Fifteen subjects (ages 5-33 years) with meningomyelocele and fecal incontinence underwent anorectal manometry followed by biofeedback conditioning of the external anal sphincter or nearby gluteal muscles. Seven of the 15 subjects undergoing biofeedback had improvement, defined as a 75% or greater decrease in the frequency of soiling, with a mean follow-up period of 23.1 months (range, 8-30 months). The seven subjects who improved with biofeedback had significantly lower thresholds of rectal sensation (smallest volume of distension perceived) than did the eight nonresponders; all responders had a sensory threshold less than or equal to 20 ml, whereas five of eight nonresponders had thresholds exceeding this level. Twenty-one children (ages 4-17 years) also underwent anorectal manometry, but not biofeedback conditioning, in order to evaluate the relationship of rectal sensation and peripheral neurological deficits. Fifteen of the 36 subjects with meningomyelocele had absent rectal sensation or thresholds exceeding 20 ml; there was no correlation of anorectal manometric parameters and peripheral sensorimotor levels. Response to biofeedback by children with meningomyelocele strongly correlates with rectal sensory thresholds which do not correlate with peripheral sensorimotor levels. If children with meningomyelocele and fecal incontinence are motivated and have adequate strength and coordination of the muscles of the lower extremities, anorectal manometry is the most accurate test to identify those who may benefit from biofeedback conditioning.