BACKGROUND: We hypothesized that functional anal incontinence with no structur al explanation comprises distinct pathophysiologic subgroups that could be ident ified on the basis of the predominant presenting bowel pattern. METHODS: Consecu tive patients (n = 80) were prospectively grouped by bowel symptoms as 1) incont inence only, 2) incontinence +constipation, 3) incontinence +diarrhea, and 4) incontinence +alternating bowel symptoms. The Hopkins Bowel Symptom Questionnai re, the Symptom Checklist 90-R, and anorectal manometry were completed. RESULTS : Significant group differences were found between subcategories of incontinent patients on the basis of symptoms. Abdominal pain was more frequent in patients with altered bowel patterns. Patients with alternating symptoms reported the hig hest prevalence of abdominal pain, rectal pain, and bloating. Basal anal pressur es were significantly higher in alternating patients (P = 0.03). Contractile pre ssures in the distal anal canal were diminished in the incontinent only and dia rrhea groups (P = 0.004). Constipated patients with incontinence exhibited eleva ted thresholds for the urge to defecate (P = 0.027). Dyssynergia was significant ly more frequent in patients with incontinence and constipation or alternating b owel patterns. CONCLUSIONS: Distinct patterns of pelvic floor dysfunction were i dentified in patient subgroups with anal incontinence, based on the presence or absence of altered bowel patterns. Physiologic assessments suggested different p athophysiologic mechanisms among the subgroups. The evaluation of patients with fecal incontinence should consider altered bowel function. BACKGROUND: We hypothesized that functional anal incontinence with no structur albucket comprises distinct pathophysiologic subgroups that could be ident ified on the basis of the predominant presenting bowel pattern. METHODS: Consecu tive patients (n = 80) were prospectively grouped by bowel symptoms as 1 ) Incontinence + constipation, 3) incontinence + diarrhea, and 4) incontinence + alternating bowel symptoms. The Hopkins Bowel Symptom Questionnai re, the Symptom Checklist 90-R, and anorectal manometry were completed. RESULTS: Significant group differences were found between subcategories of incontinent patients on the basis of symptoms. Patients with alternating symptoms reported the hig hest prevalence of abdominal pain, rectal pain, and bloating. significantly higher in alternating patients (P = 0.03). Contractile pre ssures in the distal anal canal were Dyssynergia was significantly more in patients with incontinence and constipation or alternating b owel (P = 0.004). patterns. CONCLUSIONS: Distinct patterns of pelvic floor dysfunction were i dentified in patient subgroups with anal incontinence, based on the presence or absence of altered bowel patterns. Physiologic assessments suggesting different p athophysiologic mechanisms among the subgroups. The evaluation of patients with fecal incontinence should consider altered bowel function.