Sensory Retraining Is Key to Biofeedback therapy for Formed Stool Fecal Incontinence

Chiarioni G, Bassotti G, Stegagnini S, Vantini I, Whitehead WE. The American Journal of Gastroenterology. 2002;97(1):109-117.


Pelvic Physical Therapy Distance Journal Club; February 6, 2013

Prepared by Pamela Downey, PT, DPT, WCS, BCB-PMD & Kendra Boyce, SPT ’13 University of Miami School of Medicine, Department of Physical Therapy

Primary Aim: To evaluate the efficacy of biofeedback in patients with severe solid stool incontinence, assess the importance of sensory training and strength training to the success of biofeedback, and to identify patient characteristics that predict who will respond to biofeedback.

Background: Fecal incontinence occurs in at least 2% of the population and is a major cause of social and psychological disability. Although biofeedback is often used, most studies fail to control for nonspecific treatment effects such as the benefits of education and good clinical management. The mechanism by which biofeedback training effects are achieved is controversial: some argue that the most important ingredient is sensory discrimination training while others believe that biofeedback works primarily by strengthening the EAS muscles.

Subjects: 24 patients (17 women, 7 men) referred to the Division of GI Rehabilitation of the University of Verona Valeggio sul Mincio Hospital between January, 1997-January, 1998 with complaints of solid stool incontinence. Ages ranged from 31-75 y/o (median= 54.5yrs). All patients had been prescribed conventional medical treatment (constipating agents, fiber supplementation, regular enemas) by the referring physician for at least 1 month and had failed to improve. Previous evaluation also included colonoscopy in all patients

Inclusion criteria: solid stool incontinence for at least 12 months, failure of conventional medical treatment, frequency of solid stool incontinence at least 1x/week during a 1-month run-in period

Methods: All patients underwent a careful anorectal exam both in the resting condition and while straining down.

Anorectal Manometry: With the pt. lying in the left lateral position, a six-lumen water-perfused PVC anorectal catheter was inserted into the rectum. After allowing the pressures to stabilize for 5 mins, the maximum resting and maximum squeeze pressures were determined by using the station pull-through technique. At each station the basal pressure was recorded for 1 minute, and the pt. was then asked to squeeze the sphincter as hard as he/she could for 10s.

After removal of the catheter and a 5 minute rest, a 3-balloon Schuster probe was inserted into the rectum, and the 5 cm-long rectal balloon was positioned with its distal edge 5cm from the anal verge. Rectal sensation for small volumes was assessed by rapid air distension of the rectal balloon @ 40ml/s, with volume decreasing from 60 ml to 10 ml in 10-ml decrements. The first sensation threshold was the smallest volume that consistently produced a report of sensation.

The rectoanal inhibitory reflex (RAIR) threshold was defined as the smallest volume of rectal distension producing a pressure drop of at least 3 mm Hg in the upper anal balloon pressure. Additional evaluation of the EAS function was provided by 1) the threshold of EAS contraction and 2) squeeze duration.

Rectal compliance and the urge threshold were evaluated by a cylindrical latex balloon (condom type), 10 cm in length, mounted on a stiff 0.3 mm-diameter PVC tube. The balloon was inflated with air in 50-ml increments up to the maximum tolerated volume. An inflation volume of 100 ml was tolerated well by all patients.

Defecography

With the subject lying on his/her left side, 20 ml of liquid barium was inserted into the rectum to coat the mucosa. Up to 200 ml of semisolid paste was introduced into the rectum. The injection was stopped as soon as the patient started to leak. The subjects were then asked to sit on a wooden commode. A fluoroscopic video was recorded to detect the presence of an open anal canal at rest or presence of stress anal incontinence. The patients were then asked to empty their bowel as completely as possible and were allowed up to 60 seconds. Two films, one at rest and one during defecation, were obtained for further evaluation. All patients were able to empty completely.

Intervention:

Biofeedback Training

Before biofeedback, all patients received a full explanation of the role of rectal sensation and EAS squeeze pressure in preventing incontinence. Treatment involved 3-4 training sessions: 1x/week for 3 weeks. Sensory discrimination training and sphincter strengthening training were both provided using the 3-balloon Schuster probe. Patients were trained to squeeze as soon as they perceived rectal distension. The initial distensions were large volumes, which were progressively reduced down to the sensory threshold. Intention was to decrease the sensation threshold to 10 ml of air and to have the patient squeeze within 2 seconds of distension. Patients were also trained to increase strength/duration of the EAS until they were able to contract for at least 20 seconds in response to each rectal distension using both visual and verbal reinforcement. When these techniques were mastered, both forms of feedback were gradually withheld until the same responses were consistently produced.

At home, each patient was told to pay attention to any sensation in the anorectal area and to respond to it by squeezing the anal sphincter muscles at that time. Additionally, each pt. was told to practice contracting the sphincter for 20-30 seconds, performing at least 20 squeezes/3x a day (with an assumed maximal attraction).

Primary Outcome: Treatment success was rated 3 months after the completion of initial training. A responder was defined as a patient who experienced restoration of continence or at least a 75% decrease in frequency of episodes of formed stool incontinence compared to before treatment. Subjects also were assessed through a second anorectal sensorimotor evaluation by both perfused catheter and balloon manometry 3 months after completion of initial training. Subjects took part in a 6-month and 12-month follow-up interview

Results:

• 17 of 24 of patients met the “responder” criterion

• Of the 17 responders, 13 of these patients were completely continent, 4 showed a 75% reduction in frequency of incontinence

• Frequency of solid stool incontinence decreased from 33.8 +4.7% at baseline to 11.6 +4.3 per month at follow up

• Total incontinence episodes decreased from 59.0 +6.2 per month to 23.0 + 8.0 per month

• Comparing baseline to 3-month follow up: significant improvements both in sphincter muscle strength and in sensory threshold

• 14 patients changed from abnormal to normal (10 ml) sensory thresholds

• Responders had lower threshold for first sensation at the end of treatment and lower thresholds for exhibiting an automatic sphincter contraction to rectal distension

• Responders all had final sensory thresholds of 20 ml or less

• EAS sphincter strength did not separate responders from non responders

• Baseline sensory thresholds were significantly lower in responders than in nonresponders (for both threshold for first sensation and threshold for urge to defecate)

• The threshold for the rectoanal inhibitory reflex and the threshold for automatic external anal sphincter contractions during sensory testing were also significant predictors of response to biofeedback

Interpretation:

Strengths:
• Evaluated by consistent best practice techniques (anorectal manometry, rectoanal inhibitory reflex threshold, rectal compliance, defecography)
• All pressure biofeedback sessions conducted by the same investigator
• Subjects only had 3 sessions; (feasible in PT practice)
• Researchers informed and accounted for subjects and their participation: bowel diary, etiology, treatment, reason for dropout, etc.

Weakness:
• Small sample size
• Self-reported compliance
• Published in 2002
• Only assessed biofeedback strength + sensory tx combination (would be interesting to compare a biofeedback-only group vs. biofeedback +sensory group)
• No MMT performed digitally to assess EAS vs. levator ani musculature

Conclusions:
• Sensory training appears to be more relevant than strength training to the success of biofeedback
• Responders improved their 1st rectal sensation thresholds to 20ml of air
• Improvements in EAS squeeze pressure did not distinguish responders from non-responders

Implications for clinical practice:
• Study recommends sensory discrimination training should be incorporated into all biofeedback protocols

• May be possible to identify the patients who are most likely to respond by administering a sensory threshold test (those with sensory thresholds for 1st rectal sensation above 50ml of air are less likely to improve their sensation enough to recover incontinence)

Questions for Discussion
• Does manometry reflect deeper levator ani muscle functional strength? (eg. puborectalis). Furthermore, does it help with squeeze pressure, duration, and/or identification?
 Do you typically receive manometry reports on your patients?
• Use of perianal surface vs. internal surface vs 2-channel anal electrode (eg. Tries). Muap vs pressure systems.
• Do your current biofeedback protocols involve sensory training?
• Are the materials used for sensory training in this study (3-balloon Scuster probe) available to you at your clinic?
• What are some other techniques one can use in conjunction with biofeedback to increase likelihood of success?