Journal of Reproduction & Contraception

doi: 10.7669/j.issn.1001-7844.2014.03.0133

2014 Sep.; 25(3):133-145

E-mail: randc_journal@163.com

ORIGINAL PAPER

Study on the Clinical and Endocrine Characteristics of Polycystic Ovary Syndrome with Different Ovarian Morphology
Zhi-ping HU, Tian-shu PANG, Ying WANG, Jie QIAO, Mei-zhi LI
Reproductive Medical Center, Peking University Third Hospital, Beijing 100191, China

Objective
To evaluate the differences of the clinical manifestation and endocrine
situation in patients with different ovarian morphology of polycystic ovary syndrome
(PCOS).
Methods
A total of 234 PCOS patients were enrolled according to the ovary
morphology and divided into three groups: 112 patients with B-polycystic ovary morphology (both two ovaries were PCOM, B-PCOM), 50 with U-PCOM (only one ovary was
PCOM) and 72 with N-PCOM (none was PCOM). There were 39 infertile women
without PCOS as control group. Data were analyzed by using SPSS 15.0 software.
Results There was no statistical difference in body mass index (BMI) among the three
groups of PCOS. The endometrial thickness increased in patients with B-PCOM and
decreased with N-PCOM. The levels of testosterone, androstenedione and luteinizing
hormone increased in PCOS groups, especially in N-PCOM patients. HOMA-IR
increased, HOMA-, disposition index (DI) and I60 /G60 decreased in patients with NPCOM compared with in B-PCOM and U-PCOM groups. Higher level of total cholesterol (TC) and lower level of high-density lipoprotein (HDL)-C existed in PCOS patients,
especially in N-PCOM. There were positive correlations between oligo-anovulation,
endometrial thickness, LH/FSH ratio, fasting insulin (FINS), the area under curve of
glucose(AUCGLU) and PCOM, while there was a negative correlation between HOMAIR and PCOM.
Conclusion
There are relationships among hyperandrogenism, hyperinsulinemia,
insulin resistance (IR) and ovary morphology in PCOS patients. PCOS patients without PCOM have more serious IR and hyperandrogenism.
Key words: polycystic ovary syndrome (PCOS); polycystic ovary morphology (PCOM);
hyperandrogenism (HA); insulin resistance (IR)
Corresponding author: Ying WANG; Tel: +86-13911688652; E-mail: wangying02114@bjmu.edu.cn
133

Polycystic ovary syndrome (PCOS) is one of the most common endocrine disorders,
which affects 6%-10% of reproductive-aged women[1]. This syndrome is not a well-defined
entity and shows the great variation and heterogeneity in clinic, menstrual disorders, infertility,
acne, obesity and polycystic ovary morphology (PCOM) are typical clinical manifestations.
Further more, insulin resistance (IR), hyperinsulinemia and hyperlipidemia increase with age.
Anovulation and hyperandrogenism (HA) are the main manifestations of PCOS. There
is still a controversy on the diagnosis criteria so far. In the year of 1992[2], National Institutes
of Health/National Institute of Child Health and Human Development (NIH/NICHD) drew
up the first diagnostic criteria which was widely recognized, including oligoanovulation, HA
and/or clinical manifestations of HA. Then European Society of Human Reproduction and
Embryology/American Society For Reproductive Medicine (ESHRE/ASRM) revised this
criteria in 2003 as Rotterdam Criteria [3], meeting at least 2 of the following 3 standards:
menstrual disorder (menstrual cycle >35 d and/or <21 d); clinical signs of HA (hirsutism and/
or acne and/or HA); and polycystic ovaries on ultrasound (12 or more follicles 2-9 mm in
diameter in each ovary). Other conditions, such as congenital adrenal hyperplasia, androgensecreting tumors, and Cushings syndrome, need to be ruled out on clinical examination, by
assessing testosterone and 17-hydroxyprogesterone (17-OHP) levels, and performing the
rapid dexamethasone suppression test. Participants were with 17-OHP levels higher than
25.44 nmol/L and/or testosterone levels higher than 6.94 nmol/L, and/or cortisol levels higher
than 195 nmol/L. But many scholars believe that HA is the basic pathogen of PCOS, PCOM
also exists in other diseases, particularly functional hypothalamic anovulation. However, it is
not specific. In the year 2006[4], experts in Androgen Excess Society defined HA as the
requirement of PCOS diagnosis, but not to be widely recognized. Several researchers find
that simple PCOM perhaps is the early clinical manifestation of PCOS, some patients with
only PCOM, without irregular menstruation and hirsutism, have the tendency to develop
PCOS, the levels of testosterone and androstenedione are higher than people without PCOM,
but within the normal range[4].
Whether the ovary morphology or HA can be considered as the necessary diagnostic
features for PCOS, there is a controversy. We still follows the 2003 Rotterdam Criteria[3] in
China. There are fewer researches exploring for whether there is an influence of PCOM on
clinical manifestations, endocrinology and metabolism. This is the purpose of this study.

Materials & Methods

Subject
PCOS patients were recruited in the Reproductive Medical Center of Peking Univer134

sity Third Hospital from Jan. 1, 2005 to Apr. 1, 2008 as PCOS group. The patients with
PCOS were divided into three groups according to the ovary morphology: B-PCOM group
(both two ovaries were PCOM), U-PCOM (only one ovary was PCOM) and N-PCOM
(none was PCOM). The diagnosis criteria of PCOS was according to the 2003 Rotterdam
Criteria[3]. This study was approved by the Ethics Committee of Peking University.
The diagnosis criteria of PCOM included 12 follicles or more in the both or one ovary
under B ultrasound screen, and the diameter was 2-9 mm, and / or ovary volume increases
(each ovary >10 ml, ovary volume = 0.15 multiplied by length, width and thickness of ovary
respectively).
Infertile women were recruited who were treated in Reproductive Medical Center of
Peking University Third Hospital for IVF-ET treatment with simple tubal factor as control
group. All patients in the control expressed normal menstruation or BBT biphasic and basic
endocrine determination without the organic disease in uterus and ovary.
All subjects did not have any hormone therapy or medicine treatment which could
affect endocrinology status. There was no statistical difference in age between the PCOS
groups and control group.
Obesity and overweight were determined according to Asia standard, the BMI of
obesity stage II, obesity stage I, overweight and normal weight is more than 30 kg/m2, 25.030.0 kg/m2, 23.1-24.9 kg/m2 and 18.5-23.0 kg/m2, respectively.
The diagnosis criteria of the metabolic syndrome was according to the China Diabetes
Association criteria [5], there were more than three items as follows: overweight and/or
obesity (BMI>25 kg/m2); higher glucose [fasting blood glucose (FBG)6.1 mmol/L and/or
2 h post glucose 7.8 mmol/L and/or diabetes mellitus which has been diagnosed]; hypertension (systolic/diastolic blood pressure140/90 mmHg and / or hypertension disease which
has been diagnosed); dyslipidemia [Triglyceride (TG) 17 mmol/L and/or high-density
lipoprotein (HDL)-C<1.0 mmol/L]; IR [our laboratory standards: fasting insulin (FINS)
8.9 mIU/L and/or 1 h/2 h post>100 mIU/L].
Clinical data collection
We collected detail data as follows: chief complaint, history of present illness, menstrual
history, obstetrical history, disease history, family disease history; height, weight, calculating
BMI, breast development, lactation, hirsutism, acne and pelvic examination; hormone level
determination, including estradiol (E2), luteinizing hormone (LH), follicle-stimulating hormone
(FSH), testosterone (T), androstenedione (A), FBG, FINS, total cholesterol (TC), TG,
low-density lipoprotein (LDL-C) and HDL-C; oral glucose tolerance test and insulin releasing
test; transvaginal ultrasound inspection, including ovary form and volume, number of follicles
in both ovary and endometrial thickness.
135

Oral glucose tolerance test and insulin releasing test

An oral glucose tolerance test (OGTT) was performed in the usual manner: blood
samples was drawn before the patients took 75 g of glucose in 250 ml water, and 2 more
samples were taken 60 min (G60) and 120 min (G120) later. Then the serum levels of glucose
and insulin were examed at each time point respectively.
Statistical analysis
Data were analyzed using SPSS software for Windows, version 15.0 (SPSS, Chicago,
Illinois, USA). Data were expressed as mean standard deviation (x- s) and compared
by using analysis of Variance, Chi-square test and Logisitic regression. P<0.05 was considered
significant.

Results
A total of 234 PCOS patients were recruited as PCOS group. There were 112 patients
in B-PCOM group, 50 patients in U-PCOM group and 72 patients in N-PCOM group. A
total of 39 patients were recruited as control group. There was no significant difference in
patients age between PCOS group [30.1 3.6 (22-40) years old] and the control [30.2
6.4 (23-37) years old] (P>0.05).
Baseline characteristics comparison between PCOS group and control group
Patients in B-PCOM group had higher BMI, longer menstrual cycle and thicker
endometrium than those in control group (P<0.05). Compared with the control, the BMI of
patients in B-PCOM and U-PCOM significantly increased (P<0.01, P<0.05), but there was
no significant difference either between the patients in N-PCOM and control group, or among
the three PCOS subgroups. Menstrual cycle was significantly longer in the patients of three
PCOS subgroups than that of control group (P<0.01 in B-PCOM and N-PCOM; P<0.05 in
B-PCOM and U-PCOM). Endometrial thickness was significantly thicker in B-PCOM
group than in control group (P<0.01); significantly thinner in N-PCOM group than in BPCOM group (P<0.01). The endometrial thickness in B-PCOM group was significantly
thicker than that in N-PCOM group and U-PCOM group, no difference between N-PCOM
group and control group. BMI, age and the length of menstrual cycle were not different
among the three PCOS subgroups (Table 1).
The incidence of obesity, IR, hyperlipidemia, metabolic syndrome and type 2 diabetes
mellitus in the three PCOS subgroups increased significantly compared with the control, and
there was no significant difference among the three PCOS subgroups (Table 2). Table 3
explains the method to assess insulin resistance degree.
The comparison of serum level of sex hormone between PCOS and control groups
Compared with control group, the levels of T, A, FSH, LH and the ratio of LH/FSH in
136

Table 1 Baseline characteristics of the 234 subjects enrolled for analysis (x- s)
Indicator

B-PCOM (n=112)

Age (year)

PCOS group (n=234)

U-PCOM (n=50)
N-PCOM (n=72)
30.6 3.4

29.2 3.3*

BMI (kg/m )

26.60 10.90

Menstrual cycle (d)

118.5 60.6

Endometrial thickness (cm)

**

26.40 4.60

85.0 59.5*

**

0.64 0.19
0.70 0.20
*: P<0.05, **: P<0.01, compared with control group
**

Control (n=39)

31.5 4.0

31.8 4.3

25.5 9.1

21.3 3.16

112.9 50.8**

34.8 14.8

0.61 0.30

0.59 0.30

Table 2 Comparison of obesity, IR, HA, type 2 diabetes mellitus and metabolic syndrome incidence rate
among subjects (%)
PCOS group (n=234)
Control (n=39)
B-PCOM (n=112) U-PCOM (n=50) N-PCOM (n=72)

Indicator
Obesity stage I

37.5 (19/50)**

38.7 (28/72)**

0.0 (0/39)

8.9 (10/112)

10.0 (5/50)

11.5 (8/72)

7.6 (3/39)

17.2 (19/112)

10.0 (5/50)

20.8 (15/72)

20.5 (8/39)

32.1 (36/112)**

(BMI 25.0-30.0 kg/m )

2

Obesity stage II (BMI>30.0 kg/m 2)

Over-weight
(BMI 23.1-24.9 kg/m2)
Normal (BMI 18.5-23.0 kg/m2)

41.9 (47/112)

IR

23.2 (26/112)

HA

18.75 (21/112)**

42.0 (21/50)

29.2 (21/72)

71.8 (28/39)

18.0 (9/50)

**

37.5 (27/72)

2.6 (1/39)

22.0 (11/50)**

25.0 (18/72)**

0.0 (0/39)

9.82 (11/112)

8.0 (4/50)**

13.8 (10/72)**

0.0 (0/39)

3.57 (4/112)**
**: P<0.01, compared with control group

6.0 (5/50)**

11.1 (8/72)**

0.0 (0/39)

**

Metabolic syndrome

**

Type 2 diabetes

**

Table 3 Explain method to assess IR degree

Item
Calculation methods
HOMA-IR FBG FINS/22.5
HOMA- 20 FINS/(FBG-3.5)

Assess IR degree

Significance

I60/G60

I60-FINS/G 60-FBG

The early secretion index of Islet, assess Islet -cell function and potential

ISI

ISI=1/(FBG FINS)

Fasting insulin sensitivity index, said body insulin sensitivity

AUC GLU

(FBG+G120)/2+G60

AUC INS

(FINS+I120)/2+I60

DI
I=insulin

(I60G60)/HOMA-IR

Assess Islet -cell function

Glucose disposition index

the three PCOS subgroups increased significantly (P<0.01). Among the three PCOS subgroups,
the levels of T and A were higher in N-PCOM group than in other two groups (P<0.05)
(Table 4).
The comparison of the serum level of glucose, insulin and lipid spectrum in patients
between PCOS group and control group
The levels of FINS, I60, I120, HOMA-IR and AUCINS were significantly higher in the
137

Table 4 Comparison of serum hormone among subjects (x- s)

Indicator

B-PCOM (n=112)

T (nmol/L)

PCOS group (n=234)

U-PCOM (n=50)
2.41 1.73*#

1.87 2.01*#

N-PCOM (n=72)
3.91 1.10*

Control (n=39)
0.88 0.58

A (nmol/L)

10.74 5.02

LH (IU/L)

17.10 3.10*

7.61 6.15*$

7.07 5.33*$

2.85 1.71

FSH (IU/L)

11.20 4.41

6.94 5.98

6.98 2.33

7.62 1.32

1.6 0.6*#
*: P<0.01, compared with control group

1.7 0.8*#

10.16 4.33

*#

LH/FSH

*#

14.05 4.13

5.96 2.73

0.32 0.25
0.98 0.7*
#: P<0.05, compared with N-PCOM group

$: P<0.05, compared with B-PCOM group

three PCOS subgroups than in the control, HOMA- and DI decreased significantly
(both P<0.01). FBG in N-PCOM group was higher than that in the control (P<0.05). There
was no significant difference in G60, G120 and AUCGLU in patients of PCOS subgroups compared with the control. Among the three PCOS subgroups, the levels of HOMA-IR and
FBG were significantly higher, the levels of HOMA-, I60/G60 and DI were significantly
lower in N-PCOM group than in the other groups (P<0.05); but the average value was in the
normal range (Table 5).

Table 5 Glucose metabolism comparision in subjects (x- s)

Indicator

B-PCOM (n=112)

FI (mIU/L)

9.82 8.92**

I60 (mIU/L)

156.25 58.32

I120 (mIU/L)

PCOS group (n=234)

U-PCOM (n=50)

N-PCOM (n=72)

Control (n=39)

8.81 7.82**

11.2 7.08**

8.06 6.21

**

183.28 52.15**

206.12 72.34**

48.36 21.43

93.18 39.47**

94.45 63.31**

102.31 59.23**

20.57 12.46

FBG (mmol/L)

4.74 1.40

4.97 0.76

5.13 1.21*

4.54 0.11

G60 (mmol/L)

8.48 3.12

9.01 3.29

8.58 3.06

7.95 2.09

G120 (mmol/L)

7.32 2.96

7.76 2.63

7.51 3.13

6.72 1.98

HOMA-IR

3.25 1.58**

2.82 1.91**

4.04 1.91**

HOMA-

182.25 130.30**

150.4 131.10

I60 -FI/G 60 -FG

ISI

140.52 129.30

2.07 1.26
**

287.85 114.60

49.12 57.38

22.29 10.89*

21.30 23.50*

44.23 24.39

0.70 0.57*

0.57 0.61

0.62 0.72*

0.42 0.40

AUCGLU (mmol/Lmin)

13.05 6.45

AUCINS (mIU/Lmin)

149.5 23.03**

DI

**

14.39 5.9
160.84 30.3**

11.0 11.75**
9.70 6.65**
*: P<0.05, **: P<0.01, compared with control group

14.16 4.27

13.58 2.87

135.14 93.7**

265.70 68.35

7.62 12.55**

32.21 15.86

Compared with the patients in control group, the level of TC was elevated significantly (P<0.05), the level of HDL-C decreased significantly in the three PCOS subgroups,
and the levels of HDL-C in N-PCOM group decreased more obviously (P<0.01). TG in N138

PCOM group was higher than that in control group (P<0.05). Among the three PCOS
subgroups, abnormal lipid metabolism in N-PCOM group was more obvious, particularly
HDL-C was significantly lower than that in the other groups (Table 6).
Related factors analysis of PCOM
Correlation analysis of related factors with PCOM
Oligomenorrhea and endometrial thickness were significantly positive correlated with
PCOM (P=0.04, OR=3.38; P=0.02, OR=7.84), the incidence of PCOM in PCOS with
oligomenorrhea was 3.38 times more than without oligomenorrhea. The incidence of thicker
endometrium was easier to happen in patients with PCOM of PCOS. Age was a protective
factor (P=0.01, OR=0.85)(Table 7).
Correlation analysis of hormone level with PCOM
The ratio of LH/FSH was positive correlated with PCOM (P=0.03, OR=3.25), the
higher the LH/FSH value is, the more possibility of PCOM in PCOS is, the incidence was
3.25 times of which with the lower LH/FSH value (Table 8).

Table 6 Lipid metabolism in subjects (x- s)

B-PCOM (n=112)

PCOS group (n=234)

U-PCOM (n=50)

TC (mmol/L)

4.88 0.89*

5.01 0.68*

4.82 0.84*

4.09 0.44

TG (mmol/L)

1.69 1.19

1.63 0.91

1.86 1.73*

1.30 0.66

HDL (mmol/L)

1.85 0.78*

Indicator

1.3 0.65*

LDL (mmol/L)

1.45 0.39
1.48 0.65
*: P<0.05, **: P<0.01, compared with control group

N-PCOM (n=72)

Control (n=39)

0.95 0.79**

2.21 0.37

1.48 0.48

1.13 0.30

Table 7 Correlation of Baseline characteristics with PCOM in PCOS

Indicator
Oligomenorrhea

OR

95%CI (minimum-maximum)

1.22

0.04

3.38

1.03-11.01

Amenorrhea

-0.19

0.57

0.83

0.43-1.59

Age

-0.15

0.01

0.85

0.79-0.94

BMI

0.04

0.29

1.04

0.96-1.13

Endometrial

2.06

0.02

7.84

1.34-45.89

Table 8 Correlation of hormone level with PCOM in PCOS

Indicator
T

r
0.18

OR

0.17

1.19

0.93-1.53

0.05

0.28

1.05

0.96-1.14

-0.12

0.10

0.88

0.76-1.02

1.18

0.03

3.25

1.11-9.50

LH
LH/FSH

95%CI (minimum-maximum)

139

Correlation analysis of glucose and lipid metabolism level with PCOM

FINS, AUCGLU were positive correlated with PCOM (P=0.03, OR=1.24; P=0.02,
OR=1.13), HOMA-IR was negative correlated with PCOM (P=0.02, OR=0.37), it indicates
that the more serious IR is, the lower incidence of PCOM is (Table 9).

Table 9 Correlation of glucose and lipid metabolism level and PCOM in PCOS
Indicator
FI

r
0.22

P
0.03

OR
1.24

95%CI (minimum-maximum)
1.03-1.50

I 60

0.03

0.20

1.03

0.99-1.08

I120

-0.06

0.55

0.94

0.76-1.16

FG

0.57

0.23

1.77

0.70-4.45

G 60
HOMA-IR
HOMA-

0.04

0.92

1.04

0.52-2.07

-1.00

0.02

0.37

0.16-0.85

0.00

0.94

1.00

1.00-1.01

I60 -FI/G 60 -FG

-0.02

0.49

0.98

0.93-1.03

ISI

-0.29

0.54

0.75

0.31-1.86

AUC GLU

0.13

0.02

1.13

1.02-1.26

DI

0.03

0.76

0.87

0.74-0.98

TC

0.07

0.91

1.08

0.30-3.89

TG

-0.12

0.24

0.89

0.72-1.08

HDL

0.24

0.21

1.27

0.88-1.84

LDL

-0.24

0.50

0.79

0.40-1.56

Discussion
The characteristics of PCOS include: abnormal ovarian function, increased follicle
number and ovarian hyperfunction which is the inherent feature, IR which is the common
pathophysiological mechanisms of abnormal ovarian function and glucose metabolic
abnormalities; insulin sensitizers could improve IR and ovarian function abnormalities[6].
The follicle development abnormality can lead to PCOM. There are influences for HA,
hyperinsulinemia and IR on the follicle development respectively. But PCOM is not
necessary for the diagnosis of PCOS. In the 2003 Rotterdam Criteria[3], PCOM is one of the
diagnostic criteria. But many scholars consider that HA is the basic pathogen of PCOS,
PCOM is also found in other diseases, particularly in functional hypothalamic anovulation
(FHA). In 2006, Androgen Excess Society (AES) made HA as one of necessary diagnostic
criteria, however, it is not widely recognized. Many studies[7,8] showed that simple PCOM may
be the early manifestation of PCOS, there is a tendency of varity on the hormone level in
these patients, which showed that ovarian morphology may affect on endocrine metabolism.
140

There are only several studies on ovarian morphology.

Our study showed that BMI of B-PCOM and U-PCOM increased significantly, the
incidence of stage I obesity in the three PCOS subgroups increased significantly compared
with control group, which is in according with the conclusion of previous studies. But there
was no difference in obesity incidence among PCOS patients with different ovarian morphology.
More than 50% PCOS patients are obese. IR and hyperinsulinemia are more severe in
PCOS with obesity than those without. Studies showed that losing weight could improve
ovulation and HA. Losing 5%- 10% of weight and reducing abdominal fat could improve
menstrual situation, ovulation, infertility and serum hormone level[9]. In this study, the duration
of menstrual cycle was significantly longer in the three PCOS subgroups than in the control,
but there was no significant difference in the three PCOS subgroups, which showed that
PCOM was not an important factor affecting the menstrual cycle.
Endometrial thickness was thicker in B-PCOM subgroup than in control group, and
thinner in N-PCOM group than in B-PCOM group, the difference was significant. Endometrial thickness was mainly affected by estrogen level, the traditional opinion was that follicles
of PCOS were atrophic follicles, but the research on the follicular fluid test and estradiol
synthesis in follicle granular cell showed that the function of synthesis for steroid hormones
in PCOS patients was higher than in normal female[10]. Compared with normal female, there
were higher serum level of estrodiol, more follicles and higher rate of ovarian hyperstimulation
syndrome (OHSS) in PCOS patients when ovulation induction was done using the same
amount of gonadotropin. That means the ovaries of PCOS patients are in high functional
status which estrogen synthetic is hyperfunctional and there are more follicles in basic state,
which is similarly to OHSS in ovulation induction. We could take PCOS as a basic OHSS.
Endometrial thickness was significantly thicker in B-PCOM group than in N-PCOM group
(P<0.01), which indicated that PCOM may be associated with increased levels of estrogen
and so as to estrogen dependent tumor in PCOS patients, which was a speculation by clinical
phenomenon, but it need further research and clinical trial.
Compared with the control, the serum levels of T, A, LH in the three PCOS subgroups
significantly increased (P<0.01). Among them, the serum levels of T and A were higher in NPCOM group than in the other two subgroups (P<0.05). Androgen plays a very important
role in follicular growth and development, but the influence of androgen on follicular development is like a two-edged sword. Low dose of androgen can promote follicular raising, make
more follicles from the reserve pool into growth pool, and promote theca cells and granular
cells proliferation through the receptors in the pre-antral follicles and small antral follicles,
and reduce follicle apoptosis and atresia. If androgen level is too high, there will be the side
effect, in some diseases with HA such as PCOS, high level androgen can inhibit follicular
141

selective growth, and induce follicle apoptosis and atresia, ultimately lead to ovulation dysfunction[11]. Rice et al.[12] confirmed that there was androgen receptor expression in the thecal
cell of ovary when the follicle began to grow, and the expression abundance increased with
follicular development. Study of Laven showed that serum level of anti-Mllerian hormone
(AMH) in PCOS correlated positively with follicle numbers whose diameter is 2-5 mm[13]. In
this study, serum level of androgen was statistically higher in PCOS subgroups than in the
control, and the level of androgen in N-PCOM group was higher than that in B-PCOM and
U-PCOM groups. The reason was perhaps that high androgen level inhibit follicular selective
growth, caused follicular apoptosis and atresia, eventually made disorders during the time of
follicle growth and ovulation, made the number of follicles less in N-PCOM patients than in
other PCOS patients who reached the standard of PCOM. However, how the high level of
androgen suppresses the follicular selective growth, the specific mechanism is not very clear.
Small follicles collected by androgen excess inhibit follicle to grow continuely, through the
follicle interaction to promote granular cell secreting AMH[12].
There are important effects for hyperinsulinemia and IR on the follicle development,
which is recognized widely as one of factors playing a criticial role in PCOS pathogenesis.
When insulin combine insulin receptors which located in theca cells, follicular raising level
can increase, and steroid hormone synthesis can promote in theca cells and granular cell[14].
Through 17- hydroxylase activation, insulin stimulates theca cells to produce androgen,
enhances androgen generation by LH and insulin like growth factor-1 (IGF-1) action, meanwhile reduces liver sex hormone binding globulin (SHGB) level, increases serum free
testosterone level, suppresses IGF-1 binding protein and improves the biological activity of
IGF-1. Insulin can also improve LH receptor regulation level which FSH induced in granulosa
cells, and promote progesterone reactivity on LH. The regulation disorder of signaling
pathway of the insulin receptor appeared in PCOS patients, leading to insulin sensitivity
damaged, and mediating glucose absorption rate decreased, but did not affect the steroid
hormone synthesis. So PCOS patients with IR, whether obese or not, all exhibit damaged
glucose-insulin internal environment and increased synthesis of ovarian steroid hormone. We
could deduce that the hyperinsulinemia status in PCOS patients with IR can stimulate
androgen producing, and induce granulosa cells prematurely luteinizing, lead to stagnation of
follicle proliferation and growth[15]. Some studies have found that the amount of insulin existing
in human follicle was decided by BMI and fasting serum insulin level[16]. Our study showed
that HOMA-IR was significantly higher in N-PCOM group, HOMA-, I60 /G60 and DI
were significantly lower in N-PCOM group than in the B-PCOM and U-PCOM groups. It
showed that the patients with N-PCOM had more serious insulin resistance. HOMA-,
I60 /G60 and DI were evaluation index of insulin -cell function, insulin -cell secretion
142

function deficiency may be one of the important reasons of IR in PCOS patients, particularly
in PCOS patients without PCOM. The study shows that compared to control group, TC is
significantly elevated, HDL-C was decreased significantly in the three PCOS subgroups,
and more obviously in N-PCOM group. TG in N-PCOM group was increased compared
with in the control. Among the three PCOS subgroups, abnormal lipid metabolism in NPCOM group was more obviously, particularly HDL-C. IR is also an important reason of
abnormal lipid metabolism in PCOS patients, because of IR, TG in the adipose tissue
decomposites strongly, makes overload of free fatty acid, strengthens the synthesis of
glycogen, TG biosynthesis and very low density lipoprotein cholesterol (VLDL-C) secretion
in the liver. Conversely, following VLDL-C arising in cycle, HDL-C reduces and LDL-C
particles increases, increasing the load of free fatty acid could also lead to IR of surrounding
tissue[17].
It is generally believed that there are intrinsically connection between IR and HA, but
which is exactly the first effect on PCOS is still not clear [18]. The possible mechanisms
through which hyperinsulinemia strengthens androgen synthesis are as follows: 1) the
insulin directly stimulates the pituitary to secret LH and affects the ovarian function, makes
ovarian theca cells proliferation, increases 17 alpha-hydroxylase/17,20-lyase (CYP-17)
enzyme activity and androgen synthesis excessiveness; 2) hyperinsulinemia indirectly
promotes ovarian androgen production, through its influence on the hypothalamus gonadotropin
effect; 3) high level insulin in the serum can promote IGF-1 synthesis of ovarian stromal cell
in patients with PCOS, make androgen synthesis increase; 4) high level insulin makes the
reduction of synthesis and secretion of SHBG in the liver, drops of the serum SHBG level,
increases the serum level of free testosterone (FT), makes the androgen exploitation degree
increasing[18].
In this study, PCOS patients without PCOM had higher androgen level and more severe
IR, to a certain extent, that embodies the relationship between HA and IR. Hyperinsulinemia
and IR increase serum FT level, the high level androgen suppresses the follicular selective
growth which is mentioned previously, HA, hyperinsulinemia and IR may interact with ovarian
morphology in PCOS through affecting follicular development. So in clinical practise, we
should pay more attention to PCOS patients without PCOM, who may be have serious
hyperinsulinemia or HA. We should know the HA and hyperinsulinemia are a glory and
honor, which both can lead to serious clinical consequences.
In the related factors Logistic regression analysis of PCOM, we find that there are
significant correlations between oligomenorrhea, endometrial thickness and PCOM, positive
relationships between LH/FSH and PCOM, FINS, AUCGLU and PCOM, negative relationship between HOMA-IR and PCOM. All these indicate that the more serious IR is, the lower
143

incidence of PCOM is. Results in our study show that IR and HA may result in the follicle
proliferation and growth stagnation, and affect follicular development. So in clinical practice,
HA and hyperinsulinemia should be fully treated before ovulation induction, to avoid OHSS
and ovulation induction failure, especially in the COH process.

Conclusion
In patients with PCOS, obesity, type 2 diabetes, metabolic syndrome, IR, and hyperlipidemia
incidence are significantly higher than those in the normal population. There are higher
androgen levels and IR in PCOS patients without PCOM. Therefore, HA and hyperinsulinemia
in these patients should been fully treated before ovulation induction in order to avoid OHSS.
HA, hyperinsulinemia and insulin resistance may interact with ovarian morphology in PCOS
through affecting follicular development.
In short, pathogen of PCOS is the interaction of multiple factors. PCOM under the
ultrasound may be the early status of PCOS. HA and IR are more severe in PCOS without
PCOM than with PCOM, accompanied by more obvious abnormalities in lipid metabolism,
which indicates more long-term complications such as DM, endometrial cancer, hypertension,
coronary heart disease, etc., need to be paid more clinical attentions and aggressive treatments.

References
1.

Goodarzi MO, Azziz R. Diagnosis, epidemiology, and genetics of the polycystic ovary syndrome. Best Pract
Res Clin Endocrinol Metab, 2006, 20(2):193-205.

2.

Zawdaki J, Dunaif A. Diagnostic criteria for polycystic ovary syndrome: towards a rational approach./Dunaif
A, Givens J, Haseltine F, et al. (eds). Polycystic Ovary Syndrome. Current Issues in Endocrinology and
Metabolism, Vol. 4. Boston: Blackwell Scientific, 1992:377-84.

3. The Rotterdam Eshre Asrm-sponsored PCOS consensus Workshop Group. Revised 2003 consensus on
diagnostic criteria and long-term health risks related to polycystic ovary syndrome (PCOS). Hum Reprod,
2004, 19(1):41-7.
4. Azziz R, Carmina E, Dewailly D, et al. Position statement: criteria for defining polycystic ovary syndrome
as a predominantly hyperandrogenic syndrome: an androgen excess society guideline. J Clin Endocrinol Metab,
2006, 91(11):4237-45.
5.

Chinese Medical Association Diabetes Branch Metabolic Syndrome Collaborative Group. Recommendations
on metabolic syndrome. Chin J Diabetes (in Chinese), 2004, 12(3):156-61.

6.

Practice Committee of the American Society for Reproductive Medicine. Use of insulin-sensitizing agents in
the treatment of polycystic ovary syndrome. Fertil Steril, 2008, 90(5 Suppl):S69-73.

7. Shi YH, Gao X, Sun XD, et al. Clinical and metabolic characteristics of polycystic ovary syndrome without
polycystic ovary: a pilot study on Chinese women. Fertil Steril, 2008, 90(4):1139-43.
8. Ya LQ, Kuang JQ, Feng YJ, et al. Characterization of unilateral polycystic ovary compared with polycystic
ovary syndrome. Fertil Steril, 2006, 85(3):805-8.
144

9. Norman RJ, Noakes M, Wu R, et al. Improving reproductive performance in over weight/obese women with
effective weight management. Hum Reprod Update, 2004, 10(3):267-80.
10. Jonard S, Dewailly D. The follicular excess in polycystic ovaries, due to intra-ovarian hyperandrogenism, may
be the main culprit for the follicular arrest. Hum Reprod Update, 2004, 10(2):107-17.
11. Li Y, Liang XY, Zhuang GL. The dualism effect of androgen on follicular development. J Reprod Med (in
Chinese), 2008, 17(6):423-4.
12. Rice S, Ojha K, Whitehead S, et al. Stage specific expression of androgen receptor, follicle stimulating hormone
receptor, and anti-Mllerian hormone type II receptor in single, isolated, human preantral follicles: relevance
to polycystic ovaries. Clinical Endocrinol Metab, 2007, 92(3):1034-40.
13. Laven JS, Mulders AG, Visser JA, et al. Anti-Mllerian hormone serum concent rations in normoovulatory
and anovulatory women of reproductive age. Clin Endocrinol Metab, 2004, 89(1):318-23.
14. Balen AH, Conway GS, Homburg R, et al. Polycystic Ovary Syndrome. A Guide to Clinical Management.
London, UK: Taylor &Francis, 2005:47-67.
15. Wu XK, Zhou SY, Liu JX, et al. Selective ovary resistance to insulin signaling in women with polycystic
ovary syndrome. Fertil Steril, 2003, 80(4):954-65.
16. Phy JL, Conover CA, Abbott DH, et al. Insulin and messenger ribonucleic acid expression of insulin receptor
isoforms in ovarian follicles from nonhirsute ovulatory women and polycystic ovary syndrome patients. J Clin
Endocrinol Metab, 2004, 89(7):3561-6.
17. Essah PA, Nestler JE. Metabolic syndrome in women with polycystic ovary syndrome. Fertil Steril, 2006,
86(Suppl 1):S18-19.
18. Huang HF, Li MG. Polycystic ovary syndrome and hyperandrogenism. Chin J Pract Gynecol Obstet (in
Chinese), 2002, 18(11):647-9.

(Received on June 9, 2014)

145

Conference Information
2014 Annual Scientific Update in Urogynaecology
November 3rd to 4th United Kingdom / London Obstetrics / Gynecology
Contact: Royal College of Obstetricians and Gynaecologists
Phone: 011-44-20-7772-6245
E-mail: events@rcog.org.uk
Website: http://www.rcog.org.uk/events/annual-scientific-update-urogynaecology-1
MRCOG Final Preparation: OSCE
November 3rd to 7th United Kingdom / London Obstetrics / Gynecology
Contact: Royal College of Obstetricians and Gynaecologists
Phone: 011-44-20-7772-6245
E-mail: events@rcog.org.uk
Website: http://www.rcog.org.uk/events/mrcog-final-preparation-osce-0
Obstetrics & Gynaecology in South Africa
November 4th to 15th South Africa / Johannesburg Obstetrics / Gynecology
Contact: Jon Baines Tours
Phone: 011-44-20-7223-5618
Fax: 011-44-20-7228-7290
E-mail: info@jonbainestours.co.uk
Website: http://www.jonbainestours.co.uk/tour/medical-professional-tour?tour=35
22nd Annual Fall Conference on Issues in Womens Health
November 5th to 8th Hawaii / Maui Family Medicine, General Medicine, Internal Medicine, Obstetrics /
Gynecology
Contact: Symposia Medicus
Phone: 800-327-3161 or 925-969-1789
Fax: 925-969-1795
Website: http://symposiamedicus.org/Conferences.aspx
Gynaecological Cancers
November 5th United Kingdom / London Obstetrics / Gynecology, Oncology
Contact: Education and Conference Centre, The Royal Marsden
Phone: 011-44-20-7808-2921
E-mail: conferencecentre@rmh.nhs.uk
Website: http://www.royalmarsden.nhs.uk/education/education-conference-centre/study-days-conferences/pages/
2014-gynaecological-cancers.aspx

146