Risk Factors and Postoperative Outcomes in Pouchitis Following Restorative Proctocolectomy: An 18-Year Single-Center Study

Abstract

Background/Objectives: Restorative proctocolectomy with ileo-anal pouch anastomosis (IPAA) remains the preferred surgical treatment for ulcerative colitis (UC). However, complications like pouchitis can occur. This study aimed to describe patients who underwent IPAA for inflammatory bowel disease (IBD) at Padua Hospital from 2005 to 2023 and identify risk factors for pouchitis. Secondary objectives included evaluating the effectiveness of biological therapy in chronic antibiotic-refractory pouchitis (CARP), Crohn’s disease of the pouch (CDP), and Crohn’s-like inflammation of the pouch (CDLPI), and assessing risk factors for pouch failure. Methods: This retrospective, observational study included 109 patients whose data were collected from medical records. Univariate logistic regression was used to analyze associations between preoperative and postoperative factors and outcomes such as acute pouchitis and pouch failure. The effectiveness of biological therapy was assessed by measuring changes in the Pouchitis Disease Activity Index (PDAI) and the Modified Pouchitis Disease Activity Index (mPDAI) over a 12-month treatment period. Results: Univariate logistic regression revealed significant associations between preoperative extraintestinal manifestations (OR 3.569, 95% CI 1.240–10.720), previous diagnosis of Crohn’s disease (OR 10.675, 95% CI 1.265–90.089), and transmural inflammation at cross-sectional imaging before surgery (OR 3.453, 95% CI 1.193–9.991) with an acute pouchitis risk. Pouch failure was significantly associated with a previous diagnosis of Crohn’s disease (OR 9.500, 95% CI 1.821–49.571) and post-surgical fistulas (OR 41.597, 95% CI 4.022–430.172). Biological therapy led to a significant reduction in the PDAI score in patients with CARP, decreasing from a median of 10 to 4 (p = 0.006). Similarly, in patients with CDP or CDLPI, the mPDAI score was significantly reduced from a median of 9 to 1 (p = 0.034), with remission achieved in 5/6 (83.3%) of these patients. Conclusions: This study provides valuable insights into the management of IPAA patients and highlights the importance of early identification and treatment of risk factors for pouchitis and failure. Biological therapy demonstrated significant effectiveness in reducing disease activity in patients with CARP, CDP, and CDLPI, suggesting its role as a crucial component in managing these complications.

1. Introduction

Restorative proctocolectomy (RPC), which involves a total proctocolectomy with ileal pouch–anal anastomosis (IPAA), is the preferred surgical option in patients with ulcerative colitis (UC) [1,2,3]. This surgical procedure for surgery can lead to various short- and long-term complications, classified into structural, inflammatory, or functional disorders of the pouch, as well as neoplasia [4]. Factors such as the underlying disease, surgical aspects, and anatomical changes contribute to these adverse effects [5]. A significant post-surgical issue is pouch failure which can result from such complications [6].

Pouchitis is the most common inflammatory and long-term complication following pouch surgery for UC [7]. It is defined as a non-specific inflammation of the ileal reservoir. Pouchitis can be classified as idiopathic (primary) or secondary [8]. It affects approximately 48% of patients within the first two years post-surgery, and up to 80% experience symptoms at some point after the procedure [7,9,10,11,12,13,14,15]. Reported risk factors include being a non-smoker; extensive UC [16]; backwash ileitis [17]; extraintestinal manifestations (EIMs), especially primary sclerosing cholangitis (PSC) [18,19,20]; the regular use of non-steroidal anti-inflammatory drugs (NSAIDs) [21]; and concomitant autoimmune disorders [22,23]. In recent years, our group developed and validated the Padova Prognostic Score for Colitis (PPSC) [24]. This score predicts postoperative complications such as CD, pouchitis, and anal abscess or fistula. The diagnosis of pouchitis is based on a combination of clinical symptoms and endoscopic findings [25]. Histology can further support the diagnosis by identifying inflammation in the pouch body. Fecal markers, such as fecal calprotectin, can be used as adjunct measures to further quantify pouch inflammation.

Multiple strategies are used in the treatment and prevention of pouchitis, including antibiotics, probiotics, corticosteroids, biologic therapy, and oral small-molecule drugs, with the primary goal of clinical remission. Significant advances are being made in the field, such as the development of scoring systems to better characterize patient-reported outcomes and endoscopic findings [26,27,28]. The score aims to overcome the downsides of the most commonly used Pouchitis Disease Activity Index (PDAI) [29,30].

Pouchitis can be categorized as acute or chronic, and further classified into antibiotic-responsive, antibiotic-dependent, or antibiotic-refractory phenotypes, depending on the patient’s response to standard antibiotic treatments. Chronic antibiotic therapy is recommended for recurrent pouchitis that relapses shortly after stopping antibiotics, known as chronic antibiotic-dependent pouchitis (CADP). Advanced immunosuppressive therapies, like biologics, may be used for chronic antibiotic-refractory pouchitis (CARP). CARP has been voted as one of five difficult-to-treat IBD conditions by a consensus panel of experts from the International Organization for the Study of Inflammatory Bowel Disease [31]. The recent randomized controlled trial (RCT) comparing vedolizumab with a placebo in the treatment of patients with CARP (EARNEST trial) led to the approval of vedolizumab with this indication [32].

Crohn’s disease of the pouch (CDP) or Crohn’s-like inflammation of the pouch (CDLPI) can develop post-IPAA, affecting approximately 10% of patients [33,34,35]. Although CDP and CDLPI have been used interchangeably, there is a distinction. When Crohn’s colitis is the pre-colectomy diagnosis, the condition in the pouch is considered true CDP. Conversely, if the preoperative or perioperative diagnosis was UC or indeterminate colitis, and CD is identified later, this condition is described as CDLPI [33,34].

We conducted an observational retrospective study with the aim of describing a cohort of patients who underwent IPAA for inflammatory bowel disease (IBD) at Padua Hospital from 2005 to 2023 and identifying risk factors associated with the onset of pouchitis. Secondary objectives include evaluating the effectiveness of biological therapy in CARP and CDP/CDLPI and assessing the presence of risk factors for pouch failure. Through this study, we hope to contribute to a better understanding of the complications following IPAA and inform clinical management strategies.

2. Materials and Methods

2.1. Study Aims

This study aimed to describe the cohort of patients who underwent IPAA for IBD at Padua Hospital between 2005 and 2023, with follow-up conducted at our institution. Additionally, the study sought to identify risk factors associated with the development of pouchitis and pouch failure. Furthermore, it aimed to assess the effectiveness of biological therapy in treating CARP, as well as in managing CDP and CDLPI.

2.2. PICO Statement

• Population (P): Patients who underwent restorative proctocolectomy with IPAA for IBD (UC, IBD unclassified, or CD) at Padua Hospital between 2005 and 2023;
• Intervention (I): Biological therapy administered to patients with CARP, CDP, or CDLPI;
• Comparison (C): No direct comparison group is included, as this is an observational study focusing on the effectiveness of treatment within the cohort;
• Outcome (O): Risk factors for pouchitis and pouch failure, and the effectiveness of biological therapy, measured by clinical remission defined by the PDAI and mPDAI.

2.3. Study Design

This is a retrospective, observational, monocentric study that included consecutive patients with a confirmed diagnosis of UC, IBD unclassified (IBDU), or CD according to ECCO guidelines [4], who underwent restorative proctocolectomy with IPAA at Padua Hospital between 2005 and 2023. Only patients with complete pre-, intra-, and post-surgical clinical data were included. Patients were identified through the hospital’s medical records database. For the analysis of biologic therapy effectiveness, only those with more than one year of follow-up were included, resulting in 11 patients in the CARP section and 6 patients in the CDP/CDLPI section.

2.4. Data Collection

Data were retrieved from medical records and categorized into four main sections:

1. Preoperative History and Disease Characteristics: Includes demographic information (sex, age, and date of birth), disease diagnosis and history (e.g., UC, CD, proctitis, left-sided colitis, and pancolitis), smoking status, family history of IBD, complications (abscesses, and fistulas), EIMs (number and type), diagnosis of primary sclerosing cholangitis, histological biopsy results, data from endoscopic or radiological evaluations (MRI, intestinal ultrasound, and CT abdominal scan), intestinal tumors, oncological treatments, and preoperative medical therapy details (medications, duration, and reasons for discontinuation).

2. Surgical Information: Details on surgery indications (medical failure, complications, and dysplasia), type of surgery (laparoscopic, and laparotomic), and specific surgical details (date, patient’s age, and duration).

3. Postoperative Follow-up: Includes follow-up duration and modality (outpatient visits, and endoscopies), date of last follow-up, postoperative complications (perianal fistulas, anastomotic stenosis, deep vein thrombosis, pulmonary embolism, anastomotic leak, pouchitis, pouch failure, bleeding, pelvic sepsis, and anal canal disease), and reinterventions (reasons, date, type, and outcome). Pouch failure has been defined in previous literature as either the need for a permanent stoma or both pouch excision and reconstruction. In this study, we chose to adhere to the latter definition [36].

4. Postoperative Medical Therapy: Details on postoperative medications (types, dosage, duration, and response), biological therapies (types, duration, and reasons for discontinuation), and antibiotic therapy (type, and response). The PDAI was calculated in the CARP cohort, while the Modified Pouchitis Disease Activity Index (mPDAI) was preferred in the cohorts with CDP and CDLPI due to the exclusion of histological components in mPDAI [30]. Remission was defined as a PDAI score of <7 or an mPDAI score of ≤4 and a reduction of ≥2 points from baseline [37].

2.5. Data Analysis

Statistical analyses were performed using JASP software (Version 0.19.0). The choice of statistical tests depended on the type of data. The Shapiro–Wilk test was used to assess the normality of continuous data. For categorical variables, the Chi-square test and Fisher’s exact test were employed. The Mann–Whitney U test was used for unpaired continuous data, while the Kruskal–Wallis test, followed by Dunn’s test for multiple comparisons, was used for comparing more than two groups. Paired data were analyzed using either the McNemar test or the Wilcoxon signed-rank test. Binary logistic regression was utilized to identify independent predictors of disease outcomes. A p-value of <0.05 was considered statistically significant for all analyses.

3. Results

3.1. Study Population Characteristics

The study cohort comprised 109 patients, among whom 38 (34.9%) were females. Most of the patients (100, 91.7%) had a preoperative clinical diagnosis of UC, while 8 patients (7.4%) had a diagnosis of CD and 1 patient (0.9%) of IBDU.

The median age at diagnosis was 35.5 years (interquartile range [IQR] 26.0–35.5). Regarding smoking status, only 7 patients (6.4%) were current smokers.

Preoperative complications were present in a subset of patients, with 7 patients (6.4%) having abscesses and 3 patients (2.7%) having fistulas. Endoscopic preoperative evaluations revealed backwash ileitis in 3 patients (2.7%) and endoscopic signs of discontinuous inflammation in 3 patients (2.7%).

EIMs were observed in 19 patients (17.4%), with PSC being identified in 2 patients (1.8%). Preoperative medical therapies are shown in

Table 1. Preoperative characteristics of patients undergoing total proctocolectomy with ileal pouch–anal anastomosis.

Characteristic		Value
Female (n, %)		38 (34.9%)
Preoperative clinical diagnosis
	Ulcerative colitis (n, %)	100 (91.7%)
	Crohn’s disease (n, %)	8 (7.4%)
	Inflammatory bowel disease unclassified (n, %)	1 (0.9%)
Preoperative disease location for UC
	Rectum (n, %)	0 (0.0%)
	Left colitis (n, %)	31 (28.4%)
	Pancolitis (n, %)	53 (48.6%)
Age at diagnosis (years, median, IQR)		35.5 (26.0–35.5)
Family history of IBD (n, %)		6 (5.5%)
Smoking status
	Non-smoker (n, %)	90 (82.5%)
	Current smoker (n, %)	7 (6.4%)
	Former smoker (n, %)	12 (11.0%)
Preoperative complications of IBD
	Abscesses (intra-abdominal/perianal) (n, %)	7 (6.4%)
	Fistulas intra-abdominal/perianal) (n, %)	3 (2.7%)
Preoperative endoscopy
	Backwash ileitis (n, %)	3 (2.7%)
	Endoscopic signs of discontinuous inflammation (n, %)	3 (2.7%)
Extraintestinal manifestations
	Total number of extraintestinal manifestations (n, %)	19 (17.4%)
	Primary Sclerosing Cholangitis (n, %)	2 (1.8%)
Preoperative medical therapy
	Steroids	46 (42.2%)
	Immunosuppressants	41 (37.6%)
	Biologics or small molecules	67 (61.5%)
	Antibiotics	10 (9.1%)
	Aminosalicylates	80 (73.4%)
Preoperative biological or small-molecule therapy
	Infliximab	60 (55.0%)
	Adalimumab	20 (18.3%)
	Golimumab	3 (2.7%)
	Vedolizumab	27 (24.8%)
	Ustekinumab	1 (0.9%)
	Tofacitinib	1 (0.9%)
Number of biological or small-molecule treatments before surgery (median, range)		1 (0–4)

, together with other preoperative characteristics. Notably, 46 patients (42.2%) received steroids, and 67 patients (61.5%) were treated with biologics or small molecules before surgery.

The primary surgical indications for UC patients were as follows: chronic UC unresponsive to medical therapy, affecting 61 patients (56.0%); acute severe ulcerative colitis (ASUC), affecting 10 patients (9.2%); dysplasia/neoplasia, affecting 12 patients (11.0%); and fulminant UC, affecting 4 patients (3.6%). In the cohort of CD patients, surgery was considered after a comprehensive assessment by a multidisciplinary team, which evaluated each patient’s individual condition and surgical suitability. The decision to proceed with IPAA was made following a detailed review of the patient’s disease localization, treatment history, and overall health status. The median age at the time of surgery was 45.0 years (IQR 36.0–55.5), and the mean time from diagnosis to surgery was 94.58 months (SD 93.7).

Unexpected dysplasia or neoplasia was identified in three patients (2.6%) during the histological evaluation of the surgical specimens. Of these cases, three patients were found to have low-grade dysplasia. One patient was diagnosed with a neuroendocrine neoplasm of the appendix, classified as T1a N0. Another patient was diagnosed with colorectal cancer, staged as T2 N2b M0. This individual, who was 61 years old at the time of surgery, had a left-sided UC, and the surgical indication was ASUC. The cancer diagnosis prompted a course of adjuvant chemotherapy.

Early postoperative complications (within one month) occurred in 35 patients (32.1%), whereas late postoperative complications (≥1 month) were observed in 68 patients (62.4%). Given the poorly defined diagnostic criteria for de novo CD following IPAA surgery, we categorized patients into two groups: those with CDP, diagnosed with CD before surgery, and those with CDLPI, characterized by inflammation of the pouch exhibiting typical features of CD, including the presence of a perianal or other fistula developing at least 12 months post-final IPAA stage, strictures in the pouch body or pre-pouch ileum, and pre-pouch ileitis.

Intra-abdominal reoperations were necessary for 22 patients (20.2%), with a total of 26 reoperations performed. Early reoperations (within one month) were rare, occurring in 2 patients (6.8%), while the majority were late reoperations (≥1 month), accounting for 24 cases (93.1%). The types of interventions included pouch excision (7 cases, 26.9%), pouch reconstruction (2 cases, 7.7%), fistula-related surgery (10 cases, 38.5%), surgeries to address small bowel obstruction (5 cases, 19.2%), and other types of surgeries (2 cases, 7.7%). Additionally, incisional hernia repairs were performed in 10 patients (9.1%), and endoscopic dilatation or the incision of anastomotic stenosis was necessary for 6 patients (5.5%). Additionally, incisional hernia repairs were performed in 10 patients (9.1%), and endoscopic dilatation or the incision of anastomotic stenosis was necessary for 6 patients (5.5%).

The characteristics of surgical procedures and post-operative complications are reported in

Table 2. Characteristics of surgical procedures and post-operative complications.

Variable		Value
Indication for Surgery for UC Patients
	Acute severe ulcerative colitis (ASUC)	10 (9.2%)
	Chronic UC unresponsive to medical therapy	61 (56.0%)
	Dysplasia/neoplasia	12 (11.0%)
	Fulminant UC	4 (3.6%)
Number of stages
	Two stages	84 (77.1%)
	Three stages	25 (22.9%)
Age at surgery (years, median, IQR)		45.0 (36.0–55.5)
Average time from diagnosis to surgery (months, mean, SD)		94.58 (93.7)
Post-operative complications
	Number of post-operative complications/patient (mean, DS; range)	1.3 (1, 2; 0–3)
	Patients experiencing early postoperative complications (<1 month), count	35 (32.1%)
	Patients experiencing late postoperative complications (≥1 month), count	68 (62.4%)
Type of post-operative complications
	Gastrointestinal bleeding	3 (2.7%)
	Pelvic sepsis	7 (6.4%)
	Small bowel obstruction	10 (9.1%)
	Intestinal perforation	4 (3.6%)
	Anastomotic leak	6 (5.0%)
	Pulmonary embolism/Deep vein thrombosis	1 (0.9%)
	Fistulas	18 (16.5%)
	Anal canal disease	9 (8.1%)
	Cuffitis	7 (6.4%)
	Anastomotic stenosis	7 (6.4%)
	Acute pouchitis	47 (43.1%)
	Chronic pouchitis	21 (19.3%)
	Pouch failure	9 (8.2%)
	Crohn’s-like inflammation of the pouch (CDLPI)	4 (3.6%)
	Pouch neoplasia	1 (0.9%)
	Incisional hernia	10 (9.1%)
	Post-operative diagnosis of IBD-U	1 (0.9%)
	Cancer recurrence	2 (1.8%)
Intra-abdominal reoperations
	Patients undergoing reoperation	22 (20.2%)
	Number of intra-abdominal reoperations	26
Timing of reoperation
	Early (<1 month)	2 (6.8%)
	Late (≥1 month)	24 (93.1%)
Type of reoperation
	Pouch excision	7 (26.9%)
	Pouch reconstruction	2 (7.7%)
	Fistula surgery	10 (38.5%)
	Small bowel obstruction surgery	5 (19.2%)
	Other	2 (7.7%)

.

3.2. Risk Factors for Pouchitis and Management of Pouchitis

The median time to the onset of acute pouchitis was 28 months (IQR 11.5–57.5). Pouchitis classification among the cohort revealed that 47 patients (43.1%) experienced acute pouchitis, of which 19 (17.4%) developed chronic pouchitis. Of those, 15 (13.8%) had CARP, and 4 (3.7%) had CADP. CDLPI was seen in 2 patients (1.8%). All patients with a previous diagnosis of CD developed CDP (7.34%).

Among patients undergoing treatment for pouchitis, antibiotics were the most used therapy, administered to 42 patients (89.4%). For acute pouchitis, the most frequently used antibiotics were metronidazole (11 patients, 23.4%), followed by fluoroquinolones (7 patients, 14.9%), and amoxicillin-clavulanate (2 patients, 4.3%). Local therapy with 5-ASA or steroids was utilized in 14 patients (29.8%), while biologics or small molecules were employed in 20 patients (42.5%). In cases of chronic pouchitis, biologic or small-molecule therapies included vedolizumab (14 patients, 40.0%), infliximab (9 patients, 25.7%), ustekinumab (8 patients, 22.9%), adalimumab (3 patients, 8.5%), and tofacitinib (1 patient, 2.9%). Notably, golimumab was not used in this cohort.

Additional treatments included immunomodulators (2 patients, 4.3%), steroids (3 patients, 6.4%), and 5-Aminosalicylates (12 patients, 25.5%). Surgical interventions such as pouch excision or reconstruction were required in 3 patients (6.4%). For prophylaxis, VSL#3 (VSL Pharmaceuticals, Inc., Gaithersburg, MD, USA) was used for secondary prevention in 8 patients (17.0%). The characteristics of patients with pouchitis and of medical therapies in our cohort are shown in

Table 3. Classification and treatment of pouchitis.

Variable			Value
Time to acute pouchitis onset (months, median, IQR)			28 (11.5–57.5)
Classification of pouchitis
Acute pouchitis			47 (43.1%)
Chronic pouchitis			19 (17.4%)
Chronic Antibiotic-Refractory Pouchitis (CARP)			15 (13.8%)
Chronic Antibiotic-Dependent Pouchitis (CADP)			4 (3.7%)
Crohn’s Disease of the pouch (CDP)			8 (7.34%)
Crohn’s-Like Inflammation of the Pouch (CDLPI)			4 (3.6%)
Medical treatment for pouchitis
	Local therapy (5-ASA, Steroids)		14 (29,8%)
	Antibiotics		42 (89.4%)
		Amoxicillin-clavulanate	2 (4.3%)
		Fluoroquinolones	7 (14.9%)
		Metronidazole	11 (23.4%)
	Immunomodulators		2 (4.3%)
	Steroids		3 (6.4%)
	5-Aminosalicylates		12 (25.5%)
	Biologics or small molecules		20 (42,5%)
		Infliximab	9 (25.7%)
		Adalimumab	3 (8.5%)
		Vedolizumab	14 (40.0%)
		Ustekinumab	8 (22.9%)
		Anti-JAKs	1 (2.9%)
		Golimumab	0 (0.0%)
Surgical therapy (pouch excision or reconstruction)			3 (6.4%)
Prophylaxis
	VSL#3		8 (17.0%)

.

A comparison between patients who developed acute pouchitis (n = 47) and those who did not (n = 62) revealed significant differences in several preoperative characteristics. Patients with preoperative UC had a lower incidence of acute pouchitis (83.0% vs. 98.4%, p = 0.004), while those with preoperative CD had a higher incidence (14.9% vs. 1.6%, p = 0.008). The presence of backwash ileitis at preoperative endoscopy (11.4% vs. 0%, p = 0.009) and transmural inflammation at radiological imaging performed before the surgical intervention (27.7% vs. 12.9%, p = 0.021) were more common in patients who developed acute pouchitis. Additionally, patients with acute pouchitis had a higher rate of EIMs (27.6% vs. 9.7%, p = 0.014).

Postoperative complications varied between the groups. Bowel perforation was significantly more common in patients with acute pouchitis (8.6% vs. 0%, p = 0.019). Other complications such as cuffitis, anal fissures, fistulas, anastomotic stenosis, suspected pelvic sepsis, anastomotic leak, pouch failure, pouch neoplasia, and cancer recurrence did not show significant differences between the groups.

Table 4. Characteristics of patients with and without acute pouchitis.

Variable		Patients Who Did Not Develop Acute Pouchitis (n = 62)	Patients Who Developed Acute Pouchitis (n = 47)	p-Value
Females (n, %)		19 (30.7%)	19 (40.4%)	0.293
Preoperative clinical diagnosis
	Ulcerative colitis (n, %)	61 (98.4%)	39 (83.0%)	0.004
	Crohn’s disease (n, %)	2 (1.6%)	7 (14.9%)	0.008
	Inflammatory bowel disease unclassified (n, %)	0 (0.0%)	1 (2.1%)	0.249
Preoperative UC disease location
	Rectum (n, %)	0 (0.0%)	0 (0.0%)	0.733
	Left colitis (n, %)	17 (27.4%)	10 (21.3%)
	Pancolitis (n, %)	31 (50.0%)	26 (55.3%)
Age at diagnosis (years, median, IQR)		35 (28–48)	35 (25–43)	0.202
Disease duration at surgery (months, median, IQR)		58 (28–135)	49 (22–140)	0.444
Family history of IBD (n, %)		3 (4.8%)	3 (6.3%)	0.721
Smoking status
	Non-smoker (n, %)	48 (77.4%)	42 (89.4%)	0.113
	Current smoker (n, %)	5 (8.1%)	2 (4.3%)
	Former smoker (n, %)	9 (14.5%)	3 (6.4%)
Preoperative perianal disease
	Perianal abscesses (n, %)	5 (8.1%)	2 (4.3%)	0.510
	Perianal fistulas (n, %)	2 (3.2%)	1 (2.1%)	0.809
Preoperative endoscopy
	Backwash ileitis (n, %)	0 (0.0%)	3 (6.3%)	0.044
	Endoscopic Signs of Discontinuous Inflammation (n, %)	3 (4.8%)	1 (2.1%)	0.138
Preoperative histological finding of ileal inflammation		8 (12.9%)	7 (14.9%)	0.559
Preoperative histological finding of epithelioid granulomas		2 (3.2%)	1 (2.1%)	0.813
Radiological finding of transmural inflammation		8 (12.9%)	13 (27.7%)	0.021
Extraintestinal manifestations
	Total number of extraintestinal manifestations (n, %)	6 (9.7%)	13 (27.6%)	0.014
	Primary Sclerosing Cholangitis (n, %)	1 (1.6%)	1 (2.1%)	0.845
Indication for surgery
	Acute severe ulcerative colitis (ASUC)	7 (11.3%)	3 (6.4%)	0.641
	Chronic UC unresponsive to medical therapy	37 (59.7%)	24 (51.1%)	0.454
	Dysplasia/neoplasia	8 (12.9%)	4 (8.5%)	0.793
	Fulminant UC	3 (4.8%)	1 (2.1%)	0.622
Other post-operative complications
	Cuffitis	2 (3.3%)	5 (10.7%)	0.121
	Anal fissures	1 (1.6%)	3 (3.4%)	0.193
	Fistulas	13 (21.0%)	5 (10.7%)	0.154
	Anastomotic stenosis	2 (3.3%)	5 (10.7%)	0.121
	Anastomotic leak	5 (8.1%)	1 (2.1%)	0.183
	Bowel perforation	0 (0.0%)	4 (8.6%)	0.019
	Pouch failure	6 (9.7%)	3 (6.4%)	0.542
	Pouch neoplasia	1 (1.6%)	0 (0.0%)	0.382
	Cancer recurrence	2 (3.3%)	0 (0.0%)	0.214
Patients undergoing surgical reoperations		15 (24.2%)	7 (14.9%)	0.235

presents the characteristics of patients with pouchitis as compared to patients without pouchitis.

The univariate analysis identified significant risk factors for developing acute pouchitis in patients undergoing surgery. A pre-operative diagnosis of CD was strongly associated with an increased risk, with an OR of 10.675 (95% CI: 1.265–90.089, p = 0.030). Transmural inflammation on pre-op imaging also showed a significant association, with an OR of 3.453 (95% CI: 1.193–9.991, p = 0.022). The presence of EIMs similarly indicated a higher risk, with an OR of 3.569 (95% CI: 1.240–10.720, p = 0.018). Conversely, a pre-operative diagnosis of UC was protective, with an OR of 0.080 (95% CI: 0.010–0.664, p = 0.019). Other factors assessed did not show significant associations.

Table 5. Univariate analysis of risk factors for acute pouchitis.

Variable	Odds Ratio (OR)	95% Confidence Interval (CI)	p-Value
Sex (female)	1.536	0.694–3.398	0.290
Preoperative diagnosis of CD	10.675	1.265–90.089	0.030
Preoperative diagnosis of UC	0.080	0.010–0.664	0.019
Smoking	0.594	0.307–1.147	0.121
Family history of IBD	1.341	0.258–6.963	0.727
Extra-intestinal manifestations	3.569	1.240–10.720	0.018
Abscess history before surgery	0.560	0.102–3.082	0.505
Fistulas history before surgery	0.726	0.063–8.358	0.797
Transmural inflammation at radiological imaging before surgery	3.453	1.193–9.991	0.022
Epithelioid granulomas at histologic evaluation before surgery	0.741	0.064–8.580	0.810
Ileal inflammation at histologic evaluation before surgery	1.417	0.448–4.479	0.553
Preoperative medical therapy
Preoperative immunosuppressive therapy	1.234	0.565–2.694	0.598
Preoperative biologic therapy	1.399	0.637–3072	0.402
Preoperative steroid therapy	0.645	0.297–1.402	0.268
Preoperative antibiotic therapy	0.128	0.016–1.049	0.055
Indication for surgery
ASUC	0.709	0.170–2.961	0.638
Chronic refractory UC	1.459	0.549–3.883	0.449
Dysplasia/Neoplasia	0.839	0.231–3.043	0.790
Fulminant UC	0.559	0.056–5.613	0.621

reports univariate analysis of risk factors for pouchitis.

In the multivariate logistic regression analysis, adjusting for established risk factors for pouchitis, transmural inflammation observed on preoperative radiological imaging was significantly associated with an increased likelihood of developing acute pouchitis, with an OR of 3.610 (95% CI: 1.196–10.892, p = 0.023). However, the presence of EIMs did not reach statistical significance, with an OR of 3.586 (95% CI: 0.863–14.091, p = 0.079). Conversely, smoking status did not emerge as a significant predictor in the multivariate model, with an OR of 0.499 (95% CI: 0.029–2.724, p = 0.272). The results of the multivariate analysis for these additional preoperative risk factors are summarized in

Table 6. Multivariate logistic regression for preoperative risk factors for acute pouchitis.

Variable	Odds Ratio (OR)	95% Confidence Interval (CI)	p-Value
Transmural inflammation at radiological imaging before surgery
Absent	-	-
Present	3.610	1.196–10.892	0.023
Smoking status
Non-smoker	-	-
Smoker	0.499	0.029–2.724	0.272
Extra-intestinal manifestations
Absent
Present	3.586	0.863–14.091	0.079

.

3.3. Risk Factors for Pouch Failure

Pouch failure, including both pouch excision and reconstruction, occurred in 9 out of 109 cases. The reasons for failure varied, with no cases attributed to poor pouch function or incontinence. Three cases (33.3%) required an abdominoperineal resection (Miles procedure) for oncological reasons. One case (11.1%) reported both strictures and fistulas as the reasons for pouch failure. Strictures and fistulas alone accounted for one case (11.1%) and two cases (22.2%) of pouch failure, respectively. Pelvic sepsis and CDP each accounted for one case (11.1%). Acute pouchitis was not found to be more common in these patients (see Table 4).

The incidence of pouch failure was 4.2% (3/71) in males and 15.8% (6/38) in females, a statistically significant difference (p = 0.037). Among the nine patients with pouch failure, seven (77.8%) had no EIMs, and two (22.2%) had at least one EIM (p = 0.696). None of the patients diagnosed with PSC developed pouch failure.

To explore the relationship between various inflammatory conditions of the pouch and pouch failure, we analyzed the association of CADP, CARP, CDLPI, and CDP with pouch failure. No patients with either CARP, CADP, or CDLPI experienced pouch failure. In contrast, pouch failure was seen in six patients (5.9%) without CDP, while three patients with CDP experienced pouch failure (3.7%), a statistically significant difference (p = 0.002). Since all patients with a previous diagnosis of CD developed CDP, this difference extends to the previous diagnosis of UC and CD.

The univariate analysis identified several factors associated with pouch failure.

In the univariate logistic regression analysis presented in

Table 7. Univariate logistic regression analysis of risk factors for pouch failure.

Variable	Odds Ratio (OR)	95% Confidence Interval (CI)	p-Value
Sex (Female)	4.250	0.050–4.250	0.050
Preoperative diagnosis of extensive UC	0.709	0.180–2.798	0.624
Preoperative diagnosis of CD	9.500	1.821–49.571	0.008
Preoperative diagnosis of UC	0.128	0.128–0.025	0.012
Age at diagnosis	1.015	0.966–1.066	0.555
Smoking status	1.958	0.209–18.339	0.556
Extra-intestinal manifestations, number	1.395	0.266–7.306	0.694
Others post-operative complications
Small bowel obstruction	3.286	0.583–18.526	0.178
Anastomotic leak	2.375	0.247–22.875	0.454
Anastomotic stenosis	1.958	0.209–18.334	0.556
Perianal fistulas	28.318	5.216–153.733	<0.001
Cuffitis	1.958	0.209–18.334	0.556
Acute pouchitis	0.636	0.151–2.689	0.539
Chronic pouchitis requiring biologic therapy	2.000	0.460–8.698	0.355
Anal fissures	4.042	0.376–43.462	0.249
Anal canal disease (stenosis or hemorrhoids)	24.5	3.417–175.617	0.001
Crohn’s disease of the pouch	9.500	1.821–49.571	0.008

, several risk factors were identified for pouch failure. The female sex was associated with an OR of 4.250 (95% CI: 1.000–18.051, p = 0.050), suggesting a potential increased risk. Notably, a pre-operative diagnosis of CD as well as a post-operative diagnosis of CDP significantly elevated the risk of pouch failure (OR: 9.500, 95% CI: 1.821–49.571, p = 0.008), while a pre-operative diagnosis of UC appeared protective (OR: 0.128, 95% CI: 0.025–0.656, p = 0.012). Among post-op complications, the presence of perianal fistulas emerged as a strong risk factor (OR: 28.318, 95% CI: 5.216–153.733, p < 0.001), and anal canal disease (stenosis or haemorrhoids) also demonstrated a significant association with pouch failure (OR: 24.5, 95% CI: 3.417–175.617, p = 0.001). Other factors, including smoking status and age at dx, did not show significant associations.

Table 7 presents the results of the univariate logistic regression analysis for pouch failure.

Multivariate logistic regression was performed to adjust for potential confounders and to identify independent risk factors for pouch failure. The analysis confirmed that CDP remained a significant predictor of pouch failure, with an OR of 29.539 (95% CI: 1.809–482.327, p = 0.018). The presence of postoperative perianal fistula also emerged as a strong independent risk factor (OR: 41.597, 95% CI: 4.022–430.172, p = 0.002).

Additionally, anal canal disease, including anal stenosis or hemorrhoids, approached significance in predicting pouch failure (OR: 20.590, 95% CI: 0.896–472.914, p = 0.059), suggesting that, while not statistically significant at the conventional threshold, it may still hold clinical relevance.

Table 8. Multivariate logistic regression analysis of risk factors for pouch failure.

Variable	Odds Ratio (OR)	95% Confidence Interval (CI)	p-Value
Chron’s disease of the pouch
Absent	-	-
Present	29.539	1.809–482.327	0.018
Perianal post-operative fistula
Absent	-	-
Present	41.597	4.022–430.172	0.002
Anal canal disease (exclusive of fissures)
Absent	-	-
Present	20.590	0.896–472.914	0.059

presents the results of the multivariate logistic regression analysis for pouch failure.

3.4. Effectiveness and Safety of Biological Therapy in Treating CARP

A total of 11 patients were included in the analysis, with their treatment divided among three biologic therapies: two patients received infliximab, three received ustekinumab, and six received vedolizumab.

3.4.1. Clinical Characteristics

The analysis of clinical outcomes revealed significant improvements following 12 months of biologic therapy. Specifically, the prevalence of diarrhea and rectal bleeding decreased substantially from 100% of patients before treatment to just 18.2% after treatment (p = 0.016). Similarly, urgency, which was present in 81.8% of patients before therapy, dropped to 18.2% after treatment (p = 0.016).

3.4.2. Endoscopic Features

Endoscopic assessments showed a mixed response to treatment. The resolution of edema and loss of a vascular pattern was notable, with a reduction from 100% of patients before therapy to 18.2% after therapy (p = 0.016) and from 72.7% to 9.1% (p = 0.039), respectively. However, the improvement in hyperemia, ulcerations, and exudate were not statistically significant.

3.4.3. Histologic Characteristics and PDAI Score

A histologic evaluation revealed that the proportion of patients with lymphocytic/granulocytic infiltrate decreased from 100% before therapy to 54.5% after therapy, although this change was not statistically significant (p = 0.250). The grading of lymphocytic infiltrate showed significant shifts. The number of patients with moderate infiltrate decreased from 72.7% to 36.4%, while those with severe infiltrate increased from 18.2% to 36.4 (p = 0.040). The PDAI score showed a statistically significant reduction from a median score of 10 (IQR: 9–12) before therapy to 4 (IQR: 4–6) after 12 months, with a p-value of 0.006.

Remission based on PDAI score, defined arbitrarily as a value of less than 7, was achieved in 8/11 (72%) of patients.

Figure 1 presents the distribution of PDAI scores in patients before and after 12 months of biologic therapy.

3.4.4. Laboratory Tests

The laboratory tests also reflected improvements, although not all changes were statistically significant. The median C-reactive protein (CRP) level decreased from 18.4 mg/dL (IQR: 10.4–32.5) to 1 mg/dL (IQR: 0–5.7), but this change was not statistically significant (p-value = 0.068). Fecal calprotectin levels also decreased from a median of 500.0 µg/g (IQR: 85.3–968.0) to 179.0 µg/g (IQR: 116.5–439), although this change was also not significant (p = 0.109).

Table 9. Comparison of clinical, endoscopic, histologic, and laboratory characteristics before and after 12 months of biologic therapy in patients with CARP.

	Before Biologic Therapy	After 12 Months of Biologic Therapy	p-Value
Clinical Characteristics
Diarrhea	11 (100%)	2 (18.2%)	0.016
Rectal bleeding	11 (100%)	2 (18.2%)	0.016
Urgency	9 (81.8%)	2 (18.2%)	0.016
Endoscopic features
Hyperemia	10 (90.1)	6 (54.5%)	0.125
Edema	11 (100%)	2 (18.2%)	0.016
Loss of vascular pattern	8 (72.7%)	1 (9.1%)	0.039
Ulcerations	4 (36.4%)	4 (36.4%)	1.000
Exudate	5 (45.4%)	1 (9.1%)	0.250
Histologic characteristics
Lymphocytic/granulocytic infiltrate	11 (100%)	6 (54.5%)	0. 250
Grade of Lymphocytic Infiltrate			0.040
Absent
Mild	1 (9.1%)	2 (18.2%)
Moderate	8 (72.7%)	4 (36.4%)
Severe	2 (18.2%)	4 (36.4%)
Laboratory Tests
C-reactive protein (mg/dL; median, IQR)	18.4 (10.4–32.5)	1 (0–5.7)	0.068
Fecal calprotectin (ug/g; median, IQR)	500.0 (85.3–968.0)	179.0 (116.5–439)	0.109
PDAI score (median, IQR)	10 (9–12)	4 (4–6)	0.006

shows the comparison of clinical, endoscopic, histologic, and laboratory characteristics before and after 12 months of biologic therapy.

3.5. Effectiveness and Safety of Biological Therapy in Treating Crohn’s Disease of the Pouch or Crohn’s-like Inflammation of the Pouch

In this analysis, six patients were included, comprising five with CDP and one with CDLPI. The distribution of biologic therapies among these patients was as follows: one patient received infliximab, two received ustekinumab, and three received vedolizumab.

3.5.1. Clinical Characteristics

The clinical outcomes after 12 months of biologic therapy showed notable improvements. The number of patients experiencing diarrhea decreased from six to two (p = 0.125). Rectal bleeding resolved completely, dropping from five patients before therapy to none after (p = 0.063). Fever was rare in this cohort, with only one patient experiencing it before therapy and none afterward (p = 1.000). Urgency decreased from six patients before therapy to two after (p = 0.125).

3.5.2. Endoscopic Features and mPDAI Score

Endoscopic evaluations revealed significant changes. Hyperemia was reduced from six to three patients (p = 0.250), and edema showed a significant reduction from five to two patients (p = 0.016). The loss of a vascular pattern, initially present in five patients, was completely resolved after treatment (p = 0.039). Ulcerations were reduced slightly from two to one patient (p = 1.000), and exudate resolved in all cases, dropping from five patients to none (p = 0.063).

The mPDAI demonstrated a significant reduction from baseline after 12 months, decreasing from a median score of 9 to 1 (p = 0.034). Remission, defined as an mPDAI score of ≤4 and a reduction of ≥2 points from baseline, was achieved by five out of six patients (83.3%). The distribution of the mPDAI is depicted in Figure 2.

3.5.3. Laboratory Tests

The laboratory results showed that the median CRP level decreased from 15 mg/dL (IQR: 7.5–39.0) to 1.3 mg/dL (IQR: 0.0–13.2), although this reduction was not statistically significant (p = 0.068). Fecal calprotectin levels showed a slight, non-significant increase from a median of 103 µg/g (IQR: 85–121) to 114 µg/g (IQR: 49–155) (p = 0.345).

The results in the clinical, biochemical, and endoscopic parameters are summarized in

Table 10. Comparison of clinical, endoscopic, histologic, and laboratory characteristics before and after 12 months of biologic therapy in patients with CDP or CDLPI.

	Before Biologic Therapy	After 12 Months of Biologic Therapy	p-Value
Clinical Characteristics
Diarrhea	6 (100%)	2 (33.3%)	0.125
Rectal bleeding	5 (83.3%)	0 (0.0%)	0.063
Fever (>37.5 °C)	1 (16.6%)	0 (0.0%)	1.000
Urgency	6 (100%)	2 (33.3%)	0.125
Endoscopic features
Hyperemia	6 (100%)	3 (50%)	0.250
Friability	5 (83.3%)	2 (33.3%)	0.016
Granularity	5 (83.3%)	2 (33.3%)	0.016
Edema	5 (83.3%)	2 (33.3%)	0.016
Loss of vascular pattern	5 (83.3%)	0 (0.0%)	0.039
Ulcerations	2 (33.3%)	1 (16.6%)	1.000
Exudate	5 (83.3%)	0 (0.0%)	0.063
Laboratory Tests
C-reactive protein (mg/dL; median, IQR)	15 (7.5–39.0)	1.3 (0.0–13.2)	0.068
Fecal calprotectin (ug/g; median, IQR)	103 (85–121)	114 (49–155)	0.345
Modified Pouchitis Disease Activity Index (mPDAI) (median, IQR)	9 (7.5–9.750)	1 (1–2.5)	0.034

.

4. Discussion

This study aimed to elucidate the challenges and outcomes associated with surgical interventions for patients with IBD, particularly focusing on the incidence and risk factors for postoperative complications such as pouchitis and pouch failure. The secondary objectives were to assess the effectiveness of biologic therapies in managing chronic pouchitis and to identify significant preoperative predictors of postoperative complications.

Our cohort consisted of 109 patients who underwent RPC with IPAA at a single center from 2005 to 2023 and were followed-up at our center. Most patients were diagnosed with UC (91.7%), with a smaller representation of CD and IBDU. Notably, a substantial proportion of patients were on steroids (42.2%), immunosuppressants (37.6%), and biologics (61.5%). The primary indications for surgery were chronic UC unresponsive to therapy (56.0%), ASUC (9.2%), and dysplasia/neoplasia (11.0%). Postoperative complications were predominantly late, with 32.1% of patients experiencing early complications and 62.4% facing late complications. Acute pouchitis occurred in 43.1% of patients, with preoperative transmural inflammation, and a preoperative diagnosis of CD and EIMs identified as significant risk factors. Pouch failure was notably higher among patients with CDP and post-surgical fistulas. The effectiveness of biological therapy in treating CARP and CDP/CDLPI was also assessed. Among the 11 patients with CARP included, the PDAI score showed a significant reduction (median 10 to 4, p = 0.006) and 8/11 (72%) achieved remission based on PDAI score. Of the six patients with CDP/CDLPI included, the mPDAI demonstrated a significant reduction from baseline after 12 months, decreasing from a median score of 9 to 1 (p = 0.034). Remission based on mPDAI was achieved by five out of six patients (83.3%).

Our findings indicate that patients with EIMs prior to surgery were more likely to develop acute pouchitis, consistent with previous studies [17,18,19,20]. However, our study did not find a significant association between the extent of UC or smoking status and the development of acute pouchitis. Instead, we observed that preoperative transmural inflammation, as detected via cross-sectional imaging, was associated with a higher likelihood of later pouchitis development. This observation is intriguing, as it aligns with recent evidence suggesting that UC, traditionally considered a mucosal disease, may involve deeper layers of the bowel wall such as the submucosal layer [38,39,40,41,42,43]. Scores like the Milano Ultrasound Criteria have been validated for predicting colectomy [43], but they have not been used to predict postoperative complications yet.

Pouch failure occurred in 8.2% of our cohort, a rate consistent with the literature, which reports failure rates ranging from 2% to 15% [36]. Recent studies identify pelvic sepsis, chronic pouchitis, PSC, and fistulas as key risk factors for pouch failure [44,45]. However, in our cohort, patients with EIMs or chronic pouchitis did not show a higher propensity for pouch failure. This finding could be due to the small sample size. Consistent with previous reports, all patients diagnosed with CD developed acute pouchitis and CDP, making them more prone to pouch failure [46,47]. Postoperative fistulas were more common among patients who experienced pouch failure. To determine whether these fistulas were associated with CDP, we examined their prevalence in patients with CDP, finding a higher occurrence: 25% of patients with a prior diagnosis of CD developed fistulas, compared to 15.8% of those without a CD diagnosis. However, this difference was not statistically significant. A multivariate analysis confirmed that postoperative fistulas remained a significant risk factor for pouch failure, even after accounting for a prior diagnosis of CD.

Although our study was not designed to specifically assess the effectiveness of individual therapies in patients with inflammatory disorders of the pouch, we addressed the scarcity of high-quality evidence by collecting data at baseline and after 12 months of therapy when available. In patients with CARP, biologic therapy demonstrated clinical improvement, although some endoscopic parameters, such as ulcers, did not show a significant change. The recent AGA guidelines for the management of pouchitis recommend focusing on symptom relief as the primary treatment objective, as there is insufficient evidence to support the endoscopic and/or histologic resolution of inflammation as critical treatment targets [48]. It would be important to understand if endoscopic improvement should be sought in these patients, as achieving an endoscopic resolution could potentially offer better long-term outcomes and more comprehensive disease control. The persistence of ulcers also suggests the multifactorial nature of pouchitis, with factors such as ischemia potentially playing a role.

For the six patients with CDP/CDLPI, clinical outcomes after 12 months of biologic therapy showed notable improvements but did not reach statistical significance. Endoscopic evaluations revealed significant changes in edema and a loss of a vascular pattern though hyperemia, and other features showed no significant changes. Nonetheless, the change in the mPDAI was significant, and five out of six patients achieved mPDAI-based remission. This results are in line with previous findings [46,47,49,50,51,52,53,54,55,56].

Our study has some limitations, including its retrospective design, single-center setting, and limited sample size. The heterogeneity of treatments and missing data further constrained the study, leading to a reduced sample size in the effectiveness analysis. Outcome measures such as the PDAI and mPDAI involve subjective assessments that can vary between clinicians, potentially impacting the consistency and reliability of the results [37].

Our findings provide valuable insights into the risk factors for developing pouchitis and pouch failure following restorative proctocolectomy with IPAA. Identifying preoperative factors such as transmural inflammation and EIMs can help clinicians stratify patients by risk, facilitating more personalized postoperative surveillance and early interventions. Understanding these risk factors can improve patient counselling before surgery and influence decisions on the timing and type of surgical interventions. Additionally, our study supports the clinical utility of biologic therapies in managing chronic pouchitis, particularly CARP, CDP, and CDLPI.

Future research should involve larger, multicenter studies to increase the generalizability of findings and provide more robust data on the incidence and risk factors for postoperative complications in patients undergoing RPC with IPAA. Conducting RCTs to compare the effectiveness of different biologic therapies in managing chronic pouchitis and preventing pouch failure will help establish clearer treatment guidelines. Additionally, research should evaluate the benefits of achieving endoscopic and histologic resolution of inflammation in pouchitis treatment, beyond merely alleviating symptoms. Research on pouchitis is complicated by the burden of its classifications and scoring systems, and further efforts are needed to standardize clinical research methods.

5. Conclusions

IPAA remains the preferred surgical treatment for UC, but complications such as pouchitis are common. This study identified several significant risk factors for pouchitis, including preoperative EIMs, CD, and transmural inflammation detected via cross-sectional imaging before surgery. Pouch failure was strongly associated with a prior diagnosis of CD and the presence of postoperative fistulas. While biological therapy was assessed, its impact should be interpreted cautiously due to the small sample sizes. Biological therapy showed a significant reduction in disease activity for patients with CARP and CDP/CDLPI. These findings highlight the importance of identifying risk factors preoperatively to improve patient management and outcomes in IPAA procedures.