The mode of delivery does not influence the occurrence of post-partum perianal disease flares in patients with inflammatory bowel disease

Introduction Perianal disease occurs in up to 34% of inflammatory bowel disease (IBD) patients. An estimated 25% of women will become pregnant after the initial diagnosis, thus introducing the dilemma of whether mode of delivery affects perianal disease. The aim of our study was to analyze whether a cesarean section (C-section) or vaginal delivery influence perianal involvement. We hypothesized the delivery route would not alter post-partum perianal manifestations in the setting of previously healed perianal disease. Methods All consecutive eligible IBD female patients between 1997 and 2022 who delivered were included. Prior perianal involvement, perianal flare after delivery and delivery method were noted. Results We identified 190 patients with IBD who had a total of 322 deliveries; 169 (52%) were vaginal and 153 (48%) were by C-section. Nineteen women (10%) experienced 21/322 (6%) post-partum perianal flares. Independent predictors were previous abdominal surgery for IBD (OR, 2.7; 95% CI, 1–7.2; p = 0.042), ileocolonic involvement (OR, 3.3; 95% CI, 1.1–9.4; p = 0.030), previous perianal disease (OR, 22; 95% CI, 7–69; p < 0.001), active perianal disease (OR, 96; 95% CI, 21–446; p < 0.001) and biologic (OR, 4.4; 95% CI,1.4–13.6; p < 0.011) or antibiotic (OR, 19.6; 95% CI, 7–54; p < 0.001) treatment. Negative association was found for vaginal delivery (OR, 0.19; 95% CI, 0.06–0.61; p < 0.005). Number of post-partum flares was higher in the C-section group [17 (11%) vs. 4 (2%), p = 0.002]. Conclusions Delivery by C-section section was not protective of ongoing perianal disease activity post-delivery, but should be recommended for women with active perianal involvement.


Introduction
Inflammatory bowel disease (IBD) is a chronic pathology that may present with disabling quality of life characteristics during patients' reproductive years.IBD is characterized by remissions and flares, often requiring prolonged medication, hospitalization, and surgery.Perianal involvement occurs in up to 34% of IBD patients within 10 years of diagnosis, primarily in patients with Crohn's disease (CD) with colonic or rectal involvement [1][2][3][4].Complete fistula healing can be an arduous process, involving multiple medical and surgical treatments, with significant ongoing morbidity that greatly affects the quality of life of patients due to pain, discharge, and abscess formation.In addition, the presence of a fistula indicated a more aggressive phenotype of IBD requiring more frequent hospitalizations, higher incidence of surgery, and increased utilization of corticosteroid treatment [5].
It has been estimated that up to 25% of women become pregnant after the initial diagnosis of CD [6].Given a significant proportion of patients have perianal involvement at the time of pregnancy, the impact of pregnancy and mode of delivery on perianal disease activity become important considerations [7][8][9][10][11][12].In many cases, the method of delivery in a pregnant patient with IBD involves a multidisciplinary discussion between the obstetrician, gastroenterologist, colorectal surgeon, and patient preference due to fear of perianal injury, exacerbation of perianal disease if present, non-healing wounds, fistula, pelvic floor damage, and risk of incontinence [11,13,14].Interestingly, high rates of cesarean section (C-section) are observed in IBD patients regardless of perianal status [13], despite the known potential morbidity, risk of abdominal organ injury as well as risk of fistulization in the surgical wound [12,15].The European Crohn's disease and colitis (ECCO) guidelines recommend standard vaginal delivery in patients with mild or quiescent disease and C-section only in women with active perianal disease [16].However, in many other clinical situations, such as prior or inactive perianal disease, no guidelines are currently available.The method of delivery in relation to the history of perianal disease, previous pelvic surgery, and the phenotype of the disease, continues to be a subject of debate [16,17].
The objective of our study was to analyze whether vaginal or C-section delivery impacted post-partum perianal disease in patients with IBD.We hypothesized that the use of vaginal or C-section would not alter perianal manifestations post-delivery in the setting of previous perianal disease that had healed; however, in patients with active perianal disease, C-section will be protective to prevent worsening any ongoing perianal disease activity.

Methods
Following approval by the Cleveland Clinic Institutional Review Board, all consecutive eligible female patients with IBD who delivered either vaginally or by C-section were identified in the Cleveland Clinic database from 1997 to 2022.A chart review of the electronic medical records was performed for every potential case to confirm the diagnosis of IBD, prior perianal involvement defined as the presence of one or more perianal and/or rectovaginal fistula during the course of IBD, no prior perianal involvement, perianal flare after delivery defined as an active disease in one or more perianal and/or rectovaginal fistulas, previously present or diagnosed de novo, as well as delivery method.The diagnosis of IBD was made using standard clinical, radiographic, endoscopic, and histological criteria.
Electronic medical records were reviewed for age, gender, ethnicity, age at diagnosis, smoking, previous IBD surgeries, perianal disease, location, and duration of IBD.Chart review was performed specifically for this study by the physician researchers and could be up to a few years after the pregnancy, allowing for a sufficient duration of follow-up.The Montreal classification was used to categorize disease behavior (inflammatory, structuring, or penetrating), location (ileal, colonic, ileocolonic), and severity of CD patients whereas ulcerative colitis (UC) patients were classified based on disease extent (proctitis, left-sided, or extensive pancolitis).Therapy modalities were classified as oral corticosteroid, biologics (adalimumab, certolizumab, infliximab, natalizumab, ustekinumab, vedolizumab), immunomodulator therapy What is new here?Comparative study on perianal involvement and the delivery method in inflammatory bowel disease patients treated in a world reference center with extensive experience.What is new here?Providing knowledge for the standardization of the delivery method in patients with Inflammatory Bowel Disease and perianal involvement.

Brief summary
Comparative study on perianal involvement and delivery method in inflammatory bowel disease patients treated at a world-reference center, concluding cesarean section doesn't protect from ongoing disease activity post-delivery, but it should be recommended for women with active involvement.Keywords Perianal Disease, Inflammatory bowel Disease, Cesarean section, Vaginal delivery (6-mercaptopurine, azathioprine, cyclosporine, methotrexate) and other medications (mesalamine, ozanimod, tofacitinib, antibiotic) prior, during or after delivery.
Perianal disease was classified as either a perianal or rectovaginal fistula and/or perianal abscess.Skin tags, hemorrhoids, and fissures were not included as 'perianal disease' in this series.Perianal disease was considered present following delivery if the patient had evidence of clinically active disease on physical exam (evidence of abscess, draining fistula), or radiology.The necessary treatment in these cases, whether surgical or medical, was also recorded.Delivery method was classified as either C-section or vaginal delivery.The indication for C-section was also collected, either for obstetric reasons, IBD-related, or due to patient preference.Any complications during labor were recorded, including lacerations, episiotomy, and instrumental.

Results
We identified a total of 190 patients with an established diagnosis of CD (n = 154; 81%) or UC (n = 36; 19%) who had at least 1 successful delivery at our institution between 1997 and 2022.The mean age (± SD) of IBD diagnosis was 21.5 +/-7.0 years, with a mean duration of disease of 9.0 ± 6.0 years.The mean age of delivery was 30.0 ± 5.0 years, with an average pregnancy number of 1.7 ± 0.6.More than a third of the cohort (n = 71; 37%) had had prior abdominal surgery for IBD indications, and 51 (27%) patients reported a lifetime history of smoking.Forty women (21%) had a history of perianal disease prior to delivery, with over half (n = 28; 70%) having undergone surgery for their perianal fistula in the past.(Table 1).
A total of 322 deliveries were recorded of which 169 (52%) were vaginal deliveries and 153 (48%) were by C-section.In more than half of the cases (n = 79; 51%) the indication to perform a C-section was based on an indication related to IBD (previous perianal disease, previous surgery for IBD, or by indication of the multidisciplinary team due to high risk of perianal involvement).A total of 21(6.5%)deliveries were followed by a perianal flare (Table 2).At a median of 65 (IQR 38-95) months of follow-up, 19 women (10%) experienced 21/322 (6%) post-partum perianal disease flares within two years after delivery.All patients (n = 19; 100%) who presented a flare had CD, without any UC.In the case of patients who presented a flare, 7 (37%) had a stricturing behavior, 6 (32%) had a nonpenetrating/nonstricturing behavior and 6 (32%) had a penetrating behavior in the Montreal classification.Of the total patients 15 (79%) had previously known perianal disease, 14 (74%) had ileocolonic involvement in Montreal Classification, and 11 (58%) underwent previous abdominal surgery for IBD.The median time interval between delivery and the flare was 2 (0-13) months, with the majority of cases (n = 17, 89%) occurring after the sixth week postpartum.All cases the diagnosis was made by physical examination, requiring surgical intervention in nearly all cases (n = 20; 95%) (Table 3).

Discussion
Given the peak age of diagnosis, IBD frequently affects women in their reproductive years.Thus, investigating how the mode of delivery may affect perianal phenotypes of IBD is critical [18][19][20][21][22].However, the factors that affect the evolution of perianal disease in the post-partum period, and how modes of delivery affect post-partum disease activity, are not well defined.In our retrospective cohort of 190 women at a tertiary referral center, we found 19 women (10%) experienced a perianal flare within two years post-partum.The majority of these flares occurred after a C-section.
It is important to understand the risk factors for a post-partum perianal disease flare as that may influence the mode of delivery counseled by the physician.Interestingly, we noted a significant association between abdominal involvement of IBD and the risk of developing a post-partum perianal flare.On the univariate and multivariate linear regression analyses, previous abdominal surgery for IBD and ileocolonic involvement were found as significant independent risk factors for a post-partum perianal flare; there was a significantly greater proportion of patients with ileocolonic involvement in the group that had a flare.This possibly indicates that these types of patients have more aggressive presentations of IBD and therefore greater perianal involvement.In addition, previous perianal disease was also shown to be a risk factor for experiencing a post-partum flare.These results are consistent with what has been published previously [13,23,24].Similar to our study, in a prior retrospective study of 114 women with CD, progression of perianal disease at 2 years post-partum was significantly more frequent in women with prior perianal disease as compared to women without it, respectively 7/27 (26%) and 6/87 (7%) (Odds ratio 4.7; 95% CI 1.4-15.6)[25].It can be expected that those with prior ileocolic disease and perianal disease will have increased severity of CD and their associated manifestations, thus likely to present with recurrent severe perianal disease post-partum.
Another key risk factor was the status of perianal disease involvement during pregnancy.Patients with either quiescent perianal disease or active perianal disease had significantly higher rates of post-partum perianal activity.More than half of the perianal flare group had active or quiescent disease.Those without a post-partum flare had much lower rates of active or quiescent disease.It is reasonable to suggest that if the disease is present regardless of activity, it will worsen in the post-partum period.Similar results were published by Ilnyckyji et al. in 54 vaginal births with perianal disease; 4 of 15 reported active perianal disease at birth, and all reported worsening of perianal symptoms post-partum [2].Based on this, the European Crohn's and Colitis Organization (ECCO) guidelines recommends a C-section for those women with active perianal disease [6].
Interestingly, our series found a significant increase in the risk of having a perianal disease flare when exposed biologic (OR, 4.4; 95% CI, 1.4-13.6;p < 0.011) or antibiotic (OR, 19.6; 95% CI, 7-54; p < 0.001) treatment before, during or after delivery.This finding may reflect that patients with more severe gastrointestinal involvement and perianal disease are more likely to be on medical therapy.A recent meta-analysis showed five studies (2155 patients) [20,[26][27][28][29] assessed biologic exposure during pregnancy and demonstrated a trend towards an increased odds of post-partum disease activity (OR, 1.41; 95% CI, 0.74-2.70;I2 = 73.0%;Tau2 = 0.36; χ2 = 15.05,P = 0.01) in the setting of biologic exposure [24].Again, it is difficult to determine if this is an association reflecting ongoing disease burden or, less likely, causative as a result of medication exposure.
Interestingly, vaginal delivery was identified as an independent protective factor for worse perianal disease outcomes after delivery (OR, 0.19; 95% CI, 0.06-0.61;p < 0.005).This finding should be interpreted with caution, as it is possibly influenced by the fact that patients with no perianal involvement and milder disease symptoms were those who underwent a vaginal delivery as compared to patients with ongoing disease activity were typically advised to have a C-section.Similar results were found in previous studies, showing odds of post-partum perianal CD disease activity significantly increased in those undergoing C-section delivery compared with vaginal delivery (OR, 2.34; 95% CI, 1.21-4.51;I2 = 0%) [30,31].Again, this may be because patients with severe disease underwent C-section more often than a vaginal delivery.Regardless, it is critical to underscore that a C-section is not necessarily protective of worsening perianal disease; thus, perianal disease may be independent of mode of delivery.
Finally, focusing on the comparative analysis between the cohort undergoing C-section versus vaginal delivery, there were several significant differences.First, more than half of the women in the C-section group (77.50%) had undergone previous abdominal surgery related to IBD, while only 48 (28%) vaginal deliveries had undergone previous abdominal surgery (p < 0.001).Perianal disease before delivery was also more frequent in cases in which a C-section was indicated (79% vs. 8%), and in 21% C-sections were recommended for perianal disease.Interestingly, the number of post-partum flares was significantly higher in the C-section group (11% versus 2%) which, again, maybe a reflection that patients with more severe IBD or with perianal involvement have a C-section recommended.These findings are consistent with a recent meta-analysis [24] of five studies (505 patients) assessing the impact of the mode of delivery on postpartum IBD activity [27,[29][30][31][32] and reported no difference in the odds of post-partum disease activity (OR, 1.32; 95% CI, 0.71-2.45;I2 = 50.0%;Tau2 = 0.23; χ 2 = 8.04; p = 0.09).However, when including only studies with a low risk of bias, a decreased odds of post-partum disease  We recognize several potential limitations in our study.First, since the data source is a cohort of women from a tertiary care IBD center, these patients may have a more severe course of disease with a greater number of previous surgeries and biologic therapy.Therefore, the results may not be generalizable to the entire IBD population.Second, it was a retrospective cohort study, so the results could be influenced by missing information and incomplete documentation.Due to this, our findings may underestimate the rate of post-partum flares due to inadequate documentation of care both inside and outside our system.In addition, the fact that a C-section is classically indicated for patients with perianal disease before pregnancy could influence results, so they must be interpreted with caution.Finally, we did not include nonpregnant patients as controls, since our main objectives were to identify the risk factors associated with flares  after delivery, rather than comparison with external controls.
In conclusion, C-sections may not be protective of worsening perianal disease but should be recommended for women with active perianal involvement.However, it should not be expected that a C-section will result in no perianal disease progression.

Table 1
Characteristics of the cohort before delivery

Table 1 (continued) All deliveries (n = 322) Deliveries followed by flare (n = 21) Deliveries not followed by flare (n = 301) p value
Categorical variables are number (percentage).Continuous variables are mean +/standard deviation or median (interquartile range), as appropriate

Table 2
Delivery characteristics

Table 3
Post-delivery perianal flaresCategorical variables are number (percentage).Continuous variables are mean +/standard deviation or median (interquartile range), as appropriate

Table 4
Univariate and multivariate linear regression analyses with generalized estimating equation model for factors associated with Crohn's perianal flare after delivery

Table 5
Route of delivery: C-section vs. VaginalCategorical variables are number (percentage).Continuous variables are mean +/standard deviation or median (interquartile range), as appropriate