Pouch-Related Symptoms and Quality of Life in Patients with Ileal Pouch-Anal Anastomosis

Abstract

Background

Restorative proctocolectomy with ileal pouch-anal anastomosis (IPAA) has become the standard surgical treatment for the majority of patients with inflammatory bowel disease (IBD) who require colectomy. We evaluated the prevalence of pouch-related symptoms among the Crohn’s and Colitis Foundation of America (CCFA) Partners cohort and the effect of pouch-related symptoms on Patient-Reported Outcome Measurement Information System (PROMIS) measures.

Methods

We performed analyses nested in the CCFA Partners cohort. We used bivariate analyses to compare demographics and medication use among patients with ulcerative colitis (UC) or indeterminate colitis (IC) and pouch-related symptoms and those with IPAA without symptoms. We also compared PROMIS domains (measured in T-scores) and short IBD Questionnaire (SIBDQ) quality of life scales between symptomatic pouch patients (over the past 6 months) and those without symptoms.

Results

Among 243 patients reporting a history of IPAA, 199 (82%) reported a history of pouch symptoms. Patients with recent pouch symptoms demonstrated higher mean T-scores in pain interference (53.0 vs. 45.3, p<0.001), depression (51.0 vs. 46.4, p=0.002), and fatigue (56.3 vs. 47.0, p<0.001). Symptomatic pouch patients reported lower mean scores in social role satisfaction (47.4 vs. 54.6) and SIBDQ (4.8 vs. 5.8), both p-values <0.001. These differences were all clinically meaningful.

Conclusions

In a large sample of patients with IBD, nearly all patients with IPAA reported a history of pouch symptoms. Patients experiencing symptoms within the 6 months prior to survey assessment demonstrated clinically meaningful decrements in patient-reported outcomes in multiple domains of physical and psychosocial functioning.

BACKGROUND

Approximately 20–35% of patients with ulcerative colitis (UC) eventually will require colectomy due to refractory disease or histologically proven dysplasia.^1–3 Given the ability to accomplish the major goals of eliminating the diseased segment of intestine while maintaining fecal continence, restorative proctocolectomy with ileal pouch-anal anastomosis (IPAA) has become the standard surgical treatment for the majority of patients who require colectomy. Additionally, the pouch procedure has been shown to significantly improve patients’ health related quality of life (QOL) while reducing the risk for colitis-associated neoplasia.⁴

There are known complications that can occur after IPAA surgery. Among the potential complications, pouchitis is the most common long-term complication after IPAA, affecting 50–79% of patients after IPAA for UC.^1,5–13 Several risk factors for pouchitis have been suggested in prior studies, including extensive UC,^14,15 a concurrent diagnosis of primary sclerosing cholangitis (PSC),^16–18 backwash ileitis,^18,19 non-smoking status,^15,20 the regular use of NSAIDs,^15,17 and ischemia.²¹

The majority of the studies that have evaluated pouchitis and pouch-related impact thus far have been performed in single center cohorts. However, as clinical outcomes may vary depending on the center where pouch procedures are performed,² a broader understanding of pouchitis and the effect of pouch-related symptoms on related clinical outcomes is needed. The primary objective of our study was to use the Crohn’s and Colitis Foundation of America (CCFA) Partners infrastructure to evaluate the prevalence of pouch-related symptoms and the impact of pouch-related symptoms on patient-reported outcomes (PROs) in participants with a prior history of IPAA. We used the Patient-Reported Outcomes Measurement Information System (PROMIS) measures including anxiety, depression, sleep disturbances, fatigue, pain interference, and social satisfaction to compare symptomatic pouch patients to those patients without pouch-related symptoms.

METHODS

Study Population

We identified patients with a history of IPAA within CCFA Partners, an internet based cohort of patients with IBD. The study cohort has been described in detail previously.²² Briefly, patients were recruited to enroll in this online cohort registry via a variety of means, including invitations via email, social media, and recruitment at CCFA educational events. Over 15,000 patients with self-reported IBD have enrolled in the cohort since initiation in 2011, and cohort members are followed up at 6 month intervals. Baseline and follow up surveys include a core survey with information on disease phenotype, activity, medication use, and PROs. Optional “modules” can also be completed alongside the core survey at 6 month intervals to assess specific areas of interest among patients with IBD, including pouch related complications and QOL.

For inclusion in this study, patients needed to be ≥ 18 years of age, diagnosed with UC or IC, with a history of colectomy with IPAA. Patients with a diagnosis of Crohn’s disease were excluded from this study. There were no other exclusion criteria.

Study Variables

All patients completed questionnaires regarding demographic and clinical information, in addition to validated scales including the short IBD questionnaire (SIBDQ),²³ and the PROMIS measures for pain interference, anxiety, depression, satisfaction with social role, sleep disturbance, and fatigue.^24,25 The SIBDQ has not specifically been validated in pouch populations, however the instrument includes domains with clinical relevance to this IBD sub-population including bowel symptoms, systemic symptoms, emotional function, and social function and has been used in the evaluation of patients with Crohn’s disease phenotypes of the pouch.²⁶

In addition to the survey measures described above, patients were questioned regarding any history of symptoms of pouchitis including abdominal pain, cramping, and urgent or frequent bowel movements and then separately regarding a history of pouch-related symptoms within the preceding 6 months. If a patient completed multiple surveys within the study period, all survey responses were analyzed for a positive history of pouch-related symptoms at any point. If a patient answered positively to the question regarding pouch-related symptoms in the preceding 6 months on multiple surveys, then the first survey where pouch-related symptoms were reported was analyzed for the evaluation of PROMIS measures. If a patient completed multiple surveys and never experienced pouch-related symptoms, then the last survey completed was analyzed for the evaluation of PROMIS measures. This method was chosen to allow for more complete capture of patients without pouch-related symptoms, as the question regarding pouch-related symptoms in the preceding 6 months may not have been available in the baseline survey depending on how long an individual had participated in CCFA Partners.

The PROMIS measures have previously been assessed within the CCFA Partners cohort,²⁵ and have been validated in other general and chronically ill populations for the self-reporting of outcomes.^27,28 PROMIS measures are calibrated using T-score metrics to provide mean values for each domain, with higher scores indicating higher levels of each domain being measured. Among generally healthy populations, the mean T-score for pain interference, depression, anxiety, satisfaction with one’s social role, fatigue, and sleep disturbance is 50 with a standard deviation (SD) of 10. Based on prior studies evaluating the PROMIS measures in patients with IBD and other chronic conditions such as rheumatoid arthritis,^25,29–32 a minimally important difference of 2 was assigned when analyzing PROMIS measures.

Statistical Analysis

Baseline demographics and clinical characteristics were compared using proportions and 95% confidence intervals (CI), means with SD, and medians with interquartile ranges as appropriate. Bivariate statistics using Pearson’s chi-square test statistic, Fisher’s exact, Wilcoxon rank sum, and Student’s t-test as appropriate were used to compare symptomatic pouch patients to those patients with no reported history of symptoms. Logistic regression modeling was used to compare the effect of pouch-related symptoms within the prior 6 months on individual PROs as assessed by the PROMIS measures, after controlling for covariates. When using logistic regression, the minimally important difference of 2 in each PRO scale was used as the threshold for assessment. Categorical variables for each PROMIS domain were created using a threshold of ≥ 52 for pain interference, depression, anxiety, fatigue, sleep disturbance and a threshold of ≤ 48 for satisfaction with social role. In multivariable analyses, an a priori decision was made to include age and sex in all models, given potential for confounding. All statistical analyses were performed using SAS (version 9.4) statistical software (SAS Institute, Cary, NC, USA). The study protocol was approved by the Institutional Review Board at the University of North Carolina at Chapel Hill.

RESULTS

A total of 243 participants in CCFA Partners reported a history of colectomy with IPAA. Of these, 199 (82%) reported a history of pouch-related symptoms on at least one survey at any time in the interval following their initial IPAA surgery. There were no significant differences in age, sex, or race when comparing symptomatic pouch patients to patients without a history of pouch-related symptoms (Table 1). Significant differences were noted when analyzing the number of years since diagnosis of IBD (p=0.034), with 89% of patients with colitis for more than 11 years reporting at least one episode of pouch-related symptoms (Table 1). At each time point analyzed, a greater percentage of patients reported at least one episode of pouch-related symptoms when compared to those patients with a pouch who had not experienced symptoms (Figure 1).

Table 1.

Comparison of demographics and clinical characteristics among patients from CCFA Partners with no history of pouch-related symptoms and symptomatic pouch patients

	Patients without pouch-related symptoms (n=44)	Symptomatic pouch patients (n=199)
Age (in years)
18–39	21 (48%)	109 (55%)
40–59	19 (43%)	74 (37%)
≥60	4 (9%)	16 (8%)
Female	27 (61%)	141 (71%)
Race
White	38 (86%)	182 (91%)
Black	0	3 (2%)
Other/Not Reported	6 (13%)	14 (8%)
Current Smoker	0	5 (3%)
Time since IBD Diagnosis
<1 year	3 (7%)	9 (5%)
1–5 years	17 (39%)	39 (20%)
6–10 years	11 (25%)	46 (23%)
11–15 years	4 (9%)	43 (22%)
>15 years	9 (20%)	62 (31%)
Current Therapy
Antibiotics	0	60 (30%)
Probiotics	10 (23%)	72 (36%)
Oral Steroids	1 (2%)	14 (7%)
Aminosalicylates	0	20 (10%)
Biologic Therapy	0	11 (6%)
Immunosuppressive Therapy	0	7 (4%)
Narcotic Pain Medications	5 (11%)	29 (15%)
Current GI Provider Setting	(n=30)	(n=162)
University Hospital	9 (30%)	46 (28%)
Community Practice	15 (50%)	100 (62%)
Other/Don’t Know Setting	6 (20%)	16 (10%)
Number of times a Gastroenterologist has been seen in the past year	(n=30)	(n=162)
None	7 (23%)	24 (15%)
1 or 2 times	15 (50%)	82 (51%)
3 or 4 times	4 (13%)	24 (15%)
5 or more times	4 (13%)	32 (20%)

Figure 1.

The percentage of patients with and without pouch-related symptoms compared to the number of years since diagnosis of ulcerative colitis/indeterminate colitis

Symptomatic pouch patients were more likely to report use of antibiotics (30% vs. 0, p<0.001) and aminosalicylates (10% vs. 0, p=0.030) when compared to asymptomatic patients. No significant differences were noted in use of probiotics, narcotic pain medications, oral steroids, biologic agents, or immunosuppressive therapies (Table 1). There was no significant difference in practice setting when comparing symptomatic pouch patients to those patients without symptoms, however a minority of patients in both groups reported being treated at a University hospital (Table 1).

When examining PROMIS measures, a total of 222 (91%) patients completed questionnaires regarding PROMIS and QOL measures. Of these, 187 (84%) reported pouch-related symptoms in the preceding 6 months. Symptomatic pouch patients demonstrated greater pain interference, depression, fatigue, and less satisfaction with their social role (Table 2). All of these differences were both statistically significant and clinically meaningful, as defined by the pre-specified minimally important difference. Additionally, patients with symptoms in the preceding 6 months reported decreased QOL as defined by the SIBDQ when compared to patients without pouch-related symptoms (mean SIBDQ 4.8 vs. 5.8, p<0.001).

Table 2.

Comparison of PROMIS and quality of life measures among patients from CCFA Partners with a history of pouch-related symptoms within 6 months of survey and patients with no history of pouch-related symptoms

	Patients without pouch-related symptoms (n=35)	Patients with pouch-related symptoms in the preceding 6 months (n=187)	p-value
Pain interference	45.3 (6.4)	53.0 (10.9)	<0.001
Depression	46.4 (7.1)	51.0 (9.5)	0.002
Anxiety	48.8 (7.4)	51.9 (9.7)	0.075
Satisfaction with social role	54.6 (7.6)	47.4 (9.8)	<0.001
Fatigue	47.0 (8.2)	56.3 (10.5)	<0.001
Sleep Disturbance	51.9 (3.6)	51.8 (3.8)	0.854
SIBDQ	5.8 (0.9)	4.8 (1.2)	<0.001

Scores reported as mean (Standard Deviation)

Using multivariable logistic regression to evaluate the independent association between pouch-related symptoms within the preceding 6 months on PROs, several significant associations were demonstrated. In both unadjusted analysis and adjusted analyses, symptomatic pouch patients demonstrated worse outcomes on several PROs as defined by PROMIS measures (Table 3). After adjusting for age and sex, symptomatic pouch patients in the preceding 6 months demonstrated greater pain interference [Adjusted Odds Ratio (aOR) 3.99, 95% CI 1.53 – 10.41], and fatigue (aOR 4.52, 95% CI 2.88 – 10.90). Patients with pouch-related symptoms in the preceding 6 months also reported decreased satisfaction with their social role (aOR 0.24, 95% 0.09 – 0.63).

Table 3.

Odds of experiencing worse outcomes as defined by PROMIS measures among patients from CCFA Partners with a history of pouch-related symptoms within 6 months of survey compared to patients with no history of pouch-related symptoms^a

	Unadjusted OR (95% CI)	Adjusted ORb (95% CI)
Pain interference	4.15 (1.61–10.72)	3.99 (.153–10.41)
Depression	2.27 (0.92 – 5.60)	2.07 (0.83 – 5.16)
Anxiety	2.19 (0.95 – 5.08)	2.09 (0.90 – 4.90)
Satisfaction with social role	0.24 (0.09 – 0.62)	0.24 (0.09 – 0.63)
Fatigue	4.79 (2.00 – 11.45)	4.52 (2.88 – 10.90)
Sleep Disturbance	0.82 (0.36 – 1.83)	0.82 (0.36 – 1.87)

^aFor categorical analysis, PROMIS measures were divided at the point of clinically meaningful difference (≥ 52 for pain interference, depression, anxiety, and fatigue; ≤ 48 for satisfaction with social role)

^bMultivariable analysis adjusted for age and sex

DISCUSSION

In an analysis of a large, geographically diverse cohort of patients with UC or IC reporting a history of proctocolectomy with IPAA for IBD, we evaluated the prevalence of patient reported pouchitis symptoms, which is the most common long term complication associated with IPAA. Our population was treated in multiple practice settings, with a majority of symptomatic pouch patients being treated in a community based practice. We demonstrated that pouch-related symptoms occur frequently in patients who undergo restorative proctocolectomy with IPAA for IBD related reasons, and that antibiotics and probiotics appear to be the most frequently used therapy. Symptomatic pouch patients with symptoms in the prior 6 months demonstrated worse PROs as defined by PROMIS measures, including increased pain interference, depression, and fatigue, as well as decreased satisfaction with the social role and decreased QOL as defined by the SIBDQ.

The frequency of reported pouch-related symptoms after colectomy of 82% in our cohort study is higher than previously assessed in most of the single center studies.^1,5–8 The higher incidence of pouch-related symptoms could be due to the variable window of assessment of pouchitis symptoms used in our surveys compared to other studies. Another potential explanation is that patients are also treated for pouch-related symptoms by their primary care providers and thus do not always report their symptoms to the center that performed the colectomy with IPAA. Additionally, we were unable to verify a diagnosis of self-reported pouchitis symptoms via endoscopy or review of the medical record.

Gut microbiota may be involved in the pathogenesis of pouchitis through multiple mechanisms, including an alteration in commensal bacteria.^33–36 This focus on the microbiota was reflected in medication use patterns of the patients in our study, as a significantly greater number symptomatic pouch patients reported current use of antibiotics as compared to those patients with no history of pouch-related symptoms. A greater number of patients with pouch-related symptoms also reported use of probiotics, though this result was not statistically significant. While antibiotics such as ciprofloxacin or metronidazole are viewed as the first line therapy for acute pouchitis,^4,37–39 our findings confirm that among a diverse survey of patients with IPAA, antibiotics and probiotics are being utilized more widely among symptomatic pouch patients than anti-inflammatory medications such as steroids, biologic therapy, or other immunosuppressive agents. Interestingly 11% of patients were also treated with 5-ASA for pouch-related symptoms, which has only proven to be effective for the local treatment of cuffitis but not pouchitis.^40,41

Classically, the majority of pouch surgeries have been performed in young patients with long life expectancies following colectomy with IPAA. As patients age, the risk of developing pouchitis increases,⁴² a finding that was suggested in our study. Our survey did not specifically ask the date of colectomy or ileostomy takedown, and thus we were unable to analyze the time since the last surgery related to the IPAA. In prior longitudinal evaluations where cohorts were followed for up to 15 years after IPAA, up to 47% of patients had experienced at least one episode of pouchitis.^3,42,43 While we were unable to evaluate time since IPAA, in our population, 33% of symptomatic pouch patients had been diagnosed with UC/IC greater than 15 years ago. Additionally, the majority of patients that had undergone IPAA and had been diagnosed with IBD more than 11 years prior to completing the survey reported at least one episode of pouch-related symptoms. When evaluating time since diagnosis in this population of patients with IPAA, a greater proportion of patients had experienced pouch-related symptoms at every time point analyzed as compared to patients without symptoms.

Early studies indicated that long-term QOL following ileal pouch surgery was excellent,⁴⁴ with similar cohorts demonstrating good QOL compared to the general population^45–48 despite initial surgery-specific complications in some cases.⁴⁹ Improvements in QOL have been noted as early as one month after ileostomy takedown.⁵⁰ However, other studies have demonstrated decreased QOL and functional outcomes among patients undergoing IPAA when compared to the general population,⁵¹ while another study noted no major improvement in QOL after IPAA.⁵² When QOL has been examined more specifically among patients with a history of pouchitis, the effects on QOL appear more significant. A history of pouchitis has been associated with negative effects on multiple QOL scales,⁵³ and patients experiencing chronic pouchitis have demonstrated decreased QOL and satisfaction with surgery.⁵⁴

Although the PROMIS measures have been validated among patients with IBD, another key advantage of evaluating QOL with the PROMIS measures is the ability to compare to the general population. We found that patients with a pouch who were symptomatic within the preceding 6 months demonstrated clinically meaningful worsened outcomes in several areas assessed by PROMIS measures, including pain interference, depression, satisfaction with social role, and fatigue when compared to patients without pouch-related symptoms. However, even symptomatic pouch patients demonstrated PROMIS T-scores within 1 SD of 50, indicating that while pouch-related symptoms were associated with worse outcomes in several domains of QOL, these may not be as significant as in other chronic diseases.^55,56 These findings may be particularly helpful when discussing potential outcomes in the pre-operative period while patients are considering colectomy. Additionally, those patients without pouch-related symptoms demonstrated PROMIS T-scores that were better than the mean scores of the general population in multiple domains.

Our study is the first to utilize the PROMIS measures to evaluate PROs among patients with IBD who have undergone proctocolectomy with IPAA. The PROMIS initiative of the National Institutes of Health was developed to advance the application of PROs in research and clinical practice.²⁴ While other studies have evaluated QOL among patients with a history of IPAA, we believe that the use of the PROMIS measures in a geographically diverse sample is a strength of this study.

The majority of the prior studies that have evaluated pouchitis have largely been limited to single center reports. While these studies have significantly increased our understanding of the risk factors associated with the development of pouchitis among patients undergoing IPAA and the epidemiology of pouchitis among patients with IBD undergoing restorative proctocolectomy with IPAA, the fact that these cohorts arise from single centers of excellence may indicate that these studies are less representative of the experiences of patients undergoing IPAA in the general population. Our study evaluated the experiences of patients with IPAA across practice settings, including both university hospitals and community practices. While there was no significant difference when evaluating a history of pouch-related symptoms among patients seen in differing practice settings, we view the ability to evaluate patients across a more broad geographic area and multiple practice settings as a strength.

Our study has limitations as well. Although a subset of the CCFA Partners cohort has been validated for factors such as disease diagnosis,⁵⁷ the history of pouchitis symptoms and IPAA surgery for this study are self-reported. The inability to validate patient-reported pouchitis may have led to a report of pouchitis when symptoms could have been due to other conditions such as irritable pouch syndrome or Crohn’s disease of the pouch. Given our inability to validate the diagnosis of pouchitis, we have utilized a more appropriate diagnosis of pouch-related symptoms in our analyses. As noted previously, the inability to evaluate the relationship between time since IPAA surgery and risk of development of pouch-related symptoms is a significant limitation. Although we demonstrated a trend in overall time since IBD diagnosis, the more clinically meaningful relationship between time since surgery and symptoms could not be assessed. Despite nesting this study in a cohort of over 15,000 participants, the population reporting a history of restorative protocolectomy with IPAA was relatively small. The population of patients without pouch-related symptoms was even smaller, limiting the number of variables that we could include in our final multivariable analyses.

In conclusion, in an analysis of patients with UC or IC who underwent restorative proctocolectomy with IPAA, those patients with pouch-related symptoms within the past 6 months demonstrated significantly worse PROs including pain interference, depression, satisfaction with social role, and fatigue. Given that symptomatic pouch patients demonstrated worse PROs, increased awareness by providers of the impact of pouchitis symptoms on an individual patient’s QOL appears warranted. In addition to standard treatment for pouchitis including antibiotics, symptomatic pouch patients may require alternative treatments targeting PROs such as depression, pain, and social satisfaction.