From the limited evidence available, MRSA isolation seemed to be a significant factor for treatment failure in patients with diabetic foot infections, but further research was required. Given several limitations with the evidence available, including apparently low quality and clinical and methodological heterogeneity, the authors’ cautious conclusions and recommendations for further research appear appropriate.

To identify the factors associated with treatment failure in patients with diabetic foot infections.

PubMed (January 1950 to December 2005) and Cochrane Database of Controlled Trials were searched for relevant articles. Search terms were reported. In addition, references of articles retrieved were searched manually.

Randomised controlled trials comparing different antibiotic regimens for the treatment of diabetic foot infections and reporting treatment failure, and the factors potentially associated with such failure, were eligible for inclusion. Treatment failure was defined as persistence of signs and symptoms consistent with infection of the diabetic foot with or without persistent positive cultures, despite antibiotic administration and proper surgical intervention.

Included studies used different antibiotic regimens. Most studies administered antibiotics intravenously at the beginning of therapy. Some studies assessed patients with skin and soft tissue infections, but only patients with diabetic foot infections were included in the analysis. Studies also reported the number of patients with different baseline pathogens and the presence of osteomyelitis, although most studies excluded patients with osteomyelitis.

Two reviewers independently screened studies for inclusion. Discrepancies were resolved by consensus between all authors.

The authors did not state how they assessed validity. The authors reported on the overall quality of the studies and blinding, but did not report the criteria they used to judge the overall quality.

Two reviewers independently extracted the number of treatment failures to calculate percentage failures and odds ratios, with 95% confidence intervals (CIs). Discrepancies were resolved by consensus between all authors.

Odds ratios were pooled using a fixed-effect or random-effects model, as appropriate. The number (percentage) of treatment failures were grouped by antibiotic, baseline pathogen and presence of osteomyelitis. Heterogeneity was assessed using the Χ² test. Egger’s test was used to assess for publication bias.

Eighteen randomised controlled trials (n=1,715) were included in the review. Seven randomised controlled trials were double blinded. Eleven randomised controlled trials were reported to be of low quality.

Treatment failure was reported in 22.7 per cent of patients with diabetic foot infections. Treatment failure was highest among patients receiving penicillins compared to other classes of antibiotics, odds ratio 1.33 (95% confidence interval: 1.02, 1.74). However, the difference was no longer significant when the randomised controlled trial comparing linezolid with aminopenicillins was removed. There were no other significant differences between different direct antibiotic comparisons.

Significantly fewer treatment failures were reported with carbapenems (p<0.001) and in patients with diabetic foot infections from whom an aerobic streptococcal strain was isolated (p=0.02). By comparison, patients with isolation of MRSA (methicillin resistant S. aureus) at baseline reported more treatment failures (p=0.02).

There was no significant heterogeneity between studies. Publication bias was not detected using the Egger’s test.

From the limited evidence available, MRSA isolation seemed to be a significant factor for treatment failure in patients with diabetic foot infections. Further research was required to identify additional risks for treatment failure and determine the optimal management of patients with diabetic foot infections.

The objectives of the review and supporting inclusion criteria were clearly stated. The literature search was limited to two electronic databases and one further appropriate source. Although no attempts to minimise publication or language bias were reported, tests showed no evidence of publication bias. The study selection and data extraction processes were made explicit, thus minimising the risk for reviewer error and bias. The authors stated that most studies were of low quality, but criteria used to assess studies were not reported, so it was not possible to verify the authors' findings. Although the authors reported that there was no statistically significant heterogeneity among studies, they acknowledged clinical and methodological differences. Given the limited evidence available, and the above-mentioned considerations, the authors’ cautious conclusions and recommendations for further research appear appropriate.

Practice: The authors did not state any implications for practice.

Research: The authors stated that future research should focus on severity of infections in order to make a meaningful analysis of the effect of various confounding variables on treatment failure of diabetic foot infections. Further prospective studies and randomised controlled trials were needed to further assess the potential influence of other factors such as age, type one and two diabetes mellitus, insufficiency of vascular supply and degree of glycaemic control on treatment failure in patients with diabetic foot infections.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.