International Journal of Gynecology and Obstetrics 80 (2003) 3539

Article

Estrogen levels and estrogen receptors in patients with stress

urinary incontinence and pelvic organ prolapse
J.H. Langa,*, L. Zhua, Z.J. Suna, J. Chenb
a

Department of Obstetrics and Gynecology, Peking Union Medical College Hospital, Beijing, China
b
Department of Pathology, Peking Union Medical College Hospital, Beijing, China
Received 2 April 2002; received in revised form 15 July 2002; accepted 17 July 2002

Abstract
Objectives: To investigate the histologic characteristics of tissues presumed to be the cause of urinary stress
incontinence and pelvic organ prolapse. Methods: Cardinal ligament and uterosacral ligament samples were obtained
from 73 women undergoing hysterectomy. The evaluation of estrogen receptors (ERs) by immunohistochemical
staining was semi-quantitative. Serum estrogen was determined by ELISA. Statistical analyses were performed by the
independent-sample t-test and one-way ANOVA. Results: Serum estradiol levels and ER values in the premenopausal
women with pelvic organ prolapse were significantly lower than in the control group (P-0.01). A positive correlation
was found between ERs and the number of postmenopausal years (P-0.01). ER values were similar in the cardinal
and uterosacral ligaments. Conclusions: Serum estrogen levels and ER values are significantly lower in the uterine
ligaments of premenopausal women with pelvic organ prolapse, and there was a positive correlation between ER
values in the uterine ligaments and the duration of postmenopausal years. Serum estrogen levels and ER values were
similar in the cardinal ligament and the uterosacral ligament.
2002 International Federation of Gynecology and Obstetrics. Published by Elsevier Science Ireland Ltd. All rights
reserved.
Keywords: Urinary stress incontinence; Pelvic organ prolapse; Estrogen; Estrogen receptor

1. Introduction
The incidences of pelvic organ prolapse (POP)
and urinary stress incontinence (USI) increase
significantly in postmenopausal women. It is reasonable to believe that estrogen deficiency might
be, at least in part, responsible for this condition.
Previous studies have identified estrogen receptors
(ERs) in the nuclei of connective tissue and the
smooth muscle cells of the bladder trigone, urethra,
*Corresponding author. Tel.: q31-317-483-223; fax: q8610-6512-4875.
E-mail address: zhu_julie@sina.com (J.H. Lang).

vaginal mucosa, levator ani stroma, and uterosacral

ligament w1,2x. The cardinal and uterosacral ligaments play an important role in supporting the
uterus in the normal position. They also support
the middle and upper third of the vagina. The aim
of this study was to investigate the relationship
between USI, POP and estrogen.
2. Materials and methods
2.1. Patients
A total of 73 women were recruited. They
included 12 premenopausal and 20 postmenopau-

0020-7292/03/$30.00 2002 International Federation of Gynecology and Obstetrics. Published by Elsevier Science Ireland Ltd.
All rights reserved.
PII: S 0 0 2 0 - 7 2 9 2 0 2 . 0 0 2 3 2 - 1

J.H. Lang et al. / International Journal of Gynecology and Obstetrics 80 (2003) 3539

36

sal women who formed the control group; 7

premenopausal and 18 postmenopausal women in
the POP group; and 16 postmenopausal women in
the USI group. Among the 7 POP premenopausal
women there was one first-degree case of uterine
prolapse with mild cystocele, and 6 second-degree
cases of uterine prolapse with moderate or severe
cystocele. Among the 18 POP postmenopausal
women there were 8 cases of first-degree uterine
prolapse with mild or moderate cystocele, 7 second-degree cases of uterine prolapse with moderate
or severe cystocele, and 3 third-degree cases of
uterine prolapse with severe cystocele. Among the
16 USI postmenopausal women, there were 10
cases of mild USI with first- to third-degree uterine
prolapse, and 6 cases of moderate USI with second- to third-degree uterine prolapse. Every patient
with USI also had POP. USI was defined by a
combination of medical history, gynecologic examination, stress test, Bonney test, and urodynamic
examination. Urge incontinence was excluded.
None of the women in the study had any hormonal
therapy before the study. The indications for hysterectomy were uterine prolapse or ovarian cysts.
None of women in the study showed any signs of
urinary infection or estrogen-related diseases
(endometriosis, myoma, or functional ovary
tumor). None of the women in the control group
showed any signs of POP or urinary incontinence.
2.2. Sampling and tissue preparation
A sample of 2 ml of venous blood was taken
just before hysterectomy to evaluate estrogen levels by enzyme immunoassay. The cardinal and
uterosacral ligament were identified during total
abdominal hysterectomy or total vaginal hysterectomy. A tissue sample of 500 mg was obtained
from the middle of both ligaments. Preperitoneal
fat was avoided in the biopsy. The samples were

obtained by four senior gynecologists. HE routine

staining and ER immunohistochemical staining
were performed for all the paraffin-embedded
specimens. For ER staining, after antigen retrieval
(microwave heating for 11 min) in 10 mmol of
citrate buffer (pH 5.56.0), the specimens were
incubated with monoclonal antibody (DAKO
Corp.) overnight at 4 8C, then with the second
antibody (DAKO Corp.) for 30 min at room
temperature. The DAKO antibody reacts with the
N-terminal domain (AyB region) of the receptor.
In immunoblotting, it reacts with the 67 kDa
polypeptide chain obtained by transformation of
E. coli and transfection of COS cells with plasmid
vectors expressing estrogen receptors. The antibody further reacts with the cytosolic extracts of
luteal endometrium and the human breast cancer
cell line MCF-7 w3x. The DAKO antibody reacts
with ER-a, as revealed using a DNA-binding array
w4x. DAB was used for coloration and hematoxylin
for counterstaining. A sample of breast carcinoma,
known to be positive for ER, was used as a
positive control. For negative controls, the PBS
buffer was used to replace ER monoclonal antibodies. ER staining was assessed in a semiquantitative manner. Using the H scoring system, two
observers assessed the ER staining intensity. Scores
were generated by adding together 3=% of strongly stained nuclei, 2=% of moderately stained
nuclei, and 1=% of weakly stained nuclei. The
observers were blinded to the specimen groups.
We scored more than 25 H-scope fields in each
section. When disagreements occurred between the
two observers, they were resolved using a doubleheaded microscope.
2.3. Statistics
Statistical analyses were performed by the independent-samples t-test, one-way ANOVA, and

Table 1
Comparison of two premenopausal sub-groups
Group

Cases (n)

Age (years)

Gravidity

Parity

BMI (kgym2)

Menstrual period
(follicularyluteal)

POP
Control
P-value

7
12

45.29"2.43
42.58"6.58
P)0.05

2.14"1.07
3.08"0.90
P)0.05

1.29"0.49
1.33"0.49
P)0.05

25.38"3.47
24.26"4.16
P)0.05

75%y25%
71%y29%

J.H. Lang et al. / International Journal of Gynecology and Obstetrics 80 (2003) 3539

37

Table 2
Comparison of three postmenopausal sub-groups
Group

Cases (n)

Age (years)

Gravidity

Parity

BMI (kgym2)

Postmenopause
(Years)

POP
USI
Control
P-value

18
16
20

64.22"5.41
63.81"7.09
62.10"6.41
P)0.05

4.72"1.67
4.19"1.83
4.30"1.69
Ps0.05

3.44"1.65
2.81"1.38
2.30"1.08
P)0.05

26.17"3.00
24.84"2.24
24.43"3.38
P)0.05

14.83"8.01
12.88"8.38
10.75"6.58
P)0.05

bivariate analysis. P-values less than 0.05 were

considered statistically significant.
3. Results
The two premenopausal sub-groups from the
control group and the POP group were comparable
in age, body mass index (BMI), gravidity and
parity (Table 1). There was no significant difference between the three postmenopausal sub-groups
in age, BMI, gravidity, parity, and the number of
postmenopausal years (Table 2).
Under light microscopy, the cardinal and uterosacral ligaments consisted of smooth muscle cells,
fibroblasts, fibers, blood vessels, and nerves. ER

immunohistochemical staining was performed for

all specimens. The positive nuclei for ER immunohistochemical staining were from muscle cells
and fibroblasts (Figs. 1 and 2). ER values in the
ligaments and serum estradiol levels are shown in
Table 3.
Both levels of serum estradiol and ER values in
the uterine ligaments were significantly lower in
the premenopausal women in the POP group than
in the premenopausal women in the control group
(P-0.01). There was no significant difference in
ER values among the postmenopausal USI, POP,
and control groups (P)0.05).
In the premenopausal POP and control groups,
ER values and serum estradiol levels had no

Fig. 1. ER positive staining in smooth muscle cell (=100).

J.H. Lang et al. / International Journal of Gynecology and Obstetrics 80 (2003) 3539

38

Fig. 2. ER positive staining in fibroblast (=100).

obvious correlation (P)0.05), but a positive correlation was found between the ER values in the
cardinal ligament and the length of postmenopausal years in the POP group (rs0.666; P-0.01)
and the USI group (rs0.536; P-0.05). The same
positive correlation was found for the uterosacral
ligament in the POP group (rs0.662; P-0.01)
and in the USI group (rs0.521; P-0.05).
ER values for the cardinal and uterosacral ligaments in the POP, USI, and control groups are
shown in Table 3. ER values for both ligaments

were significantly positively correlated in each

group (r)0.5; P-0.01).
4. Discussion
ERs were first reported by Shyamala and Gorski
in 1969. Much research on ERs has been done
since. Mosselman w3x found ER-b in 1996, and
traditional ERs were thus named ER-a. The antibody used in this study was for the antigen
determinant in the amino end of ER-a. ERs exist

Table 3
Serum estradiol and ER levels in uterine ligaments
Sub-groups

Groups

E2 (pgyml)

Cardinal
ligament (%)

Uterosacral
ligament (%)

Premenopause

Control
POP

73.83"52.72a
47.29"38.30a9

9.58"2.39b
4.21"0.80b9

11.64"2.68c
4.51"0.90c9

Postmenopause

Control
POP
USI

10.00"0.00
10.00"0.00
10.00"0.00

10.89"2.54e
11.85"2.89e9
9.64"5.01e0

13.09"3.03f
13.21"3.46f9
11.12"6.13f0

a: a9 P-0.01; b: b9 P-0.01; c: c9 P-0.01; e: e9: e0 P)0.05; f: f9: f0 P)0.05.

J.H. Lang et al. / International Journal of Gynecology and Obstetrics 80 (2003) 3539

in many organs other than the reproductive organs,

such as brain, mammary glands, skin, bones, bladder, urethra, and others. ER is a kind of intranuclear receptor, and ERs express different
characteristics in different tissues. However, the
ERs in the vagina are not influenced by menstrual
periods, and there was no obvious difference in
ER values before and after menopause. ERs in the
vagina are in a negative correlation with serum
estrogen w4x. This is the first time that the correlation of ERs in uterine ligaments and estrogen
are reported. ER values and serum estrogen levels
had no obvious correlation in premenopausal women, and there was no significant difference in the
ER values before and after menopause in the
control group. A significantly positive correlation
was found between ER values and the number of
postmenopausal years for postmenopausal women
in both the POP and USI groups. ER values in the
uterine ligaments rose as the patients number of
postmenopausal years increased. This might suggest that ER values are in a negative correlation
with serum estrogen. We therefore speculate that
ERs in the uterine ligaments are different from
those in the endometrium, that they are perhaps
similar to those in the vagina.
Although many studies have been done about
the effects of estrogen on USI, large double-blind,
placebo-controlled trials are few, the use and dosage of hormone therapy are not uniform, objective
criteria on cure and improvement are absent, and
the results are different w5x. The effect of hormone
replacement therapy (HRT) in the treatment of
USI and POP is not clear. The presence of ERs is
generally considered a prerequisite for a tissue to
recognize estrogen and to manifest a response to
hormonal therapy. ERs were detected in the female
pelvic floor muscles and urogenital ligaments by
Smith w6x. The findings provide a rationale for
estrogen treatment. From our data, the levels of
serum estradiol and ERs in the uterine ligaments
were significantly lower in the premenopausal
women in the POP group than those in the control
group. Therefore, deficiency in estrogen and ERs

39

may play a role in premenopausal USI and POP.

A significantly positive correlation was found
between the ER values and the number of postmenopausal years in postmenopausal women with
POP or USI. ER values in the uterine ligaments
rose as the patients number of postmenopausal
years increased. This suggests that HRT may be
useful in the treatment of USI.
The cardinal ligament is a condensation of
subserous fascia extending from the uterus to the
lateral pelvic wall. The uterosacral ligament is a
condensation of subserous fascia extending from
the sacrum around the rectum to the cervix. Both
ligaments consist of smooth muscle cells, fibroblasts, fibers, blood vessels, and nerve. The serum
estradiol levels and ER values of the two ligaments
were similar in each group. A correlation analysis
indicated a significant correlation between the
cardinal ligament and the uterosacral ligament,
which can be regarded as a whole. This supports
the theory of a uterosacralcardinal ligament
complex.
The etiology and pathogenesis of USI and POP
are complex, involving multiple factors such as
delivery injury. Much work remains to be done.
References
w1x Rechberger T, Donical H, Baranowski W, Jakowicki J.
Female urinary stress incontinence in terms of connective tissues biochemistry. Eur J Obstet Gynecol Reprod
Biol 1993;49:187 191.
w2x Franz H, Wendler D, Oettling G. Immunohistochemical
assessment of steroid hormone receptors in tissues of
the anal canal implications for anal incontinence. Acta
Obstet Gynecol Scand 1996;75:892 895.
w3x Mosselman S, Polman J, Dijkema R. Identification and
characterization of a novel human estrogen receptor.
FEBS Lett 1996;392:49 53.
w4x Perez-Lopez FR, Lopez CC, Alos L, Juste G, Ibanez

F,

Martinez-Hernandez
H. Estrogen and progesterone
receptors in the human vagina during the menstrual
cycle, pregnancy and postmenopause. Maturitas
1993;16:139 144.
w5x Klutke J, Bergman A. Hormonal influence on the
urinary tract. Urol Clin North Am 1994;22:629 639.
w6x Smith P, Heimer G, Norgren A, Ulmsten U. Steroid
hormone receptors in pelvic muscles and ligaments in
women. Gyncol Obstet Invest 1990;30:27 30.