Clinical Trial: Effectiveness and Safety of a Novel Anal Insert Device for Treatment of Faecal Incontinence.

Nicholas R. Oblizajek: writing – review and editing, investigation. Joel G. Fletcher: investigation, writing – review and editing. Misha Gautam: investigation, writing – review and editing. Adil E. Bharucha: conceptualization, investigation, funding acquisition, writing – original draft, methodology, writing – review and editing, project administration, supervision. Ramesh Lamichhane: investigation, visualization, writing – review and editing, data curation, formal analysis. Kent R. Bailey: writing – review and editing, formal analysis. Lawrence A. Szarka: investigation, writing – review and editing, supervision, project administration.

This work was supported by Minnesota Medical Technologies.

This study was approved by the Mayo Clinic Institutional Review Board and was conducted at Mayo Clinic, Rochester, Minnesota. All participants provided written informed consent. All authors had access to the study data and have reviewed and approved the final manuscript. Eligible participants were aged ≥ 18 years, had FI for ≥ 6 months, experienced ≥ 4 episodes per month, and persistent symptoms despite conservative medical therapy as appropriate for FI [ 3 , 7 ]. Key exclusion criteria included a history of inflammatory bowel disease with active proctosigmoiditis, anorectal pathology (e.g., perianal abscess or fistula, faecal impaction, or clinically significant rectocele) or faecal impaction with overflow diarrhoea, rectal surgery within the past 6 months, ileoanal pouch, clinically significant acute or chronic illness affecting study outcomes, use of selected drugs (e.g., warfarin), suspected upper or lower gastrointestinal obstruction, or any other condition that, in the investigator's judgement, could harm participants or interfere with data interpretation. The StaySure anal insert device is a soft, flexible, single‐use, nonsterile device filled with mineral oil and petrolatum and coated with petrolatum USP (Figure  1A ). It is self‐inserted into the anal canal where a liquid‐filled bulb forms a seal at the anorectal junction to prevent FI (Figure  1B ). The liquid‐filled shaft fills the anal canal, allowing sphincter pressure to maintain the seal, while an external retainer flare prevents migration (Figure  1B ). It comes in 2 sizes of 10 and 13 mm diameter that weigh 14.1 and 25.1 g. Together, these features are designed to prevent FI. Device and Study Design. (A) The StaySure anal insert device includes a bulb, a shaft, and an external retainer. The liquid‐filled shaft fills the anal canal. Compression displaces liquid from the bulb to the shaft, maintaining the anal seal until defecation is convenient. (B) Patients self‐insert the device into the rectum, then remove the retainer. The inner bulb and external flange prevent both external and internal displacement. (C) The study comprised baseline, fitting, and treatment periods. The study included baseline, fitting, and treatment periods (Figure  1C ). In the original protocol, participants' ability to retain the device was assessed only at the start of the fitting period. The protocol was later amended to assess retainability earlier in the study, either at the screening visit or before participants began the baseline diary period. Retainability was assessed by determining whether participants could retain the smaller (10 mm) device while ambulating a short distance, and if necessary, the larger (13 mm) device. Participants received standardized in‐person training on device insertion, positioning, and removal. Training included verbal instruction, demonstration by study personnel, and supervised participant practice to confirm appropriate placement and comfort with device use. After the modification was approved, only participants able to retain a 10‐ or 13‐mm device completed a 4‐week baseline diary; others were withdrawn (Figure  2 ). Those with ≥ 4 episodes of moderate or full bowel movement (BM) FI advanced to fitting. Per FDA guidance, FI was defined as leakage of moderate amount of stool or a full BM, excluding staining. CONSORT Diagram. Figure shows the flow of participants through the study. Unable to retain device: 3/124 during screening, 3/98 during baseline, 2/71 during fitting. ITT indicates intention‐to‐treat; mITT cohort defined as participants who used the device for ≥ 7 days ( n  = 58). Results were similar to the ITT cohort. During fitting and treatment, patients wore the device continuously, removing it only for defecation, and replacing it afterward. Fitting began with the smallest retained size (Figure  2 ). Those who could only retain the 13‐mm device used it throughout the study. Participants who retained the 10 mm device began with that size. After 2 weeks, response and comfort were assessed. Participants with (1) ≥ 75% reduction in FI frequency or (2) < 75% reduction who were dissatisfied with device comfort were advanced to treatment, as they were considered unlikely to tolerate the 13 mm device. Those with < 75% reduction who were satisfied with comfort switched to the 13 mm device for the next 2 weeks, then chose the more comfortable size for treatment. Comfort was assessed with the Quebec User Evaluation of Satisfaction with Assistive Technology (QUEST 2.0) (scores of 1–2 = dissatisfied; 3–5 = satisfied). Participants recorded each BM and FI episode in a daily bowel diary throughout the study [ 15 ]. For each FI episode, participants recorded leakage volume (staining only, moderate amount, or full BM), consistency (solid or liquid, or both), and level of awareness prior to the episode (none, normal warning, or strong urge). Stool form was classified using the Bristol Stool Form (BSF) scale: hard (BSF 1–2), normal (BSF 3–5), and loose (BSF 6–7) [ 16 ]. During the 4‐week baseline and treatment periods, participants rated overall relief of FI symptoms weekly on a 7‐point scale: 0 (“completely relieved”), 1 (“considerably relieved”), 2 (“somewhat relieved”, 3 (“unchanged”), 4 (“somewhat worse”). 5 (“considerably worse”) and 6 (“significantly worse”). Scores were averaged within each period and relative change (%) was calculated as 100 × (Baseline‐Treatment)/Baseline) scores. QoL, anxiety, and depression were assessed at the end of the baseline and treatment periods using the Faecal Incontinence Quality of Life (FI‐QoL) scale [ 17 ], and the PROMIS (Patient Reported Outcomes Measurement Information System) Short Forms for Anxiety and Depression [ 18 ]. Ease of use, usability, and comfort were assessed using the QUEST instrument during the fitting/run‐in period and at the end of treatment [ 19 ]. Responses of “more or less”, “quite” or “very” satisfied were categorized as satisfied; “not satisfied at all” or “not very satisfied” were categorized as unsatisfied. Anorectal pressures and rectal sensation were measured using left‐lateral manometry (Manoscan; Medtronic Inc) and interpreted with sex‐ and age‐appropriate normal values [ 20 , 21 ]. A rectal balloon expulsion time < 60 s was considered normal [ 22 ]. Anoscopy was performed at baseline in all participants and at the end of the study in 28 participants. Twenty‐nine participants (49%) also underwent anorectal magnetic resonance imaging (MRI) [ 23 , 24 ]. The anal sphincters and puborectalis were imaged using an endorectal coil (MRInnervu; Medrad Inc.), followed by magnetic resonance proctography in the supine position after instilling ultrasound gel (180 mL) into the rectum. Interactive single‐shot fast spin‐echo images were obtained at rest, squeeze, and evacuation [ 23 , 24 , 25 , 26 ]. A patulous canal was defined as anterior–posterior anal mucosal separation along the full anal canal length [ 27 ]. A single radiologist (J.G.F.) evaluated sphincter and puborectalis abnormalities and assessed perineal descent and rectal evacuation using established normative criteria [ 23 , 25 , 26 , 28 ]. Outcomes were assessed at 1 month after initiating treatment. Throughout the trial, the primary treatment response outcome was defined as a ≥ 50% reduction in FI frequency, as compared with baseline. Secondary efficacy outcomes were a ≥ 75% reduction in FI frequency, frequency of urge and passive FI, duration for which defecation could be deferred, proportion of complete bowel movements, loperamide use, symptom relief score, FIQoL, and anxiety and depression scores. The sample size was determined based on a randomized controlled trial in which the mean (SD) number of weekly FI episodes over a 4‐week baseline was 2.64 (1.07) in the placebo group [ 29 ]. Assuming a mean baseline frequency of 2.6 episodes/week and a conservative SD of change (post–pre) of 1.3, a sample size of 13 participants provides 90% power (two‐sided paired t ‐test, α  = 0.05) to detect a 50% reduction (from 2.6 to 1.3 episodes/week). Therefore, a sample size of 30 participants per group (10‐ and 13‐mm inserts) is more than sufficient. For each participant, bowel diary outcomes were summarized for baseline and treatment (details in Supporting Information ). The proportion of complete BMs was calculated as complete BMs divided by total BMs. Days with FI and days with loperamide use were defined as days with ≥ 1 event. Missing diary values were imputed using the overall mean value from participants with non‐missing data for that endpoint. Primary analyses were conducted in the intention‐to‐treat (ITT) cohort, defined as all participants who used the device during the treatment period. Continuous variables were summarized as Mean (SD) and categorical variables as counts and percentages. Primary and secondary binary endpoints were summarized as proportions with exact binomial CIs. Quantitative diary endpoints were analysed using one‐sample t ‐tests on absolute change or percent relative difference. Because the ratio baseline/treatment urge FI episodes was skewed, 0.2 was added to each count to avoid division by zero, and analyses used a one‐sample t ‐test of the adjusted ratio (ln [No. of baseline urge FI episodes + 0.2/No. of treatment urge FI episodes + 0.2]). To assess whether observed changes were clinically meaningful, we estimated the minimum clinically important difference (MCID) using the widely accepted distribution‐based threshold of 0.48 times the baseline standard deviation for all BM and most FI variables in this cohort [ 30 , 31 , 32 ], supplemented by published FIQoL data in a cohort of 260 patients with FI [ 5 ], and published data for the PROMIS Anxiety and Depression instruments [ 33 , 34 ]. Logistic regression models were used to assess whether the number of moderate and full BM FI episodes were associated with the likelihood of treatment response at both the ≥ 50% and ≥ 75% improvement thresholds. Adverse events were summarized for all 124 enrolled participants. Symptom data from the fitting period were used only for device‐sizing decisions. All analyses were performed with SAS v9.4 statistical software (SAS Institute Inc).

Of 124 enrolled participants, 98, 71, and 61 entered the baseline, fitting, and treatment periods, respectively. Device retention/tolerance issues accounted for 10 exclusions (Figure  2 ). Eight participants were excluded because they were unable to retain the device during a one‐time assessment conducted at screening ( n  = 3), baseline ( n  = 3), or fitting ( n  = 2). The remaining two participants were excluded because they were uncomfortable with the device or with using it independently. All results are based on the ITT cohort of 59 participants (97%) who reported using devices during the treatment period (Figure  2 ). In the ITT cohort, participants used devices for a mean (SD) of 23.5 (5.8) days. Among 59 participants, 55 (93%) were women; 38 (64%) participants were aged < 65 years (Table  1 ). A total of 45 participants (76%) had 1 or more organic conditions that are associated with diarrhoea and/or anorectal sensorimotor dysfunctions: 23 (39%) had only 1, 18 (31%) had 2, and 4 (7%) had 3 risk factors (Table  1 ). Forty‐six percent had actual or potential obstetric trauma, including forceps‐assisted deliveries or possible anal sphincter injury, as reported in medical records or by the patient (tears unspecified or documented as 3rd or 4th degree). Prior procedures for FI included sacral nerve stimulation ( n  = 11), anal sphincteroplasty ( n  = 4), or anal injection of dextranomer ( n  = 2). Baseline demographic and clinical characteristics in the ITT cohort. Abbreviations: BMI, body mass index; FI, faecal incontinence; ITT, intention‐to‐treat; LARS, low anterior resection syndrome; MRI, magnetic resonance imaging; NA, not available. Data available for 57 participants. Number of participants with ≥ 1 risk factor in this category. Includes staining, moderate leakage, and full bowel movement. Anorectal location measured as perpendicular distance from pubococcygeal line. This is an imputed value while the number of moderate and full BM FI episodes are actual values. In the bowel diary, the mean (SD) daily stool frequency and BSF were 2.6 (1.4) and 4.1 (0.8), respectively. A total of 17 participants (29%) had > 3 stools daily and 43 (73%) had loose stools (BSF ≥ 5) at least 25% of the time. Based on the prespecified imputed diary analysis, participants had a mean (SD) of 5.8 (6.8) weekly moderate or full BM FI episodes. In the observed diary data, participants reported a mean of 4.2 (5.5) weekly moderate FI episodes, 1.0 (2.0) weekly full FI episodes, and 3.3 (2.8) weekly staining episodes. At baseline, FI involving a moderate amount of stool or a full BM occurred in 28.8% of BMs, whereas FI involving staining, a moderate amount of stool, or a full BM occurred in 48.7% of BMs. Of all FI episodes, 61% involved semi‐formed stools (BSF 5–7). The anal resting pressure alone, squeeze pressure increment alone, or both were reduced in 6 (10%), 17 (29%), and 12 (21%) participants, respectively (Table  1 ). One or more rectal sensory thresholds were increased, indicative of hyposensitivity in 15 participants, and decreased, indicative of hypersensitivity, in 10 participants. Among the 28 participants, 17 (61%) had a patulous anal canal. Clinically significant internal sphincter, external sphincter, or puborectalis injury was apparent in 6, 14, and 8 participants, respectively. Seven patients had reduced anorectal motion during squeeze and 11 had increased motion (i.e., increased perineal descent) during evacuation. The weekly FI frequency declined by 69% (SD 32%) from 5.4 (6.9) episodes at baseline to 2.1 (5.1) episodes during treatment ( p  < 0.001) (Table  2 ). This improvement was not significantly different between participants treated with 10‐ and 13‐mm devices. Compared to baseline, the frequency of FI declined by ≥ 5 0% in 45 participants (76%, p < 0 .001) (i.e., treatment responders) and by ≥ 75% in 32 participants (54%, p = 0 .302). Primary and secondary outcomes with StaySure anal insert device (intention‐to‐treat). 3.3 (2.4, 4.2) 69.2 (60.9, 77.7) 76.3 (45/59) (63.4, 86.4) 76.7 (23/30) (57.7, 90.1) 75.9 (22/29) (56.5, 89.7) 66.1 (39/59) (52.6, 77.9) 91.5 (54/59) (81.3, 97.2) 74.6 (44/59) (61.6, 85.0) 2.4 (0.5) Abbreviations: FI, faecal incontinence; FI‐QoL, faecal incontinence quality of life; PROMIS, Patient Reported Outcomes Measurement Information System. p value from one‐sided binomial exact test. p value from one sample t ‐test on absolute change. Duration of urgency deferment on % relative change. (ln [No. of baseline urge FI episodes + 0.2/No. of treatment urge FI episodes + 0.2]). Values meet or exceed the MCID. MCID = (0.485 × Baseline SD) or reference provided. During treatment, the number of days with FI and frequency of passive FI declined by 51% and 59% respectively ( p <  0.001) (Table  2 ). The number of weekly urge FI episodes declined from 2.0 (2.1) to 0.93 (1.1) ( p <  0.001). The mean proportion of complete BMs increased by an absolute value of 6.1% (95% CI, 1.6–10.5) during treatment vs. baseline ( p  = 0.008). In contrast, the composition of stool leakage ( p  = 0.182) and defecation deferral time were not significantly different between treatment and baseline. Twenty‐nine participants used loperamide at baseline and during treatment. The absolute reduction of days and tablets of loperamide used per week declined by 0.3 ( p  = 0.301) and 0.1 ( p  = 0.125), respectively from baseline to treatment. The symptom relief score declined from 3.5 (1.5) at baseline to 1.9 (1.1) during the treatment period, indicating improved symptoms ( p  < 0.001). Among the 56 participants who completed the QUEST instrument, satisfaction rates were as follows: dimensions (83%), weight (94%), ease of adjustment (96%), safety and security (87%), durability (96%), ease of use (94%), comfort (85%), and effectiveness (66%) (Table  3 ). Of 56 participants, 50 (89.3%) ranked effectiveness, 41 (73%) ranked comfort, and 35 (62.5%) ranked ease of use as the most important factors that influenced their satisfaction with the StaySure Anal insert Device on the QUEST instrument. These variables were not associated with age (Table  3 ). Comparison of satisfaction with device in younger and older participants. “Satisfied” calculated as subjects that reported “More or Less”, “Quite” or “Very” satisfied. “Unsatisfied” calculated as subjects that reported “Not satisfied at all” or “Not very satisfied.” p value calculated from two‐sided Fisher's Exact Test. Compared to baseline, all four subscores of the FI‐QoL were significantly greater, reflecting better QoL during the treatment period (Table  2 , Figure  3 ). The absolute change in PROMIS Anxiety and Depression T‐scores declined by 3.0 points ( p  = 0.01) and 2.6 points ( p  = 0.025). Comparison of FI‐QoL scores during the baseline and treatment periods. (A) The overall FI‐QoL score increased in all participants. (B–E) FI‐QoL subscores also increased during treatment versus baseline. FI‐QoL indicates faecal incontinence quality of life. The treatment‐associated improvement in several outcomes met or exceeded the MCID at 1 month (Table  3 ). This includes the primary outcome for the 10 mm device size (i.e., frequency of any FI) and several secondary outcomes (i.e., frequency of urge FI, number of days with FI, duration for which FI could be deferred, number of loperamide tablets, scores for symptom relief, FIQoL depression/self‐perception score, and PROMIS—Anxiety T‐Score). There were no serious adverse events. During the fitting period, 4 of 71 participants (5.6%) reported adverse events. One participant reported a single episode of device over‐insertion into the rectum. A nurse grasped the external flange and removed the device. One participant each had perianal pain and bleeding related to a flare of lichen sclerosis, abdominal cramps, and perianal soreness. There were no adverse events during the treatment period. Baseline frequency of large‐volume FI episodes was significantly inversely associated with treatment response in both logistic regression models. In this model, the logistic regression coefficient for the number of full BM episodes at baseline was −0.67 (SE 0.27, p =  0.13). The estimated probability of response (≥ 50% reduction in FI frequency) fell from 87% with no baseline large‐volume FI episodes to 77% with 1 episode and 47% with 3 episodes. The estimated probability of a ≥ 75% reduction in FI frequency fell from 65% with no baseline large‐volume FI episodes to 52% with 1 episode and 38% with 3 episodes. Thus, the likelihood of a treatment response decreased with increasing baseline large‐volume FI episodes but was not significantly associated with the frequency of moderate FI (data not shown). The anal resting pressure was not different between 50% responders (51 [24] mmHg) and non‐responders (47 [28] mmHg) ( p  = 0.52). The squeeze increment was not different ( p  = 0.84) between 50% responders (71 [55] mmHg) and non‐responders (71 [67] mmHg; p  = 0.84). Compared to the ITT cohort, participants who withdrew due to device‐related issues had lower anal resting pressure (26 [14] vs 50 [25] mmHg, p  = 0.02), whereas the anal squeeze increment did not differ significantly (108 [114] 71 vs. [57] mmHg, p  = 0.73). These findings are based on available data from 5 of 10 participants who withdrew because of device‐related issues and 58 of 59 participants in the ITT cohort (Figure  2 ).

These study participants are representative of the broader population of patients with moderate/severe FI refractory to medical and behavioural therapy [ 5 , 35 ]. Some participants had undergone anal dextranomer injection, sacral nerve stimulation, and/or anal sphincteroplasty, and 60% had reduced anal resting and/or squeeze pressures. Using detailed bowel diaries [ 15 , 36 ], the observed frequency of staining, moderate FI, and large FI was 8.5 episodes per week, which is comparable to prior studies: ~9.4 weekly episodes in the pivotal clinical trial of sacral nerve stimulation [ 37 ], a median of 15 episodes over 2 weeks in the pivotal trial of anal dextranomer injection [ 38 ], and ~9.9 weekly episodes in the Faecal Incontinence Treatment (FIT) trial of dextranomer versus anorectal biofeedback therapy for FI [ 9 ]. In the FIT Trial, moderate or full BM FI accounted for 49% of FI episodes [ 9 , 39 ] compared with 61% in this study. Nevertheless, the ≥ 50% responder rate at 1 month was higher in this study (76%) than the 3‐month response (47%) after biofeedback or dextranomer injection in the FIT trial [ 9 ]. Because the outcomes were only based on moderate or full BM FI, they are arguably more rigorous that prior studies that also considered staining FI. Even so, the 50% responder rate (76%) for moderate or full BM FI was comparable to the 79% rate at 1 month for the Eclipse vaginal bowel control device [ 14 ], and numerically greater than rates for any FI (i.e., staining, moderate, and severe) reported in other studies: 62% at 3 months for the Renew anal device [ 13 ], and 48% at 3 months for dextranomer [ 38 ]. However, these cross‐trial comparisons should be interpreted cautiously, as differences in patient populations, outcome definitions, follow‐up duration, and study design preclude direct comparisons and do not permit conclusions regarding comparative efficacy or superiority. On average, all 4 FI‐QoL sub scores increased significantly by 0.3–0.4 units at 1 month, comparable to improvements with biofeedback therapy and anal dextranomer injections at 3 months [ 9 ]. Global relief also improved. Most participants were satisfied with the device, which was well tolerated. The device prevented leakage without impairing evacuation. Indeed, the proportion of complete bowel movements increased during the study. For most outcomes, including FI frequency, days with FI, ability to defer defecation, global relief, and overall FI‐QoL, the treatment effects exceeded 0.485 × SD of the baseline value, a conservative estimate of the MCID [ 30 , 31 , 32 , 40 ]. Across diverse diseases and instruments, the MCID is remarkably consistent and approximates one‐half a standard deviation. However, although this distribution‐based approach is widely used and provides a standardized estimate of clinically meaningful change, it does not directly reflect patient‐perceived improvement. Consequently, distribution‐based MCIDs may not fully align with patient‐defined treatment success, which can vary substantially between individuals. Nevertheless, these findings indicate that the benefits were not only statistically significant but also clinically meaningful. The baseline frequency of large‐volume FI episodes was significantly predictive of the treatment response. Participants with more frequent large‐volume leakage at baseline were substantially less likely to achieve either ≥ 50% or ≥ 75% improvement. When expressed as a linear function of frequency of large‐volume leakage, the probability of a response declined progressively as the number of large‐volume leakage episodes increased. These findings suggest that patients with more severe FI may derive less benefit from the device, potentially because they have more severe diarrhoea and/or underlying anroectal dysfunction(s) that is less amenable to treatment. At the same time, clinically meaningful improvement remained possible among participants with moderate symptom burden, supporting the potential utility of the device across a range of FI severities. Because the model used the raw count of large‐volume leakage episodes, the estimated decline in response probability becomes particularly steep at higher values. Although alternative modelling approaches (e.g., logarithmic transformation of baseline leakage frequency) may warrant exploration in future studies, the present analysis provides an intuitive clinical interpretation: increasing baseline large‐volume FI was consistently associated with lower likelihood of treatment success. A total of 8 of 10 participants who withdrew from the study before the treatment period did so because they were unable to retain the device. Among five of these eight participants for whom manometry data were available, mean anal resting pressure was 50% lower than in the ITT cohort. In a separate cohort of 81 patients with FI in whom resting pressure was measured using the same method, the mean (SD) anal resting pressure was 66 (25) mmHg and the 10th percentile value was 31 mmHg [ 21 ]. In another cohort of 49 patients with a patulous anal canal, the median anal resting pressure was 36 mmHg (IQR 23–58) [ 27 ]. These observations suggest that only a minority of patients with FI, likely fewer than 25%, have sufficiently low anal resting pressure to impair retention of this device. Only 4 participants (7%) experienced mild adverse effects compared with 43% and 23% rates of mild and moderate device‐related adverse events, respectively during the fitting and treatment periods with the Eclipse device. The StaySure device is simpler to use than the Eclipse device, which requires clinician fitting, vaginal insertion, balloon inflation after placement, and deflation prior to removal. Device displacement into the rectum did not occur compared with 24% of patients using the Renew anal device [ 13 ]. In a phantom experiment, the StaySure device was visible on radiography (unpublished observations). Taken together, these findings support the use of the StaySure anal insert in patients with FI who are refractory to medical and behavioural therapy [ 7 ]. Although participants used the device daily, some patients prefer intermittent use—for example, during travel, social events, or when toilet access is uncertain. Patients who find the device comfortable, effective, and easy to use may elect to continue long‐term use. This device is approved in several European countries and by the United States FDA. Because the device physically occludes the anal canal, its use may be particularly beneficial in patients with passive FI and/or a patulous anal canal that permits stool leakage, provided the canal is not so patulous that the device is expelled [ 27 ]. Although anal resting and squeeze pressures did not predict the response to therapy, patients unable to retain the device—presumably reflecting a more patulous canal—were excluded before the treatment period. We acknowledge some limitations. The treatment period was limited to 4 weeks. As with studies of the Renew [ 13 ] and Eclipse devices [ 14 ], this study was uncontrolled because an acceptable placebo device is unavailable. However, improvements for most outcomes exceeded the MCID. Conducting a methodologically rigorous RCT comparing this device with standard care would be challenging [ 15 ]. In this study, FI frequency declined by 50% in 76% and by 75% in 54% of participants at 4 weeks. We acknowledge that the observed improvement may partly reflect a placebo response, regression to the mean, or behavioural modifications perhaps prompted by tracking of symptoms in a diary. However, these response rates are substantially greater than the 50% responder rate (i.e., 32% [ 29 ] and 53% [ 15 ]) and 75% responder rate (13%) [ 15 ] in patients with moderate/severe FI who were treated with placebo. The 4‐week treatment period limits assessment of long‐term adherence, durability of benefit, and safety. Accordingly, our findings should be interpreted as reflecting short‐term outcomes only. Although no serious device‐related adverse events or mucosal complications were observed during the study period, longer‐term use could potentially be associated with declining adherence, device fatigue/intolerance, or mucosal effects that were not assessable in this trial. Additional longer‐duration studies are necessary to evaluate sustained efficacy, long‐term tolerability, adherence, and safety. To conclude, this StaySure device appears to be a safe and effective option for patients with FI refractory to medical and behavioural therapy.

One in 12 adults has faecal incontinence (FI) [ 1 ], which can impair quality of life (QoL) and poses a considerable health care burden [ 2 , 3 ]. Bowel disturbances, predominantly diarrhoea and anorectal sensorimotor dysfunctions—primarily anal weakness, which is commonly due to irreversible obstetric injury—are the main risk factors for FI [ 4 , 5 , 6 ]. Stepwise management, starting with dietary changes, bowel modifiers, and pelvic floor exercises and progressing as needed to anorectal biofeedback, sacral neuromodulation, or anal dextranomer injection is recommended; rarely, a colostomy or anal sphincteroplasty is required [ 3 , 6 , 7 ]. Among patients with severe FI (average of 9 episodes/week), only 15% responded to medical and behavioural therapy, defined as a ≥ 75% reduction in FI frequency [ 8 ]. Among non‐responders, approximately 28% each responded to subsequent treatments, including anorectal biofeedback and dextranomer injection [ 9 ]. Additional safe and effective treatments for FI are needed [ 7 ]. Anal barrier devices (plugs) are infrequently used to manage FI. Early devices were uncomfortable and associated with hypersensitivity [ 10 ]. In 2012, the US Food and Drug Administration (FDA) approved the Renew (Renew Medical UK Limited) anal barrier, which has a rectal volume of < 1 mL versus 8–20 mL for older devices [ 11 , 12 ]. In the pivotal Renew study, FI frequency declined by ≥ 50% in 77% of 73 completers and 62% of 91 intention‐to‐treat (ITT) participants [ 13 ]. This device is not currently marketed in the United States. The FDA‐approved Eclipse (Laborie Medical Technologies Corp) device, which has an inflatable vaginal balloon that compresses the rectum, is available in the US [ 14 ]. In an open‐label study of 110 women with FI, 65 were successfully fitted with the Eclipse device, and 79% achieved a ≥ 50% reduction in FI at 1 month. This study aimed to evaluate the effectiveness and safety of the StaySure anal insert device in patients with FI.

Adil E. Bharucha, MBBS, MD, jointly holds a patent for the anal insert device used in this study. Dr. Bharucha and Mayo Clinic have contractual rights to receive royalties from the licensing of this technology. Mayo Clinic holds equity in the company to which the technology is licensed. This research has been reviewed by the Mayo Clinic Conflicts of Interest Review Board and is being conducted in compliance with Mayo Clinic Conflict of Interest policies. The following authors have no conflicts of interest to declare: Ramesh Lamichhane, Joel G. Fletcher, Nicholas R. Oblizajek, Misha Gautam, Kent R. Bailey, and Lawrence A. Szarka.

Table S1: Summarizing Data.