The objectives of this pilot study were to identify the bacterial flora of neurotrophic ulcers, and to determine whether treatment with ceftriaxone improves the clinical picture. Twenty-five patients (15 male, 10 female, mean age 70 +/- 11 years) with mal perforant were treated with 2 g ceftriaxone once daily until the ulcer was healed, or for a period of 6 weeks. Biopsies were taken from the ulcer floor before and after therapy. Initially, the following bacteria were found, in order of frequency: Staphylococcus aureus, enterobacteria, coagulase-negative staphylococci, enterococci and streptococci as well as Pseudomonas aeruginosa. Under therapy with ceftriaxone, the lesions of 11 patients were healed; in 5 patients improvement > 50% was observed and in 4 patients improvement was < 50%. In 3 patients there was no change in the condition. Two patients had to stop treatment due to diarrhea. Asepsis was achieved in 5 patients despite persistence of the ulcer. The control group (25 patients without antibiotic therapy) showed the following results: healing in 6, improvement > 50% in 5, improvement < 50% in 10, and no change in 4. The results for the treatment group have to be compared with a historical control group since, for ethical reasons, the implementation of a double-blind study would not be feasible.