Pelvic floor disorders, including urinary and fecal incontinence and pelvic organ prolapse, are more prevalent among women who have delivered at least one child.1–3 Determining which aspects of childbirth contribute most to the risk of pelvic floor disorders and investigating how obstetric care can be modified to reduce their incidence is an area of intense investigation. Cesarean delivery has been suggested as a possible prevention strategy.
Research on the relationship between childbirth and pelvic floor disorders is challenging because of the long latency for these disorders. In the first year after childbirth, some studies have suggested that cesarean birth reduces the prevalence of bothersome incontinence symptoms.4–7 However, cross-sectional studies of parous women later in life2,8 suggest that women who have delivered vaginally have a higher prevalence of pelvic floor symptoms than women who report only cesarean deliveries. This association has been most convincingly demonstrated for stress incontinence2,8 with much less evidence for other pelvic floor disorders.
Epidemiologic studies of prolapse are particularly challenging because of the expense and logistical complexity of performing gynecologic examinations across large populations. As a result, studies of the relationship between childbirth and prolapse have typically relied on surrogate measures of prolapse such as symptoms8,9 or surgical treatment.3,10,11 However, symptoms are weakly correlated with objective measures of prolapse.12,13 Also, given that thresholds for surgical intervention may vary considerably, the incidence of surgical treatment might be an unreliable indicator of the true incidence of prolapse. Therefore, the relative incidence of prolapse after vaginal and cesarean delivery remains unknown.
To address the question of whether delivery mode influences the later development of pelvic floor disorders, we are conducting a longitudinal cohort study of pelvic floor disorders after childbirth. Because of the long latency for these disorders, women are recruited to participate 5–10 years after their first delivery. This analysis, based on enrollment data, addresses the specific question of whether delivery mode is associated with pelvic floor disorders, including pelvic organ prolapse, 5–10 years after delivery.
The Mothers' Outcomes After Delivery study is a prospective cohort study of pelvic floor outcomes in women recruited 5–10 years after delivery of their first child. This study has been designed and conducted by investigators from the Johns Hopkins Medical Institutions and the Greater Baltimore Medical Center in Baltimore, Maryland. Institutional review board approval was obtained at both institutions. All participants provided written informed consent.
Recruitment for this longitudinal cohort study began in 2008 and is ongoing. To be eligible, women must have given birth to their first child (index birth) at Greater Baltimore Medical Center 5–10 years before enrollment. Participants were identified from obstetric hospital discharge records using discharge diagnoses. Hospital charts were reviewed by trained personnel to verify eligibility and to confirm delivery type. Potential participants were also screened for eligibility by telephone interview. Exclusion criteria (applied to the index birth) included: maternal age younger than 15 or older than 50 years, delivery at less than 37 weeks of gestation, placenta previa, multiple gestation, known fetal congenital anomaly, stillbirth, prior myomectomy, and abruption. Women who developed these events during subsequent pregnancies were not excluded.
Based on chart review, 8,285 eligible women were identified. Each eligible delivery was classified as either a vaginal birth or cesarean birth. Cesarean births were further classified as either unlabored cesarean deliveries or labored cesarean deliveries. Unlabored cesarean delivery was defined as cesarean delivery performed before the onset of active labor, defined as regular contractions with cervical dilation of 3 cm or greater. Each delivery was also classified by the woman's age at delivery (organized into 5-year strata) and the number of years from first delivery (in quarter-year strata).
Based on the objectives of the study, the recruitment goal was 1,000 women, including 200 unlabored cesareans, 400 labored cesarean deliveries, and 400 vaginal births (classified by index birth). Our recruitment strategy was designed to ensure that age at first delivery and years since first delivery were similar across these three birth groups. To ensure a sufficient sample of unlabored cesarean deliveries, we recruited all eligible women in this group. For the other groups, women within each stratum were randomly assigned a “rank” and recruitment proceeded according to rank until the number of women recruited in the labored cesarean and vaginal birth groups was twice the number of women in the unlabored cesarean group. As a result of these matching criteria, 5,215 of the 8,285 eligible births were approached for recruitment.
We were able to contact 2,510 of 5,215 women (48.1%). Fifty-seven women were excluded because they were ineligible, were currently pregnant, had delivered a child or reported pelvic surgery within the past 6 months, or were unable complete written questionnaires in English. Of the 2,453 who were contacted and eligible, 1,271 (51.8%) declined to participate. This analysis is based on enrollment data from 1,011 women who had completed the baseline assessment at the time of this analysis.
Obstetric exposures were derived from abstraction of all delivery records for each participant. Of 1,951 deliveries reported by the 1,011 participants, multiparous women reported 70 deliveries that occurred at other hospitals (eg, after the index birth). For these deliveries, we substituted maternal recall of delivery events for a formal review of the obstetric records.
Based on obstetric history, women were classified into five groups to more fully characterize their obstetric exposures over all deliveries. The reference group (ie, unexposed) comprised women who had delivered all their children by unlabored cesarean. The four “exposed” groups were: 1) cesarean delivery after the onset of active labor but before complete cervical dilation; 2) cesarean delivery after complete cervical dilation; 3) spontaneous vaginal birth; and 4) operative vaginal birth. It was hypothesized that the harm to the pelvic floor increased across these groups, with the least harm in the reference group (group 0) and the greatest harm in group 4. A woman's group was determined by considering all of her deliveries; women were placed in the group corresponding to the delivery that was likely to cause the most harm to the pelvic floor. For instance, any woman with an operative delivery was placed in that group regardless of her other delivery types. In 96%, the first birth was the birth most likely to cause the most harm to the pelvic floor.
In addition to obstetric exposures, we considered the following confounders: race, maternal age at the time of first delivery, interval (in years) between delivery and study enrollment, obesity, and cigarette smoking. Race was self-reported. Obesity was determined at study enrollment. Specifically, each participant's weight and height were measured and body mass index (BMI) was calculated (as weight (kg)/[height (m)]2). Obesity was defined as a BMI of 30 or greater. Cigarette smoking was classified as “never” or “ever” based on whether a woman had smoked at least 100 cigarettes in her life.
The presence or absence of pelvic floor disorders was evaluated at the enrollment visit. Symptoms of pelvic floor disorders were assessed using the Epidemiology of Prolapse and Incontinence Questionnaire, a validated self-administered questionnaire.14 This questionnaire generates scores for four pelvic floor disorders: stress urinary incontinence, overactive bladder, anal incontinence and pelvic organ prolapse. In each case, a validated threshold is used to define women who meet criteria for the disorder. Scores greater than these threshold values have been shown to correspond to significant bother from pelvic floor symptoms.14 In this research, we used the published thresholds14 to distinguish women with and without each pelvic floor disorder.
In addition to the research questionnaire, a gynecologic examination was performed to assess pelvic organ support using the Pelvic Organ Prolapse Quantification examination system.15 The examination was performed by physicians and a research nurse, each of whom demonstrated competency in performing the research examination before the study; competency was reconfirmed throughout the study. Women were classified as having objective evidence of prolapse if the most dependent point of the vaginal wall or the cervix came to or beyond the hymen.13,16,17
At enrollment into our cohort study, participants were asked about prior treatment for pelvic floor disorders, including surgery. Participants were also asked about current therapy, including medications for urinary incontinence or current pessary use for treatment of prolapse. We also considered current or prior pelvic muscle exercises, but only if the program was supervised by a therapist. For the purposes of this analysis, women who reported prior surgery, prior supervised pelvic muscle exercises, or any current therapy for a specific pelvic floor disorder were considered to have that condition regardless of current symptoms.
This cross-sectional analysis is based on enrollment data. Contingency tables were used to estimate the univariable associations between each pelvic floor disorder and each exposure. In these analyses, each of the five pelvic floor disorders of interest (stress urinary incontinence, overactive bladder, anal incontinence, pelvic organ prolapse symptoms, and pelvic organ prolapse by examination) was considered separately. P values were obtained using a Fisher's exact test.
Logistic regression was used to provide measures of association between obstetric exposures and prevalent pelvic floor disorders. Group 0 (women who had delivered exclusively by cesarean before labor) was the reference group. Univariable (ie, unadjusted) and multivariable (ie, adjusted) analyses were performed. The multivariable model was adjusted for African American race, maternal age older than 35 years at first delivery, multiparity, and obesity (BMI greater than 30). Relative odds were used to summarize the strength of the associations and Wald-based 95% confidence intervals (CIs)18 were used to determine statistical significance.
In planning this research, we estimated that at least 10% of women would demonstrate pelvic floor disorders at the time of enrollment. We selected our sample size to detect a doubling of the relative odds, which we considered a clinically important difference. With that assumption, we calculated that a sample size of 200 women in the reference group (unlabored cesarean) and 400 women in each comparison group would provide 80% power to reject the null hypothesis (assuming a type I error of 0.05).
At enrollment, the median age was 39.5 years, 82% were white, and 28% were older than 35 years at the time of their first delivery. Also, 72% were multiparous and 26% were classified as obese. The median interval between first delivery and enrollment was 7.4 years (interquartile range, 6.3–9.0); because of delays in recruiting, 4% were more than 10 years from first delivery at enrollment. The cohort included 205 women whose index birth was an unlabored cesarean, 388 whose index birth was a labored cesarean, and 418 women whose index birth was a vaginal delivery. Participation was similar across these three birth groups (P=.63).
The characteristics of participants, by obstetric exposure, are shown in Table 1. Of 1,011 participants, 192 women had delivered exclusively by cesarean before active labor (group 0). Of the 368 women who had no prior vaginal births and at least one cesarean birth after the onset of active labor, 228 had experienced at least one cesarean delivery in the first stage of labor (group 1) and 140 had experienced at least one cesarean delivery in the second stage of labor (group 2). Of the 451 women who had at least one vaginal birth, 325 women had no operative vaginal births (group 3) and 126 women had experienced at least one operative vaginal birth (group 4). Among those who delivered by operative vaginal birth, 50 had delivered by vacuum extraction, 71 by forceps, and five had experienced both vacuum and forceps deliveries. As expected by study design, the obstetric exposure groups were similar with respect to age at first delivery and the interval from first delivery; the groups were also similar with respect to parity and smoking. Differences across groups were noted for age at enrollment, race, and obesity.
The most common pelvic floor disorders were stress urinary incontinence (11%), overactive blader (8%), and anal incontinence (11%). Bothersome symptoms of prolapse were reported by 3% compared with the 7% who demonstrated prolapse to or beyond the hymen based on physical examination. As shown in Table 2, women who had experienced at least one vaginal birth were significantly more likely to report stress incontinence (P<.001) and were significantly more likely to demonstrate prolapse to or beyond the hymen (P<.001). Symptoms of prolapse were uncommon across all groups but were significantly associated with operative delivery (P=.012). Symptoms of anal incontinence did not differ significantly across obstetric exposure groups (P=.195).
As noted, symptoms of prolapse were less prevalent than anatomic evidence of prolapse. Specifically, of 75 women with prolapse to or beyond the hymen, only 14 (19%) reported bothersome symptoms. Thus, the majority of women with prolapse on examination were asymptomatic or minimally symptomatic. Among the 75 women who had prolapse on examination, those who delivered their first child after age 35 years were more likely to report bothersome prolapse symptoms (P=.045). Moreover, among the 26 women reporting symptoms of prolapse, obese women were significantly less likely to demonstrate prolapse on examination (P<.001). Thus, the symptoms of prolapse were expressed differently by obese women and by women who delivered a first child at an older age.
The relative odds for each pelvic floor disorder by delivery group are presented in Table 3. After adjusting for African American race, maternal age at first delivery, multiparity, smoking history, and obesity, we found that stress incontinence and prolapse to or beyond the hymen were significantly more common among women who had delivered by vaginal birth (either spontaneous or operative) than among women who delivered exclusively by cesarean (groups 0 to 2). The other pelvic floor disorders were significantly associated with operative vaginal birth but not with spontaneous vaginal birth. For women with a history of at least one operative delivery, the adjusted odds of stress incontinence and overactive bladder were more than quadrupled (stress urinary incontinence: relative odds 4.45, 95% CI 2.14–9.27; overactive bladder: relative odds 4.89, 95% CI 2.23–10.74). The odds of prolapse to or beyond the hymen were increased almost eightfold among women who had a history of at least one operative birth (relative odds 7.50; 95% CI 2.70–20.87). To put this into perspective, we calculated the number needed to harm, a measure of the magnitude of risk associated with an exposure. Assuming a causal relationship between delivery type and prolapse, our results suggest that, relative to cesarean deliveries, 8.9 spontaneous vaginal births would lead to one additional case of prolapse. Similarly, 6.8 operative births would lead to one additional case of prolapse.
In contrast, we did not see a clinical or statistical difference in the odds of any pelvic floor disorder among the three cesarean groups. Specifically, among women who delivered all their children by cesarean, there was no significant difference in pelvic floor disorders between women who entered active labor before at least one cesarean birth and women who had never been exposed to active labor. However, given the relatively low prevalence for each pelvic floor disorder in the reference group (less than 10%), the study had limited power (less than 80%) to detect less than a doubling of the odds of pelvic floor disorders among the three subgroups of women with only cesarean deliveries.
The most important finding from this study is that 5–10 years after a first birth, pelvic floor disorders were dramatically increased among women with a history of at least one operative vaginal birth. Our results are consistent with published studies indicating that operative delivery is associated with subsequent fecal incontinence.9,19 Our findings also demonstrate an increase in urinary incontinence after operative delivery, a finding that has been inconsistently observed in prior research.5,9,20,21 We are unaware of any prior studies investigating the association between operative vaginal birth and prolapse. We found a very strong association in this regard and we speculate that this association may be mediated by undetected injury to the levator ani. It has been suggested that operative delivery is a risk factor for levator ani muscle injury22 and that levator injuries may contribute to prolapse.23 Further research is needed to identify whether levator ani injuries are the mechanism responsible for the association we observed between operative birth and prolapse.
Another important finding from this research is the absence of any association between active labor and pelvic floor disorders among women delivering by cesarean. Prior studies had suggested that labored cesarean delivery is a risk factor for pelvic floor disorders compared with unlabored cesarean delivery.8,24–26 We specifically designed our recruitment to include a sufficient number of women to test for differences between these groups. Our findings are similar to those of Boyles,4 who found no difference in urinary incontinence 1 year after either elective cesarean delivery, cesarean delivery in labor, or cesarean delivery in the second stage of labor. Our results build on these findings with a longer duration of follow-up and an assessment of other pelvic floor disorders.
An important observation is that 75% of women with prolapse to or beyond the hymen were asymptomatic or minimally symptomatic. This finding is consistent with observations made by Bradley, who found that 84% of postmenopausal women with this degree of prolapse did not see or feel a vaginal bulge.13 The prognosis for women with asymptomatic prolapse is unclear. The conventional wisdom is that these women develop worsening prolapse and increasing bother over time. However, the natural history of prolapse is poorly understood.27,28 We do not know whether asymptomatic prolapse is more or less likely to progress (or regress). Longitudinal observation of this cohort will allow us to characterize prolapse progression or remission and to investigate the incidence of bothersome symptoms as the population ages.
One limitation of this study is that the prevalence of all pelvic floor disorders in this population was low, limiting our power to investigate some associations and contributing to wide CIs for relative odds of the least common outcomes. However, we intentionally recruited women 5–10 years from first childbirth in the hopes of capturing the onset of pelvic floor disorders with longitudinal observation (eg, as the population ages). Thus, the relative difference between these groups is expected to grow with time and will be a major focus of the ongoing longitudinal study. In addition, this is an observational study and we therefore cannot with certainty ascribe the incidence of pelvic floor disorders to obstetric events. We cannot exclude the possibility that unmeasured characteristics of the population or other exposures were relevant to the development of pelvic floor disorders. Finally, participation in this research was only 50%. However, this participation rate is similar to other studies investigating pelvic floor disorders after childbirth.4,8,9
A strength of this study is that obstetric exposures were defined by obstetric hospital records, rather than maternal recall, for 96.5% of all deliveries. Another strength of the study is the assessment of pelvic floor disorders with a validated symptom questionnaire and a structured examination to assess pelvic organ support. The examination for pelvic organ support is a relatively unique aspect of this study. Prior studies defined prolapse either by symptoms8 or the incidence of surgical intervention.3,10,11 In one of the few prior cohort studies that included an anatomic assessment of pelvic organ support after childbirth,16 prolapse to or beyond the hymen was noted in two of 39 (5%) women after unlabored cesarean delivery compared with 15 of 105 (14%) after vaginal birth. The present study expands on those results by including a more adequate sample size as well as a longer duration of follow-up.
In summary, these results provide important evidence of a strong association between vaginal childbirth and pelvic floor disorders. In particular, our results suggest that operative vaginal delivery is associated with a large increase in the relative odds of pelvic floor disorders 5–10 years after delivery. Longitudinal data to be collected as this study proceeds will establish whether vaginal birth and operative delivery have a different effect than cesarean birth on the long-term progression of pelvic floor disorders over time. Our future research will establish to what extent obstetric exposures affect changes over time in symptom burden and anatomic support. These data will be useful to women and their obstetric providers as they weigh childbirth options.
1. Nygaard I, Barber MD, Burgio KL, Kenton K, Meikle S, Schaffer J, et al.; Pelvic Floor Disorders Network. Prevalence of symptomatic pelvic floor disorders in US women. JAMA 2008;300:1311–6.
2. Rortveit G, Daltveit AK, Hannestad YS, Hunskaar S; Norwegian EPINCONT Study. Urinary incontinence after vaginal delivery or cesarean section. N Engl J Med 2003;348:900–7.
3. 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.
4. Boyles SH, Li H, Mori T, Osterweil P, Guise JM. Effect of mode of delivery on the incidence of urinary incontinence in primiparous women. Obstet Gynecol 2009;113:134–41.
5. Schytt E, Lindmark G, Waldenström U. Symptoms of stress incontinence 1 year after childbirth: prevalence and predictors in a national Swedish sample. Acta Obstet Gynecol Scand 2004;83:928–36.
6. Borello-France D, Burgio KL, Richter HE, Zyczynski H, Fitzgerald MP, Whitehead W, et al.; Pelvic Floor Disorders Network. Fecal and urinary incontinence in primiparous women. Obstet Gynecol 2006;108:863–72.
7. Hannah ME, Whyte H, Hannah WJ, Hewson S, Amankwah K, Cheng M; Term Breech Trial Collaborative Group. Maternal outcomes at 2 years after planned cesarean section versus planned vaginal birth for breech presentation at term: the international randomized Term Breech Trial. Am J Obstet Gynecol 2004;191:917–27.
8. Lukacz ES, Lawrence JM, Contreras R, Nager CW, Luber KM. Parity, mode of delivery, and pelvic floor disorders. Obstet Gynecol 2006;107:1253–60.
9. Dolan LM, Hilton P. Obstetric risk factors and pelvic floor dysfunction 20 years after first delivery. Int Urogynecol J Pelvic Floor Dysfunct 2010;21:535–44.
10. Larsson C, Källen K, Andolf E. Cesarean section and risk of pelvic organ prolapse: a nested case-control study. Am J Obstet Gynecol 2009;200:243.e1–4.
11. Leijonhufvud A, Lundholm C, Cnattingius S, Granath F, Andolf E, Altman D. Risks of stress urinary incontinence and pelvic organ prolapse surgery in relation to mode of childbirth. Am J Obstet Gynecol 2011;204:70.e1–7.
12. Gutman RE, Ford DE, Quiroz LH, Shippey SH, Handa VL. Is there a pelvic organ prolapse threshold that predicts pelvic floor symptoms? Am J Obstet Gynecol 2008;199:683.e1–7.
13. Bradley CS, Nygaard IE. Vaginal wall descensus and pelvic floor symptoms in older women. Obstet Gynecol 2005;106:759–66.
14. Lukacz ES, Lawrence JM, Buckwalter JG, Burchette RJ, Nager CW, Luber KM. Epidemiology of prolapse and incontinence questionnaire: validation of a new epidemiologic survey. Int Urogynecol J Pelvic Floor Dysfunct 2005;16:272–84.
15. Bump RC, Mattiasson A, Bø K, Brubaker LP, DeLancey JO, Klarskov P, et al.. The standardization of terminology of female pelvic organ prolapse and pelvic floor dysfunction. Am J Obstet Gynecol 1996;175:10–7.
16. Handa VL, Nygaard I, Kenton K, Cundiff GW, Ghetti C, Ye W, et al.; Pelvic Floor Disorders Network. Pelvic organ support among primiparous women in the first year after childbirth. Int Urogynecol J Pelvic Floor Dysfunct 2009;20:1407–11.
17. Whitcomb EL, Rortveit G, Brown JS, Creasman JM, Thom DH, Van Den Eeden SK, et al.. Racial differences in pelvic organ prolapse. Obstet Gynecol 2009;114:1271–7.
18. Cox DR, Snell EJ. Analysis of binary data. 2nd ed. London (UK): Chapman and Hall; 1989.
19. Macarthur C, Glazener C, Lancashire R, Herbison P, Wilson D, Grant A. Faecal incontinence and mode of first and subsequent delivery: a six-year longitudinal study. BJOG 2005;112:1075–82.
20. MacArthur C, Glazener CM, Wilson PD, Lancashire RJ, Herbison GP, Grant AM. Persistent urinary incontinence and delivery mode history: a six-year longitudinal study. BJOG 2006;113:218–24.
21. Goldberg RP, Abramov Y, Botros S, Miller JJ, Gandhi S, Nickolov A, et al.. Delivery mode is a major environmental determinant of stress urinary incontinence: results of the Evanston-Northwestern Twin Sisters Study. Am J Obstet Gynecol 2005;193:2149–53.
22. Shek KL, Dietz HP. Intrapartum risk factors for levator trauma. BJOG 2010;117:1485–92.
23. DeLancey JO, Morgan DM, Fenner DE, Kearney R, Guire K, Miller JM, et al.. Comparison of levator ani muscle defects and function in women with and without pelvic organ prolapse. Obstet Gynecol 2007;109:295–302.
24. Groutz A, Rimon E, Peled S, Gold R, Pauzner D, Lessing JB, Gordon D. Cesarean section: does it really prevent the development of postpartum stress urinary incontinence? A prospective study of 363 women one year after their first delivery. Neurourol Urodyn 2004;23:2–6.
25. Sze EH, Sherard GB 3rd, Dolezal JM. Pregnancy, labor, delivery, and pelvic organ prolapse. Obstet Gynecol 2002;100:981–6.
26. Chin HY, Chen MC, Liu YH, Wang KH. Postpartum urinary incontinence: a comparison of vaginal delivery, elective, and emergent cesarean section. Int Urogynecol J Pelvic Floor Dysfunct 2006;17:631–5.
27. Handa VL, Garrett E, Hendrix S, Gold E, Robbins J. Progression and remission of pelvic organ prolapse: a longitudinal study of menopausal women. Am J Obstet Gynecol 2004;190:27–32.
28. Bradley CS, Zimmerman MB, Qi Y, Nygaard IE. Natural history of pelvic organ prolapse in postmenopausal women. Obstet Gynecol 2007;109:848–54.