- Email: [email protected]
Percutaneous tibial nerve stimulation versus sham electrical stimulation for the treatment of faecal incontinence in adults (CONFIDeNT): a double-blind, multicentre, pragmatic, parallel-group, randomised controlled trial
Articles
Percutaneous tibial nerve stimulation versus sham electrical stimulation for the treatment of faecal incontinence in adults (CONFIDeNT): a double-blind, multicentre, pragmatic, parallel-group, randomised controlled trial Charles H Knowles*, Emma J Horrocks*, Stephen A Bremner, Natasha Stevens, Christine Norton, P Ronan O’Connell, Sandra Eldridge, on behalf of the CONFIDeNT study group†
Summary Background Percutaneous tibial nerve stimulation (PTNS) is a new ambulatory therapy for faecal incontinence. Data from case series suggest it has beneficial outcomes in 50–80% patients; however its effectiveness against sham electrical stimulation has not been investigated. We therefore aimed to assess the short-term efficacy of PTNS against sham electrical stimulation in adults with faecal incontinence. Methods We did a double-blind, multicentre, pragmatic, parallel-group, randomised controlled trial (CONtrol of Faecal Incontinence using Distal NeuromodulaTion [CONFIDeNT]) in 17 specialist hospital units in the UK that had the skills to manage patients with faecal incontinence. Eligible participants aged 18 years or older with substantial faecal incontinence for whom conservative treatments (such as dietary changes and pelvic floor exercises) had not worked, were randomly assigned (1:1) to receive either PTNS (via the Urgent PC neuromodulation system) or sham stimulation (via a transcutaneous electrical nerve stimulation machine to the lateral forefoot) once per week for 12 weeks. Randomisation was done with permuted block sizes of two, four, and six, and was stratified by sex and then by centre for women. Patients and outcome assessors were both masked to treatment allocation for the 14-week duration of the trial (but investigators giving the treatment were not masked). The primary outcome was a clinical response to treatment, which we defined as a 50% or greater reduction in episodes of faecal incontinence per week. We assessed this outcome after 12 treatment sessions, using data from patients’ bowel diaries. Analysis was by intention to treat, and missing data were multiply imputed. This trial is registered with the ISRCTN registry, number 88559475, and is closed to new participants. Findings Between Jan 23, 2012, and Oct 31, 2013, we randomly assigned 227 eligible patients (of 373 screened) to receive either PTNS (n=115) or sham stimulation (n=112). 12 patients withdrew from the trial: seven from the PTNS group and five from the sham group (mainly because they could not commit to receiving treatment every week). Two patients (one in each group) withdrew because of an adverse event that was unrelated to treatment (exacerbation of fibromyalgia and rectal bleeding). 39 (38%) of 103 patients with full data from bowel diaries in the PTNS group had a 50% or greater reduction in the number of episodes of faecal incontinence per week compared with 32 (31%) of 102 patients in the sham group (adjusted odds ratio 1·28, 95% CI 0·72–2·28; p=0·396). No serious adverse events related to treatment were reported in the trial. Seven mild, related adverse events were reported in each treatment group, mainly pain at the needle site (four in PTNS, three in sham). Interpretation PTNS given for 12 weeks did not confer significant clinical benefit over sham electrical stimulation in the treatment of adults with faecal incontinence. Further studies are warranted to determine its efficacy in the long term, and in patient subgroups (ie, those with urgency).
Published Online August 18, 2015 http://dx.doi.org/10.1016/ S0140-6736(15)60314-2 See Online/Comment http://dx.doi.org/10.1016/ S0140-6736(15)60508-6 *Contributed equally †Listed at the end of the report National Centre for Bowel Research and Surgical Innovation (Prof C H Knowles PhD, E J Horrocks MBBS) and Pragmatic Clinical Trials Unit (S A Bremner PhD, N Stevens MSc, Prof S Eldridge PhD), Blizard Institute, Queen Mary University of London, London, UK; Florence Nightingale Faculty of Nursing and Midwifery, King’s College London, London, UK (Prof C Norton PhD); and School of Medicine and Medical Science, University College Dublin, Dublin, Ireland (Prof P R O’Connell MD) Correspondence to: Prof Charles Knowles, National Centre for Bowel Research and Surgical Innovation, London E1 2AT, UK [email protected]
Funding National Institute for Health Research.
Introduction Faecal incontinence is a substantial health problem with major effects on physical and emotional wellbeing. Quality of life for patients with this disorder is further reduced by psychological disability, stigmatisation, and social exclusion.1 The prevalence and severity of faecal incontinence increases with age and it is a common cause of admission to residential care, where more than 50% of residents are affected.2 Management of faecal incontinence is challenging because of a widespread shortage of skills to deal with the
disorder and the complex interaction of its causes. Patients for whom conservative management strategies (such as diet, bowel retraining, and drugs) do not work, often need surgical intervention.3 However, surgical procedures are invasive, have variable success rates, and are associated with a risk of significant morbidity.4 In view of these limitations of surgical approaches, the use of neuromodulation to treat faecal incontinence is now well established.5 Chronic, electrical stimulation of the sacral root (sacral nerve stimulation)5,6 is now the first-line surgical intervention.3 This technique is safe,
www.thelancet.com Published online August 18, 2015 http://dx.doi.org/10.1016/S0140-6736(15)60314-2
1
Articles
Research in context Evidence before this study Percutaneous tibial nerve stimulation (PTNS) is a new neuromodulatory therapy for faecal incontinence. We did a systematic review of tibial nerve stimulation (TNS) outcomes for the treatment of faecal incontinence of studies published from Jan 1, 2003, to Nov 1, 2014, with the search terms “tibial nerve stimulation faecal incontinence”, “tibial nerve”, “tibial nerve stimulation”, “faecal incontinence”, and “fecal incontinence”, with no language restrictions. Included studies could be of any design but should have provided selected review outcome data (both baseline and post intervention) for at least ten patients. No exclusions were placed on study centre or patients, in terms of age, sex, ethnic origin, or cause of faecal incontinence. We identified 20 eligible studies, 13 of which studied PTNS. These included ten case series of PTNS, a comparative case matched study of PTNS versus sacral nerve stimulation (SNS), a prospective clinical audit of SNS and PTNS, and a randomised study of PTNS versus transcutaneous tibial nerve stimulation versus sham. From these studies, the success rate of PTNS in the treatment of faecal incontience (with the same primary outcome as the CONFIDeNT Trial) ranged from 52 to 82%.
modifiable, and reversible, but needs two operations and has high direct equipment costs. Tibial nerve stimulation is a minimally invasive alternative to sacral nerve stimulation. Stimulation of the tibial nerve is thought to lead to similar changes in anorectal neuromuscular function as with sacral nerve stimulation because of shared sacral segmental innervation. It is an outpatient treatment, which makes it cheaper than sacral nerve stimulation. Of the two delivery methods of sacral nerve stimulation—percutaneous tibial nerve stimulation (PTNS) and transcutaneous tibial nerve stimulation— PTNS seems to have greater clinical effect.7–9 Data from six case series of PTNS, and one randomised trial, including a total of 371 patients7 have shown improvements in several clinical outcome measures (bowel diary, summative symptom scores, and quality of life) compared with baseline results, leading to clinical success in 62–83% patients in the short-term. However, the outcome of PTNS versus sham stimulation has not been compared in a large, adequately powered, multicentre randomised trial. In this study (CONtrol of Faecal Incontinence using Distal NeuromodulaTion [CONFIDeNT]), we aimed to compare the short-term efficacy of PTNS versus sham electrical stimulation in adults with faecal incontinence.
Methods
Added value of this study We did a multicentre, parallel-group, double-blind, randomised controlled trial of 227 patients with faecal incontinence who were given either PTNS or sham electrical stimulation. The trial did not meet the primary outcome: 39 (38%) patients in the PTNS group had a 50% or greater reduction in episodes of faecal incontinence per week compared with 32 (31%) in the sham group (OR 1·28, 95% CI 0·72–2·28, p=0·396). No serious adverse events related to treatment were reported. Implications of all the available evidence The results of this study suggest that PTNS does not have significant clinical benefit over sham electrical stimulation for the patient population studied. This information is contrary to previous observational data from systematic reviews and might change practice and policy. Further studies are needed to establish the efficacy of PTNS in the long term and within specific subgroups; eg, patients with urgency. Health-service research could also identify how PTNS promotes or contributes to other mixed packages of nurse-led care.
granted ethics approval (10/H0703/25) and local research and development approval at each centre. Patients aged 18 years or older were eligible for inclusion if they had faecal incontinence sufficiently severe to warrant intervention (as recommended by the principal investigator at each site) and for whom medically supervised conservative therapies (a combination of diet, pelvic-floor exercises, biofeedback, and antidiarrhoeal drugs) had not worked. Exclusion criteria included patients who had relevant neurological diseases, such as diabetic neuropathy, multiple sclerosis and Parkinson’s disease; pregnancy; anatomical limitations that would prevent placement of the needle electrode; other medical disorders precluding stimulation (such as bleeding disorders, cardiac pacemakers, peripheral vascular disease or ulcers, or lower leg cellulitis); congenital anorectal anomalies or a complete absence of native rectum due to surgery (eg, anterior resection); cloacal defect; full-thickness rectal prolapse; other previous rectal surgery (rectopexy or resection <12 months before the study [<24 months for cancer]); stoma in situ; chronic bowel diseases with chronic uncontrolled diarrhoea; having had sacral nerve stimulation or PTNS previously; and insufficient knowledge of written or spoken English. Patients provided written informed consent for the study and women of childbearing potential had to provide a negative pregnancy test at baseline on site.
Study design and participants For protocol see http://www. thelancet.com/protocolreviews/11PRT-4126
2
We did a UK-based, double-blind, multicentre, pragmatic, parallel group, randomised controlled trial and enrolled patients from 17 specialist hospital units with the skills to manage patients with faecal incontinence. This trial was
Randomisation and masking Patients were randomly assigned to receive either PTNS or sham electrical stimulation. Allocation was done on an equal basis (1:1) with randomly permuted block sizes
www.thelancet.com Published online August 18, 2015 http://dx.doi.org/10.1016/S0140-6736(15)60314-2
Articles
of varying length (two, four, and six) used to preserve allocation concealment, and initial stratification by sex, then stratification of female patients by centre. With this stratification we aimed to reduce potential confounding effects of variation in disease pathophysiology and outcomes between male and female patients, and to ensure equal allocation to PTNS and sham groups at each centre. Because males only represent 10% of the population with faecal incontinence, only females were allocated by centre. Allocation was managed with an automated, real-time, central web-based system (at Nottingham University Clinical Trials UNit), and done by a local researcher after collection of baseline data, immediately before delivering the first treatment. For the duration of each patient’s involvement with the trial, all patients and investigators doing clinical outcome assessments were masked to treatment allocation, but investigators who delivered the treatment were not masked. To achieve masking, all patients had an identical equipment set-up, which was hidden from their view with a stool covered with a sheet from before set-up to after treatment. A lead wire was attached to patients’ legs in both treatment groups so that treatments were indistinguishable from each other; we also ensured that both groups received identical auditory stimuli. After the final data collection, patients were unmasked, and those in the sham group offered active treatment on an open-label basis.
Procedures Baseline data and eligibility were assessed at a visit 2 weeks before treatments were assigned. PTNS was delivered via the Urgent PC neuromodulation system (Uroplasty, Minnetonka, MN, USA) according to the manufacturer’s instructions. Sham electrical stimulation A
was delivered by transcutaneous electrical nerve stimulation (TENS) to the lateral forefoot (ie, distant and contralateral to the tibial nerve) by modification of the validated sham technique used in the pivotal trial of PTNS for overactive bladder.10 Both groups used the recommended 12 outpatient stimulations lasting 30 min each (one per week).7 Protocol tolerance stipulated a minimum of ten treatments, no less than 5 days or greater than 10 days apart, to be completed in 13 weeks. Treatments were given in individual treatment rooms to patients lying, bare-legged, supine on a couch, and applied to the right leg (unless this was medically contraindicated). A standard text was read to the patient before each treatment (appendix). All PTNS practitioners had training with a minimum of three mentored full 12-session courses of treatment before trial certification. In-trial quality assurance measures were also instigated (appendix). For PTNS, the site of needle insertion was identified and cleaned (5 cm cephalad to and 2 cm behind the medial malleolus). The needle was inserted, advanced 2 cm, and connected to the lead wire, which in turn was connected to the stimulator. The calcaneal reference electrode was attached (figure 1). The lead wire was then taped to the patient’s leg (to mimic sham equipment set-up). The TENS machine (Biostim M7 TENS unit, Biomedical Life Systems, Vista, CA, USA) was connected to two electrodes, one placed under and one on top of the fifth toe (figure 1) but not turned on. The setting for PTNS therapy was established by increasing the current while recording the patient’s sensory response (the appropriate response being in the great toe or sole of foot) or motor response (plantar flexion of the foot or great toe). For the sham group, the same protocol was followed as for PTNS; however, the needle was inserted only 2 mm
See Online for appendix
B
Figure 1: Set-up for procedures (A) Percutaneous tibial nerve stimulation needle and calcaneal electrode. (B) Transcutaneous electrical nerve stimulation surface electrode placements.
www.thelancet.com Published online August 18, 2015 http://dx.doi.org/10.1016/S0140-6736(15)60314-2
3
Articles
373 patients assessed for eligibility
146 ineligible 1 withdrew consent 3 lost to follow-up 96 declined to participate 46 did not have eligibility criteria 13 no faecal incontinence 9 previous sacral nerve stimulation 8 neurological disease 5 symptoms not severe enough 4 unable to speak English 3 sphincter repair <12 months 1 conservative management 1 heart problem 1 unable to complete bowel diary 1 ulcerative colitis
227 patients enrolled and randomised
115 assigned to PTNS
4 discontinued treatment 1 lost to follow-up 1 could not commit to weekly treatment* 2 perceived no efficacy
111 completed six treatments 111 gave mid-treatment data
3 discontinued treatment 2 could not commit to weekly treatment* 1 because of an adverse event unrelated to treatment (exacerbation of fibromyalgia)
108 completed treatment 110 gave follow-up data 115 included in intention-to-treat analysis
112 assigned to sham
2 discontinued treatment 1 lost to follow-up 1 because of an adverse event unrelated to treatment (rectal bleeding)
110 completed six treatments 109 gave mid-treatment data
3 discontinued treatment 1 lost to follow-up 2 could not commit to weekly treatment*
107 completed treatment 109 gave follow-up data 112 included in intention-to-treat analysis
Figure 2: Trial profile PTNS=percutaneous tibial nerve stimulation. Sham=sham electrical stimulation. *Withdrew only from the treatment, not from the study or follow-up.
into the skin. The lead was taped to the patient’s leg near to the needle. After equipment set-up, the practitioner picked up the Urgent PC machine and the TENS machine and turned both machines on. The TENS machine was set to a pulse frequency of 10 Hz and a pulse width of 200 μs. Then, pressing buttons simultaneously on both machines, the practitioner increased the current setting for TENS therapy by observing any sensory or motor response in the toe or ankle. The TENS machine was used to provide the electrical stimulation and the Urgent PC used only to provide the auditory stimulus. Patients completed three bowel diaries: two 2-week diaries, one immediately before and one immediately after 4
a 12-week course of treatment, and a 1-week bowel diary after the sixth treatment. Bowel diaries recorded the number of incontinent bowel movements per day (recorded as rush or urge episodes and passive episodes separately). A masked investigator also administered questionnaires at clinical visits 2 weeks before the first treatment and 2 weeks after the final treatment. These recorded changes in symptom severity scores (St Mark’s Incontinence Score),11 patient-reported outcomes (a short-form version from the International Consultation on Incontinence Questionnaire—Bowel Symptoms [ICIQ-BS]),12 changes in diseasespecific quality of life scores (Gastrointestinal Quality of Life Index [GIQoL]13 and Quality of Life scale for Faecal Incontinence [FIQoL]),14 changes in generic quality of life with the Short Form 36 health survey (SF-36),15 and changes in patients’ health status and overall health with European Quality of Life-5 Dimensions 3 levels (EQ-5D-3L)16 instruments. Patients also completed a further case-report form at 2 weeks after the final treatment. This recorded a Likert scale of global rating of success, change in use of incontinence pads, and drugs for constipation, change in urinary symptoms, general impression of the treatment, and perceived treatment allocation.
Outcomes The primary objective defined patients as responders or non-responders, where a responder achieved a 50% or greater reduction in the mean number of episodes of faecal incontinence per week compared with baseline.6,17 The primary objective was met if PTNS produced a significant (p<0·05) difference in the proportion of responders compared with the sham group. Mean episodes of faecal incontinence per week were calculated from the 2-week bowel diaries by dividing the total number of uncontrolled bowel movements by two. Secondary endpoints included the percentage change in the number of episodes of faecal incontinence per week (ie, the proportion of patients achieving reductions of at least 25%, 75%, or 100% in the mean number of episodes per week) and the change in mean episodes of faecal incontinence per week as a continuous measure. Other secondary endpoints were changes in symptom severity scores, patient-reported outcomes, changes in disease-specific and generic quality of life scores, changes in patients’ health status and overall health, global rating of success, changes in use of incontinence pads, drugs for constipation, changes in urinary symptoms, general impression of the treatment, and perceived treatment allocation (PTNS or sham). Adverse events and concomitant drugs were also recorded at each visit.
Statistical analysis On the basis of 80% power at the 5% significance level, we calculated that 212 patients would be needed to detect the primary endpoint with an estimated success rate (based on available scientific literature10,17,18) of 55% in the PTNS group and 35% in the sham group.
www.thelancet.com Published online August 18, 2015 http://dx.doi.org/10.1016/S0140-6736(15)60314-2
Articles
Data obtained at baseline and after treatment were used to assess the effect of PTNS compared with sham electrical stimulation. All patients randomly assigned to treatment were included in the intention-to-treat analysis, including patients who had no episodes of faecal incontinence at baseline (who, by definition, did not respond to treatment). A statistical analysis plan was agreed before unmasking and analysis. We did the analysis with Stata version 12.1, interfacing with Realcom Impute for multiple imputation of data for missing outcomes and baseline covariates.18 After a burn in of 1000 runs of the Monte Carlo Markov Chain (MCMC) sampler, missing values were filled every 500th run to create a total of ten completed datasets for analysis, the results of which were pooled by Rubin’s rules.19 For all outcomes, we used mixed-effects models, adjusting for outcomes measured at baseline, sex, and a random effect for study centre.20 Linear or logistic models were used, depending on the nature of the outcome, with results presented as effect sizes (odds ratios [OR] or differences in means) with 95% CIs. We also did sensitivity analyses for (1) patients who had ten treatment sessions within 13 weeks with no less than 5 and no greater than 10 days between sessions (per-protocol analysis), (2) excluding any patients who had reported no episodes of faecal incontinence in the baseline bowel diary, and (3) excluding centres who randomly assigned fewer than five patients. Preplanned subgroup analyses for the primary outcome were done by sex, severity of faecal incontinence (participants with at least seven episodes of faecal incontinence per week vs patients with fewer than seven episodes per week), age (<40, 40–60, and >60 years) and type of faecal incontinence (urge faecal incontinence only vs passive faecal incontinence only vs both). A data monitoring committee oversaw the study. This trial is registered with the ISRCTN registry, number 88559475.
Role of the funding source The funder of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication.
Results Between Jan 23, 2012, and Oct 31, 2013, we randomly assigned 227 eligible patients (of 373 screened) to receive either PTNS or sham stimulation (figure 2). 12 patients withdrew from the trial: seven from the PTNS group and five from the sham group (mainly because they could not commit to receiving treatment every week). Two patients (one in each group) withdrew because of an adverse event that was unrelated to treatment (exacerbation of fibromyalgia and rectal bleeding). Baseline demographic and clinical data were similar between groups (table 1). Baseline and follow-up data for outcomes are in table 2. After 12 weeks of intervention, no significant differences were noted between treatment groups for the primary
PTNS (n=115)
Sham (n=112)
104 (90%)
101 (90%)
Sex Female Male
11 (10%)
11 (10%)
Age (years)
58 (50–67)
58 (48–65)
Duration of symptoms (months)
60 (24–168)
48 (24–108)
Parous*
95 (91%)
96 (95%)
Vaginal deliveries only†
90 (95%)
96 (100%)
Caesarean sections only†
5 (5%)
0 (0%)
78 (87%)
82 (85%)
Passive faecal incontinence
88 (77%)
86 (77%)
Urge faecal incontinence
94 (82%)
93 (83%)
Flatus incontinence
74 (64%)
83 (74%)
Evacuatory difficulties
44 (38%)
49 (44%)
Straining
34 (30%)
37 (33%)
Digitation
12 (10%)
15 (13%)
Obstetric history
Episiotomies or tears‡ Bowel function history
Bladder function history Urinary symptoms
70 (61%)
72 (64%)
Urinary urgency
50 (43%)
49 (44%)
Urinary urge incontinence
39 (34%)
42 (38%)
Previous treatments for faecal incontinence Antidiarrhoeal medications
77 (67%)
67 (60%)
Biofeedback
56 (49%)
59 (53%)
Pelvic-floor exercises
37 (32%)
36 (32%)
Fibre supplementation
18 (16%)
30 (27%)
Laxatives, suppositories, or irrigation
20 (17%)
16 (14%)
Anal-sphincter repair Other anal surgery
4 (3%)
4 (4%)
11 (10%)
8 (7%)
Advice about defecation
9 (8%)
7 (6%)
Other
5 (4%)
8 (7%)
30 (29%)
24 (24%)
Relevant medical history Hysterectomy* Vaginal operation*
3 (3%)
2 (2%)
Pelvic operation*
19 (18%)
16 (16%)
Abdominal operation
28 (24%)
30 (27%)
Anal operation
6 (5%)
9 (8%)
Neck or back pain
15 (13%)
21 (19%)
Overactive bladder
15 (13%)
7 (6%)
Diverticular disease
4 (3%)
6 (5%)
Irritable-bowel syndrome
1 (1%)
4 (4%)
Data are n (%) and median (IQR) for 227 patients. PTNS=percutaneous tibial nerve stimulation. Sham=sham electrical stimulation. *Females only. †Percentage calculated from parous females only. ‡Percentage calculated from females with vaginal deliveries only.
Table 1: Baseline characteristics of the intention-to-treat population
outcome: 39 (38%) of 103 patients with full data from bowel diaries in the PTNS group had a 50% or greater reduction in the number of episodes of faecal incontinence per week compared with 32 (31%) of 102 patients with full data in the sham group (unadjusted OR 1·33; adjusted OR [with missing data imputed] 1·28, 95% CI 0·72–2·28; p=0·396).
www.thelancet.com Published online August 18, 2015 http://dx.doi.org/10.1016/S0140-6736(15)60314-2
5
Articles
Baseline
PTNS
After treatment
Sham
Differences in effect at study end (PTNS vs sham; 95% CIs)
PTNS
p value
Sham
Bowel diary data Episodes of faecal incontinence per week
6·0 (2·0–14·0); 9·9 (11·2)
6·9 (2·5–16·0); 10·4 (10·9)
3·5 (1·0–10·0); 6·4 (7·6)
4·8 (1·5–12·8); 9·1 (10·7)
–2·262 (–4·185 to –0·339)
0·021
Episodes of urge faecal incontinence per week
3·0 (0·9–8·0); 5·3 (7·2)
2·5 (0·5–7·0); 4·8 (5·9)
1·5 (0·0–4·5); 3·0 (4·2)
1·5 (0·5–5·5); 4·4 (6·5)
–1·456 (–2·693 to –0·219)
0·021
Episodes of passive faecal incontinence per week
2·0 (0·0–7·5); 4·6 (6·0)
3·0 (0·0–8·0); 5·7 (7·6)
1·5 (0·0–5·0); 3·4 (4·6)
1·5 (0·0–6·5); 4·7 (6·6)
–0·635 (–1·668 to 0·397)
0·228
St Mark’s Incontinence Score*
14·0 (12·0–17·0); 14·4 (3·7)
16·0 (13·0–18·0); 15·4 (4·1)
14·0 (11·0–17·0); 13·9 (4·3)
–0·047 (–1·033 to 0·939)
0·925
15·0 (11·0–18·0); 14·6 (4·6)
Faecal incontinence quality-of-life scale scores Lifestyle†
2·7 (1·8–3·4); 2·6 (0·9)
2·5 (1·7–3·6); 2·6 (1·0)
3·0 (2·2–3·7); 2·8 (0·9)
2·9 (1·9–3·7); 2·8 (1·0)
0·086 (–0·075 to 0·248)
0·290
Coping and behaviour†
1·7 (1·2–2·3); 1·9 (0·7)
1·6 (1·1–2·6); 1·9 (0·9)
1·9 (1·3–2·6); 2·0 (0·8)
1·7 (1·2–2·9); 2·0 (1·0)
0·013 (–0·171 to 0·197)
0·889
Depression and self-perception‡
3·1 (2·0–3·4); 2·8 (0·9)
2·6 (2·0–3·7); 2·7 (0·9)
3·1 (2·2–3·7); 2·9 (1·0)
2·6 (2·0–3·9); 2·8 (1·0)
0·014 (–0·297 to 0·324)
0·927
Embarrassment†
2·0 (1·7–2·7); 2·2 (0·8)
2·0 (1·3–2·7); 2·1 (0·8)
2·7 (1·7–3·0); 2·4 (0·8)
2·3 (1·7–3·0); 2·3 (0·9)
0·036 (–0·151 to 0·223)
0·706
8·9 (7·8–9·8); 8·5 (1·6)
9·2 (8·3–10·0); 8·7 (1·7)
8·4 (6·9–9·4); 7·8 (2·0)
9·3 (7·6–10·0); 8·4 (2·1)
–0·545 0·047 (–1·081 to –0·008)
Patient-centred outcomes§ Gastrointestinal quality of life¶
130·0 (113·0–41·0); 126·7 (18·8)
126·5 (109·0–39·0); 123·8 (20·2)
135·0 (115·0–48.0); 132·0 (20·6)
134·0 (120·0–146·0; 131·6 (20·5)
–1·300 (–5·168 to 2·568)
0·506
70·0 (45·0–90·0); 65·7 (27·4)
65·0 (40·0–85·0); 61·4 (28·4)
75·0 (47·5–90·0); 67·1 (27·7)
70·0 (45·0–90·0); 63·8 (29·0)
–1·854 (–6·992 to 3·284)
0·479
SF-36 scores (%) Physical functioning Role-physical functioning
50·0 (0·0–100·0); 46·4 (42·1)
25·0 (0·0–75·0); 36·4 (41·4)
62·5 (0·0–100·0); 54·4 (44·1)
25·0 (0·0–100·0); 46·2 (44·8)
1·113 0·826 (–8·866 to 11·092)
Bodily pain
60·0 (40·0–90·0); 61·3 (30·0)
57·5 (32·5–90·0); 58·2 (31·5)
67·5 (45·0–90·0); 64·3 (28·3)
67·5 (35·0–90·0); 62·1 (31·0)
–1·026 (–6·815 to 4·764)
0·728
General health
50·0 (35·0–70·0); 51·2 (23·4)
50·0 (30·0–70·0); 50·3 (23·8)
55·0 (30·0–75·0); 52·8 (24·6)
50·0 (35·0–70·0); 50·6 (23·9)
–0·158 (–4·749 to 4·433)
0·946
Vitality
45·0 (30·0–57·5); 43·9 (22·1)
50·0 (30·0–60·0); 42·7 (22·8)
50·0 (25·0–60·0); 45·6 (22·2)
50·0 (35·0–65·0); 46·7 (23·1)
–3·142 (–8·129 to 1·845)
0·215
Social functioning
62·5 (37·5–75·0); 58·4 (28·8)
62·5 (37·5–87·5); 59·3 (31·6)
75·0 (50·0–87.5); 66·4 (28·6)
62·5 (37·5–87.5); 60·6 (31·7)
5·209 (–0·740 to 11·157)
0·087
Role-emotional function
66·7 (0·0–100·0); 58·4 (28·8)
33·3 (0·0–100·0); 59·3 (31·6)
100·0 (0·0–100·0); 61·7 (45·3)
–4·815 (14·802 to 5·171)
0·343
Mental health
60·0 (44·0–76·0); 60·3 (21·0)
–0·509 (–4·831 to 3·814)
0·817
–0·017 (–0·078 to 0·044)
0·583
EQ-5D index score||
0·73 (0·62–0·85); 0·69 (0·27)
64·0 (48·0–76·0); 60·8 (21·6) 0·73 (0·62–0·85); 0·63 (0·34)
64·0 (48·0–84·0); 62·7 (25·1) 0·76 (0·62–0·85); 0·68 (0·28)
83·3 (0.0–100.0); 60·2 (44·1) 64·0 (52·0–76·0); 63·0 (21·4) 0·73 (0·56–0·85); 0·65 (0·34)
Data are unadjusted medians (IQR) and means (SD). PTNS=percutaneous tibial nerve stimulation. Sham=sham electrical stimulation. *Scores range from 0 (best) to 24 (worst). †Scores range from 1 (best) to 4 (worst). ‡Scores range from 1 (best) to 4·4 (worst). §Scores range from 1 (best) to 10 (worst). ¶Scores range from 36 (worst) to 180 (best). ||Scores range from –0·594 (worst) to 1 (best).
Table 2: Outcome measures
No significant differences were noted between PTNS and sham in the proportions of patients achieving more than 25%, more than 75%, and 100% reduction in mean episodes of faecal incontinence per week (figure 3, table 3). However, the mean total episodes of faecal incontinence per week in the PTNS group were significantly decreased at the study end compared with the sham group (difference in means –2·26, 95% CI –4·19 to –0·34; p=0·02, figure 3). Similarly, a significant reduction in the mean number of episodes of 6
urge faecal incontinence per week (–1·46, –2·69 to –0·22; p=0·02) was noted, but not passive faecal incontinence (adjusted difference –0·64, –1·67 to 0·40, p=0·23; figure 3). Reductions in episodes of faecal incontinence were already manifest after 6 weeks of treatment. No significant differences were noted in St Mark’s Incontinence Score between PTNS and sham (difference in means –0·05, 95% CI –1·03 to 0·94; p=0·93), or any significant differences in any domains of disease-specific
www.thelancet.com Published online August 18, 2015 http://dx.doi.org/10.1016/S0140-6736(15)60314-2
Articles
A
PTNS
Sham
Adjusted odds ratio (95% CI) for PTNS vs sham
p value
5·0
Odds ratios
4·0
3·0
≥25% ≥50% ≥75% 100% Reduction in episodes of faecal incontinence per week (%)
B 1·0 Change in mean number of episodes of faecal incontinence per week
46/102 (45%)
1·264 (0·730–2·190)
0·404
39/103 (38%)
32/102 (31%)
1·283 (0·722–2·281)
0·396
≥75% reduction
26/103 (25%)
17/102 (17%)
1·615 (0·770–3·388)
0·205
100% reduction
11/103 (11%)
7/102 (7%)
1·635 (0·592–4·514)
0·344
Table 3: Participants with reductions in episodes of faecal incontinence at follow-up (n=227)
1·0
0
PTNS (n=115)
Sham (n=112)
Pain at needle site
4 (3%)
3 (3%)
Bruising at needle site
2 (2%)
1 (<1%)
Altered sensation at needle site
1 (<1%)
0
Bleeding at needle site
0
2 (2%)
Altered sensation in toe
0
1 (<1%)
Data are n (%). All events were classed as mild by both the patient and the researcher. PTNS=percutaneous tibial nerve stimulation. Sham=sham electrical stimulation.
–1·0
Table 4: Treatment-related adverse events –2·0
–3·0
–4·0
–5·0
Total
Urge
Passive
Episodes of faecal incontinance
C 12 Episodes of faecal incontinence per week
51/103 (50%)
≥50% reduction (primary outcome)
Data are n/N (%). We calculated adjusted odds ratios, CIs, and p values based on imputed and adjusted data. PTNS=percutaneous tibial nerve stimulation. Sham=sham electrical stimulation.
2·0
0
≥25% reduction
Median Mean
10 8 6 4 2 0
Baseline
6 weeks PTNS
14 weeks
Baseline
6 weeks
14 weeks
Sham
Figure 3: Changes in episodes of faecal incontinence per week (A) Categorical reductions; data are odds ratios (95% CI) for PTNS versus sham at 14 weeks. (B) Adjusted differences in mean episodes of faecal incontinence per week (95% CI) for PTNS versus sham at 14 weeks. (C) Frequency of episodes per week throughout the study (PTNS: baseline median 6·0 [IQR 2·0–14·0], mean 9·9 [SD 11·2], 6 weeks median 4·0 [1·0–9·0], mean 6·1 [7·6], 12 weeks median 3·5 [1·0–10·0], mean 6·4 [7·6]; sham: baseline median 6·9 [IQR 2·5–16·0], mean 10·4 [SD 10·9], 6 weeks median 6·0 [1·0–14·0], mean 9·6 [11·5], 12 weeks median 4·8 [1·5–12·8], mean 9·1 [10·7]). PTNS=percutaneous tibial nerve stimulation. Sham=sham electrical stimulation.
FIQOL or generic SF-36 quality of life measures (appendix). Similarly no significant differences between groups in the GIQOL score (–1·30, –5·17 to 2·57; p=0·51) or the EQ5D index scores (–0·02, –0·08 to 0·04; p=0·58) were noted. A significant improvement in patient-reported outcomes was recorded for PTNS compared with sham (difference in means –0·55, 95% CI –1·08 to –0·01; p=0·05), but not for patients’ global impression of success (0·81, –0·06 to 1·67; p=0·07). Of patients who used loperamide at baseline, most (33 (67%) of 49 in PTNS and 32 (84%) of 38 in sham) reported no change in use at study end. Similar proportions in each group (2 [4%] in PTNS vs 2 [5%] in sham) reported increasing loperamide use; however, a higher proportion of patients reduced their loperamide use in the PTNS group compared with the sham group (14 [29%] vs 4 [11%]). In the PTNS group, 57 (54%) of 107 patients thought they had received PTNS and 48 (46%) of 107 thought they had received sham. In the sham group, 32 (31%) of 103 patients thought they had received PTNS and 71 (69%) of 103 thought they had received sham. Most patients in both groups reported no change in use of incontinence pads, and only a few in each group had slight or substantial improvements in urinary symptoms (appendix); no differences were noted between groups. The per-protocol analysis included 197 patients, and did not change interpretation of the results (for the primary outcome of a 50% or greater reduction in episodes of faecal incontinence per week OR 1·27, 95% CI 0·67–2·34; p=0·45), and neither did any of the other subgroup or
www.thelancet.com Published online August 18, 2015 http://dx.doi.org/10.1016/S0140-6736(15)60314-2
7
Articles
sensitivity analyses (appendix). There was also no significant centre effect (appendix). No serious adverse events related to treatment were reported in the trial. Seven mild, related adverse events were reported in each treatment group, mainly pain at the needle site (four in PTNS, three in sham (table 4).
Discussion PTNS did not result in a significant treatment response based on the primary endpoint. However, PTNS did result in a significant reduction in the mean number of total episodes of faecal incontinence per week and mean number of episodes of urge faecal incontinence per week, compared with sham. Although this result is significant, the clinical interpretation of these findings is uncertain, since a reduction from a mean of 6·0 (IQR 2·0–14·0) episodes of faecal incontinence per week to 3·5 (1·0–10·0) might or might not be helpful to patients. The clinical significance of a seemingly beneficial effect of PTNS compared with sham on urge versus passive faecal incontinence is also complex, because only a few (23% in the trial) patients had isolated urge faecal incontinence. Although PTNS also resulted in an improvement in the patient-centred outcomes (a short-form derivative of the validated ICIQ-B),13 this result was not supported by other outcome measures: no significant improvements were noted for summative symptom scores, disease-specific, or generic quality-of-life measures. The results of our trial are contrary to results from published studies of PTNS including six case series,7 one small single-centre, randomised single-blind trial (PTNS vs transcutaneous tibial nerve stimulation vs sham),8 one comparative case-matched study (PTNS vs sacral nerve stimulation)21 and one prospective clinical audit comparing PTNS and sacral nerve stimulation.22 These showed 63–82% response rates with the same primary outcome, which are substantially higher than the 38% reported here. This disparity is probably because of differences in trial design. The inclusion of a sham group helps to negate the effects of natural change in disease status over time, and controls for the well recognised effect of nurse-led face-to-face interventions.23 Placebo responses are disproportionately high in patients with chronic debilitating gastrointestinal disorders, because of a high level of expectation;24 however, our study correctly estimated the response (estimated 35% vs 31% shown). Also, without the rigour of an independently managed trial, bowel diaries can also introduce bias if they are unmasked and subject to investigator interpretation.25 The effect size seen in the PTNS group is nearer to that reported by a recent pilot randomised study (n=40) of PTNS versus sacral nerve stimulation, in which 47% of patients had treatment success in the PTNS group at 3 months.26 These results are similar to those from a large randomised trial of transcutaneous tibial nerve stimulation versus sham electrical stimulation for the same indication, 8
for which no superiority of transcutaneous tibial nerve stimulation was noted.24 Our study had several limitations that might have contributed to the negative outcome. The continuing dilemma of no perfect or universally accepted outcome measure for faecal incontinence is well documented.25 However, for the same reasons as other pivotal, therapeutic faecal incontinence trials,6,17 and to allow comparisons to be drawn between treatment modalities, the 50% or greater reduction was chosen. Quality assurance measures aimed to minimise variation in intervention quality but this could have contributed to differences from previously published reports in single specialist centres.7,27 However, although disparity existed in the proportions of centre recruitment (range 1–45 patients), a sensitivity analysis excluding low-volume centres did not change the primary outcome. Additionally, no significant centre effect was recorded in analysis of the primary outcome. Patient selection is also important. All patients received previous conservative therapy but this was not formally standardised, showing that a consensus has not been reached on what constitutes a minimum UK standard. Thus, a pragmatic view was adopted to balance some heterogeneity in baseline characteristics (eg, degree of treatment refraction) with exclusion of other specific patient subgroups that might have differing response rates. Some efficacy was shown in the reduction of episodes of urge faecal incontinence, and this has been previously documented7,27 and is akin to the approved indication for PTNS to treat overactive bladder.10 However, to determine whether the targeted selection of patients with this baseline characteristic (or others) would lead to a different conclusion would need validation from an appropriately designed study. Other exclusions were required for patients in whom PTNS was contraindicated. Overall, we acknowledge that the results can only be judged as valid for the population studied at the point of intervention in the algorithm of care. For similar pragmatic reasons, use of antidiarrhoeal drugs was not prohibited (because of concerns that unrecorded use would continue). Almost three times as many patients in the PTNS group reduced loperamide compared with the sham group. However, post-hoc χ² testing showed only weak evidence of a difference in loperamide use (p=0·06). Although this result could have confounded the effect size, it would have been unlikely to affect the overall trial conclusion. In conclusion, CONFIDeNT supports the importance of well designed, randomised, controlled trials to define the clinical benefit of treatments for faecal incontinence. On the basis of the evidence presented here, however, to recommend PTNS therapy for the patient population studied here would be difficult. Within the context of audit or research,28 PTNS is already embedded within recommendations for management of faecal incontinence in many centres in the UK and European Union. Further studies are required to establish the efficacy of PTNS in
www.thelancet.com Published online August 18, 2015 http://dx.doi.org/10.1016/S0140-6736(15)60314-2
Articles
the long-term and within specific subgroups; eg, those with urgency. Health-service research could also establish how PTNS promotes or contributes to other mixed packages of nurse-led care. Contributors CHK researched and did the study design with EJH, SAB, NS, CN, PRO’C, and SE, acquired the funding, interpreted the data with EJH, SAB, and NS, wrote the report with EJH, SAB, NS, CN, PRO’C, and SE, and supervised a research student. EJH obtained the data and analysed the data with SAB and SE. SAB did the figures, and NS managed the trial. The CONFIDeNT Study Group Chair of Data and Safety Monitoring Committee: Dion Morton Aintree University Hospitals NHS Foundation Trust: Paul Skaife, Sandra Blythin, Rachel Smith, Ching Way Medical Centre: Pasquale Giordano, Salvatore Lamberti Sandwell and West Birmingham Hospitals NHS Trust: Katherine Gill, Julie Collie, United Lincolnshire Hospitals NHS Trust: Pradeep Aggarwal, Kimberley Netherton, University Hospital of South Manchester NHS Foundation Trust: Karen Telford, James Nicholson, Claire Molyneux, Sheffield Teaching Hospitals NHS Foundation Trust: Steve Brown, Lynne Smith, University College London Hospitals NHS Foundation Trust: Anton Emmanuel, Ahsan Alam, Guy’s and St Thomas’ NHS Foundation Trust: Alexis Schizas, Emma Collins, Nottingham University Hospitals NHS Trust: Charles Maxwell-Armstrong, Rudra Maitra, Poole Hospital NHS Foundation Trust: Andrew Clarke, Steve Perring, Emma Jones, University Hospital Southampton NHS Foundation Trust: Karen Nugent, Janet Hicks, Leeds Teaching Hospitals NHS Trust: Dermot Burke, Clare Liddell Hull and East Yorkshire Hospitals NHS Trust: Graham Duthie, Philip Waudby, North West London Hospitals NHS Trust: Carolynne Vaizey, Sherill Tripoli, University Hospitals of Leicester NHS Trust: Justin Yeung University Hospitals Bristol NHS Foundation Trust: Paul Durdey. Declaration of interests We declare no competing interests. Acknowledgments The trial was under the auspices of the Chief Investigator and the Pragmatic Clinical Trials Unit (PCTU), Barts and The London School of Medicine and Dentistry (sponsor: Queen Mary University of London). It was funded by the UK NIHR Health Technology Assessment (HTA) programme (HTA 09/104/16). The views and opinions expressed in this report are those of the authors and are not necessarily those of the HTA programme, NIHR, the UK National Health Service, or the UK Department of Health. We would like to acknowledge the following people in relation to the CONFIDeNT study: Deborah Gilbert (Patient and Public Involvement representative), Norman Williams (clinical adviser), Sandy Smith, Mike Waring, Lara Edwards, Anitha Manivannan, Glenn Poon, Syed Arafath, and Polly Jordan (the Data Management and Quality Assurance team), and Elaine Denny and Daniel Altmann (the Data and Safety Monitoring committee). References 1 Nelson R, Norton N, Cautley E, et al. Community-based prevalence of anal incontinence. JAMA 1995; 274: 559–61. 2 Whitehead WE, Borrud L, Goode PS, et al. Fecal incontinence in US adults: epidemiology and risk factors. Gastroenterology 2009; 137: 512–17. 3 National Institute for Health and Care Excellence. Faecal incontinence: the management of faecal incontinence in adults. NICE guideline CG 49. London: National Institute for Health and Care Excellence, 2007. 4 Brown SR, Wadhawan H, Nelson RL. Surgery for faecal incontinence in adults. Cochrane Database Syst Rev 2013; 7: CD001757. 5 Matzel KE, Stadelmaier U, Hohenfellner M, et al. Electrical stimulation of sacral spinal nerves for treatment of faecal incontinence. Lancet 1995; 346: 1124–27. 6 Wexner SD, Coller JA, Devroede G, et al. Sacral nerve stimulation for fecal incontinence: results of a 120-patient prospective multicenter study. Ann Surg 2010; 251: 441–49.
7
8
9
10
11 12
13
14
15
16 17
18
19
20
21
22
23
24
25 26
27
28
Horrocks EJ, Thin N, Thaha MA, et al. Systematic review of tibial nerve stimulation to treat faecal incontinence. Br J Surg 2014; 101: 457–68. George AT, Kalmar K, Sala S, et al. Randomized controlled trial of percutaneous versus transcutaneous posterior tibial nerve stimulation in faecal incontinence. Br J Surg 2013; 100: 330–38. Leroi AM, Siproudhis L, Etienney I, et al. Transcutaneous electrical tibial nerve stimulation in the treatment of fecal incontinence: a randomized trial. Am J Gastroenterol 2012; 107: 1888–96. Peters KM, Carrico DJ, Perez-Marrero RA, et al. Randomized trial of percutaneous tibial nerve stimulation versus sham efficacy in the treatment of overactive bladder syndrome: results from the SUmiT trial. J Urol 2010; 183: 1438–43. Vaizey CJ, Carapeti E, Cahill JA, et al. Prospective comparison of faecal incontinence grading systems. Gut 1999; 44: 77–80. Cotterill N, Norton C, Avery KNL, et al. Psychometric evaluation of a new patient-completed questionnaire for evaluating anal incontinence symptoms and impact on quality of life: the ICIQ-B. Dis Colon Rectum 2011; 54: 1235–50. Eypasch E, Williams JI, Wood-Dauphinee S, et al. Gastrointestinal Quality of Life Index: development, validation and application of a new instrument. Br J Surg 1995; 82: 216–22. Rockwood TH, Church JM, Fleshman JW, et al. Fecal Incontinence Quality of Life Scale: quality of life instrument for patients with fecal incontinence. Dis Colon Rectum 2000; 43: 9–17. Stewart AL, Hays RD, Ware JE Jr. The MOS short-form general health survey. Reliability and validity in a patient population. Med Care 1988; 26: 724–35. EuroQoL Group. EuroQol—a new facility for the measurement of health-related quality of life. Health Policy 1990; 19: 199–208. Graf W, Mellgren A, Matzel KE, et al. Efficacy of dextranomer in stabilised hyaluronic acid for treatment of faecal incontinence: a randomised, sham-controlled trial. Lancet 2011; 377: 997–1003. Enck P, Horing B, Weimer K, Klosterhalfen S. Placebo responses and placebo effects in functional bowel disorders. Eur J Gastroenterol Hepatol 2012; 24: 1–8. Carpenter JR, Goldstein H, Kenward MG. REAL COM-IMPUTE software for multilevel multiple imputation with mixed response types. J Stat Soft 2011; 45: 1–12. Kahan BC, Morris TP. Analysis of multicentre trials with continuous outcomes: when and how should we account for centre effects? Stat Med 2013; 32: 1136–49. Asari SA, Meurette G, Mantoo S, et al. Percutaneous tibial nerve versus sacral nerve stimulation for faecal incontinence: a comparative case-matched study. Colorectal Dis 2014; 16: O393–99. Hotouras A, Murphy J, Allison M, et al. Prospective clinical audit of two neuromodulatory treatments for fecal incontinence: sacral nerve stimulation (SNS) and percutaneous tibial nerve stimulation (PTNS). Surg Today 2014; 44: 2124–30. Pacheco-Lopez G, Engler H, Niemi MB, et al. Expectations and associations that heal: immunomodulatory placebo effects and its neurobiology. Brain Behav Immun 2006; 20: 430–46. Musial F, Klosterhalfen S, Enck P. Placebo responses in patients with gastrointestinal disorders. World J Gastroenterol 2007; 13: 3425–29. Vaizey CJ. Faecal incontinence: standardizing outcome measures. Colorectal Dis 2014; 16: 156–58. Thin NN, Taylor SJ, Bremner SA, et al. Randomised clinical trial of sacral and percutaneous tibial nerve stimulation in patients with faecal incontinence. Br J Surg 2015; 102: 349–58. Hotouras A, Thaha MA, Boyle D, et al. Short-term outcome following percutaneous tibial nerve stimulation (PTNS) for faecal incontinence: a single-centre prospective study. Colorectal Dis 2011; 14: 1101–05. National Institute for Health and Care Excellence. NICE Interventional Procedure Guideline (IPG395). London: National Institute for Health and Care Excellence, 2011.
www.thelancet.com Published online August 18, 2015 http://dx.doi.org/10.1016/S0140-6736(15)60314-2
9
Percutaneous tibial nerve stimulation versus sham electrical stimulation for the treatment of faecal incontinence in adults (CONFIDeNT): a double-blind, multicentre, pragmatic, parallel-group, randomised controlled trial Charles H Knowles*, Emma J Horrocks*, Stephen A Bremner, Natasha Stevens, Christine Norton, P Ronan O’Connell, Sandra Eldridge, on behalf of the CONFIDeNT study group†
Summary Background Percutaneous tibial nerve stimulation (PTNS) is a new ambulatory therapy for faecal incontinence. Data from case series suggest it has beneficial outcomes in 50–80% patients; however its effectiveness against sham electrical stimulation has not been investigated. We therefore aimed to assess the short-term efficacy of PTNS against sham electrical stimulation in adults with faecal incontinence. Methods We did a double-blind, multicentre, pragmatic, parallel-group, randomised controlled trial (CONtrol of Faecal Incontinence using Distal NeuromodulaTion [CONFIDeNT]) in 17 specialist hospital units in the UK that had the skills to manage patients with faecal incontinence. Eligible participants aged 18 years or older with substantial faecal incontinence for whom conservative treatments (such as dietary changes and pelvic floor exercises) had not worked, were randomly assigned (1:1) to receive either PTNS (via the Urgent PC neuromodulation system) or sham stimulation (via a transcutaneous electrical nerve stimulation machine to the lateral forefoot) once per week for 12 weeks. Randomisation was done with permuted block sizes of two, four, and six, and was stratified by sex and then by centre for women. Patients and outcome assessors were both masked to treatment allocation for the 14-week duration of the trial (but investigators giving the treatment were not masked). The primary outcome was a clinical response to treatment, which we defined as a 50% or greater reduction in episodes of faecal incontinence per week. We assessed this outcome after 12 treatment sessions, using data from patients’ bowel diaries. Analysis was by intention to treat, and missing data were multiply imputed. This trial is registered with the ISRCTN registry, number 88559475, and is closed to new participants. Findings Between Jan 23, 2012, and Oct 31, 2013, we randomly assigned 227 eligible patients (of 373 screened) to receive either PTNS (n=115) or sham stimulation (n=112). 12 patients withdrew from the trial: seven from the PTNS group and five from the sham group (mainly because they could not commit to receiving treatment every week). Two patients (one in each group) withdrew because of an adverse event that was unrelated to treatment (exacerbation of fibromyalgia and rectal bleeding). 39 (38%) of 103 patients with full data from bowel diaries in the PTNS group had a 50% or greater reduction in the number of episodes of faecal incontinence per week compared with 32 (31%) of 102 patients in the sham group (adjusted odds ratio 1·28, 95% CI 0·72–2·28; p=0·396). No serious adverse events related to treatment were reported in the trial. Seven mild, related adverse events were reported in each treatment group, mainly pain at the needle site (four in PTNS, three in sham). Interpretation PTNS given for 12 weeks did not confer significant clinical benefit over sham electrical stimulation in the treatment of adults with faecal incontinence. Further studies are warranted to determine its efficacy in the long term, and in patient subgroups (ie, those with urgency).
Published Online August 18, 2015 http://dx.doi.org/10.1016/ S0140-6736(15)60314-2 See Online/Comment http://dx.doi.org/10.1016/ S0140-6736(15)60508-6 *Contributed equally †Listed at the end of the report National Centre for Bowel Research and Surgical Innovation (Prof C H Knowles PhD, E J Horrocks MBBS) and Pragmatic Clinical Trials Unit (S A Bremner PhD, N Stevens MSc, Prof S Eldridge PhD), Blizard Institute, Queen Mary University of London, London, UK; Florence Nightingale Faculty of Nursing and Midwifery, King’s College London, London, UK (Prof C Norton PhD); and School of Medicine and Medical Science, University College Dublin, Dublin, Ireland (Prof P R O’Connell MD) Correspondence to: Prof Charles Knowles, National Centre for Bowel Research and Surgical Innovation, London E1 2AT, UK [email protected]
Funding National Institute for Health Research.
Introduction Faecal incontinence is a substantial health problem with major effects on physical and emotional wellbeing. Quality of life for patients with this disorder is further reduced by psychological disability, stigmatisation, and social exclusion.1 The prevalence and severity of faecal incontinence increases with age and it is a common cause of admission to residential care, where more than 50% of residents are affected.2 Management of faecal incontinence is challenging because of a widespread shortage of skills to deal with the
disorder and the complex interaction of its causes. Patients for whom conservative management strategies (such as diet, bowel retraining, and drugs) do not work, often need surgical intervention.3 However, surgical procedures are invasive, have variable success rates, and are associated with a risk of significant morbidity.4 In view of these limitations of surgical approaches, the use of neuromodulation to treat faecal incontinence is now well established.5 Chronic, electrical stimulation of the sacral root (sacral nerve stimulation)5,6 is now the first-line surgical intervention.3 This technique is safe,
www.thelancet.com Published online August 18, 2015 http://dx.doi.org/10.1016/S0140-6736(15)60314-2
1
Articles
Research in context Evidence before this study Percutaneous tibial nerve stimulation (PTNS) is a new neuromodulatory therapy for faecal incontinence. We did a systematic review of tibial nerve stimulation (TNS) outcomes for the treatment of faecal incontinence of studies published from Jan 1, 2003, to Nov 1, 2014, with the search terms “tibial nerve stimulation faecal incontinence”, “tibial nerve”, “tibial nerve stimulation”, “faecal incontinence”, and “fecal incontinence”, with no language restrictions. Included studies could be of any design but should have provided selected review outcome data (both baseline and post intervention) for at least ten patients. No exclusions were placed on study centre or patients, in terms of age, sex, ethnic origin, or cause of faecal incontinence. We identified 20 eligible studies, 13 of which studied PTNS. These included ten case series of PTNS, a comparative case matched study of PTNS versus sacral nerve stimulation (SNS), a prospective clinical audit of SNS and PTNS, and a randomised study of PTNS versus transcutaneous tibial nerve stimulation versus sham. From these studies, the success rate of PTNS in the treatment of faecal incontience (with the same primary outcome as the CONFIDeNT Trial) ranged from 52 to 82%.
modifiable, and reversible, but needs two operations and has high direct equipment costs. Tibial nerve stimulation is a minimally invasive alternative to sacral nerve stimulation. Stimulation of the tibial nerve is thought to lead to similar changes in anorectal neuromuscular function as with sacral nerve stimulation because of shared sacral segmental innervation. It is an outpatient treatment, which makes it cheaper than sacral nerve stimulation. Of the two delivery methods of sacral nerve stimulation—percutaneous tibial nerve stimulation (PTNS) and transcutaneous tibial nerve stimulation— PTNS seems to have greater clinical effect.7–9 Data from six case series of PTNS, and one randomised trial, including a total of 371 patients7 have shown improvements in several clinical outcome measures (bowel diary, summative symptom scores, and quality of life) compared with baseline results, leading to clinical success in 62–83% patients in the short-term. However, the outcome of PTNS versus sham stimulation has not been compared in a large, adequately powered, multicentre randomised trial. In this study (CONtrol of Faecal Incontinence using Distal NeuromodulaTion [CONFIDeNT]), we aimed to compare the short-term efficacy of PTNS versus sham electrical stimulation in adults with faecal incontinence.
Methods
Added value of this study We did a multicentre, parallel-group, double-blind, randomised controlled trial of 227 patients with faecal incontinence who were given either PTNS or sham electrical stimulation. The trial did not meet the primary outcome: 39 (38%) patients in the PTNS group had a 50% or greater reduction in episodes of faecal incontinence per week compared with 32 (31%) in the sham group (OR 1·28, 95% CI 0·72–2·28, p=0·396). No serious adverse events related to treatment were reported. Implications of all the available evidence The results of this study suggest that PTNS does not have significant clinical benefit over sham electrical stimulation for the patient population studied. This information is contrary to previous observational data from systematic reviews and might change practice and policy. Further studies are needed to establish the efficacy of PTNS in the long term and within specific subgroups; eg, patients with urgency. Health-service research could also identify how PTNS promotes or contributes to other mixed packages of nurse-led care.
granted ethics approval (10/H0703/25) and local research and development approval at each centre. Patients aged 18 years or older were eligible for inclusion if they had faecal incontinence sufficiently severe to warrant intervention (as recommended by the principal investigator at each site) and for whom medically supervised conservative therapies (a combination of diet, pelvic-floor exercises, biofeedback, and antidiarrhoeal drugs) had not worked. Exclusion criteria included patients who had relevant neurological diseases, such as diabetic neuropathy, multiple sclerosis and Parkinson’s disease; pregnancy; anatomical limitations that would prevent placement of the needle electrode; other medical disorders precluding stimulation (such as bleeding disorders, cardiac pacemakers, peripheral vascular disease or ulcers, or lower leg cellulitis); congenital anorectal anomalies or a complete absence of native rectum due to surgery (eg, anterior resection); cloacal defect; full-thickness rectal prolapse; other previous rectal surgery (rectopexy or resection <12 months before the study [<24 months for cancer]); stoma in situ; chronic bowel diseases with chronic uncontrolled diarrhoea; having had sacral nerve stimulation or PTNS previously; and insufficient knowledge of written or spoken English. Patients provided written informed consent for the study and women of childbearing potential had to provide a negative pregnancy test at baseline on site.
Study design and participants For protocol see http://www. thelancet.com/protocolreviews/11PRT-4126
2
We did a UK-based, double-blind, multicentre, pragmatic, parallel group, randomised controlled trial and enrolled patients from 17 specialist hospital units with the skills to manage patients with faecal incontinence. This trial was
Randomisation and masking Patients were randomly assigned to receive either PTNS or sham electrical stimulation. Allocation was done on an equal basis (1:1) with randomly permuted block sizes
www.thelancet.com Published online August 18, 2015 http://dx.doi.org/10.1016/S0140-6736(15)60314-2
Articles
of varying length (two, four, and six) used to preserve allocation concealment, and initial stratification by sex, then stratification of female patients by centre. With this stratification we aimed to reduce potential confounding effects of variation in disease pathophysiology and outcomes between male and female patients, and to ensure equal allocation to PTNS and sham groups at each centre. Because males only represent 10% of the population with faecal incontinence, only females were allocated by centre. Allocation was managed with an automated, real-time, central web-based system (at Nottingham University Clinical Trials UNit), and done by a local researcher after collection of baseline data, immediately before delivering the first treatment. For the duration of each patient’s involvement with the trial, all patients and investigators doing clinical outcome assessments were masked to treatment allocation, but investigators who delivered the treatment were not masked. To achieve masking, all patients had an identical equipment set-up, which was hidden from their view with a stool covered with a sheet from before set-up to after treatment. A lead wire was attached to patients’ legs in both treatment groups so that treatments were indistinguishable from each other; we also ensured that both groups received identical auditory stimuli. After the final data collection, patients were unmasked, and those in the sham group offered active treatment on an open-label basis.
Procedures Baseline data and eligibility were assessed at a visit 2 weeks before treatments were assigned. PTNS was delivered via the Urgent PC neuromodulation system (Uroplasty, Minnetonka, MN, USA) according to the manufacturer’s instructions. Sham electrical stimulation A
was delivered by transcutaneous electrical nerve stimulation (TENS) to the lateral forefoot (ie, distant and contralateral to the tibial nerve) by modification of the validated sham technique used in the pivotal trial of PTNS for overactive bladder.10 Both groups used the recommended 12 outpatient stimulations lasting 30 min each (one per week).7 Protocol tolerance stipulated a minimum of ten treatments, no less than 5 days or greater than 10 days apart, to be completed in 13 weeks. Treatments were given in individual treatment rooms to patients lying, bare-legged, supine on a couch, and applied to the right leg (unless this was medically contraindicated). A standard text was read to the patient before each treatment (appendix). All PTNS practitioners had training with a minimum of three mentored full 12-session courses of treatment before trial certification. In-trial quality assurance measures were also instigated (appendix). For PTNS, the site of needle insertion was identified and cleaned (5 cm cephalad to and 2 cm behind the medial malleolus). The needle was inserted, advanced 2 cm, and connected to the lead wire, which in turn was connected to the stimulator. The calcaneal reference electrode was attached (figure 1). The lead wire was then taped to the patient’s leg (to mimic sham equipment set-up). The TENS machine (Biostim M7 TENS unit, Biomedical Life Systems, Vista, CA, USA) was connected to two electrodes, one placed under and one on top of the fifth toe (figure 1) but not turned on. The setting for PTNS therapy was established by increasing the current while recording the patient’s sensory response (the appropriate response being in the great toe or sole of foot) or motor response (plantar flexion of the foot or great toe). For the sham group, the same protocol was followed as for PTNS; however, the needle was inserted only 2 mm
See Online for appendix
B
Figure 1: Set-up for procedures (A) Percutaneous tibial nerve stimulation needle and calcaneal electrode. (B) Transcutaneous electrical nerve stimulation surface electrode placements.
www.thelancet.com Published online August 18, 2015 http://dx.doi.org/10.1016/S0140-6736(15)60314-2
3
Articles
373 patients assessed for eligibility
146 ineligible 1 withdrew consent 3 lost to follow-up 96 declined to participate 46 did not have eligibility criteria 13 no faecal incontinence 9 previous sacral nerve stimulation 8 neurological disease 5 symptoms not severe enough 4 unable to speak English 3 sphincter repair <12 months 1 conservative management 1 heart problem 1 unable to complete bowel diary 1 ulcerative colitis
227 patients enrolled and randomised
115 assigned to PTNS
4 discontinued treatment 1 lost to follow-up 1 could not commit to weekly treatment* 2 perceived no efficacy
111 completed six treatments 111 gave mid-treatment data
3 discontinued treatment 2 could not commit to weekly treatment* 1 because of an adverse event unrelated to treatment (exacerbation of fibromyalgia)
108 completed treatment 110 gave follow-up data 115 included in intention-to-treat analysis
112 assigned to sham
2 discontinued treatment 1 lost to follow-up 1 because of an adverse event unrelated to treatment (rectal bleeding)
110 completed six treatments 109 gave mid-treatment data
3 discontinued treatment 1 lost to follow-up 2 could not commit to weekly treatment*
107 completed treatment 109 gave follow-up data 112 included in intention-to-treat analysis
Figure 2: Trial profile PTNS=percutaneous tibial nerve stimulation. Sham=sham electrical stimulation. *Withdrew only from the treatment, not from the study or follow-up.
into the skin. The lead was taped to the patient’s leg near to the needle. After equipment set-up, the practitioner picked up the Urgent PC machine and the TENS machine and turned both machines on. The TENS machine was set to a pulse frequency of 10 Hz and a pulse width of 200 μs. Then, pressing buttons simultaneously on both machines, the practitioner increased the current setting for TENS therapy by observing any sensory or motor response in the toe or ankle. The TENS machine was used to provide the electrical stimulation and the Urgent PC used only to provide the auditory stimulus. Patients completed three bowel diaries: two 2-week diaries, one immediately before and one immediately after 4
a 12-week course of treatment, and a 1-week bowel diary after the sixth treatment. Bowel diaries recorded the number of incontinent bowel movements per day (recorded as rush or urge episodes and passive episodes separately). A masked investigator also administered questionnaires at clinical visits 2 weeks before the first treatment and 2 weeks after the final treatment. These recorded changes in symptom severity scores (St Mark’s Incontinence Score),11 patient-reported outcomes (a short-form version from the International Consultation on Incontinence Questionnaire—Bowel Symptoms [ICIQ-BS]),12 changes in diseasespecific quality of life scores (Gastrointestinal Quality of Life Index [GIQoL]13 and Quality of Life scale for Faecal Incontinence [FIQoL]),14 changes in generic quality of life with the Short Form 36 health survey (SF-36),15 and changes in patients’ health status and overall health with European Quality of Life-5 Dimensions 3 levels (EQ-5D-3L)16 instruments. Patients also completed a further case-report form at 2 weeks after the final treatment. This recorded a Likert scale of global rating of success, change in use of incontinence pads, and drugs for constipation, change in urinary symptoms, general impression of the treatment, and perceived treatment allocation.
Outcomes The primary objective defined patients as responders or non-responders, where a responder achieved a 50% or greater reduction in the mean number of episodes of faecal incontinence per week compared with baseline.6,17 The primary objective was met if PTNS produced a significant (p<0·05) difference in the proportion of responders compared with the sham group. Mean episodes of faecal incontinence per week were calculated from the 2-week bowel diaries by dividing the total number of uncontrolled bowel movements by two. Secondary endpoints included the percentage change in the number of episodes of faecal incontinence per week (ie, the proportion of patients achieving reductions of at least 25%, 75%, or 100% in the mean number of episodes per week) and the change in mean episodes of faecal incontinence per week as a continuous measure. Other secondary endpoints were changes in symptom severity scores, patient-reported outcomes, changes in disease-specific and generic quality of life scores, changes in patients’ health status and overall health, global rating of success, changes in use of incontinence pads, drugs for constipation, changes in urinary symptoms, general impression of the treatment, and perceived treatment allocation (PTNS or sham). Adverse events and concomitant drugs were also recorded at each visit.
Statistical analysis On the basis of 80% power at the 5% significance level, we calculated that 212 patients would be needed to detect the primary endpoint with an estimated success rate (based on available scientific literature10,17,18) of 55% in the PTNS group and 35% in the sham group.
www.thelancet.com Published online August 18, 2015 http://dx.doi.org/10.1016/S0140-6736(15)60314-2
Articles
Data obtained at baseline and after treatment were used to assess the effect of PTNS compared with sham electrical stimulation. All patients randomly assigned to treatment were included in the intention-to-treat analysis, including patients who had no episodes of faecal incontinence at baseline (who, by definition, did not respond to treatment). A statistical analysis plan was agreed before unmasking and analysis. We did the analysis with Stata version 12.1, interfacing with Realcom Impute for multiple imputation of data for missing outcomes and baseline covariates.18 After a burn in of 1000 runs of the Monte Carlo Markov Chain (MCMC) sampler, missing values were filled every 500th run to create a total of ten completed datasets for analysis, the results of which were pooled by Rubin’s rules.19 For all outcomes, we used mixed-effects models, adjusting for outcomes measured at baseline, sex, and a random effect for study centre.20 Linear or logistic models were used, depending on the nature of the outcome, with results presented as effect sizes (odds ratios [OR] or differences in means) with 95% CIs. We also did sensitivity analyses for (1) patients who had ten treatment sessions within 13 weeks with no less than 5 and no greater than 10 days between sessions (per-protocol analysis), (2) excluding any patients who had reported no episodes of faecal incontinence in the baseline bowel diary, and (3) excluding centres who randomly assigned fewer than five patients. Preplanned subgroup analyses for the primary outcome were done by sex, severity of faecal incontinence (participants with at least seven episodes of faecal incontinence per week vs patients with fewer than seven episodes per week), age (<40, 40–60, and >60 years) and type of faecal incontinence (urge faecal incontinence only vs passive faecal incontinence only vs both). A data monitoring committee oversaw the study. This trial is registered with the ISRCTN registry, number 88559475.
Role of the funding source The funder of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication.
Results Between Jan 23, 2012, and Oct 31, 2013, we randomly assigned 227 eligible patients (of 373 screened) to receive either PTNS or sham stimulation (figure 2). 12 patients withdrew from the trial: seven from the PTNS group and five from the sham group (mainly because they could not commit to receiving treatment every week). Two patients (one in each group) withdrew because of an adverse event that was unrelated to treatment (exacerbation of fibromyalgia and rectal bleeding). Baseline demographic and clinical data were similar between groups (table 1). Baseline and follow-up data for outcomes are in table 2. After 12 weeks of intervention, no significant differences were noted between treatment groups for the primary
PTNS (n=115)
Sham (n=112)
104 (90%)
101 (90%)
Sex Female Male
11 (10%)
11 (10%)
Age (years)
58 (50–67)
58 (48–65)
Duration of symptoms (months)
60 (24–168)
48 (24–108)
Parous*
95 (91%)
96 (95%)
Vaginal deliveries only†
90 (95%)
96 (100%)
Caesarean sections only†
5 (5%)
0 (0%)
78 (87%)
82 (85%)
Passive faecal incontinence
88 (77%)
86 (77%)
Urge faecal incontinence
94 (82%)
93 (83%)
Flatus incontinence
74 (64%)
83 (74%)
Evacuatory difficulties
44 (38%)
49 (44%)
Straining
34 (30%)
37 (33%)
Digitation
12 (10%)
15 (13%)
Obstetric history
Episiotomies or tears‡ Bowel function history
Bladder function history Urinary symptoms
70 (61%)
72 (64%)
Urinary urgency
50 (43%)
49 (44%)
Urinary urge incontinence
39 (34%)
42 (38%)
Previous treatments for faecal incontinence Antidiarrhoeal medications
77 (67%)
67 (60%)
Biofeedback
56 (49%)
59 (53%)
Pelvic-floor exercises
37 (32%)
36 (32%)
Fibre supplementation
18 (16%)
30 (27%)
Laxatives, suppositories, or irrigation
20 (17%)
16 (14%)
Anal-sphincter repair Other anal surgery
4 (3%)
4 (4%)
11 (10%)
8 (7%)
Advice about defecation
9 (8%)
7 (6%)
Other
5 (4%)
8 (7%)
30 (29%)
24 (24%)
Relevant medical history Hysterectomy* Vaginal operation*
3 (3%)
2 (2%)
Pelvic operation*
19 (18%)
16 (16%)
Abdominal operation
28 (24%)
30 (27%)
Anal operation
6 (5%)
9 (8%)
Neck or back pain
15 (13%)
21 (19%)
Overactive bladder
15 (13%)
7 (6%)
Diverticular disease
4 (3%)
6 (5%)
Irritable-bowel syndrome
1 (1%)
4 (4%)
Data are n (%) and median (IQR) for 227 patients. PTNS=percutaneous tibial nerve stimulation. Sham=sham electrical stimulation. *Females only. †Percentage calculated from parous females only. ‡Percentage calculated from females with vaginal deliveries only.
Table 1: Baseline characteristics of the intention-to-treat population
outcome: 39 (38%) of 103 patients with full data from bowel diaries in the PTNS group had a 50% or greater reduction in the number of episodes of faecal incontinence per week compared with 32 (31%) of 102 patients with full data in the sham group (unadjusted OR 1·33; adjusted OR [with missing data imputed] 1·28, 95% CI 0·72–2·28; p=0·396).
www.thelancet.com Published online August 18, 2015 http://dx.doi.org/10.1016/S0140-6736(15)60314-2
5
Articles
Baseline
PTNS
After treatment
Sham
Differences in effect at study end (PTNS vs sham; 95% CIs)
PTNS
p value
Sham
Bowel diary data Episodes of faecal incontinence per week
6·0 (2·0–14·0); 9·9 (11·2)
6·9 (2·5–16·0); 10·4 (10·9)
3·5 (1·0–10·0); 6·4 (7·6)
4·8 (1·5–12·8); 9·1 (10·7)
–2·262 (–4·185 to –0·339)
0·021
Episodes of urge faecal incontinence per week
3·0 (0·9–8·0); 5·3 (7·2)
2·5 (0·5–7·0); 4·8 (5·9)
1·5 (0·0–4·5); 3·0 (4·2)
1·5 (0·5–5·5); 4·4 (6·5)
–1·456 (–2·693 to –0·219)
0·021
Episodes of passive faecal incontinence per week
2·0 (0·0–7·5); 4·6 (6·0)
3·0 (0·0–8·0); 5·7 (7·6)
1·5 (0·0–5·0); 3·4 (4·6)
1·5 (0·0–6·5); 4·7 (6·6)
–0·635 (–1·668 to 0·397)
0·228
St Mark’s Incontinence Score*
14·0 (12·0–17·0); 14·4 (3·7)
16·0 (13·0–18·0); 15·4 (4·1)
14·0 (11·0–17·0); 13·9 (4·3)
–0·047 (–1·033 to 0·939)
0·925
15·0 (11·0–18·0); 14·6 (4·6)
Faecal incontinence quality-of-life scale scores Lifestyle†
2·7 (1·8–3·4); 2·6 (0·9)
2·5 (1·7–3·6); 2·6 (1·0)
3·0 (2·2–3·7); 2·8 (0·9)
2·9 (1·9–3·7); 2·8 (1·0)
0·086 (–0·075 to 0·248)
0·290
Coping and behaviour†
1·7 (1·2–2·3); 1·9 (0·7)
1·6 (1·1–2·6); 1·9 (0·9)
1·9 (1·3–2·6); 2·0 (0·8)
1·7 (1·2–2·9); 2·0 (1·0)
0·013 (–0·171 to 0·197)
0·889
Depression and self-perception‡
3·1 (2·0–3·4); 2·8 (0·9)
2·6 (2·0–3·7); 2·7 (0·9)
3·1 (2·2–3·7); 2·9 (1·0)
2·6 (2·0–3·9); 2·8 (1·0)
0·014 (–0·297 to 0·324)
0·927
Embarrassment†
2·0 (1·7–2·7); 2·2 (0·8)
2·0 (1·3–2·7); 2·1 (0·8)
2·7 (1·7–3·0); 2·4 (0·8)
2·3 (1·7–3·0); 2·3 (0·9)
0·036 (–0·151 to 0·223)
0·706
8·9 (7·8–9·8); 8·5 (1·6)
9·2 (8·3–10·0); 8·7 (1·7)
8·4 (6·9–9·4); 7·8 (2·0)
9·3 (7·6–10·0); 8·4 (2·1)
–0·545 0·047 (–1·081 to –0·008)
Patient-centred outcomes§ Gastrointestinal quality of life¶
130·0 (113·0–41·0); 126·7 (18·8)
126·5 (109·0–39·0); 123·8 (20·2)
135·0 (115·0–48.0); 132·0 (20·6)
134·0 (120·0–146·0; 131·6 (20·5)
–1·300 (–5·168 to 2·568)
0·506
70·0 (45·0–90·0); 65·7 (27·4)
65·0 (40·0–85·0); 61·4 (28·4)
75·0 (47·5–90·0); 67·1 (27·7)
70·0 (45·0–90·0); 63·8 (29·0)
–1·854 (–6·992 to 3·284)
0·479
SF-36 scores (%) Physical functioning Role-physical functioning
50·0 (0·0–100·0); 46·4 (42·1)
25·0 (0·0–75·0); 36·4 (41·4)
62·5 (0·0–100·0); 54·4 (44·1)
25·0 (0·0–100·0); 46·2 (44·8)
1·113 0·826 (–8·866 to 11·092)
Bodily pain
60·0 (40·0–90·0); 61·3 (30·0)
57·5 (32·5–90·0); 58·2 (31·5)
67·5 (45·0–90·0); 64·3 (28·3)
67·5 (35·0–90·0); 62·1 (31·0)
–1·026 (–6·815 to 4·764)
0·728
General health
50·0 (35·0–70·0); 51·2 (23·4)
50·0 (30·0–70·0); 50·3 (23·8)
55·0 (30·0–75·0); 52·8 (24·6)
50·0 (35·0–70·0); 50·6 (23·9)
–0·158 (–4·749 to 4·433)
0·946
Vitality
45·0 (30·0–57·5); 43·9 (22·1)
50·0 (30·0–60·0); 42·7 (22·8)
50·0 (25·0–60·0); 45·6 (22·2)
50·0 (35·0–65·0); 46·7 (23·1)
–3·142 (–8·129 to 1·845)
0·215
Social functioning
62·5 (37·5–75·0); 58·4 (28·8)
62·5 (37·5–87·5); 59·3 (31·6)
75·0 (50·0–87.5); 66·4 (28·6)
62·5 (37·5–87.5); 60·6 (31·7)
5·209 (–0·740 to 11·157)
0·087
Role-emotional function
66·7 (0·0–100·0); 58·4 (28·8)
33·3 (0·0–100·0); 59·3 (31·6)
100·0 (0·0–100·0); 61·7 (45·3)
–4·815 (14·802 to 5·171)
0·343
Mental health
60·0 (44·0–76·0); 60·3 (21·0)
–0·509 (–4·831 to 3·814)
0·817
–0·017 (–0·078 to 0·044)
0·583
EQ-5D index score||
0·73 (0·62–0·85); 0·69 (0·27)
64·0 (48·0–76·0); 60·8 (21·6) 0·73 (0·62–0·85); 0·63 (0·34)
64·0 (48·0–84·0); 62·7 (25·1) 0·76 (0·62–0·85); 0·68 (0·28)
83·3 (0.0–100.0); 60·2 (44·1) 64·0 (52·0–76·0); 63·0 (21·4) 0·73 (0·56–0·85); 0·65 (0·34)
Data are unadjusted medians (IQR) and means (SD). PTNS=percutaneous tibial nerve stimulation. Sham=sham electrical stimulation. *Scores range from 0 (best) to 24 (worst). †Scores range from 1 (best) to 4 (worst). ‡Scores range from 1 (best) to 4·4 (worst). §Scores range from 1 (best) to 10 (worst). ¶Scores range from 36 (worst) to 180 (best). ||Scores range from –0·594 (worst) to 1 (best).
Table 2: Outcome measures
No significant differences were noted between PTNS and sham in the proportions of patients achieving more than 25%, more than 75%, and 100% reduction in mean episodes of faecal incontinence per week (figure 3, table 3). However, the mean total episodes of faecal incontinence per week in the PTNS group were significantly decreased at the study end compared with the sham group (difference in means –2·26, 95% CI –4·19 to –0·34; p=0·02, figure 3). Similarly, a significant reduction in the mean number of episodes of 6
urge faecal incontinence per week (–1·46, –2·69 to –0·22; p=0·02) was noted, but not passive faecal incontinence (adjusted difference –0·64, –1·67 to 0·40, p=0·23; figure 3). Reductions in episodes of faecal incontinence were already manifest after 6 weeks of treatment. No significant differences were noted in St Mark’s Incontinence Score between PTNS and sham (difference in means –0·05, 95% CI –1·03 to 0·94; p=0·93), or any significant differences in any domains of disease-specific
www.thelancet.com Published online August 18, 2015 http://dx.doi.org/10.1016/S0140-6736(15)60314-2
Articles
A
PTNS
Sham
Adjusted odds ratio (95% CI) for PTNS vs sham
p value
5·0
Odds ratios
4·0
3·0
≥25% ≥50% ≥75% 100% Reduction in episodes of faecal incontinence per week (%)
B 1·0 Change in mean number of episodes of faecal incontinence per week
46/102 (45%)
1·264 (0·730–2·190)
0·404
39/103 (38%)
32/102 (31%)
1·283 (0·722–2·281)
0·396
≥75% reduction
26/103 (25%)
17/102 (17%)
1·615 (0·770–3·388)
0·205
100% reduction
11/103 (11%)
7/102 (7%)
1·635 (0·592–4·514)
0·344
Table 3: Participants with reductions in episodes of faecal incontinence at follow-up (n=227)
1·0
0
PTNS (n=115)
Sham (n=112)
Pain at needle site
4 (3%)
3 (3%)
Bruising at needle site
2 (2%)
1 (<1%)
Altered sensation at needle site
1 (<1%)
0
Bleeding at needle site
0
2 (2%)
Altered sensation in toe
0
1 (<1%)
Data are n (%). All events were classed as mild by both the patient and the researcher. PTNS=percutaneous tibial nerve stimulation. Sham=sham electrical stimulation.
–1·0
Table 4: Treatment-related adverse events –2·0
–3·0
–4·0
–5·0
Total
Urge
Passive
Episodes of faecal incontinance
C 12 Episodes of faecal incontinence per week
51/103 (50%)
≥50% reduction (primary outcome)
Data are n/N (%). We calculated adjusted odds ratios, CIs, and p values based on imputed and adjusted data. PTNS=percutaneous tibial nerve stimulation. Sham=sham electrical stimulation.
2·0
0
≥25% reduction
Median Mean
10 8 6 4 2 0
Baseline
6 weeks PTNS
14 weeks
Baseline
6 weeks
14 weeks
Sham
Figure 3: Changes in episodes of faecal incontinence per week (A) Categorical reductions; data are odds ratios (95% CI) for PTNS versus sham at 14 weeks. (B) Adjusted differences in mean episodes of faecal incontinence per week (95% CI) for PTNS versus sham at 14 weeks. (C) Frequency of episodes per week throughout the study (PTNS: baseline median 6·0 [IQR 2·0–14·0], mean 9·9 [SD 11·2], 6 weeks median 4·0 [1·0–9·0], mean 6·1 [7·6], 12 weeks median 3·5 [1·0–10·0], mean 6·4 [7·6]; sham: baseline median 6·9 [IQR 2·5–16·0], mean 10·4 [SD 10·9], 6 weeks median 6·0 [1·0–14·0], mean 9·6 [11·5], 12 weeks median 4·8 [1·5–12·8], mean 9·1 [10·7]). PTNS=percutaneous tibial nerve stimulation. Sham=sham electrical stimulation.
FIQOL or generic SF-36 quality of life measures (appendix). Similarly no significant differences between groups in the GIQOL score (–1·30, –5·17 to 2·57; p=0·51) or the EQ5D index scores (–0·02, –0·08 to 0·04; p=0·58) were noted. A significant improvement in patient-reported outcomes was recorded for PTNS compared with sham (difference in means –0·55, 95% CI –1·08 to –0·01; p=0·05), but not for patients’ global impression of success (0·81, –0·06 to 1·67; p=0·07). Of patients who used loperamide at baseline, most (33 (67%) of 49 in PTNS and 32 (84%) of 38 in sham) reported no change in use at study end. Similar proportions in each group (2 [4%] in PTNS vs 2 [5%] in sham) reported increasing loperamide use; however, a higher proportion of patients reduced their loperamide use in the PTNS group compared with the sham group (14 [29%] vs 4 [11%]). In the PTNS group, 57 (54%) of 107 patients thought they had received PTNS and 48 (46%) of 107 thought they had received sham. In the sham group, 32 (31%) of 103 patients thought they had received PTNS and 71 (69%) of 103 thought they had received sham. Most patients in both groups reported no change in use of incontinence pads, and only a few in each group had slight or substantial improvements in urinary symptoms (appendix); no differences were noted between groups. The per-protocol analysis included 197 patients, and did not change interpretation of the results (for the primary outcome of a 50% or greater reduction in episodes of faecal incontinence per week OR 1·27, 95% CI 0·67–2·34; p=0·45), and neither did any of the other subgroup or
www.thelancet.com Published online August 18, 2015 http://dx.doi.org/10.1016/S0140-6736(15)60314-2
7
Articles
sensitivity analyses (appendix). There was also no significant centre effect (appendix). No serious adverse events related to treatment were reported in the trial. Seven mild, related adverse events were reported in each treatment group, mainly pain at the needle site (four in PTNS, three in sham (table 4).
Discussion PTNS did not result in a significant treatment response based on the primary endpoint. However, PTNS did result in a significant reduction in the mean number of total episodes of faecal incontinence per week and mean number of episodes of urge faecal incontinence per week, compared with sham. Although this result is significant, the clinical interpretation of these findings is uncertain, since a reduction from a mean of 6·0 (IQR 2·0–14·0) episodes of faecal incontinence per week to 3·5 (1·0–10·0) might or might not be helpful to patients. The clinical significance of a seemingly beneficial effect of PTNS compared with sham on urge versus passive faecal incontinence is also complex, because only a few (23% in the trial) patients had isolated urge faecal incontinence. Although PTNS also resulted in an improvement in the patient-centred outcomes (a short-form derivative of the validated ICIQ-B),13 this result was not supported by other outcome measures: no significant improvements were noted for summative symptom scores, disease-specific, or generic quality-of-life measures. The results of our trial are contrary to results from published studies of PTNS including six case series,7 one small single-centre, randomised single-blind trial (PTNS vs transcutaneous tibial nerve stimulation vs sham),8 one comparative case-matched study (PTNS vs sacral nerve stimulation)21 and one prospective clinical audit comparing PTNS and sacral nerve stimulation.22 These showed 63–82% response rates with the same primary outcome, which are substantially higher than the 38% reported here. This disparity is probably because of differences in trial design. The inclusion of a sham group helps to negate the effects of natural change in disease status over time, and controls for the well recognised effect of nurse-led face-to-face interventions.23 Placebo responses are disproportionately high in patients with chronic debilitating gastrointestinal disorders, because of a high level of expectation;24 however, our study correctly estimated the response (estimated 35% vs 31% shown). Also, without the rigour of an independently managed trial, bowel diaries can also introduce bias if they are unmasked and subject to investigator interpretation.25 The effect size seen in the PTNS group is nearer to that reported by a recent pilot randomised study (n=40) of PTNS versus sacral nerve stimulation, in which 47% of patients had treatment success in the PTNS group at 3 months.26 These results are similar to those from a large randomised trial of transcutaneous tibial nerve stimulation versus sham electrical stimulation for the same indication, 8
for which no superiority of transcutaneous tibial nerve stimulation was noted.24 Our study had several limitations that might have contributed to the negative outcome. The continuing dilemma of no perfect or universally accepted outcome measure for faecal incontinence is well documented.25 However, for the same reasons as other pivotal, therapeutic faecal incontinence trials,6,17 and to allow comparisons to be drawn between treatment modalities, the 50% or greater reduction was chosen. Quality assurance measures aimed to minimise variation in intervention quality but this could have contributed to differences from previously published reports in single specialist centres.7,27 However, although disparity existed in the proportions of centre recruitment (range 1–45 patients), a sensitivity analysis excluding low-volume centres did not change the primary outcome. Additionally, no significant centre effect was recorded in analysis of the primary outcome. Patient selection is also important. All patients received previous conservative therapy but this was not formally standardised, showing that a consensus has not been reached on what constitutes a minimum UK standard. Thus, a pragmatic view was adopted to balance some heterogeneity in baseline characteristics (eg, degree of treatment refraction) with exclusion of other specific patient subgroups that might have differing response rates. Some efficacy was shown in the reduction of episodes of urge faecal incontinence, and this has been previously documented7,27 and is akin to the approved indication for PTNS to treat overactive bladder.10 However, to determine whether the targeted selection of patients with this baseline characteristic (or others) would lead to a different conclusion would need validation from an appropriately designed study. Other exclusions were required for patients in whom PTNS was contraindicated. Overall, we acknowledge that the results can only be judged as valid for the population studied at the point of intervention in the algorithm of care. For similar pragmatic reasons, use of antidiarrhoeal drugs was not prohibited (because of concerns that unrecorded use would continue). Almost three times as many patients in the PTNS group reduced loperamide compared with the sham group. However, post-hoc χ² testing showed only weak evidence of a difference in loperamide use (p=0·06). Although this result could have confounded the effect size, it would have been unlikely to affect the overall trial conclusion. In conclusion, CONFIDeNT supports the importance of well designed, randomised, controlled trials to define the clinical benefit of treatments for faecal incontinence. On the basis of the evidence presented here, however, to recommend PTNS therapy for the patient population studied here would be difficult. Within the context of audit or research,28 PTNS is already embedded within recommendations for management of faecal incontinence in many centres in the UK and European Union. Further studies are required to establish the efficacy of PTNS in
www.thelancet.com Published online August 18, 2015 http://dx.doi.org/10.1016/S0140-6736(15)60314-2
Articles
the long-term and within specific subgroups; eg, those with urgency. Health-service research could also establish how PTNS promotes or contributes to other mixed packages of nurse-led care. Contributors CHK researched and did the study design with EJH, SAB, NS, CN, PRO’C, and SE, acquired the funding, interpreted the data with EJH, SAB, and NS, wrote the report with EJH, SAB, NS, CN, PRO’C, and SE, and supervised a research student. EJH obtained the data and analysed the data with SAB and SE. SAB did the figures, and NS managed the trial. The CONFIDeNT Study Group Chair of Data and Safety Monitoring Committee: Dion Morton Aintree University Hospitals NHS Foundation Trust: Paul Skaife, Sandra Blythin, Rachel Smith, Ching Way Medical Centre: Pasquale Giordano, Salvatore Lamberti Sandwell and West Birmingham Hospitals NHS Trust: Katherine Gill, Julie Collie, United Lincolnshire Hospitals NHS Trust: Pradeep Aggarwal, Kimberley Netherton, University Hospital of South Manchester NHS Foundation Trust: Karen Telford, James Nicholson, Claire Molyneux, Sheffield Teaching Hospitals NHS Foundation Trust: Steve Brown, Lynne Smith, University College London Hospitals NHS Foundation Trust: Anton Emmanuel, Ahsan Alam, Guy’s and St Thomas’ NHS Foundation Trust: Alexis Schizas, Emma Collins, Nottingham University Hospitals NHS Trust: Charles Maxwell-Armstrong, Rudra Maitra, Poole Hospital NHS Foundation Trust: Andrew Clarke, Steve Perring, Emma Jones, University Hospital Southampton NHS Foundation Trust: Karen Nugent, Janet Hicks, Leeds Teaching Hospitals NHS Trust: Dermot Burke, Clare Liddell Hull and East Yorkshire Hospitals NHS Trust: Graham Duthie, Philip Waudby, North West London Hospitals NHS Trust: Carolynne Vaizey, Sherill Tripoli, University Hospitals of Leicester NHS Trust: Justin Yeung University Hospitals Bristol NHS Foundation Trust: Paul Durdey. Declaration of interests We declare no competing interests. Acknowledgments The trial was under the auspices of the Chief Investigator and the Pragmatic Clinical Trials Unit (PCTU), Barts and The London School of Medicine and Dentistry (sponsor: Queen Mary University of London). It was funded by the UK NIHR Health Technology Assessment (HTA) programme (HTA 09/104/16). The views and opinions expressed in this report are those of the authors and are not necessarily those of the HTA programme, NIHR, the UK National Health Service, or the UK Department of Health. We would like to acknowledge the following people in relation to the CONFIDeNT study: Deborah Gilbert (Patient and Public Involvement representative), Norman Williams (clinical adviser), Sandy Smith, Mike Waring, Lara Edwards, Anitha Manivannan, Glenn Poon, Syed Arafath, and Polly Jordan (the Data Management and Quality Assurance team), and Elaine Denny and Daniel Altmann (the Data and Safety Monitoring committee). References 1 Nelson R, Norton N, Cautley E, et al. Community-based prevalence of anal incontinence. JAMA 1995; 274: 559–61. 2 Whitehead WE, Borrud L, Goode PS, et al. Fecal incontinence in US adults: epidemiology and risk factors. Gastroenterology 2009; 137: 512–17. 3 National Institute for Health and Care Excellence. Faecal incontinence: the management of faecal incontinence in adults. NICE guideline CG 49. London: National Institute for Health and Care Excellence, 2007. 4 Brown SR, Wadhawan H, Nelson RL. Surgery for faecal incontinence in adults. Cochrane Database Syst Rev 2013; 7: CD001757. 5 Matzel KE, Stadelmaier U, Hohenfellner M, et al. Electrical stimulation of sacral spinal nerves for treatment of faecal incontinence. Lancet 1995; 346: 1124–27. 6 Wexner SD, Coller JA, Devroede G, et al. Sacral nerve stimulation for fecal incontinence: results of a 120-patient prospective multicenter study. Ann Surg 2010; 251: 441–49.
7
8
9
10
11 12
13
14
15
16 17
18
19
20
21
22
23
24
25 26
27
28
Horrocks EJ, Thin N, Thaha MA, et al. Systematic review of tibial nerve stimulation to treat faecal incontinence. Br J Surg 2014; 101: 457–68. George AT, Kalmar K, Sala S, et al. Randomized controlled trial of percutaneous versus transcutaneous posterior tibial nerve stimulation in faecal incontinence. Br J Surg 2013; 100: 330–38. Leroi AM, Siproudhis L, Etienney I, et al. Transcutaneous electrical tibial nerve stimulation in the treatment of fecal incontinence: a randomized trial. Am J Gastroenterol 2012; 107: 1888–96. Peters KM, Carrico DJ, Perez-Marrero RA, et al. Randomized trial of percutaneous tibial nerve stimulation versus sham efficacy in the treatment of overactive bladder syndrome: results from the SUmiT trial. J Urol 2010; 183: 1438–43. Vaizey CJ, Carapeti E, Cahill JA, et al. Prospective comparison of faecal incontinence grading systems. Gut 1999; 44: 77–80. Cotterill N, Norton C, Avery KNL, et al. Psychometric evaluation of a new patient-completed questionnaire for evaluating anal incontinence symptoms and impact on quality of life: the ICIQ-B. Dis Colon Rectum 2011; 54: 1235–50. Eypasch E, Williams JI, Wood-Dauphinee S, et al. Gastrointestinal Quality of Life Index: development, validation and application of a new instrument. Br J Surg 1995; 82: 216–22. Rockwood TH, Church JM, Fleshman JW, et al. Fecal Incontinence Quality of Life Scale: quality of life instrument for patients with fecal incontinence. Dis Colon Rectum 2000; 43: 9–17. Stewart AL, Hays RD, Ware JE Jr. The MOS short-form general health survey. Reliability and validity in a patient population. Med Care 1988; 26: 724–35. EuroQoL Group. EuroQol—a new facility for the measurement of health-related quality of life. Health Policy 1990; 19: 199–208. Graf W, Mellgren A, Matzel KE, et al. Efficacy of dextranomer in stabilised hyaluronic acid for treatment of faecal incontinence: a randomised, sham-controlled trial. Lancet 2011; 377: 997–1003. Enck P, Horing B, Weimer K, Klosterhalfen S. Placebo responses and placebo effects in functional bowel disorders. Eur J Gastroenterol Hepatol 2012; 24: 1–8. Carpenter JR, Goldstein H, Kenward MG. REAL COM-IMPUTE software for multilevel multiple imputation with mixed response types. J Stat Soft 2011; 45: 1–12. Kahan BC, Morris TP. Analysis of multicentre trials with continuous outcomes: when and how should we account for centre effects? Stat Med 2013; 32: 1136–49. Asari SA, Meurette G, Mantoo S, et al. Percutaneous tibial nerve versus sacral nerve stimulation for faecal incontinence: a comparative case-matched study. Colorectal Dis 2014; 16: O393–99. Hotouras A, Murphy J, Allison M, et al. Prospective clinical audit of two neuromodulatory treatments for fecal incontinence: sacral nerve stimulation (SNS) and percutaneous tibial nerve stimulation (PTNS). Surg Today 2014; 44: 2124–30. Pacheco-Lopez G, Engler H, Niemi MB, et al. Expectations and associations that heal: immunomodulatory placebo effects and its neurobiology. Brain Behav Immun 2006; 20: 430–46. Musial F, Klosterhalfen S, Enck P. Placebo responses in patients with gastrointestinal disorders. World J Gastroenterol 2007; 13: 3425–29. Vaizey CJ. Faecal incontinence: standardizing outcome measures. Colorectal Dis 2014; 16: 156–58. Thin NN, Taylor SJ, Bremner SA, et al. Randomised clinical trial of sacral and percutaneous tibial nerve stimulation in patients with faecal incontinence. Br J Surg 2015; 102: 349–58. Hotouras A, Thaha MA, Boyle D, et al. Short-term outcome following percutaneous tibial nerve stimulation (PTNS) for faecal incontinence: a single-centre prospective study. Colorectal Dis 2011; 14: 1101–05. National Institute for Health and Care Excellence. NICE Interventional Procedure Guideline (IPG395). London: National Institute for Health and Care Excellence, 2011.
www.thelancet.com Published online August 18, 2015 http://dx.doi.org/10.1016/S0140-6736(15)60314-2
9