A total of 203 patients were randomized into a prospective trial to compare short (SC) versus long courses (LC) of systemic antibiotic for acute cholecystitis treated by early cholecystectomy. The initial pre-operative management was the same and all patients received 2 g of cefamandole intravenously just before operation. Two further doses of cefamandole 500 mg were given 6 and 12 h later for patients on SC while the antibiotic was continued at 500 mg at 6 h intervals for 7 days for patients on LC. Seven patients developed wound infection on SC compared with five patients with wound infection and an additional patient with a subphrenic abscess on LC (P greater than 0.05). Thrombophlebitis related to intravenous antibiotic injections was more common in patients on LC (P less than 0.05). Also, patients on LC had to stay statistically longer in hospital in order to complete the course of antibiotic (P less than 0.05). We therefore recommend a SC to be used, as it is more cost-effective and causes fewer complications.