Randomized, Placebo-Controlled Trial of Biofeedback for the Treatment of Rumination

OBJECTIVES: We previously demonstrated that rumination is produced by an unperceived, somatic response to food ingestion, and we developed an original biofeedback technique based on electromyography (EMG)-guided control of abdomino-thoracic muscular activity. Our aim was to demonstrate the superiority of biofeedback vs. placebo for the treatment of rumination. METHODS: Randomized, placebo-controlled trial performed in a referral center. Consecutive patients who fulfilled the Rome III criteria for rumination (18 women, 6 men; 19-79 years age) were selected and all included in the study; 1 patient assigned to placebo withdrew because of an unrelated accident. Abdomino-thoracic muscle activity after a challenge meal was recorded by EMG. The patients in the biofeedback group were shown the signal and instructed to control muscle activity, whereas the patients in the placebo group were not shown the signal and were given oral simethicone. Each patient underwent 3 sessions over a 10-day period. Main outcome: number of rumination events as measured by questionnaires for 10 consecutive days before and after intervention. RESULTS: Patients on biofeedback (n = 12) but not on placebo (n = 11) effectively learned to reduce intercostal activity (by 51 +/- 6% vs. 10 +/- 7% increment on placebo; P < 0.001) and anterior wall muscle activity (by 52 +/- 4% vs. 9 +/- 2% increment on placebo; P < 0.001). Biofeedback treatment resulted in a 74 +/- 6% reduction in rumination activity (from 29 +/- 6 before to 7 +/- 2 daily events after intervention) vs. 1 +/- 14% on placebo; P = 0.001 (from 21 +/- 2 before to 21 +/- 4 daily events after intervention). CONCLUSIONS: Rumination can be effectively corrected by biofeedback-guided control of abdomino-thoracic muscular activity.