To assess the cost-effectiveness of Dermagraft(R) (human dermal replacement) in the treatment of the diabetic foot ulcer, compared to standard treatment. A Markov model was developed, to simulate, over a 52-week period, the health status of a cohort of 100 patients with a diabetic foot ulcer treated either with conventional therapy or with Dermagraft(R). The considered health states were: healed, same site recurrence, unhealed not infected, cellulitis, osteomyelitis, amputation and death. Each week, the patient may progress among states according to a set of transition probabilities directly derived from the original clinical trial conducted in the USA. The cost of each health state was estimated by a Delphi panel of French diabetologists (direct costs only, valuated from a societal perspective). A sensitivity analysis was performed. The total number of healed ulcers included first ulcers healed (76.38% for Dermagraft(R) vs. 69.35% for standard treatment; median time to heal is 14-15 weeks for Dermagraft(R) compared with 28-29 weeks for standard treatment) plus recurrences which are subsequently healed within the 52-week period (14.29 for Dermagraft(R) vs. 25.09 for standard treatment; median time to heal is 3-4 weeks for Dermagraft(R) compared with 5-6 weeks for standard treatment). The average expected cost per treated patient (C/E) using standard treatment for the considered 52-week period is 47,418 FF vs. 54,384 FF for Dermagraft(R) (including 18,200 FF for Dermagraft(R) acquisition and 36,184 FF for standard treatment). Because Dermagraft(R) heals more ulcers within 52 weeks, the average cost per healed ulcer is lower (53,522 FF vs. 56,687 FF for standard treatment). The incremental cost-effectiveness ratio of Dermagraft(R) (DeltaC/DeltaE) equals 38,784 FF, indicating the extra investment that the decision-maker has to accept for an additional ulcer healed with Dermagraft(R) compared with conventional treatment.