ORIGINAL ARTICLES: REPRODUCTIVE ENDOCRINOLOGY

Utility of ultrasound in the diagnosis

of polycystic ovary syndrome
in adolescents
Michal Youngster, M.D.,a Valerie L. Ward, M.D., M.P.H.,c Emily A. Blood, Ph.D.,a,d Carol E. Barnewolt, M.D.,c
S. Jean Emans, M.D.,a and Amy D. Divasta, M.D., M.M.Sc.a,b
a
Division of Adolescent/Young Adult Medicine, Department of Medicine, b Division of Gynecology, Department of Surgery,
c
Department of Radiology, and d Department of Pediatrics, Boston Children's Hospital, Harvard Medical School, Boston,
Massachusetts

Objective: To determine the utility of transabdominal pelvic ultrasound in the diagnosis of polycystic ovary syndrome (PCOS) during
adolescence.
Design: Retrospective case-control study.
Setting: Academic tertiary care pediatric hospital.
Patient(s): A case group of 54 patients (mean age, 15.2 years) with PCOS based on the National Institutes of Health criteria and a com-
parison group of 98 patients (mean age, 14.6 years) with acute appendicitis.
Intervention(s): Transabdominal ultrasound (TAUS) images were evaluated in the two groups of adolescents, with data collected on
quality of the images, ovarian volume, ovarian follicle count, and endometrial thickness.
Main Outcome Measure(s): Sonographic modied Rotterdam criteria (volume >10 mL and/or follicle number per section R10) for
polycystic ovaries (PCO).
Result(s): Among the 54 patients with PCOS and 98 comparison subjects with usable images, the sonographic modied Rotterdam
criteria for PCO morphology (PCOM) were met more frequently in the PCOS group than in the comparison group (65% vs. 11%). The
vast majority of images were of adequate quality for diagnosis (PCOS 94% and comparison 91%), even in the presence of obesity.
Conclusion(s): The prevalence of ovarian morphology meeting the sonographic modied Rotterdam criteria by TAUS in girls with
PCOS was markedly higher than in the adolescents serving as a comparison group. PCOM ndings by the sonographic modied
Rotterdam criteria were uncommon in the nongynecologic comparison group, in contrast to previous reports. TAUS may provide useful
information in the evaluation of PCOS during adolescence, even in obese adolescents. (Fertil
Steril 2014;102:14328. 2014 by American Society for Reproductive Medicine.) Use your smartphone
Key Words: Polycystic ovary syndrome, adolescents, transabdominal ultrasound, Rotterdam to scan this QR code
criteria, endometrial thickness and connect to the
discussion forum for
this article now.*
Discuss: You can discuss this article with its authors and with other ASRM members at http://
fertstertforum.com/youngsterm-ultrasonography-diagnosis-pcos-adolescents/ * Download a free QR code scanner by searching for QR
scanner in your smartphones app store or app marketplace.

P
olycystic ovary syndrome on pelvic ultrasound (US). Associated treatments to prevent at least part of
(PCOS) is the most common conditions include obesity, diabetes, the late sequelae (4).
endocrine disorder, affecting cardiovascular disease, infertility, and Several criteria have been proposed
6%15% of women of reproductive increased risk of endometrial hyperpla- in adults to make a diagnosis of PCOS
age (1). Manifestations of PCOS include sia and cancer (2, 3). Early diagnosis is with the concomitant exclusion of
irregular menses, hyperandrogenism, important, as awareness can promote other disorders. The 1990 National In-
and the presence of polycystic ovaries lifestyle modications or medical stitutes of Health (NIH) criteria require
menstrual irregularities and clinical or
Received February 8, 2014; revised July 21, 2014; accepted July 22, 2014; published online September biochemical hyperandrogenism. The
16, 2014. Rotterdam criteria (RC), the product of
M.Y. has nothing to disclose. V.L.W. has nothing to disclose. E.A.B. has nothing to disclose. C.E.B. has
nothing to disclose. S.J.E. has nothing to disclose. A.D.D. has nothing to disclose.
a consensus workshop held in 2003 by
This study was supported by the National Institutes of Health grant nos. R01HD066963, R01DA033974, the European Society for Human Repro-
and R34DA030353-01A1 (to E.A.B., not related to this work). duction and Embryology (ESHRE) and
Reprint requests: Michal Youngster, M.D., Division of Adolescents/Young Adults, 300 Longwood
Avenue, Boston, Massachusetts 02115 (E-mail: michal.youngster@childrens.harvard.edu). the American Society for Reproductive
Medicine (ASRM), broadened the de-
Fertility and Sterility Vol. 102, No. 5, November 2014 0015-0282/$36.00
Copyright 2014 American Society for Reproductive Medicine, Published by Elsevier Inc.
nition to include two out of three of
http://dx.doi.org/10.1016/j.fertnstert.2014.07.1241 the following criteria: oligomenorrhea

1432 VOL. 102 NO. 5 / NOVEMBER 2014


and/or anovulation, hyperandrogenism, and polycystic
ovaries (PCO) by transvaginal US (5) (either ovarian volume
>10 mL and/or R12 follicles measuring 29 mm) (6). The
Androgen Excess-PCOS Society criteria require the presence
of hyperandrogenism (clinical and/or biochemical) and
ovarian dysfunction (oligoanovulation and/or PCO) (7).
The application of these diagnostic criteria for PCOS in
the adolescent girl is more challenging, given the expected
anovulatory menses and hormone changes characteristic of
the early postmenarchal years. Menstrual and hormonal uc-
tuations that occur normally during adolescence may mimic
the clinical characteristics included in the adult PCOS diag-
nostic criteria, making them less reliable as a diagnostic
tool in this age group. After menarche, anovulatory cycles
are common, may persist for several years (1, 8), and do not
necessarily correspond to clinical or biochemical
hyperandrogenism (9). Clinical hyperandrogenism is dened
primarily by the degree of hirsutism in adolescents because
acne is common in this age group and is usually transient
(10, 11). Although some studies suggest relying primarily on
biochemical markers when making the diagnosis of PCOS
(12, 13), measurements of serum androgens can be
challenging to interpret owing to the variability among
laboratories and the use of different methods of determining
free and total T (14).
The usefulness of pelvic US in supporting the suspected
diagnosis of PCOS has been controversial. The nding of PCO
on US has been reported to be nonspecic, overlapping with
image ndings seen in up to 40% of the normal population
(9, 1517). In addition, most pelvic US exams in adolescents
are performed transabdominally rather than transvaginally,
resulting in lower resolution and thus less accurate
observation of ovarian morphology (6). The prevalence of
obesity in the PCOS population may potentially make the
transabdominal US (TAUS) even less ideal to dene ovarian
morphology. Furthermore, the multifollicular appearance
characteristically seen during puberty as a result of follicular
growth without consistent recruitment of a dominant follicle
may also be a source of confusion and misdiagnosis of PCO
in this age group (6). As a result, some investigators have
proposed replacing ovarian sonographic imaging with
ovarian magnetic resonance imaging or serum antim ullerian
hormone (AMH) levels, especially in the adolescent
population (1821). The most recent Endocrine Society
guidelines cautioned against the use of PCO morphology
(PCOM) as a diagnostic criteria for adolescents (22).
Given the controversies and questions regarding the use
of US in an adolescent PCOS population, we analyzed pelvic
US images obtained in a cohort of adolescents with PCOS and
a comparison adolescent population to assess the prevalence
of PCOM by a modied version of the Rotterdam US criteria
and to examine whether there were differences in endometrial
thickness.
MATERIALS AND METHODS
We performed a retrospective, case-control cohort study. The
protocol was approved by the Boston Children's Hospital
Committee on Clinical Investigation.
VOL. 102 NO. 5 / NOVEMBER 2014
Fertility and Sterility
Study Populations
US images were evaluated in two cohorts: [1] a sample of
adolescent girls who all were diagnosed with PCOS by the
1990 NIH criteria (menstrual irregularities and clinical and/
or biochemical hyperandrogenism) between 2006 and 2008
at the Boston Children's Hospital's Reproductive Endocrine
Clinic in the Division of Adolescent/Young Adult Medicine;
and [2] a comparison group of adolescents with surgically
conrmed appendicitis who had a pelvic TAUS in the Emer-
gency Department (ED) (20052013) before the operative pro-
cedure. Patients with PCOS taking hormonal medications
(such as birth control pills) that could affect ovarian
morphology were excluded. Hormonal and physical charac-
teristics of PCOS patients were assessed (Table 1). Hirsutism
was considered clinically signicant if the Ferriman-Gallwey
score was >7. None of the girls had acne as the only sign of hy-
perandrogenism. The biochemical evaluation included serum
concentrations of total T (high-performance liquid chroma-
tography tandem mass spectrometry, Esoterix, Inc.; the upper
limit of the normal values ranged between 32 and 55 ng/dL,
depending on Tanner stage) and free T (equilibrium dialysis,
Esoterix, Inc.; normal <6.3 pg/mL). The comparison group
was identied using an institutional database. Medical charts
were reviewed including the initial ED visit and subsequent
clinic visits in our hospital. In total, 398 patients were initially
identied. Patients were excluded if they were premenarchal
or if they had a past medical history suggesting a diagnosis
of PCOS or potentially affecting ovarian morphology (e.g., en-
docrinopathies, use of hormonal medications, known adnexal
pathology; Fig. 1). The data collected included weight and last
menstrual period (LMP) before the US.
US Data
All examinations were performed on one of three units, with
multiple, size-appropriate, 25 MHz transducers: Siemens,
TABLE 1
Menstrual and hormonal characteristics of girls with PCOS (n [ 54).
n (%a) Mean valueb (SD)
Age at menarche (y) 12.0 (1.6)
Gynecologic age 3.5 (2)
Oligoanovulation
Primary amenorrhea 1 (2)
Secondary amenorrhea 7 (12)
Oligomenorrhea 45 (84)
Polymenorrhea 1 (2)
Hirsutism (n 54) 46 (85)
Acne (n 54) 37 (69)
Total T (ng/dL) (n 49) 46.7 (25.3)
Elevated total T (n 53) 17 (32)
Free T (pg/mL) (n 45) 9.6 (8.2)
Elevated free T (n 47) 17 (36)
DHEAS (mg/dL) (n 45) 296.2 (138)
Elevated DHEAS (n 48) 19 (40)
a
Percentages are from those patients with complete data on a given variable.
b
Mean is calculated for patients with laboratory values obtained at our institution. Patients
whose labs were performed before the initial visit (at facilities that used different laboratory
methods with a variable range and cutoff) were included in calculations of the frequency of
elevated androgen concentrations.
Youngster. US evaluation of PCO morphology in adolescents. Fertil Steril 2014.
1433
ORIGINAL ARTICLE: REPRODUCTIVE ENDOCRINOLOGY
FIGURE 1
Study participant eligibility and selection.
Youngster. US evaluation of PCO morphology in adolescents. Fertil Steril 2014.
Sequoia; Philips, iU22; or GE, Logiq E9. All studies for both
groups were performed in the ultrasound lab in the Depart-
ment of Radiology. Studies were initially performed by
trained pediatric sonographers, followed by review and re-
scanning, if necessary, by a pediatric radiologist. Our protocol
for pelvic sonography in girls with suspected appendicitis
included longitudinal and transverse views of the two ovaries
and uterus (including measurement of endometrial stripe), all
obtained transabdominally through a full urinary bladder.
For the purpose of this study, a pediatric radiologist and a
gynecologist independently reviewed the still TAUS images of
the PCOS patients and comparison group; the US was
excluded if there was a dominant follicle >10 mm. Ovarian
volume (OV) was calculated using the traditional method of
a simplied formula for a prolate ellipse (0.5  length  width
 thickness) (6). The US images were evaluated for the num-
ber of antral follicles (29 mm) in a single cross-sectional
plane (follicle number per section [FNPS]) for each ovary.
The RC use an OV cutoff of >10 mL and a follicle number
per ovary (FNPO) cutoff of R12 follicles as seen in cineloops
or real time US images. Since we reviewed still images ob-
tained by TAUS, we dened PCOM for the purposes of this
1434
study using a modied Rotterdam US criteria (mRC), with
an OV cutoff of >10 mL and FNPS threshold of 10 follicles,
as previously described (2325). Endometrial thickness was
also assessed. In the comparison group, patients were
stratied into the following subgroups: US occurring during
the rst 2 weeks of the cycle (follicular phase) or US
occurring during weeks 3 and 4 since LMP (luteal phase).
The usability of the US for ovarian imaging was evaluated;
the US was excluded if it was technically nonreadable for
either ovary. The two clinicians initially reviewed 30 images
to ensure similar techniques for measurement and
calculations.
Data Analysis
Descriptive statistics were used to characterize the partici-
pant data. Tests of associations between variables within
the PCOS group and comparison group were followed by
comparisons between two groups. For participant-level
data (age, weight, maximum OV, mRC PCOM, and US us-
ability), tests of association between continuous variables
were carried out via correlation coefcients, associations
VOL. 102 NO. 5 / NOVEMBER 2014
 Fertility and Sterility

between continuous and binary variables were carried out
via nonparametric Wilcoxon rank sum tests, and associa-
tions between two categorical variables were carried out
via Fisher's exact test. Since data from participants were
collected at the ovary level (OV for both right and left ovary)
and sagittal versus transverse view level (number of follicles
for both views for both right and left ovary) in addition to
the participant level, statistical procedures that account for
the correlation within person for ovary-level variables and
correlation within person and within ovary for view-level
variables were used. The associations between age and
weight and number of follicles were performed via nonlinear
and linear mixed-effects models that included a random in-
dividual and ovary random effects. For the PCOS group, as-
sociations between age and weight and the dichotomous
variable number of follicles R10 were performed and
nonlinear models were used. For the comparison group (in
which few participants had 10 or more follicles), the associ-
ation between age and weight and continuous number of
follicles was performed and linear models were used. The
ovary-level association between OV and age was tested
via generalized estimating equations. These models account
for the correlation of volume measurements between the
right and left ovary within a participant. All analyses were
carried out with SAS version 9.3; P< .05 was considered sta-
tistically signicant at the two-sided level.
RESULTS
Ovarian Data
Among 70 girls with PCOS initially identied by the NIH
criteria, two US exams were not obtainable. Fourteen girls
were excluded from the analyses owing to nondiagnostic
quality of the imaging or follicular size >10 mm.
Among the 54 remaining cases (mean age, 15.2 
1.8 years; mean weight, 79.0  19.8 kg), 35 met the sono-
graphic mRC for PCOM on TAUS (65%); 14/35 (40%) by
number of follicles R10 alone, 5/35 (14%) by volume alone,
and 16/35 (46%) by both number and volume (Table 2). Of the
21 cases who met OV criteria, 13 (62%) had enlargement of
both ovaries and eight (38%) had unilateral enlargement.
There were no associations between age and maximal OV
(P .15) or meeting the mRC (P .29). Increased age was not
associated with a reduced likelihood of having R10 follicles
(P .08). There was no relationship between weight and
meeting the sonographic criteria (P .46). Weight was not
associated with the likelihood of having R10 follicles
(P .11) or with mean OV (P .17).
The comparison group included 158 patients who met the
criteria for inclusion. After review of the images, 60 patients
were excluded from further analysis owing to inadequate
ovarian imaging or ovarian follicle greater than 10 mm. Of
the 98 girls in the comparison group (mean age, 14.7 
1.7 years; mean weight, 56.6  11.7 kg) included in the nal
analysis, 11 met the criteria for PCOM (11%): four by follicle
criteria alone, two by volume alone, and ve by both volume
and number of follicles. The increased OV was unilateral in all
instances. The girls who met the sonographic mRC for PCOM
tended to be older (P .04). Older girls had larger maximal OV
(P .03) and mean OV (P .01). Age was not signicantly
associated with an increased number of follicles (P .10).
Increased weight was associated with increased OV (P .02)
but not with increased follicle number (P .50) or with
meeting mRC for PCOM (P .25).
The vast majority of images were of adequate quality for
diagnosis (PCOS 94% and comparison 91%; P .45).
Among the 54 PCOS patients and 98 comparisons with usable
images, the sonographic mRC for PCOM was signicantly

TABLE 2

Physical and US characteristics of girls with PCOS (n [ 54) and comparison group (n [ 98).
PCOS Comparison group P value
Age (y) 15.2  1.8 (range, 1220) 14.7  1.7 (range, 1118) .09
Weight (kg) 79.0  19.8 (range, 44122) 56.6  11.7 (range, 3194) < .0001
Race/ethnicity (%) .57
Non-Hispanic white 31 (57) 55 (56)
Non-Hispanic black 4 (7) 8 (8)
Hispanic 6 (11) 19 (19)
Asian 4 (7) 5 (5)
Other 1 (2) 6 (6)
No information 8 (15) 5 (5)
Meeting mRC for PCOM (%) 35 (65) 11 (11) < .0001
Volume only (% of PCOM) 5/35 (14) 2/11 (18)
No. of follicles only (% of PCOM) 14/35 (40) 4/11 (36)
Both volume and follicle number (% of PCOM) 16/35 (46) 5/11 (45)
Total meeting mRC by volume (% of PCOM) 21/35 (60) 7/11 (64)
Total meeting mRC by no. of follicles (% of PCOM) 30/35 (86) 9/11 (82)
Both ovaries > 10 mL (% of PCOM) 13/35 (37) 0
Maximal OV (patient's largest ovary, mL) 9.5  3.9 (range, 2.719.7) 6.9  2.3 (range, 1.613.0) < .0001
Mean OV (mL) 8.2  3.9 (range, 1.619.7) 5.96  2.4 (range, 1.213.0) < .0001
(n 110 ovaries) (n 196 ovaries)
No. of images with a follicle > 10 mm (%) 10/68 (15) 46/158 (29) .02
Note: Data are presented as mean  SD (continuous variables) or n (%) (categorical variables). Ethnicity was self-reported by the patients at the time of clinic registration. MRC: OV >10 mL and/or
FNPS R10 in either ovary.
Youngster. US evaluation of PCO morphology in adolescents. Fertil Steril 2014.

VOL. 102 NO. 5 / NOVEMBER 2014 1435

ORIGINAL ARTICLE: REPRODUCTIVE ENDOCRINOLOGY
more prevalent in the PCOS group than in the comparison
group (65% vs. 11%; P< .0001). Similarly, maximal OV was
greater in the PCOS group (P< .0001). Mean weight in the
PCOS group was higher than in the comparison cohort
(79.0 kg vs. 56.6 kg; P< .0001), but there was no signicant
association between weight and US usability in either group
(P .52 for the PCOS group and .66 for the comparison group).
Endometrial Thickness
Endometrial thickness was greater in the comparison group
than in the PCOS group (P .005; Table 3). In the comparison
group, 80 patients had documentation of their LMP; these pa-
tients had their US further stratied as occurring during the
follicular phase or luteal phase. Girls in the follicular phase
had a thinner endometrium compared with girls in the
PCOS group (P .05), while girls in the luteal phase had a
thicker endometrium than girls in the PCOS group
(P< .0001). No association was found between endometrial
thickness and PCOM by US criteria in either group (P .99
and P .12 for the PCOS and comparison group, respectively).
DISCUSSION
In the current study, we found that most adolescents with
PCOS diagnosed by the 1990 NIH criteria did meet the sono-
graphic mRC for PCOM using TAUS. In our sample, the com-
parison cohort had a lower prevalence of PCOM than in
previous studies (15, 17). The incidence of PCOM was
signicantly higher in adolescents with PCOS than in the
comparison group. Although the evaluation of ovarian
morphology continues to be part of the diagnostic criteria
for adults, the utility of TAUS has been widely debated for
adolescents. The results of our study suggest that TAUS may
be a useful diagnostic tool and that OV >10 mL and follicle
number R10 using FNPS can be determined using TA
imaging even in overweight patients. The measurement of
endometrial thickness did not yield additional diagnostic
information.
Recent guidelines for the evaluation of PCOS in adoles-
cents have questioned the usefulness of TAUS. The recently
published Australian evidence-based guidelines for assess-
ment and management of PCOS recommended against using
US as a rst-line diagnostic tool in the adolescent patient.
This recommendation was based on expert opinion, as the in-
vestigators found insufcient data generated from controlled
studies to support its use (26). Another recent publication
TABLE 3
Endometrial thickness in the PCOS group and comparison group.
Endometrial
thickness (mm) PCOS (n [ 53) (%)
<5 28 (53)
510 17 (32)
>10 8 (15)
Note: One patient in the PCOS group and four patients in the comparison group were missing data on endometrial thickness. Fourteen patients in the comparison group did not have the date of the
LMP documented.
Youngster. US evaluation of PCO morphology in adolescents. Fertil Steril 2014.
1436
from the Endocrine Society similarly suggested that while
basing the diagnosis of PCOS in adult women on the RC is rec-
ommended, the workup of the adolescent should be ap-
proached differently and the diagnosis should be based on
hyperandrogenism and amenorrhea, while limiting the use
of US (22).
The reluctance to include the use of US as a diagnostic
tool for PCOS in adolescence is based on three assumptions
that our study was structured to address: the prevalence of
PCOM in the general adolescent population; the evolution
of ovarian morphology; and the imaging quality of TAUS,
particularly in obese adolescents. First, studies have demon-
strated that PCOM is found in up to 40% of the healthy adult
and adolescent population and thus questioned the validity of
the diagnosis of PCOS using the revised RC (15, 27). A recent
Australian study found a prevalence of PCOM of 35% in 244
unselected adolescents aged 1416 years, mainly attributed to
increased OV rather than an increased number of follicles (9).
Whether a proportion of the study populations had a variant
of PCOS is unknown (16). Moreover, women presenting with
isolated PCOM have higher levels of AMH compared with
control subjects, suggesting the presence of a granulosa cell
abnormality in PCO similar to that observed in PCOS (28).
In contrast, using a comparison group with a proven
surgical diagnosis and no documentation of irregular
menses or clinical signs of hyperandrogenism, we found the
prevalence of PCO in the general adolescent population to
be only 11%, which is compatible with the data presented
in several other studies (2, 29). Because we do not have
laboratory data available for the control subjects, we cannot
exclude the possibility of biochemical hyperandrogenism in
this cohort; thus the percentage of PCOM in the control
population may be lower.
Second, investigators have raised concerns about the util-
ity of sonography as ovarian morphology may evolve during
adolescence and early adulthood (30), specically the nding
of a multifollicular appearance. In our study, no association
was found between age and PCOM in girls with PCOS. In
the comparison cohort, which had a young mean age of
14.6 years, older age was associated with the presence of
PCOM. This result was attributed to the increase in OV, as ex-
pected with increased age, and not to an increase in the num-
ber of follicles. We found no association between older age
and the likelihood of having a reduced number of follicles.
If a patient is evaluated in early adolescence and does not
have features of PCO on imaging, she may be reevaluated at
Comparison group Comparison follicular phase Comparison luteal phase
(n [ 94) (%) group (n [ 47) (%) group (n [ 33) (%)
29 (31) 23 (51) 3 (9)
56 (60) 21 (47) 24 (73)
9 (10) 1 (2) 6 (18)
VOL. 102 NO. 5 / NOVEMBER 2014

a later time. However, if PCOM is present, specically based
on OV, then studies suggest this nding is unlikely to regress
(30, 31).
A third concern repeatedly raised about the use of sonog-
raphy as a criterion for PCOS is the technical challenge asso-
ciated with accurately visualizing the ovaries on TAUS in this
frequently obese population. In our study, we were able to
visualize and calculate the ovarian and endometrial values
for the vast majority of our patients regardless of weight,
consistent with other studies (9, 15, 29). When the ovaries
are adequately visualized, OV can reliably be assessed (32,
33). The similarity between the prevalence of PCOM in our
case group and previous studies (16) strengthens the
validity of our results and underscores the high usability of
TAUS in our study. Recent advances in technology have
improved detection rates of antral follicles, especially when
using endovaginal sonography with FNPO count (25, 34,
35), leading to the suggestion that follicle count cutoffs be
increased. However, in adolescents, most imaging is
performed transabdominally, and despite the improvement
in US machine resolution, several studies have
demonstrated a remarkably similar threshold for OV and
FNPS when comparing new versus older reports using a
single cross-sectional view of the ovary (25, 36). The
investigators concluded that a combined metric of FNPS
(using a cutoff of nine to 10 follicles) and OV provides
signicant predictive power in detecting PCOS comparable
with that of FNPO alone (25). In situations in which image
quality is reduced, the investigators suggest using the
volume criterion alone (sensitivity of 81% and specicity of
84% in distinguishing between PCOS and non-PCOS patients)
(36). These studies suggest that the use of TAUS in obese girls,
even with concerns for suboptimal image quality, has a sig-
nicant predictive value in the diagnosis of PCOS using the
OV and FNPS, as was used in our study.
Endometrial thickness did not add diagnostic informa-
tion. Most girls in the PCOS group had an endometrial thick-
ness <5 mm, consistent with previous reports (37). Nine girls
in the PCOS group (15%) had an endometrium >10 mm.
In adulthood, PCOS and anovulation are risk factors for endo-
metrial hyperplasia and cancer (38, 39), which typically occur
in cases in which the endometrium is thick and heterogeneous
(40). Unopposed estrogen secretion and anovulation should
be treated to limit endometrial overgrowth and prevent this
increased risk in the future.
Study limitations should be acknowledged. This study
was a single-center review of imaging, laboratory, and clin-
ical data. However, the study group was selected based on
clinical criteria ascertained in real time, and the imaging
was reviewed specically for these analyses. The comparison
group was carefully selected to include adolescents who were
evaluated for a nongynecological diagnosis. As their main
complaint was acute abdominal pain, the evaluation was per-
formed to exclude any ovarian pathology according to a uni-
ed pelvic exam protocol used in our institution. As expected,
more US studies were excluded from analysis for the presence
of a dominant follicle >10 mm from the comparison group. In
addition, menstrual history and the use of contraceptives
VOL. 102 NO. 5 / NOVEMBER 2014
Fertility and Sterility
were addressed in the vast majority of cases as these factors
have implications for the differential diagnosis of abdominal
pain in adolescents.
We were able to compare weights between the two
groups. Obesity is a prevalent clinical feature in PCOS pa-
tients, and the mean weight was higher in the PCOS group
than in the comparison group. We did not nd a correlation
between weight and US usability in terms of image quality;
thus, weight did not affect our ability to review the images.
We did not have a sufcient number of height measurements
for the comparison group as this is not a part of the routine
evaluation in the ED, thus a comparison of body mass index
between the groups was not presented.
It is still possible that some ndings of PCOS, such as the
presence of hirsutism or high androgen levels, could have
been present and not documented in the 11 patients with
PCOM in the comparison group; however, exclusion of any
patient included in that group would only further decrease
the percentage of PCOM identied in the normal comparison
group. It is also possible that relevant patient medical history
including recent past medical use of oral contraceptives that
might affect ovarian morphology was not documented.
Sonographic interpretation was performed by review of
previously obtained still frames; both clinicians remeasured
all relevant parameters and recalculated all volumes. It is
possible that prospective assessment of real-time images
and/or cine clips might be more precise. A recent task force
recommended using FNPO on transvaginal US with a higher
threshold as the most accurate diagnostic criteria for PCOM
when using advanced US equipment (41). With older US
equipment, the task force recommended using the volume
criteria. As noted above, an evaluation based on a combined
metric of FNPS and OV can provide predictive power in de-
tecting PCOM comparable with that of FNPO alone (25).
Most pelvic US exams in adolescent girls are performed
with a lower resolution TA transducer and use FNPS counts,
emphasizing the need to develop targeted criteria for this
age group. In the current study, we used a combined metric
of FNPS and OV measured transabdominally that requires
further validation in future investigations. Images were re-
viewed independently by two different clinicians with a
high level of interobserver agreement for follicle number
and OV.
In conclusion, the prevalence of ovarian morphology
meeting the mRC by TAUS in girls diagnosed with PCOS ac-
cording to the NIH criteria, in our study, was markedly higher
than that in adolescents who served as a comparison group.
Our study found the prevalence of PCOM using a sonographic
mRC in a nongynecologic cohort to be lower than previously
reported, supporting the usefulness of US in adolescents as
part of the diagnosis of PCOS. Based on our results, TAUS
may provide useful information in the evaluation of PCOM,
even in obese girls. Studies have indicated that adolescence
may be the most appropriate time to intervene in PCOS pa-
tients, as many cardiovascular risk factors are present in early
adulthood (42). Continued efforts to dene diagnostic criteria
in adolescents that potentially include TAUS are warranted so
that patient-specic early interventions can be initiated.
1437
ORIGINAL ARTICLE: REPRODUCTIVE ENDOCRINOLOGY
REFERENCES
1. Fauser BC, Tarlatzis BC, Rebar RW, Legro RS, Balen AH, Lobo R, 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;97:2838.e25.
2. Rahmanpour H, Jamal L, Mousavinasab SN, Esmailzadeh A, Azarkhish K. As-
sociation between polycystic ovarian syndrome, overweight, and metabolic
syndrome in adolescents. J Pediatr Adolesc Gynecol 2012;25:20812.
3. Michaliszyn SF, Lee S, Tfayli H, Arslanian S. Polycystic ovary syndrome and
nonalcoholic fatty liver in obese adolescents: association with metabolic
risk prole. Fertil Steril 2013;100:174551.
4. Shayya R, Chang RJ. Reproductive endocrinology of adolescent polycystic
ovary syndrome. Br J Obstet Gynecol 2010;117:1505.
5. Revised 2003 consensus on diagnostic criteria and long-term health risks
related to polycystic ovary syndrome (PCOS). Hum Reprod 2004;19:417.
6. Balen AH, Laven JS, Tan SL, Dewailly D. Ultrasound assessment of the poly-
cystic ovary: international consensus denitions. Hum Reprod Update 2003;
9:50514.
7. Azziz R, Carmina E, Dewailly D, Diamanti-Kandarakis E, Escobar-
Morreale HF, Futterweit W, et al. The Androgen Excess and PCOS Society
criteria for the polycystic ovary syndrome: the complete task force report.
Fertil Steril 2009;91:45688.
8. Metcalf MG, Skidmore DS, Lowry GF, Mackenzie JA. Incidence of ovulation
in the years after the menarche. J Endocrinol 1983;97:2139.
9. Hickey M, Doherty DA, Atkinson H, Sloboda DM, Franks S, Norman RJ, et al.
Clinical, ultrasound and biochemical features of polycystic ovary syndrome
in adolescents: implications for diagnosis. Hum Reprod 2011;26:146977.
10. Olutunmbi Y, Paley K, English JC. Adolescent female acne: etiology and
management. J Pediatr Adolesc Gynecol 2008;21:1716.
11. Jeffrey Chang R, Cofer MS. Polycystic ovary syndrome: early detection in
the adolescent. Clin Obstet Gynecol 2007;50:17887.
12. Carmina E, Obereld SE, Lobo RA. The diagnosis of polycystic ovary syn-
drome in adolescents. Am J Obstet Gynecol 2010;203:201.e15.
13. Blank SK, Helm KD, McCartney CR, Marshall JC. Polycystic ovary syndrome
in adolescence. Ann N Y Acad Sci 2008;1135:7684.
14. Rosner W, Auchus RJ, Azziz R, Sluss PM, Raff H. Position statement: utility,
limitations, and pitfalls in measuring testosterone: an Endocrine Society po-
sition statement. J Clin Endocrinol Metab 2007;92:40513.
15. Johnstone EB, Rosen MP, Neril R, Trevithick D, Sternfeld B, Murphy R, et al.
The polycystic ovary post-Rotterdam: a common, age-dependent nding in
ovulatory women without metabolic signicance. J Clin Endocrinol Metab
2010;95:496572.
16. Mortensen M, Ehrmann DA, Littlejohn E, Roseneld RL. Asymptomatic vol-
unteers with a polycystic ovary are a functionally distinct but heterogeneous
population. J Clin Endocrinol Metab 2009;94:157986.
17. Codner E, Villarroel C, Eyzaguirre FC, Lo pez P, Merino PM, Perez-Bravo F,
et al. Polycystic ovarian morphology in postmenarchal adolescents. Fertil
Steril 2011;95:7026.e12.
18. Pigny P, Jonard S, Robert Y, Dewailly D. Serum anti-Mullerian hormone as a
surrogate for antral follicle count for denition of the polycystic ovary syn-
drome. J Clin Endocrinol Metab 2006;91:9415.
19. Yoo RY, Sirlin CB, Gottschalk M, Chang RJ. Ovarian imaging by magnetic
resonance in obese adolescent girls with polycystic ovary syndrome: a pilot
study. Fertil Steril 2005;84:98595.
20. Iliodromiti S, Kelsey TW, Anderson RA, Nelson SM. Can anti-Mullerian hor-
mone predict the diagnosis of polycystic ovary syndrome? A systematic re-
view and meta-analysis of extracted data. J Clin Endocrinol Metab 2013;
98:333240.
21. Siow Y, Kives S, Hertweck P, Perlman S, Fallat ME. Serum Mu llerian-inhibit-
ing substance levels in adolescent girls with normal menstrual cycles or with
polycystic ovary syndrome. Fertil Steril 2005;84:93844.
22. Legro RS, Arslanian SA, Ehrmann DA, Hoeger KM, Murad MH, Pasquali R,
et al. Diagnosis and treatment of polycystic ovary syndrome: an Endocrine So-
ciety clinical practice guideline. J Clin Endocrinol Metab 2013;98:456592.
1438
23. Adams J, Franks S, Polson DW, Mason HD, Abdulwahid N, Tucker M, et al.
Multifollicular ovaries: clinical and endocrine features and response to pulsa-
tile gonadotropin releasing hormone. Lancet 1985;2:13759.
24. Atiomo WU, Pearson S, Shaw S, Prentice A, Dubbins P. Ultrasound criteria in
the diagnosis of polycystic ovary syndrome (PCOS). Ultrasound Med Biol
2000;26:97780.
25. Christ JP, Willis AD, Brooks ED, Vanden Brink H, Jarrett BY, Pierson RA, et al. Fol-
licle number, not assessments of the ovarian stroma, represents the best ultraso-
nographic marker of polycystic ovary syndrome. Fertil Steril 2014;101:2807.e1.
26. Boyle J, Teede HJ. Polycystic ovary syndromean update. Aust Fam Physi-
cian 2012;41:7526.
27. Mortensen M, Roseneld RL, Littlejohn E. Functional signicance of
polycystic-size ovaries in healthy adolescents. J Clin Endocrinol Metab
2006;91:378690.
28. Catteau-Jonard S, Bancquart J, Poncelet E, Lefebvre-Maunoury C, Robin G,
Dewailly D. Polycystic ovaries at ultrasound: normal variant or silent polycy-
stic ovary syndrome? Ultrasound Obstet Gynecol 2012;40:2239.
29. Van Hooff MH, Voorhorst FJ, Kaptein MB, Hirasing RA, Koppenaal C,
Schoemaker J. Hyperandrogenism: polycystic ovaries in adolescents and
the relationship with menstrual cycle patterns, luteinizing hormone, andro-
gens, and insulin. Fertil Steril 2000;74:49.
30. Venturoli S, Porcu E, Fabbri R, Pluchinotta V, Ruggeri S, Macrelli S, et al. Lon-
gitudinal change of sonographic ovarian aspects and endocrine parameters
in irregular cycles of adolescence. Pediatr Res 1995;38:97480.
31. Panidis D, Tziomalos K, Macut D, Delkos D, Betsas G, Misichronis G, et al.
Cross-sectional analysis of the effects of age on the hormonal, metabolic,
and ultrasonographic features and the prevalence of the different pheno-
types of polycystic ovary syndrome. Fertil Steril 2012;97:494500.
32. Chen Y, Yang D, Li L, Chen X. The role of ovarian volume as a diagnostic cri-
terion for Chinese adolescents with polycystic ovary syndrome. J Pediatr
Adolesc Gynecol 2008;21:34750.
33. Herter LD, Magalhaes JA, Spritzer PM. Relevance of the determination of
ovarian volume in adolescent girls with menstrual disorders. J Clin Ultra-
sound 1996;24:2438.
34. Dewailly D, Gronier H, Poncelet E, Robin G, Leroy M, Pigny P, et al. Diagnosis
of polycystic ovary syndrome (PCOS): revisiting the threshold values of folli-
cle count on ultrasound and of the serum AMH level for the denition of
polycystic ovaries. Hum Reprod 2011;26:31239.
35. Kristensen SL, Ramlau-Hansen CH, Ernst E, Olsen SF, Bonde JP, Vested A,
et al. A very large proportion of young Danish women have polycystic
ovaries: is a revision of the Rotterdam criteria needed? Hum Reprod 2010;
25:311722.
36. Lujan ME, Jarrett BY, Brooks ED, Reines JK, Peppin AK, Muhn N, et al. Up-
dated ultrasound criteria for polycystic ovary syndrome: reliable thresholds
for elevated follicle population and ovarian volume. Hum Reprod 2013;
28:13618.
37. Shah B, Parnell L, Milla S, Kessler M, David R. Endometrial thickness, uterine,
and ovarian ultrasonographic features in adolescents with polycystic ovarian
syndrome. J Pediatr Adolesc Gynecol 2010;23:14652.
38. Cheung AP. Ultrasound and menstrual history in predicting endometrial hy-
perplasia in polycystic ovary syndrome. Obstet Gynecol 2001;98:32531.
39. Wild S, Pierpoint T, Jacobs H, McKeigue P. Long-term consequences of poly-
cystic ovary syndrome: results of a 31 year follow-up study. Hum Fertil
(Camb) 2000;3:1015.
40. Peri N, Levine D. Sonographic evaluation of the endometrium in patients
with a history or an appearance of polycystic ovarian syndrome. J Ultrasound
Med 2007;26:558. quiz 5960.
41. Dewailly D, Lujan ME, Carmina E, Cedars MI, Laven J, Norman RJ, et al. De-
nition and signicance of polycystic ovarian morphology: a task force report
from the Androgen Excess and Polycystic Ovary Syndrome Society. Hum Re-
prod Update 2014;20:33452.
42. Vural B, Caliskan E, Turkoz E, Kilic T, Demirci A. Evaluation of metabolic syn-
drome frequency and premature carotid atherosclerosis in young women
with polycystic ovary syndrome. Hum Reprod 2005;20:240913.
VOL. 102 NO. 5 / NOVEMBER 2014