Faecal incontinence with concurrent disorders of gut‐brain interaction: A worse outcome

Abstract

Background

Faecal incontinence is a common debilitating condition associated with poor quality of life that generates substantial economic strain on healthcare systems.

Objectives

We aimed to evaluate, in a tertiary referral population presenting with faecal incontinence, the impact of suffering additional disorders of gut‐brain interaction (DGBI) on symptom severity, anxiety, depression and quality of life.

Methods

Design: Retrospective cohort study. Setting: Tertiary referral Neurogastroenterology centre. Patients: All patients presenting with faecal incontinence from 2007 to 2020 were included. Main Outcome Measures: The results from structured medical and surgical questionnaires including Rome III Integrative Questionnaire, Faecal Incontinence Severity Index, Hospital Anxiety and Depression Scale, SF‐36, and anorectal physiology were analysed using Stata version 17. Patients were categorised into 3 groups: 0–1 additional DGBI, 2 DGBIs, and 3+ DGBI. Statistical significance was defined as p < 0.05 (two‐tailed).

Key Results

Faecal incontinence patients (n = 249; mean age 63.4 ± 12.6 years; 93.6% female, 48.1% urge subtype) met diagnostic criteria for mean 2.2 additional DGBI each, mostly affecting bowel (n = 231, 42.4%) and anorectal (n = 150, 27.5%) regions. A greater number of DGBIs was associated with higher faecal incontinence symptom severity (p < 0.001), higher anxiety (p = 0.002) and depression (p = 0.003), and worse quality of life in areas of mental health (p = 0.037) and social effect (p < 0.001). Patients with a greater number of concurrent DGBI demonstrated a greater family history of gastrointestinal problems (p = 0.004). There were no associations found between a greater amount of DGBIs and anorectal physiology.

Conclusions and Inferences

A greater number of additional DGBIs in faecal incontinence patients was associated with worse faecal incontinence symptoms, higher anxiety and depression scores, and worse quality of life but was unrelated to physiology. This highlights the need to proactively search for comorbid DGBI in patients presenting with faecal incontinence.

Key summary.

Summarise the established knowledge on this subject?

• In community studies, a greater number of concurrent DGBIs has been associated with increased somatisation, greater healthcare utilisation and reduced quality of life.

• Whether these findings hold true for a tertiary referral DGBI population, especially one with a well‐defined physiological basis such as faecal incontinence, is unclear.

What are the significant and/or new findings of this study?

• For a tertiary referral population of faecal incontinence, a greater number of additional DGBIs is associated with higher faecal incontinence symptom severity, worse anxiety and depression and poorer quality of life.

• A greater number of additional DGBI in faecal incontinence patients is not, however, associated with more abnormal anorectal physiology

INTRODUCTION

According to Rome IV publications, functional gastrointestinal disorders encompass an extensive range of conditions of the GI tract and are characterised by a dysregulation of the gut‐brain axis in the absence of organic or structural disease. ¹ Beyond Rome IV these disorders are now referred to as disorders of gut‐brain interaction (DGBI). DGBI are common and impact over 40% of the wider community, disproportionally more in women ² , ³ and there is a higher prevalence in older age groups. ² In community‐based population studies, it has been shown that suffering a higher number of concurrent DGBI is associated with worse outcomes such as increased somatisation of symptoms, greater utilisation of healthcare and reduced quality of life. ⁴ , ⁵ Whether this holds true (or is exaggerated) for patients presenting to doctors with a DGBI, such as faecal incontinence, remains unclear.

Faecal incontinence (FI) is defined as the loss of voluntary control of stool (either solid or liquid) for three or more months in an individual over 4 years of age by current Rome IV diagnostic criteria. ¹ Faecal incontinence is associated with significant impairment in quality of life and leads to considerable cost to society. For example, the mean cost burden of treating faecal incontinence surpasses $3000USD per patient per year, and the cost rises proportional to symptom severity. ⁶ There is increasing evidence recognising FI as highly prevalent in irritable bowel syndrome (IBS) patients, and this concurrent diagnosis being a prognosticator of IBS symptom severity. ⁷ Whether or not patients presenting specifically with faecal incontinence also suffer additional concurrent DGBI and to what extent, and whether they too demonstrate worse outcomes such as poorer quality of life or worse psychological health with a higher number of overlapping DGBI remains unstudied. Faecal incontinence is a unique DGBI as there are clear physiological abnormalities such as anal sphincter hypotension or rectal hyposensitivity. ⁸ Whether having additional DGBI relates to higher symptom severity or more abnormal physiology in faecal incontinence also remains unstudied.

Therefore, our aims were to report the prevalence and nature of DGBI overlap in patients presenting specifically with faecal incontinence and to evaluate whether an increasing number of DGBI is associated with a change in symptom severity, mental health, quality of life, or altered anorectal physiology.

MATERIALS AND METHODS

282 patients (mean age 52 years, SD ± 13.1 yrs; 92.9% female) referred by a colorectal surgeon or gastroenterologist with faecal incontinence to a tertiary referral Neurogastroenterology clinic between 2007 and 2020 and who underwent anorectal biofeedback therapy were included in the study. Exclusion criteria included patients with known spinal cord injury and previous pelvic radiation therapy. Ethics approval was granted by the Human Research Ethics Committee at Royal North Shore Hospital on 15 September 2020 (2019/STE10108). All participants provided written, informed consent for their data to be retrospectively studied. Data collection was performed in compliance with the National Health and Medical Research Council National Statement on Ethical Conduct in Human Research (2007) and the ethical guidelines of the 1975 Declaration of Helsinki.

Questionnaires and clinical evaluation

A clinical evaluation and a series of questionnaires were administered prospectively. These included a structured questionnaire for medical and surgical history and medication use, a 1 week stool diary, the Rome III Integrative Questionnaire, ⁹ Hospital Anxiety and Depression Scale (HAD), ¹⁰ modified SF‐36 questionnaire ¹¹ and Faecal Incontinence Severity Index. ¹² The patient reported measures included global bowel satisfaction, impact of bowel dysfunction on quality of life and feeling of control over bowel movements, each of which was assessed on 10‐point visual analogue scales (VAS).

Anorectal physiology testing

Each patient completed anorectal manometry, sensory testing and the balloon expulsion test using a high resolution a 7‐lumen water‐perfused manometry catheter with 5 mm spaced side holes radially arranged (Dentsleeve International, Canada). The following physiological parameters were recorded largely in accordance with the London protocol ¹³ : anal sphincter pressures: resting anal sphincter pressure, maximum anal sphincter squeeze pressure, cough pressure, squeeze duration; defecation manoeuvre: rectal pressure on push, anal relaxation and absent or paradoxical anal contraction on strain, duration for successful expulsion of a balloon filled with 50 mL of warm water from the rectum, where the maximum time limit was 180s; rectal perception: first, urge and maximum‐tolerated threshold.

Statistical analysis

Statistical analyses were performed using Stata version 17. ¹⁴ Quantitative measures were described using mean and standard deviation (SD) and compared between DGBI burden groups via the Kruskal‐Wallis test due to the non‐normal distribution of some measures. Qualitative measures were described using count and percentage and compared between DGBI burden groups via the Pearson chi – square test. Statistical significance was defined as p < 0.05 (two‐tailed) and was not adjusted for multiple inference. We therefore regard these findings as being in need of replication.

RESULTS

Participant details

282 consecutive patients presenting for assessment and management of FI were evaluated. Of these, 33 (11.7%) were excluded, 20 due to exclusion criteria, and 13 had incomplete Rome III questionnaire data, leaving 249 patients. Additionally, there were 32 patients with incomplete data (≥4 data points missing) for the secondary SF‐36 and HAD Questionnaires. The decision was made to include these patients as the missing data would not significantly impact the findings, leaving 249 subjects for analysis.

The mean age of the total study sample was 63.4 years (SD 12.6), and 93.6% of the population were female. 34.0% of participants reported being employed, with 34.4% retirees, 9.8% unemployed, and 16.7% full‐time homemakers. Patients were divided into subtypes of urge (48.1%), non‐urge (36.1%) and both urge and non‐urge (15.9%) by responses in the FI questionnaire. The mean length of time patients suffered symptoms was 6.1 years (SD 7.5), with a mean FISI score of 24.5 (SD 11.3), with the maximum score of 61 indicating the highest perceived severity of FI symptoms. 44.6% reported having a family history of bowel or stomach problems. Detailed demographics and clinical characteristics are outlined in Table 1.

TABLE 1.

Total population: demographics and clinical characteristics (n = 249).

	Mean (SD)	Number of patients (%)
Age (years)	63.4 (12.6)
Gender
Female		233 (93.6%)
Male		16 (6.4%)
Symptom duration (years)	6.1 (7.5)
FISI score/61	24.5 (11.3)
FI subtype (n = 233)
Urge		112 (48.1%)
Non‐urge/passive		84 (36.1%)
Both urge and non‐urge		37 (15.9%)
Employment (n = 215)
Employed		73 (34.0%)
Retiree		74 (34.4%)
Unemployed		21 (9.8%)
Full‐time homemaker		36 (16.7%)
Other		11 (5.1%)
Family history of bowel/stomach problems (n = 239)
Yes		111 (44.6%)
No		128 (51.4%)
HAD scores
Anxiety	5.81 (3.7)
Depression	3.36 (3.1)
Home location from Neurogastro clinic
Within 5 km radius		104 (41.8%)
More than 5 km radius		145 (58.2%)

Note: Where noted, total numbers per section vary due to small amounts of missing data.

Abbreviations: FI, faecal incontinence; HAD, Hospital Anxiety and Depression Scale.

Prevalence and overlap of DGBI

In the total population (n = 249) there were 545 DGBIs by Rome III Criteria in addition to FI, with the mean number of additional DGBI being 2.2 (SD 1.57) per patient. Only 3 patients were reported to have FI with no concurrent DGBI. The most common concurrent diagnosis was functional defecation disorder (35.6%), followed by irritable bowel syndrome (27.1%) and then functional constipation (20.7%). Classifying these DGBIs by region affected, the most prevalent functional disorders involved the bowel with 42.4% (n = 231), followed by anorectal disorders (n = 150, 27.6%) (Table 2). There were no functional gallbladder or Sphincter of Oddi disorders recorded in the population.

TABLE 2.

Breakdown of diagnosable DGBI by anatomical region within Rome III classification.

Functional gastrointestinal disorders by anatomical region by Rome III classification	Number	% total
Anorectal	150	27.5
Chronic proctalgia	13	2.4
Proctalgia fugax	49	9.0
Functional defecation disorders	88	16.1
Bowel	231	42.4
Irritable bowel syndrome	66	12.1
Functional bloating	37	6.8
Functional constipation	51	9.4
Functional diarrhoea	30	5.5
Unspecified functional bowel disorder	47	8.6
Gastroduodenal	97	17.8
Postprandial distress syndrome	16	2.9
Epigastric pain syndrome	0	0.0
Rumination syndrome	5	0.9
Chronic idiopathic nausea	6	1.1
Functional vomiting	1	0.2
Belching disorders	29	5.3
Cyclic vomiting syndrome	8	1.5
Functional dyspepsia	32	5.9
Oesophageal	67	12.3
Globus	10	1.8
Functional heartburn	29	5.3
Functional chest pain of presumed oesophageal origin	12	2.2
Functional dysphagia	16	2.9
Total number of diagnoses	545

Abbreviation: DGBI, disorders of gut‐brain interaction.

Correlation of family history and number of DGBI

Those patients with a higher number of additional DGBI were noted to have a higher prevalence of a family history of bowel or stomach problems documented as 27.0% of the 0–1 DGBI group versus 55.1% of the 3+ DGBI group (p = 0.004), n = 239 (see Figure 1).

FIGURE 1.

Family history of stomach or bowel problems in patients with faecal incontinence and an increasing number of additional DGBI. DGBI, disorders of gut‐brain interaction.

Correlation of symptom severity and number of DGBI

An increase in concurrent DGBI was associated with worse symptom severity as evidenced by a higher total FISI score (p < 0.002) (Figure 2). In addition, an increase in the FISI subcategory score of mucus stool incontinence was associated with an increase in concurrent DGBI (p = 0.02). Detailed components of the FISI subscores are shown in supplementary Table S1.

FIGURE 2.

Symptom severity in patients with faecal incontinence and an increasing number of additional DGBI. DGBI, disorders of gut‐brain interaction.

Correlation of anxiety, depression and quality of life and number of DGBI

Patients who reported a greater number of concurrent DGBI reported higher depression (p = 0.003) and anxiety (p = 0.002) scores as measured by the HAD questionnaire (Figure 3).

FIGURE 3.

Anxiety and depression in patients with faecal incontinence and increasing number of additional DGBI. DGBI, disorders of gut‐brain interaction.

Comparing SF‐36 subsection scores, faecal incontinence patients with a greater number of concurrent DGBI indicated a worse quality of life (QOL) in the fields of social impact of disease (p = 0.0009) and mental health (p = 0.04) (supplementary Table S2). There were no other significant findings with regard to other QOL measures.

Bowel specific QOL measures were undertaken using a 10‐point visual analog scale (VAS). Compared to patients with 0–2 additional DGBI, patients with 3+ DGBI reported a decrease in the feeling of control over bowel movements (p = 0.03), greater impact of bowel dysfunction on QOL (p = 0.008) and decreased satisfaction with bowel movements (p = 0.043).

Correlation of anorectal physiology and number of DGBI

The number of DGBIs was not correlated with any parameters of anorectal physiology, including resting and squeeze anal sphincter pressures, the duration of squeeze, first and urge sensation threshold, maximum tolerated‐volume threshold and mean time to balloon expulsion (p = NS for all values), shown in supplementary Table S3.

Correlation between the above outcome measures and anorectal physiology

By way of contrast, there was no correlation between the outcome measures of total FISI score, QOL, patient satisfaction and anorectal physiology (resting and squeeze anal sphincter pressures, duration of squeeze, rectal pressure, rectal sensation or balloon expulsion time) (supplementary Table S4). This is in comparison to the significant correlation between an increase in concurrent DGBI with worse outcomes detailed above.

DISCUSSION

This study has highlighted the significance of having multiple overlapping DGBI in patients presenting with faecal incontinence. In particular, we show that having concurrent DGBI is highly prevalent (mean of 2.2 extra DGBI) and that having a greater number of DGBI is associated with greater FI symptom severity, higher levels of anxiety and depression and a worse QOL. Our study has echoed the findings of Sperber et al ² and Aziz et al, ⁴ who studied multiple concurrent DGBI in a community‐based population; however, many of the findings are exaggerated in FI patients.

Faecal incontinence is a unique disorder of gut‐brain interaction with somewhat well‐defined disruption of peripheral motor and sensory physiology. In addition, there is a strong correlation of FI with stool form. In the current study, we had the opportunity to correlate additional DGBI with physiology, not possible in questionnaire studies such as those by Sperber and Aziz. ² , ⁴ However, a greater number of DGBIs was not associated with worse anorectal physiology. Furthermore, increased severity of physiological abnormalities was not associated with worse outcomes. This disconnect between physiology and disease burden adds weight to the concept of FI being more a disease of gut brain interaction than primarily a disorder of disrupted anorectal physiology.

It is no surprise that a chronic condition like FI may erode the psychosocial wellbeing of patients, and it has been extensively demonstrated that patients with DGBI report an inferior QOL compared with healthy subjects. ⁸ , ¹⁵ FI patients have also been shown to be four times as likely to be diagnosed with anxiety and depression, and the majority report a sense of frustration and humiliation as a by‐product of their symptoms. ¹⁶ Previous community‐based studies have identified weak correlations between increasing DGBI and increasing anxiety and depression; ⁵ however, these findings appear magnified in our FI disease population. Undisputedly, individual anxiety and depression scores increased in a step‐wise fashion (indicating a worse outcome) with increasing number of DGBI in our cohort.

Our findings of significant impact on mental health with increasing number of DGBIs are of great clinical relevance, exemplified by the evidence of increased suicidal ideation in FI patients. ¹⁷ Anxiety in FI patients may manifest as a maladaptive locus of control, and that catastrophising and anticipatory concerns may perpetuate symptoms. ¹⁸ , ¹⁹ , ²⁰ Psychological intervention has been shown to be effective in spinal cord injury with FI ²¹ and counselling support has been shown to improve symptoms in urinary incontinence patients. ¹⁶ These parallel findings suggest that FI patients with multiple DGBI could also potentially reduce their symptom burden and quality of life with early psychosocial intervention.

By way of contrast, anorectal physiology including manometry, rectal sensation test and the balloon expulsion test, was not different in patients with more DGBI compared with those with fewer DGBI. In addition, a greater number of DGBIs correlated more significantly with symptoms and quality of life rather than any association between physiology (such as anal squeeze pressure). This is a crucial piece of the puzzle in ascertaining the relative contribution of the “end organ” such as the anorectum with abnormal sensory or motor function compared to the brain or abnormal brain‐gut axis interactions with regard to symptom development in DGBI. Put another way, higher symptom severity in faecal incontinence may be due to a greater derangement of the brain‐gut axis rather than the end organ, even though FI is a disorder with well‐defined peripheral motor and sensory physiology disruption.

In our study, we noted an intriguing association of family history with a more severe clinical presentation of FI in our cohort. Previous studies from Locke et al established what was described as a “familial aggregation” ²² of functional symptoms in IBS and dyspepsia candidates. However, they were yet to determine whether the contributions were more‐so genetic or environmental. Increased health‐seeking behaviour in adulthood can be a product of learned family influence in childhood rather than through inheritance of pathology. ²³ Thus, it is important to acknowledge that our cohort's observable “family aggregation” may also be multifactorial in nature, with either genetic or early life environmental factors predominating and our findings highlight a promising area for future research. Locke et al's paper also queried the validity of analysing the Rome question of “family [history] of bowel or stomach problems”, ²² which was used as a data point for our study (Q15 in Rome III). Locke and coauthors suggested that as bowel symptoms are frequent and can be conflated in laymen reporting, the Rome questionnaire would be more useful in establishing a greater family linkage of FI if it focused on a history of diagnosed bowel problems as opposed to the experience of symptoms. ²² Indeed, this was a reflected in a change in the Rome Questionnaire for family history in the Rome IV version of the questionnaire (Q93).

In this study, we evaluated all patients referred with the clinical problem of FI diagnosed by gastroenterologists or colorectal surgeons. We chose to include patients who did not meet strict Rome IV criteria (which stipulates at least 2 episodes each 4 weeks). This may be a limitation to our study; however, we consider it to provide a thorough insight into the clinical entity of faecal incontinence, especially to capture the low frequency, high impact FI, an entity increasingly realised as important to include in further iterations of Rome diagnostic criteria. ²⁰ In addition, examining the relationship between stool form and faecal incontinence, which was not within our original aims and hypotheses, is an important pathophysiology of faecal incontinence worthy of closer examination. A further limitation is that we excluded spinal cord injury and pelvic radiotherapy in an attempt to study predominant DGBI patients, and that our patient group were undergoing biofeedback, limiting the generalisability of our findings. Finally, as noted in the statistical methods section, we did not adjust our definition of statistical significance for the number of hypothesis tests conducted. This means that the rejection of statistically null hypotheses in this study should be considered indicative rather than definitive, and needs replication. We do note however that other statistical elements, such as effect size measures, are unaffected by multiple hypothesis testing and can be interpreted at face value.

FI remains a common and debilitating condition with multiple likely contributing pathophysiologies, both in the area of the gut, the brain and the brain‐gut axis as well as the contribution of stool form. We found that a greater number of additional DGBIs was associated with higher severity of FI symptoms, higher degree of anxiety and depression, and worse quality of life; however, it was unrelated to anal sphincter dysfunction. This novel finding highlights that the pathophysiology of FI, particularly those with high symptom severity and high health care utilisation, may include a dominant contribution from dysfunction of the brain‐gut axis rather than the historical view that faecal incontinence is primarily due to a weak anal sphincter. Clinicians treating FI should also be aware of these associations, ask for the presence of this symptom in all patients with DGBI, and incorporate multimodality therapy (including psychological treatments) for optimum therapeutic outcomes.

CONFLICT OF INTEREST STATEMENT

The authors have no conflicts of interest to declare.

Supporting information

Supporting Information S1

Li SG, Mazor Y, Park CJ, Jones MP, Malcolm A. Faecal incontinence with concurrent disorders of gut‐brain interaction: a worse outcome. United European Gastroenterol J. 2024;12(4):496–503. 10.1002/ueg2.12541

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.