Results of collagen plug occlusion of anal fistula: a multicentre study of 126 patients

Aim The Biodesign (R) anal fistula plug was introduced as a means of obliterating the fistula tract and promoting healing through biocompatibility. The results demonstrated unexplained variations from good to bad. This report analysed the results of a retrospective multicentre study. Method All plug procedures performed in four Stockholm hospitals between June 2006 and June 2010 were identified and studied using a common protocol. The outcome after the first plug-insertion procedure was assessed by chart review performed a minimum of 8 months after plug insertion. Cox proportional-hazards models were used to assess the associations of various factors with fistula healing. Results One-hundred and twenty-six patients (mean age 47 years) were deemed suitable for the plug procedure. Eighty-five per cent of fistulae were cryptoglandular, 64% of patients were male and a mean of 2.9 previous fistulae procedures had been performed. All patients, except four, had an indwelling seton at the time of the plug procedure, which was performed in accordance with previously established principles of day surgery. After a median of 13 months, 30 (24%) fistulae had closed with no discomfort or secretion reported. The outcome in the four hospitals varied from 13% to 33% with similar numbers of patients in each hospital. A success rate of 12% was observed for patients with anterior fistula compared with 32% for those with posterior tracks [hazard ratio (HR) for successful healing = 2.98; 95% CI: 1.01-8.78) and 41% for those with a lateral internal opening (HR = 3.76; 95% CI: 1.03-13.75). Age, sex and number of previous procedures were not associated with healing. Conclusion Four independent patient groups showed low success rates after the first plug-insertion procedure. Anterior fistulae were much less likely to heal compared with fistulae in other locations.