https://orcid.org/0000-0002-2522-9798
Abstract
Approximately 1 in 10 patients with an ileal pouch–anal anastomosis is diagnosed with Crohn’s disease of the pouch (CDP). However, this diagnosis may be established inappropriately, as alternative underlying causes necessitating an alternative treatment approach, such as long-term surgical sequelae, may mimic CDP. In this study, we aimed to identify patients diagnosed with and treated for CDP with a (concurrent) alternative diagnosis.
Ulcerative colitis and inflammatory bowel disease unclassified patients who underwent ileal pouch–anal anastomosis surgery in a tertiary center between 1990 and 2017 were retrospectively reviewed. Patients with a postoperative diagnosis of CDP for which medical treatment was initiated were identified. Presence of pouchitis, prepouch ileitis, stricture, and fistulas was assessed and histopathological reports were evaluated. Thereafter, cross-sectional images of the pouch in CDP patients were re-evaluated to identify potential long-term surgical sequelae (ie, chronic presacral sinus or perianastomotic fistulas).
After a median postoperative follow-up of 6.2 (interquartile range, 2.3-13.5) years, 47 (10%) of 481 patients were diagnosed with CDP. CDP patients had pouchitis (n = 38 [81%]), prepouch ileitis (n = 34 [74%]), strictures (n = 17 [36%]), fistulas (n = 15 [32%]), or a combination. Multiple granulomas were found in 1 pouch resection specimen. Re-evaluation of 40 (85%) patients who underwent magnetic resonance imaging revealed presence of long-term surgical sequelae in 17 (43%) patients. Six (15%) patients demonstrated isolated nonanastomotic fistulas.
Re-evaluation of cross-sectional imaging of the pouch revealed that potential alternative causes were found in nearly half of CDP patients. Cross-sectional imaging is therefore recommended early in the diagnostic pathway to exclude an alternative diagnosis.
Lay Summary
Crohn’s disease of the pouch is frequently diagnosed in patients with an ileal pouch–anal anastomosis for ulcerative colitis. However, we have found that nearly half of all Crohn’s disease of the pouch patients have an underlying long-term surgical sequelae as an alternative diagnosis.
What is already known?
Approximately 10% of patients with an ileal pouch–anal anastomosis for ulcerative colitis develop Crohn’s disease of the pouch, a refractory phenotype with a high rate of pouch failure.
What is new here?
After re-evaluation of magnetic resonance imaging images in patients who have received a diagnosis of Crohn’s disease of the pouch in clinical care, nearly half had a surgical sequelae (ie, perianastomotic fistula or chronic sinus) that may explain the clinical picture.
How can this study help patient care?
Physicians should remain vigilant on surgical sequelae when a patient presents with a Crohn’s-like phenotype of the pouch, as it has significant implications for the management of these patients, either medical or surgical.
Introduction
Ileal pouch–anal anastomosis (IPAA) surgery is the treatment of choice to restore bowel continuity after proctocolectomy in patients with ulcerative colitis (UC) or familial adenomatous polyposis. To a lesser extent, patients with inflammatory bowel disease unclassified (IBD-U) or Crohn’s disease (CD) limited to the colon may be considered for IPAA as well.
Approximately 1 of 10 patients undergoing IPAA surgery for UC of IBD-U is diagnosed with Crohn’s disease of the pouch (CDP) postoperatively.1 However, the etiology of CDP remains unknown. Several hypotheses have been postulated: (1) misinterpretation of the original UC diagnosis before colectomy, (2) de novo CD that has been developed after IPAA, or (3) a new disease entity with Crohn’s-like features.2,3 CDP is generally based on the presence of fistulas, strictures, and/or treatment refractory pouchitis, extending to the prepouch ileum (prepouch ileitis).1,4,5 In patients diagnosed with CDP, pouch failure (PF) rates as high as 89% have been reported.6-8 A chronic sinus or perianastomotic fistulas are other common causes of PF, which are generally considered to be long-term sequelae of an untreated anastomotic leakage.9-12 The distinction between CDP and these long-term surgical sequelae remains a challenge, as they may manifest with similar clinical and endoscopic features. Step-up medical therapies (eg, immunomodulators and biologicals) are frequently initiated in patients with a diagnosis of CDP, whereas patients with long-term surgical sequelae may require a surgical approach. Furthermore, changing the original diagnosis of UC to CD may have a major emotional impact on patients’ well-being.13 It is thus of great importance to rule out long-term surgical sequelae before establishing a diagnosis of CDP. Although pouchoscopy could be helpful in the diagnostic process, this modality may not diagnose extraluminal collections of the pouch. Cross-sectional imaging is therefore the optimal manner for distinction between CDP and surgical sequelae.5
In the current study, patients who were diagnosed with and treated for CDP were identified. The aim of this study was to analyze the proportion of patients with a (concurrent) alternative diagnosis by re-evaluation of cross-sectional imaging in CDP patients.
Methods
Design and Patients
In this single-center retrospective cohort study, we collected data from a prospectively maintained database including all consecutive patients with an initial diagnosis of UC, IBD-U, or CD of the colon who underwent proctocolectomy and IPAA in a tertiary referral center (Amsterdam University Medical Center, location AMC) between January 1990 and December 2017.
Surgical Procedure
Baseline demographics such as initial diagnosis before colectomy, sex, (previous) smoking, age at diagnosis, and intraoperative techniques were collected from electronic patient records. Histology reports of the colectomy and/or proctectomy specimen were collected. IPAA surgery was performed in 1 or multiple stages, with a preference for a modified 2-staged procedure. A multistage pouch procedure was preferred to diminish the negative impact of medical therapy on (anastomotic) healing. Patients with an upfront diagnosis of CD were allowed to have a pouch in the absence of or history of perianal fistulas or small bowel disease localization.
Identification of Patients with CDP
A validated clinical, endoscopic, or histologic definition of CDP is lacking. In the current study, patients with a diagnosis of CDP were defined as follows: all patients were retrospectively reviewed for a diagnosis of CDP that was established in daily clinical practice for which medical treatment was initiated. We listed the features that led to a diagnosis of CDP, including endoscopic lesions (ie, presence of erythema, edema, erosions, or ulcerations) of the pouch (pouchitis) and prepouch ileum (prepouch ileitis), strictures, and fistulas. Strictures were classified by location: either proximal of the pouch inlet or at the pouch–anal anastomosis. In CDP patients, biopsies of the pouch and prepouch ileum as well as resection specimens in patients who underwent pouch excision were assessed for presence of (chronically active) inflammation on histopathology and/or granulomas.11 If granulomas were found in the rectal stump only, this was not considered confirmative of CD, as this finding is nonspecific.
Medical treatment initiated for CDP included corticosteroids, immunomodulators (thiopurines, methotrexate), biologicals (infliximab, adalimumab, vedolizumab, ustekinumab), or small molecules (tofacitinib). Patients with a diagnosis of CDP were compared with the rest of patients that underwent IPAA surgery without a postoperative diagnosis of CDP.
Re-evaluation of Cross-Sectional Imaging of CDP Patients
In patients with a diagnosis of CDP, we assessed if (postoperative) cross-sectional imaging (magnetic resonance imaging [MRI] computed tomography with rectal contrast) was performed. Cross-sectional images were reassessed for findings indicative of long-term surgical sequelae. Long-term surgical sequelae were defined as a chronic presacral sinus or fistulas originating from the pouch–anal anastomosis, chronic pelvic sepsis, and/or per-pouch fluid collections on cross-sectional imaging. Fistulas that did not originate from the pouch–anal anastomosis were listed separately.
Statistical Analyses
Continuous variables were expressed as mean ± SD or median and interquartile range (IQR), according to distribution. Categorical variables were expressed as percentage and were compared using the Pearson chi-square test. Normally distributed continuous data were compared by using the independent-sample t test. If continuous data were not normally distributed, the Mann-Whitney U test was applied. Variables with missing data exceeding 20% of all cases were excluded from analyses. Data were analyzed through complete case analyses. SPSS 28.0 (IBM Corporation) was used to perform the statistical analyses, with a predefined 2-sided significance limit of P < .05. Reporting on this study was done according to the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines.
Ethical Approval
The institutional review board of Amsterdam University Medical Center waived the need for an informed consent procedure. All patients were given the opportunity to decline participation through an opt-out procedure.
Results
Patient Demographics
A total of 481 patients underwent IPAA surgery between January 1990 and December 2017. Out of 481 included patients, 468 (97%) were diagnosed with UC and 13 (3%) with IBD-U before colectomy. The median age at diagnosis was 28.7 (IQR, 20.6-37.6) years and 51% were male; the median age at time of IPAA construction was 35.9 (IQR, 27.9-44.7) years. The overall postoperative follow-up period was 6.2 (IQR, 2.3-13.5) years. In total, 47 (10%) patients were diagnosed with CDP with a median follow-up period of 7.4 (IQR, 1.4-14.1) years after IPAA surgery. The postoperative follow-up in CDP patients was significantly longer patients when compared with patients without a diagnosis of CDP (22.9 [IQR, 12.1-27.4] years vs 5.4 years [IQR, 2.0-11.9]) (P < .01). Forty (85%) patients with a diagnosis of CDP underwent IPAA surgery before 2000. Patients with CDP demonstrated to have a numerically higher proportion of colectomy specimen fitting CD when compared with patients without CDP (n = 8 of 47 [17%] vs 27 of 434 [6%]; P = .051). All baseline characteristics are demonstrated in Table 1.
Baseline characteristics (N = 481).
| All patients (N = 481) | CDP (n = 47) | No CDP (n = 434) | P value | Missing | |
|---|---|---|---|---|---|
| Postoperative follow-up, y | 6.2 (2.3-13.5) | 22.9 (12.1-27.4) | 5.4 (2.0-11.9) | <.01a | 0 (0) |
| Female | 236 (49) | 19 (40) | 217 (50) | .212 | 0 (0) |
| Diagnosis before colectomy | .798 | 0 (0) | |||
| UC | 468 (97) | 46 (98) | 422 (97) | ||
| IBD-U | 13 (3) | 1 (2) | 12 (27) | ||
| Age at diagnosis, y | 28.7 (20.6-37.6) | 24.6 (18.8-32.3) | 29.5 (20.7-38.0) | .036a | 60 (12) |
| Disease duration at time of colectomy, y | 4.1 (1.5-8.0) | 3.3 (1.2-5.8) | 4.1 (1.5-8.3) | .126 | 62 (13) |
| Histology report resection specimenb | .051 | 94 (19) | |||
| Fitting UC | 289 (60) | 29 (62) | 260 (60) | ||
| Indeterminate | 63 (13) | 5 (11) | 58 (13) | ||
| Fitting CD | 35 (7) | 8 (17) | 27 (6) | ||
| Colectomy setting | .963 | 74 (15) | |||
| Elective | 253 (53) | 25 (53) | 228 (53) | ||
| (Semi)urgent | 154 (32) | 15 (32) | 139 (32) | ||
| Age at time of IPAA y | 35.9 (27.9-44.7) | 30.3 (23.3-38.4) | 36.7 (28.4-45.4) | <.01a | 0 (0) |
| Stages IPAA surgery | .737 | 3 (1) | |||
| 1 | 140 (29) | 15 (32) | 125 (29) | ||
| 2 | 80 (17) | 10 (21) | 70 (16) | ||
| Modified 2 | 213 (44) | 18 (38) | 195 (45) | ||
| 3 | 45 (9) | 4 (9) | 41 (9) | ||
| Smoking | .345 | 149 (31) | |||
| No | 260 (54) | 14 (30) | 246 (57) | ||
| Previously | 25 (5) | 3 (6) | 22 (5) | ||
| Yes | 47 (10) | 4 (9) | 43 (10) | ||
| Pouch excision during follow-up | 18 (4) | 6 (13) | 12 (3) | <.01a | 0 (0) |
| All patients (N = 481) | CDP (n = 47) | No CDP (n = 434) | P value | Missing | |
|---|---|---|---|---|---|
| Postoperative follow-up, y | 6.2 (2.3-13.5) | 22.9 (12.1-27.4) | 5.4 (2.0-11.9) | <.01a | 0 (0) |
| Female | 236 (49) | 19 (40) | 217 (50) | .212 | 0 (0) |
| Diagnosis before colectomy | .798 | 0 (0) | |||
| UC | 468 (97) | 46 (98) | 422 (97) | ||
| IBD-U | 13 (3) | 1 (2) | 12 (27) | ||
| Age at diagnosis, y | 28.7 (20.6-37.6) | 24.6 (18.8-32.3) | 29.5 (20.7-38.0) | .036a | 60 (12) |
| Disease duration at time of colectomy, y | 4.1 (1.5-8.0) | 3.3 (1.2-5.8) | 4.1 (1.5-8.3) | .126 | 62 (13) |
| Histology report resection specimenb | .051 | 94 (19) | |||
| Fitting UC | 289 (60) | 29 (62) | 260 (60) | ||
| Indeterminate | 63 (13) | 5 (11) | 58 (13) | ||
| Fitting CD | 35 (7) | 8 (17) | 27 (6) | ||
| Colectomy setting | .963 | 74 (15) | |||
| Elective | 253 (53) | 25 (53) | 228 (53) | ||
| (Semi)urgent | 154 (32) | 15 (32) | 139 (32) | ||
| Age at time of IPAA y | 35.9 (27.9-44.7) | 30.3 (23.3-38.4) | 36.7 (28.4-45.4) | <.01a | 0 (0) |
| Stages IPAA surgery | .737 | 3 (1) | |||
| 1 | 140 (29) | 15 (32) | 125 (29) | ||
| 2 | 80 (17) | 10 (21) | 70 (16) | ||
| Modified 2 | 213 (44) | 18 (38) | 195 (45) | ||
| 3 | 45 (9) | 4 (9) | 41 (9) | ||
| Smoking | .345 | 149 (31) | |||
| No | 260 (54) | 14 (30) | 246 (57) | ||
| Previously | 25 (5) | 3 (6) | 22 (5) | ||
| Yes | 47 (10) | 4 (9) | 43 (10) | ||
| Pouch excision during follow-up | 18 (4) | 6 (13) | 12 (3) | <.01a | 0 (0) |
Values are median (interquartile range) or n (%).
Abbreviations: CD, Crohn’s disease; CDP, Crohn’s disease of the pouch; IBD-U, inflammatory bowel disease unclassified; IPAA, ileal pouch–anal anastomosis; UC, ulcerative colitis.
aDenotes that P value reaches statistical significance.
bDerived from colonic (n = 336) or rectal stump (n = 51) specimen histopathology reports.
Baseline characteristics (N = 481).
| All patients (N = 481) | CDP (n = 47) | No CDP (n = 434) | P value | Missing | |
|---|---|---|---|---|---|
| Postoperative follow-up, y | 6.2 (2.3-13.5) | 22.9 (12.1-27.4) | 5.4 (2.0-11.9) | <.01a | 0 (0) |
| Female | 236 (49) | 19 (40) | 217 (50) | .212 | 0 (0) |
| Diagnosis before colectomy | .798 | 0 (0) | |||
| UC | 468 (97) | 46 (98) | 422 (97) | ||
| IBD-U | 13 (3) | 1 (2) | 12 (27) | ||
| Age at diagnosis, y | 28.7 (20.6-37.6) | 24.6 (18.8-32.3) | 29.5 (20.7-38.0) | .036a | 60 (12) |
| Disease duration at time of colectomy, y | 4.1 (1.5-8.0) | 3.3 (1.2-5.8) | 4.1 (1.5-8.3) | .126 | 62 (13) |
| Histology report resection specimenb | .051 | 94 (19) | |||
| Fitting UC | 289 (60) | 29 (62) | 260 (60) | ||
| Indeterminate | 63 (13) | 5 (11) | 58 (13) | ||
| Fitting CD | 35 (7) | 8 (17) | 27 (6) | ||
| Colectomy setting | .963 | 74 (15) | |||
| Elective | 253 (53) | 25 (53) | 228 (53) | ||
| (Semi)urgent | 154 (32) | 15 (32) | 139 (32) | ||
| Age at time of IPAA y | 35.9 (27.9-44.7) | 30.3 (23.3-38.4) | 36.7 (28.4-45.4) | <.01a | 0 (0) |
| Stages IPAA surgery | .737 | 3 (1) | |||
| 1 | 140 (29) | 15 (32) | 125 (29) | ||
| 2 | 80 (17) | 10 (21) | 70 (16) | ||
| Modified 2 | 213 (44) | 18 (38) | 195 (45) | ||
| 3 | 45 (9) | 4 (9) | 41 (9) | ||
| Smoking | .345 | 149 (31) | |||
| No | 260 (54) | 14 (30) | 246 (57) | ||
| Previously | 25 (5) | 3 (6) | 22 (5) | ||
| Yes | 47 (10) | 4 (9) | 43 (10) | ||
| Pouch excision during follow-up | 18 (4) | 6 (13) | 12 (3) | <.01a | 0 (0) |
| All patients (N = 481) | CDP (n = 47) | No CDP (n = 434) | P value | Missing | |
|---|---|---|---|---|---|
| Postoperative follow-up, y | 6.2 (2.3-13.5) | 22.9 (12.1-27.4) | 5.4 (2.0-11.9) | <.01a | 0 (0) |
| Female | 236 (49) | 19 (40) | 217 (50) | .212 | 0 (0) |
| Diagnosis before colectomy | .798 | 0 (0) | |||
| UC | 468 (97) | 46 (98) | 422 (97) | ||
| IBD-U | 13 (3) | 1 (2) | 12 (27) | ||
| Age at diagnosis, y | 28.7 (20.6-37.6) | 24.6 (18.8-32.3) | 29.5 (20.7-38.0) | .036a | 60 (12) |
| Disease duration at time of colectomy, y | 4.1 (1.5-8.0) | 3.3 (1.2-5.8) | 4.1 (1.5-8.3) | .126 | 62 (13) |
| Histology report resection specimenb | .051 | 94 (19) | |||
| Fitting UC | 289 (60) | 29 (62) | 260 (60) | ||
| Indeterminate | 63 (13) | 5 (11) | 58 (13) | ||
| Fitting CD | 35 (7) | 8 (17) | 27 (6) | ||
| Colectomy setting | .963 | 74 (15) | |||
| Elective | 253 (53) | 25 (53) | 228 (53) | ||
| (Semi)urgent | 154 (32) | 15 (32) | 139 (32) | ||
| Age at time of IPAA y | 35.9 (27.9-44.7) | 30.3 (23.3-38.4) | 36.7 (28.4-45.4) | <.01a | 0 (0) |
| Stages IPAA surgery | .737 | 3 (1) | |||
| 1 | 140 (29) | 15 (32) | 125 (29) | ||
| 2 | 80 (17) | 10 (21) | 70 (16) | ||
| Modified 2 | 213 (44) | 18 (38) | 195 (45) | ||
| 3 | 45 (9) | 4 (9) | 41 (9) | ||
| Smoking | .345 | 149 (31) | |||
| No | 260 (54) | 14 (30) | 246 (57) | ||
| Previously | 25 (5) | 3 (6) | 22 (5) | ||
| Yes | 47 (10) | 4 (9) | 43 (10) | ||
| Pouch excision during follow-up | 18 (4) | 6 (13) | 12 (3) | <.01a | 0 (0) |
Values are median (interquartile range) or n (%).
Abbreviations: CD, Crohn’s disease; CDP, Crohn’s disease of the pouch; IBD-U, inflammatory bowel disease unclassified; IPAA, ileal pouch–anal anastomosis; UC, ulcerative colitis.
aDenotes that P value reaches statistical significance.
bDerived from colonic (n = 336) or rectal stump (n = 51) specimen histopathology reports.
Diagnostic Characteristics in CDP Patients
All 47 patients with CDP presented with persisting complaints such as high stool frequency, abdominal pain, rectal blood loss, fecal incontinence, obstructive symptoms, perianal abscess, purulent discharge from the anus, fecal leakage from the vagina, or a combination of these symptoms. Corticosteroids, immunomodulators, and biologicals and were initiated in 12 (26%), 22 (47%), and 38 (80%) of patients with a diagnosis of CDP, respectively. Endoscopic assessment of the pouch was performed in all patients with a diagnosis of CDP. The reported endoscopic findings were pouchitis (n = 38 [81%]), prepouch ileitis (n = 34 [72%]), strictures (n = 17 [36%]), fistulas (n = 15 [32%]), or a combination of these findings, as demonstrated in Figure 1. Strictures were located either proximal of the pouch or at the pouch–anal anastomosis in 9 (19%) and 8 (17%) patients, respectively. Cross-sectional imaging of the ileoanal pouch by MRI was performed in 40 (85%) patients with CDP, a median of 41 (IQR, 7-100) months after the diagnosis of CDP. Fistulas were mentioned in initial MRI reports of 20 (50%) patients, of which 4 combined with a chronic sinus. An isolated chronic sinus was reported in 3 other patients, adding up to a total of 7 (10%) patients with a chronic sinus. Histopathological assessment was performed in 43 (91%) CDP patients (pouch biopsy in 43 and resection specimen in 6 patients). Chronically active inflammation was found in all patients, but granulomas were only found in 1 (2%) pouch excision specimen. The endoscopic and histological findings of all 47 patients with CDP are demonstrated in Supplementary Appendix 1.
Flowchart demonstrating Crohn’s disease of the pouch (CDP) characteristics. IPAA, ileal pouch–anal anastomosis;.
Secondary Assessment of Cross-Sectional Imaging in CDP Patients
MRI re-evaluation confirmed fistulas in 17 (43%) patients, of which 10 (25%) included a fistula originating from the pouch–anal anastomosis. A chronic presacral sinus was found in another 7 (18%) patients, adding up to a total of 17 (43%) patients with surgical sequelae (perianastomotic fistula or chronic sinus) after re-evaluation of MRI images. Perianal (n = 4 [10%]), enteroenteral (n = 1 [3%]), or pouch–vaginal (n = 1 [3%]) fistulas with absence of perianastomotic fistula or chronic sinus were found in 6 (15%) patients. The time to diagnosis of CDP was numerically shorter for patients with surgical sequelae when compared with patients without surgical sequelae (3.7 [IQR, 1.0-11.3] years vs 8.7 [IQR, 3.0-15.1] years; P = .2). Surgical sequelae were found in 4 (50%) of 8 CDP patients with a colectomy specimen histopathologically fitting CD. Findings after re-evaluation of MRI are demonstrated in Figure 2 and Supplementary Appendix 1. Figure 3 demonstrates an MRI image of a patient initially diagnosed with CDP with a chronic presacral sinus likely to be a long-term surgical sequela.
Flowchart demonstrating findings of re-evaluation of magnetic resonance imaging (MRI) images. CDP = Crohn’s disease of the pouch.
Fat-suppressed T2-weighted magnetic resonance image of a patient with a chronic presacral sinus (red arrow) originating from the pouch–anal anastomosis. The green arrow indicates the pouch lumen.
Discussion
A diagnosis of CDP during the postoperative follow-up after IPAA surgery has a reported incidence of approximately 10%.1 However, a validated definition of CDP is still lacking.1,5 Current diagnostic modalities such as endoscopy or histology do not allow an accurate distinction between CDP and other pathology that may present with CD-like features, such as long-term surgical sequelae (eg, chronic presacral sinus and/or a fistulas originating from the pouch–anal anastomosis) or a therapy refractory chronic pouchitis.14,15 In the current study, 10% of patients received a diagnosis of CDP in daily clinical practice after a median postoperative follow-up period of 7.4 years, which is well within range of previously published literature.1 After re-evaluation of cross-sectional imaging, long-term surgical sequelae were identified on MRI in nearly half of CDP patients (43%). It is of utmost importance to separate patients with long-term surgical sequelae from those with CDP, as management strategies are different. Patients with CDP require immunosuppressive therapy, while patients with surgical sequelae initially require surgical management. A causal relation between long-term surgical sequelae and clinical manifestation of CDP cannot be established, and surgical sequelae may present concomitantly with CDP. However, overdiagnosis of CDP is plausible in at least a proportion of patients with surgical sequelae on cross-sectional imaging. Misdiagnosis of CDP in patients with surgical sequelae could lead to overtreatment with immunosuppressive therapy, while alternative (surgical) treatment is indicated instead.9 We recommend implementation of routine cross-sectional imaging by MRI early during follow-up after IPAA, particularly within the diagnostic workup to CDP, to eliminate surgical complications and/or sequelae. In this way, misdiagnosis of CDP may be avoided.
Histopathological preference of CD in the (procto)colectomy specimen may have shaped the postoperative diagnosis of CDP in case of pouch dysfunction. Half of the CDP patients with a (procto)colectomy specimen histopathologically fitting CD had surgical sequelae on MRI. Vigilance is warranted when concluding CDP based on the initial histopathological specimen only, as surgical complications may be present.
All CDP patients who underwent histopathological assessment of pouch biopsies or pouch resection specimen had chronically active inflammation. However, only 1 (2%) CDP patient was demonstrated to have specific histopathological findings confirmative of CD. A study by Lightner et al14 reported a histological diagnosis of CD in only 1 in 5 patients with an excised pouch for presumed CDP. We based our findings on available pathology reports of pouch biopsies and excision specimens. Presence of a granuloma was the only confirmative parameter for CD in the current study, as other histological features are generally nonspecific.16 The inability of histopathologic confirmation of CD impairs proper distinction between CDP and alternative diagnoses.
As CDP is one of the most common causes for pouch failure, it is of great importance to initiate appropriate treatment options. Clinically, CDP is often suspected based on the presence of inflammation in the pouch and/or prepouch ileum. In the current study, these endoscopic findings were observed in the majority of patients with a diagnosis of CDP. Inflammation in the pouch and/or prepouch ileum may be indicative, yet not confirmative of CDP, as (severe) pouchitis may appear with similar features.4,17-20 In this study, we did not evaluate patients diagnosed with a chronic and/or severe pouchitis. Although distinction between CDP and chronic pouchitis can also be challenging, both require medical therapeutic management. As surgical therapy is recommended for surgical sequelae, the distinction between surgical sequelae and CDP may be of greater clinical relevance.
Interestingly, both age at diagnosis and disease duration and age at IPAA were (numerically) lower in CDP patients as compared with patients with no diagnosis of CDP. This difference may be caused by a more severe disease phenotype in patients, contributing to the suspicion and diagnosis of CDP.17,21 Correspondingly, the median postoperative follow-up was significantly longer for patients with CDP. We hypothesize that this difference may be partly due to a more intensive and prolonged postoperative follow-up practice. Recent increased awareness of septic surgical sequelae may also have decreased CDP diagnoses over time.
Our study has several strengths. We included a large cohort of patients from a tertiary IBD center who underwent IPAA surgery with a median postoperative follow-up of 6.2 years. All CDP patients received endoscopic assessments, and the majority underwent histopathological assessment of the pouch. Furthermore, we provided a complete overview of all IPAA patients with a postoperative diagnosis of CD and the disease characteristics they presented with.
Several limitations need to be addressed for the current study. First of all, the absence of a gold standard to diagnose CDP precluded a fully reliable cohort composition. Moreover, missing data were inevitable due to the retrospective nature of the study. Information on extraintestinal manifestations and disease extent were particularly difficult to obtain in the majority of patients, even though these variables may be predictive for developing chronic pouchitis or CDP.22 Although the majority of CDP patients underwent cross-sectional imaging, absence of imaging precluded re-evaluation in 15% of CDP patients. As patients with CDP or surgical complications might have been lost to follow-up, the incidence of CDP may have been underestimated. However, as patients who present with therapy refractory disease or complications of the pouch are usually referred back to the initial surgery center, it is likely that most patients with CDP and surgical complications were identified. The median time to diagnosis of CDP exceeded the median follow-up length of the control group, which may also have caused underestimation of the incidence of CD.
Conclusions
CDP was diagnosed in 10% of patients after IPAA surgery. After reassessment of cross-sectional imaging, nearly half of patients diagnosed with CDP demonstrated findings compatible with long-term surgical sequelae. Implementation of cross-sectional imaging in the diagnostic workup of symptomatic patients after pouch surgery may prevent a misdiagnosis of CDP and inappropriate subsequent treatment.
Supplementary data
Supplementary data is available at Inflammatory Bowel Diseases online.
Author Contribution
Study design: M.R., D.d.J., E.W., C.B., M.D., G.D., C.P. Data collection: D.d.J., M.R., E.W. Revision of magnetic resonance imaging images: C.B., R.H. Statistical analysis and interpretation of data: D.d.J., M.R. Drafting of the manuscript: D.d.J., M.R. All coauthors revised and approved the manuscript for important intellectual content.
Conflicts of Interest
D.d.J. has received speaker fees from Johnson and Johnson. W.B. has received grants from Braun and Vifor, and has served as speaker for Takeda, Johnson and Johnson and Braun. C.P. has served as advisor for Pliant, Shire, and Takeda, has received grants from Gilead and Perspectum, and has received speaker’s fees from Tillotts. G.D. has served as advisor for AbbVie, Ablynx, Alimentiv, Amgen, AM Pharma, Biogen, Bristol Meiers Squibb, Boehringer Ingelheim, Celgene/Receptos, Celltrion, Cosmo, Covidien/Medtronic, Ferring, Dr Falk Pharma, Eli Lilly, Engene, Galapagos, Genentech/Roche, Gilead, GlaxoSmithKline, Immunic, Johnson and Johnson, Lamepro, Lument, Mitsubishi Pharma, Merck Sharp Dome, Mundipharma, Nextbiotics, Novo Nordisk, Otsuka, Pfizer, Polpharm, Prometheus Laboratories/Nestlé, Procise Diagnostics, Protagonist, Salix, Samsung Bioepis, Sandoz, Setpoint, Shire, Takeda, Tigenix, Tillotts, Topivert, Versant and Vifor; and received speaker fees from AbbVie, Biogen, Ferring, Johnson and Johnson, Merck Sharp Dome, Mundipharma, Norgine, Pfizer, Samsung Bioepis, Shire, Takeda, Tillotts, and Vifor. R.H. has served as a consultant for Applied Medical; received research grant support from Stryker; and received speaker fees Johnson and Johnson and Medtronic. C.B. has received unrestricted grant support from Boehringer Ingelheim and Roche; and received consultancy fees or speaker honoraria from Merck Sharp & Dohme, Tillotts, Janssen, and Takeda. M.D. has served as advisor for Echo Pharma and Robarts Clinical Trials; received nonfinancial support from Dr Falk Pharma; and received speaker fees from Janssen, Merck & Co., Pfizer, Takeda, and Tillotts Pharma. All other authors disclose no conflicts.
References
Author notes
Authors share first authorship.
