Abstract
Background:
Many patients with quiescent inflammatory bowel disease (IBD) suffer from irritable bowel syndrome (IBS)-like symptoms. Although these symptoms cause significant reductions in quality of life, evidence-based treatments are lacking as risk factors and pathophysiology of these symptoms are not clearly defined. We aimed to identify risk factors for development of IBS-like symptoms in IBD patients with quiescent disease.
Methods:
We performed a single-center retrospective cohort study of adults with IBD from 2015 to 2021. Quiescent IBD was defined by a fecal calprotectin level <250 μg/g of stool or endoscopic evidence of quiescent disease. Cox regression was performed to identify variables that were independently associated with the incident development of IBS-like symptoms in IBD patients.
Key Results:
A total of 368 IBD patients were included for analysis, including 278 patients with UC and 88 with Crohn’s disease. 15.5% of quiescent IBD patients developed IBS symptoms, with an incidence rate of (95% CI 48.0–82.0) 63.3 per 1000 person-years. In the multivariate model, mood disorders (including anxiety and depression) and Crohn’s disease were associated with increased risk for developing IBS symptoms. Male sex and higher iron levels conferred lower risk for developing IBS symptoms. Results from the multivariable model were similar in sensitivity analysis with quiescent IBD defined by fecal calprotectin level <150 mcg/g.
Conclusions & Inferences:
Mood disorder and Crohn’s disease were positively associated with IBS-like symptoms in quiescent IBD, whereas male sex and iron levels were protective. Our results were robust to different fecal calprotectin levels, arguing against inflammation as a mechanism for IBS-like symptoms. This data suggests non-inflammatory mechanisms may be important in the pathogenesis of IBS-like symptoms in quiescent IBD. Future work may address whether modifying these risk factors may alter disease course.
Keywords: brain–gut axis, Crohn’s disease, IBD, IBS-like symptoms, ulcerative colitis
1 |. INTRODUCTION
Inflammatory bowel disease (IBD), including Crohn’s disease (CD) and ulcerative colitis (UC), are chronic, relapsing inflammatory conditions of the gastrointestinal tract that affect over 3 million people in the United States and their incidence is increasing worldwide.1 Irritable bowel syndrome (IBS) symptoms are prevalent in IBD patients, even in those without active inflammation. A meta-analysis reported IBS-like symptoms in up to 41% of patients with quiescent IBD.2 Furthermore, IBS-like symptoms result in a significant deterioration in quality of life3 and are an independent risk-factor for future opioid use in IBD.4,5 A study from 2022 found that about 1/3 of IBD patients in remission met Rome III criteria for IBS-like symptoms at some point in time in a 12-month period that they were followed with significant impacts on psychological health and quality of life.6 Despite the prevalence of IBS symptoms in IBD, there are currently no evidence-based therapies, likely due to lack of mechanistic understanding of this disease process.
IBS-like symptoms in quiescent IBD have previously been attributed to presence of subclinical inflammation in the gut.7 However, subclinical inflammation cannot fully explain residual symptoms as some patients have persistent symptoms even when they achieve endoscopic and histologic resmission.8,9 Accumulating evidence suggests that residual symptoms in controlled IBD may be due to varied factors including fibrosis and surgery-related GI dysmotility and reduced intestinal compliance. Much data links residual GI symptoms with the gut–brain axis, but whether this is a dominant pathway is unclear. For example, observational prospective studies have shown that baseline IBD activity was associated with a nearly 6-fold increased risk for incident anxiety. Similarly, baseline symptoms of mental disorders, including anxiety and depression, were associated with nearly 2-fold increased risk for IBD flare.10–12
Although disturbances in brain–gut interaction likely mediate incident mental disorders and disease activity in IBD, risk factors for development of persistent symptoms despite normalization of inflammation in IBD are not fully understood. The primary aim of this study was to identify baseline risk factors, including structural IBD-related and/or disturbances in brain–gut interactions, for development of IBS symptoms in patients with quiescent IBD. Secondly, we aimed to determine the cumulative incidence of IBS symptoms in quiescent IBD.
2 |. MATERIALS AND METHODS
2.1 |. Study design and population
We performed a single-center retrospective cohort study of adult patients ≥18 years old with quiescent IBD, including ulcerative colitis (UC) and Crohn’s disease (CD) between the years 2015–2021. Inclusion criteria required biochemical evidence of quiescent disease defined by a fecal calprotectin level ≤250 μg/g of stool as supported by international consensus,13 and/or endoscopic evidence of disease remission. Patients were excluded if they had history of a total colectomy, end ileostomy or colostomy, active GI symptoms at baseline, or prior diagnosis of irritable bowel syndrome.
2.2 |. Search Methods
For cohort discovery, EMERSE (electronic medical record search engine, a tool developed at the University of Michigan for electronic medical record searches) was employed to search the University of Michigan electronic medical record (MiChart) for eligible subjects with quiescent inflammatory bowel disease between the years of 2015 and 2021. Search terms including “quiescent IBD”, “quiescent crohns disease”, “quiescent UC”, “crohns in remission”, “UC in remission”, and “IBD in remission” were employed in EMERSE to identify our cohort.
Chart review was done by two investigators (MA and AP) while discrepancies were adjudicated by a third reviewer (AL). Investigators searched the chart to see if the subject met the inclusion/exclusion criteria. Clinical notes were also reviewed for evidence of clinical, biochemical, and/or endoscopic evidence of remission. Medical records, including inpatient and outpatient notes, as well as patient problem lists were reviewed to exclude previous history of irritable bowel syndrome. Patients that fit the criteria were then eligible for further data collection. The index date was defined as the initial time when the subject had evidence of quiescence by fecal calprotectin or endoscopy confirming remission. Demographic, laboratory data, medication history, and medical history were collected. The institutional review board at the University of Michigan gave ethical approval for this work.
2.3 |. Outcomes
The primary outcome was to identify predictors for developing IBS-like symptoms in quiescent IBD. IBS-like symptoms was defined either by Rome IV criteria14 (i.e., new onset abdominal pain associated with change in bowel frequency/form and/or related to defecation for at least 3 months) and/or if the patient was given a new diagnosis of IBS by their medical provider. Quiescent inflammation was defined by fecal calprotectin less than 250 μg/g of feces and/or absence of mucosal ulceration on colonoscopy. As the presence of IBD technically excludes patients from meeting Rome criteria for IBS, we therefore have defined patients with IBD that meet Rome criteria as “IBS-like.” The secondary outcome was the cumulative incidence of IBS-like symptoms in quiescent IBD in our patient population. Patient charts were reviewed through the gastroenterology clinical notes to assess for development of IBS-like symptoms. Patients were censored if they were lost to follow-up, had an acute IBD flare with elevated fecal calprotectin and/or evidence of mucosal ulceration on colonoscopy, underwent total colectomy, or if there were alternative causes for symptoms (e.g., infectious diarrhea).
2.4 |. Covariates
Covariates of interest included demographic variables, including age, sex, race, ethnicity, and body mass index (BMI). In addition, covariates related to structural/anatomic IBD-related issues included diagnosis of IBD (CD or UC), duration of IBD, extent of UC, CD location and behavior, extraintestinal manifestations, celiac disease status, prior IBD surgery, current IBD therapy, presence of small intestinal bacterial overgrowth (SIBO), and method of establishing quiescence (fecal calprotectin vs. endoscopy). Laboratory values, including C-reactive protein, white blood count, hemoglobin, platelets, AST, ALT, bilirubin, albumin, vitamin D level, vitamin B12 level, iron level, iron saturation, ferritin, and fecal calprotectin, were also extracted +/− 30 days of the index date. Finally, covariates related to potential disturbances in brain–gut interactions included tobacco use, tobacco pack-year history, alcohol use, marijuana use, opioid use, history of mood disorder, current use of psychotropic medication, antibiotic use in the last 3 months, and probiotic use within the past month.
2.5 |. Statistical analysis
Continuous variables were described as means and standard deviation (SD), while categorical and ordinal variables were summarized as absolute frequencies and proportions. T-tests and chi-square tests were employed to compare baseline differences between quiescent IBD patients with and without IBS-like symptoms. Simple and multiple Cox proportional hazard models were fit to determine baseline predictors of developing IBS-like symptoms while adjusting for potential confounders. A data-driven approach was taken to determine baseline predictors of interest. Covariates with a p-value <0.05 by unadjusted analyses were included in multivariable models. A final multivariable model was selected by forward stepwise regression method which resulted in the best goodness of fit statistics by Akaike information criteria (AIC) for non-nested or partial likelihood ratio tests for nested models. Results were presented as hazard ratios (HR) with 95% confidence intervals (CI). A two-tailed p-value of <0.05 was considered statistically significant. All statistical analyses were performed in R version 4.0.2 (R Foundation, Vienna, Austria).
3 |. RESULTS
A total of 366 patients were included for analysis, including 278 (75.5%) with ulcerative colitis and 88 (24.0%) with Crohn’s disease. The mean duration of follow-up was 29.4 months. In total, 57 (15.5%) of patients developed incident IBS-like symptoms, which included 25 (43.9%) with CD and 32 (56.1%) with UC. Of the 57 patients who developed IBS-like symptoms, 43 (75.4%) had diarrhea-predominant, 8 (14.0%) had constipation-predominant, while six had mixed-type IBS. Incidence rate of IBS-like symptoms was found to be 63.3 (95% CI 48.0–82.0) per 1000 person-years. In CD incidence was found to be 136.6 (CI 95% 88.4–201.7) per 1000 person-years, whereas the incidence in UC was 44.7 (95% CI 30.6–63.2) per 1000 person-years. Baseline characteristics are shown in Table 1.
TABLE 1.
Baseline demographics.
| No IBS-like Symptoms (N = 311) | IBS-like Symptoms (N = 57) | Total (N = 368) | p Value | |
|---|---|---|---|---|
| Age | ||||
| Mean (SD) | 43.625 (14.971) | 41.228 (14.575) | 43.251 (14.915) | 0.266 |
| BMI | ||||
| Mean (SD) | 27.899 (6.468) | 27.924 (7.603) | 27.903 (6.647) | 0.979 |
| Sex | ||||
| Female | 159 (51.5%) | 48 (84.2%) | 207 (56.6%) | <0.001 |
| Male | 150 (48.5%) | 9 (15.8%) | 159 (43.4%) | |
| Race | ||||
| Caucasian | 275 (90.5%) | 54 (94.7%) | 329 (91.1%) | 0.297 |
| Non-Caucasian | 29 (9.5%) | 3 (5.3%) | 32 (8.9%) | |
| Ethnicity | ||||
| Non-Hispanic | 296 (98.3%) | 54 (96.4%) | 350 (98.0%) | 0.344 |
| Hispanic/Latino | 5 (1.7%) | 2 (3.6%) | 7 (2.0%) | |
| Tobacco use | ||||
| Never | 215 (69.6%) | 37 (64.9%) | 252 (68.9%) | 0.484 |
| Current/Former | 94 (30.4%) | 20 (35.1%) | 114 (31.1%) | |
| Alcohol use | ||||
| Never | 108 (35.1%) | 19 (33.3%) | 127 (34.8%) | 0.030 |
| Heavy | 58 (18.8%) | 7 (12.3%) | 65 (17.8%) | |
| Light | 67 (21.8%) | 7 (12.3%) | 74 (20.3%) | |
| Moderate | 75 (24.4%) | 24 (42.1%) | 99 (27.1%) | |
| Marijuana use | ||||
| No | 275 (89.0%) | 44 (77.2%) | 319 (87.2%) | 0.014 |
| Former/Current | 34 (11.0%) | 13 (22.8%) | 47 (12.8%) | |
| Opioid use | ||||
| None | 219 (70.9%) | 34 (59.6%) | 253 (69.1%) | 0.030 |
| Former | 10 (3.2%) | 6 (10.5%) | 16 (4.4%) | |
| Current | 80 (25.9%) | 17 (29.8%) | 97 (26.5%) | |
| History of mood disorder | ||||
| No | 218 (70.8%) | 17 (29.8%) | 235 (64.4%) | <0.001 |
| Yes | 90 (29.2%) | 40 (70.2%) | 130 (35.6%) | |
| IBD diagnosis | ||||
| Ulcerative colitis | 246 (79.1%) | 32 (56.1%) | 278 (75.5%) | <0.001 |
| Crohn's disease | 63 (20.3%) | 25 (43.9%) | 88 (23.9%) | |
| Extent ulcerative colitis | ||||
| Left sided | 62 (27.3%) | 8 (26.7%) | 70 (27.2%) | 0.940 |
| Pan colitis | 165 (72.7%) | 22 (73.3%) | 187 (72.8%) | |
| CD location | ||||
| Colonic | 16 (23.2%) | 1 (4.2%) | 17 (18.3%) | <0.001 |
| Ileo-colonic | 38 (55.1%) | 8 (33.3%) | 46 (49.5%) | |
| Other | 15 (21.7%) | 15 (62.5%) | 30 (32.3%) | |
| Extraintestinal manifestations | ||||
| None | 203 (66.6%) | 30 (53.6%) | 233 (64.5%) | 0.069 |
| Other | 53 (17.4%) | 10 (17.9%) | 63 (17.5%) | |
| Rheumatologic | 49 (16.1%) | 16 (28.6%) | 65 (18.0%) | |
| Prior IBD surgery | ||||
| No | 281 (91.2%) | 44 (77.2%) | 325 (89.0%) | 0.002 |
| Yes | 27 (8.8%) | 13 (22.8%) | 40 (11.0%) | |
| Current IBD therapy | ||||
| None | 21 (6.8%) | 7 (12.3%) | 28 (7.7%) | 0.005 |
| 5-asa | 100 (32.4%) | 10 (17.5%) | 110 (30.1%) | |
| Biologic | 68 (22.0%) | 18 (31.6%) | 86 (23.5%) | |
| Combination | 66 (21.4%) | 19 (33.3%) | 85 (23.2%) | |
| Immunomodulator | 54 (17.5%) | 3 (5.3%) | 57 (15.6%) | |
| Current psychotropic medication | ||||
| No | 223 (72.2%) | 24 (42.1%) | 247 (67.5%) | <0.001 |
| Yes | 86 (27.8%) | 33 (57.9%) | 119 (32.5%) | |
| Antibiotic use (within 3months) | ||||
| No | 244 (79.2%) | 37 (64.9%) | 281 (77.0%) | 0.018 |
| Yes | 64 (20.8%) | 20 (35.1%) | 84 (23.0%) | |
| Probiotic use (within lmonth) | ||||
| No | 223 (76.9%) | 27 (58.7%) | 250 (74.4%) | 0.009 |
| Yes | 67 (23.1%) | 19 (41.3%) | 86 (25.6%) | |
| CRP | ||||
| Mean (SD) | 0.465 (0.552) | 0.689 (0.918) | 0.509 (0.645) | 0.039 |
| WBC | ||||
| Mean (SD) | 6.844 (2.464) | 7.419 (1.866) | 6.942 (2.379) | 0.131 |
| Hemoglobin | ||||
| Mean (SD) | 13.996 (1.809) | 13.340 (1.642) | 13.884 (1.796) | 0.022 |
| Platelet | ||||
| Mean (SD) | 251.406 (68.403) | 277.348 (65.933) | 255.793 (68.568) | 0.019 |
| AST | ||||
| Mean (SD) | 40.192 (105.540) | 26.510 (12.882) | 37.709 (95.746) | 0.357 |
| ALT | ||||
| Mean (SD) | 32.947 (54.610) | 27.706 (16.234) | 31.989 (49.866) | 0.498 |
| Total bilirubin | ||||
| Mean (SD) | 1.019 (4.136) | 0.600 (0.469) | 0.942 (3.745) | 0.471 |
| Albumin | ||||
| Mean (SD) | 4.422 (0.294) | 4.400 (0.276) | 4.418 (0.290) | 0.624 |
| 25-hydroxyvitamin D | ||||
| Mean (SD) | 32.326 (16.903) | 33.229 (9.993) | 32.455 (16.075) | 0.813 |
| Vitamin B12 | ||||
| Mean (SD) | 695.250 (389.883) | 606.750 (356.596) | 674.827 (380.902) | 0.486 |
| Iron | ||||
| Mean (SD) | 87.010 (41.210) | 67.778 (34.429) | 84.000 (40.696) | 0.065 |
| Iron saturation | ||||
| Mean (SD) | 25.680 (12.533) | 19.833 (10.945) | 24.765 (12.437) | 0.067 |
| Ferritin | ||||
| Mean (SD) | 83.371 (88.568) | 54.289 (47.117) | 77.616 (82.654) | 0.171 |
| Fecal calprotectin | ||||
| Mean (SD) | 47.251 (48.351) | 38.941 (33.470) | 45.368 (45.422) | 0.388 |
| Length of follow-up time (months) | ||||
| Mean (SD) | 32.095 (21.166) | 14.403 (13.728) | 29.354 (21.172) | <0.001 |
| Method to establish quiescent disease | ||||
| Fecal calprotectin only | 72 (23.2%) | 24 (42.1%) | 96 (26.1%) | 0.003 |
| Fecal calprotectin and negative colonoscopy | 239 (76.8%) | 33 (57.9%) | 272 (73.9%) | |
By unadjusted Cox proportional hazard models, we found that age; male sex (Figure 1); current use of 5-ASA or immunomodulators for IBD therapy; lower hemoglobin levels; and higher iron levels were protective against developing IBS-like symptoms (Table 2). By contrast, Crohn’s disease (Figure 2); involvement of Crohn’s disease outside of the ileum or colon; prior IBD surgery; rheumatologic extraintestinal manifestations; marijuana use current or prior; alcohol use >3 drinks a week; any mood disorder (Figure 3); use of psychotropic medication; elevated CRP level; and elevated platelets all were positively associated with the development of incident IBS-like symptoms in quiescent IBD (Table 2).
FIGURE 1.
Kaplan–Meier curve for developing IBS-like symptoms in male versus female patients with quiescent IBD. Estimated risk of developing IBS-like symptoms for female (red curve) and male patients (blue curve) with quiescent IBD. Time is presented in days.
TABLE 2.
Estimates from unadjusted and adjusted Cox proportional hazards model.
| Variable | Univariable Hazard ratio (95% CI) | p Value | Multivariable Hazard ratio (95% CI) | p Value |
|---|---|---|---|---|
| Age | 0.98 (0.96–1.00) | 0.02 | - | |
| Sex | ||||
| Female | Ref | |||
| Male | 0.22 (0.11–0.45) | <0.001 | 0.43 (0.20–0.91) | 0.03 |
| IBD type | ||||
| UC | Ref | |||
| Crohn’s disease | 2.90 (1.72–4.90) | <0.001 | 2.89 (1.68–4.96) | <0.001 |
| Crohn’s disease location | - | |||
| Colonic | Ref | |||
| Ileo-colonic | 3.33 (0.41–26.8) | 0.30 | ||
| Other | 11.8 (1.54–89.8) | 0.017 | ||
| Extra-intestinal manifestations | - | |||
| None | Ref | |||
| Rheumatologic | 1.92 (1.04–3.52) | 0.04 | ||
| Other | 1.14 (0.56–2.34) | 0.70 | ||
| Prior IBD surgery | 2.46 (1.32–4.56) | 0.004 | - | |
| Disease duration | ||||
| UC | 1.0 (0.95–1.04) | 0.90 | ||
| Crohn’s disease | 0.96 (0.91–1.02) | 0.20 | ||
| Current IBD therapy | - | |||
| None | Ref | |||
| 5-ASA | 0.28 (0.11–0.74) | 0.01 | ||
| Immunomodulator | 0.16 (0.04–0.63) | 0.009 | ||
| Biologic | 0.82 (0.34–1.96) | 0.70 | ||
| Combination therapy | 0.83 (0.35–1.97) | 0.70 | ||
| Former/Current Marijuana use | 2.18 (1.17–4.06) | 0.01 | - | |
| Former/Current Opioid use | 2.36 (1.25–4.47) | 0.008 | - | |
| Mood disorder | 4.77 (2.70–8.42) | <0.001 | 3.82 (2.14–6.82) | <0.001 |
| C-reactive protein (per 1-unit increase) | 1.49 (1.05–2.10) | 0.03 | - | |
| Hemoglobin (per 1-unit increase) | 0.79 (0.65–0.95) | 0.01 | - | |
| Platelet (per 50-unit increase) | 1.28 (1.07–1.54) | 0.008 | - | |
| Iron (per 10-unit increase) | 0.87 (0.76–0.99) | 0.04 | 0.88 (0.79–0.98) | 0.02 |
FIGURE 2.
Kaplan–Meier curve for developing IBS-like symptoms in quiescent Ulcerative colitis versus Crohn’s disease patients. Estimated risk of developing IBS-like symptoms for ulcerative colitis (red curve) compared with Crohn’s disease (blue curve). Time is presented in days.
FIGURE 3.
Kaplan–Meier curve for developing IBS-like symptoms in quiescent IBD patients with and without history of mood disorder. Estimated risk of developing IBS-like symptoms for quiescent IBD patients without history of mood disorder (red curve) compared to those with history of mood disorder (blue curve). Time is presented in days.
By multivariable Cox regression, we found that Crohn’s disease versus ulcerative colitis (HR 2.89, 95% CI 1.68–4.96) and presence of mood disorder (HR 3.82, 95% CI 2.14–6.82) significantly increased risk while male sex (HR 0.43, 95% CI 0.20–0.91) and iron levels (HR 0.88 for every 10-unit increase, 95% CI 0.79–0.98) decreased risk for IBS-like symptoms in quiescent IBD (Figure 4).
FIGURE 4.
Risk factors for development of irritable bowel syndrome (IBS)-like symptoms in quiescent inflammatory bowel disease (IBD). Results are shown as estimate and 95% confidence interval from multivariable Cox regression model. Decreased risk of developing IBS-like symptoms are depicted in blue while increased risk is shown in red.
3.1 |. Sensitivity analysis
To determine whether the fecal calprotectin level may have affected our results, we performed a sensitivity analysis where we restricted the definition of quiescent IBD to include patients with a fecal calprotectin level < = 150 mcg/g stool. A total of 123 patients met this definition for quiescent IBD, including 38 patients with CD and 84 with UC. Twenty-nine (23.6%) patients with fecal calprotectin level < = 150 mcg/g stool developed IBS-like symptoms while 94 (76.4%) remained asymptomatic (Table S1). Results from the multivariable Cox model were similar in this sensitivity analysis using a fecal calprotectin level < = 150 mcg/g stool. Male sex (HR 0.19, 95% CI 0.02–1.71) and iron level (HR 0.92, 95% CI 0.75–1.12) were protective against while presence of mood disorder (HR 3.15, 95% CI 0.99–10.03) and CD versus UC diagnosis (HR 4.71, 95% CI 1.40–15.91) were significantly associated with development of IBS-like symptoms.
3.2 |. Risk factors in UC versus CD
We next sought to determine whether risk factors were different in UC versus CD. Analyzing only patients with UC, by unadjusted Cox proportional hazards models, we found that age (HR 0.98 [95% CI 0.95–1.0], p = 0.09); and male sex (HR 0.24 [95% CI 0.1–0.6], p = 0.001) were protective against developing IBS-like symptoms in quiescent UC. By contrast, Hispanic/Latino ethnicity (HR 4.16 [95% CI 0.99–17.4], p = 0.05); rheumatologic extraintestinal manifestations (HR 2.3 [95% CI 1.0–5.2], p < 0.05); marijuana use (HR 2.1 [95% CI 0.89–4.7], p = 0.09); opioid use (HR 3.6 [95% CI 1.5–8.2], p = 0.003); history of mood disorder (HR 7.2 [95% CI 3.1–16.5], p < 0.001); use of psychotropic medications (HR 4.6 [95% CI 2.2–9.6], p < 0.001); as well as CRP (HR 1.7 [95% CI 1.1–2.5], p = 0.008); WBC (HR 1.1 [95% CI 1.0–1.2], p = 0.04); and platelet levels (HR 1.26 [95% CI 0.99–1.6] per every 50-unit increase, p = 0.06) were positively associated with development of IBS-like symptoms in quiescent UC. By multivariable Cox regression, we found that male sex (HR 0.38 [95% CI 0.15–0.93], p = 0.03) decreased risk while history of mood disorder (HR 5.2 [95% CI 2.2–12.3], p = 0.0001); and CRP level (HR 1.52 [95% CI 1.0–2.3], p = 0.04) significantly increased risk for IBS-like symptoms in quiescent UC (Figure 5A).
FIGURE 5.
Risk factors for developing IBS-like symptoms in quiescent UC versus CD. Results are shown as estimate and 95% confidence interval from multivariable Cox regression model when stratifying IBD patients by (A) UC and (B) CD. Decreased risk of developing IBS-like symptoms are depicted in blue while increased risk is shown in red.
Similarly, we evaluated risk factors specifically in CD patients. By unadjusted Cox proportional hazard models, male sex (HR 0.26 [95% CI 0.08–0.89], p = 0.03); and increasing hemoglobin level (HR 0.8 [95% CI 0.6–1.0], p = 0.08) were associated with lower risk for development of IBS-like symptoms in quiescent CD. In contrast, involvement of CD outside of the ileum or colon (HR 9.3 [95% CI 1.2–70.8], p = 0.03); marijuana use (HR 2.3 [95% CI 0.9–5.7], p = 0.08); moderate/heavy alcohol use (HR 2.0 [95% CI 0.9–4.5], p = 0.08); history of mood disorder (HR 3.2 [95% CI 1.4–7.1], p = 0.005); and increased platelet levels (HR 1.4 [95% CI 1.0–2.0], p = 0.03) were associated with increased risk for IBS-like symptoms in quiescent CD. By multivariable Cox regression, increasing iron levels (HR 0.6 [95% CI 0.5–0.8] per every 10-unit increase, p < 0.0001) decreased risk, while moderate-heavy alcohol use (HR 4.0 [95% CI 1.5–10.7], p = 0.006); and involvement of CD outside of the ileum or colon (HR 20.1 [95% CI 2.5–160.7], p = 0.005) were associated with increased risk for IBS-like symptoms in quiescent CD (Figure 5B).
4 |. DISCUSSION
Although IBS-like symptoms are prevalent in IBD patients with quiescent disease, risk factors for developing IBS-like symptoms in quiescent IBD are poorly understood. In this retrospective cohort study, we demonstrated that the incidence of IBS-like symptoms based on Rome IV criteria in patients with quiescent disease was 15.5% over an average of 29 months of follow-up. Notably, the incidence of IBS-like symptoms was much higher in CD compared with UC (28.4% vs. 11.5%). Importantly, these patients met our definition of quiescence at baseline as well as at the time of developing IBS-like symptoms. Furthermore, our results show that Crohn’s disease and presence of mood disorders increased risk of developing IBS-like symptoms in quiescent inflammatory bowel disease, while male sex and increasing iron levels were associated with decreased risk. These results were robust even with varying thresholds for fecal calprotectin levels. Finally, we found that predictors of IBS-like symptoms were different in patients with UC versus CD. Specifically, presence of mood disorders was important in UC while disease location portended greatest risk in CD.
Prior research has demonstrated IBS-like symptoms to be prevalent in the quiescent IBD population. A 2020 study in Medicine found the prevalence of IBS-like symptoms in quiescent IBD was 32% in UC patients and 35% of CD in remission.15 An updated systematic review looking at prospective studies from 2012 to 2020 found a pooled prevalence of 32.5%.16 However, incidence of IBS-like symptoms in quiescent IBD has been less well defined. One study demonstrated that IBD patients with quiescent disease may experience active symptoms in up to 54% of patients over 6 months while another study suggested 20.8% had active symptoms over mean follow-up of 935 days. While our estimate of incidence is lower than either study, this is likely due to differences in study design. Active symptoms in these prior studies included patients with active inflammation with approximately 1/3rd of the cohort in the study by Sexton et al. having evidence of inflammation17 while longitudinal assessment of inflammation was not performed in the study by Gracie et al.12 In contrast, our study design required subjects to have evidence of quiescent inflammation at baseline as well as throughout longitudinal follow-up and thus provides a more realistic estimate of incidence of IBS-like symptoms in quiescent IBD.
Our results show that anxiety and depression were associated with increased risk for incident IBS-like symptoms in quiescent IBD. These findings are consistent with prior studies, which suggest that disordered gut–brain interactions play an important role in the pathogenesis of IBS-like symptoms.6,10 Murine models have shown that induction of depression activates inflammation in the gut.18 There are also common immune-inflammatory markers found in both depression and IBD, suggesting a strong link of common mental disorders and gut inflammation.19 A recent study by Fairbrass et al. suggested an association between anxiety and depression with IBS-like symptoms in quiescent IBD patients.20 Similarly, Perera et al. found psychiatric diagnoses and earlier age onset of IBD were associated with development of IBS-like symptoms in IBD patients using Rome III criteria.21 These findings are consistent with our results showing anxiety and/or depressive disorders as a risk factor of IBS-like symptoms in IBD patients.
We also demonstrated that Crohn’s disease increased risk for developing IBS-like symptoms compared with ulcerative colitis. Prior studies have shown a higher prevalence of IBS-like symptoms in CD than in UC.2 Although the mechanism for this has not been determined, we speculate that comorbid conditions in CD, such as SIBO22 and bile acid malabsorption,23 as well as possibly fibrostenotic disease may increase risk for IBS-like symptoms in CD relative to UC. Consistent with this hypothesis, we found that involvement of CD outside of the ileum or colon carried the largest risk for development of IBS-like symptoms in quiescent CD. In contrast, presence of mood disorders was most important in quiescent UC patients. This suggests that structural/anatomic issues related to IBD may be more important in CD while disturbances in brain–gut interactions may carry greater risk for UC. Future studies should clarify specific mechanisms by which CD increases risk for IBS-like symptoms compared with UC.
We further demonstrated that female sex was a risk factor for incident IBS-like symptoms in quiescent IBD. While female sex is a known risk factor for the development of IBS in non-IBD patients,24 the increased risk of development of IBS-like symptoms in female IBD patients has not previously been described. A recent study by Janssen et al found that females with IBD in remission have an increased risk of abdominal pain compared to males; however, IBS criteria were not used.25 It is possible that the development of IBS-like symptoms in IBD patients may be similar to what is currently known for the general development of IBS, which is an epigenetic phenomenon with varied social experiences between female and male patients or sex-specific hormones altering the brain–gut axis with pain, motility, microbiota, and intestinal permeability.24,26
A novel finding in our study is the association between increased iron levels and decreased risk for IBS-like symptoms in quiescent IBD. It remains unclear if this finding suggests iron is directly involved in the development of IBS-like symptoms in IBD, or if it is secondary to altered intake in the setting of IBS-like symptoms. Recent studies have identified that many patients with IBS without IBD tend to have relative lower or deficient micronutrient values, including iron.27,28 This is important to consider when thinking about treatment options and restrictive diets, including the low FODMAP diet, to ensure adequate micronutrient intake. While we cannot exclude the possibility that elevated iron level may be a manifestation of subclinical inflammation playing a role in the development of IBS-like symptoms, our results remained robust even when we decreased our threshold for fecal calprotectin levels to 150 μg/g or even 100 μg/g. Thus, this suggests that our results are likely not driven by inflammation. However, we cannot exclude the possibility of deeper bowel wall inflammation even with lower fecal calprotectin levels and/or evidence of mucosal healing on colonoscopy.29 However, as the hazard ratio and the 95% confidence interval for iron levels approached 1, these findings should be interpreted cautiously and requires validation in an independent cohort.
Our study has several strengths, including inclusion of patients with both ulcerative colitis and Crohn’s disease. We used rigorous criteria for IBS-like symptoms which were defined by Rome IV criteria, while patients were excluded from analysis if they carried a prior diagnosis of IBS. Additionally, subjects had to meet evidence of quiescence at baseline as well as longitudinally using both biochemical and/or endoscopic criteria for quiescence. We also evaluated risk factors specific to UC versus CD.
Despite these strengths, there were potential limitations to our study. First, this was a retrospective study using data collected as part of routine clinical care, which may introduce bias. Although patients with prior diagnosis of IBS were specifically excluded from analysis, we cannot exclude the possibility that some patients may have previously been diagnosed with IBS. In addition, Rome IV criteria were not designed to be applied in a retrospective manner and may be affected by variations in documentation across providers. Secondly, this was a single center study with only 57 patients developing IBS-like symptoms, making the outcome relatively uncommon in our cohort. In addition, our CD cohort is small compared to our UC cohort, which limits interpretation of CD results. Thirdly, as quiescent IBD was defined by FCP, we cannot exclude the possibility that our results may have been influenced by microscopic inflammation. Additional studies are required to validate these findings in larger, independent cohorts.
5 |. CONCLUSION
We found that mood disorders and Crohn’s disease were associated with increased risk of developing IBS-like symptoms in quiescent inflammatory bowel disease, while male sex and iron levels were associated with decreased risk. This data suggests presence of altered brain–gut interactions may be important in the pathogenesis of IBS-like symptoms in quiescent IBD, and there may also be a role of subclinical inflammation. Future work may examine whether addressing mood disorder may alter disease course.
Supplementary Material
Key points.
What is already known?: IBS-like symptoms are prevalent in quiescent IBD and are associated with poor outcomes and poor quality of life.
What is new here?: Risk factors leading to the development of these symptoms and incidence of these symptoms are unknown.
How can this study help patient care?: It helps by focusing future research efforts. By identifying baseline risk factors in this patient population, structured investigations can take place to determine if modification of these risk factors and awareness of patients at risk of development of IBS-like symptoms improves outcomes.
FUNDING INFORMATION
This study was supported by grants from the National Institutes of Health grants DK124567 (to AAL), HS027431 (to KR), and DK123403 (to SB).
Footnotes
CONFLICT OF INTEREST STATEMENT
The authors declare no conflict of interest exist.
SUPPORTING INFORMATION
Additional supporting information can be found online in the Supporting Information section at the end of this article.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.
REFERENCES
- 1.Molodecky NA, Soon IS, Rabi DM, et al. Increasing incidence and prevalence of the inflammatory bowel diseases with time, based on systematic review. Gastroenterology. 2012;142:46–54.e42. [DOI] [PubMed] [Google Scholar]
- 2.Halpin SJ, Ford AC. Prevalence of symptoms meeting criteria for irritable bowel syndrome in inflammatory bowel disease: systematic review and meta-analysis. Am J Gastroenterol. 2012;107:1474–1482. [DOI] [PubMed] [Google Scholar]
- 3.Gracie DJ, Williams CJM, Sood R, et al. Negative effects on psychological health and quality of life of genuine irritable bowel syndrome–type symptoms in patients with inflammatory bowel disease. Clin Gastroenterol Hepatol. 2017;15:376–384.e5. [DOI] [PubMed] [Google Scholar]
- 4.Long MD, Barnes EL, Herfarth HH, Drossman DA. Narcotic use for inflammatory bowel disease and risk factors during hospitalization. Inflamm Bowel Dis. 2012;18:869–876. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Anderson A, Click B, Ramos-Rivers C, et al. The association between sustained poor quality of life and future opioid use in inflammatory bowel disease. Inflamm Bowel Dis. 2018;24:1380–1388. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Barberio B, Fairbrass KM, Gracie DJ, Ford AC. Natural history and impact of irritable bowel syndrome-type symptoms in inflammatory bowel disease during 12 months of longitudinal follow-up. Neurogastroenterol Motil. 2024;36:e14713. [DOI] [PubMed] [Google Scholar]
- 7.Keohane J, O’Mahony C, O’Mahony L, O’Mahony S, Quigley EM, Shanahan F. Irritable bowel syndrome-type symptoms in patients with inflammatory bowel disease: a real association or reflection of occult inflammation? Am J Gastroenterol. 1788;2010(105):1789–1794. [DOI] [PubMed] [Google Scholar]
- 8.Gracie DJ, Williams CJM, Sood R, et al. Poor correlation between clinical disease activity and mucosal inflammation, and the role of psychological comorbidity, in inflammatory bowel disease. Am J Gastroenterol. 2016;111:541–551. [DOI] [PubMed] [Google Scholar]
- 9.Henriksen M, Høivik ML, Jelsness-Jørgensen L-P, Moum B, IBSEN Study Group. Irritable bowel-like symptoms in ulcerative colitis are as common in patients in deep remission as in inflammation: results from a population-based study [the IBSEN study]. J Crohns Colitis. 2018;12:389–393. [DOI] [PubMed] [Google Scholar]
- 10.Fairbrass KM, Lovatt J, Barberio B, Yuan Y, Gracie DJ, Ford AC. Bidirectional brain–gut axis effects influence mood and prognosis in IBD: a systematic review and meta-analysis. Gut. 2022;71:1773–1780. [DOI] [PubMed] [Google Scholar]
- 11.Fairbrass KM, Gracie DJ, Ford AC. Relative contribution of disease activity and psychological health to prognosis of inflammatory bowel disease during 6.5 years of longitudinal follow-up. Gastroenterology. 2022;163:190–203.e5. [DOI] [PubMed] [Google Scholar]
- 12.Gracie DJ, Guthrie EA, Hamlin PJ, Ford AC. Bi-directionality of brain–gut interactions in patients with inflammatory bowel disease. Gastroenterology. 2018;154:1635–1646.e3. [DOI] [PubMed] [Google Scholar]
- 13.Rogler G, Aldeguer X, Kruis W, et al. Concept for a rapid point-of-care calprotectin diagnostic test for diagnosis and disease activity monitoring in patients with inflammatory bowel disease: expert clinical opinion. J Crohns Colitis. 2013;7:670–677. [DOI] [PubMed] [Google Scholar]
- 14.Lacy BE, Mearin F, Chang L, et al. Bowel Disorders. Gastroenterology. 2016;150:1393–1407.e5. [DOI] [PubMed] [Google Scholar]
- 15.Ozer M, Bengi G, Colak R, Cengiz O, Akpinar H. Prevalence of irritable bowel syndrome-like symptoms using Rome IV criteria in patients with inactive inflammatory bowel disease and relation with quality of life. Medicine. 2020;99:e20067. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Fairbrass KM, Costantino SJ, Gracie DJ, Ford AC. Prevalence of irritable bowel syndrome-type symptoms in patients with inflammatory bowel disease in remission: a systematic review and meta-analysis. Lancet Gastroenterol Hepatol. 2020;5:1053–1062. [DOI] [PubMed] [Google Scholar]
- 17.Sexton KA, Walker JR, Graff LA, et al. Evidence of bidirectional associations between perceived stress and symptom activity: a prospective longitudinal investigation in inflammatory bowel disease. Inflamm Bowel Dis. 2017;23:473–483. [DOI] [PubMed] [Google Scholar]
- 18.Ghia J, Blennerhassett P, Deng Y, et al. Reactivation of inflammatory bowel disease in a mouse model of depression. Gastroenterology. 2009;136:2280–2288.e4. [DOI] [PubMed] [Google Scholar]
- 19.Martin-Subero M, Anderson G, Kanchanatawan B, Berk M, Maes M. Comorbidity between depression and inflammatory bowel disease explained by immune-inflammatory, oxidative, and nitrosative stress; tryptophan catabolite; and gut–brain pathways. CNS Spectr. 2016;21:184–198. [DOI] [PubMed] [Google Scholar]
- 20.Fairbrass KM, Hamlin PJ, Gracie DJ, Ford AC. Natural history and impact of irritable bowel syndrome-type symptoms in inflammatory bowel disease during 6 years of longitudinal follow-up. Aliment Pharmacol Ther. 2022;56:1264–1273. [DOI] [PubMed] [Google Scholar]
- 21.Perera LP, Radigan M, Guilday C, et al. Presence of irritable bowel syndrome symptoms in quiescent inflammatory bowel disease is associated with high rate of anxiety and depression. Dig Dis Sci. 2019;64:1923–1928. [DOI] [PubMed] [Google Scholar]
- 22.Shah A, Morrison M, Burger D, et al. Systematic review with meta-analysis: the prevalence of small intestinal bacterial overgrowth in inflammatory bowel disease. Aliment Pharmacol Ther. 2019;49:624–635. [DOI] [PubMed] [Google Scholar]
- 23.Vijayvargiya P, Gonzalez Izundegui D, Calderon G, et al. Increased fecal bile acid excretion in a significant subset of patients with other inflammatory diarrheal diseases. Dig Dis Sci. 2022;67:2413–2419. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Kim YS, Kim N. Sex-gender differences in irritable bowel syndrome. J Neurogastroenterol Motil. 2018;24:544–558. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Janssen LM, Rezazadeh Ardabili A, Romberg-Camps MJL, et al. Abdominal pain in patients with inflammatory bowel disease in remission: a prospective study on contributing factors. Aliment Pharmacol Ther. 2023;58:1041–1051. [DOI] [PubMed] [Google Scholar]
- 26.Mulak A, Taché Y. Sex difference in irritable bowel syndrome: do gonadal hormones play a role? Gastroenterol Pol. 2010;17:89–97. [PMC free article] [PubMed] [Google Scholar]
- 27.Bek S, Teo YN, Tan X-H, Fan KHR, Siah KTH. Association between irritable bowel syndrome and micronutrients: a systematic review. J Gastroenterol Hepatol. 2022;37:1485–1497. [DOI] [PubMed] [Google Scholar]
- 28.Roth B, Larsson E, Ohlsson B. Poor intake of vitamins and minerals is associated with symptoms among patients with irritable bowel syndrome. J Gastroenterol Hepatol. 2022;37:1253–1262. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Laharie D, D’Haens G, Nachury M, et al. Steroid-free deep remission at one year does not prevent Crohn’s disease progression: Long-term data from the TAILORIX trial. Clin Gastroenterol Hepatol. 2022;20:2074–2082. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.