Is vaginal delivery the only relevant predictor of symptomatic pelvic organ prolapse – a matched cohort study in women aged 40 to 64 years.

Gyhagen M1, Åkervall S1, Molin M2, Milsom I1

Research Type


Abstract Category

Pelvic Organ Prolapse

Abstract 256
Best Urogynaecology
Scientific Podium Session 14
Thursday 5th September 2019
10:00 - 10:15
Hall K
Female Prolapse Symptoms Pelvic Floor Prevention Questionnaire
1.Gothenburg Continence Research Centre, Institute of Clinical Sciences, Sahlgrenska Academy at Gothenburg University., 2.Statistical Consultancy Group

Maria Gyhagen



Hypothesis / aims of study
Pelvic organ prolapse (POP) poses a huge and still largely unaddressed disease burden for many women throughout the world. In welfare states more than every fifth elderly parous woman has undergone surgery for POP and/or urinary incontinence (1). Surgery however does not guarantee success as there is a one in three risk of failure after a primary non-mesh procedure. There is a widespread belief that it usually takes many years, even decades, after the first childbirth, before symptoms and signs of POP manifest themselves, before women present for evaluation and ultimately seek adequate treatment. Surgical treatment eventually peaks after 70 years of age. This long induction period has caused misinterpretations of results from short-term studies and has made a complete understanding of the natural course of this condition a major challenge. The potential protective effect of cesarean section against genital prolapse is therefore poorly investigated and understood. The aim of this study was therefore to estimate the relative effect of vaginal delivery, pregnancy, ageing, and body mass index (BMI) on the age-dependent change of symptomatic pelvic organ prolapse prevalence in nulliparous, cesarean and vaginal delivered women between the ages of 40 to 64 years. A secondary aim was to calculate the protective capacity of cesarean section during this period, using a matched cohort design and a restricted definition of symptomatic POP.
Study design, materials and methods
Women intended for matching were obtained from three large, nationwide and randomly selected source cohorts of nulliparous women and one-para, vaginal and cesarean delivered women (in total 14,335 were available for matching). The Swedish Medical Birth Register recruited women with one vaginal delivery (n = 3,787) or one cesarean section (n =1,412) (2). These women were surveyed in 2008, all 20 years after childbirth. Inclusion criteria were one singleton birth and no further births. The nulliparous cohort comprised 9,136 women and was selected and surveyed in 2014 (3). The cesarean section group (smallest) served as index against which nulliparous and vaginal delivered women were matched. The matching procedure succeeded in 2635 out of 2640 women (99.8%), resulting in an adequate distribution of age and BMI between the matched groups. The number of women in the matched cohorts were 1320, 1320 and 1315 respectively. The surveys used validated questions to identify pelvic floor dysfunction. For sPOP, a more restricted definition was used by excluding women having a “bulging” symptom only “Infrequently”. Difference in outcome between groups was analyzed with Fisher’s exact test for categorical variables and Mann–Whitney U test for continuous variables. Trend between ordered independent groups was analyzed with Mantel–Haenszel statistics. Predicted age-related values of sPOP were obtained from the logistic regression analysis. Statistical significance was set at p <0.05.
In nulliparous women (2.2-2.6%) and in women after cesarean section (0.5-3.9) the prevalence of sPOP was low, relatively stable and similar from 40 up to 64 years of age (p = 0.89) (Figure 1). In comparison the prevalence of sPOP increased successively from 2.1 to 10.1% in women from 44 to 64 years of age who had previously undergone a vaginal delivery (Figure 1). Predicted values from the logistic regression analysis showed that after cesarean section the rate of sPOP was 4.2% (95% CI 2.1-8.2%) at 40 years of age and 1.1% (95% CI 0.4-2.5%) at 64 years (Figure 2). After vaginal delivery, sPOP increased consistently during the whole age interval. Predicted values of sPOP showed an almost fourfold increase from 3.8% (95% CI 2.2-6.2%) at 40 years to 13.4% (95% CI 9.4-18.9%) at 64 years of age, being almost doubled for each 10-year period (OR10 years 1.86; 95% CI 1.28-2.72) (Figure 2). BMI was not a risk factor for sPOP in any group (p = 0.17, 0.26, 0.63 respectively). The induction period (time from vaginal delivery to occurrence of sPOP) seemed to be about 20 years among women giving birth in their early twenties. The derived calculated protective effect of cesarean section against sPOP after vaginal delivery was therefore substantial and can be expressed as the age-dependent difference of sPOP between vaginal delivery and cesarean section. The derived calculated protective effect of cesarean section increased continuously up to age 64 year, at which age the reduction was 95.0%.
Interpretation of results
Across an age span of 25 years up to age 64 years, nulliparous and cesarean delivered women did not seem to add incident sPOP to any significant degree. In sharp contrast, after one vaginal delivery the prevalence of sPOP increased exponentially, almost doubling for each 10-year period. Of the four potential risk factors analyzed (age, BMI, pregnancy and vaginal delivery), only vaginal delivery was a clinically relevant predictor for symptomatic pelvic organ prolapse.
Concluding message
The personal suffering caused by genital prolapse is borne by women following vaginal delivery, which in this study was the only relevant predictor of symptomatic pelvic organ prolapse. These findings indicate the need for a more critical attitude towards policies, attitudes and practices in obstetric care.
Figure 1 Figure 1. The age–related change of symptomatic pelvic organ prolapse in nulliparous, cesarean and vaginally delivered women.
Figure 2 Figure 2. The age-related prevalence of symptomatic pelvic organ prolapse "Sometimes and more often" in nulliparous, cesarean and vaginally deliverd women.
  1. Haya et al. Am J Obstet Gynecol 2015;212:755.e1-755.e27.
  2. Gyhagen et al. BJOG. 2013;120:152-60
  3. Al-Mukhtar Othman et al. Am J Obstet Gynecol. 2017;216:149.e1-149.e11
Funding The study was supported by a national grant (nr 11315) and from the Healthcare Committee, Region Västra Götaland. Grants from the Göteborg Medical Society, Hjalmar Svensson Fund, and Alice Swenzon Fund. The funding source had no role in study design, data analysis, data interpretation, or writing of the report. Clinical Trial No Subjects Human Ethics Committee Ethical approval for the study was obtained from the Regional Ethical Review Board in Gothenburg, Sweden (reference no. 381–07; August 13, 2007 and no. 776–13; November 18, 2013) and the National Review Board (no. 34–9148; October 26, 2007). All women received written information and gave their written consent before participation in the study. Helsinki Yes Informed Consent Yes