Purpose: Although anismus has been considered to be the principal cause of anorectal outlet obstruction, it is doubtful whether contraction of the puborectalis muscle during straining is paradoxical. The present study was conducted to answer this question.
Methods: During the first part of the study, we retrospectively reviewed 121 patients with constipation and/or obstructed defecation (male:female, 10/111; median age, 51 years). All of these patients underwent electromyography (EMG) of the pelvic floor and the balloon expulsion test (BET) in the left lateral position. Evacuation proctography was performed in all of these patients in the sitting position. Both the posterior anorectal angle and the central anorectal angle were measured. EMG and BET were also performed in ten controls (male:female, 4/6; median age, 47). In 147 patients with fecal incontinence (male:female, 24/123; median age, 58) only EMG activity was recorded. Criteria for anismus during straining were increase or insufficient (<20 percent) decrease of EMG activity, failure to expel an air-filled balloon on BET, and decrease or insufficient (<5 percent) increase of anorectal angle on evacuation proctography. Between June 1994 and March 1995, we conducted a second prospective study in a consecutive series of 49 patients with constipation and/or obstructed defecation and 28 patients with fecal incontinence. Both groups were compared with 19 control subjects. In this study, all three tests were performed. EMG and BET were performed both in the left lateral position and in the sitting position.
Results: The retrospective study was undertaken by comparing the constipated patients with the incontinent patients and the controls, and the anismus detected by EMG was found in, respectively, 60, 46, and 60 percent. Failure to expel the air-filled balloon was observed in 80 constipated patients (66 percent) and in 9 control subjects (90 percent). Based on posterior anorectal angle and central anorectal angle measurements, anismus was diagnosed in, respectively, 21 and 35 percent of constipated patients. In the prospective study, none of the tests showed significant differences regarding the prevalence of anismus between the two subgroups of patients and the control subjects. The prevalence of anismus only differed between constipated and incontinent patients when the diagnosis was based on BET in the sitting position (67 vs. 32 percent; P < 0.005). Our study shows that contraction of the puborectalis muscle during straining is not exclusively found in patients with constipation and/or obstructed defecation. The three tests most commonly used for the diagnosis of anismus showed an extremely poor agreement.
Conclusion: Based on these findings, we doubt the clinical significance of anismus.