Objective: To study the accuracy of endoanal ultrasound in pre-operative assessment of cryptoglandular anal fistulas, with respect to the site of the internal opening, type and depth of the fistula tract.
Patients and methods: A consecutive series of 151 patients with anal sepsis underwent pre-operative endoanal ultrasound assessment of a suspected anal fistula. Hydrogen peroxide was used to define the tract when there was doubt as to the course of the fistula. All patients subsequently had surgical exploration under anaesthesia, irrespective of findings at sonography. The site of the internal opening, depth and type of fistula were recorded at surgery, and concordance with the ultrasound was assessed.
Results: One hundred and forty-five patients were subsequently shown to have a fistula at surgical exploration. Type of fistula: Two thirds were transsphincteric (63%) and one third were inter sphincteric (32%), with a few submucosal, and supra sphincteric fistulas. Ultrasound correctly predicted surgical findings in 82% of patients (124/151). Concordance was highest for transsphincteric fistulas (87%). Internal opening: Accuracy of predicting the site of the internal opening was 93% (140/151). The commonest site for the internal opening was the midline posteriorly (49%), followed by the midline anteriorly (25%), the rest lay laterally. Fistula depth: Ultrasound and surgical assessment of the depth of fistulas was concordant in 120 of 145 patients (83%).
Conclusions: Endoanal ultrasound has a high accuracy of predicting the site of internal opening of an anal fistula. Endoanal ultrasound is able to assess the type and depth of a fistula. This information is useful for pre-operative planning of fistula treatment.