Tone et al. (1) are to be awarded for performing the first randomized trial that compares 6-week and 12-week antibiotic treatment duration in patients with diabetes and osteomyelitis with neuropathy without vascular disease. The study has two strong points: cultures of transcutaneous bone biopsy material were obligatory for inclusion and, in almost all cases, there was an antibiotic-free period between suspicion of infection and the start of antibiotic therapy.
We have, however, three concerns about the methodology of this study that we would like to address. First, we missed a clear description of which methods were used for pressure offloading, a very important component in the treatment of infected neuropathic ulcers (2). Differences in methods used for pressure offloading may have influenced the study results, especially when considering that the study had an open-label design.
Second, it remains unclear whether a noninferiority or superiority design was used. In case a noninferiority design was chosen, the conclusion that both treatment durations result in similar remission rates should be supported by confidence intervals of the difference in remission rates in both groups. In case a superiority design was chosen, the confidence intervals should also have been reported for readers to judge whether a possible type 2 error occurred.
Third, the absence of a power calculation in the article prohibits the reader from evaluating whether a type 2 error could have been present (3). The first entry on clinical trials was on 15 April 2014. The start date of the study was June 2007 and the completion date was September 2012. It was a missed opportunity, given the unique design of this study, not to have preregistered the trial, for instance on ClinicalTrials.gov, before the results of the study were available.
Article Information
Duality of Interest. No potential conflicts of interest relevant to this article were reported.
