RESUMO
AIM: Precise information regarding the location of an anal fistula and its relationship to adjacent structures is necessary for selecting the best surgical strategy. Retrospective and cross-sectional studies were performed to determine predictive factors for recurrence of anal fistula from preoperative examination by three-dimensional endoanal ultrasound (3D-EAUS). METHOD: Patients in our tertiary centre and in a private centre specialized in proctology undergoing preoperative 3D-EAUS for cryptoglandular anal fistulae between 2002 and 2012 were included. A questionnaire was sent in September 2013 to assess the patient's condition with regard to recurrence. Variables checked for association with recurrence were gender, type of centre, previous fistula surgery, secondary track formation and classification of the fistula. RESULTS: There were 143 patients of whom 96 had a low fistula treated by fistulotomy, 28 a high fistula treated by fistulectomy and 19 a high fistula treated by fistulectomy combined with a mucosal advancement flap. The median duration of follow-up was 26 (2-118) months. The fistula recurred in 40 (27%) patients. Independent risk factors included the presence of secondary track formation [hazard ratio 2.4 (95% CI 1.2-51), P = 0.016] and previous fistula surgery [hazard ratio 1.2 (95% CI 1.0-4.6), P = 0.041]. Agreement between the 3D-EAUS examination and the evaluation under anaesthesia regarding the site of the internal opening, classification of the fistula and the presence of secondary tracks was 97%, 98% and 78%. CONCLUSION: The identification of secondary tracks by preoperative 3D-EAUS examination was the strongest independent risk factor for recurrence. This stresses the importance of preoperative 3D-EAUS in mapping the pathological anatomy of the fistula and a thorough search for secondary track formation during surgery.