DOI: 10.1111/1471-0528.

12020
Epidemiology
www.bjog.org

Prevalence and risk factors for pelvic organ

prolapse 20 years after childbirth: a national
cohort study in singleton primiparae after
vaginal or caesarean delivery
M Gyhagen,a M Bullarbo,a,b TF Nielsen,a,b I Milsoma
a
Department of Obstetrics and Gynaecology, Sahlgrenska Academy at Gothenburg University, Gothenburg, Sweden b Department of
Obstetrics and Gynecology, Södra Älvsborgs Hospital, Borås, Sweden
Correspondence: I Milsom, Department of Obstetrics and Gynaecology, Sahlgrenska University Hospital, SE-416 85 Gothenburg, Sweden.
Email ian.milsom@gu.se

Accepted 12 September 2012. Published Online 2 November 2012.

Objective To investigate prevalence and risk factors for
symptomatic pelvic organ prolapse (sPOP) and sPOP concomitant
with urinary incontinence (UI) in women 20 years after one
vaginal delivery or one caesarean delivery.
Design Registry-based national cohort study.
Setting Women who returned a postal questionnaire in 2008
(response rate 65.2%).
Population Singleton primiparae with a birth in 1985–88 and no
further births (n = 5236).
Methods The SWEPOP study used validated questionnaires about
sPOP and UI.
Main outcome measures Prevalence rate and risk of sPOP with or
without concomitant UI.
Results Prevalence of sPOP was higher after vaginal delivery
compared with caesarean section (14.6 versus 6.3%, odds ratio
[OR] 2.55; 95% confidence interval [95% CI] 1.98–3.28) but was
not increased after acute compared with elective caesarean section.
Episiotomy, vacuum extraction and second-degree or more
laceration were not associated with increased risk of sPOP
compared with spontaneous vaginal delivery. Symptomatic POP
increased 3% (OR 1.03; 95% CI 1.01–1.05) with each unit
increase of current BMI and by 3% (OR 1.03; 95% CI 1.02–1.05)
for each 100 g increase of infant birthweight. Mothers £160 cm
who delivered a child with birthweight ‡4000 g had a doubled
prevalence of sPOP compared with short mothers who delivered
an infant weighing <4000 g (24.2 versus 13.4%, OR 2.06; 95% CI
1.19–3.55). Women with sPOP had UI and UI > 10 years more
often than women without prolapse.
Conclusion The prevalence of sPOP was doubled after vaginal
delivery compared with caesarean section, two decades after one
birth. Infant birthweight and current BMI were risk factors for
sPOP after vaginal delivery.
Keywords Body mass index, caesarean section, epidemiology,
episiotomy, long-term risk factor, pelvic organ prolapse, perineal
laceration, questionnaire, urinary incontinence, vacuum extraction,
vaginal delivery.

Please cite this paper as: Gyhagen M, Bullarbo M, Nielsen T, Milsom I. Prevalence and risk factors for pelvic organ prolapse 20 years after childbirth: a
national cohort study in singleton primiparae after vaginal or caesarean delivery. BJOG 2013;120:152–160.

reported to be 11–19% in welfare states.6–8 Life expectancy

Introduction
Pelvic organ prolapse (POP) is considered to be one of the
major anatomical sequelae of childbirth.1 Globally up to
half of all parous women have some degree of clinical pro-
lapse and 10–20% are symptomatic (sPOP).1,2 In addition
to mechanical discomfort, POP may negatively affect sexu-
ality, body image and quality of life1,3,4 and is one of the
most common reasons for gynaecological surgery5 peaking
in upper midlife.6 Today the lifetime risk of POP surgery is
of women worldwide is increasing and women throughout
the world are living longer after childbirth and therefore
we must further increase our efforts to minimise possible
long-term morbidity due to childbirth through preventive
strategies based on a better knowledge of risk factors.
Pelvic organ prolapse is a rare condition in nulliparous
women and in women after one or several caesarean sec-
tions, indicating that mode of delivery is more important
than pregnancy alone.9,10 Obstetric trauma resulting from

152 ª 2012 The Authors BJOG An International Journal of Obstetrics and Gynaecology ª 2012 RCOG

the passage of the fetus through the pelvic floor during
vaginal delivery and increasing parity have been implicated
as important risk factors for the development of POP.1,11,12
The assessment of the influence of childbirth on POP later
in life has been hampered by the heterogeneity of study
populations. Women of different ages and varying body
weights have been included after a variable number of
pregnancies often with different modes of delivery.
The aim of the present study was to determine the prev-
alence of sPOP and concomitant urinary incontinence (UI)
in women 20 years after one single pregnancy terminating
either in a vaginal or a surgical delivery and to analyse the
relative importance of obstetric and non-obstetric risk
factors.
Methods
A national survey of pelvic floor dysfunction, the SWEPOP
(SWEdish Pregnancy, Obesity and Pelvic floor) study was
conducted in 2008 assessing pelvic floor function in women
20 years after one single pregnancy terminating either in a
vaginal or a surgical delivery. The population studied and
their obstetric data were obtained from the Swedish Medi-
cal Birth Register and complemented with information
from a postal questionnaire sent to the women 20 years
after the single delivery. A detailed description of the study
population and methods, including an analysis of the non-
responders, has been published previously.13
The questionnaire included 31 questions about current
height and weight, urinary or anal incontinence and genital
prolapse, menstrual status, hysterectomy, the menopause
and hormone treatment. Symptomatic prolapse was diag-
nosed according to a validated five-item questionnaire:14
sPOP was defined by the key symptom ‘feeling of a vaginal
bulge’ (often/sometimes/infrequently). This symptom was
sufficient for classification as having POP. In the absence
of this symptom, a combination of the symptoms ‘vaginal
pain/discomfort’ (often), ‘worsening upon heavy lifting’
and ‘need of manual reduction of the anterior vaginal wall’
(often/sometimes/infrequently), and urge incontinence
(often/sometimes/infrequently) could also lead to a POP
classification, but no women reported this combination,
and in practice, all had a feeling of a vaginal bulge in the
validation study performed by Tegerstedt et al.14 Urinary
incontinence was defined according to the International
Continence Society15 and by the question ‘Do you have
involuntary loss of urine?’ Participants reporting UI were
grouped according to the duration of UI (UI < 5 years,
5–10 years, or >10 years).
Body mass index (BMI; kg/m2) was categorised as normal
(<25), overweight (‡25–29.9) and obese (‡30) according to
the World Health Organization (WHO) classification16 and
was calculated for each woman according to weight and
ª 2012 The Authors BJOG An International Journal of Obstetrics and Gynaecology ª 2012 RCOG
Pelvic organ prolapse 20 years after childbirth
height measurements in early pregnancy at 8–10 weeks of
gestation (early pregnancy BMI), at delivery (delivery BMI),
and 20 years after delivery (current BMI). Perineal tears were
diagnosed by the midwife or obstetrician and classified into
four degrees according to the WHO International Classifica-
tion of Diseases. A first-degree tear involves the forchet, the
perineal skin, vaginal epithelium but not the underlying fas-
cia and muscles. A second-degree tear also involves the fas-
cia, muscles, perineal body but not the anal sphincter. A
third-degree tear involves the anal sphincter, but does not
extend through the rectal mucosa. A fourth-degree tear is
defined as extending through the rectal mucosa.17
Statistical analysis
Statistical analysis was performed with sas 9.1 (SAS Insti-
tute Inc., Cary, NC, USA). The influence of potential risk
factors for sPOP and sPOP concomitant with UI in
women 20 years after one delivery was compared according
to mode of delivery. Chi-square test was used to compare
categorical variables and Student’s t test was used to com-
pare continuous variables. Original data were sometimes
stratified in the analyses, e.g. when comparing women
delivered by the vaginal route with women delivered by
caesarean section. Stratification was performed according
to the following: Maternal age groups: <23 years; 23–
29 years; 30–34 years; ‡35 years; BMI: categories of BMI
were the same as the WHO classification; Infant head cir-
cumference: dichotomised according to £35 cm or >35 cm;
Infant birthweight: we chose the most commonly used
stratification in the literature (starting at <3000 g and with
intervals of 500 g); Maternal height and infant birthweight:
maternal height was dichotomised into £160 cm or
>160 cm and the birthweight was dichotomised into
<4000 g or ‡4000 g. Logistic regression analysis was used
to demonstrate independent risk factors for sPOP while
controlling for potential confounding factors. Potential risk
factors used in the analysis were mode of delivery, mater-
nal age at delivery, current BMI, hysterectomy, hormone
replacement therapy, gestational age, infant birthweight
and infant head circumference. Odds ratios (ORs) and
their 95% confidence intervals (95% CIs) were calculated
from the model. The Wald Odds Ratio test was used to
test for multiple parameters simultaneously. Adjusted prev-
alence rates and odds ratios were calculated using a covari-
ance analysis model for maternal age, infant birthweight,
current BMI and infant head circumference (only for the
subgroup analysis of the vaginal delivery group). These
variables were considered potential confounders and effect
modifiers on the basis of a combination of clinical consid-
erations and the significance of risk factors identified in
the logistic regression analysis. Because the regression mod-
els require that there be no missing data for the dependent
variable and all of the independent variables, the number
153
 Gyhagen et al.

of women included in each analysis varies. A P value of

<0.05 was considered statistically significant. Nonlinearity
and possible threshold effects of the stratified variables
(current BMI, infant birthweight and maternal age) were
analysed for vaginal delivery and caesarean section sepa-
rately. To address some specific obstetric events (vacuum
extraction, disproportion, episiotomy) associated with vagi-
nal delivery, subgroup analysis was performed and pre-
sented separately. The prevalence data permitted the
calculation of the number of caesarean sections needed to
avoid one case of sPOP using the number-needed-to-treat
principle. The number-needed-to-treat was calculated as
the inverse of the absolute risk reduction, where risk
reduction was the difference of adjusted sPOP prevalence
between vaginal delivery and caesarean section.
Results
The questionnaire was returned by 65.2% of 9423 women
invited to participate and the basic characteristics of those
included, grouped according to mode of delivery, have
been described in detail previously.13
The overall prevalence of sPOP was 12.8% (663/5199).
The number of women with sPOP in the vaginal delivery
group was 588/3995 and in the caesarean group was 75/
1204 (48/766 elective caesarean section and 27/438 acute
caesarean section; Table 1). The prevalence of sPOP was
more than doubled after vaginal delivery compared with
caesarean section (14.6 versus 6.3%, OR 2.55; 95% CI
1.98–3.28) (Table 1). One percent of the women delivered
vaginally had received treatment for pelvic organ prolapse.

Table 1. The difference in crude and adjusted prevalence rates of sPOP between vaginal delivery (VD) and caesarean section (CS) in relation to
stratified risk factors

CS (%) VD (%) Crude OR (95% CI) CS (%) VD (%) Adjusted* OR (95% CI)

All 6.2 14.7 2.60 (2.02–3.33) 6.3 14.6 2.55 (1.98–3.28)

Infant birthweight (g)
<3000 7.9 11.7 1.53 (0.95–2.47) 7.9 11.8 1.57 (0.97–2.56)
n = 808 n = 353 n = 455
3000–3499 5.4 14.1 2.89 (1.82–4.61) 5.0 14.2 3.17 (1.95–5.15)
n = 1698 n = 390 n = 1308
3500–3999 5.7 14.6 2.82 (1.61–4.95) 5.9 14.6 2.73 (1.55–4.80)
n = 1430 n = 245 n = 1185
4000–4499 4.3 16.1 4.27 (1.96–9.29) 3.7 15.8 4.85 (2.10–11.17)
n = 1078 n = 163 n = 915
‡4500 9.6 23.3 2.84 (1.04–7.81) 7.8 23.8 3.67 (1.22–11.03)
n = 181 n = 52 n = 129
Infant head circumference (cm)
£35 7.4 14.7 2.15 (1.60–2.88) 8.0 14.4 1.93 (1.45–2.57)
n = 3490 n = 756 n = 2734
>35 4.1 15.1 4.14 (2.45–7.00) 3.7 15.1 4.70 (2.70–8.21)
n = 1593 n = 390 n = 1203
Current BMI (kg/m2)
<25 4.7 12.2 2.80 (1.86–4.22) 5.1 12.1 2.57 (1.73–3.83)
n = 2477 n = 572 n = 1905
25–29 7.8 15.5 2.19 (1.44–3.31) 7.8 15.5 2.18 (1.44–3.30)
n = 1643 n = 361 n = 1282
‡30 7.4 19.4 3.03 (1.82–5.07) 7.4 19.4 3.04 (1.82–5.08)
n = 977 n = 245 n = 732
Maternal age at delivery (years)
<23 10.3 15.4 1.59 (0.85–2.99) 7.1 15.5 2.41 (1.02–5.70)
n = 869 n = 116 n = 753
23–29 6.8 12.9 2.02 (1.27–3.24) 8.5 12.9 1.59 (1.05–2.40)
n = 1735 n = 308 n = 1427
30–34 6.8 14.9 2.39 (1.53–3.73) 6.7 14.9 2.45 (1.55–3.86)
n = 1471 n = 351 n = 1120
‡35 4.2 17.3 4.76 (2.86–7.95) 4.6 17.3 4.39 (2.66–7.25)
n = 1124 n = 429 n = 695

Chi-square test was used for statistical comparison between groups.

*Adjusted for current BMI, infant birthweight and maternal age.

154 ª 2012 The Authors BJOG An International Journal of Obstetrics and Gynaecology ª 2012 RCOG

The corresponding figure for women delivered by caesarean
section was 0.1%.
The results indicated that 12 caesarean sections needed
to be performed to avoid one case of sPOP. There was
however no significant difference in the prevalence or odds
of sPOP after acute caesarean section compared with elec-
tive caesarean section 6.0 versus 6.3% (OR 0.95; 95% CI
0.58–1.57). The difference in crude and adjusted prevalence
rates of sPOP between vaginal delivery and caesarean sec-
tion in relation to stratified risk factors (infant birthweight,
head circumference, current BMI and maternal age at
delivery) are also shown in Table 1.
Concomitant UI and sPOP
The prevalence of UI (61.8 versus 34.8%, OR 3.02; 95% CI
2.54–3.59) was higher in women with sPOP and the preva-
lence of UI > 10 years more than doubled (16.7 versus
7.6%, OR 2.43; 95% CI 1.93–3.07) in women with sPOP
compared with those without sPOP. Vaginally delivered
women with sPOP had an almost tripled prevalence of UI
compared with women delivered by caesarean section with
sPOP (9.1 versus 2.8%, OR 3.48; 95% CI 2.41–4.99)
(Table 2). Furthermore, women who were delivered vagi-
nally and who had sPOP reported UI > 10 years five times
more often compared with women delivered by caesarean
section with sPOP (2.6 versus 0.5%, OR 5.22; 95% CI
2.29–11.92; Table 2). Of the 383 women with a combina-
tion of sPOP and UI, 351(91.6%) had a vaginal delivery
and of 110 women with the combination sPOP and
UI > 10 years 105 (95.5%) had a vaginal delivery. The
prevalence of either sPOP or UI or sPOP + UI combined
was 45.5% in women who had a vaginal delivery. The cor-
responding prevalence of either sPOP or UI or sPOP + UI
combined for women delivered by caesarean section was
33.2%. These results indicate that eight caesarean sections
Table 2. Crude and adjusted prevalence rates of sPOP and concomitant UI and concomitant UI persisting for more than 10 years (UI > 10 years)
according to mode of delivery
CS VD
(%) (%)
sPOP 6.2 14.7
n = 663 n = 75 n = 588
sPOP and UI 2.9 9.2
n = 383 n = 32 n = 351
sPOP and UI > 10 years 0.4 2.6
n = 110 n=5 n = 105
CS, caesarean section; VD, vaginal delivery.
n = 5159, missing n = 40.
Chi-square test was used for statistical comparison between groups.
*Adjusted for current weight, maternal age, infant birth weight and head circumference.
ª 2012 The Authors BJOG An International Journal of Obstetrics and Gynaecology ª 2012 RCOG
Pelvic organ prolapse 20 years after childbirth
have to be performed to avoid one case of either UI or
sPOP or the combination of UI and sPOP.
Risk factors for sPOP
Risk factors for sPOP were analysed using logistic regres-
sion analysis (Table 3). Vaginal delivery, infant birthweight
and current maternal BMI were shown to be risk factors
for sPOP. There were no effects demonstrated for gesta-
tional length, infant head circumference, or factors such as
hysterectomy or estrogen treatment (Table 3). Age was not
a risk factor for sPOP (OR 1.01; 95% CI 0.99–1.02).
The odds of sPOP increased 3% (OR 1.03; 95% CI
1.01–1.05) for each unit increase of BMI (Table 3). With
normal current BMI as reference the odds of sPOP
increased significantly for both overweight and obese
women in the vaginal delivery group (Table 4).
There was also a 3% increased odds of sPOP (OR 1.03;
95% CI 1.01–1.05) for each 100-g increase of infant birth-
weight (Table 3). Analysis of birthweight after vaginal
delivery showed a numerical trend towards higher rates of
prolapse with increasing birthweight, which became signif-
icantly higher for the birthweight group ‡4500 g com-
pared with birthweight group <3000 g. This effect of
birthweight was not observed after caesarean section
(Table 4).
The importance of infant birthweight in conjunction
with maternal height was also investigated. The combined
factor of birthweight and maternal height together was
found to be important for the development of sPOP after
vaginal delivery (Table 5). Mothers £160 cm who delivered
a child with birthweight ‡4000 g (n = 91) after vaginal
delivery had a higher prevalence of sPOP compared with
short mothers who delivered an infant weighing <4000 g
(n = 484; 24.2 versus 13.4%, OR 2.06; 95% CI 1.19–3.55)
after adjusting for maternal age, current BMI and infant
Crude CS VD Adjusted*
OR (95% CI) (%) (%) OR (95% CI)
2.60 (2.02–3.33) 6.3 14.6 2.55 (1.98–3.28)
3.36 (2.37–4.79) 2.8 9.1 3.48 (2.41–4.99)
6.47 (2.63–15.9) 0.5 2.6 5.22 (2.29–11.92)
155
 Gyhagen et al.

Table 3. Logistic regression of risk factors for sPOP including all the
cantly associated with an increased prevalence or odds of
variables of the full model (n = 4066) sPOP. The prevalence of sPOP after spontaneous vaginal
delivery was similar to the prevalence of sPOP after birth
OR (95% CI) P assisted by vacuum extraction, between women who had
an episiotomy compared with those without episiotomy
Vaginal delivery (yes vs no)* 2.36 (1.76–3.17) <0.001 and between women who had at least second degree lacera-
Age at delivery (years) 1.01 (0.99–1.02) 0.57
tions and those without or with a less than second-degree
Infant birthweight (0.1 kg) 1.03 (1.01–1.05) 0.02
lacerations (Table 6).
Infant head circumference 0.97 (0.92–1.03) 0.34
(£35 vs >35 cm)**
Gestational length (weeks) 0.99 (0.92–1.06) 0.71
Discussion
BMI current (kg/m2) 1.03 (1.01–1.05) <0.001
Hysterectomy (yes vs no)*** 0.75 (0.48–1.17) 0.21 The odds of sPOP 20 years after birth increased by 255%
Estrogen therapy (yes vs no)**** 1.36 (0.92–2.00) 0.13 after vaginal delivery compared with caesarean section.
References: categorical variables *caesarean section; **infant head There was however no difference in the prevalence or odds
circumference £35 cm; ***no hysterectomy; ****no oestrogen of sPOP after acute caesarean section compared with elec-
therapy. tive caesarean section. The BMI was a significant risk factor
for prolapse and for each unit increase of BMI the odds of
sPOP increased by 3%. Infant birthweight ‡4500 g signifi-
Table 4. Adjusted* odds ratio for sPOP in relation to stratified risk cantly increased the rate and odds of sPOP compared with
factors and according to mode of delivery infants <3000 g after vaginal delivery. This was not
observed after caesarean section. Shorter mothers <160 cm
Caesarean section Vaginal delivery
who delivered a child with infant birthweight ‡4000 g had
a doubled prevalence of sPOP compared with shorter
Current BMI (kg/m2)
<25 1.0 (ref) 1.0 (ref)
mothers <160 cm who delivered a child <4000 g. The prev-
25–29.9 1.70 (0.99–2.94) 1.33 (1.08–1.63) alence of UI almost doubled to over 60% in women with
‡30 1.60 (0.86–2.96) 1.74 (1.38–2.18) prolapse compared with those without and the combined
Age at delivery (years) risk factor of vaginal delivery and sPOP was even stronger
<23 1.0 (ref) 1.0 (ref) for UI that had persisted for more than 10 years. It is also
23–29 1.43 (0.53–3.83) 0.83 (0.63–1.10) important to note that we found no increase in the preva-
30–34 1.14 (0.42–3.10) 1.01 (0.75–1.34)
lence of sPOP after vacuum extraction or in women with
‡35 0.72 (0.26–2.00) 1.20 (0.88–1.63)
Infant birthweight (g)
vaginal tears of second degree or greater or an episiotomy
<3000 1.0 (ref) 1.0 (ref) compared with a normal vaginal delivery.
3000–3499 0.65 (0.36–1.18) 1.26 (0.91–1.75) The main strengths of this study are the high response
3500–3999 0.73 (0.38–1.43) 1.28 (0.92–1.77) rate and the large study population included according to
4000–4499 0.53 (0.23–1.25) 1.39 (0.99–1.95) strict inclusion criteria from the national cohort of single-
‡4500 0.96 (0.32–2.86) 2.09 (1.26–3.47) ton primiparae who delivered during 1985–1988. A detailed
*When applicable adjusted for current BMI, age at delivery and summary and discussion of the strengths and limitations of
infant birthweight. this study design using a national cohort of primiparous
women was presented in the paper evaluating urinary
incontinence.13 Some specific limitations of the present
head circumference (Table 5). This effect of birthweight investigation evaluating sPOP must also be considered.
was not observed for mothers >160 cm. The prevalence First, symptomatic women have been shown to be more
and odds of concomitant sPOP and UI was doubled and predisposed to participate in studies and therefore sPOP
the prevalence and odds of concomitant sPOP and might have been overestimated.18 Second, the symptoms of
UI > 10 years was tripled after giving birth to a child with POP were self-reported. However, several studies have
birthweight ‡4000 g in women of short stature (£160 cm). shown that self-reported symptoms are consistent19 and
The prevalence and odds of concomitant sPOP and valid when they exist at the time of report.20,21 Third, it
UI > 10 years was also doubled for mothers >160 cm who has been shown in a community-based population14 that
gave birth to a child with birthweight ‡4000 g (Table 5). sensitivity and specificity for the question ‘feeling a bulge’
No effects of maternal height and birthweight were seen in to correctly identify genital prolapse according to the
the caesarean section group. POPQ system were 66.5 and 94.2%, respectively. However,
None of the obstetric events, vacuum extraction, episiot- it should be noted that although every third woman with
omy, or second-degree or greater laceration was signifi- prolapse was a false negative, most of the missed cases

156 ª 2012 The Authors BJOG An International Journal of Obstetrics and Gynaecology ª 2012 RCOG
 Pelvic organ prolapse 20 years after childbirth

Table 5. Prevalence and odds ratio of sPOP, sPOP combined with UI, and sPOP combined with UI for more than 10 years (>10 years) in women
delivered vaginally grouped according to maternal height and fetal weight

Maternal height/ sPOP sPOP and UI sPOP and

infant birthweight UI >10 years

% OR (95% CI) % OR (95% CI) % OR (95% CI)

£160 cm/<4000 g (n = 484) vs 13.4 vs 24.2 2.06 (1.19–3.55) 8.4 vs 15.4 1.96 (1.02–3.78) 2.1 vs 6.6 3.35 (1.18–9.45)
£160 cm/‡4000 g (n = 91)
>160 cm/<4000 g (n = 2364) vs 12.3 vs 14.7 1.12 (0.90–1.38) 8.9 vs 9.5 1.08 (0.83–1.40) 2.1 vs 4.0 1.93 (1.25–2.98)
>160 cm/‡4000 g (n = 923)

Chi-square test was used for statistical comparison between groups. Prevalence and odds ratios were adjusted for current weight, maternal age,
head circumference.

Table 6. Adjusted* prevalence and odds ratio of sPOP in relation to

obstetric events in vaginally delivered women
(%) OR (CI 95%)
SVD vs VE 14.8 vs 14.0 0.94 (0.74–1.19)
(n = 3283/n = 712)
No episiotomy vs episiotomy 14.8 vs 13.7 0.92 (0.70–1.20)
(n = 3418/n = 498)
Less than second-degree 14.5 vs 16.9 1.20 (0.80–1.80)
tear vs second-degree
tear or greater
(n = 3744/n = 172)
SVD, spontaneous vaginal delivery; VE, vacuum extraction.
Chi-square test was used for statistical comparison between groups.
*Adjusted for current BMI, maternal age, infant birth weight and
infant head circumference.
(72.5%) were stage I prolapses only. In our study, as also
in the study of Tegerstedt et al.,14 all the women who
reported a combination of symptoms indicating sPOP in
fact also reported the single symptom ‘feeling a bulge’ indi-
cating the discriminatory importance of this single
symtom.
The reported wide range (15–48%) in the prevalence of
POP after childbirth is mainly a result of differences in
study populations and varying classification of POP.1,10,22
The influence of specific obstetric events is difficult to
interpret in the majority of studies because of the inclusion
of several pregnancies and information about the preva-
lence of sPOP after one single vaginal or surgical delivery is
scarce.23 In this study the odds of sPOP after vaginal deliv-
ery were more than double those after caesarean section,
which is in contrast to the findings of Lukacz et al.24 who
found no difference in prolapse rates between women
delivered by acute caesarean section and vaginal delivery.
We did not find any difference in the prevalence of sPOP
between women delivered by acute caesarean section or
elective caesarean section. We interpret our results to indi-
cate that it is not until the final stage of delivery when the
fetus passes through the pelvic hiatus during vaginal deliv-
ery (at the end of the second stage of delivery) that the
structural damage to the pelvic floor occurs causing POP.
This is also consistent with the findings on the occurrence
of UI in the SWEPOP study, which also did not show any
difference in the prevalence of UI between acute caesarean
section or elective caesarean section.13
Whether or not infant birthweight is a risk factor for
prolapse is still controversial. In the study of Tegerstedt
et al.10 birthweight was not a risk factor for sPOP after
adjustment for age and parity, but groups were probably
too small. Our results are however consistent with those of
two other studies indicating that birthweight is a risk fac-
tor. In the cross-sectional population study of Samuelsson
et al.25 a strong correlation was found between odds of
POP and maximum birthweight. Timonen et al.26 reported
that POP was found in more than one-third of women
who delivered a child ‡4000 g compared with 9.5% in the
general population. Similarly, the analysis of prolapse prev-
alence in relation to stratified infant birthweights in the
present study showed that sPOP more than doubled for
infants ‡4500 g in comparison with infants <3000 g after
vaginal delivery. In contrast this was not observed after cae-
sarean section. The logistic regression analysis confirmed a
3% risk increase in the prevalence of sPOP for each 100-g
increase of birthweight. However, from the analysis of
stratified infant birthweight the relationship did not seem
to be linear and there seemed to be a threshold effect for a
birthweight around 4500 g.
When analysing the effect of maternal height we found a
significant increased prevalence and odds of sPOP among
mothers who delivered an infant ‡4000 g compared with an
infant <4000 g after vaginal delivery in mothers £160 cm.
This infant-birthweight-dependent effect was not observed
for mothers >160 cm and no effect of maternal height and
birthweight was seen in the caesarean section group. Data

ª 2012 The Authors BJOG An International Journal of Obstetrics and Gynaecology ª 2012 RCOG 157
 Gyhagen et al.
in the literature on the effect of incongruity between mother
and child as a risk factor for sPOP are lacking for compari-
son. From our data it is obviously not possible to determine
whether high birthweight in combination with shorter
mother itself is causative for sPOP or is associated with
other underlying factors that predispose to pelvic floor
trauma. Other studies have shown that neuromuscular
damage is associated with prolonged second stage of labour
and extended voluntary straining,27,28 which may be linked
to disproportion. Our results suggest that maternal and
infant body characteristics may be of significance and
should perhaps also be taken into account at the time of
scheduled vaginal delivery to minimise pelvic floor damage.
Short-term follow-up magnetic resonance imaging stud-
ies have shown that levator ani muscle injury is found in
10% of women after spontaneous vaginal delivery, in 25%
after vacuum extraction and in 66% of women after for-
ceps delivery, but are not seen in nulliparous women.29
Opening of the genital hiatus and weakening of the pelvic
plate have also been shown to occur more often in women
with prolapse than in those with normal pelvic support.30
Risk for subsequent long-term development of POP that
could be attributed to specific obstetric events during vaginal
delivery is however poorly analysed in the literature. Teger-
stedt et al.10 found that episiotomy or instrumental delivery
(proportions of vacuum extraction and forceps were not
specified) was neither positively nor negatively associated
with sPOP, which is consistent with the outcome shown in
our study. Handa et al.23 showed a strong association
between operative vaginal delivery and POP despite a low
prevalence rate (3%) in their study. Tegerstedt et al.10 also
found that vaginal delivery in combination with ‘extensive
vaginal rupture’ more than doubled the odds of sPOP (in
connection with at least one delivery). The diagnosis how-
ever was based on patient recall. This is contrary to our find-
ing that a tear of second degree or more, after adjustment
for birthweight, maternal age, current BMI and infant head
circumference, was not significantly associated with sPOP.
Whether current BMI is a risk factor for POP or not is
still controversial. In our study current BMI was an inde-
pendent but, in comparison with mode of delivery, less
important risk factor with a weak effect on odds of pro-
lapse. By each unit increase of BMI sPOP increased 3%. A
positive association between BMI and odds of sPOP has
also been shown in the study by Miedel et al.12 and in the
Women’s Health Initiative.31 Even if the analysis techniques
used in these two studies differed from our study, the esti-
mated risk increase by each unit increase of BMI from the
stratified data in these two studies was approximately in
the same range as ours. However it should be noted that
studies by Rortveit et al.9 and Samuelsson et al.25 did not
establish BMI as a significant risk factor for sPOP and clin-
ical POP, respectively.
158 ª 2012 The Authors BJOG An International Journal of Obstetrics and Gynaecology ª 2012 RCOG
Advancing age is a widely accepted independent risk fac-
tor for POP.1 In two cross-sectional studies on sPOP, prev-
alence increased with greater age.31,32 Miedel et al.12 also
found a significant age-dependent (parity-adjusted) increase
of sPOP for women of all age groups ‡50 years (OR 2.0–
2.3) compared with women aged 30–39 years old, after one
or more vaginal and/or surgical deliveries. Contrary to
these findings we did not find age to be a significant risk
factor for sPOP, neither in the logistic regression analysis
nor in the covariance analysis of stratified age groups. One
explanation for differences in the reported results may be
related to population and time period bias. In our study
the women at the time of assessment in 2008 were aged
35–67 years and the interval after delivery had been fixed
to 20 years. Our findings may reflect a generally good
access to modern antenatal care and obstetric practices in
the 1980s, a more favourable health status among pregnant
women and the narrow age range of our cohort. Cross-sec-
tional studies have included elderly women who gave birth
earlier under perhaps less optimal birth conditions. In sup-
port of our results the Oxford FPA Study also concluded
that age was a much less important risk factor, once parity
and calendar period was taken into account.33
The observation that POP is associated with a higher fre-
quency of UI was first reported by Olsen et al.6 in 1985 and
was later confirmed in the EPIQ study34 and by Buchs-
baum.35 All three studies reported that about 60% of
women with POP are also diagnosed with UI. These find-
ings are very close to our results that 61.8% of all women
with sPOP also reported UI compared with 34.8% of
women with no sPOP. Further, it was found that in women
with sPOP, as opposed to women without sPOP, UI with a
duration of more than 10 years was more prevalent.
The combination of vaginal delivery and sPOP was a
strong risk factor for concomitant UI and even stronger for
women with UI of longer duration (>10 years). The preva-
lence of sPOP in our study was more than twice as common
in women who had undergone vaginal delivery compared
with caesarean section and in addition the prevalence of
concomitant UI was tripled after vaginal delivery compared
with caesarean section in women who had sPOP. It should
also be noted that the prevalence of UI > 10 year was more
than five times greater after vaginal delivery compared with
caesarean section in women who had sPOP. This indicates a
protective effect of caesarean section for the occurrence of
sPOP and the protective effect of caesarean section was even
greater for sPOP + UI and sPOP + UI > 10 year. It is rea-
sonable therefore to assume that women with concomitant
sPOP and UI and especially those with sPOP and
UI > 10 year more often have pelvic floor damage as a
result of vaginal delivery compared with women delivered
by caesarean section. It is also important to note that there
may be other differences in the pathogenesis of prolapse

occuring after caesarean section compared with vaginal
delivery (e.g. due to other non-obstetric factors).
In conclusion, the results from this study clearly demon-
strate that in primiparous women the single most impor-
tant long-term risk factor for genital prolapse is delivery
via the vaginal route. Taking into account the prevalence of
POP stated for nulliparous women in the literature, caesar-
ean section seems to be protective. The results indicated
that 12 caesarean sections needed to be performed to avoid
one case of sPOP and that eight caesarean sections have to
be performed to avoid one case of either UI or sPOP or
the combination of UI and sPOP. In mid- and upper mid-
life other long-term risk factors are much less important
although current BMI is a significant risk factor in com-
bination with a previous vaginal delivery. The obstetric
events episiotomy, vacuum extraction and tear of second-
degree or more were not found to be associated with an
increased risk of sPOP. In contrast, other parameters were
more important. Infant birthweight and the combined risk
factor maternal height and birthweight were stronger risk
factors for sPOP, but only in association with vaginal
delivery. There seemed to be a threshold effect of a birth-
weight around 4500 g and the combination of short
mother and large baby (‡4000 g) more than doubled sPOP
prevalence. Prolapse itself seemed to be a risk factor for UI
and UI > 10 years. Women with sPOP after vaginal deliv-
ery fared significantly worse than women with sPOP after
caesarean section.
In the western world the incidence of caesarean section
on maternal request for nonmedical reasons is steadily
increasing.36 Women and healthcare providers frequently
report prevention of pelvic floor dysfunction as one of the
main reasons for elective caesarean section.37 For the pro-
fession, giving priority to pelvic floor dysfunction alone
against all other maternal and neonatal outcomes, seems
inappropriate. Women who contemplate planned caesarean
section should be able to base their request on scientific
data concerning the long-term effects of vaginal delivery.
The State-of-the-Science Conference on Caesarean Delivery
on Maternal Request of the National Institutes of Health in
200638 could not find any high-quality data on pelvic floor
outcomes supporting either type of delivery. Our results
may contribute to a better informed and balanced evalua-
tion and discussion of vaginal versus surgical delivery.
Disclosure of interests
We declare that we have no conflict of interests.
Contribution to authorship
All authors were involved in the conception and design of
the study, acquisition of data and interpretation of the
results as well as the writing of the manuscript. All authors
approved the final version of the submitted manuscript.
ª 2012 The Authors BJOG An International Journal of Obstetrics and Gynaecology ª 2012 RCOG
Pelvic organ prolapse 20 years after childbirth
MG and IM take full responsibility for the integrity of the
data and the accuracy of the data analysis.
Details of ethics approval
Ethical approval for the SWEPOP study was obtained from
the Regional and the National Ethic Review Boards (the
Ethics Committee at Sahlgrenska Academy, Gothenburg
University, ref no.381-07, 13 August 2007 and the National
Board of Health and Welfare, ref no. 34-9148/2007, 26
October 2007).
Funding
The study was supported by a National LUA/ALF grant no.
11315 and the Region of Västra Götaland, grants from The
Göteborg Medical Society and Hjalmar Svenssons Fund,
The funding source had no role in the study design, data
analysis, data interpretation or writing of the report. MG
and IM had full access to all study data and had final
responsibility for the decision to submit for publication.
Acknowledgements
We thank Ms Marianne Sahlén and Ms Anja Andersson for
help with data registration and Björn Areskoug MSc for
expertise in statistical programming. j
References
1 Milsom I, Altman D, Herbison P, Lapitan MC, Nelson R, Sillén U, et al.
Epidemiology of urinary (UI) and faecal (FI) Incontinence and pelvic
organ prolapse (POP). In: Abrams P, Cardozo L, Khoury S, Wein A,
editors. Incontinence. Paris: Health Publications Ltd; 2009. 35–111 pp.
2 Bidmead J, Cardozo LD. Pelvic floor changes in the older woman. Br
J Urol 1998;82(Suppl. 1):18–25.
3 Mouritsen L, Larsen JP. Symptoms, bother and POPQ in women
referred with pelvic organ prolapse. Int Urogynecol J Pelvic Floor
Dysfunct 2003;14:122–7.
4 Ozel B, White T, Urwitz-Lane R, Minaglia S. The impact of pelvic
organ prolapse on sexual function in women with urinary inconti-
nence. Int Urogynecol J Pelvic Floor Dysfunct 2006;17:14–7.
5 Brown JS, Waetjen LE, Subak LL, Thom DH, Van den Eeden S,
Vittinghoff E. Pelvic organ prolapse surgery in the United States,
1997. Am J Obstet Gynecol 2002;186:712–6.
6 Olsen AL, Smith VJ, Bergstrom JO, Colling JC, Clark AL. Epidemiol-
ogy of surgically managed pelvic organ prolapse and urinary inconti-
nence. Obstet Gynecol 1997;89:501–6.
7 Luber KM, Boero S, Choe JY. The demographics of pelvic floor disor-
ders: current observations and future projections. Am J Obstet Gyne-
col 2001;184:1496–501; discussion 501–3.
8 Smith FJ, Holman CD, Moorin RE, Tsokos N. Lifetime risk of under-
going surgery for pelvic organ prolapse. Obstet Gynecol
2010;116:1096–100.
9 Rortveit G, Brown JS, Thom DH, Van Den Eeden SK, Creasman JM,
Subak LL. Symptomatic pelvic organ prolapse: prevalence and risk
factors in a population-based, racially diverse cohort. Obstet Gynecol
2007;109:1396–403.
10 Tegerstedt G, Miedel A, Maehle-Schmidt M, Nyren O, Hammarstrom
M. Obstetric risk factors for symptomatic prolapse: a population-
based approach. Am J Obstet Gynecol 2006;194:75–81.
159
 Gyhagen et al.
11 Chiaffarino F, Chatenoud L, Dindelli M, Meschia M, Buonaguidi A,
Amicarelli F, et al. Reproductive factors, family history, occupation
and risk of urogenital prolapse. Eur J Obstet Gynecol Reprod Biol
1999;82:63–7.
12 Miedel A, Tegerstedt G, Maehle-Schmidt M, Nyren O, Hammarstrom
M. Nonobstetric risk factors for symptomatic pelvic organ prolapse.
Obstet Gynecol 2009;113:1089–97.
13 Gyhagen M, Bullarbo M, Nielsen TF, Milsom I. The prevalence of uri-
nary incontinence 20 years after child birth: a national cohort study
in singleton primiparae after vaginal or caesarean delivery. BJOG
2012; doi: 10.111/j.1471-0528.2012.03301.x [Epub ahead of print].
14 Tegerstedt G, Miedel A, Maehle-Schmidt M, Nyren O, Hammarstrom
M. A short-form questionnaire identified genital organ prolapse.
J Clin Epidemiol 2005;58:41–6.
15 Abrams P, Cardozo L, Fall M, Griffiths D, Rosier P, Ulmsten U, et al.
The standardisation of terminology in lower urinary tract function:
report from the standardisation sub-committee of the International
Continence Society. Urology 2003;61:37–49.
16 World Health Organization. BMI classification. 2006[http://apps.-
who.int/bmi/index.jsp?introPage=intro_3.html]. Accessed 2 February
2011.
17 World Health Organization. International classification of deseases, 9
th revision, clinical modification (ICD-9-CM). Geneva, Switzerland:
WHO, 1996.
18 Thom DH, van den Eeden SK, Ragins AI, Wassel-Fyr C, Vittinghof E,
Subak LL, et al. Differences in prevalence of urinary incontinence by
race/ethnicity. J Urol 2006;175:259–64.
19 Grodstein F, Fretts R, Lifford K, Resnick N, Curhan G. Association of
age, race, and obstetric history with urinary symptoms among
women in the Nurses’ Health Study. Am J Obstet Gynecol
2003;189:428–34.
20 Diokno AC, Brown MB, Brock BM, Herzog AR, Normolle DP. Clinical
and cystometric characteristics of continent and incontinent noninsti-
tutionalized elderly. J Urol 1988;140:567–71.
21 Herzog AR, Fultz NH. Prevalence and incidence of urinary inconti-
nence in community-dwelling populations. J Am Geriatr Soc
1990;38:273–81.
22 Slieker-ten Hove MC, Pool-Goudzwaard AL, Eijkemans MJ, Steegers-
Theunissen RP, Burger CW, Vierhout ME. The prevalence of pelvic
organ prolapse symptoms and signs and their relation with bladder
and bowel disorders in a general female population. Int Urogynecol
J Pelvic Floor Dysfunct 2009;20:1037–45.
23 Handa VL, Blomquist JL, Knoepp LR, Hoskey KA, McDermott KC,
Munoz A. Pelvic floor disorders 5–10 years after vaginal or caesar-
ean childbirth. Obstet Gynecol 2011;118:777–84.
160 ª 2012 The Authors BJOG An International Journal of Obstetrics and Gynaecology ª 2012 RCOG
24 Lukacz ES, Lawrence JM, Contreras R, Nager CW, Luber KM. Parity,
mode of delivery, and pelvic floor disorders. Obstet Gynecol
2006;107:1253–60.
25 Samuelsson EC, Victor FT, Tibblin G, Svardsudd KF. Signs of genital
prolapse in a Swedish population of women 20 to 59 years of age
and possible related factors. Am J Obstet Gynecol 1999;1:299–
305.
26 Timonen S, Nuoranne E, Meyer B. Genital prolapse. Etiological fac-
tors. Ann Chir Gynaecol Fenn 1968;57:363–70.
27 Allen RE, Hosker GL, Smith AR, Warrell DW. Pelvic floor damage
and childbirth: a neurophysiological study. Br J Obstet Gynaecol
1990;97:770–9.
28 Snooks SJ, Swash M, Henry MM, Setchell M. Risk factors in child-
birth causing damage to the pelvic floor innervation. Int J Colorectal
Dis 1986;1:20–4.
29 Kearney R, Miller JM, Ashton-Miller JA, DeLancey JO. Obstetric fac-
tors associated with levator ani muscle injury after vaginal birth.
Obstet Gynecol 2006;107:144–9.
30 Singh K, Jakab M, Reid WM, Berger LA, Hoyte L. Three-dimensional
magnetic resonance imaging assessment of levator ani morphologic
features in different grades of prolapse. Am J Obstet Gynecol
2003;188:910–5.
31 Hendrix SL, Clark A, Nygaard I, Aragaki A, Barnabei V, McTiernan A.
Pelvic organ prolapse in the Women’s Health Initiative: gravity and
gravidity. Am J Obstet Gynecol 2002;186:1160–6.
32 Risk factors for genital prolapse in non-hysterectomized women
around menopause. Results from a large cross-sectional study in
menopausal clinics in Italy. Progetto Menopausa Italia Study Group.
Eur J Obstet Gynecol Reprod Biol 2000;93:135–40.
33 Mant J, Painter R, Vessey M. Epidemiology of genital prolapse:
observations from the Oxford Family Planning Association Study. Br
J Obstet Gynaecol 1997;104:579–85.
34 Lawrence JM, Lukacz ES, Nager CW, Hsu JW, Luber KM. Prevalence
and co-occurrence of pelvic floor disorders in community-dwelling
women. Obstet Gynecol 2008;111:678–85.
35 Buchsbaum GM. Urinary incontinence and pelvic organ prolapse.
Minerva Urol Nefrol 2006;58:311–9.
36 Barber EL, Lundsberg LS, Belanger K, Pettker CM, Funai EF, Illuzzi JL.
Indications contributing to the increasing caesarean delivery rate.
Obstet Gynecol 2011;118:29–38.
37 Wax JR, Cartin A, Pinette MG, Blackstone J. Patient choice caesar-
ean – the Maine experience. Birth 2005;32:203–6.
38 National Institutes of Health state-of-the-science conference state-
ment: Caesarean delivery on maternal request March 27–29, 2006.
Obstet Gynecol 2006;107:1386–97.