Crohn's-like Ileal Pouch Illness and Ileal Pouch Salvage Strategies

Abstract

De novo Crohn's disease (CD) of the pouch or Crohn's-like Ileal Pouch Illness (CLIPI) is an increasingly common occurrence in an ever-growing ileal pouch population. Although currently undetermined if a subset of classic CD or a completely new entity, it primarily affects the prepouch afferent limb, pouch, and rectal cuff. Symptoms can mimic other more common disorders, such as pouchitis, and requires a thorough workup, including pouchoscopy with biopsy and often cross-sectional imaging, for the diagnosis to be made. There is an increased risk of long-term pouch failure in this population. Treatment is typically dependent upon the disease phenotype with surgical management considered in a step-up fashion. Medical management is primarily performed with “biologics,” such as antitumor necrosis factor agents, although data are limited due to the lack of randomized controlled trials. Surgical management for CLIPI can include endoscopic, anorectal, and abdominal approaches to assist as “pouch-salvage strategies.” The performance of advanced pouch-salvage techniques in the CLIPI population requires careful patient selection and should preferably be performed at high-volume pouch centers.

Since first described by Sir Alan Parks in 1978, the ileal pouch-anal anastomosis (IPAA) has been the preferred surgical treatment for patients with ulcerative colitis (UC) or familial adenomatous polyposis (FAP) who wish to avoid a permanent ileostomy. ^{1 2} In appropriately selected patients, the outcomes of an IPAA have been mostly positive. ^{3 4 5 6} Long-term failure is also relatively low, with one meta-analysis suggesting around only 9% pouch failure rates with >10 years follow-up. ⁷ The J-pouch is the most common modern pouch procedure, and it allows for a relatively normal quality of life (QoL) with frequent but predictable pouch emptying. ²

While IPAA has excellent outcomes in UC, patients with Crohn's disease (CD) have inferior outcomes, with as high as a five-fold increased risk of pouch dysfunction and failure. ⁸ Therefore, an ileal pouch is felt to be relatively contraindicated in this population. ¹ IPAA has been performed in highly selected CD patients with acceptable results, though it is not a standard practice. ⁹

A recent meta-analysis examined patients with isolated colon-only CD who received an intentional IPAA. ¹⁰ The results showed equivocal functional outcomes and short-term complications but a higher risk of pouch failure, up to 15% at a mean follow-up time of 69 months. ¹⁰ Pouch failure rates are worse in those with incidental CD found on colectomy histopathology or de novo CD.

Occasionally, an IPAA may be created inadvertently in CD patients initially diagnosed with UC or indeterminate colitis (IC). ^{2 11} CD may be incidentally found on postoperative histopathology or upon development of hallmark characteristics of CD, such as perianal disease, after initial surgery. ¹¹ Some patients may present with onset of CD symptoms months to years after IPAA and reversal of the diverting loop ileostomy.

Crohn's-like Ileal Pouch Illness

As the ileal pouch population grows, one important phenomenon is de novo CD of the pouch, also referred to as Crohn's-like Ileal Pouch Illness (CLIPI). Although unclear if this is a subset of classic CD or a completely new entity, CLIPI is defined as new onset Crohn's-like ulcerative small bowel inflammation or anorectal disease after ileal pouch creation. It is an increasingly common contributor of Crohn's-like symptoms after pouch creation. Though difficult to distinguish from other postoperative IPAA sequalae, CLIPI occurs in approximately 10 to 20% of patients and is associated with significant morbidity. ¹² It is characterized by symptoms such as inflammation, stricturing, and/or fistulizing disease of the pouch and afferent limb. ⁹

It is important to recognize and understand its true incidence, presentation, and treatment options. CLIPI most commonly affects the prepouch afferent limb, pouch, and rectal cuff, though some patients develop CD in other parts of the gastrointestinal tract as well. ^{9 12} Timing of presentation can be variable. Early fistulas and strictures may be iatrogenic rather than due to CD in the acute post-operative period. However, presentation of symptoms >1 year after after IPAA creation and anatomic distance from the anastomosis are suggestive of CLIPI. ^{11 12} In addition to the development of fistulae or nonanastomotic intestinal strictures, other common characteristics of CLIPI include systemic symptoms such as malnutrition, weight loss, persistent nausea and vomiting, concurrent autoimmune disorders, and extraintestinal manifestations such as uveitis and arthralgias. ⁹ Of note, endoscopic and histologic markers of CLIPI may mimic other inflammatory bowel disorders, which decreases their diagnostic utility. ⁹

Pathophysiology

Patients with CLIPI are classified based on phenotype into inflammatory (nonstricturing, nonpenetrating), fibrostenotic (stricturing), or fistulizing (penetrating) disease according to the Vienna and Montreal Classifications. ^{13 14} The pathophysiology of pouch-specific disease is still poorly understood, but current evidence suggests an interplay between postpouch genetic changes, microbial changes, and environmental insults. There is evidence of altered gene expression and dysbiosis of the pouch that may predispose to or cause CLIPI. ^{15 16} Transcriptional profiles of pouches in patients showed evidence of increased colon-associated genes and decreased ileum-associated genes compared to FAP pouches. ¹⁷ Fecal and anaerobic bacteria also demonstrate a time-dependent increase in the pouch microbiome. ¹⁸ Finally, there is evidence of chronic inflammation on histologic samples even prior to clinical symptoms. ^{18 19} Family history of CD also predisposes to developing CLIPI, suggesting a genetic component. ²⁰ Different phenotypes of CLIPI also have different risk factors, although these data are still limited, suggesting multiple pathophysiologic processes. ¹⁴ Ultimately, CLIPI is likely a result of complex interaction between the pouch construct, microbiome, and environmental influences in genetically susceptible patients.

Chronic inflammation leads to excessive and sustained repair response characterized by remodeling of the extracellular matrix and fibrosis. ^{21 22} While inflammation is known to be necessary for stricture formation, the transition from inflammation into fibrosis and stricture is poorly understood. As strictures develop, two processes occur in parallel; expansion of the smooth muscle and fibrosis. ²¹ Collagen deposition is seen in all layers of the involved bowel while smooth muscle expansion is responsible for the majority of the wall thickening. Epigenetic changes including DNA methylation leads to modification of gene expression furthering intestinal fibrosis. Factors in strictured tissues are believed to serve as damage signals that perpetuate fibroblast activation, proliferation, and expansion without normal apoptosis of fibroblasts and myofibroblasts. ²¹ The constitutively active fibroblasts and myofibroblasts are believed to cause a continuous cycle of matrix protection and degradation that ultimately leads to fibrosis. Fibrotic changes that develop during episodes of inflammation will persist even after inflammation resolves. ²¹

Penetrating CD also occurs in the background of inflammation but is characterized by the presence of fistulas, which are believed to originate from epithelial defects from ongoing inflammation. ²³ Intestinal epithelial cells undergo transition to mesenchymal cells which gives them enhanced migratory capacity. This in turn allows for penetration through the intestinal wall and localized tissue damage. ²³ The epithelial-to-mesenchymal transition also enhances activation of matrix metalloproteinase (MMP-3 and MMP-9) which exacerbates tissue damage. ²³ Additional soluble mediators, including tumor necrosis factor (TNF) and IL-13, lead to expression of molecules that promote cell invasiveness while inducing their own continued expression. ²³

Together, the nonphysiologic pouch microbiome, transcriptional changes, and chronic inflammation may manifest with a clinical presentation mirroring that of CD. ^{24 25} Further efforts to characterize this pathologic process are still ongoing.

Related Pouch Disorders

Rigorous diagnostic workups to distinguish CLIPI from other pouch disorders is critical as misdiagnosis can lead to treatment and possibly pouch failure. ¹¹ The most common complication is pouchitis, defined as inflammation of the mucosal lining of the pouch ( Fig. 1 ). It is seen in more than 50% of pouches and is generally responsive to antibiotics. Up to 10 to 20% of pouchitis patients will have resistance to antibiotics and should have CLIPI included in their differential diagnosis. ⁹

Fig. 1.

Pouchitis on endoscopic view.

Additionally, prepouch ileitis, defined as chronic active inflammation of the afferent limb of the ileal pouch with friability, erosions, or occasionally ulcers, should be distinguished from CLIPI. ²⁶ Between 4 and 6% of IPAA patients will develop prepouch ileitis, and it primarily presents in patients with existing pouchitis. ^{27 28} Notably, prepouch ileitis is associated with a worse prognosis than pouchitis. ²⁸ Several studies have shown a close correlation between prepouch ileitis and an increased likelihood of developing CD-like complications, thus arguing for its interconnectedness. ^{26 27} Some have even concluded that prepouch ileitis should be considered a defining feature of CLIPI. ²⁶ Although the exact nature of prepouch ileitis is unclear, its presence should raise clinical suspicion for CLIPI and be considered in treatment algorithms. ²⁶

Diagnosis

Initial Workup

Patients suspected of developing CLIPI should have a thorough history and physical examination performed. Features consistent with the development of CD of the pouch, including abdominal pain, weight loss, fatigue, and diarrhea should be identified. Known risk factors also include a family history of CD, longer duration of IPAA, smoking, a prepouch diagnosis of IC, and seropositivity of anti- Saccharomyces cerevisiae or anti-CBir1 flagellin antibodies. ²⁹ In addition, patients who develop early pouchitis (within 1 year of ileostomy closure) are associated with CLIPI development. ²⁴ Finally, the timing of symptoms in reference to pouch formation may also help to differentiate between CLIPI and other disease presentations. For instance, fistulas or strictures in the immediate perioperative period may be sequelae from surgical course rather than newly developed CD.

Pouchoscopy may be a useful tool to identify pouch pathology, though up to 50% of asymptomatic IPAA patients may have abnormal findings on endoscopy. ³⁰ Certain endoscopic phenotypes in different anatomic areas can be suggestive of different diagnoses (e.g., pouchitis vs. cuffitis). ²⁹ Biopsies should be taken of any suspicious lesions during a pouchoscopy as well as from the rectal cuff. Depending on the presentation, the urgency of the scope can vary from emergent to elective.

Cross-sectional imaging is another important adjunct in the evaluation of CLIPI. CT or MRI pelvis can be effective, particularly in the evaluation of fistulas or additional pelvic or perianal abscesses. A pouchogram can be also performed to evaluate for compliance, emptying, function, and strictures/fistulas. Finally, anorectal manometry or ultrasound can help with evaluation of sphincter or pelvic floor dysfunction and can be considered when working up functional problems of the pouch. ³¹

Diagnostic Criteria

A previous consortium has suggested the following features as suggestive of CLIPI: prepouch ileitis; late development of fistulas or abscesses (6–12 months after ileostomy takedown); segmental/skip lesions or strictures in pouch and small bowel; noncaseating, noncrypt rupture-associated granulomas on biopsy. ³¹ Notably, strictures located outside of the anastomosis, staple, or suture lines is also suggestive of CLIPI. ³² These findings depend on a mixture of clinical, histological, and radiologic findings to confirm a diagnosis. Transmural inflammation on its own is considered insufficient for diagnosis of CD of the pouch.

Differential Diagnosis

Pouchitis

Pouchitis from infectious or immune-mediated sources should be distinguished from CLIPI. A Clostridium difficile PCR, serum/tissue CMV, anti-IgG4, and pANCA assays are useful in identifying these etiologies when classic treatment with antibiotics fails. Pouchoscopy with biopsies should be performed as well. The Pouchitis Disease Activity Index can also be useful, as clinical symptoms, endoscopy, and histology do not independently correlate with a pouchitis diagnosis. ³³

Cuffitis

Inflammation of the rectal cuff causes cuffitis, which may present with similar symptoms to pouchitis. It is generally categorized as classic cuffitis, defined as remnant UC at the rectal cuff, or nonclassic, defined as non-UC etiologies. Risk of cuffitis is higher with cuffs greater than 2 cm. It is further classified by its response to treatment, categorized as topical mesalazine and corticosteroid-responsive, dependent, or refractory phenotypes. Refractory cuffitis is more likely to be nonclassic, with etiologies including pouch prolapse, CD, or ischemia. ³²

Prepouch Ileitis

Defined as inflammation of the distal 10 cm of afferent limb, isolated prepouch ileitis can be distinguished from CLIPI. Endoscopic findings include friability, erosions, or ulcers in the afferent limb of the pouch. These patients also tend to have acute ileitis, occasionally with chronic inflammation on histology. The vast majority of prepouch ileitis patients have co-occurring pouchitis, resulting in significant symptom overlap. The association of prepouch ileitis with CLIPI is evolving; it may be considered either a risk factor for CLIPI or part of its diagnostic criteria. ²⁶ In one study, resected afferent limb specimens in prior UC patients found that 56% had confirmed CD on pathology with 69% with a change of diagnosis to CD. ³⁴

Pouch Sinus

Pouch sinuses are thought of as contained anastomotic leaks that are confined to a blind-ending tract. They are frequently located at the pouch-anal anastomosis in the presacral space. Patients may present with perianal pain, pelvic pressure/discomfort, and tailbone pain. More complex sinuses can have systemic symptoms such as fever, weight loss, and anemia. ³⁵ On pouchoscopy, careful attention at the anastomosis and tip of the “J” are important for identification of sinuses. Pouchograms with water-soluble contrast or MRIs are also useful imaging modalities to identify sinuses. ³⁵

Irritable Pouch Syndrome

Functional disorders of the pouch may present with pouch dysfunction and additional symptoms. Often these patients present with isolated clinical symptoms without any endoscopic or histologic abnormalities. In these patients, the dysfunction is not related to pouch stricture. ³⁶

Pouch Failure in Crohn's-like Ileal Pouch Illness

Pouch failure in a patient is defined as post-IPAA patients who undergo pouch resection, permanent diversion, or a redo pouch procedure. ⁷ Overall pouch failure rates range from 5 to 9.1% within 10 years and 9 to 18% after 10 years. ^{7 37 38} Despite higher complication rates than end ileostomy, many patients will opt for IPAA due to its contributions to improved body image and QoL. ³⁹ The presentation of pouch failure is variable and depends on the etiology, such as infection, mechanical/functional difficulties, or CLIPI. ⁴⁰ The development of pouch functional scores has also contributed to quantifiable measurements of QoL that may be used to identify pouch failure. ⁴¹

The most common cause of pouch failure historically was pelvic sepsis, such as from anastomotic leak, especially in the era when pouches were not routinely diverted at creation. ⁴² Additional risk factors include preoperative CD, strictures, chronic pouchitis, fistulas, and persistent pouch dysfunction. ^{42 43} Preoperative biologic use, CLIPI, and prior pouch revision have also been found to be associated with higher failure rates. ⁴³

More recently, CLIPI is being identified as an increasingly frequent cause of pouch failure at high-volume centers. In fact, pouch failure occurs in up to 60% of pouch patients with CD, especially fistulizing disease with perianal involvement. ¹¹ Another study reported that approximately half of CLIPI patients do not respond to standard medical (biologic) therapy and ultimately require definitive pouch excision. ⁹ Regardless of phenotype, close monitoring of treatment response is essential, should timely escalation of care be needed.

CLIPI is a cause of both early and late pouch failure. Early pouch failure is generally considered to be within 5 years of IPAA creation and diverting loop ileostomy reversal although there are no consensus guidelines. One such study in very long-term post-IPAA follow-up reported an average failure time of 11.4 years. ⁴⁴ Development of late CLIPI may be due to evolving disease phenotypes rather than a missed CD diagnosis. In one study, 40% of patients who developed CLIPI had dual pathologist agreement on UC as a diagnosis at time of colectomy. ⁴⁴ This is indicative of the vast range in timing at which CLIPI can both present and cause pouch failure.

Surveillance

The overall risk of pouch neoplasia is low at approximately 0.5 to 1%. However, certain factors such as a history of primary sclerosing cholangitis or chronic pouchitis increase this risk. ^{45 46} Given this, a pouchoscopy should be performed annually in patients with CLIPI for both disease activity assessment and neoplasia surveillance. ⁴⁶ Consideration should be made to note the extent of disease and obtain biopsies of the prepouch ileum, pouch body, and cuff (if present). Dependent on the extent of involvement of the prepouch ileum and the presence of strictures or fistulae, surveillance may also include imaging of the abdomen and pelvis annually.

Treatments

Similar to intestinal CD, CLIPI should be treated according to its phenotypical presentation, which may be inflammatory (9–82%), fibrostenotic (20–38%), and/or fistulizing/penetrating (29–63%). ¹¹ The inflammatory phenotype is likely to present with diarrhea and abdominal discomfort. Fibrostenotic CD is more likely to present with chronic and progressing obstructive symptoms. Fistulizing or penetrating CD can present with very complex disease including enteroenteric fistulas, abscess or perforation, or perianal disease. Treatment options include many traditional CD medications such as antibiotics, immunomodulators, TNF inhibitors, or non-TNF biologics (such as vedolizumab or ustekinumab). ^{11 32} In cases of medically refractory disease, surgery may be performed and is often performed in a staged approach. ¹¹ Patients with fibrostenotic or fistulizing/penetrating disease are more likely to need endoscopic or surgical procedures. ³⁵

Medical Management of Crohn's-like Ileal Pouch Illness

Recommendations for management of CLIPI are challenging due to the lack of randomized controlled trial data. A number of studies have evaluated the efficacy of biologic therapies for the management of CLIPI, but these have largely been limited by sample size and the inclusion of patients with chronic antibiotic refractory pouchitis (CARP).

Antitumor Necrosis Factor Therapy: Adalimumab and Infliximab

In a systematic review and meta-analysis of 21 studies which included 313 patients with CLIPI or CARP, adalimumab and infliximab achieved clinical remission in 52% of patients at 12 months. ⁴⁷ Specifically, remission rates at the end of induction and maintenance were 64 and 57% in patients with CLIPI, respectively. Remission was not significantly different among patients treated with infliximab versus adalimumab, and a subsequent meta-analysis noted no significant difference between infliximab or adalimumab monotherapy and combination therapy (infliximab or adalimumab plus immunomodulator). ⁴⁸ On the other hand, in a single-center study of 83 patients, those exposed to anti-TNF therapy pre- and post-IPAA had lower rates of clinical remission and higher rates of pouch failure. ⁴⁹ Judicious use of adalimumab and infliximab post-IPAA is overall recommended with consideration for past exposures to biologic agents and the risk of immunogenicity and nonresponse. ⁴⁹

Anti-Interleukin Therapy: Ustekinumab

In a retrospective study of 56 patients with CLIPI ( n  = 47) or CARP ( n  = 9), ustekinumab achieved clinical response in 83% at 6 months. Although ustekinumab did not achieve clinical remission in any patient, it did achieve antibiotic-free remission in 60% at 6 months. ⁵⁰ In a subsequent retrospective study of 46 patients with CLIPI ( n  = 36), CARP ( n  = 6), or cuffitis ( n  = 4), ustekinumab achieved clinical response in 80.4% at 8 to 16 weeks. Dose frequency intensification was required in 50% of patients—a median of 223 days after induction due to breakthrough symptoms on standard 8-week dosing. ⁵¹ Further data regarding the efficacy of selective anti-interleukin therapies, such as Risankizumab, are lacking at this time.

Anti-Integrin Therapy: Vedolizumab

In a retrospective study of 83 patients with CLIPI ( n  = 54) or CARP ( n  = 29), vedolizumab achieved clinical response in 71.1% and clinical remission in 19.3%. ⁵² There was no difference in clinical response or remission among patients with CLIPI or CARP, but patients who developed symptoms within 1 year of IPAA were less likely to have response to vedolizumab. ⁵²

Small Molecule Therapy: Tofacitinib

In a case series of 14 patients with CLIPI ( n  = 11) or CARP ( n  = 4), tofacitinib achieved clinical response in three (23%) patients at 3 months. An additional three patients (23%) achieved clinical response during follow-up, however nine (64%) patients ultimately discontinued tofacitinib. ⁵³ Further data regarding the efficacy of selective small molecule therapies, such as upadacitinib, are lacking at this time.

Surgical Salvage Strategies in Crohn's-like Ileal Pouch Illness

Endoscopic Interventions

Pouch stricture is a known complication for patients with CLIPI and can occur at either the afferent limb (pouch inlet), the pouch body, or at the pouch-anal anastomosis (pouch outlet). ⁵⁴ The principles of management for CD-related fibrostenotic strictures also apply to pouch strictures. Endoscopic approaches are less invasive with the goal of improving symptoms, delaying operative treatments, and preventing complications like obstruction. Outlet strictures can be treated mechanically with digital dilation or with Hegar or anoscopic dilators. For inlet or pouch body strictures, endoscopic interventions can be attempted by trained advanced endoscopists in conjunction with medical therapy or when medical therapies fail. ³² These interventions include endoscopic balloon dilation (EBD), endoscopic stricturotomy, endoscopic electroincision, and endoscopic stricturoplasty with clip placement.

EBD of pouch strictures is considered in patients with strictures less than 4 to 5 cm and can be performed every 3 to 12 months. ^{55 56} Consensus guidelines from the Global Interventional Inflammatory Bowel Disease Group include using a small caliber endoscope, performing stepwise dilation with balloons up to a maximum of 18 to 20 mm in size, using wire-guided balloons and retrograde dilation, avoiding stricture dilation if deep ulcers or fistula are present, and obtaining biopsies to rule out malignancy. ⁵⁶ A systematic review and pooled analysis of EBD for Crohn's strictures found that the risk of surgery increased by 8% for every 1 cm of additional stricture length balloon dilated. ⁵⁷ EBD for pouch strictures was first described in 2004 in a series of 19 patients with inlet or outlet strictures, including 11 patients with CLIPI, who all successfully underwent EBD with improved QoL and symptom scores up to 16 weeks' postdilation. ⁵⁸ One patient in this series with CLIPI had both inlet and outlet strictures and ultimately required pouch excision and diversion. Outcomes after EBD were further examined by the same research group in a larger study of 150 patients, including 62 patients with CLIPI, with a median follow-up on 9.6 (6–17) years. ⁵⁹ Among 406 total dilations performed, major adverse events were 1.5%, including bleeding ( n  = 4) and perforation ( n  = 2), with no procedure-related mortality. Overall, patients had high rates of pouch retention, up to 85.6% at 25 years. Another retrospective review identified 20 patients with 88 EBD performed over a 5-year period. ⁶⁰ Rate of technical success was 98%, and only one patient had a stricture-related pouch failure requiring surgery during the follow-up period. Among patients with outlet strictures, mechanical dilation can have a longer duration to repeat dilation compared to endoscopic dilation, but the two methods have similar rates of effectiveness and safety profile. ⁶¹ Overall, EBD is considered safe in Crohn's pouch strictures, but many patients require repeat dilation. Routine use of intralesional steroid injection following EBD is not recommended. ⁵⁶

Endoscopic stricturotomy involves using a transendoscopic needle knife or insulated-tip knife with or without electrocautery to incise the stricture radially or circumferentially. ^{55 62} There are limited studies that assess outcomes of endoscopic stricturotomy specifically for pouch strictures and none that specifically look at CLIPI. One case series reported on 85 patients who had a total of 127 pouch strictures treated with endoscopic stricturotomy with needle knife. There was a 100% technical success rate, but 77 (60.6%) patients required repeat procedures during the follow-up period (median 0.98 [interquartile range (IQR) 10.3–1.8] years) and 13 (15.3%) patients ultimately required surgery to address the stricture or endoscopic complications. ⁶³ In one study comparing patients who underwent endoscopic stricturotomy ( n  = 40) to patients treated with EBD ( n  = 160) for either pouch inlet or afferent limb strictures, postprocedural complication rates were the same between the two groups (21–22%). ⁶⁴ However, endoscopic stricturotomy had a higher risk of postprocedural bleeding. In contrast, EBD had a higher risk of perforation. No patients who underwent endoscopic stricturotomy had pouch failure at the end of the study period. Therefore, endoscopic stricturotomy is a potential nonsurgical approach to address strictures that are refractory to EBD.

Surgical Stricturoplasty

Surgical stricturoplasty is an effective treatment for fibrostenotic CD and indicated when strictures are refractory to endoscopic therapy or for long-segment strictures greater than 5 cm in length. ^{65 66} There are a few retrospective studies on surgical stricturoplasty for ileal pouch strictures. ³² One study compared outcomes among 164 patients with a J- or S-pouch who underwent either surgical stricturoplasty or EBD. ⁶⁷ Surgical stricturoplasty was performed in 16 patients via the Heineke–Mikulicz technique or transanal stricturoplasty using the Heineke–Mikulicz principle. Patients who underwent surgical stricturoplasty had a longer time interval to stricture recurrence or pouch failure compared to the EBD group. Overall, surgical stricturoplasty could be utilized in carefully selected patients as a strategy for pouch salvage in the setting of fibrostenotic strictures.

Anorectal Interventions

Perianal abscesses and fistula are frequently seen in CD of the pouch and can be difficult to treat. Antibiotic treatment is frequently used but there is no good evidence to support the efficacy of this treatment. Despite lack of data, long-term antibiotics and drainage are recommended in the consensus guidelines developed by the International Ileal Pouch Consortium. ³²

When present, perianal and/or peripouch abscesses should be drained, particularly before initiation of immunomodulators. This is done surgically, when possible, with the placement of setons or mushroom-tip drains in the abscess cavity to ensure complete drainage. ³² If collections are too deep to be safely accessed for incision and drainage, assistance from interventional radiology should be requested for percutaneous drainage. Fistulas, are initially managed medically but often require surgical treatment. If there is an associated abscess, setons are placed until resolution of infection. Transanal repairs of fistulas, including advancement flaps, have been described after resolution of inflammation; however these repairs are often avoided due to complications with poor wound healing and pouch dysfunction. ⁶⁸ Treatment of pouch-to-pouch fistula with endoscopic fistulotomy has also been described and may have efficacy in select patients. ³² This approach is best for short, shallow fistulas.

Injection of mesenchymal stem cells is a new treatment that shows promise in the treatment of perianal fistulizing CD for patients with IPAA. Patients are treated in a minimally invasive fashion, with direct local injection of mesenchymal stem cells. One phase II randomized clinical trial of 22 patients with CLIPI showed promising results for this treatment modality. The treatment group had a 31% rate of complete clinical and radiographic healing compared to 20% in controls. ⁶⁸ It had efficacy for perianal fistulas (57% complete healing rate) but no response for anovaginal fistulas (0% complete healing rate). ⁶⁸ Improvements to storage and transport of mesenchymal stem cells in addition to further phase trials will be needed prior to increased usage of this treatment modality.

Pouch–Vaginal Fistula Interventions

Pouch–vaginal fistulas are notoriously difficult to manage. A retrospective study of patients with CD-associated pouch–vaginal fistula treated with local repair, including advancement flap or transvaginal repair, showed low rates of fistula healing (22%) and high rates of pouch failure (52.7%). ³⁵ Use of gracilis muscle flaps has also been reported, with disappointing results. Due to limited efficacy, transanal flap or pouch advancement procedures are not recommended for CD-associated pouch–vaginal fistula. ³² Transvaginal repair advancement flaps may be used as alternative repair, with the benefit of direct access to the fistula without risk of sphincter damage. ⁶⁹

Afferent Limb Disease Interventions

When dilation is unsuccessful or not technically possible, afferent limb strictures can be treated with resection or bypass. Resection of the stricture and reanastomosis of a new afferent limb to the pouch is used to remove the stricture without having to revise the IPAA. ⁷⁰ Surgical technique should be tailored to intraoperative findings. One study describes creation of neo-ileal–ileal pouch anastomosis using a circular stapler to create an end-to-end anastomosis. ⁷⁰ When long segments of strictured bowel or acute angulations are encountered, bypass without resection can be considered. Distal pouch strictures, when present, can be dilated and proximal strictures can be addressed with stricturoplasty in the same operation. Bypass is preferred in cases with high risk of pouch devascularization or long strictures that may prevent the creation of a tension-free anastomosis. Sellers et al reported 100% pouch retention at a mean follow-up of 36.5 months using this technique. ⁷⁰

Surgical Fecal Diversion

For disease refractory to medical, endoscopic, and pouch-preserving surgical techniques, ileostomy is indicated with or without pouch excision. Between 17 and 57% of patients with CLIPI will ultimately develop pouch failure with increased rates seen in patients with fistulizing disease. ^{9 71} In a study by Gu et al of 65 patients with de novo CD of the pouch, pouch retention rate was 77% in patients with inflammatory presentation, 100% in patients with fibrostenotic phenotype, and 42% with fistulizing pattern. ⁷¹ The decision to excise the pouch is controversial, as there is risk associated with either approach. Diversion without pouch excision allows for the persistence of symptomatic perianal fistula and may lead to the development of diversion pouchitis. ⁷² However, excision of the pouch can lead to pelvic fibrosis, which limits future surgical options. ⁷² Furthermore, recurrence of CD has been documented in the small bowel, perianal area, and upper gastrointestinal tract leading to persistent perineal sinuses, leaks, abscesses, and pain. ⁷²

Pouch Advancement and Redo Ileal Pouch-Anal Anastomosis

When considering revisional surgery, pouch viability and capacity must be considered. Revision should not be considered in patients with poor anal sphincter function, active anal or small bowel CD, or in those with limited length residual terminal ileum. The pouch should be examined and measured before the decision is made whether to excise and recreate the pouch versus revision using the existing pouch. Repair of the existing pouch requires mobilization and disconnection from the anus with repair and redo anastomosis. When revision is not possible, permanent fecal diversion with end ileostomy should be performed.

In rare cases and in the hands of experienced surgeons/centers, a redo pouch or pouch advancement can be offered to carefully selected patients with CD. Redo surgery for ileal pouches can include creating a new J, S, or W pouch with or without mucosectomy. One retrospective analysis of a prospectively maintained pouch database compared outcomes among patients with either UC or CD who underwent a three-stage redo pouch (diverting loop ileostomy, redo pouch, ileostomy closure). ⁷³ There were 30 patients with CD in the study, including 7 with pouch stricture, and all required mucosectomy with handsewn anastomosis during pouch recreation. Pouch failure at 5 years was 55% among the CD group. Overall, patients with CD were more likely to have redo pouch failure compared to patients with UC.

Conclusion

CLIPI is an increasingly common phenomenon that may compromise long-term IPAA success rates. It is critical to differentiate this disease process from other common pouch disorders, particularly because of its association with pouch failure and symptomatic impact on patients. It remains unclear whether it is a subset of CD or a new entity. Medical and surgical management mimic the treatment of other CD manifestations, though rigorous data is still somewhat lacking. Patients with CLIPI are at an increased risk for pouch failure.