In a recent article (The Diabetic Foot Journal 12: 39–43), Martin Turns provided an interesting case report in which healing of a diabetic foot ulcer occurred concurrent to the use of glucose oxidase dressings.
In the background to the report, and in justification of the use of this dressing, the author reported the results obtained with the dressing in three observational studies, called trials.
I am concerned that the numbers of people with diabetes and foot ulcers are so small in the combined studies that they provide no clear justification for the use of this dressing. The combined number of cases reported were 19: in one case the ulcer worsened, in three cases we are given no further information, and in 15 cases the wound improved or healed. Indeed, the occurrence of an adverse event in one of the studies may rather be a reason not to use the dressing. What is surely needed for any new dressing is a randomised controlled trial with a large number of participants.
In an individual case report, such as reported by Turns, assuming ethical permission, what would help is an n=1 study. In this, the healing rate is assessed for 1 week on the standard therapy and the healing rate measured, then the treatment is changed to the trial dressing for 1 week and the healing rate is measured again. The treatment is again changed to standard for the next week and further measurement taken, with the sequence continuing until the wound is healed. By comparing the healing rates between the two dressings, the hypothesis that the new dressing is superior to the standard therapy can be tested. This would also allow for a power calculation for a formal randomised controlled trial.
AC Felix Burden, Community Diabetologist, Birmingham