Key words
hirsutism - hyperandrogenemia - polycystic ovary syndrome - body weight - FSH
Introduction
Polycystic ovarian syndrome (PCOS) is a heterogenous endocrinal disorder and one of
the leading causes of infertility worldwide [1]. It is also associated with many serious long-term complications
including metabolic disorders such as type 2 diabetes mellitus, hypertension,
cardiovascular, cerebrovascular disorders, and endometrial carcinoma, if left
untreated [2]
[3]. The prevalence of PCOS in women all over the world shows wide
variation due to the different diagnostic criteria used, that is, according to
National Health Institute (NIH), 1990 it is 3.39%, Rotterdam’s,
2003, 11.04%, and Androgen excess society (AES), 2006 it is 8.03%
[4]. Despite its common prevalence in
today’s era the exact pathophysiology of PCOS remains unclear due to
complicated interwoven factors playing role in its development including genetic and
epigenetic factors, ovarian abnormalities, neuroendocrine modifications,
hyperinsulinemia, insulin resistance, endocrine and metabolic changes, adiposity,
adrenal factors, etc. [5].
The classical clinical presentation of PCOS includes features of hyperandrogenism,
chronic anovulation/oligomenorrhoea, and ultrasonographic morphology of the
ovary [6]. It was observed that the main
pathology lies with ovarian steroidogenesis leading to increased androgen production
and ovarian dysfunction [7]. Of many criteria
used for the diagnosis of PCOS, Rotterdam and AES criteria are most commonly used.
According to the Rotterdam criteria, the women should fulfill two of the following:
oligo-ovulation or anovulation, clinical or biochemical signs of hyperandrogenism,
and/or polycystic ovarian morphology [8], whereas the AES has further broadened the criteria as a woman having
clinical or biochemical hyperandrogenism, ovarian dysfunction (oligo-anovulation
and/or polycystic ovaries), and exclusion of other androgen excess or
related disorders [9]. Changing lifestyles,
better diagnostic facilities, and increasing awareness, have led to a rise in cases
of PCOS all over the world. Furthermore, early diagnosis and management of this
chronic condition can help in preventing the various long-term complications
associated with it. Hence, the present study was conducted to assess the early
clinical, biochemical, and radiological features in obese, non-obese young women
with PCOS as compared to non-PCOS women.
Subjects and Methods
Study design
This is a prospective comparative study involving healthy young women and women
with suspected PCOS. Diagnosis of PCOS was made if the woman presents with two
of three features of Rotterdam criteria [8]. Hirsutism was considered with a score of≥8 by
Ferriman-Gallwey scoring system [10]. For
the Asian and South Asian populations, obesity is defined as
BMI≥25 kg/m2
[11].
Study setting
The present study was conducted in the Department of Obstetrics and Gynaecology
of a rural tertiary care center of Northern India over one year (January 2017 to
January 2018).
Participants
Inclusion criteria
All young women between 18–22 years of age with or without features of
oligo/amenorrhoea, clinical signs of hyperandrogenism, and, or
polycystic ovarian morphology on ultrasound were enrolled as study participants.
A total of 80 participants fulfilling Rotterdam’s criteria for PCOS were
included as cases. Based on the BMI, they were further divided into two groups:
Group I as obese PCOS (BMI≥25 kg/m2) and
Group II as non-obese PCOS (BMI<25 kg/m2)
with 40 participants in each case group. There were 40 control participants
(both obese and non-obese) not fulfilling Rotterdam’s criteria for
PCOS.
Exclusion criteria
Women<18 or>22 years, those on oral contraceptive pills or any
other drug that can affect gonadal function, adrenal or ovarian tumors,
congenital adrenal hyperplasia, Cushing syndrome, acromegaly, hypogonadotropic
hypogonadism, endocrine abnormalities like hyperprolactinemia, thyroid
disorders, androgen-secreting tumors, women with Mullerian anomalies,
Asherman’s syndrome, Turner’s syndrome, addicted to
alcohol/smoking/drugs and those refusing to participate were
excluded from the study.
Sampling procedure
Consecutive sampling of participants fulfilling the inclusion criteria was done
for the entire duration of the study.
Data sources/measurements
After informed written consent from all the participants in their vernacular
language various sociodemographic parameters like age, religion, education, and
socio-economic status were recorded, followed by the anthropometric measurement
and clinical examination including weight, height, BMI, waist and hip
circumference, menstrual cycle details, acne, acanthosis nigricans (as
identified by the presence of dark, velvety discoloration of skin folds and
creases around armpits, neck or groin) and hirsutism were all recorded on a
preformed data collection sheet by trained nursing staff. A detailed medical,
surgical, and family history of all the participants was recorded, followed by a
thorough physical examination and investigations of the participants.
Anthropometric measurements
The height was measured in an upright position using a stadiometer, closest to
0.5 cm. The participants were asked to remove their hair ornaments,
ponytails, buns, braids, shoes and were made to stand on the floor with the
heels of both feet together and the toes pointed slightly outward. Then they
were asked to take a deep breath and stand as tall as possible with their eyes
looking straight ahead and to hold this position for a while till the horizontal
headboard is brought down firmly on top of the head. The reading in cm was
recorded at the level where the headboard is aligned with the vertical
scale.
The weight of the participants was measured in light clothing without shoes using
a digital weighing machine in kilograms to the nearest 0.1 kg. BMI
(kg/m2) was calculated by using the formula: weight in
kg/height in meters2.
Waist circumference (WC) was measured in the horizontal plane between the lower
margin of the last rib to the iliac crest using an inch tape to the nearest
0.1 cm. Hip circumference (HC) was taken at the widest portion of the
buttocks by wrapping the inch tape around it. This was followed by the
calculation of waist to hip ratio (WHR) by dividing the two parameters
(WC/HC).
Scoring of hirsutism
Hirsutism was assessed at a total of nine points; upper lip, chin, chest, upper
and lower back, upper and lower abdomen, thigh, and upper arm using a modified
Ferriman–Gallwey score. The minimum score was zero and the maximum 36
according to this scale. The scoring was done as 8=no hirsutism;
8–16=mild hirsutism; 17–30=moderate
hirsutism;≥25=severe hirsutism [12]. All the participants who had a score of≥8 were
considered clinically hirsute.
Following anthropometric measurements and clinical examination, baseline
endocrinal investigations, and transabdominal pelvic ultrasonography for ovarian
volume, the number, and arrangement of ovarian follicles were performed in the
early follicular phase (day 2 or 3) of the spontaneous or induced menstrual
cycle.
Biochemical test
The hormonal tests were performed after overnight fasting, with around
8–10 ml of blood drawn in a plain vial using a heparinized
syringe. The following biochemical and hormonal assays were done: Serum
Thyroid-stimulating hormone (TSH), Prolactin, Follicle-stimulating hormone
(FSH), Luteinizing Hormone (LH), Total Testosterone, fasting, and postprandial
sugar levels. The blood collected in test tubes was centrifuged to separate the
serum, which was then stored at –20°C for hormonal assays. All
the hormonal tests were carried out in duplicate in the same assay by
Radioimmunoassay (RIA) in the laboratory of the Biochemistry department of the
institute. Serum FSH, LH, and total Testosterone were determined by the sandwich
ELISA technique.
Statistical analysis
Statistical analysis of data was performed using Statistical Package for Social
Sciences (SPSS) software version 22.0. Comparison of the groups or continuous
measures was done using a Student’s t-test and F-test was used to
calculate the equity of two sample variations. Comparison between groups or
nominal measures was done using Fisher’s exact test. Results were
reported as mean±SD wherever required. A p-value<0.05 was
considered statistically significant.
Ethical issues
The present research involving human subjects was conducted following the ethical
standards of all applicable national and institutional committees and the World
Medical Association’s Helsinki Declaration. It was conducted after
informed written consent from the participants and ethical approval from the
Institutional Ethical Committee (IEC number: 2016/841).
Results
The mean age and age of onset of menarche in obese PCOS was 20.43±1.53 years
and 13.05±1.84 years, in non-obese PCOS 20.25±1.45 years and
13.18±0.93 years, and in controls 20.58±1.47 years and
13.43±0.90 years respectively. The mean height, weight, BMI, WHR in obese
PCOS, non-obese PCOS, and controls were 1.52±0.07 m,
65.88±8.88 kg, 28.36±2.47 kg/m2,
and 0.95±0.08 versus 1.59±0.07 m,
57.23±7.83 kg, 22.58±2.10 kg/m2
and 0.91±0.09 versus 1.53±0.09 m,
61.45±9.56 kg, 26.25±3.63 kg/m2
and 0.89±0.09, respectively. A significant difference was observed between
the anthropometric measurements of obese and non-obese PCOS cases (p<0.05).
The comparison of clinical features including acne, acanthosis nigricans, and
hirsutism between obese, non-obese PCOS and controls is depicted in [Table 1]. Clinical features including acne,
acanthosis nigricans, and hirsutism were more in obese PCOS cases as compared to
non-obese PCOS and controls. On ultrasonography the mean ovarian volume and the
number of follicles<9 mm were significantly higher in both obese and
non-obese PCOS cases as compared to controls. Furthermore, the peripheral
arrangement of these follicles in the ovary on ultrasound was more commonly
associated with cases as compared to controls (p=0.000). The comparison of
ultrasound findings in cases and controls is depicted in [Table 2]. Of all the biochemical tests
performed a significant difference was observed in serum LH, FSH, and total
Testosterone between cases and controls, with obese PCOS cases having significantly
higher LH (10.04±1.60 vs. 8.93±2.40 mIU/ml) and total
testosterone levels (2.71±0.39 vs. 2.21±0.39 pg/ml)
as compared to non-obese PCOS cases. Furthermore, the LH: FSH ratio of 3:1 was
significantly associated with cases as compared to controls. The comparison of
biochemical parameters between cases and controls is depicted in [Table 3].
Table 1 Comparison of clinical features of cases (obese and
nonobese PCOS) and controls.
|
Clinical feature
|
Obese PCOS
|
Nonobese PCOS
|
Non-PCOS controls
|
|
Acne
|
|
|
|
|
Present
|
24 (60.0%)
|
13 (32.5%)
|
10 (25.0%)
|
|
Absent
|
16 (40.0%)
|
27 (67.5%)
|
30 (75.0%)
|
|
Acanthosis nigricans
|
|
|
|
|
Present
|
14 (35.0%)
|
4 (10.0%)
|
1 (2.5%)
|
|
Absent
|
26 (65.0%)
|
36 (90.0%)
|
39 (97.5%)
|
|
Grading of Hirsutism
|
|
|
|
|
Absent (<8)
|
5 (12.5%)
|
7 (17.5%)
|
25 (62.5%)
|
|
Mild (8–16)
|
25 (62.5%)
|
30 (75.0%)
|
11 (27.5%)
|
|
Moderate (17–24)
|
10 (25.0%)
|
3 (7.5%)
|
4 (10.0%)
|
|
Severe (≥25)
|
0 (0%)
|
0 (0%)
|
0 (0%)
|
Table 2 Comparison of ultrasound findings of cases (obese and
nonobese PCOS) and controls.
|
Ultrasonographic findings
|
Obese PCOS
|
Nonobese PCOS
|
Non-PCOS controls
|
|
Ovarian volume (mean±SD)
|
9.37±2.77
|
9.26±2.62
|
5.05±1.92
|
|
Number of follicles (mean±SD)
|
14.80±3.17
|
15.10±3.06
|
4.13±1.52
|
|
Arrangement of follicles
|
|
|
|
|
Central
|
0 (0%)
|
0 (0%)
|
33 (82.5%)
|
|
Peripheral
|
40 (100%)
|
40 (100%)
|
7 (17.5%)
|
Table 3 Comparison of biochemical parameters of cases (obese
and nonobese PCOS) and controls.
|
Biochemical parameters (mean±SD)
|
Obese PCOS
|
Nonobese PCOS
|
Non-PCOS controls
|
|
TSH (uIU/ml)
|
3.14±0.84
|
3.20±0.85
|
3.10±0.91
|
|
Fasting blood sugar (mg/dl)
|
71.20±6.98
|
69.35±6.32
|
72.65±6.78
|
|
Postprandial blood sugar (mg/dl)
|
135.33±10.56
|
136.55±9.33
|
134.93±10.13
|
|
Serum prolactin (ng/ml)
|
10.02±6.11
|
9.54±5.95
|
11.03±6.22
|
|
Serum FSH (IU/l)
|
3.71±0.95
|
3.52±0.99
|
5.16±1.56
|
|
Serum LH (IU/l)
|
10.04±1.60
|
8.93±2.40
|
7.51±2.85
|
|
Serum total testosterone (mmol/l)
|
2.71±0.39
|
2.21±0.39
|
1.79±0.93
|
|
LH:FSH ratio [n (%)]
|
|
|
|
|
1:1
|
0 (0%)
|
4 (10.0%)
|
23 (57.5%)
|
|
2:1
|
9 (22.5%)
|
11 (27.5%)
|
10 (25.0%)
|
|
3:1
|
31 (77.5%)
|
25 (62.5%)
|
7 (17.5%)
|
Discussion
PCOS is one of the most common endocrinal diseases of multifactorial origin affecting
women of all age groups. There is a strong correlation between obesity and PCOS
[13]
[14], though it can occur in non-obese women also. In the present study,
the early clinical, biochemical, and radiological features in obese and non-obese
PCOS cases were compared with the non-PCOS young women.
No significant difference in age at presentation and age of menarche was observed
in
our study between the cases and controls. This was similar to the results of a study
that also reported insignificant differences between cases and controls regarding
age at presentation and age of menarche [15].
Menstrual irregularity was the most common presenting complaint in obese and
non-obese PCOS cases as it was considered one of the criteria for diagnosis of PCOS.
Similar results were reported by other studies also [16]
[17]. In the present study, a
significant difference was observed between the anthropometric measurements of obese
and non-obese PCOS cases [height: 1.52±0.07 m vs.
1.59±0.07 m (p=0.000); weight: 65.88±8.88 kg
vs. 57.23±7.83 kg (p=0.000); BMI:
28.36±2.47 kg/m2 vs.
22.58±2.10 kg/m2 (p=0.000); WHR:
0.95±0.08 vs. 0.91±0.09 (0.018)]. Furthermore, the mean BMI and WHR
were significantly higher in cases as compared to controls. This was similar to a
study that observed, women with PCOS had significantly higher central fat mass
(waist, waist-hip ratio, and upper/lower fat ratio) and BMI as compared to
controls [18]. Another study reported that
lean PCOS cases despite low BMI have abdominal obesity [19].
In the present study, obese PCOS cases were significantly associated with acne,
acanthosis nigricans, and hirsutism as compared to non-obese PCOS and controls. A
recent study reported that PCOS women and higher BMI had increased hair growth as
assessed by modified Ferriman-Gallwey scores that were found to be 2.96-fold higher
as compared to healthy-normal BMI counterparts [20]. Another study compared hirsutism in normal weight and overweight
PCOS cases and reported that hirsutism was more commonly (p=0.009) seen in
overweight PCOS cases [21]. Similar results of
increased clinical hyperandrogenism in obese PCOS cases as compared to nonobese
cases were reported by other studies also [22]
[23]. Another study reported
that acanthosis nigricans was more common in obese women with higher waist
circumference [24]. Several studies have
reported that cutaneous manifestations including hirsutism, acne, seborrhea,
acanthosis nigricans, and acrochordons are more common in PCOS cases, especially
obese PCOS, and were associated with high serum fasting insulin levels or insulin
resistance [25]
[26].
On ultrasonography the mean ovarian volume and the number of
follicles<9 mm with peripheral arrangement were significantly higher
in both obese and non-obese PCOS cases as compared to controls in the present study.
A similar study found a positive association between mean ovarian volume and WHR
[27]. Another recent study reported that
the mean ovarian volume was statistically higher in obese PCOS cases as compared to
non-obese counterparts [28]. A similar study
reported a significant increase in ovarian volume and the number of ovarian
follicles (<9 mm) on ultrasound in PCOS cases as compared to
controls [29]. On the other hand, a recent
study observed no significant impact of high BMI on antral follicle count, ovarian
volume, or serum androgens [20].
In our study, of all the biochemical tests performed, a significant difference was
observed in serum LH, FSH, and total Testosterone between cases and controls, with
obese PCOS cases having significantly higher LH (10.04±1.60 vs.
8.93±2.40 mIU/ml) and total testosterone levels (2.71±0.39
vs. 2.21±0.39 pg/ml) as compared to non-obese PCOS cases.
Furthermore, the LH:FSH ratio of 3:1 was significantly associated with cases as
compared to controls. Similar results of significantly higher serum LH and LH:FSH
ratio in obese PCOS as compared to non-obese PCOS and controls was reported by a
recent study [29]. Another study reported that
obese PCOS cases had significantly higher levels of basal total testosterone as
compared to non-obese PCOS cases [30]. Similar
results of significantly higher serum testosterone and androstenedione levels in
obese PCOS as compared to nonobese PCOS cases were reported by another study [23]. Other recent studies have observed no
significant difference between obese and non-obese PCOS groups in terms of the
LH/FSH ratio. Furthermore, they observed no significant correlation between
BMI and LH/FSH ratio [23]
[31]. Contrary to our results a study reported
significantly higher serum levels of LH in lean PCOS women as compared to obese or
overweight PCOS women [32].
Conclusion
PCOS is a heterogeneous disorder of multifactorial origin with symptoms,
ultrasonographic features, and biochemical parameters being more severely deranged
in obese cases as compared to nonobese cases and controls. The cutaneous
manifestation including acne, hirsutism, acanthosis nigricans, alopecia, and
seborrhea is more commonly associated with increased BMI, and insulin resistance.
Obese cases also have higher serum total testosterone, LH levels, and LH:FSH ratio
of≥3:1. Therefore, obese with PCOS are at a higher risk of having long-term
complications and hence need early and vigorous management.
