AACE/ACE Disease State Clinical Review

Neil F. Goodman, MD, FACE1; Rhoda H. Cobin, MD, MACE2; Walter Futterweit, MD, FACP, FACE3;
Jennifer S. Glueck, MD4; Richard S. Legro, MD, FACOG5; Enrico Carmina, MD6
EXECUTIVE SUMMARY

Polycystic Ovary Syndrome (PCOS) is recognized as
the most common endocrine disorder of reproductive-aged
women around the world. This document, produced by
the collaboration of the American Association of Clinical
Endocrinologists (AACE) and the Androgen Excess and
PCOS Society (AES) aims to highlight the most important
clinical issues confronting physicians and their patients
with PCOS. It is a summary of current best practices in
2015.

Submitted for publication March 26, 2015

Accepted for publication August 6, 2015
From the 1University of Miami Miller School of Medicine, Miami, Florida,
2Icahn School of Medicine at Mount Sinai, New York, New York, 3Icahn
School of Medicine at Mount Sinai, Department of Medicine, Division of
Endocrinology, New York, New York, 4Reproductive and Endocrine Health,
Highland Park, Illinois, 5Penn State University College of Medicine,M.S.
Hershey Medical Center, Hershey, Pennsylvania, and 6Department of
Society, Law and Sport Sciences, University of Palermo, Palermo, Italy.
Address correspondence to Dr. Neil F. Goodman, 9150 SW 87th Ave, Ste 210,
Miami, FL 33176. E-mail: [email protected]
DOI: 10.4158/EP15748.DSC
To purchase reprints of this article, please visit: www.aace.com/reprints.
Copyright 2015 AACE.
The opinions represented in the AACE/ACE Disease State Clinical Review:
Guide to the Best Practices in the Evaluation and Treatment of Polycystic
Ovary Syndrome are the expressed opinions of the Reproductive Endocrinology
Scientific Committee of the American Association of Clinical Endocrinologists
and the Androgen Excess and PCOS Society. AACE/ACE Disease State Clinical
Reviews are systematically developed documents written to assist health care professionals in medical decision making for specific clinical conditions, but are in no way
a substitute for a medical professionals independent judgment and should not be
considered medical advice. Most of the content herein is based on literature reviews.
In areas of uncertainty, professional judgment of the authors was applied.
This review article is a working document that reflects the state of the field at the
time of publication. Because rapid changes in this area are expected, periodic revisions are inevitable. We encourage medical professionals to use this information in
conjunction with, and not a replacement for, their best clinical judgment. The presented recommendations may not be appropriate in all situations. Any decision by
practitioners to apply these guidelines must be made in light of local resources and
individual patient circumstances.
Copyright 2015 AACE.

PCOS has been defined using various criteria,

including menstrual irregularity, hyperandrogenism, and polycystic ovary morphology (PCOM).

General agreement exists among specialty society guidelines that the diagnosis of PCOS must be
based on the presence of at least two of the following three criteria: chronic anovulation, hyperandrogenism (clinical or biological) and polycystic
ovaries.

There is need for careful clinical assessment of

womens history, physical examination, and laboratory evaluation, emphasizing the accuracy and
validity of the methodology used for both biochemical measurements and ovarian imaging.

Free testosterone (T) levels are more sensitive

than the measurement of total T for establishing
the existence of androgen excess and should be
ideally determined through equilibrium dialysis
techniques. Value of measuring levels of androgens other than T in patients with PCOS is relatively low.
New ultrasound machines allow diagnosis of
PCOM in patients having at least 25 small follicles (2 to 9 mm) in the whole ovary. Ovarian size
at 10 mL remains the threshold between normal
and increased ovary size.
Serum 17-hydroxyprogesterone and anti-Mllerian hormone are useful for determining a diagnosis of PCOS.

Correct diagnosis of PCOS impacts on the likelihood of associated metabolic and cardiovascular risks and leads to appropriate intervention,
depending upon the womans age, reproductive
status, and her own concerns. The management of
women with PCOS should include reproductive
function, as well as the care of hirsutism, alopecia, and acne.

ENDOCRINE PRACTICE Vol 21 No. 11 November 2015 1291

1292 PCOS Best Practices, Endocr Pract. 2015;21(No. 11)

Cycle length >35 days suggests chronic anovulation, but cycle length slightly longer than normal (32 to 35 days) or slightly irregular (32 to
35-36 days) needs assessment for ovulatory dysfunction. Ovulatory dysfunction is associated with
increased prevalence of endometrial hyperplasia
and endometrial cancer, in addition to infertility.
In PCOS, hirsutism develops gradually and intensifies with weight gain. In the neoplastic virilizing
states, hirsutism is of rapid onset, usually associated with clitoromegaly and oligomenorrhea.
Girls with severe acne or acne resistant to
oral and topical agents, including isotretinoin
(Accutane), may have a 40% likelihood of developing PCOS.
Hair loss patterns are variable in women with
hyperandrogenemia, typically the vertex, crown
or diffuse pattern, whereas women with more
severe hyperandrogenemia may see bitemporal
hair loss and loss of the frontal hairline.
Oral contraceptives (OCPs) can effectively
lower androgens and block the effect of androgens via suppression of ovarian androgen production and by increasing sex hormonebinding
globulin.
Physiologic doses of dexamethasone or prednisone can directly lower adrenal androgen output.
Anti-androgens can be used to block the effects of
androgen in the pilosebaceous unit or in the hair
follicle. Anti-androgen therapy works through
competitive antagonism of the androgen receptor
(spironolactone, cyproterone acetate, flutamide)
or inhibition of 5-reductase (finasteride) to
prevent the conversion of T to its more potent
form, 5-dihydrotestosterone. The choice of antiandrogen therapy is guided by symptoms.

The diagnosis of PCOS in adolescents is particularly challenging given significant age and
developmental issues in this group. Management
of infertility in women with PCOS requires an
understanding of the pathophysiology of anovulation as well as currently available treatments.

Many features of PCOS, including acne, menstrual irregularities, and hyperinsulinemia, are
common in normal puberty. Menstrual irregularities with anovulatory cycles and varied cycle
length are common due to the immaturity of the
hypothalamic-pituitary-ovarian axis in the 2- to
3-year time period post-menarche. Persistent
oligomenorrhea 2 to 3 years beyond menarche
predicts ongoing menstrual irregularities and
greater likelihood of underlying ovarian or adrenal dysfunction.

In adolescent girls, large, multicystic ovaries

are a common finding, so ultrasound is not a firstline investigation in women <17 years of age.
Ovarian dysfunction in adolescents should
be based on oligomenorrhea and/or biochemical evidence of oligo/anovulation, but there are
major limitations to the sensitivity of T assays in
ranges applicable to young girls.
Metformin is commonly used in young girls and
adolescents with PCOS as first-line monotherapy
or in combination with OCPs and anti-androgen
medications. In lean adolescent girls, a dose as
low as 850 mg daily may be effective at reducing
PCOS symptoms; in overweight and obese adolescents, dose escalation to 1.5 to 2.5 g daily is
likely required.
Anti-androgen therapy in adolescents could
affect bone mass, although available short-term
data suggest no effect on bone loss. (Endocr
Pract. 2015;21:1291-1300)

Abbreviations:
17OHP = 17 hydroxyprogesterone; 5R =
5-reductase; AA = androgenic alopecia; AES =
Androgen Excess Society; AMH = anti-Mllerian
hormone; BMI = body mass index; CV = cardiovascular; DHT = 5-dihydrotestosterone; FG = FerrimanGallwey; LH = luteinizing hormone; MetS = metabolic
syndrome; MS = mass spectrometry; NIH = National
Institutes of Health; OCP = oral contraceptive; PCOM
= polycystic ovary morphology; PCOS = polycystic
ovary syndrome; RIA = radioimmunoassay; SHBG =
sex hormonebinding globulin; SPA = spironolactone;
T = testosterone
INTRODUCTION

The past decade of research into polycystic ovary syndrome (PCOS) has produced important new insights into
the evaluation and treatment of this disorder. This document is intended as a guide, highlighting the most current
clinical information that a health care provider can use in
managing patients with this disorder. It is not intended as a
guideline but is written in a question and answer format by
experts in clinical practice.
DEFINING PCOS

The Rotterdam criteria for PCOS have been endorsed
by the National Institutes of Health (NIH). However,
Androgen Excess Society (AES) guidelines may correspond better to the pathogenesis of this disorder, as the AES
emphasizes the importance of clinical and/or biochemical hyperandrogenism and placing less importance on

PCOS Best Practices, Endocr Pract. 2015;21(No. 11) 1293

polycystic ovary morphology (PCOM). Table 1 lists a comparison of criteria for the diagnosis of PCOS. Nevertheless,
there is a general agreement that the diagnosis of PCOS
must be based on the presence of at least two of the following three criteria: chronic anovulation, hyperandrogenism
(clinical or biological), and polycystic ovaries. Despite this
consensus, many doubts remain for the clinician who has
to establish the existence of the criteria. In this section, we
will discuss the possible issues that must be solved by the
clinician in defining PCOS.
Establishing Hyperandrogenism
1. What Androgens Should Be Measured?
The issue of which serum androgen should be measured for diagnosis of PCOS remains controversial. Ideally,
assessments of free testosterone (T) levels are more sensitive than the measurement of total T for establishing the
existence of androgen excess (1). That said, although freeT measurements require equilibrium dialysis techniques,
many commercial laboratories use direct analogue radioimmunoassay (RIA), which is notoriously inaccurate (2,3).
Consequently, if the clinician is uncertain regarding the
quality of the free-T assay, it may be preferable to rely on
calculated free T, which has a good concordance and correlation with free T as measured by equilibrium dialysis
methods (4).

The value of measuring the levels of androgens other
than T in patients with PCOS is relatively low. Although
levels of dehydroepiandrosterone sulfate (DHEAS)
are increased in about 30 to 35% of PCOS patients (5),
its measurement does not add significantly to the diagnosis, and in the majority of the patients, free and total
T are also increased (5,6). Indeed, it has been estimated
that only 5% of patients with PCOS have an exclusive
increase in DHEAS (5). Similarly, measurements of
either 11-hydroxyandrostenedione or androstenedione

reportedly add only a few patients and are thus generally

not needed in clinical use (7).
2. What Technology Should Be Standard?

Based on the experience of Centers for Disease Control
laboratories (8), many commercial diagnostic laboratories
have switched to mass spectrometry (MS) coupled with
liquid chromatography (LC) (LC/MS) assays, which have
high sensitivity and specificity and provide accurate results.
Moreover, MS is the reference method at centers such as
the National Institute of Standards and Technology and is
considered to be the gold standard for measurement of
a variety of compounds. It should be noted, however, that
in PCOS, good RIA methods that incorporate purification
provide similar results (9). Methods that do not involve
purification should be avoided, as they tend to overestimate values of total T in an unpredictable manner and do
not generate reliable results (10). As previously discussed,
measurement of free T requires equilibrium dialysis techniques. If these techniques are not available, determination
of the free androgen index may be preferred in clinical
practice. In conclusion, the clinician should be aware of
the methods used by the laboratory and should avail the
services of laboratories using LC/MS assays or RIA with
purification.
3. Are There Normative Data References,
and Are Any Useful?

Using conventional RIAs with purification steps, normal values of T are generally lower than 60 ng/dL. Although
no normative values for T in women measured using LC/
MS technology are available, the greater specificity of MS
assays would anticipate lower total T levels (e.g., <50 ng/
dL). Although normal values with direct RIAs are much
higherup to 70 to 80 ng/dL and in some instances 100 ng/
dLthese should be avoided, as it becomes very difficult to
distinguish normal women from hyperandrogenic women.

Table 1
Criteria for the Diagnosis of Polycystic Ovary Syndrome
(Other Hormonal or Androgen Excess Conditions Being Previously Excluded)a
NIH/NICHD
(must meet both criteria)

ESHRE/ASRM
(Rotterdam criteria) 2004

Androgen Excess Society 2006

Includes all of the following:

Includes two of the following:

Includes all of the following:

Clinical and/or biochemical

hyperandrogenism

Clinical and/or biochemical

hyperandrogenism

Clinical and/or biochemical

hyperandrogenism

Menstrual dysfunction

Oligo-ovulation or anovulation
Polycystic ovaries

Ovarian dysfunction and/or

polycystic ovaries

Abbreviations: ESHRE/ASRM = European Society for Human Reproduction and Embryology/American Society for Reproductive
Medicine; NIH/NICH = National Institutes of Health/National Institute of Child Health and Human Disease.
a Adapted from Clin Epidemiol. 2014;6:1-13.

1294 PCOS Best Practices, Endocr Pract. 2015;21(No. 11)

4. To What Extent Should Hirsutism, Acne,

and Alopecia Be Discussed?
In adult women, hirsutism, alopecia, and acne are
good substitutes of biochemical hyperandrogenism and
should be considered as indicating a condition of excess
androgen production. This is not, however, the case during
adolescence, when acne is very common and often reversible, whereas alopecia is uncommon and generally has
other causes. During adolescence, therefore, only hirsutism
should be considered a substitute of biochemical hyperandrogenism (11).
Establishing Ovulatory Dysfunction
1. How Can Ovulatory Dysfunction Be Defined?
If cycle length is >35 days, it may be assumed that
chronic anovulation is present and no special tests are
needed. However, if cycle length is only slightly longer
than normal (32 to 35 days), or if cycles are slightly irregular, ranging from 32 to 35 to 36 days, ovulation should be
assessed. In addition, when the patients are hyperandrogenic (i.e., biochemical or clinical hyperandrogenism), the
possibility that apparently normal cycles are anovulatory
should be ruled out. Several studies have shown that 10
to 15% of hyperandrogenic women with apparently normal cycles are anovulatory (12,13). In contrast, the finding of anovulation is very uncommon in normoandrogenic
women with normal menses.
Measurement of serum progesterone during the
midluteal phase (days 21 to 22) is the best way to assess
ovulation. Whereas progesterone levels >2.5 ng/mL may
indicate ovulation, values 7 ng/mL are generally needed
for regular luteal function (14). Some physicians have
proposed using 3 consecutive luteal determinations with a
total serum value of 15 ng/mL to indicate normal luteal
function (15). Alternatives to progesterone measurement
(e.g., basal body temperature charts, urinary luteinizing
hormone [LH] kits or timed endometrial biopsies) may be
used, but they do not give sufficient information about the
luteal phase.
2. What Length of Menstrual Cycle Defines the
Threshold for Oligomenorrhea?

In adult women, a cycle length of 35 days is the
threshold for oligomenorrhea. During adolescence, the
threshold is higher, and a cycle length up to 40 days may be
considered normal, with longer menstrual cycles indicating
oligomenorrhea.
3. What Are the Clinical Implications of
Ovulatory Dysfunction in PCOS?

Infertility is, of course, the main clinical implication of
ovulatory dysfunction in PCOS. However, ovulatory dysfunction in association with other characteristics of PCOS,
such as obesity, is also associated with increased prevalence

of endometrial hyperplasia and endometrial cancer (16).

Oral contraceptive (OCP) treatment may be useful for
reducing endometrial cancer in PCOS, but, in any case,
prolonged absence of a menstrual cycle should be avoided.
In the recent joint meeting of the European Society of
Human Reproduction and Embryology and the American
Society for Reproductive Medicine in Amsterdam, induction of menstrual flow was recommended when the duration of the menstrual cycle is >3 months (17).
Polymenorrhea and/or hypermenorrhea are relatively
uncommon in PCOS, but when present, they may induce
iron deficiency anemia. Again, in such patients, OCP
therapy may be useful for reducing the risk of anemia.
Altered quality of life is another consequence of PCOS,
but this seems to depend more on obesity and hirsutism
than on altered menstrual cycles (18). However, in particular patients, psychological disturbances may be linked
to altered menstrual cycles, and this should be considered
when deciding the treatment of a PCOS patient.
Establishing PCOM
1. What Is the Standard Technology for
Evaluating Ovarian Morphology?

Ovarian morphology must be assessed by transvaginal
ultrasound. Although transabdominal ultrasounds may be
used in the case of young girls or patients from a particular cultural background, ovarian morphology assessment,
and, in particular, calculation of the number of small follicles, are less accurate using this technique. In the past few
years, rapid improvements in ultrasound technology have
increased by at least 2-fold the number of small follicles
that may be observed inside a single ovary, and the clinician should be aware that the results of ovarian sonography
strongly depend on the sensitivity of the ultrasound. Recent
guidelines from the AES have suggested that machines
using new software for automatic follicle numbering and
probes with a frequency of at least 8 mHz are needed for
optimal evaluation of ovarian morphology (19).
2. Are There Any Evidence-Based Criteria for
Defining PCOM by Ultrasound?

The existing Rotterdam guidelines suggest that PCOM
is indicated by the presence of at least 12 follicles measuring 2 to 9 mm in the whole ovary or by the finding
of increased ovarian size (>10 mL) (20). These criteria
were based on a completely different ultrasound technology however, and the unjustified increase in diagnoses of
PCOS that has arisen from their use in conjunction with
new ultrasound technology indicates that they are no longer appropriate in the clinic (21,22). New AES guidelines,
which are based upon a review of the data published using
new ultrasound technology, have increased the threshold
count of small ovarian follicles to 25 (19). When using the
new ultrasound machines, therefore, diagnosis of PCOM

PCOS Best Practices, Endocr Pract. 2015;21(No. 11) 1295

is possible in patients having at least 25 small follicles

(2 to 9 mm) in the whole ovary. Ovarian size threshold
has not been influenced by new technologies, and 10 mL
remains the threshold between normal and increased ovary
size. In certain populations and during adolescence or in
aging, however, a different threshold for ovarian size may
be needed (23).
It is now absolutely necessary for clinicians to know
what technology is used for assessment of ovarian morphology in their patients. If the clinician is not sure about
which ultrasound technology is used, the diagnosis of
PCOM should not be based on follicle count but only on
ovarian size and eventually, bearing in mind the assay
problems, on anti-Mllerian hormone (AMH) values (see
the Reproductive and Genetic Issues in PCOS section
in Part 2 of this Disease State Clinical Review, to be published in the next issue of Endocrine Practice)
3. What Other Laboratory Tests Should Be
Performed in Evaluating PCOS?
In addition to androgens and (eventually) progesterone, only assessment of serum 17-hydroxyprogesterone
(17OHP) and AMH are useful for determining a diagnosis
of PCOS. Because patients with nonclassic 21-hydroxylase deficiency may present as PCOS (hyperandrogenism, anovulation, and PCOM [24]), evaluation of serum
17OHP should be always included in a diagnostic study. In
these patients, a finding of 17OHP >10 ng/mL indicates the
existence of a 21-hydroxylase deficiency, whereas values
between 2 and 10 ng/mL suggest the need of further testing
with adrenocorticotropic hormone stimulus (24). Finally,
elevated AMH values (>4.5 ng/mL) may be useful as a
substitute for ovarian morphology when no accurate ovarian ultrasound is available (25).
HYPERANDROGENISM AND PCOS
1. When Are Hirsutism, Acne, and Alopecia Defined as
Clinical Signs of Hyperandrogenism?
Hirsutism

Hirsutism is defined as excessive hair growth in
women in a pattern consistent with androgen sensitivity.
Hair growth in certain anatomic areas is driven by androgens, and the development of excessive body hair in these
areas is a major clinical sign of hyperandrogenism. The
hair type present in most women with a hormonal hyperandrogenic disorder is coarse, thickened, pigmented, and
long and is called terminal hair. It differs from vellus hair,
which is fine, soft, unpigmented, and present in areas
where hair growth is not androgen dependent. Typically,
the onset of hirsutism in PCOS follows menarche, although
a minority of premenarche girls have earlier onset of pubic
hair development and some degree of hirsutism. Although

25 to 33% of white women have terminal hairs on the

upper lip and periareolar area, as well as linea alba areas,
the hirsutism in PCOS and a variety of other hyperandrogenic disorders is more pronounced. The presence of
substantial numbers of terminal hairs over the chin, neck,
lower face, and sideburns (particularly if extending medially) indicates the presence of androgen excess. Similarly,
excessive hair growth on the lower back, sternum, abdomen, shoulders, buttocks, perineal area, and inner thighs
is considered abnormal. Several clinical assessment scales
have been used in the grading of hirsutism. Although the
Hatch modification of the Ferriman-Gallwey (FG) scale
has been the most widely used, it is limited by its subjective nature and failure to include the sideburn, perineal, or
buttock areas (26). In PCOS, hirsutism develops gradually and intensifies with weight gain, whereas the neoplastic virilizing states involve a fairly rapid onset of severe
hirsutism, usually associated with clitoromegaly and
oligomenorrhea.

It should be noted that ethnic differences in the number of hair follicles present and individual skin sensitivity
of the pilosebaceous unit to androgens are major determinants of the presence of hirsutism, as well as acne and
androgenic alopecia (AA) (27). This emphasizes the wellknown fact that there are women with minimal or no hirsutism who may have increased serum T levels, whereas
other women with significant hirsutism may have normal
or only slightly increased androgen levels.
Acne
During adolescence, acne should not be considered
a substitute of hyperandrogenism (11), although girls
with severe acne or acne that is resistant to oral and topical agents, including isotretinoin (Accutane), may have a
40% likelihood of developing PCOS (28,29). When acne
persists after adolescence or is exacerbated in the mid-20s
or -30s, hyperandrogenemia is common, and acne may be
considered a clinical sign of hyperandrogenism. Those presenting with acne alone may have serum free T levels as
high as those seen in hyperandrogenic disease states, demonstrating hirsutism without acne (30).
Alopecia
In the normal hair cycle, the anagen or growth phase
lasts for 2 to 3 years and accounts for 85 to 90% of scalp
hairs. In the setting of androgen excess, androgen-sensitive
hair follicles shorten during the anagen phase, resulting in
miniaturization of the scalp hair, less scalp coverage, and
alopecia. The pattern of hair loss in women with hyperandrogenism is variable. For example, although hair loss patterns in women with hyperandrogenemia typically involve
the vertex, crown, or a diffuse pattern, women with more
severe hyperandrogenemia may experience bitemporal
hair loss and loss of the frontal hairline (31).

1296 PCOS Best Practices, Endocr Pract. 2015;21(No. 11)

2. What Are the Best Treatment Options for

Women With Hyperandrogenism?
Pharmacologic strategies to control the dermatologic
symptoms of hyperandrogenism are aimed both at lowering androgen levels and controlling the effect of androgens
at the tissue level. OCPs can effectively lower androgens
and block the effect of androgens via suppression of ovarian androgen production and by increasing sex hormone
binding globulin (SHBG). Moreover, physiologic doses of
dexamethasone or prednisone can directly lower adrenal
androgen output, and anti-androgens can be used to block
the effects of androgen in regulating the expression of
target genes in the pilosebaceous unit or in the hair follicle. The mechanistic basis of anti-androgen therapy is
competitive antagonism of the androgen receptor (spironolactone, cyproterone acetate, flutamide) or inhibition
of 5-reductase (5R) (finasteride) to prevent the conversion of T to its more potent form, 5-dihydrotestosterone
(DHT). The choice of anti-androgen therapy is guided by
which symptoms are most bothersome to the patient, as
well as whether the androgens are thought to be primarily
of ovarian or adrenal origin.

The mainstay of first-line therapy for all dermatologic
symptoms of hyperandrogenism is OCP therapy. OCPs contain estrogen (almost exclusively ethinyl estradiol) and a progestin. A daily dose of 20 to 35 g of ethinyl estradiol effectively suppresses pituitary-ovarian communication, thereby
decreasing ovarian androgen production. Newer progestins
have been developed with an emphasis on greater progestogenic and less androgenic effects. Although the ideal progestins to use in PCOS are those with the lowest androgenic
profile, such as chlormadinone and drospirenone, these may
induce a higher number of venous thrombosis events and
may be contraindicated in patients with severe obesity. The
androgenic symptoms that may be attenuated by OCP treatment include hirsutism and acne, primarily through the ability of OCPs to raise SHBG and lower free T levels. OCPs
pass through the liver, causing increased synthesis of SHBG,
and they may be more effective at controlling hirsutism and
acne than transdermal or vaginal ring preparations. OCPs as
monotherapy are not very effective in arresting mild to moderate alopecia or hirsutism and are preferably combined with
an anti-androgen to achieve a better response when targeting
hirsutism and alopecia.
Spironolactone (SPA) is the most commonly used
androgen blocker in the United States. It is relatively effective in the treatment of all dermatologic signs of PCOS, in
particular acne and hirsutism (32,33). In addition to being
an aldosterone antagonist, it competes with DHT for binding to the androgen receptor, although its effect on the
latter is minimal compared to DHT. SPA also has several
other effects, including a moderate local blocking of 5R
activity, competing with androgens for binding to SHBG,
blocking conversion of T to DHT in dermal papilla cells,
and antagonizing the androgenic effect of DHT on the hair

follicle. Its progestational activity may also reduce levels

of gonadotropin-releasing hormone and LH, thereby attenuating the LH effect on androgen steroidogenesis. The dosage of SPA is 100 to 200 mg daily, given in 2 divided doses.
Because SPA may induce hyperkalemia, serum potassium
should be monitored. Although headaches and dizziness
are relatively frequent, the patient should increase water
and salt intake in hot weather. Intermenstrual spotting may
occur in almost half of the women taking SPA as monotherapy, and breast discomfort, dry skin, and gastritis are
also noted in some women on this drug. As with other antiandrogens, SPA has the potential for teratogenicity (specifically, inadequate masculinization of male genitalia),
and its combination with a nonandrogenic OCP is recommended. Although flutamide and cyproterone acetate are
potent anti-androgens, they are not available in the United
States.
Finasteride is 5R inhibitor available in the United
States that has been shown to exert a 50 to 60% reduction in DHT levels. As such, it is an important therapeutic
agent for AA at a daily dosage of 5 mg (34), which has
been shown to effect a significant reduction in FG scores
in the majority of women after 6 months of treatment (35).
Treatment of hirsutism is targeted at reducing the production and bioavailability of T, as well as blocking target tissue androgen action. An assessment of the data indicates
that the use of 5R inhibition therapy should be considered when prior therapy with OCPs and SPA are relatively
ineffective for severe hirsutism (36). The effects of 5 mg
of finasteride for premenopausal women with AA are variably favorable, particularly when used with an ethinyl
estradiol/drospirenone OCP (37). Although the limited literature has made it difficult to fully assess dutasteride, it
appears to have a somewhat better success rate in the treatment of alopecia (38). Dutasteride reduces plasma DHT
more significantly than finasteride, with a concomitant
rise in plasma T of approximately 25% due to blockage of
the conversion of T to DHT by inhibition both 5R isoenzymes. Consequently, the aromatase pathway is enhanced,
resulting in an increase in circulating estrogens. Finally,
metformin is an alternative therapy for hirsutism in women
with PCOS who have other indications for metformin use.
Although it is useful for metabolic and glycemic abnormalities and for improving menstrual irregularities, metformin
is less effective than anti-androgens in treating hirsutism or
acne (36,39).
PCOS IN THE ADOLESCENT
1. What Makes the Diagnosis of PCOS Challenging in
Adolescents? Is It Important to Differentiate PCOS
at This Stage, or Is It Acceptable to Delay Diagnosis
Until Adulthood?
Because PCOS commonly presents during adolescence, it is important to diagnose the condition as early as

PCOS Best Practices, Endocr Pract. 2015;21(No. 11) 1297

possible to evaluate and treat metabolic and cardiovascular

(CV) risks, as well as the psychologic and dermatologic
issues. There are complicating factors in diagnosing PCOS
in adolescent girls, and there is a distinct potential for both
over- and underdiagnosis of the syndrome (40). On the one
hand, many of the cardinal features of PCOS, including
acne, menstrual irregularities, and hyperinsulinemia, are
common issues in normal puberty and accordingly are
difficult to distinguish from a true underlying disorder.
On the other hand, diagnostic features such as hirsutism
may not be fully realized in adolescent girls presenting
earlier in the continuum. Furthermore, clinical definitions of excessive hair growth and biochemical parameters for hyperandrogenism are based upon standards and
definitions derived from adults. As previously discussed,
PCOS as defined by the AES criteria requires evidence
of ovarian dysfunction (either based on clinical evidence
of oligo/anovulation or PCOM on ultrasound) and either
biochemical or clinical evidence of hyperandrogenism (1).
In addition to excluding alternative causes of these issues,
including thyroid, adrenal, and pituitary dysfunction, the
definition of each of the AES parameters in adolescents
presents specific obstacles. For example, menstrual irregularities with anovulatory cycles and varied cycle length
are common due to the immaturity of the hypothalamicpituitary-ovarian axis during the 2- to 3-year time period
postmenarche. Moreover, hyperandrogenemia leading to
acne and hirsutism could also be related to normal puberty
rather than underlying PCOS, and hyperinsulinemia is a
feature of normal puberty. PCOS may be a trigger when
obesity and concomitant insulin resistance act to exacerbate otherwise normal hormonal changes of puberty.

In terms of defining ovarian dysfunction in adolescents, a widely accepted approach is to follow the patient
clinically for 2 to 3 years postmenarche before beginning
a thorough investigation for underlying PCOS (41). This
is based on the knowledge that persistent oligomenorrhea
2 to 3 years beyond menarche predicts ongoing menstrual
irregularities and offers a greater likelihood of uncovering true underlying ovarian or adrenal dysfunction. Adult
ultrasound criteria for PCOM are not applicable to adolescent girls, in whom large, multicystic ovaries are a common finding due to the natural history of ovarian development at menarche. Parenthetically, there is a great deal
of heterogeneity in ovarian morphology in adolescents,
and there are therefore minimal normative data on which
to base ultrasound criteria for PCOM in this population.
For these reasons, ultrasound is not a first-line investigation in women <17 years of age, and ovarian dysfunction
in adolescents should be based on oligomenorrhea and/or
biochemical evidence of oligo/anovulation.

Defining clinical and biochemical hyperandrogenism
in adolescents has its own set of challenges. The cornerstone of defining clinical hyperandrogenism in women
is the presence of hirsutism, whereas acne and alopecia

indicate a possibility of underlying hyperandrogenism.

Acne has a high background prevalence in adolescents and
may not be related to underlying PCOS. On the other hand,
diffuse hirsutism and abnormal hair growth defined by an
elevated FG score may not be fully manifested in adolescence. Furthermore, the FG scoring system was determined in adult women and is not applicable to younger,
perimenarchal age groups. Due to these issues, biochemical evidence of hyperandrogenism is extremely important
to evaluate in this group. As we have previously alluded,
in adult women, there are major limitations to the sensitivity of T assays in ranges applicable to young girls, with
no established normal ranges. The lower limit of normal
T in young girls may be lower than in adult women. A
recent Australian study defined the upper 5 and 10% levels
of free T in an unselected population of girls at 1.3 and
1.0 ng/dL, respectively (42), with a median total T of
35 ng/dL. A study from the United Kingdom showed
that total T and SHBG levels progressively increase and
decrease, respectively, throughout puberty, resulting in
higher free T. By Tanner stage 5, median T is reportedly
40.3 ng/dL, and SHBG is reportedly 43 nmol/L (40). Rising
free T levels in puberty are in part secondary to a decrease
in SHBG from physiologic hyperinsulinemia.
Because the signs and symptoms are heterogeneous
in adolescents and may vary over time, the diagnosis of
PCOS may be overlooked. However, adolescence is a crucial time for diagnosis because this is a time frame when
many patients with PCOS start gaining weight. Adolescents
should therefore be followed carefully to confirm a diagnosis of PCOS and reduce the frequency of later complications in the CV system and type 2 diabetes mellitus.
2. What Is the Role of Metformin in Adolescents with
PCOS, and What Is the Optimal Timing and Dose?

Metformin is commonly used in young girls and adolescents with PCOS as first-line monotherapy or in combination with OCPs and anti-androgen medications (43).
Metformin is currently used to target hyperandrogenemia
and symptoms of androgen excess, to restore normal menses, to aid in weight reduction, and to intervene in metabolic parameters of insulin resistance, with the goal of
preventing long-term metabolic and CV complications. In
addition, a body of evidence supports the use of metformin
to prevent or delay the progression to PCOS in high-risk
prepubertal girls.

As in adult women, evidence supports the use of metformin in obese/overweight and lean adolescent girls to
help reduce androgen excess and improve ovarian function
(44). Many adolescents who do not want to be on an OCP
may derive symptom benefit from metformin monotherapy.
In a small randomized, placebo-controlled study of obese
adolescents and baseline hyperinsulinemia, metformin
significantly improved biochemical hyperandrogenemia,
restored menses, and improved high-density lipoprotein

1298 PCOS Best Practices, Endocr Pract. 2015;21(No. 11)

(HDL) cholesterol, although there were no improvements

in insulin sensitivity measurements or weight loss (45).
The prevalence of metabolic syndrome (MetS) in
adolescents with PCOS appears to be very high. A U.S.
study using AES criteria to define PCOS reported a MetS
prevalence of 10.8%, compared with 1.7% in those without
PCOS (46). The study also reported a higher proportion of
metabolic abnormalities in subjects with PCOS compared
with adolescents without PCOS, even after excluding body
mass index (BMI). Insulin resistance and the components
of MetS are therefore important targets of therapy in adolescent girls with PCOS. A recent retrospective study compared the effects of metformin monotherapy and metformin with OCPs on lipids in overweight/obese adolescents
with PCOS (47). In the monotherapy group, there was a
significant decrease in BMI and improvement in total and
low-density lipoprotein (LDL) cholesterol over the 10
months of treatment, with no significant changes in triglycerides (TGs). In the group taking metformin with an OCP,
total and LDL cholesterol were not significantly altered,
whereas there was a significant increase in HDL cholesterol and a trend toward an increase in TGs. Lean girls
with PCOS may also benefit from metformin monotherapy. For example, a small study of 10 normal-weight girls
with a history of precocious puberty and PCOS showed
that metformin treatment decreased biochemical and clinical hyperandrogenism, improved cyclic menses, decreased
hyperinsulinemia, and improved lipid profiles (48).

In lean adolescent girls, a dose as low as 850 mg daily
may be effective at reducing PCOS symptoms, although in
overweight and obese adolescents, dose escalation to 1.5
to 2.5 g daily is likely required (44). In terms of timing,
certain studies have suggested that early intervention in
high-risk prepubertal and early menarchal girls can prevent
the metabolic and clinical sequelae of PCOS. It should be
pointed out, however, that these results were obtained in
a well-defined patient population of girls with low birth
weight and history of premature pubarche (49,50).
3. Are Anti-Androgen Therapies Appropriate in
Adolescents With PCOS? Are There Additional
Concerns Regarding Bone and CV Health
Associated with the Use of These Medications in
Young Girls?

SPA and finasteride are often added to OCPs in adolescents with PCOS to more aggressively address the androgen excess symptoms of acne, hirsutism, and alopecia. A
major risk of both of these medications is teratogenesis,
and ongoing OCP use must be emphasized. Although the
safety and efficacy of these medications has been studied
in adult women, this has not been the case in adolescents,
and there is a paucity of literature addressing their use in
this population. Other anti-androgens, such as cyproterone
acetate and flutamide, have been studied in the adolescent
population with good efficacy and side effect profiles (51).

A specific concern of anti-androgen therapy in adolescents

is the effect on bone mass. A recent retrospective study
of adolescents treated with metformin or metformin plus
an anti-androgenic OCP and flutamide for at least 1 year
found no differences between the groups with respect to
BMI, abdominal fat composition, and insulin sensitivity.
Similarly, there were no differences in bone density and
bone geometry parameters measured using peripheral
quantitative computed tomography (52). A recent small,
prospective study of younger (mean age, 22 years), lean
women with PCOS treated with an OCP containing ethinyl
estradiol/drospirenone and SPA found a statistically significant increase in C-reactive protein and homocysteine
levels after 6 months of this treatment, although there were
no changes in insulin parameters or lipid profile (53).
DISCLOSURE
Dr. Neil F. Goodman is on the Speakers Bureau for
Abbie. The other authors have no multiplicity of interest to
disclose.
REFERENCES
1. Azziz R, Carmina E, Dewailly D, et al. The Androgen
Excess and PCOS Society criteria for the polycystic ovary
syndrome: the complete task force report. Fertil Steril.
2009;91 456-488.
2. Rosner W. Errors in the measurement of plasma free testosterone. J Clin Endocrinol Metab. 1997;82:2014-2015.
3. Rosner W. An extraordinarily inaccurate assay for free testosterone is still with us. J Clin Endocrinol Metab. 2001;86:
2903.
4. Vermeulen A, Verdonck L, Kaufman JM. A critical
evaluation of simple methods for the estimation of free
testosterone in serum. J Clin Endocrinol Metab. 1999;84:
3666-3672.
5. Carmina E. Ovarian and adrenal hyperandrogenism. Ann
N Y Acad Sci. 2006;1092:130-137.
6. Carmina E, Lobo RA. Prevalence and metabolic characteristics of adrenal androgen excess in hyperandrogenic
women with different phenotypes. J Endocrinol Invest.
2007;30:111-116.
7. Stanczyk FZ, Chang L, Carmina E, Putz Z, Lobo RA. Is
11 beta-hydroxyandrostenedione a better marker of adrenal
androgen excess than dehydroepiandrosterone sulfate? Am
J Obstet Gynecol. 1991;165(6 Pt 1):1837-1842.
8. Vesper HW, Bhasin S, Wang C, et al. Interlaboratory
comparison study of serum total testosterone [corrected]
measurements performed by mass spectrometry methods.
Steroids. 2009;74:498-503.
9. Janse F, Eijkemans MJ, Goverde AJ, et al. Assessment
of androgen concentration in women: liquid chromatography-tandem mass spectrometry and extraction RIA show
comparable results. Eur J Endocrinol. 2011;165:925-933.
10. Rosner W, Auchus RJ, Azziz R, Sluss PM, Raff H.
Position statement: utility, limitations, and pitfalls in measuring testosterone: an Endocrine Society position statement. J Clin Endocrinol Metab. 2007;92:405-413.
11. Carmina E, Oberfield SE, Lobo RA. The diagnosis of
polycystic ovary syndrome in adolescents. Am J Obstet
Gynecol. 2010;203:201.e1-e5.

PCOS Best Practices, Endocr Pract. 2015;21(No. 11) 1299

12. Azziz R, Waggoner WT, Ochoa T, Knochenhauer ES,
Boots LR. Idiopathic hirsutism: an uncommon cause of
hirsutism in Alabama. Fertil Steril. 1998;70:274-278.
13. Carmina E, Lobo RA. Do hyperandrogenic women with
normal menses have polycystic ovary syndrome? Fertil
Steril. 1999;71:319-322.
14. Hull MG, Savage PE, Bromham DR, Ismail AA, Morris
AF. The value of a single serum progesterone measurement
in the midluteal phase as a criterion of a potentially fertile
cycle (ovulation) derived from treated and untreated conception cycles. Fertil Steril. 1982;37:355-360.
15. Abraham GE, Maroulis GB, Marshall JR. Evaluation of
ovulation and corpus luteum function using measurements
of plasma progesterone. Obstet Gynecol. 1974;44:522-525.
16. Dumesic DA, Lobo RA. Cancer risk and PCOS. Steroids.
2013;78:782-785.
17. Fauser BC, Tarlatzis BC, Rebar RW, et al. Consensus
on womens health aspects of polycystic ovary syndrome
(PCOS): the Amsterdam ESHRE/ASRM-sponsored 3rd
PCOS consensus workshop group. Fertil Steril. 2012;97:
28-38.e25.
18. Veltman-Verhulst SM, Boivin J, Eijkemans MJ, Fauser
BJ. Emotional distress is a common risk in women with
polycystic ovary syndrome: a systematic review and metaanalysis of 28 studies. Hum Reprod Update. 2012;18:
638-651.
19. Dewailly D, Lujan ME, Carmina E, et al. Definition and
significance of polycystic ovarian morphology: a task force
report from the Androgen Excess and Polycystic Ovary
Syndrome Society. Hum Reprod Update. 2014;20:334-352.
20. Balen AH, Laven JS, Tan SL, Dewailly D. Ultrasound
assessment of the polycystic ovary: international consensus
definitions. Hum Reprod Update. 2003;9:505-514.
21. Duijkers IJ, Klipping C. Polycystic ovaries, as defined by
the 2003 Rotterdam consensus criteria, are found to be very
common in young healthy women. Gynecol Endocrinol.
2010;26:152-160.
22. Johnstone EB, Rosen MP, Neril R, et al. The polycystic
ovary post-rotterdam: a common, age-dependent finding in
ovulatory women without metabolic significance. J Clin
Endocrinol Metab. 2010;95:4965-4972.
23. Chen Y, Li L, Chen X, et al. Ovarian volume and follicle number in the diagnosis of polycystic ovary syndrome in Chinese women. Ultrasound Obstet Gynecol.
2008;32:700-703.
24. Carmina E. Hirsutism: investigation and management.
Expert Rev Endocrinol Metab. 2010;5:189-195.
25. Dewailly D, Gronier H, Poncelet E, et al. Diagnosis of
polycystic ovary syndrome (PCOS): revisiting the threshold values of follicle count on ultrasound and of the serum
AMH level for the definition of polycystic ovaries. Hum
Reprod. 2011;26:3123-3129.
26. Hatch R, Rosenfield RL, Kim MH, Tredway D.
Hirsutism: implications, etiology, and management. Am J
Obstet Gynecol. 1981;140:815-830.
27. Rosenfield RL, Deplewski D. Role of androgens in the
developmental biology of the pilosebaceous unit. Am J
Med. 1995;98:80S-88S.
28. Borgia F, Cannav S, Guarneri F, Cannav SP, Vaccaro
M, Guarneri B. Correlation between endocrinological
parameters and acne severity in adult women. Acta Derm
Venereol. 2004;84:201-204.
29. Timpatanapong P, Rojanasakul A. Hormonal profiles
and prevalence of polycystic ovary syndrome in women
with acne. J Dermatol. 1997;24:223-229.

30. Lucky AW, McGuire J, Rosenfield RL, Lucky PA, Rich

BH. Plasma androgens in women with acne vulgaris. J
Invest Dermatol. 1983;81:70-74.
31. Goodman NF, Bledsoe MB, Cobin RH, et al. American
Association of Clinical Endocrinologists medical guidelines
for the clinical practice for the diagnosis and treatment of
hyperandrogenic disorders. Endocr Pract. 2001;7:120-134.
32. Brown J, Farquhar C, Lee O, Toomath R, Jepson RG.
Spironolactone versus placebo or in combination with steroids for hirsutism and/or acne. Cochrane Database Syst
Rev. 2009;2:CD000194.
33. Yemisci A, Gorgulu A, Piskin S. Effects and side-effects
of spironolactone therapy in women with acne. J Eur Acad
Dermatol Venereol. 2005;19:163-166.
34. Yeon JH, Jung JY, Choi JW, et al. 5 mg/day finasteride
treatment for normoandrogenic Asian women with female
pattern hair loss. J Eur Acad Dermatol Venereol. 2011;25:
211-214.
35. Lakryc EM, Motta EL, Soares JM Jr, Haidar MA, de
Lima GR, Baracat EC. The benefits of finasteride for hirsute women with polycystic ovary syndrome or idiopathic
hirsutism. Gynecol Endocrinol. 2003;17:57-63.
36. Paparodis R, Dunaif A. The Hirsute woman: challenges
in evaluation and management. Endocr Pract. 2011;17:
807-818.
37. Iorizzo M, Vincenzi C, Voudouris S, Piraccini BM, Tosti
A. Finasteride treatment of female pattern hair loss. Arch
Dermatol. 2006;142:298-302.
38. Olszewska M, Rudnicka L. Effective treatment of female
androgenic alopecia with dutasteride. J Drugs Dermatol.
2005;4:637-640.
39. Legro RS, Arslanian SA, Ehrmann DA, et al. Diagnosis
and treatment of polycystic ovary syndrome: an Endocrine
Society clinical practice guideline. J Clin Endocrinol
Metab. 2013;98:4565-4592.
40. Williams RM, Ong KK, Dunger DB. Polycystic ovarian syndrome during puberty and adolescence. Mol Cell
Endocrinol. 2013;373:61-67.
41. Hardy TS, Norman RJ. Diagnosis of adolescent polycystic ovary syndrome. Steroids. 2013;78:751-754.
42. Hickey M, Doherty DA, Atkinson H, et al. Clinical,
ultrasound and biochemical features of polycystic ovary
syndrome in adolescents: implications for diagnosis. Hum
Reprod. 2011;26:1469-1477.
43. Hsia Y, Dawoud D, Sutcliffe AG, Viner RM, Kinra S,
Wong IC. Unlicensed use of metformin in children and adolescents in the UK. Br J Clin Pharmacol. 2012;73:135-139.
44. Geller DH, Pacaud D, Gordon CM, Misra M. State of
the art review: emerging therapies: the use of insulin sensitizers in the treatment of adolescents with polycystic ovary
syndrome (PCOS). Int J Pediatr Endocrinol. 2011;2011:9.
45. Bridger T, MacDonald S, Baltzer F, Rodd C. Randomized
placebo-controlled trial of metformin for adolescents with
polycystic ovary syndrome. Arch Pediatr Adolesc Med.
2006;160:241-246.
46. Roe AH, Prochaska E, Smith M, Sammel M, Dokras A.
Using the androgen excess-PCOS society criteria to diagnose polycystic ovary syndrome and the risk of metabolic
syndrome in adolescents. J Pediatr. 2013;162:937-941.
47. Bredella MA, McManus S, Misra M. Impact of metformin monotherapy versus metformin with oestrogen-progesterone on lipids in adolescent girls with polycystic ovarian syndrome. Clin Endocrinol (Oxf). 2013;79:199-203.
48. Ibez L, Dimartino-Nardi J, Potau N, Saenger P.
Premature adrenarche--normal variant or forerunner of
adult disease? Endocr Rev. 2000;21:671-696.

1300 PCOS Best Practices, Endocr Pract. 2015;21(No. 11)

49. Ibez L, Ferrer A, Ong K, Amin R, Dunger D, de
Zegher F. Insulin sensitization early after menarche prevents progression from precocious pubarche to polycystic
ovary syndrome. J Pediatr. 2004;144:23-29.
50. Ibez L, Lpez-Bermejo A, Daz M, Marcos MV, de
Zegher F. Early metformin therapy (age 8-12 years) in
girls with precocious pubarche to reduce hirsutism, androgen excess, and oligomenorrhea in adolescence. J Clin
Endocrinol Metab. 2011;96:E1262-E1267.
51. Ibez L, Jaramillo A, Ferrer A, de Zegher F. Absence
of hepatotoxicity after long-term, low-dose flutamide in

hyperandrogenic girls and young women. Hum Reprod.

2005;20:1833-1836.
52. Bechtold S, Dalla Pozza R, Putzker S, et al. Effect of
antiandrogen treatment on bone density and bone geometry
in adolescents with polycystic ovary syndrome. J Pediatr
Adolesc Gynecol. 2012;25:175-180.
53. Harmanci A, Cinar N, Bayraktar M, Yildiz BO. Oral
contraceptive plus antiandrogen therapy and cardiometabolic risk in polycystic ovary syndrome. Clin Endocrinol
(Oxf). 2013;78:120-125.