The clinical and histological effects of Dermagraft (R) in the healing of chronic venous leg ulcers

Objective: Pilot study to assess the safety and effectiveness of Dermagraft(R) when used in conjunction with multi-layer compression bandage therapy (Profore(TM)) compared with multilayer compression only in the treatment of chronic venous leg ulcers. Design: Open-label, prospective, multicentre, randomized, controlled clinical trial. Methods: Patients aged at least 18 years with leg ulceration of venous aetiology were screened for inclusion in the trial. Patients with arterial disease (ankle brachial pressure index <0.7) and causes of ulceration other than venous disease were excluded. Patients were randomized into four groups. Three active treatment groups received Dermagraft(R) in combination with compression bandaging, Profore(TM): Group 1, 12 pieces of Dermagraft (one per week for 12 weeks); Group 2, four pieces of Dermagraft (Weeks 0, 1, 4, 8); Group 3, one piece of Dermagraft (Week 0). The control group was treated with compression bandaging alone (Group 4). The study was not powered to detect statistical differences in healing rates. Adverse events were reported according to national guidelines. There was no statistical analysis of adverse event data. Results: In all 53 patients were randomized, of whom 47 completed the study per protocol. At 12 weeks complete healing of the ulcer, analysed by 'intention-to-treat' (ITT) was 15% using Profore alone, 7% using single application Dermagraft and Profore (Group 3), compared with 38% using multiple applications of Dermagraft and Profore (Groups 1 and 2). At study discontinuation, the median percentage reduction in wound area was 81.4% for Group 1, 88.6% for Group 2, 59.4% for Group 3 and 78.1% for Group 4. No major safety issues were identified during the course of the study. Conclusions: The results of the pilot study indicate that four pieces of Dermagraft are the optimal application frequency to take forward to further clinical trials. Further studies are required to confirm these data, and these should be powered to detect whether there are statistical differences in healing rates and safety between different regimens. (C) 2003 Royal Society of Medicine Press.