A randomized, controlled trial of fibrin glue vs. conventional treatment for anal fistula

PURPOSE: Fibrin glue is a novel treatment for anal fistulas and possesses many advantages in the treatment of difficult high fistulas. Fibrin glue treatment is simple and repeatable; failure does not compromise further treatment options; and sphincter function is preserved. We aimed to compare the outcomes of patients with low and high anal fistulas randomly assigned to either fibrin glue or conventional treatment. METHODS: Patients with simple fistulas (low fistulas) and complex fistulas (high, Crohn's, and low fistulas with compromised sphincters) were randomly assigned to either fibrin glue or conventional treatment (fistulotomy or loose seton insertion with or without subsequent advancement flap). Patients with rectovaginal fistulas and anal fistulas associated with chronic cavities, acute sepsis, and side branches were excluded. The primary end point was fistula healing. Secondary end points were complications, changes in preoperative continence score, changes in maximum resting and squeeze pressure, satisfaction scores, and pain scores and time off work (simple fistulas only). RESULTS: Patients in the fibrin glue and conventional treatment arms were well matched for gender, median age, duration of fistula symptoms, and follow-up. Fibrin glue healed three (50 percent) of six and fistulotomy seven (100 percent) of seven simple fistulas (difference, 50 percent; confidence interval, 10 to 90 percent; P = 0.06, Fisher's exact probability test). There was no change in baseline incontinence score, maximum resting pressures, or squeeze pressures between the study arms. Return to work was quicker in the glue arm, but pain scores were similar and satisfaction scores higher in the fistulotomy group. Fibrin glue healed 9 (69 percent) of 13 and conventional treatment 2 (13 percent) of 16 complex fistulas (difference, 56 percent; 95 percent confidence interval, 25.9 to 86.1 percent; P = 0.003, Fisher's exact probability test). There was no change in baseline incontinence score, maximum resting pressures, or squeeze pressures in either study arm. Satisfaction scores were higher in the fibrin glue group. CONCLUSIONS: No advantage was found for fibrin glue over fistulotomy for simple fistulas, but fibrin glue healed more complex fistulas than conventional treatment and with higher patient satisfaction.