Pelvic pain and mode of delivery

Pelvic pain is common, accounting for 10% of all ambulatory referrals to a gynecologist 1 . Prevalence rates range from 4–15% depending on the population studied 2 , 3 . Several previous studies have suggested that mode of delivery may have an impact on the development of pelvic pain. During a vaginal delivery, descent of fetal head and stretching of the perineum can cause direct tissue and neuromuscular injury. Prior to vaginal birth, the area of the levator hiatus is 6–36 cm 2 . The area of the average fetal head in the plane of minimal diameter measures 70–90 cm 2 . Stretching of the tissues to accommodate the fetal head may result in tears of the levator ani muscles as well as the perineum. Quinn et al recently found an association between levator ani lacerations and unexplained chronic pelvic pain 4 . Cesarean delivery has also been advocated as a cause of pelvic pain. Twelve percent of women continue to report pain related to their cesarean delivery 10.2 months postoperatively 5 . Potential causes of pain related to cesarean delivery include ilioinguinal or iliohypogastric nerve entrapment 6 , uterine scar defect 7 , and pelvic adhesions 8 . Almeida et al reported that cesarean delivery is a risk factor for chronic pelvic pain with an Odd Ratio (OR) of 3.7 and 95% Confidence Limit (CL) 1.7 to 7.7 8 . The only comparative study on the topic is the Term Breech Trial 9 . A subset of participants was asked about pelvic pain 2 years or more postpartum. Based on intent to treat, there was no difference in pain during sex, pain on the outside of the abdomen, pain deep inside the abdomen, and pain in the genital areas between women planning a vaginal birth and those planning a cesarean birth. It is important t o note that 50% of the women planning a vaginal birth actually delivered by cesarean delivery. Therefore, this study leaves unanswered questions with regards to pain and mode of delivery. The objective of this study was to assess the effect of mode of delivery on pelvic pain 6–11 years after the first delivery. Specifically, in the setting of a longitudinal cohort study, we investigated whether mode of delivery (vaginal birth ever vs. cesarean delivery only) is associated with dysmenorrhea, dyspareunia, and pelvic pain not associated with menses or intercourse. We also assessed the effect of other maternal and obstetrical characteristics on pelvic pain.

This was a planned supplementary analysis of data collected for the Mothers’ Outcomes after Delivery (MOAD) study, a longitudinal cohort study of maternal health outcomes after childbirth. The study began recruitment in 2008 and is ongoing. The study details have been previously described 10 . Briefly, women are recruited 5–10 years after delivery of their first child at Greater Baltimore Medical Center. Recruitment is based on delivery type (cesarean without labor, cesarean with labor, vaginal birth) and matched for maternal age and years since delivery. As previously reported 10 , participation in this research was similar across birth groups. Participants are seen on an annual basis at which time a physical exam is performed and questionnaires are completed. Institutional review board approval was obtained for this research and all participants provided informed consent. Questions about pelvic pain are asked at the first follow-up MOAD visit, approximately one year after a woman’s baseline visit, and 6–11 years after her first delivery. The Oxfordshire Women’s Health Study 11 Questionnaire is used to assess prevalence, severity and frequency of pelvic pain. The questionnaire is a validated tool and has previously been used to study gynecologic pain populations. 11 , 12 At the time of this analysis, 1,497 women were enrolled in the MOAD study, and 1,416 (95%) were eligible for the first follow-up survey. Among those who were eligible, 1129 (80%) completed the survey (including the Oxfordshire Questionnaire), 230 (15%) did not participate in this follow up study, and 57 (4%) were still pending at the time of this analysis. Of the 1129 who completed the survey, fourteen women were excluded from the analysis because they reported having surgery for pelvic pain prior to their first delivery, leaving 1,115 for the analysis. The primary outcome for this analysis was moderate or severe pelvic pain as determined by the Oxfordshire Questionnaire. The questionnaire explores 3 types of pelvic pain: dysmenorrhea (pain associated with menses), dyspareunia (pain associated with sexual intercourse), and pelvic pain not associated with menses or intercourse. Women who are still menstruating are asked about dysmenorrhea. Those who report dysmenorrhea are asked how often the painful periods occurred over the past year: ‘Never’, ‘Occasionally (with 50% of my periods)’, or ‘Always (with every period).’ They are also asked to classify their pain as ‘None’, ‘Mild’, ‘Moderate’ or ‘Severe.’ Women who report having intercourse in the past year are asked similar questions regarding dyspareunia. In addition to the frequency and severity questions, women reporting dyspareunia are asked whether their pain occurred during intercourse, in the 24 hours after intercourse, or both during and in the 24 hours after intercourse. All women participating in the survey are asked about pelvic pain not associated with menses and intercourse, and those who report having pain are asked to classify the severity of their pain in the same fashion as mentioned above. The main exposure of interest was vaginal birth. Women who had at least one vaginal delivery were compared with women who delivered all of their children by cesarean. The delivery records for every GBMC delivery were reviewed by a MOAD physician and the following obstetrical exposures were recorded: vaginal birth, prolonged second stage (greater than two hours), vaginal delivery with birth weight greater than or equal to 4,000 grams, perineal laceration, episiotomy, anal sphincter laceration, and operative vaginal delivery (vacuum or forceps). The 1,115 women in this analysis contributed 2,126 delivery records to the study; of those, only 107 (5%) deliveries occurred at outside hospitals. In cases where the woman delivered somewhere other than GBMC or the information was missing on the chart, the woman’s self-reported history was used. In addition to the obstetrical exposures, we also considered the following confounders: age at first follow-up, years from first delivery to first follow-up, race, maternal age at first delivery, multiparity, and obesity at time of first follow-up. Obesity was defined as a body mass index of 30 kg/m 2 or greater. In order to compare maternal characteristics by mode of delivery, p-values were calculated using a Fisher’s Exact Test for categorical variables and a Kruskall-Wallis test for continuous variables. Chi-squared tests were run to detect differences in pelvic pain prevalence, severity and frequency by mode of delivery. To identify differences in prevalence of moderate to severe pain by maternal characteristics and obstetrical exposures, we used multivariate logistic regression, modeling pelvic pain on the characteristic of interest, age greater than 40 years and BMI greater than or equal to 30 kg/m 2 . All analysis was performed using SAS 9.2 © statistical software. Statistical significance was defined at the α=0.05 level.

One thousand, one hundred and fifteen women completed the first follow-up visit and are included in this analysis. There was no difference in participation between women who had delivered vaginally versus those who had delivered exclusively by cesarean. Five hundred and seventy-seven delivered by cesarean delivery only and 538 had at least one vaginal birth. The maternal characteristics, by delivery group, are listed in Table 1 . Women who delivered by cesarean were significantly more likely to be obese, and women with at least one vaginal delivery were more likely to have experienced a prolonged second stage of labor. Of those women who experienced at least one vaginal birth, 14% (75) delivered at least one infant ≥4kg vaginally, 62% (331) had at least one perineal laceration, 55% (297) had at least one episiotomy, 18% (99) had at least one anal sphincter lacerations, and 25% (132) had at least one operative delivery (data not shown). Pelvic pain by mode of delivery is summarized in Table 2 . Women who delivered by cesarean delivery only were more likely to report moderate to severe dysmenorrhea (p=0.020). We found no other significant differences between delivery groups in the prevalence or severity of pelvic pain. Eighty one (7%) of women experienced both a vaginal and cesarean delivery, and therefore would carry risk factors for both types of deliveries. In a separate sensitivity analysis that excluded these 81 women, our primary conclusions were unchanged. Overall, dysmenorrhea was common in our population with 42% of women reporting pain with menses. Dyspareunia was also relatively common (21%), however, very few women reported moderate to severe dyspareunia or a decrease in frequency of intercourse due to dyspareunia. One hundred and ninety one (17%) women had talked to a healthcare provider about pain, and nineteen women (2%) reported having surgery for pelvic pain. The median age of women at the time of their surgery was 36 years (IQR: 32, 39), with no difference noted between groups for prevalence or age at surgery. Moderate to severe pelvic pain based on maternal and obstetrical characteristics is presented in Table 3 (adjusted for age >40 years and BMI ≥ 30 kg/m 2 ). Dysmenorrhea was reported more frequently by younger and obese women and less frequently among women using hormonal contraception. Although moderate or severe dysmenorrhea was significantly more common among women who delivered by cesarean delivery ( Table 2 ), this relationship loses significance after adjusting for age and obesity ( Table 3 ). In a post-hoc analysis, our study had had 80% power to detect a ≥9% difference in dysmenorrhea. Among those who had at least one vaginal birth, moderate to severe dyspareunia was more common among those who experienced a vaginal delivery of a baby ≥4kg and those who had at least one forcep delivery. Moderate or severe pain unrelated to menses or intercourse was not associated with any of the maternal or obstetrical characteristics studied. In order to better distinguish risk factors that contribute to pelvic pain, moderate to severe dyspareunia was modeled on age greater than forty, BMI greater than or equal to 30 kg/m 2 , vaginal delivery of a baby ≥4kg, perineal trauma status (history of both episiotomy and laceration, at least one episiotomy and no lacerations, at least one laceration and no episiotomies, or neither), and history of forceps delivery. The model, restricted to sexually active women with a history of vaginal delivery, revealed that after adjusting for the other risk factors, only vaginal delivery of a baby ≥4kg (OR (95% CL):2.45 (1.02, 5.93)) and forceps delivery (OR (95% CL): 4.04 (1.69, 9.66)) significantly predicted moderate to severe dyspareunia.

The most important finding of this research is that the prevalence of pelvic pain was similar between women who delivered vaginally and those who delivered by cesarean delivery. We found no association between route of delivery (vaginal vs. cesarean) and report of dysmenorrhea, dyspareunia, or pelvic pain not associated with menses or intercourse 6–11 years after a first delivery. As noted, our study had had 80% power to detect a ≥9% difference in the proportion of women with dysmenorrhea. Dysmenorrhea was more common among younger women and obese women, but was not associated with mode of delivery or any of the obstetrical characteristics studied. One previous study 7 reported an association between dysmenorrhea and uterine scar dehiscence based on transvaginal ultrasound studies. The number of previous cesarean deliveries correlated with size of the defect and increased risk of dysmenorrhea and pelvic pain. Others have suggested that an incision in the uterus may predispose one to adenomyosis and thus dysmenorrhea. Our study does not suggest an association between cesarean birth and dysmenorrhea. Among women who delivered vaginally, dyspareunia was reported more frequently among women who experienced at least one forcep delivery and among those who experienced a vaginal delivery of at least one child ≥4kg. We found no conclusive association between perineal trauma and pelvic pain. Perineal trauma is a known cause of immediate postpartum pain and dysparuenia 12 , 13 . For example, in a retrospective cohort study of 615 primiparous women, Signorello et al 13 showed that dyspareunia at 3 months postpartum was associated with the degree of perineal trauma, with a second degree episiotomy or tear doubling the odds for dyspareunia (OR (95% CL): 1.8 (1.2,2.8)) and a third or fourth degree tear increasing the odds almost four-fold (OR (95% CL): 3.7 (1.7,7.7)) compared to no perineal trauma. The difference dissipated at 6 months postpartum, suggesting that this association pertains only to the first few months after delivery. Systematic reviews comparing episiotomy to spontaneous laceration have suggested that episiotomy is associated with more perineal trauma. However, episiotomy has not been shown to be associated with dyspareunia 14 – 16 . Similarly, our study shows no association between dyspareunia and perineal trauma, episiotomy, or anal sphincter laceration. This finding may be related to the long interval between delivery and questionnaire in our study compared to the short term outcome in other studies. Our study showed an increase in dysparuenia among women who delivered at least one child with forceps. In addition to an increased risk of perineal trauma, forcep deliveries have been shown to be associated with levator ani lacerations 17 – 19 . Likewise, levator ani trauma has been shown to be associated with pelvic pain. In a case control study, Quinn 4 showed a higher rate of levator ani injury in women with unexplained chronic pelvic pain (20/26) compared to controls (2/32). We speculate that levator ani injury may contribute to increased pain seen in our participants who delivered with forceps. The strengths of our study include the large sample size, use of a validated questionnaire, long term follow-up, and review of all hospital records to confirm maternal and obstetrical exposures. In addition, our groups are based on actual route of delivery rather than intent to treat as with the Term Breech Trial. One of the weaknesses of this study is that we have information about pelvic pain at only one point in time. For example, a prior history of dysmenorrhea was not assessed among women who were no longer menstruating (15% of cesarean delivery and 11% of vaginal delivery participants). Thus, we do not know if these women had dysmenorrhea at any other time since delivery. Similarly, if a woman did not have intercourse over the prior year, she was not asked about dysparuenia. In addition, we do not have information about other causes of pelvic pain such as endometriosis, pelvic inflammatory disease, and fibroids. Finally, because our participants agreed to participate in a longitudinal cohort study, they may differ from other parous women. The most notable example is that these participants volunteered for research that requires an annual gynecologic examination. We speculate that women with severe pain and dyspareunia may be less likely to participate in this type of research. However, this would be unlikely to threaten the internal validity of our investigation. In conclusion, our study shows that vaginal birth is not associated with a higher prevalence of dysmenorrhea, dyspareunia, or pelvic pain not associated with menses or intercourse as compared to cesarean delivery. However, we did find a higher rate of dyspareunia among women who experienced at least one forcep delivery and women who delivered at least one baby ≥4kg vaginally. Physicians will find this information helpful in counseling patients as they prepare for childbirth.