Use of endoanal ultrasound in patients with rectovaginal fistulas

The purpose of our study was to define the role of endoanal ultrasound in the evaluation and management of patients with rectovaginal fistula.

A retrospective review was performed of all patients with rectovaginal fistula who were evaluated by endoanal ultrasound at Barnes-Jewish Hospital at Washington University from 1992 to 1997.

Twenty-five females underwent endoanal ultrasound before rectovaginal fistula repair. Mean age was 34 years. Rectovaginal fistulas were caused by obstetric trauma (19 patients; 76 percent), cryptoglandular disease (5 patients; 20 percent), and Crohn's disease (1 patient; 4 percent). Previous rectovaginal fistula repair had been performed in ten patients (40 percent). A history of anal incontinence was present in ten patients (40 percent). Rectovaginal fistula location was above (15 patients), at (7 patients), or below (3 patients) the dentate line. Rectovaginal fistula size was <5 mm (19 patients; 76 percent) or >5 mm (6 patients; 24 percent). Anal manometry revealed decreased sphincter pressures (resting or squeeze) in 12 patients (48 percent). Pudendal nerve latency was abnormal in three patients (9 percent). Endoanal ultrasound identified the rectovaginal fistula in 7 patients (28 percent) and an anterior sphincter defect in 23 patients (92 percent). At surgery sphincter injuries were identified in 23 patients (92 percent). Treatment was either sliding flap repair with anal sphincter reconstruction (22 patients; 88 percent) or sliding flap repair alone (3 patients; 12 percent). Repair of the rectovaginal fistula was successful in 23 patients (92 percent). Complications occurred in 11 patients (44 percent): two recurrent rectovaginal fistulas, five infections, two skin separations, one ectropion, and one hematoma. The two patients with recurrent rectovaginal fistula had prior repairs, and both were subsequently repaired successfully. Of the 11 patients with preoperative anal incontinence, 6 patients (54 percent) were continent and 2 (18 percent) improved after surgery. Cause, size, location, and previous repair of fistula had no effect on final outcome.

Noncontrast endoanal ultrasound was not useful in imaging rectovaginal fistulas and cannot be recommended as a diagnostic or screening tool for the identification of a rectovaginal fistula. However, we recommend that endoanal ultrasound be performed preoperatively in all patients with known rectovaginal fistulas to identify and map occult sphincter defects. Concomitant anal sphincter reconstruction should be considered strongly in patients with rectovaginal fistula and an endoanal ultrasound-documented sphincter defect.