Biologic Therapies for the Treatment of Post-ileal Pouch Anal Anastomosis Surgery Chronic Inflammatory Disorders: Systematic Review and Meta-analysis

Abstract

Background

Chronic inflammatory disorders after ileal pouch-anal anastomosis (IPAA) surgery are common. These include chronic pouchitis (CP), Crohn’s disease (CD) of the pouch, prepouch ileitis (PI) and rectal cuff inflammation (cuffitis). The aim of this study was to evaluate the efficacy of biologic therapies in treating these disorders.

Method

Systematic review of all published studies from inception to August 1, 2021 was performed to investigate the efficacy of biologic therapies for post-IPAA chronic inflammatory disorders. The primary outcome was the efficacy of biologic therapies in achieving complete clinical response in patients with IPAA.

Results

A total of 26 studies were identified including 741 patients. Using a random-effect model, the efficacy of infliximab in achieving complete clinical response in patients with CP was 51% (95% CI, 36 to 66), whereas the efficacy of adalimumab was 47% (95% CI, 31 to 64). The efficacies of ustekinumab and vedolizumab were 41% (95% CI, 06 to 88) and 63% (95% CI, 35 to 84), respectively. In patients with CD/PI, the efficacy of infliximab in achieving complete clinical response was 52% (95% CI, 33 to 71), whereas the efficacy of adalimumab was 51% (95% CI, 40 to 61). The efficacies of ustekinumab and vedolizumab were 42% (95% CI, 06 to 90) and 67% (95% CI, 38 to 87), respectively. Only one study involved patients with cuffitis.

Conclusion

Ustekinumab, infliximab, vedolizumab and adalimumab are effective in achieving complete clinical response in post-IPAA surgery chronic inflammatory disorders. More studies are needed to determine the efficacy of biologics in cuffitis.

INTRODUCTION

Proctocolectomy remains the therapy of choice for patients with ulcerative colitis (UC) who are refractory to medical therapy, and develop unresectable colonic dysplasia or colorectal malignancy (1–3). The lifetime risk of colectomy in patients with UC remains between 10% and 30% despite the era of biologic therapy (4). Proctocolectomy followed by ileal pouch anal anastomosis (IPAA) is one of the preferred surgical approaches (5). However, chronic inflammatory complications following (IPAA) surgery are not uncommon (6,7).

Chronic pouchitis (CP), Crohn’s disease (CD) of the pouch, prepouch ileitis (PI) and rectal cuff inflammation (cuffitis) are some of the chronic inflammatory disorders post-IPAA. Although inflammatory disorders can be idiopathic, it is important to recognize that these are a wide spectrum of diseases and conditions which can emerge in the pouch (8). Distinguishing between pouchitis and other pouch inflammatory disorders can be challenging and not always clearly described in the available literature.

Apart from conventional treatment with antibiotics, 5-ASAs or steroids, biologic therapies have been used to treat these conditions that are refractory to initial therapy. However, the efficacy of each biologic for each of these conditions remains unclear. We performed a systematic review and meta-analysis to better understand the magnitude of the effect of each biologic therapy to tailor our practice accordingly.

METHODS

This systematic review and meta-analysis were conducted using the methods described in the Cochrane Handbook of Systematic Reviews and reported according to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) statement (9). MOOSE Meta-analyses Of Observational Studies in Epidemiology guidelines were also followed (10) (Supplementary Files 2 and 3).

Eligibility Criteria

Randomized, placebo-, or active comparator-controlled trials, cohort studies, observational studies and case series reporting biologic therapies used post-IPAA surgery were included. Adult patients (age ≥18 years) with post-IPAA chronic inflammatory disorders who were refractory to 4 weeks of conventional therapy were included. Our study focus is biological pharmacological treatments alone or in combination with other agents. We excluded trials studying only paediatric patients (age, <18 years), case reports, editorials and correspondences, studies that did not evaluate a biological agent, and those where data could not be extracted.

Definitions and Outcomes

CP, CD of the pouch, PI and rectal cuff inflammation were considered to be post-IPAA chronic inflammatory disorders. CP was defined as inflammation of the pouch refractory to 4 weeks of first-line therapy (antibiotics, 5-ASAs and steroids), whereas CD of the pouch was defined as the presence of one of the following: non-anastomotic fistula involving the perineum or small bowel, non-anastomotic stricture of the pouch, skip ulcerations of the pouch with or without extension to the afferent limb, or histological features suggestive of CD. PI was defined as chronic inflammation of the afferent limb of the pouch. For the purpose of this study, we have reported and analyzed data reporting outcomes in CD and PI jointly.

The primary outcome measure was the efficacy of biologics in inducing complete clinical response, as defined by each study, after the failure of 4 weeks of first-line therapy. The main outcome was evaluated at 26 weeks or less. The definitions of complete clinical response used in each study were used for the extraction of the data. Patients who experienced partial responses were not included.

Search Strategy, Data Extraction and Outcomes

Literature searches were conducted by two authors (M.S. and F.A.) using MEDLINE, Embase, Scopus and Cochrane Central Register of Controlled Trials databases were searched from inception to August 1, 2021, using predefined strategies (Supplementary Table 1). Our search strategies were designed with the help of a librarian. The search was restricted to English language publications involving humans. English conference proceedings were searched [World Congress of Gastroenterology, American College of Gastroenterology, Canadian Digestive Disease Week, Digestive Disease Week, European Crohn’s and Colitis Organization congress, and United European Gastroenterology Week]. Furthermore, clinical trials databases (www.clinicaltrials.gov and International Randomized Standard Clinical Trial [IRSCT] Register) were searched. Google scholar was also searched for unindexed trials.

Data extraction and quality control were done independently by two reviewers (M.S. and F.A.). Discrepancies were resolved by a third reviewer (T.B.). The same two authors extracted information from the studies. Extracted information included baseline characteristics, study design, risk of bias, year and country of publication, intervention, outcomes and mean follow-up duration using a standardized excel spreadsheet.

Statistical Analysis

Meta-analysis methods were used to pool the percentage of patients who achieved complete clinical responses from the various studies. Statistical analysis was conducted using Stata software (version 13; StataCorp, College Station, TX, USA). Prevalence and 95% confidence interval were estimated using random-effects models assuming between- and within-study variability. The analysis was performed for all studies combined and then separately for each type of inflammatory disorder.

I² statistic, which ranges from 0% to 100%, was used to quantify the relative amount of observed heterogeneity. An I² value less than 30% indicates low heterogeneity, whereas a range of 30% to 75% indicates moderate heterogeneity and high heterogeneity was defined as I² > 75%. Sources of heterogeneity were explored by performing multiple sensitivity analyses.

Sensitivity Analysis

To identify potential sources of heterogeneity, different sensitivity analyses were performed: (i) location where the study was performed (within North America compared to outside countries), (ii) Quality of studies: moderate to high-quality studies were only included, defines as modified Newcastle-Ottawa Scale (mNOS) (11) of 4 and above and (iii) study published after May 2014, as vedolizumab and ustekinumab were approved by U.S. Food and Drug Administration (FDA) after that time period. It is possible that patients on these medications may have received anti-TNF therapy, i.e., are anti-TNF experience.

Risk of Bias and Study Quality

To assess the risk of bias and quality of the included studies, two authors (M.S. and F.A.) independently used the Cochrane risk of bias tool for randomized controlled trial and ROBINS-I for assessing the risk of bias in non-randomized studies of interventions (12). By using these assessment tools, studies were classified as being unclear or low or high risk of bias. Seven domains: random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective outcome reporting and other potential sources of bias are included in this tool.

The quality of all included studies was assessed using the modified Newcastle-Ottawa Scale (mNOS). Three domains were assessed by using mNOS: selection, compatibility and outcome. Study quality was defined as low (score of 0 to 3), moderate (score of 4 to 6) and high (score of 7 and 8).

RESULTS

Search Results

Our search identified 611 studies (Figure 1). After exclusion of duplicates and selection based on our prespecified inclusion and exclusion criteria, 26 studies were included in the main analysis. All studies were observational except for one, which was a randomized control trial. Thirteen studies were published in North America, and the rest were published in different countries in Europe.

Figure 1.

PRISMA flow chart outlining the search process for selecting the studies included in this systematic review with meta-analysis.

Patient Characteristics

The selected studies included a total of 741 patients diagnosed with one of the post-IPAA chronic inflammatory disorders (Table 1). Among those patients, 373 had CP, 277 had CD, 48 had PI and 4 had cuffitis.

Table 1.

Summary of included studies

Study	Total number of patients	Male gender (n)	Mean age	Biologic agent	Disease type*	Follow-up (months)
Infliximab
Barreiro-de Acosta et al. (1)	33	18	45	Infliximab	CP	13
Segal et al.	34	19	25	Infliximab	CP, CD/PI	9.3
Viscido et al.	7	3	30	Infliximab	CP	2.5
Calabrese et al.	10	6	39.2	Infliximab	CP	7.3
Colombel et al.	26	7	32	Infliximab	CD/PI	21.5
Ricart et al.	7	2	38	Infliximab	CD/PI	5
Haveran et al.	22	14	27.8	Infliximab	CD/PI	97
Kelly et al.	42	11	32.6	Infliximab	CP	12
Viazis et al.	7	3	37.1	Infliximab	CP	36
Ferrante et al.	28	14	26	Infliximab	CP, CD/PI	Not reported
Segal et al. (ileitis)	29	17	53	Adalimumab/infliximab	CD/PI	21
Adalimumab
Shen et al.	17	12	36	Adalimumab	CP, CD/PI	2
Li et al.	48	29	25.3	Adalimumab	CD/PI	25
Kjaer et al.	13	2	40.1	Adalimumab	CP	3
Barreiro-de Acosta et al. (2)	8	5	42	Adalimumab	CP	13
Verstockt et al.	33	13	34.4	Adalimumab	CP	3.5
Robbins et al.	92	50	32	Adalimumab	CD/PI	Not reported
Gionchetti et al.	12	7	32.6	Adalimumab	CP	Not reported
Ustekinumab
Dalal et al.	46	24	39.6	Ustekinumab	CP, CD/PI, cuffitis	4
Ollech et al.	24	14	35.6	Ustekinumab	RP	12.6
Weaver et al.	56	24	44.1	Ustekinumab	CP, CD/PI,	12
Vedolizumab
Khan et al.	12	3	41	Vedolizumab	CD/PI	Not reported
Harrison et al.	13	Not reported	50	Vedolizumab	CP	Not reported
Gregory et al.	83	38	42.1	Vedolizumab	CP	12
Bar et al.	20	12	22.5	Vedolizumab	CP	Not reported
Singh et al.	19	10	26.7	Vedolizumab	CP	3

CP, Chronic pouchitis; CD/PI, Crohn’s disease of the pouch/prepouch ileitis.

The mean age was 35 (±7), and 366 (53.7%) participants were males. The mean disease duration was 9.9 years, whereas the mean follow-up duration was 16.4 months. Two hundred sixty-nine patients were treated with infliximab, 143 patients with vedolizumab, 164 patients with adalimumab and 66 patients with ustekinumab. The rest received different non-biologic treatments (controls) or were lost to follow up.

Main Outcomes

The definition of complete clinical response varied from study to study, ranging from a complete resolution of symptoms or return to normal function with non-bloody bowel movements without pain, urgency or frequency to complete mucosal healing of the pouch as per endoscopic findings.

Using a random-effect model, the efficacy of infliximab in achieving complete clinical response in patients with CP was 51% (95% CI, 36 to 66), whereas the efficacy of adalimumab was 47% (95% CI, 31 to 64). The efficacies of ustekinumab and vedolizumab were 41% (95% CI, 06 to 88) and 63% (95% CI, 35 to 84), respectively (Figure 2).

Figure 2.

Forest plot showing the overall efficacy of biologic therapies in achieving complete clinical response in patients with chronic pouchitis.

In patients with CD/PI, the efficacy of infliximab in achieving complete clinical response was 52% (95% CI, 33 to 71), whereas the efficacy of adalimumab was 51% (95% CI, 40 to 61). The efficacies of ustekinumab and vedolizumab were 42% (95% CI, 06 to 90) and 67% (95% CI, 38 to 87), respectively (Figure 3).

Figure 3.

Forest plot showing the efficacy of biologic therapies in achieving complete clinical response in patients with Crohn’s disease of the pouch or prepouch ileitis.

Sensitivity Analysis and Heterogeneity

The overall I² value was 67% for CD/PI analysis, which is considered medium heterogeneity (13). However, the overall I² value was 76% for CP analysis. Due to high heterogeneity among studies, we performed an additional sensitivity analysis, post hoc analysis, by removing outliers’ studies. After performing sensitivity analysis by removing outlier’s studies, the overall I² value decreased to 49% for CD/PI and 51% for CP. After restricting the analysis to studies performed in North America, study period, or high-quality studies the pooled efficacy of individual biologic therapies did not differ substantially as described below:

Patients with CP:

• i.After removing outliers studies, pooled efficacy of infliximab in achieving complete clinical response in patients with CP was 54% (95% CI, 41 to 66). Specifically, the pooled efficacy of ustekinumab in achieving complete clinical response was 68% (95% CI, 44 to 85), whereas the efficacy of vedolizumab was 72% (95% CI, 60 to 81). Finally, the efficacy of adalimumab was 47% (95% CI, 31 to 64) (Figure 4).
• ii.Moderate or high-quality studies only: the efficacy of infliximab in achieving complete clinical response in patients with CP was 44% (95% CI, 28 to 62). The pooled efficacy of adalimumab in achieving complete response was 46% (95% CI, 29 to 65), whereas the efficacy of ustekinumab was 41% (95% CI, 6 to 88). Finally, the efficacy of vedolizumab was 31% (95% CI, 7 to 74) (Supplementary Figure 1).
• iii.North America only: the pooled efficacy of infliximab in achieving complete clinical response in patients with CP was 54% (95% CI, 43 to 64). Specifically, the pooled efficacy of ustekinumab in achieving complete clinical response was 41% (95% CI, 6 to 88), whereas the efficacy of vedolizumab was 46% (95% CI, 15 to 81) (Supplementary Figure 2).
• iv.Studies published after May 2014 only: Sensitivity analysis performed based on study period showed that pooled efficacy of infliximab in achieving complete clinical response in patients with CP was 47% (95% CI, 38 to 57). Specifically, the pooled efficacy of ustekinumab in achieving complete clinical response was 41% (95% CI, 6 to 88), whereas the efficacy of vedolizumab was 63% (95% CI, 35 to 84). Finally, the efficacy of adalimumab was 47% (95% CI, 31 to 64) (Supplementary Figure 3).

Figure 4.

Forest plot after performing sensitivity analysis by removing outliers’ studies (chronic pouchitis).

Patients with CD/PI:

• i.After removing outliers studies, pooled efficacy of infliximab in achieving complete clinical response in patients with CP was 57% (95% CI, 44 to 70). Specifically, the efficacy of ustekinumab in achieving complete clinical response was 80% (95% CI, 66 to 91), whereas the efficacy of vedolizumab was 67% (95% CI, 35 to 90). Finally, the efficacy of adalimumab was 51% (95% CI, 40 to 61) (Figure 5).
• ii.Moderate or high-quality studies only: after removing the low-quality studies (mNOS of 3 or less), the efficacy of infliximab in achieving complete clinical response in patients with CD/PI was 39% (95% CI, 21 to 60). The pooled efficacy of adalimumab in achieving complete response was 47% (95% CI, 36 to 59), whereas the efficacy of ustekinumab was 42% (95% CI, 6 to 90) (Supplementary Figure 4).
• iii.North America only: after removing the low-quality studies (mNOS of 3 or less), pooled efficacy of infliximab in achieving complete clinical response in patients with CD/PI was 63% (95% CI, 50 to 75). Specifically, the pooled efficacy of ustekinumab in achieving complete clinical response was 42% (95% CI, 6 to 90), whereas the efficacy of adalimumab was 53% (95% CI, 41 to 64) (Supplementary Figure 5).
• iv.Studies published after May 2014 only: Sensitivity analysis performed based on the study period showed that pooled efficacy of infliximab in achieving complete clinical response in patients with CD/PI was 34% (95% CI, 12 to 67). Specifically, the pooled efficacy of ustekinumab in achieving complete clinical response was 42% (95% CI, 6 to 90). Finally, the efficacy of adalimumab was 54% (95% CI, 27 to 79) (Supplementary Figure 6).

Figure 5.

Forest plot after performing sensitivity analysis by removing outliers’ studies (Crohn’s disease of the pouch or prepouch ileitis).

Study Quality and Risk of Bias

The median mNOS score was 4, with scores ranging from 2 to 8 (Supplementary Table 2). In terms of risk of bias, most studies were judged to have a low risk of bias using the Cochrane risk of bias tool for randomized controlled trials and ROBINS-I, while six studies had a moderate risk of bias (Supplementary Table 3).

Publication Bias

Supplementary Figure 7 shows a funnel plot of publication bias. Based on the visual examination of the plot, the symmetrical distribution of the studies on the funnel plot suggests no publication bias.

DISCUSSION

There are several complications that could occur following ileal pouch-anal anastomosis (IPAA), of which CP is the most common. A study showed that the 5-year cumulative incidence of CP was approximately 40% (14). The total number of patients diagnosed with CP included in our study was 50.3%. A recent study evaluated the burden of inflammatory conditions of the pouch in a tertiary hospital and found that up to 30% of patients developed CP that required reinitiation of biologic therapy (15). Furthermore, the same study showed that even among patients originally diagnosed with UC, 10% of patients can be diagnosed with CD of the pouch. Patients with CD were found to require early biologic therapy soon after IPAA to prevent further complications (16). This emphasizes the importance of understanding the effectiveness of the different biologic therapies in inducing a complete response in patients with inflammatory disorders of the pouch.

In this meta-analysis, the overall efficacy of biologics in achieving complete clinical response in patients with CP was 53%, and the efficacy of infliximab was 51%. These results are similar to the findings of a meta-analysis by Huguet et al.(17). Another meta-analysis by Segal et al. (18) explored the efficacies of multiple therapies in achieving remission in patients with CP, and similar to our study, the authors found an overall remission rate with biologic therapy of 53%. Segal et al.’s study also suggested that antibiotics have higher efficacy in the treatment of CP. While antibiotics are generally considered first-line therapy, the treatment of CP is often a challenge and requires multiple therapies to achieve remission. A more recent meta-analysis (19) investigated the efficacy of biologics in patients with CP. Their findings demonstrate that biological therapy is a safe and effective short-term as well as a long-term treatment. Clinical improvement in symptoms was reported by 62% of patients treated with infliximab, 58% of patients treated with adalimumab. Because our definition of clinical response was stricter and included patients with complete response only, the efficacies of biologics in patients with CP were lower with infliximab and adalimumab efficacies at 51% and 47%, respectively in patients with CP.

In addition to investigating the efficacy of biologic therapy in patients with CP, our study explored the efficacy of biologic therapy in other inflammatory disorders of the pouch, including PI and CD of the pouch. Patients with PI accounted for 5.4% of the total patients included in our meta-analysis, a number similar to what was seen in two other studies that reported a 6% frequency of this complication among patients with IPAA (20,21). A study compromising 66 patients with PI and pouchitis found that the presence of PI is associated with an increased risk of development of CD-like complications that manifest as pouch failure, need for endoscopic or surgical interventions and need for immunosuppressive therapy (22). The authors of this study attributed the higher need for immunosuppressive therapy and surgical or endoscopic procedure in patients with PI, compared with pouchitis alone, to higher CD-like complications in this group. These findings underline the strong association of PI with disease complications and emphasize the need to treat early with the most effective therapy. Our analysis found that the pooled efficacy of biologic therapy in patients with CD/PI is 52%. PI in patients with IPAA is an uncommon occurrence and finding a large group of such patients to study is challenging. Conversely, it is debatable whether PI lies on the pouchitis spectrum or represents a more aggressive inflammatory pouch disease phenotype, like CD. Unlike Syal et al.’s study, other studies have shown that PI is not associated with the development of CD-like complications (20,21,23).

Gregory et al. (24) was a multicenter, retrospective study, that included 83 patients with CP treated vedolizumab. The authors of the study assessed partial clinical response as their primary outcome and complete clinical response as one of their secondary outcomes. Complete clinical response was defined as a complete return to normal function with reported normal, non-bloody bowel movements without pain, urgency, or increased nocturnal bowel movements and the absence of a fistula. About 40% of the enrolled patients achieved clinical response, however, only 12% achieved complete clinical response. It is important to note that more than half of the included patients had received at least one anti-TNF therapy before vedolizumab therapy. This study contributed to high heterogeneity in our analysis; however, after removing it from the analysis, heterogeneity decreased significantly. Furthermore, a recent phase 4, randomized, double-blind, placebo-controlled, multicenter study investigated the efficacy and safety of intravenous vedolizumab in patients with UC who developed chronic or recurrent pouchitis after IPAA. The primary outcome measure was clinical remission as measured by a reduction in the modified Pouchitis Disease Activity Index (mPDAI). The study included 102 patients and mPDAI clinical remission rates at week 34 were 35.5% with vedolizumab versus 17.6% in the placebo group (P = 0.043). The authors concluded that over 34 weeks, vedolizumab showed efficacy and tolerability in patients with CP (25).

One of the complications is a chronic inflammation of this rectal cuff, known as cuffitis. This is most often attributed to the recurrence of UC within the rectal cuff and occurs in up to 13% of patients after IPAA (26). A limited number of studies have been published regarding this inflammatory disorder, with even less studies focusing on treatment. This was reflected in our systematic review. Only one study (27) involved patients with cuffitis, limited to four patients treated with ustekinumab. Therefore, it was hard to draw any conclusion based on the limited data we found. Nonetheless, it is important to keep in mind that cuffitis and pouchitis may often coexist.

To the best of our knowledge, this is the largest and only systematic review and meta-analysis that explored biologic therapies' efficacy in all inflammatory disorders post-IPAA surgery, including PI and cuffitis. In addition, this study was performed by adhering to the highest standards, including MOOSE guideline and PRISMA statement. Furthermore, the inclusion of good quality studies with detailed extraction of data, and rigorous evaluation of study quality lend great credibility and strength to our systematic review and meta-analysis. Finally, this study is the first to investigate multiple biologic therapies other than anti-TNF agents. However, there are some limitations to this study. All studies, except one, were observational studies, with 19 out of 26 studies performed retrospectively. This can lead to a high risk of confounding and selection bias as well as high heterogeneity among the studies. To address it, we performed four sensitivity analyses to explore potential causes of high heterogeneity. Unsurprisingly, heterogeneity decreased after performing these sensitivity analyses (Figures 4 and 5). In addition, the definition of clinical response varied from study to study, which ranged from symptoms resolution to change in the Pouchitis Disease Activity Index (PDAI); hence it was difficult to make comparisons across studies. We addressed this limitation in our study by including complete clinical response as defined by each study, rather than partial response or remission. Finally, follow-up time varied greatly in the included studies. To limit this, we reported our outcomes in a standardized timeline of up to 26 weeks. Nevertheless, our study remains the most up to date and comprehensive review that evaluated the efficacy of biologics in all inflammatory disorders of the pouch.

CONCLUSION

Ustekinumab, infliximab, vedolizumab and adalimumab are effective in achieving complete clinical response post-IPAA chronic inflammatory disorders. Randomized control trials are needed to determine the precise efficacy of biologics in patients with chronic inflammatory disorders post-IPAA surgery. In addition, studies are needed to determine the efficacy of biologics in cuffitis.

Funding

No funding was received for this article.

Conflict of Interest

All authors declare no conflict of interest or financial support.

Author Contributions

F.A. and M.S. performed the search, A.C. and T.B. were involved in planning and supervised the work, A.C. was the third reviewer of the data collection and extraction, F.A. and M.S. performed the analysis, drafted the manuscript and designed the figures. A.C. and T.B. aided in interpreting the results and worked on the manuscript. All authors discussed the results and commented on the manuscript.

Data Availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.