ABSTRACT
BACKGROUND: Surgical intervention was once the mainstay of treatment for Buruli ulcer disease, a neglected tropical disease caused by Mycobacterium ulcerans. Since the introduction of streptomycin and rifampicin for 8 weeks as standard care, surgery has persisted as an adjunct therapy, but its role is uncertain. We investigated the effect of delaying the decision to operate to 14 weeks on rates of healing without surgery. METHODS: In this randomised controlled trial, we enrolled patients aged 3 years or older with confirmed disease at one hospital in Lalo, Benin. Patients were randomly assigned (1:1) to groups assessing the need for excision surgery 8 weeks (standard care) or 14 weeks after initiation of antimicrobial treatment. The primary endpoint was the number of patients healed without the need for surgery (not including skin grafting), assessed in all patients in follow-up at 50 weeks (or last observation for those healed for >10 weeks). A doctor masked to treatment assignment checked the indications for surgery according to predefined criteria. This study is registered with ClinicalTrials.gov, number NCT01432925. FINDINGS: Between July 1, 2011, and Jan 15, 2015, 119 patients were enrolled, with two patients per group lost to follow-up. 55 (96%) of 57 participants in the delayed-decision group and 52 (90%) of 58 participants in the standard-care group had healed lesions 1 year after start of antimicrobial treatment (relative risk [RR] 1·08, 95% CI 0·97-1·19). 37 (67%) of 55 patients in the delayed-decision group had their lesions healed without surgical intervention, as did 25 (48%) of 52 in the standard-care group (RR 1·40, 95% CI 1·00-1·96). The time to heal and residual functional limitations did not differ between the two groups (median time to heal 21 weeks [IQR 10-27] in the delayed-decision group and 21 weeks [10-39] in the standard-care group; functional limitations in six [11%] of 57 and three [5%] of 58 patients; p=0·32). Postponing the decision to operate resulted in reduced median duration of hospitalisation (5 days [IQR 0-187] vs 131 days [0-224]; p=0·024) and wound care (153 days [IQR 56-224] vs 182 days [94-307]; p=0·036). INTERPRETATION: In our study, patients treated for Buruli ulcer benefited from delaying the decision to operate. Even large ulcers can heal with antibiotics alone, without delaying healing rate and without an increase in residual functional limitations. FUNDING: NWO-VENI grant 241500, BUG Foundation, and UBS OPTIMUS.
