Prospective cohort study of high‐volume transanal irrigation in patients with constipation and/or faecal incontinence

The aim of this work was to determine the clinical efficacy of high‐volume transanal irrigation (TAI) in patients with constipation and/or faecal incontinence using validated symptom and quality of life questionnaires.


INTRODUC TI ON
Treatment of faecal incontinence and chronic constipation with transanal irrigation (TAI) is gaining popularity, and an increasing number of studies have been published during the past two decades reporting success rates in both adults and children [1,2]. A systematic review and meta-analysis conducted in 2010 reported excellent success rates in children [1]. In adults, TAI was shown to be successful in 45% of patients with chronic constipation, in 47% of patients with faecal incontinence and in 59% of patients with coexistent symptoms (data based on 1229 patients in 17 studies) [1].
Similar results have been reported in a recent systematic review in adults [2]. Further, a large study in 348 patients with constipation and/or faecal incontinence showed successful treatment in 47% at a mean follow-up of 21 months [3]. This observation was confirmed by a study in 507 patients, of whom 216 (43%) still used TAI at a followup of 12 months [4].
A number of clinical trials have been performed in specific patient groups. A randomized controlled trial compared TAI with conservative management in 87 patients with spinal cord injury [5]. Improvement of constipation, faecal incontinence and related quality of life (QoL) was demonstrated. Three clinical trials showed that TAI improves symptoms and QoL in patients with low anterior resection syndrome [6][7][8]. A recent report of the Chronic Constipation Treatment Pathway (CapaCiTY) research programme showed a reduction in the Patient Assessment of Constipation Quality of Life score (PAC-QoL) at 3 months, which was larger in the group of patients using high-volume irrigation (which was preferred over lowvolume irrigation by the majority of patients) [9].
In contrast to stronger evidence in these specific patient groups, most other studies to date had methodological limitations, reporting cross-sectional or retrospectively collected data over a short follow-up period (several weeks to months) or reporting on small patient groups. Further, validated scoring systems to report on QoL were not routinely used, which makes it hard to interpret and compare the results of different studies. There is a need for prospective studies using validated symptom severity and QoL measures to validate the success rate. In this prospective cohort study, the primary objective was to determine the effect of high-volume TAI on reduction of symptoms of constipation and/or faecal incontinence in those who continued TAI at 52 weeks' follow-up. Secondary objectives included: (1) continuation rates of TAI, (2) the effect of TAI on QoL and (3) evaluation of predictors for continuation.

Study population
We prospectively recruited 114 consecutive adult patients (≥18 years old) naive to TAI with either the Navina™ Classic or Smart system,  [10] and/or the Rome IV criteria for faecal incontinence (episodes to solid or liquid stools occurring more than monthly) [11]. All participating patients underwent treatment with TAI according to routine care. Patients were allowed to decide to use either the Classic (manual pump) or Smart (electronic pump) system. During the first face-to-face visit, all patients were instructed by a conservative management nurse to start TAI daily with 500-1000 mL of water. The frequency of TAI and the volume of water were adjusted during follow-up (frequent telephone clinics).

Study questionnaire
Data were collected using a comprehensive questionnaire completed at five different time points: baseline (just before starting the first irrigation at a face-to-face visit with the conservative management nurse) and at 4, 12, 26 and 52 weeks' follow-up. The baseline questionnaire incorporated questions on demographics, obstetric, surgical and medical history and previous treatment. Validated symptom severity and QoL scores were completed at both baseline and during follow-up, including the Rome IV criteria for functional constipation [10] and faecal incontinence [11], the Cleveland Clinic Constipation Score (CCCS; 0-30; a higher score indicates more severe symptoms) [12], the St Marks Incontinence Score (SMIS; 0-24; a higher score indicates more severe symptoms) [13], PAC-QoL (1-4; a higher score indicates worse QoL) [14], the Fecal Incontinence Quality of Life Scale (FI-QoL; 1-4; a higher

What does this paper add to the literature?
This is one of the first cohort studies in which the clinical efficacy of high-volume transanal irrigation in patients with constipation and/or faecal incontinence was evaluated using validated symptom and quality of life questionnaires. Further, we provide detailed results on irrigation parameters, which can be helpful for therapists to guide patients. score indicates better QoL) score [15] and the Bristol stool form scale [16]. QoL questionnaires (PAC-QoL and FI-QoL) were not included in the questionnaire at 12 weeks' follow-up. Completeness of evacuation after irrigation [visual analogue scale (VAS) score], treatment satisfaction (VAS score) and side effects were also recorded at each follow-up appointment. In case of discontinuation of TAI, patients were asked to provide the main reason for discontinuation. At follow-up, irrigation parameters [frequency, volume (mL) and duration (min) of irrigation] were evaluated. The questionnaire data were collected on paper (baseline visit) or digitally via a secured data management platform (Castor) depending on patient preference. All data were stored on the data management platform.

Sample size
The sample size was based on the primary endpoint of the study: reduction of symptoms at 52 weeks measured using the CCCS and SMIS. According to previously published results, we estimated that patients had a median SMIS of 9 (SD 4.5) and a median CCCS of 13 (SD 4.5) at baseline [5]. Patients were divided into three groups: primary symptoms of constipation, primary symptoms of faecal incontinence and symptoms of both constipation and faecal incontinence (based on the Rome IV criteria for functional constipation [10] and/ or faecal incontinence [11]; these symptoms frequently coexist [17]).
A 30% scale reduction with a variance estimate conservatively set at a SD of 4.5 was considered clinically relevant. A power of 90% and a significance level of 5% was used. To detect a median change of 3.9 in the CCCS pre-versus postirrigation, 17 patients with primary symptoms of constipation had to be included. To detect a median change of 2.7 in the SMIS pre-versus postirrigation, 32 patients with primary symptoms of FI had to be included. To detect a median change of 6.6 in CCCS and SMIS pre-versus postirrigation, eight patients with symptoms of both constipation and faecal incontinence had to be included. Different studies (including a study from our centre) have reported a dropout of approximately 50% of patients after 1 year of follow-up [3,16,18]. Therefore, we aimed to recruit a total of 34 patients with constipation, 64 patients with faecal incontinence and 16 patients with both constipation and faecal incontinence (a total of 114 patients).

Data analysis
All data are presented in the total group as well as per Rome IV classification. Descriptive statistics were used to report the re-  (4) device group (Navina™ Classic vs. Smart). Reduction in symptoms was only analysed in those who continued TAI at 52 weeks' follow-up. Reduction in symptoms was measured per entire group (median, IQR) and per individual (median, IQR) to confirm that symptom improvement was not secondary to drop-out of patients with more severe symptoms. Multivariable logistic regression was used to identify independent factors associated with continuation of TAI at 52 weeks, with results presented as effect sizes (odds ratio) with 95% CIs. Statistical analysis was performed using GraphPad Prism 9.0 and R Studio.

Study participants
Demographics and clinical characteristics of the total study group and subgroups according to the Rome IV criteria for functional constipation and faecal incontinence are detailed in Table 1 Men were most often referred for treatment of functional constipation in isolation. The median body mass index in the whole group was 25.1 kg/m 2 . Women with faecal incontinence in isolation or coexistent symptoms were more likely to be parous than women with functional constipation in isolation (p < 0.0001). Pelvic surgery was more often performed in patients with faecal incontinence in isolation than in patients with functional constipation (65.4% vs. 33.9%; p = 0.007). A history of anal/perineal surgery was frequently reported in patients with faecal incontinence in isolation or coexistent symptoms (34.6% and 37.9%, respectively). Neurological conditions were most common in patients with functional constipation (32.2%).
Two-thirds of all patients underwent pelvic floor physiotherapy (± biofeedback) and six patients (5.3%) tried TAI with another device before entering the study.

Symptomology at baseline
Bowel symptoms at baseline in the total group and per individual Rome IV criterion are shown in Table 2. Remarkably, most patients with faecal incontinence in isolation reported symptoms of incomplete rectal evacuation (88.5%). The median CCCS was 15  in patients with functional constipation and 13 (IQR 11-  c Episodes more often than monthly.

Continuation of TAI
Continuation rates throughout the study are shown in Figure 1.

Symptomology and QoL
Symptom severity and QoL scores at baseline and follow-up are detailed in Table 4.

Constipation
In  of constipation did not occur might be related to the outcome measure which was used for the primary outcome (CCCS) [12]. This questionnaire was also used in the first randomized controlled trial in patients using TAI [5]    Note: 25 patients were excluded from the analyses due to reasons of discontinuation unrelated to insufficient symptom improvement/side effects, or lost to follow-up.Abbreviation: BMI, body mass index.

TA B L E 5
Multivariate regression analyses of demographics and clinical characteristics predicting continuation of transanal irrigation at 52 weeks in 89 patients.

FU N D I N G I N FO R M ATI O N
The study was partly funded by Wellspect. Wellspect was not involved in the study design, the analyses or writing of the manuscript.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.