To test the efficacy of surgical treatment of non-infected neuropathic foot ulcers compared to conventional non-surgical management, a group of diabetic outpatients attending our diabetic foot clinic were studied. All patients who came to the clinic for the first time from January to December 1995 inclusive with an uncomplicated neuropathic ulcer were randomized into two groups. Group A received conservative treatment, consisting of relief of weight-bearing, regular dressings; group B underwent surgical excision, eventual debridement or removal of bone segments underlying the lesion and surgical closure. Healing rate, healing time, prevalence of infection, relapse during a 6-month period following intervention and subjective discomfort were assessed. Twenty-four ulcers in 21 patients were treated in group A (17 Type 2 DM/3 Type 1 DM, age 63.24 +/- 13.46 yr, duration of diabetes 18.2 +/- 8.41 yr, HbA1c 9.5 +/- 3.8%) and 22 ulcers in 21 patients in group B (19 Type 2 DM/2 Type 1 DM, age 65.53 +/- 9.87yr, duration of diabetes 16.84 +/- 10.61 yr; HbA1c 8.9 +/- 2.2%). Healing rate was lower (79.2% = 19/24 ulcers) in group A than in group B (95.5% = 21/22 ulcers; p < 0.05), and healing time was longer (128.9 +/- 86.60 days vs 46.73 +/- 38.94 days; p < 0.001). Infective complications occurred significantly more often in group A patients (3/24, 12.5% vs 1/22, 4.5%; p < 0.05), as did relapses of ulcerations (8 vs 3; p < 0.01). There were only two minor perioperative complications in group B patients. Patients reported a higher degree of satisfaction in group B (p < 0.01) as well as lower discomfort (p < 0.05) and restrictions (p < 0.05). Thus surgical treatment of neuropathic foot ulcers in diabetic patients proved to be an effective approach compared to conventional treatment in terms of healing time, complications, and relapses, and can be safely performed in an outpatient setting.