Rectovaginal fistula: Risk factors for failure after graciloplasty-A bicentric retrospective European study of 61 patients.

AIM: Graciloplasty (GP) is indicated in the case of recurrent rectovaginal fistula (RVF) after failure of previous local treatments. The aim of this study was to assess risk factors for GP failure performed for RVF.

METHODS: This is a retrospective study based on a prospective database on GP, coming from two expert centres.

RESULTS: Sixty-one patients undergoing a first GP for RVF (n = 51) or ileal-vaginal fistula after ileal pouch anal anastomosis (n = 10), with a mean age of 42 years (range 24-72), were analysed. After a mean follow-up of 56 ± 48 months (range 1-183), failure of GP (considered as persistent stoma and/or clinical RVF) was noted in 24/61 patients (39%). The failure rate was 43% (13/30) in the case of Crohn's disease, 38% (3/8) in the case of ileal-vaginal fistula after ileal pouch anal anastomosis for ulcerative colitis, 30% (3/10) in the case of obstetrical RVF, 33% (1/3) in the case of post radiotherapy RVF and 40% (4/10) for other causes (not significant). Two risk factors for failure of GP were found on univariate analysis: (1) absence of postoperative antibiotic prophylaxis-only 3/24 (13%) patients with failure of GP received postoperative antibiotic prophylaxis versus 18/37 (49%) patients with success of GP (P = 0.0053); (2) postoperative perineal infection-11/23 (48%) with failure of GP developed postoperative perineal infection versus only 4/37 (10%) patients with success of GP (P = 0.0021).

CONCLUSIONS: Failure of GP for RVF is observed in approximately 40% of the patients whatever the aetiology of the fistula. A reduced failure rate was associated with systematic postoperative antibiotic prophylaxis.