Short-term treatment with infliximab in chronic refractory pouchitis and ileitis


Dr C. Calabrese, Department of Internal Medicine and Gastroenterology, University of Bologna, Via Massarenti n. 9, 40138 Bologna, Italy.


Background  Chronic refractory pouchitis is a long-term complication after ileal pouch-anal anastomosis and it may be associated with ileal inflammation.

Aim  To determine the efficacy of infliximab in treatment of chronic refractory pouchitis complicated by ileitis, using a wireless capsule endoscopy.

Methods  Sixteen patients with chronic refractory pouchitis complicated by ileitis were enrolled. Pouchitis was diagnosed by clinical, endoscopic and histological criteria. Ileitis was documented using wireless capsule endoscopy. Duodenum–jejunum and proximal–middle ileum were evaluated and the presence of small lesions and large lesions were noted. Crohn’s disease, intestinal infections were excluded in all patients. Patients were treated with infliximab and clinical response was recorded. Wireless capsule endoscopy was repeated at week 10 and Pouchitis Disease Activity Index score was determined.

Results  Ten patients were enrolled and completed the study. Clinical remission was achieved in nine patients. At wireless capsule endoscopy and pouch endoscopy, a complete recovery of lesions was observed in eight patients. One patient presented four small lesions of the ileum at the 6 weeks of treatment and one patient did not show any modification. Clinical and endoscopic remission was maintained in these eight patients at least 6 months.

Conclusion  These preliminary results indicate that infliximab may be recommended for the treatment of chronic refractory pouchitis complicated by ileitis.


The surgical treatment of choice for most patients with severe ulcerative colitis (UC) is restorative proctocolectomy with ileal pouch anal anastomosis (IPAA).1 Pouchitis is a nonspecific idiopathic inflammation of the ileal reservoir and could be a common long-term complication after pouch surgery for UC.2 Symptoms of this pathology are dysfunctions associated with endoscopic and histological evidence of acute inflammation.3, 4 It is still very difficult to determine the true incidence of pouchitis, and reported incidence rates are between 10% and 59% with UC.5, 6 Patients with preoperative extra-intestinal manifestations and primary sclerosing cholangitis have more risk to develop a pouchitis.7, 8 As in UC, smoking may protect against the development of pouchitis,9 whereas different surgical techniques do not influence the frequency of pouchitis.10, 11

The aetiology of pouchitis is still unknown and its pathogenesis is still poorly understood.12 The hypotheses could be mucosal ischaemia of the pouch, a missed diagnosis of Crohn’s disease (CD) or a novel form of inflammatory bowel disease.13 Nonsteroidal anti-inflammatory drug (NSAID) use, concurrent Clostridium difficile or cytomegalovirus (CMV) infection, coeliac disease, cuffitis and irritable pouch syndrome could be other possible causes.14

There are no universally accepted diagnostic criteria for pouchitis. Semi-objective assessments to diagnose pouchitis have been proposed using composite score such as the Pouchitis Disease Activity Index (PDAI).15

The disease activity of pouchitis can be so defined: remission, mild–moderate (increased stool frequency, urgency and infrequent incontinence) or severe (dehydration and frequent incontinence). On the basis of the duration of disease, pouchitis can also be defined as acute (<4 weeks) or chronic (>4 weeks). Another way to classify this syndrome considers the following patterns: infrequent (a single or two acute episodes), relapsing (more than three acute episodes) or chronic (a treatment responsive form requiring a maintenance therapy or a treatment-resistant form). Approximately 10–15% of patients with pouchitis develop a chronic pouchitis. Patients with chronic refractory pouchitis (CRefP) do not respond to conventional available therapies, and continue to suffer symptoms. This condition is a common cause of pouch failure. Medical treatment of patients with CRP is particularly difficult and disappointing. Misdiagnosed CD, anastomotic stricture, infectious aetiology, cuffitis and irritable pouch syndrome should be excluded as alternative diagnosis, in case of a refractory patient.16

Until the advent of wireless capsule endoscopy (WCE), there were no endoscopic devices that allowed a complete visualization of the small bowel. WCE could have a role to asses suspected small bowel CD, differentiating to indeterminate colitis and defining the endoscopic estimation of disease activity and response to therapy in patients with known small bowel CD.

Recently, it has been demonstrated that CRefP may be associated with an unknown form of inflammatory disease of the ileum.17

The aim of this study was to evaluate the efficacy of infliximab as treatment of the CRefP complicated by ileitis diagnosed using WCE.

Material and methods

Study design

This was a single-blind, prospective, cohort study carried out in a referral centre for inflammatory bowel disease patients. Of 600 patients affected by UC who underwent IPAA, 240 (40%) had at least one episode of acute pouchitis. Of the 240 patients, 29 (12%) were affected by CRefP. The patients were routinely followed up by our Department since the diagnosis of UC was made.

We prospectively included 16 patients with CRefP between April 2005 and November 2006. All patients with pouchitis had the diagnosis confirmed according to PDAI,18 based on the combination of clinical, endoscopic and histological criteria. Patients underwent a pouch endoscopy, a small bowel follow-through and after 2 weeks at WCE. Pouchitis was defined as a total of PDAI score ≥7 points.

Inclusion criteria

Patients were included if they had a CRefP, defined as no response to at least 4 weeks of standard antibiotic therapies (ciprofloxacin 1 g twice daily or metronidazole 500 mg t.d.s.) and if presented lesions in the jejunum–ileum at WCE study.

Exclusion criteria

Patients were excluded from entry if they:

  • 1had taken NSAIDs within 4 weeks;
  • 2had a perianal disease (including abscess, fissure, stricture or anal-sphincter weakness);
  • 3had CD or cuffitis;
  • 4had a severe cardiovascular, respiratory, hepatic (including primary sclerosing cholangitis) or renal conditions;
  • 5were pregnant or breast feeding and
  • 6were treated with infliximab or immunosuppressive therapy before surgery.

Crohn’s disease was at first excluded using a careful re-examination of the colonic surgical and histological specimens. Every relapse was documented endoscopically and histologically.

Moreover, histological and laboratory data excluded the diagnosis of coeliac disease. Serological tests for Shigella, Salmonella, Toxoplasma, Epstein–Barr and CMV were negative. At histology, bacteria including Campylobacter and acid fast bacilli, protozoa, including Giardia, fungi were excluded with Giemsa, Wartin-Starry, Ziehl-Nielsen and periodic acid-Schiff stains. Morphologically, no nuclear changes suggestive of viral infection were seen. Stool examinations for bacteria and parasites were negative. Antineutrophil cytoplasm autoantibodies and the anti-Saccharomyces cerevisiae mannan antibodies were performed at entry.

Two weeks before WCE, all patients underwent to a SBFT and pouch endoscopy. Multiple biopsies were taken in the pouch and in the terminal ileum as far as 30 cm from the pouch.

The study was performed in accordance with the Declaration of Helsinki and was approved by the local ethical committee. All patients received detailed information regarding the procedure and written and informed consent was obtained.

The WCE procedure

Wireless capsule endoscopy was performed using a standard Pillcam capsule, which takes two pictures per second in bowel (M2A, Given Imaging, Yoqneam, Israel), as previously described,17 with the following modifications: (i) we routinely administered an oral purging solution (2 L of polyethylene glycol) 10–12 h before the ingestion of the capsule; (ii) the patient fasted overnight and, on the morning of the procedure, swallowed the M2A capsule, together with a small amount of water; (iii) patients were allowed to start drinking, respectively, 2 and 4 h after capsule ingestion.

Evaluation of WCE took approximately 1.5 h per patient. A standardized evaluation form was completed immediately after each study by the investigator (CC). He was blinded to the results of pouch endoscopy, SBFT and clinical features.

Duodenum–jejunum and proximal–middle ileum were evaluated separately. For the purpose of this study, we arbitrarily classified:

Small lesions: aphtha, erosion, erythema/oedema, atrophia, scar and polyp (<5 mm).

Large lesions: cobblestone pattern and deep/fissural ulcers.

Lesion type, number and location were also noted.


Patients were then treated with infliximab (5 mg/kg) at 0, 2 and 6 weeks. At week 10 (±4 days) WCE and pouch endoscopy were repeated and PDAI score determined. Clinical assessment was done also at weeks 2, 6, 12, 24 and 48.

Macroscopic remission was defined as the disappearance of all large lesions and the persistence of ≤3 small lesions, at WCE and pouch endoscopy.


Sixteen patients (nine male; mean age 38.7 ± 6.6; range: 23–56 years) were studied with WCE and all of them presented lesions in the jejunum–ileum.

At SBFT, among the 16 patients with CRefP, there was no evidence of minimal lesions, while only two (13%) patients presented a substenosis of the pouch at the level of the ileal-pouch anastomosis, and a focal ectasia of the middle ileum.

Six of 16 patients resulted not eligible to the treatment with infliximab because five had a positive Mantoux test (>5 mm) and one had severe essential lymphopenia (HIV negative).

Characteristics of patients are shown in Table 1. All the patients completed the study. Infliximab was well tolerated and no side effects were recorded, apart from a single case of slight and transient skin rush, which appeared at the beginning of the second infusion and disappeared after reduction of the flow.

Table 1.   Demographic, epidemiological and clinical data
  1. UC, ulcerative colitis; IPAA, ileal pouch anal anastomosis; CRefP, chronic refractory pouchitis; ASCA, anti-Saccharomyces cerevisiae mannan antibody; ANCA, antineutrophil cytoplasm autoantibody; PDAI, Pouchitis Disease Activity Index.

Gender (male/female)6/4
Mean age (±s.d.)39.2 ± 7.1
Time (year) from diagnosis of UC to IPAA6.3 ± 5.2
Mean follow-up (since IPAA)7.3 ± 3.1
Smoking (%)1
Albumin (g/dL) 4.0 ± 0.8
CRefP (mg/dL)2.2 ± 1.8
PDAI13 ± 2.6

At entry, WCE investigation showed both small and large lesions of the ileum. In all 10 patients (six male; mean age 39.2 ± 7.1, range: 23–56) particularly, morphological findings detected were: apthae (8%), erosions (18%), erythema/oedemas (21%), atrophia (8%), scars (6%), polyps (1%), cobblestone pattern (10%) and deep-fissural ulcers (28%) (Figure 1, Table 2). The mean value of the number of the small lesions was 16.1 ± 6.2 (range: 8–30) and of the large lesions was 7 ± 3.3 (range: 3–12). Small lesions were more frequently presented in the jejunum, while the larger lesions were more frequent in the ileum. In two patients with large lesions in the duodenum, an upper gastrointestinal endoscopy was performed and biopsies were taken. Histopathology showed a mucosal inflammation not diagnostic of CD.

Figure 1.

 Wireless capsule endoscopy before treatment.

Table 2.   Frequency of lesions occurring in patients with chronic refractory pouchitis with a diagnostic capsule examination
Small bowel segmentPatients with small lesions (n)Patients with large lesions (n)
Proximal ileum1010
Middle ileum810

Pouch endoscopy revealed severe pouchitis in seven patients and moderate in three. Histopathology of the pouch and of the terminal ileum showed degrees of villous atrophy and mucosal chronic inflammatory cell infiltrate with crypt abscesses.

At the end of therapy, a clinical remission was observed in nine of 10 patients with a significant decrease in PDAI from 13 (range: 11–16) to 5 (range: 3–8) (Figure 2). Median stool frequency in these patients was 9 (range: 7–11) at baseline and 4 (range: 3–6) after the treatment with infliximab.

Figure 2.

 Changes in total Pouchitis Disease Activity Index score in all patients.

Eight of 10 patients presented a complete regression of the jejunum and ileum lesions at WCE, and of the pouch lesions at endoscopy (Figure 3).

Figure 3.

 Wireless capsule endoscopy after treatment.

One patient presented at WCE a reduction of the jejunum–ileum lesions but four small lesions at the middle ileum after therapy were noted and at pouch endoscopy was observed a reduction of the deep ulcer.

Only one patient did not show modification of endoscopic appearance both at WCE and pouch endoscopy. Clinical and endoscopic remission was maintained in these eight patients at least 6 months (range: 8–16).


Ten to 15% of patients with pouchitis experience a chronic pouchitis either ‘treatment responsive’ or ‘treatment refractory’. Patients with treatment-responsive chronic pouchitis respond to therapy, but when the therapy is stopped, pouchitis relapses. Patients with treatment-refractory pouchitis do not respond to conventional available therapies, and continue to suffer symptoms.

Treatment of chronic pouchitis is largely empirical and only few small placebo-controlled trials have been conducted. Recently, budesonide was shown to be effective in CRefP not responding after 1 month of antibiotic treatment.18

We previously used WCE to study the small bowel in patients with CRefP, we used WCE in a single-blind prospective cohort study. All 16 patients studied, showed diffuse lesions from duodenum to ileum. Lesions found were suggestive of CD, but histology showed a mucosal inflammation not diagnostic for CD.17 Moreover, CD was also excluded using a careful re-examination of the colonic surgical and multiple histological specimens.

In this study, we evaluated the efficacy of infliximab in 10 eligible patients with chronic pouchitis and extensive ileal involvement. In this small open study, short-term treatment of infliximab determined a clinical remission in 90% of patients and an endoscopic/histological healing of lesions in 80% of patients. Only one patient was unresponsive to therapy without any clinical and endoscopic improvement, probably related to a long-standing history of disease.

In conclusion, this is the first report as regards treatment of chronic pouchitis with extensive ileal involvement, suggesting that short-therapy with infliximab could be recommended for the treatment of CRefP complicated by ileitis. The meaning of these small bowel lesions is still unclear, and may suggest to the development of an unknown form of inflammatory disease.


Declaration of personal and funding interests: None.