsch,R. syndrome on chromosome 2p16.3, 2p21 and 9q33.3. hyperandrogenism. Front. Horm. Res. 40, 2227
&Reinehr,T. Effect of lifestyle intervention on features Nat. Genet. 43, 5559 (2011). (2013).
of polycystic ovarian syndrome, metabolic syndrome, This is the first genome-wide association study 95. Padmanabhan,V. etal. Dynamics of bioactive
and intimamedia thickness in obese adolescent girls. ofPCOS. folliclestimulating hormone secretion in women
J.Clin. Endocrinol. Metab. 96, 35333540 (2011). 73. Du,J. etal. Family-based analysis of INSR withpolycystic ovary syndrome: effects of estradiol
51. Chang,A.Y. etal. Polycystic ovarian syndrome polymorphisms in Chinese PCOS. Reprod. Biomed. and progesterone. Fertil. Steril. 75, 881888 (2001).
andsubclinical atherosclerosis among women Online 29, 239244 (2014). 96. Webber,L.J. etal. Formation and early development
ofreproductive age in the Dallas heart study. 74. Zhao,S. etal. Family-based analysis of eight of follicles in the polycystic ovary. Lancet 362,
Clin.Endocrinol. (Oxf.) 74, 8996 (2011). susceptibility loci in polycystic ovary syndrome. 10171021 (2003).
52. Dahlgren,E., Janson,P.O., Johansson,S., Lapidus,L. Sci.Rep. 5, 12619 (2015). 97. Nestler,J.E. etal. Insulin stimulates testosterone
& Oden,A. Polycystic ovary syndrome and risk for 75. Hayes,M.G. etal. Genome-wide association of biosynthesis by human thecal cells from women
myocardial infarction. Evaluated from a risk factor polycystic ovary syndrome implicates alterations withpolycystic ovary syndrome by activating its own
model based on a prospective population study of ingonadotropin secretion in European ancestry receptor and using inositolglycan mediators as the
women. Acta Obstet. Gynecol. Scand. 71, 599604 populations. Nat. Commun. 6, 7502 (2015). signal transduction system. J.Clin. Endocrinol. Metab.
(1992). 76. Day,F.R. etal. Causal mechanisms and balancing 83, 20012005 (1998).
53. Carmina,E., Campagna,A.M. & Lobo,R.A. selection inferred from genetic associations with 98. Visser,J.A., Schipper,I., Laven,J.S.
Emergence of ovulatory cycles with aging in women polycystic ovary syndrome. Nat. Commun. 6, 8464 &Themmen,A.P. Anti-Mllerian hormone: anovarian
with polycystic ovary syndrome (PCOS) alters the (2015). reserve marker in primary ovarian insufficiency.
trajectory of cardiovascular and metabolic risk factors. 77. Azziz,R. PCOS in 2015: new insights into the genetics Nat.Rev. Endocrinol. 8, 331341 (2012).
Hum. Reprod. 28, 22452252 (2013). of polycystic ovary syndrome. Nat. Rev. Endocrinol. 12, 99. TeixeiraFilho,F.L. etal. Aberrant expression of
54. RamezaniTehrani,F. etal. Trend of cardio-metabolic 7475 (2016). growth differentiation factor9 in oocytes of women
risk factors in polycystic ovary syndrome: 78. Cui,L. etal. Genotypephenotype correlations of with polycystic ovary syndrome. J.Clin. Endocrinol.
apopulationbased prospective cohort study. PCOS susceptibility SNPs identified by GWAS in a large Metab. 87, 13371344 (2002).
PLoSONE 10, e0137609 (2015). cohort of Han Chinese women. Hum. Reprod. 28, 100. Bergman,R.N., Ider,Y.Z., Bowden,C.R. & Cobelli,C.
55. BaireyMerz,C.N. etal. Cardiovascular disease 538544 (2013). Quantitative estimation of insulin sensitivity.
and10year mortality in postmenopausal women with 79. Welt,C.K. etal. Variants in DENND1A are associated Am.J.Physiol. 236, E667E677 (1979).
clinical features of PCOS. J.Womens Health (Larchmt) with polycystic ovary syndrome in women of 101. OMeara,N.M. etal. Defects in beta-cell function
6June 2016 [epub ahead of print]. Europeanancestry. J.Clin. Endocrinol. Metab. 97, infunctional ovarian hyperandrogenism. J.Clin.
56. Hart,R. & Doherty,D.A. The potential implications of E1342E1347 (2012). Endocrinol. Metab. 76, 12411247 (1993).
a PCOS diagnosis on a womans long-term health using 80. Saxena,R. etal. Han Chinese polycystic ovary 102. Ehrmann,D.A. etal. Insulin secretory defects in
data linkage. J.Clin. Endocrinol. Metab. 100, 911919 syndrome risk variants in women of European ancestry: polycystic ovary syndrome. Relationship to insulin
(2015). relationship to FSH levels and glucose tolerance. sensitivity and family history of non-insulin-dependent
57. Cibula,D. etal. Increased risk of non-insulin dependent Hum.Reprod. 30, 14541459 (2015). diabetes mellitus. J.Clin. Invest. 96, 520527
diabetes mellitus, arterial hypertension and coronary 81. Cui,L. etal. Polycystic ovary syndrome susceptibility (1995).
artery disease in perimenopausal women with a history single nucleotide polymorphisms in women with 103. Dunaif,A. & Finegood,D.T. Beta-cell dysfunction
of the polycystic ovary syndrome. Hum. Reprod. 15, asingle PCOS clinical feature. Hum. Reprod. 30, independent of obesity and glucose intolerance in the
785789 (2000). 732736 (2015). polycystic ovary syndrome. J.Clin. Endocrinol. Metab.
58. Mani,H. etal. Diabetes and cardiovascular events 82. Shim,U. etal. Pathway analysis based on a genome- 81, 942947 (1996).
inwomen with polycystic ovary syndrome: a20year wide association study of polycystic ovary syndrome. 104. DiamantiKandarakis,E. & Dunaif,A. Insulin
retrospective cohort study. Clin. Endocrinol. (Oxf.) 78, PLoS ONE 10, e0136609 (2015). resistance and the polycystic ovary syndrome
926934 (2013). 83. Jones,M.R. etal. Systems genetics reveals the revisited: anupdate on mechanisms and implications.
59. Krentz,A.J., vonMuhlen,D. & BarrettConnor,E. functional context of PCOS loci and identifies genetic Endocr. Rev. 33, 9811030 (2012).
Searching for polycystic ovary syndrome in and molecular mechanisms of disease heterogeneity. 105. Pasquali,R. etal. Insulin and Cpeptide levels in obese
postmenopausal women: evidence of a doseeffect PLoS Genet. 11, e1005455 (2015). patients with polycystic ovaries. Horm. Metab. Res.
association with prevalent cardiovascular disease. 84. Banaszewska,B., Spaczynski,R.Z., Pelesz,M. 14, 284287 (1982).
Menopause 14, 284292 (2007). &Pawelczyk,L. Incidence of elevated LH/FSH ratio 106. Mahabeer,S. etal. Insulin and Cpeptide secretion
60. Wild,S., Pierpoint,T., McKeigue,P. & Jacobs,H. inpolycystic ovary syndrome women with normo- innon-obese patients with polycystic ovarian disease.
Cardiovascular disease in women with polycystic ovary andhyperinsulinemia. Rocz. Akad. Med. Bialymst. 48, Horm. Metab. Res. 21, 502506 (1989).
syndrome at long-term followup: aretrospective 131134 (2003). 107. Dunaif,A., Xia,J., Book,C.B., Schenker,E. & Tang,Z.
cohort study. Clin. Endocrinol. (Oxf.) 52, 595600 85. Dale,P.O., Tanbo,T., Vaaler,S. & Abyholm,T. Body Excessive insulin receptor serine phosphorylation
(2000). weight, hyperinsulinemia, and gonadotropin levels incultured fibroblasts and in skeletal muscle.
61. Iftikhar,S. etal. Risk of cardiovascular events in inthe polycystic ovarian syndrome: evidence of two Apotential mechanism for insulin resistance in
patients with polycystic ovary syndrome. Neth. J.Med. distinct populations. Fertil. Steril. 58, 487491 (1992). thepolycystic ovary syndrome. J.Clin. Invest. 96,
70, 7480 (2012). 86. Burt Solorzano,C.M., McCartney,C.R., Blank,S.K., 801810 (1995).
62. Carmina,E. Oral contraceptives and cardiovascular Knudsen,K.L. & Marshall,J.C. Hyperandrogenaemia 108. Dunaif,A. etal. Evidence for distinctive and intrinsic
risk in women with polycystic ovary syndrome. in adolescent girls: origins of abnormal gonadotropin- defects in insulin action in polycystic ovary syndrome.
J.Endocrinol. Invest. 36, 358363 (2013). releasing hormone secretion. BJOG 117, 143149 Diabetes 41, 12571266 (1992).
63. Bird,S.T., Hartzema,A.G., Brophy,J.M., Etminan,M. (2010). 109. Ciaraldi,T.P. etal. Cellular mechanisms of insulin
& Delaney,J.A. Risk of venous thromboembolism 87. Dumesic,D.A. etal. Scientific statement on the resistance in polycystic ovarian syndrome. J.Clin.
inwomen with polycystic ovary syndrome: diagnostic criteria, epidemiology, pathophysiology, Endocrinol. Metab. 75, 577583 (1992).
apopulationbased matched cohort analysis. CMAJ andmolecular genetics of polycystic ovary syndrome. 110. Corbould,A., Zhao,H., Mirzoeva,S., Aird,F.
185, E115E120 (2013). Endocr. Rev. 36, 487525 (2015). &Dunaif,A. Enhanced mitogenic signaling in skeletal
64. Anderson,S.A., Barry,J.A. & Hardiman,P.J. Risk of This paper provides a comprehensive overview of muscle of women with polycystic ovary syndrome.
coronary heart disease and risk of stroke in women with the complexity of PCOS and also provides some Diabetes 55, 751759 (2006).

16 | 2016 | VOLUME 2 www.nature.com/nrdp



2
0
1
6
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
PRIMER

111. Rajkhowa,M. etal. Insulin resistance in polycystic 133. Shroff,R., Syrop,C.H., Davis,W., VanVoorhis,B.J. 153. Skubleny,D. etal. The impact of bariatric surgery on
ovary syndrome is associated with defective regulation &Dokras,A. Risk of metabolic complications in polycystic ovary syndrome: asystematic review and
of ERK1/2 by insulin in skeletal muscle invivo. thenew PCOS phenotypes based on the Rotterdam meta-analysis. Obes. Surg. 26, 169176 (2016).
Biochem.J. 418, 665671 (2009). criteria. Fertil. Steril. 88, 13891395 (2007). 154. Tang,T., Lord,J.M., Norman,R.J., Yasmin,E.
112. Book,C.B. & Dunaif,A. Selective insulin resistance 134. Daan,N.M. etal. Cardiovascular and metabolic &Balen,A.H. Insulin-sensitising drugs (metformin,
inthe polycystic ovary syndrome. J.Clin. Endocrinol. profiles amongst different polycystic ovary syndrome rosiglitazone, pioglitazone, dchiro-inositol) for women
Metab. 84, 31103116 (1999). phenotypes: who is really at risk? Fertil. Steril. 102, with polycystic ovary syndrome, oligo amenorrhoea
113. Corbould,A. etal. Insulin resistance in the skeletal 14441451.e3 (2014). and subfertility. Cochrane Database Syst. Rev. 5,
muscle of women with PCOS involves intrinsic and 135. Daskalopoulos,G. etal. Excess metabolic and CD003053 (2012).
acquired defects in insulin signaling. Am. J.Physiol. cardiovascular risk is not manifested in all phenotypes 155. Naderpoor,N. etal. Metformin and lifestyle
Endocrinol. Metab. 288, E1047E1054 (2005). of polycystic ovary syndrome: implications for diagnosis modification in polycystic ovary syndrome: systematic
114. Eriksen,M. etal. Insulin resistance is not conserved and treatment. Curr. Vasc. Pharmacol. 13, 788800 review and meta-analysis. Hum. Reprod. Update 21,
inmyotubes established from women with PCOS. (2015). 560574 (2015).
PLoSONE 5, e14469 (2010). 136. Tehrani,F.R., Rashidi,H., Khomami,M.B., Tohidi,M. 156. Du,Q. etal. A systematic review and meta-analysis
115. Ciaraldi,T.P., Aroda,V., Mudaliar,S., Chang,R.J. & Azizi,F. The prevalence of metabolic disorders ofrandomized controlled trials comparing pioglitazone
&Henry,R.R. Polycystic ovary syndrome is associated invarious phenotypes of polycystic ovary syndrome: versus metformin in the treatment of polycystic ovary
with tissue-specific differences in insulin resistance. acommunity based study in Southwest of Iran. syndrome. Curr. Med. Res. Opin. 28, 723730
J.Clin. Endocrinol. Metab. 94, 157163 (2009). Reprod. Biol. Endocrinol. 12, 89 (2014). (2012).
116. Chazenbalk,G. etal. Regulation of adiponectin 137. Ibanez,L., LopezBermejo,A., Diaz,M., Marcos,M.V. 157. Li,X.J. etal. Metformin versus thiazolidinediones
secretion by adipocytes in the polycystic ovary & deZegher,F. Early metformin therapy fortreatment of clinical, hormonal and metabolic
syndrome: role of tumor necrosis factor-. J.Clin. (age812years) in girls with precocious pubarche characteristics of polycystic ovary syndrome: ameta-
Endocrinol. Metab. 95, 935942 (2010). toreduce hirsutism, androgen excess, and analysis. Clin. Endocrinol. (Oxf.) 74, 332339 (2011).
117. MannerasHolm,L. etal. Adipose tissue has aberrant oligomenorrhea in adolescence. J.Clin. Endocrinol. 158. Goodman,N.F. etal. American Association Of Clinical
morphology and function in PCOS: enlarged Metab. 96, E1262E1267 (2011). Endocrinologists, American College Of Endocrinology,
adipocytes and low serum adiponectin, but not 138. Legro,R.S. etal. Diagnosis and treatment of polycystic And Androgen Excess And PCOS Society Disease State
circulating sex steroids, are strongly associated ovary syndrome: anEndocrine Society clinical practice clinical review: guide to the best practices in the
withinsulin resistance. J.Clin. Endocrinol. Metab. 96, guideline. J.Clin. Endocrinol. Metab. 98, 45654592 evaluation and treatment of polycystic ovary syndrome
E304E311 (2011). (2013). part 2. Endocr. Pract. 21, 14151426 (2015).
118. Ek,I., Arner,P., Bergqvist,A., Carlstrom,K. 139. Fauser,B.C. etal. Consensus on womens health 159. Formuso,C., Stracquadanio,M. & Ciotta,L.
&Wahrenberg,H. Impaired adipocyte lipolysis in aspects of polycystic ovary syndrome (PCOS): Myoinositol versus dchiro inositol in PCOS treatment.
nonobese women with the polycystic ovary syndrome: theAmsterdam ESHRE/ASRM-Sponsored 3rd PCOS Minerva Ginecol. 67, 321325 (2015).
apossible link to insulin resistance? J.Clin. Endocrinol. Consensus Workshop Group. Fertil. Steril. 97, 160. Nestler,J.E. & Unfer,V. Reflections on inositol(s)
Metab. 82, 11471153 (1997). 2838.e25 (2012). forPCOS therapy: steps toward success.
119. Rosenbaum,D., Haber,R.S. & Dunaif,A. Insulin 140. Azziz,R. etal. Positions statement: criteria for defining Gynecol.Endocrinol. 31, 501505 (2015).
resistance in polycystic ovary syndrome: decreased polycystic ovary syndrome as a predominantly 161. Zhang,Y.Y., Hou,L.Q. & Zhao,T.Y. Effects of acarbose
expression of GLUT4 glucose transporters in hyperandrogenic syndrome: anAndrogen Excess on polycystic ovary syndrome: ameta-analysis.
adipocytes. Am. J.Physiol. 264, E197E202 (1993). Society guideline. J.Clin. Endocrinol. Metab. 91, Exp.Clin. Endocrinol. Diabetes 122, 373378 (2014).
120. Garvey,W.T. etal. Pretranslational suppression of a 42374245 (2006). 162. Gao,L., Zhao,F.L. & Li,S.C. Statin is a reasonable
glucose transporter protein causes insulin resistance 141. Moran,L.J. etal. Sleep disturbances in a treatment option for patients with polycystic ovary
inadipocytes from patients with non-insulin-dependent communitybased sample of women with polycystic syndrome: ameta-analysis of randomized controlled
diabetes mellitus and obesity. J.Clin. Invest. 87, ovary syndrome. Hum. Reprod. 30, 466472 trials. Exp. Clin. Endocrinol. Diabetes 120, 367375
10721081 (1991). (2015). (2012).
121. Carvalho,E., Jansson,P.A., Nagaev,I., Wenthzel,A.M. 142. Thomson,R.L. etal. Lifestyle management improves 163. Sun,J. etal. An investigation into the therapeutic
& Smith,U. Insulin resistance with low cellular IRS1 quality of life and depression in overweight and obese effects of statins with metformin on polycystic ovary
expression is also associated with low GLUT4 women with polycystic ovary syndrome. Fertil. Steril. syndrome: ameta-analysis of randomised controlled
expression and impaired insulin-stimulated glucose 94, 18121816 (2010). trials. BMJ Open 5, e007280 (2015).
transport. FASEBJ. 15, 11011103 (2001). 143. Moran,L.J., Hutchison,S.K., Norman,R.J. 164. Khera,R. etal. Association of pharmacological
122. Carlson,C.J., Koterski,S., Sciotti,R.J., Poccard,G.B. &Teede,H.J. Lifestyle changes in women with treatments for obesity with weight loss and adverse
& Rondinone,C.M. Enhanced basal activation of polycystic ovary syndrome. Cochrane Database Syst. events: asystematic review and meta-analysis. JAMA
mitogen-activated protein kinases in adipocytes Rev. 7, CD007506 (2011). 315, 24242434 (2016).
fromtype2 diabetes: potential role of p38 in the A meta-analysis of the data reported on lifestyle 165. DiamantiKandarakis,E., Katsikis,I., Piperi,C.,
downregulation of GLUT4 expression. Diabetes 52, changes, and summarizes results that may be Alexandraki,K. & Panidis,D. Effect of long-term
634641 (2003). obtained and problems that arise with this orlistat treatment on serum levels of advanced
123. Chen,Y.H. etal. miRNA93 inhibits GLUT4 and is treatment. glycation end-products in women with polycystic ovary
overexpressed in adipose tissue of polycystic ovary 144. Carmina,E. PCOS: metabolic impact and long-term syndrome. Clin. Endocrinol. (Oxf.) 66, 103109
syndrome patients and women with insulin resistance. management. Minerva Ginecol. 64, 501505 (2012). (2007).
Diabetes 62, 22782286 (2013). 145. Kiddy,D.S. etal. Improvement in endocrine and 166. Jia,X.Z. etal. Effect of vitaminD on clinical and
124. Ciaraldi,T.P. Molecular defects of insulin action in the ovarian function during dietary treatment of obese biochemical parameters in polycystic ovary syndrome
polycystic ovary syndrome: possible tissue specificity. women with polycystic ovary syndrome. women: ameta-analysis. J.Obstetr. Gynaecol. Res. 41,
J.Pediatr. Endocrinol. Metab. 13 (Suppl.5), Clin.Endocrinol. (Oxf.) 36, 105111 (1992). 17911802 (2015).
12911293 (2000). 146. Moran,L.J. etal. Dietary composition in restoring 167. Mathur,R., Levin,O. & Azziz,R. Use of
125. Ciaraldi,T.P. etal. Cellular insulin resistance in reproductive and metabolic physiology in overweight ethinylestradiol/drospirenone combination in patients
adipocytes from obese polycystic ovary syndrome women with polycystic ovary syndrome. J.Clin. with the polycystic ovary syndrome. Ther. Clin. Risk
subjects involves adenosine modulation of insulin Endocrinol. Metab. 88, 812819 (2003). Manag. 4, 487492 (2008).
sensitivity. J.Clin. Endocrinol. Metab. 82, 14211425 147. Moran,L.J. etal. Dietary composition in the treatment 168. Conway,G. etal. The polycystic ovary syndrome:
(1997). of polycystic ovary syndrome: asystematic review to aposition statement from the European Society of
126. Chuang,T.Y. etal. MicroRNA223 expression is inform evidence-based guidelines. J.Acad. Nutr. Diet. Endocrinology. Eur. J.Endocrinol. 171, 129 (2014).
upregulated in insulin resistant human adipose tissue. 113, 520545 (2013). 169. Yildiz,B.O. Approach to the patient: contraception
J.Diabetes Res. 2015, 943659 (2015). 148. Goss,A.M. etal. Effects of a eucaloric reduced- inwomen with polycystic ovary syndrome. J.Clin.
127. McAllister,J.M., Legro,R.S., Modi,B.P. carbohydrate diet on body composition and fat Endocrinol. Metab. 100, 794802 (2015).
&Strauss,J.F.3rd. Functional genomics of PCOS: distribution in women with PCOS. Metabolism 63, 170. Centers for Disease Control and Prevention (CDC).
from GWAS to molecular mechanisms. Trends 12571264 (2014). U.S.medical eligibility criteria for contraceptive use,
Endocrinol. Metab. 26, 118124 (2015). 149. Gower,B.A. etal. Favourable metabolic effects 2010. MMWR Recomm. Rep. 59, 186 (2010).
128. Yildiz,B.O. & Azziz,R. The adrenal and polycystic of a eucaloric lower-carbohydrate diet in women with 171. Azziz,R. etal. Leuprolide and estrogen versus oral
ovary syndrome. Rev. Endocr. Metab. Disord. 8, PCOS. Clin. Endocrinol. (Oxf.) 79, 550557 (2013). contraceptive pills for the treatment of hirsutism:
331342 (2007). 150. Roessler,K.K., Birkebaek,C., Ravn,P., aprospective randomized study. J.Clin. Endocrinol.
129. Deeks,A.A., GibsonHelm,M.E. & Teede,H.J. Andersen,M.S. & Glintborg,D. Effects of exercise Metab. 80, 34063411 (1995).
Anxiety and depression in polycystic ovary syndrome: andgroup counselling on body composition and 172. Naka,K.K. etal. Effect of the insulin sensitizers
acomprehensive investigation. Fertil. Steril. 93, VO2max in overweight women with polycystic ovary metformin and pioglitazone on endothelial function
24212423 (2010). syndrome. Acta Obstetricia Gynecol. Scand. 92, inyoung women with polycystic ovary syndrome:
130. Moran,L., GibsonHelm,M., Teede,H. & Deeks,A. 272277 (2013). aprospective randomized study. Fertil. Steril. 95,
Polycystic ovary syndrome: abiopsychosocial 151. Harrison,C.L., Lombard,C.B., Moran,L.J. 203209 (2011).
understanding in young women to improve knowledge &Teede,H.J. Exercise therapy in polycystic ovary 173. Donesky,B.W. & Adashi,E.Y. Surgically induced
and treatment options. J.Psychosomat. Obstetr. syndrome: asystematic review. Hum. Reprod. Update ovulation in the polycystic ovary syndrome: wedge
Gynaecol. 31, 2431 (2010). 17, 171183 (2011). resection revisited in the age of laparoscopy.
131. GibsonHelm,M.E., Lucas,I.M., Boyle,J.A. 152. Harrison,C.L., Stepto,N.K., Hutchison,S.K. Fertil.Steril. 63, 439463 (1995).
&Teede,H.J. Womens experiences of polycystic ovary &Teede,H.J. The impact of intensified exercise 174. Farquhar,C., Brown,J. & Marjoribanks,J.
syndrome diagnosis. Fam. Pract. 31, 545549 (2014). training on insulin resistance and fitness in overweight Laparoscopic drilling by diathermy or laser for
132. Kitzinger,C. & Willmott,J. The thief of womanhood: and obese women with and without polycystic ovary ovulation induction in anovulatory polycystic ovary
womens experience of polycystic ovarian syndrome. syndrome. Clin. Endocrinol. (Oxf.) 76, 351357 syndrome. Cochrane Database Syst. Rev. 6,
Soc. Sci. Med. 54, 349361 (2002). (2012). CD001122 (2012).

NATURE REVIEWS | DISEASE PRIMERS VOLUME 2 | 2016 | 17



2
0
1
6
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
PRIMER

175. Abu Hashim,H., AlInany,H., DeVos,M. 196. Nahuis,M.J. etal. Long-term outcomes in women characterised by impaired glucose tolerance, reduced
&Tournaye,H. Three decades after Gjonnaesss with polycystic ovary syndrome initially randomized insulin sensitivity and related metabolic defects.
laparoscopic ovarian drilling for treatment of PCOS; toreceive laparoscopic electrocautery of the ovaries Diabetologia 50, 24242432 (2007).
what do we know? An evidence-based approach. orovulation induction with gonadotrophins. 217. Fraser,G.L., Ramael,S., Hoveyda,H.R., Gheyle,L.
Arch.Gynecol. Obstetr. 288, 409422 (2013). Hum.Reprod. 26, 18991904 (2011). &Combalbert,J. The NK3 receptor antagonist
176. Koulouri,O. & Conway,G.S. A systematic review 197. Cha,K.Y. etal. Obstetric outcome of patients ESN364 suppresses sex hormones in men and women.
ofcommonly used medical treatments for hirsutism in withpolycystic ovary syndrome treated by invitro J.Clin. Endocrinol. Metab. 101, 417426 (2016).
women. Clin. Endocrinol. (Oxf.) 68, 800805 (2008). maturation and invitro fertilizationembryo transfer. 218. Azziz,R., Dumesic,D.A. & Goodarzi,M.O. Polycystic
177. Yildiz,B.O. Assessment, diagnosis and treatment of Fertil. Steril. 83, 14611465 (2005). ovary syndrome: anancient disorder? Fertil. Steril. 95,
apatient with hirsutism. Nat. Clin. Prac. Endocrinol. 198. Siristatidis,C. etal. Invitro maturation in women 15441548 (2011).
Metab. 4, 294300 (2008). withversus without polycystic ovarian syndrome: 219. SirPetermann,T. etal. Early metabolic derangements
178. Swiglo,B.A. etal. Clinical review: antiandrogens for asystematic review and meta-analysis. PLoS ONE 10, in daughters of women with polycystic ovary syndrome.
the treatment of hirsutism: asystematic review and e0134696 (2015). J.Clin. Endocrinol. Metab. 92, 46374642 (2007).
metaanalyses of randomized controlled trials. J.Clin. 199. Walls,M.L. etal. Invitro maturation as an alternative 220. SirPetermann,T. etal. Metabolic and reproductive
Endocrinol. Metab. 93, 11531160 (2008). to standard invitro fertilization for patients diagnosed features before and during puberty in daughters
179. vanZuuren,E.J., Fedorowicz,Z., Carter,B. with polycystic ovaries: acomparative analysis of fresh, ofwomen with polycystic ovary syndrome. J.Clin.
&Pandis,N. Interventions for hirsutism frozen and cumulative cycle outcomes. Hum. Reprod. Endocrinol. Metab. 94, 19231930 (2009).
(excludinglaser and photoepilation therapy alone). 30, 8896 (2015). 221. Maliqueo,M. etal. Adrenal function during childhood
CochraneDatabase Syst. Rev. 4, CD010334 (2015). 200. Shi,Y. etal. Live birth after fresh embryo transfer and puberty in daughters of women with polycystic
180. Venturoli,S. etal. A prospective randomized trial versus elective embryo cryopreservation/frozen embryo ovary syndrome. J.Clin. Endocrinol. Metab. 94,
comparing low dose flutamide, finasteride, transfer in women with polycystic ovary syndrome 32823288 (2009).
ketoconazole, and cyproterone acetate-estrogen undergoing IVF (FreFro-PCOS): study protocol for a 222. Battaglia,C. etal. Polycystic ovaries in childhood:
regimens in the treatment of hirsutism. J.Clin. multicenter, prospective, randomized controlled clinical acommon finding in daughters of PCOS patients.
Endocrinol. Metab. 84, 13041310 (1999). trial. Trials 15, 154 (2014). Apilot study. Hum. Reprod. 17, 771776 (2002).
181. EscobarMorreale,H.F. etal. Epidemiology, diagnosis 201. Tang,T., Glanville,J., Orsi,N., Barth,J.H. 223. Ibanez,L. etal. Postpubertal outcome in girls
and management of hirsutism: aconsensus statement &Balen,A.H. The use of metformin for women diagnosed of premature pubarche during childhood:
by the Androgen Excess and Polycystic Ovary withPCOS undergoing IVF treatment. Hum. Reprod. increased frequency of functional ovarian
Syndrome Society. Hum. Reprod. Update 18, 21, 14161425 (2006). hyperandrogenism. J.Clin. Endocrinol. Metab.
146170 (2012). 202. Tso,L.O., Costello,M.F., Albuquerque,L.E., 76, 15991603 (1993).
182. Townsend,K.A. & Marlowe,K.F. Relative safety Andriolo,R.B. & Freitas,V. Metformin treatment 224. Meas,T. etal. Endocrine consequences of premature
andefficacy of finasteride for treatment of hirsutism. before and during IVF or ICSI in women with polycystic pubarche in post-pubertal Caucasian girls.
Ann.Pharmacother. 38, 10701073 (2004). ovary syndrome. Cochrane Database Syst. Rev. 2, Clin.Endocrinol. (Oxf.) 57, 101106 (2002).
183. Moghetti,P. etal. Comparison of spironolactone, CD006105 (2009). 225. McCartney,C.R. etal. The association of obesity
flutamide, and finasteride efficacy in the treatment 203. McHorney,C.A. Health status assessment methods for andhyperandrogenemia during the pubertal transition
ofhirsutism: arandomized, double blind, placebo- adults: past accomplishments and future challenges. in girls: obesity as a potential factor in the genesis
controlled trial. J.Clin. Endocrinol. Metab. 85, 8994 Annu. Rev. Public Health 20, 309335 (1999). ofpostpubertal hyperandrogenism. J.Clin. Endocrinol.
(2000). 204. Li,Y. etal. Polycystic ovary syndrome is associated Metab. 91, 17141722 (2006).
This randomized controlled trial clearly withnegatively variable impacts on domains of health- 226. vanHooff,M.H. etal. Predictive value of menstrual
demonstrates the positive effect and comparability related quality of life: evidence from a meta-analysis. cycle pattern, body mass index, hormone levels
of androgen blockade on hirsutism in PCOS. Fertil. Steril. 96, 452458 (2011). andpolycystic ovaries at age 15years for oligo-
184. Balfour,J.A. & McClellan,K. Topical eflornithine. Am. 205. Cronin,L. etal. Development of a health-related amenorrhoea at age 18years. Hum. Reprod. 19,
J.Clin. Dermatol. 2, 197201; discussion 202 (2001). quality-oflife questionnaire (PCOSQ) for women with 383392 (2004).
185. Lizneva,D., GavrilovaJordan,L., Walker,W. & Azziz,R. polycystic ovary syndrome (PCOS). J.Clin. Endocrinol. This study demonstrates the predictive nature of
Androgen excess: investigations and management. Metab. 83, 19761987 (1998). the degree of menstrual dysfunction, particularly
Best Pract. Res. Clin. Obstet. Gynaecol. http:// The authors of this article developed the first for persistent menstrual dysfunction in adolescents.
dx.doi.org/10.1016/j.bpobgyn.2016.05.003 (2016). PCOS-dedicated health-related QOL questionnaire 227. WikstenAlmstromer,M., Hirschberg,A.L.
186. Boomsma,C.M. etal. A meta-analysis of pregnancy for PCOS. &Hagenfeldt,K. Prospective followup of menstrual
outcomes in women with polycystic ovary syndrome. 206. Barnard,L. etal. Quality of life and psychological well disorders in adolescence and prognostic factors. Acta
Hum. Reprod. Update 12, 673683 (2006). being in polycystic ovary syndrome. Hum. Reprod. 22, Obstetricia Gynecol. Scand. 87, 11621168 (2008).
187. Legro,R.S. etal. Clomiphene, metformin, or both 22792286 (2007). 228. Fanelli,F. etal. Androgen profiling by liquid
forinfertility in the polycystic ovary syndrome. N.Engl. 207. Hahn,S. etal. Metformin treatment of polycystic ovary chromatography-tandem mass spectrometry
J.Med. 356, 551566 (2007). syndrome improves health-related quality-oflife, (LCMS/MS) in healthy normal-weight ovulatory
This is the first large randomized controlled trial emotional distress and sexuality. Hum. Reprod. 21, andanovulatory late adolescent and young women.
todemonstrate the relative inferiority of metformin 19251934 (2006). J.Clin. Endocrinol. Metab. 98, 30583067 (2013).
as a primary ovulatory agent in PCOS. 208. Ladson,G. etal. The effects of metformin with lifestyle 229. Villarroel,C. etal. Hirsutism and oligomenorrhea are
188. Legro,R.S. etal. Letrozole versus clomiphene for therapy in polycystic ovary syndrome: arandomized appropriate screening criteria for polycystic ovary
infertility in the polycystic ovary syndrome. N.Engl. double-blind study. Fertil. Steril. 95, 10591066.e7 syndrome in adolescents. Gynecol. Endocrinol. 31,
J.Med. 371, 119129 (2014). (2011). 625629 (2015).
189. Zain,M.M., Jamaluddin,R., Ibrahim,A. 209. HarrisGlocker,M., Davidson,K., Kochman,L., 230. Ibanez,L. etal. Polycystic ovaries in nonobese
&Norman,R.J. Comparison of clomiphene citrate, Guzick,D. & Hoeger,K. Improvement in quality-oflife adolescents and young women with ovarian androgen
metformin, or the combination of both for first-line questionnaire measures in obese adolescent females excess: relation to prenatal growth. J.Clin. Endocrinol.
ovulation induction, achievement of pregnancy, and with polycystic ovary syndrome treated with lifestyle Metab. 93, 196199 (2008).
live birth in Asian women with polycystic ovary changes and oral contraceptives, with or without 231. Cresswell,J.L. etal. Fetal growth, length of gestation,
syndrome: arandomized controlled trial. Fertil. Steril. metformin. Fertil. Steril. 93, 10161019 (2010). and polycystic ovaries in adult life. Lancet 350,
91, 514521 (2009). 210. Cinar,N., Harmanci,A., Demir,B. & Yildiz,B.O. 11311135 (1997).
190. Sagle,M.A., HamiltonFairley,D., Kiddy,D.S. Effectof an oral contraceptive on emotional distress, 232. DeFronzo,R.A., Bonadonna,R.C. & Ferrannini,E.
&Franks,S. A comparative, randomized study of anxiety and depression of women with polycystic ovary Pathogenesis of NIDDM. A balanced overview.
lowdose human menopausal gonadotropin and syndrome: aprospective study. Hum. Reprod. 27, Diabetes Care 15, 318368 (1992).
follicle-stimulating hormone in women with polycystic 18401845 (2012).
ovarian syndrome. Fertil. Steril. 55, 5660 (1991). 211. Deeks,A.A., GibsonHelm,M.E., Paul,E. Acknowledgements
191. Polson,D.W., Mason,H.D., Saldahna,M.B. &Teede,H.J. Is having polycystic ovary syndrome The authors thank Y.-H. Chen, Augusta University, Georgia,
&Franks,S. Ovulation of a single dominant follicle apredictor of poor psychological function including USA.
during treatment with low-dose pulsatile follicle anxiety and depression? Hum. Reprod. 26,
stimulating hormone in women with polycystic ovary 13991407 (2011). Author contributions
syndrome. Clin. Endocrinol. (Oxf.) 26, 205212 212. Elsenbruch,S. etal. Determinants of emotional Introduction (R.A.); Epidemiology (R.A.); Mechanisms/
(1987). distress in women with polycystic ovary syndrome. pathophysiology (R.A., E.C., Z.-J.C., A.D., J.S.E.L., H.J.T. and
192. Homburg,R. etal. Clomifene citrate or low-dose FSH Hum. Reprod. 21, 10921099 (2006). B.O.Y.); Diagnosis, screening and prevention (R.A. and H.J.T.);
for the first-line treatment of infertile women with 213. Liu,D.M. etal. Evidence for gonadotrophin secretory Management (R.A., E.C., Z.-J.C., R.S.L., D.L. and B.O.Y.);
anovulation associated with polycystic ovary and steroidogenic abnormalities in brothers of women Quality of life (E.C. and B.O.Y.); Outlook (R.A., B.N.-H., D.L.
syndrome: aprospective randomized multinational with polycystic ovary syndrome. Hum. Reprod. 29, and R.S.L.); Overview of Primer (R.A.).
study. Hum. Reprod. 27, 468473 (2012). 27642772 (2014).
193. Solnik,M.J. in Te Lindes Atlas of Gynecologic Surgery 214. Coviello,A.D., Sam,S., Legro,R.S. & Dunaif,A. High Competing interests
(eds Cundiff,G.W., Azziz,R. & Bristow,R.E.) prevalence of metabolic syndrome in first-degree male R.A. has a consulting agreement with KinDex Pharmaceutical
217222 (Lippincott Williams & Wilkins, 2013). relatives of women with polycystic ovary syndrome is Inc., is on the advisory board of Global PET Imaging, and has
194. Thessaloniki ESHRE/ASRM-Sponsored PCOS related to high rates of obesity. J.Clin. Endocrinol. a consulting appointment with Selge Holdings and Ventures.
Consensus Workshop Group. Consensus on infertility Metab. 94, 43614366 (2009). J.S.E.L. has received unrestricted research grants from
treatment related to polycystic ovary syndrome. 215. Recabarren,S.E. etal. Metabolic profile in sons Ferring, Merck-Serono, MSD, Schering Plough, Serono and
Fertil.Steril. 89, 505522 (2008). ofwomen with polycystic ovary syndrome. J.Clin. Okganon. R.S.L. is a consultant for Takeda, KinDex,
195. Api,M. Is ovarian reserve diminished after Endocrinol. Metab. 93, 18201826 (2008). Euroscreen and Ferring, and has received research funding
laparoscopic ovarian drilling? Gynecol. Endocrinol. 216. Baillargeon,J.P. & Carpentier,A.C. Brothers from and is a consultant for Ferring. All other authors declare
25, 159165 (2009). ofwomen with polycystic ovary syndrome are no competing interests.

18 | 2016 | VOLUME 2 www.nature.com/nrdp



2
0
1
6
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.

Das könnte Ihnen auch gefallen