To determine if pattern analysis of defecography can predict the responsiveness of biofeedback therapy in patients with chronic functional constipation.

Over a two-year period, 104 patients with chronic functional constipation underwent defecography and biofeedback therapy. Two blinded readers analyzed the defecographic findings and classified them into six types; I = normal defecation, II = hypertonic lower anal sphincter (poor anal opening due to a persistent contraction of the lower anal sphincter), III = dyskinetic puborectal sling (inadequate laxity of the puborectal sling), IV = spastic pelvic floor syndrome (persistent contraction of both the puborectal sling and the lower anal sphincter), V = unclassified (including paradoxical contraction of the anal sphincter), VI = anatomical obstruction. In addition, the degree of rectal contraction during defecation was scored (grade 0 to 3). After biofeedback therapy, the differences in the defecography patterns or rectal contractions between the two groups, the responsive or non-responsive group, were analyzed.

The defecograms revealed that the type IV of the spastic pelvic floor syndrome was most common (50 of 104 patients, 48%), followed by II (21/104, 20%), III (12/104, 11.5%), V (9/104, 9%) and VI (12/104, 11.5%). Biofeedback therapy showed a therapeutic response in 71 out of 104 patients (68%) but failed in 33 patients (32%). However, there were no significant differences in the defecographic pattern between the responsive and non-responsive groups (p=0.630). The defecograms revealed rectal contractions in 78 patients (75%) and moderate to vigorous contractions (more than grade 2) in 66 patients. Most of the biofeedback-responsive group showed rectal contractions (66 of 71 patients, 93%, p<0.001).

In patients with chronic functional constipation, there was no significant difference in the morphological patterns of the defecogram between the responsive and non-responsive biofeedback groups. However, the presence of rectal contractions during defecation was strongly associated with the therapeutic response after biofeedback therapy.