Anorectal Manometry in Patients With Fecal Incontinence After Ileal Pouch–Anal Anastomosis for Ulcerative Colitis: A Cohort Study

Abstract

Background

Fecal incontinence commonly occurs in patients with ulcerative colitis and ileal pouch–anal anastomosis. There is a paucity of manometric data in pouch patients. We aimed to better define manometric parameters in pouch patients with fecal incontinence.

Methods

We compared clinical and manometric variables in ulcerative colitis patients with pouch and fecal incontinence to ulcerative colitis patients with pouch without fecal incontinence and to non-ulcerative colitis patients with fecal incontinence. Manometric data for the 3 cohorts were compared to established normative data. An independent-samples t-test was performed for continuous variables, and chi-square test was used for categorical variables. Logistic regression was performed to identify predictors of incontinence in pouch patients (P < .05).

Results

Among 26 pouch patients with fecal incontinence (73% female), 26 pouch patients without fecal incontinence (35% female), and 84 patients with fecal incontinence without ulcerative colitis (68% female), there were no differences in anorectal pressures between patients with fecal incontinence. Lower pressures were observed in pouch patients with fecal incontinence compared to those without fecal incontinence. Resting pressure was similar between pouch patients with fecal incontinence and healthy controls (60.9 ± 36.1 mmHg vs. 66.9 ± 3.2 mmHg, P = .40). Female sex (P = .019) and defecatory disorders (P = .033) each independently predicted fecal incontinence in pouch patients.

Conclusions

Pouch patients with fecal incontinence have lower anorectal pressures compared to pouch patients without incontinence, though have similar pressures to non-ulcerative colitis patients with fecal incontinence. Pouch patients with fecal incontinence have similar resting pressures as healthy controls. Distinct manometric normative values for pouch patients are needed.

Graphical Abstract

Graphical Abstract.

Introduction

Fecal incontinence is common in patients with inflammatory bowel disease (IBD) and occurs in up to 24% in ulcerative colitis (UC).^1–6 Symptoms include passive and urge incontinence, as well as incontinence to gas, liquid, and solid stools.³ Fecal incontinence (FI) in UC is often associated with disease-induced inflammation, repeated surgeries, disease duration, and altered rectal sensitivity.^1,6–8 Prior manometric studies of IBD patients with FI have demonstrated lower resting and squeeze pressures, rectal hypersensitivity, and increased external anal sphincter fatigability.^1,8 These changes are associated with disease activity, though can persist even in the absence of active inflammation.^7,9

Ileal pouch–anal anastomosis (IPAA) is the preferred surgery for patients with UC requiring colectomy.¹⁰ It is estimated that up to 10%–15% of patients with UC will require IPAA for medically refractory disease or dysplasia.¹¹ The benefits of IPAA are numerous and include avoidance of a permanent stoma, preservation of the normal route of defecation, clinical improvement without immunosuppressive medications, and decreased colorectal cancer risk.^10,12 However, FI occurs in up to 25%–70% of patients following IPAA.^13–15

Dyssynergic defecation (DD) is characterized by excessive straining and incomplete evacuation in the absence of structural outlet obstruction.¹⁶ Prior studies have demonstrated that DD is common in patients with IPAA, occurring anywhere from 25% to 75% of patients.¹⁷ High-resolution anorectal manometry (HRAM) is used to evaluate both DD and FI in patients, though current normative values are based on patients with intact rectums.¹⁸ Comparatively, manometric parameters are not well defined in patients with IPAA. A recent study found that even in patients with healthy pouches, nearly 30% have previously undocumented functional symptoms and/or structural changes seen on imaging.¹⁹ It has been posited that different normative HRAM values may be needed to account for the postsurgical changes following IPAA, which include both lower resting and squeeze pressures and sensory dysfunction, changes that are also associated with FI.^19–22 While studies have sought to identify manometric values in patients with IPAA, none have specifically compared anorectal manometry findings in UC patients with IPAA and FI to non-IBD patients with FI.

In this study, we compare clinical characteristics, anorectal function, and manometric findings between UC patients with IPAA and FI to 2 cohorts—UC patients with IPAA without FI and non-IBD patients with FI. With these 2 relevant comparator groups, the aim of our study is to better characterize HRAM parameters in IPAA patients with FI, a population where manometry is not currently well studied, defined, or standardized. We hypothesized that patients with FI would have lower resting and squeeze pressures and decreased rectal sensitivity compared to healthy controls without FI and that these changes would be similar in patients with FI with or without IPAA.

Materials and Methods

Patient Characteristics

Retrospective review was conducted on patients undergoing outpatient HRAM from January 2017 to July 2022 at NYU Langone Medical Center (NYULMC). Patients were identified through the NYU Gastroenterology Motility Lab High-Resolution Anorectal Motility database. Inclusion criteria were (1) adult patients ages 18 and above, (2) at least 1 HRAM evaluation at NYULMC performed by a board-certified gastroenterologist specializing in motility (SS), and (3) those with FI defined as at least 2–4 episodes of involuntary loss of solid, mucus, or liquid stools in 4 weeks with symptom onset for ≥6 months based on Rome IV Criteria.²³ For patients with IPAA without FI, HRAM was obtained for assessment of suspected defecatory disorder. Those who underwent HRAM for alternate reasons (rectal prolapse, rectal pain, tenesmus, incomplete evacuation, fecal urgency), and those with IPAA for non-IBD indications (familial adenomatous polyposis, severe constipation) were also excluded. Patients presenting for HRAM had alternate etiologies of diarrhea excluded prior to referral, including small intestinal bacterial overgrowth, malabsorption, infection, malignancy, and endocrinopathy. For IPAA patients with a history of acute pouchitis, antibiotics were trialed prior to referral to HRAM.

Patients were divided into 3 cohorts: UC patients with IPAA and FI (IPAA-FI), UC patients with IPAA without FI (IPAA-no-FI), and non-IBD patients with FI (non-IBD-FI). We compared demographic, clinical, and manometric variables between the IPAA-FI and IPAA-no-FI cohorts, as well as the IPAA-FI and non-IBD-FI cohorts. A regression analysis was performed to identify predictors of FI among IPAA patients. In a secondary analysis, HRAM data for all 3 cohorts were compared to established gender-controlled data of healthy controls based on manometry findings of 45 healthy subjects conducted by Rao et al.¹⁸

Anorectal Manometry

HRAM catheters were manufactured by Medtronic, Inc., formerly Given Imaging (Yoqneam, Israel, 2015). The solid-state catheter (ManoScan 360AR catheter) has an outer diameter of 4.2 mm with 12 circumferential sensors, including 10 sensors at 6-mm intervals along the anal canal and 2 sensors in the rectal balloon (3.3 cm long, maximum capacity of 400 mL). Data were recorded utilizing a proprietary system and data acquisition software (Manoscan 360AR; Medtronic, Inc.). Four anorectal functions were evaluated: recto-anal reflex activity, anal sphincter function, recto-anal coordination during simulated defecation, and rectal sensory function.²⁴ HRAM pressures included resting, squeeze, cough, and push pressures. For cough and push pressures, 2 readings were obtained for each maneuver, of which the attempt associated with the greatest increase in pressure was recorded. For squeeze pressures, 2–3 readings were recorded—1–2 squeezes each lasting 5 seconds and 1 squeeze lasting 30 seconds of which the best was recorded.²⁵

Rectal sensation was measured through gradual inflation of the catheter balloon with air at a rate of 10 cc/s. The catheter balloon was inflated initially 10 cc at a time until 60 cc, then 30 cc at a time to 300 cc or maximum tolerable volume, with complete balloon deflation in between each volume. Sensory function included first sensation, constant sensation, desire to defecate, urgency to defecate, and maximum tolerable volume. Recto-anal inhibitory reflex (RAIR) was assessed by inflating the intrarectal balloon with 50 cc of air. Sphincter length was measured in centimeters, and the balloon expulsion test (BET) measured timed expulsion of a rectal balloon filled with 50 cc of water with normal defined as ≤1 minute.

Defecatory disorders, or disorders of recto-anal coordination, were also identified with HRAM. Dyssynergia is defined as inappropriate pelvic floor contraction with attempted defecation. Due to difficulty distinguishing between healthy controls and patients with constipation on HRAM, there has been a notable change in nomenclature away from the traditional subtypes of dyssynergia (types I–IV).^25–27 Consistent with this change, following the London Classification of anorectal disorders and International Anorectal Physiology Working Group consensus, we utilize “defecatory disorders” to refer to poor propulsion, inadequate anal sphincter relaxation, and/or dyssynergia (specifically paradoxical contraction) observed during HRAM.

Clinical and Demographic Variables

Demographic and clinical variables included age, sex, race, body mass index (BMI), tobacco use, diabetes, obstetric/gynecologic history, prior perianal or pelvic floor surgery, IBD, and urinary incontinence. Stool characteristics, specifically stool frequency (number of bowel movements/day), FI type (solid, liquid, gas), FI subtype (passive, urge, mixed), and daytime versus nocturnal FI were collected. Medications included antimotility agents (loperamide, diphenoxylate/atropine, opium tincture), antibiotics, and anti-inflammatory agents (5-aminosalicylic acid, topical/oral steroids, biologics, and immunomodulators). Surgical history included IPAA indication, surgery year, IPAA type, index versus redo pouch, pouch age, and history of pouchitis based on clinical symptoms, endoscopic findings, and/or histologic features resulting in a Pouchitis Disease Activity Index (PDAI) ≥7 or a modified PDAI score of ≥5.^28,29 Acute pouchitis was defined as clinical improvement to antibiotics within 4 weeks of HRAM. Chronic pouchitis was characterized into antibiotic-dependent, antibiotic-refractory, and Crohn’s disease of the pouch.³⁰

Statistical Methods

Data were collected into a password-protected standardized database program (Microsoft Excel, Seattle, WA). An independent-samples t-test was performed for continuous variables and chi-square test for categorical variables. Cases with missing data were excluded. Univariate logistic regression analyses were performed to identify predictors of FI among IPAA patients (P < .05). Statistically significant variables were then entered into a multivariate logistic regression analysis. Variables trending towards significance (P < .1) were adjusted for in the multivariate analysis to account for their effects on the outcome (FI in IPAA patients). All statistical analyses were performed in IBM SPSSv25 (Chicago, IL).

Ethical Considerations

The study was performed according to Good Clinical Practice guidelines, and approval was obtained from the NYU Langone Health Institutional Review Board. Individual informed consent was not required for this study. All authors had access to the study data and reviewed and approved the final manuscript.

Results

In total, 617 patients underwent HRAM between January 2017 and July 2022 at NYULMC. One hundred thirty-six patients met the inclusion criteria: 84 patients were in the non-IBD-FI cohort, 26 in the IPAA-FI cohort, and 26 in the IPAA-no-FI cohort (Figure 1). All 52 IPAA patients had biopsy-proven UC, with 3 also having suspected Crohn’s disease of the pouch.

Figure 1.

Flow diagram of patients meeting study inclusion criteria. FI, fecal incontinence; HRAM, high-resolution anorectal manometry; IBD, inflammatory bowel disease; IPAA, ileal pouch–anal anastomosis; UC, ulcerative colitis.

Baseline patient characteristics are presented in Table 1. The average age of IPAA-FI patients was 45.7 ± 14.1 years compared to 38.5 ± 13.8 years in IPAA-no-FI patients (P = .07) and 65.4 ± 15.0 years in non-IBD-FI patients (P < .001 compared to IPAA-FI). Seventy-three percent of IPAA-FI patients were female compared to 35% of IPAA-no-FI patients (P = .012). There were no differences in BMI, diabetes, tobacco use, pregnancy rates, or obstetric/gynecologic surgical history between the 3 cohorts. Notably, 31% of IPAA-FI patients had prior perianal surgery (including hemorrhoidectomy, fissurectomy, fistulotomy, perianal abscess drainage, rectal prolapse repair, and stricture dilation) compared to only 7% of non-IBD-FI patients (P = .004).

Table 1.

Baseline clinical characteristics between the IPAA-FI, IPAA-no-FI, and non-IBD-FI cohorts.

	IPAA-FI (n = 26)	IPAA-no-FI (n = 26)	P valuea	Non-IBD-FI (n = 84)	P valueb
Age (years)	45.7 ± 14.1	38.5 ± 13.8	.07	65.4 ± 15.0	<.001***
Female sex	19/26 (73%)	9/26 (35%)	.012*	57/84 (68%)	.81
BMI (kg/m2)	24.3 ± 4.4	23.6 ± 4.0	.56	25.8 ± 6.0	.23
Diabetes	4/26 (15%)	3/26 (12%)	1.00	21/84 (25%)	.42
Tobacco use	6/26 (23%)	7/26 (27%)	1.00	33/84 (39%)	.16
Pregnancy history	11/19 (58%)	5/9 (56%)	1.00	40/55 (73%)	.26
Obstetric/gynecologic surgery history	11/17 (65%)	5/9 (56%)	.69	36/52 (69%)	.77
Daytime FI	25/26 (96%)	—	—	84/84 (100%)	0.24
Nocturnal FI	21/26 (81%)	—	—	20/84 (24%)	<.001***
Stool frequencyc	11.9 ± 4.6	7.8 ± 5.0	0.004**	2.3 ± 2.3	<.001***
Urinary incontinence	3/26 (12%)	0/26 (0%)	.24	35/84 (42%)	.005**
Perianal surgery historyd	8/26 (31%)	4/26 (15%)	.32	6/84 (7%)	.004**
Treatment with antimotility agentse	19/26 (73%)			13/84 (16%)	<.001***
Antibiotics within 30 days ARM	3/26 (12%)	5/26 (19%)	.70	2/84 (2%)	.09
Redo pouch	13/26 (50%)	5/26 (19%)	.04*	—	—
Pouch age (in months)	77.2 ± 118.6	73.0 ± 87.5	.89	—	—
Pouch age ≤ 12 months	10/26 (39%)	6/26 (23%)	.37	—	—
Acute pouchitisf	2/16 (13%)	1/18 (6%)	.59	—	—
Chronic pouchitisg	10/15 (67%)	8/15 (53%)	.71	—	—
Any pouchitis history	22/26 (85%)	20/26 (77%)	.73	—	—

Abbreviations: BMI, body mass index; FI, fecal incontinence; HRAM, high-resolution anorectal manometry; IBD, inflammatory bowel disease; IPAA, ileal pouch–anal anastomosis.

*denotes statistical significance (P < .05); **denotes statistical significance (P < .01); ***denotes statistical significance (P < .001).

^a P value comparing the IPAA-FI cohort to the IPAA-no-FI cohort.

^b P value comparing the IPAA-FI cohort to the non-IBD-FI cohort.

^cStool frequency is defined as number of bowel movements per day.

^dHistory of perianal surgery includes hemorrhoidectomy, fissurectomy, fistulotomy, perianal abscess drainage, rectal prolapse repair, and perianal stricture dilation.

^eTreatment with antimotility agents includes loperamide, diphenoxylate/atropine, and opium tincture trialed at any point in time prior to HRAM.

^fAcute pouchitis was defined as clinical improvement to antibiotics in IPAA patients within 4 weeks of HRAM.

^gChronic pouchitis included IPAA patients with antibiotic-dependent pouchitis, antibiotic-refractory pouchitis, and Crohn’s disease of the pouch.

Bolded text denotes analyses found to be statistically significant (P < .05).

IPAA-FI patients had an average of 11.9 ± 4.6 bowel movements per day compared to 7.8 ± 5.0 in the IPAA-no-FI cohort (P = .004) and 2.3 ± 2.3 in the non-IBD-FI group (P < .001). Eighty-one percent of IPAA-FI patients had nocturnal FI compared to 24% of non-IBD-FI patients (P < .001), though there was no difference in daytime FI between the 2 cohorts. Non-IBD-FI patients had higher rates of urinary incontinence at 42% versus 12% of IPAA-FI patients (P = .005).

Among IPAA patients, 73% were trialed on antimotility agents (loperamide, diphenoxylate/atropine, and opium tincture) compared to only 16% of non-IBD-FI patients (P < .001). With regards to acute pouchitis, there was no difference in antibiotic use within 30 days of HRAM between IPAA-FI patients and IPAA-no-FI or non-IBD-FI patients. There was no difference in pouch age, rates of acute pouchitis, or chronic pouchitis, though rates of redo pouch were higher in the IPAA-FI group at 50% versus 19% in IPAA-no-FI patients (P = .04).

Manometric Outcomes

IPAA patients with and without FI

Manometric outcomes between the 3 patient cohorts are presented in Table 2. Resting pressure was lower in IPAA-FI patients at 60.9 ± 36.1 mmHg compared to 84.3 ± 32.1 mmHg in IPAA-no-FI patients (P = .017). Max squeeze pressure was also lower at 156.2 ± 52.3 mmHg in the IPAA-FI cohort compared to 200.3 ± 71.1 mmHg in the IPAA-no-FI cohort (P = .014), as well as sustained squeeze pressure, which was 92.9 ± 35.2 mmHg in IPAA-FI patients versus 118.5 ± 43.4 mmHg among IPAA-no-FI patients (P = .023). The IPAA-FI cohort also had lower anorectal pressures during cough, with a recto-anal pressure gradient of -81.0 ± 48.8 mmHg compared to −113.6 ± 63.9 mmHg in IPAA-no-FI patients (P = .046). There were no differences in RAIR, sphincter length, or sensation thresholds between IPAA-FI and IPAA-no-FI patients. Higher rates of defecatory disorders were observed in IPAA-FI patients at 84% compared to 54% of IPAA-no-FI patients (P = .034).

Table 2.

Comparison of HRAM outcomes between the IPAA-FI, IPAA-no-FI, and non-IBD-FI cohorts.

	IPAA-FI (n = 26)	IPAA-no-FI (n = 26)	P valuea	Non-IBD-FI (n = 84)	P valueb
RAIRc	6/21 (29%)	6/24 (25%)	1.00	69/70 (99%)	<.001***
Sphincter length (cm)	2.6 ± 1.0	3.0 ± 0.9	.12	3.1 ± 0.7	.012*
Resting pressure (mmHg)	60.9 ± 36.1	84.3 ± 32.1	.017*	54.3 ± 23.0	.39
Max squeeze pressure (mmHg)	156.2 ± 52.3	200.3 ± 71.1	.014*	153.7 ± 85.3	.86
Sustained squeeze pressure (mmHg)	92.9 ± 35.2	118.5 ± 43.4	.023*	84.8 ± 45.7	.41
Rectal pressure, cough (mmHg)	44.4 ± 20.4	60.9 ± 29.2	.024*	66.8 ± 25.5	<.001***
Anal pressure, cough (mmHg)	125.4 ± 51.9	172.0 ± 72.7	.012*	126.0 ± 63.7	.97
Recto-anal pressure gradient, cough (mmHg)	−81.0 ± 48.8	−113.6 ± 63.9	.046*	−59.2 ± 53.2	.07
Rectal pressure, push (mmHg)	37.5 ± 22.7	39.4 ± 25.3	.77	46.4 ± 21.1	.07
Anal residual pressure, push (mmHg)	62.5 ± 31.7	73.7 ± 31.8	.21	65.0 ± 41.5	.78
First sensation (cc)	40.0 ± 44.8	46.4 ± 53.8	.65	22.8 ± 23.1	.07
Constant sensation (cc)	57.7 ± 47.7	65.6 ± 61.7	.61	36.4 ± 34.0	.014*
Desire to defecate (cc)	92.3 ± 65.2	125.0 ± 91.2	.16	86.0 ± 62.9	.66
Urge to defecate (cc)	128.1 ± 72.6	173.5 ± 93.0	.06	130.6 ± 69.8	.87
Max tolerable volume (cc)	175.4 ± 83.9	217.5 ± 89.0	.09	174.3 ± 72.0	.95
BETd	0/18 (0%)	2/15 (13%)		53/82 (65%)
Defecatory disorders on HRAMe	22/26 (84%)	14/26 (54%)	.034*	63/84 (75%)	.42

Abbreviations: BET, balloon expulsion test; FI, fecal incontinence; HRAM, high-resolution anorectal manometry; IBD, inflammatory bowel disease; IPAA, ileal pouch–anal anastomosis; RAIR, recto-anal inhibitory reflex.

*denotes statistical significance (P < .05); **denotes statistical significance (P < .01); ***denotes statistical significance (P < .001).

^a P value comparing the IPAA-FI cohort to the IPAA-no-FI cohort.

^b P value comparing the IPAA-FI cohort to the non-IBD-FI cohort.

^cRAIR = recto-anal inhibitory reflex. RAIR was unable to be assessed in 5 IPAA-FI patients, 2 IPAA-no-FI patients, and 14 non-IBD-FI patients.

^dBET assesses the ability of a patient to expulse the balloon catheter within 1 minute. BET was not assessed in 8 IPAA-FI patients, 11 IPAA-no-FI patients, and 2 non-IBD-FI patients.

^eDefecatory disorders include dyssynergia (specifically paradoxical contraction), poor propulsive force, and inadequate anal sphincter relaxation during the push maneuver on HRAM.

Bolded text denotes analyses found to be statistically significant (P < .05).

IPAA and non-IBD patients with FI

Sphincter length in IPAA-FI patients was shorter at 2.6 ± 1.0 cm compared to 3.1 ± 0.7 cm in the non-IBD-FI group (P = .012). Recto-anal inhibitory reflex was present in 99% of non-IBD-FI patients, while only 29% of IPAA-FI patients had intact RAIR (P < .001). Although rectal pressure during cough was lower in the IPAA-FI group at 44.4 ± 20.4 mmHg compared to 66.8 ± 25.5 mmHg in the non-IBD-FI group (P < .001), there was no difference in the recto-anal pressure gradient during cough between the 2 cohorts. There were also no differences in resting, squeeze, and push pressures between IPAA-FI and non-IBD-FI patients. Constant sensation in the IPAA-FI cohort was at higher volumes of 57.7 ± 47.7 cc compared to 36.4 ± 34.0 cc in the non-IBD-FI cohort (P = .014). There were no significant differences in first sensation, desire to defecate, urge to defecate, maximum tolerable volume, or rates of defecatory disorders between the 2 groups.

Although BET is standard in HRAM, the balloon frequently could not be expulsed through the ileoanal anastomotic ring in IPAA patients. This limited the test’s utility in this cohort and thus was not routinely performed. Statistical analyses of BET rates between the patient cohorts were deferred, given the limited data for IPAA patients.

Secondary Analysis: Comparison With Healthy Controls

HRAM data for the 3 cohorts were compared to established gender-controlled data of anorectal function in healthy controls.¹⁸ Compared to normative data, significantly shorter sphincter length was observed in all 3 cohorts (Table S1). There was no difference in resting pressure between IPAA-FI patients and normative data. However, IPAA-no-FI patients had higher resting pressures at 84.3 ± 32.1 mmHg compared to 69.6 ± 3.4 mmHg in healthy controls (P = .026). Non-IBD-FI patients had lower resting pressure than healthy controls at 54.3 ± 23.0 mmHg versus 67.3 ± 3.3 mmHg (P < .001). All 3 patient cohorts had negative recto-anal pressure gradients during cough, though patients with IPAA, regardless of FI, had significantly more negative recto-anal pressure gradients compared to normative data. Lower rectal pressures and higher anal residual pressures during push were observed across all 3 cohorts compared to normative data.

Both IPAA-FI and IPAA-no-FI patients had higher first sensation thresholds at 40.0 ± 44.8 and 46.4 ± 53.8 cc, respectively, versus 19.3 ± 0.5 and 19.7 ± 0.5 cc in healthy controls. For IPAA-FI patients, urge to defecate was 128.1 ± 72.6 versus 176.2 ± 5.4 cc in normative data, and maximum tolerable volume was 175.4 ± 83.9 versus 235.1 ± 8.6 cc in normative data. Non-IBD-FI patients also had lower sensation thresholds for desire to defecate, urge to defecate, and maximum tolerable volume compared to normative data (Table S1).

Associations of Clinical and Manometric Parameters With IPAA

On univariate logistic regression analysis, compared to IPAA-no-FI patients, IPAA-FI patients were more likely to be female (odds ratio [OR] 5.1, 95% confidence interval [CI] 1.6–16.8, P = .007), have increased stool frequency (OR 1.2, 95% CI 1.1–1.4, P = .009), have a redo pouch (OR 4.2, 95% CI 1.2–14.5, P = .024), and experience higher rates of defecatory disorders (OR 4.7, 95% CI 1.3–17.6, P = .021; Table 3). Multivariate logistic regression analysis adjusting for age, sex, defecatory disorders, and history of redo pouch surgery was then performed to identify associations with FI among IPAA patients. The multivariate logistic regression demonstrated that both female sex (P = .019) and defecatory disorders (P = .033) were independently associated with FI among IPAA patients. The model was highly significant (P = .002) and explained 48.1% of the data variance.

Table 3.

Univariate and multivariate logistic regression analyses associated with FI among IPAA patients: a comparison between IPAA-FI and IPAA-no-FI patients.

	Univariate analysis			Multivariate analysisa
	Odds ratio	95% CI	P value	Odds ratio	95% CI	P value
Age	1.0	1.0–1.1	.07	1.0	1.0–1.1	.65
Female sex	5.1	1.6–16.8	.007**	6.4	1.3–30.1	.019*
Diabetes	1.4	0.3–7.0	.69
BMI (kg/m2)	1.0	0.9–1.2	.55
Tobacco use	0.8	0.2–2.9	.75
Stool frequencyb	1.2	1.1–1.4	.009**	1.2	1.0–1.3	.06
Pregnancy history	1.1	0.2–5.4	.91
History of perianal surgeryc	2.4	0.6–9.5	.20
Pouch age (in months)	1.0	1.0–1.01	.88
Pouch age <12 months	2.1	0.6–7.0	.23
Redo pouch	4.2	1.2–14.5	.024*	2.6	0.6–11.7	.22
Acute pouchitis	2.1	0.2–24.5	.56
Chronic pouchitis	1.8	0.4–7.7	.46
Any pouchitis	1.7	0.4–6.7	.48
Defecatory disordersd	4.7	1.3–17.6	.021*	6.1	1.2–32.6	.033*
Sphincter length (cm)	0.6	0.4–1.1	.12
RAIR	1.2	0.3–4.5	.79

Abbreviations: BMI, body mass index; CI, confidence interval; FI, fecal incontinence; HRAM, high-resolution anorectal manometry; IPAA, ileal pouch–anal anastomosis; RAIR, recto-anal inhibitory reflex.

*denotes statistical significance (P < .05); **denotes statistical significance (P < .01); ***denotes statistical significance (P < .001).

^aMultivariate logistic regression analysis evaluating age, sex, defecatory disorders, and history of redo pouch surgery demonstrated significance with FI (P = .002) and explained 48% of the data variance.

^bStool frequency is defined as number of bowel movements per day.

^cHistory of perianal surgery includes hemorrhoidectomy, fissurectomy, fistulotomy, perianal abscess drainage, rectal prolapse repair, and perianal stricture dilation.

^dDefecatory disorders include dyssynergia (specifically paradoxical contraction), poor propulsive force, and inadequate anal sphincter relaxation during the push maneuver on HRAM.

Bolded text denotes analyses found to be statistically significant (P < .05).

Discussion

In this study of UC patients with IPAA, those with FI have lower resting, squeeze, and cough pressures and higher rates of defecatory disorders compared to IPAA patients without FI. However, IPAA-FI patients had similar resting pressures compared to established normative data. Patients with FI, regardless of IPAA, had similar pressure gradients and sensation thresholds, though rates of RAIR were lower in IPAA-FI patients compared to non-IBD-FI patients. Female sex and defecatory disorders were independently associated with FI among IPAA patients. Non-IBD-FI patients had higher rates of urinary incontinence compared to IPAA-FI patients, suggesting more global pelvic floor dysfunction in this cohort.

Though first sensation thresholds were higher among IPAA-FI and non-IBD-FI patients compared to normative controls, lower thresholds were observed for the other sensation parameters, suggesting overall rectal hypersensitivity. Furthermore, IPAA-FI and non-IBD-FI patients had lower squeeze and push pressures compared to established normative data. Interestingly, our results indicate that resting pressure in IPAA-FI patients was not significantly different than that of healthy controls based on current normative values.¹⁸ However, IPAA patients clinically differ from healthy controls and experience more liquid stools and higher stool frequency, which may explain why IPAA patients with normal resting pressures are more likely to have FI. In clinical practice, higher resting pressures may be needed in IPAA patients to decrease the risk of postsurgical fecal incontinence in this population. Our results further support the need for standardized HRAM values in IPAA patients to ensure a more accurate manometric assessment following IPAA. This is corroborated by a study by Quinn et al., where new HRAM values for IPAA patients with normal, healthy pouches were established to account for the rectal hypersensitivity and decreased compliance observed in pouches.¹⁹ Ultimately, our findings highlight the importance of counseling patients on postoperative risks following IPAA, specifically with regard to decreased anorectal pressures and their clinical association with FI.

Our results are consistent with a recent review of anorectal manometry (ARM) in patients with IPAA by Luo et al., where FI was characterized by decreased mean and maximal resting pressures and pouch hyposensitivity.³¹ Of the included studies, the largest sample size of IPAA patients with FI was 22 patients, all with nocturnal FI, while other studies had a range of only 2–8 IPAA patients with FI.^31–35 Only 2 studies had a comparison group to IPAA patients without FI, though data was obtained in the 1990s prior to advances in HRAM technology.^32,33 In contrast, our study is larger and includes 26 IPAA-FI patients experiencing daytime FI, nocturnal FI, or both, which more holistically encompasses the spectrum of FI symptoms in this population. Further, our comparison of ARM values between IPAA-FI patients to IPAA-no-FI and non-IBD-FI patients serves as the first and largest study to utilize different comparator groups to better characterize the association of postsurgical changes on continence in a population where manometry is not well defined.

Prior studies have found no significant difference between resting and squeeze pressures in IBD patients with fecal urgency and FI compared to those without symptoms, though none of the patients had a history of IPAA.⁸ Among IPAA patients, one study in Japan consisting of 10 patients with IPAA and FI and 15 patients with IPAA and no FI demonstrated similar findings with lower resting and squeeze pressures and higher sensory threshold volumes in those with FI, though measurements were obtained with an older ARM probe consisting of three unidirectional pressure sensors compared to the HRAM probe containing 12 circumferential pressure sensors utilized in our study for more comprehensive assessment.²⁰

Among patients with IPAA, on multivariate regression analysis, a history of redo pouch was not significantly associated with FI. This could be due in part to lower resting pressures returning to baseline as early as 6 months postoperatively and potentially improving continence.³⁶ Interestingly, although pouchitis is associated with FI, rates of acute and chronic pouchitis were similar between the IPAA-FI and IPAA-no-FI cohorts, suggesting that inflammation may not be the predominant driver for FI in this population.³⁷

Our results suggest differing mechanisms of FI between IPAA-FI and non-IBD-FI patients. Higher rates of urinary incontinence and older age in non-IBD-FI patients may indicate that age-related global pelvic floor dysfunction is a potential driver for FI in this population, consistent with prior studies, which demonstrate increased urinary, fecal, and dual incontinence with age.^38,39 These age-related changes are due to structural factors, specifically decreased anal sphincter tone, decreased compliance of the rectal vault, reduced rectal sensation, and increased perineal descent, and extrinsic factors such as immobility, diet, comorbidities, and medication side effects.^40,41

While pelvic floor dysfunction is more common in multiparous women, there were no difference in rates of pregnancy, vaginal deliveries, or obstetric surgeries between IPAA-FI and non-IBD-FI patients.⁴² While obesity is associated with increased FI, there was no difference in BMI between the IPAA-FI and non-IBD-FI cohorts.⁴³ Taken together, our findings suggest that neuromuscular changes associated with age, rather than with obesity or multiparity, likely drive urinary incontinence and FI in patients without UC or IPAA.

Comparatively, IPAA-FI patients are younger and have absent RAIR, suggesting that postsurgical changes following surgery impact continence in these patients. Though the technique for IPAA has been refined over the years, nerve injury to the anorectum can occur during surgery, potentially damaging the S2, S3, and S4 parasympathetic nerves and affecting rectal sensation.^13,40 While human perception of stool content in the anorectum is not fully understood, it is thought that sensation is most likely transmitted along the S2, S3, and S4 parasympathetic nerves.^40,44 Consistent with prior studies, RAIR was absent in the majority of IPAA patients regardless of FI, likely due to postsurgical denervation and/or sphincter damage. Other factors impacting sphincter tone, such as parity and prior obstetric/gynecologic history, were similar across the cohorts.³¹ Given that RAIR is preserved in non-IBD-FI patients despite older age in this population, denervation in patients with IPAA during surgery could lead to impaired sensation during rectal distension, thus contributing to FI, specifically with regard to nocturnal symptoms.³⁴ This is further corroborated by prior studies demonstrating that lower pressures and absent RAIR correlate with more nocturnal FI and that sacral nerve stimulation and neuromodulation are highly efficacious in treating FI and hypersensitivity in IPAA patients.^34,45–48

In clinical practice, although BET is considered a standard adjunct to HRAM for the interpretation of manometric outcomes, it may not be appropriate for patients with IPAA. Consistent with prior studies, the majority of patients in our study with IPAA could not expulse the balloon past the anastomotic ring.¹⁹ Even in patients with normal functioning pouches with no evidence of pelvic floor dysfunction on magnetic resonance (MR) defecography, nearly 40% are unable to expel the balloon in <60 seconds, suggesting that BET should be interpreted with caution in this cohort due to the fixed anastomosis.¹⁹ In addition to different normative HRAM values for patients with IPAA, dynamic imaging with MR defecography and electromyography (EMG) may be helpful to assess pouch function.^19,31 Further studies are needed to assess the multiple diagnostic modalities for patients with IPAA to identify defecatory disorders.

To our knowledge, this study is the largest HRAM analysis to date of IPAA patients with FI, a cohort where manometric data remain poorly defined. The strength of our study lies in having 2 comparator groups for the IPAA-FI cohort in order to better inform the need for separate normative values in this population. Furthermore, all IPAA patients had biopsy-proven UC, and the HRAM studies were consistently performed by 1 motility specialist with the latest manometric technology to ensure no variability in interpretation. Ultimately, data from this study demonstrate important clinical and HRAM features in IPAA patients with FI, a group where manometric studies remain limited. Understanding manometric parameters in IPAA patients may guide clinical practice, perioperative counseling, and management in this population.

Several limitations in this study exist. This was a retrospective cohort study, which allows for associations but not assessment of causal or temporal relationships. Standardized fecal incontinence scores were not able to be calculated retrospectively. Pouch volume was not measured for each patient, which could potentially affect pouch sensation measures. Data on bulking agents were not consistently collected, as these are common over-the-counter medications that do not require prescriptions. The normative values of healthy subjects were obtained with an older ARM catheter compared to the HRAM catheter utilized in our study. While newer studies have examined normative values utilizing HRAM catheters on patient populations in countries outside of the United States, these data may not be generalizable to our population. Thus, normative data from Rao et al. remain the available, standardized comparison until new normative values are established in the United States for patients with IPAA.^18,19,49,50 Furthermore, the IPAA cohorts included patients with pouchitis, which can affect FI, though notably, rates of acute and chronic pouchitis were similar among the IPAA-FI and IPAA-no-FI cohorts. Higher rates of perianal surgery in the IPAA-FI cohort could also affect sphincter tone and subsequent FI. Patient recruitment from a single urban tertiary care center restricts our findings and limits the generalizability of our results. Also, our small sample size limits study power for subgroup analysis. As anorectal pressures and volumes may vary by age and sex, updated normative data utilizing HRAM catheters, as well as a larger, prospective, multicenter study is ultimately needed to confirm our findings.

Conclusions

On anorectal manometry, UC patients with IPAA and FI had lower resting, squeeze, and cough pressures compared to IPAA patients without FI, though they were similar to non-IBD patients with FI. Female sex and defecatory disorders were independently associated with FI in UC patients with IPAA. Resting pressures for patients with IPAA appear to be similar to those of healthy controls based on available normative data, despite higher stool frequency and more liquid bowel movements, suggesting the potential need for different and/or updated normative values to accurately diagnose defecatory disorders and account for postoperative changes. Furthermore, our data suggest that IPAA-FI, IPAA-no-FI, and non-IBD-FI patients may benefit from distinct manometric ranges that underlie each of their unique characteristics. Our results can help guide perioperative counseling. Prospective studies are warranted to further examine the association of clinical and manometric variables with FI in IPAA patients and to determine the predictive value for long-term outcomes.

Author Contributions

Concept and study design: SY, SS, SC, JA; Methodology: SY, SS, SC, JA; Data acquisition: SY, BL; Formal analysis: SY; Writing original draft: SY; Article revisions: all authors; Approval of final manuscript: all authors

Funding

This work was supported by grants from the Crohn’s and Colitis Foundation, the Judith & Stewart Colton Center for Autoimmunity, and the National Institutes of Health/National Institute of Diabetes and Digestive and Kidney Diseases [grant number K23DK124570].

Conflicts of Interest

S.Y., B.L., and S.S.: none to declare. S.C. discloses consulting fees from Pfizer, AbbVie, and Bristol Myers Squibb. J.A. has received research grants from BioFire Diagnostics; consultancy, advisory board fees, or honorarium from BioFire Diagnostics, Adiso, Fresenius, Abbvie, Pfizer, Bristol Myers Squibb, and Janssen.

Data Availability

Data are not publicly available. Deidentified data can be provided upon request to the corresponding author.