Evidence of the Month

Commentaries on both new and classic studies of importance for the treatment of diabetes are posted here monthly.

N. C. Schaper
Department of Internal Medicine
Division of Endocrinology
University Hospital Maastricht
Maastricht, the Netherlands

Click here to download PowerPoint slides of this commentary.

Click here for the programme evaluation and post-test.

Spreading infection with extensive tissue loss is one of the main reasons for a lower leg amputation in individuals with diabetes mellitus with a foot ulcer. Early diagnosis with prompt institution of an appropriate antibiotic and, if necessary, surgical treatment are the cornerstones of managing these infections. However, diagnosing diabetic foot infections is a clinical challenge as typical signs and symptoms of infection, such as pain, redness, or elevated circulating inflammatory markers, can be absent in individuals with neuropathic or neuroischaemic ulcers. Wound cultures are in this phase less helpful as, like all chronic wounds, these ulcers tend to become colonised by various micro-organisms that may or may not subsequently invade the surrounding tissues depending on factors such as virulence, wound characteristics, and host defence. Staphylococcus aureus is the most frequently isolated pathogen from infected ulcers, in monoculture or in combination with other pathogens. Treating all clinically uninfected diabetic foot ulcers with antibiotics would result in excessive costs and would undoubtedly further contribute to the development of multi-resistant micro-organisms, such as methicillin-resistant S aureus (MRSA). To solve this clinical dilemma, we urgently need new strategies to diagnose infection more early, and ideally we need a technique that can detect pathogenic micro-organisms in a foot ulcer before they invade the surrounding tissues.

In the August 2007 issue of Diabetes Care, Sotto and co-workers present the results of a French pilot study on the use of a miniaturized oligonucleotide DNA array, covering the genes encoding for virulence and resistance factors, with the aim to differentiate between colonised and infected wounds. As stated by Benjamin Lipsky in the accompanying editorial, this study shows us a glimpse of what we may expect in the future for clinical microbiology.

Specimens for microbiological examination were obtained from 338 patients with diabetes without a recent history of antibiotic therapy and who were admitted to the hospital with a foot ulcer. Infection was classified according to the clinical criteria of the Infectious Diseases Society of America/International Working Group on the Diabetic Foot. Samples for bacterial culture were taken after debridement in all patients, and only patients with a monoculture of S aureus were included in the study. Using oligonucleotide arrays, profiles of resistance and virulence genes were obtained from each S aureus strain. A repeat sample was taken after 1 month or earlier in case of worsening of the wound, and all patients were followed for 6 months.

S aureus was isolated as monoculture in 72 patients: 35 were methicillin resistant, and 37 were methicillin sensitive. This high number of MRSA isolates is in line with earlier reports from other countries on the growing prevalence of MRSA in diabetic foot ulcers. MRSA was, in particular, observed in all patients with a recurrent ulcer; these patients had been hospitalised and treated with antibiotics for an infected foot ulcer in the past. In contrast, only 1 patient who had a foot ulcer for the first time was MRSA positive. Results of the resistance genes obtained with the DNA array were in agreement with the conventional susceptibility data.

Of the patients with S aureus monoculture, 50 had clinical signs of infection, and 22 had clinically an uninfected wound. The prevalence of MRSA was lower in the uninfected wounds compared with the infected wounds, and virulence genes were more prevalent in the patients with infected wounds (98%) than in the uninfected wounds (9%). The 22 patients with clinically uninfected wounds were not treated with antibiotics; during the 6-month follow-up period, 8 healed and 14 worsened with development of clinical infection. Two patients developed a clinical infection within 10 days, and the DNA array at baseline showed S aureus virulence genes in their wound samples; the same S aureus genotype was observed in the sample taken at follow-up. In contrast, 12 patients developed infection more than 3 weeks after initial presentation, and no virulence genes were observed in their initial DNA array at baseline. S aureus was isolated again in the majority of these patients with a late infection at follow-up, but these isolates had a different genotype.

Aggressive treatment of infection is one of the key elements in the treatment of diabetic foot ulcers. Approximately half of the patients have signs of infection at initial presentation, and several patients with initially uninfected wounds will develop an infection during follow-up. The study by Sotto et al highlights the problems we face in diagnosing and treating diabetic foot infections. Nearly 50% of all S aureus isolated were MRSA-positive, and this high prevalence was related to previous hospitalisation and antibiotic treatment, stressing the importance of limiting the use of broad-spectrum antibiotics. In countries such as the Netherlands, with a restrictive antibiotic policy, MRSA positive diabetic foot ulcers are relatively rare. Although S aureus was isolated from clinically infected and uninfected wounds, both virulence and resistance factors were more frequently present in clinically infected wounds. Moreover, isolation of S aureus from a clinically uninfected wound but with an unfavourable resistance and virulence profile was associated with rapid development of clinical signs of infection in a small number of patients. If confirmed in studies with a sufficient number of patients, DNA arrays could help us to detect more virulent colonisers and/or early infection before clinical signs can be observed. The results of the S aureus DNA arrays from the patients in the study by Sotto et al were in agreement with the susceptibility data obtained from conventional culture and are in line with earlier publications on arrays and PCR tests, as described by Lipsky in the accompanying editorial.

At initial presentation, broad-spectrum antibiotics are usually prescribed in individuals with a severe or extensive infected foot ulcer, as it takes a few days before the results of the wound culture are available. If adequate surgical drainage is obtained, most individuals will respond favourably to such treatment, and clinicians may be reluctant to narrow the antibiotic treatment based on these culture results. With PCR methods, it is already possible to identify gram-negative antibiotic-resistant micro-organisms, and it is likely that in the near future these and other techniques will enable the clinician to identify S aureus and, in particular, antimicrobial resistance within 24 hours, stimulating the prescription of antibiotics with a spectrum as narrow as possible. Only with such new approaches can we stem the growing epidemic of multi-resistant micro-organisms in individuals with infected diabetic foot ulcers.

This Website Feature is funded by an unrestricted educational grant from Pfizer Inc.