OBJECTIVETo extend our previous work on evaluating the use of oligonucleotide arrays to discriminate colonization from infection owing to Staphylococcus aureus in diabetic foot ulcers (DFUs).RESEARCH DESIGN AND METHODSPatients admitted to 14 French diabetic foot departments for a DFU were screened for entry into the study. At admission, ulcers were classified based on clinical examination according to the Infectious Diseases Society of America system. Only patients with monomicrobial culture for S. aureus were included. In persons with an uninfected ulcer, a second wound bacterial specimen was obtained 1 month later. Using oligonucleotide arrays, S. aureus resistance and virulence genes were determined, and each isolate was affiliated to a clonal complex (CC).RESULTSS. aureus was initially isolated from 75 uninfected and 120 infected ulcers; 35 were methicillin resistant. A total of 44 (59%) strains from uninfected DFUs belonged to CC5/CC8 clones vs. 6 (5%) from infected DFUs (P < 0.001). During follow-up, 57 (76%) of uninfected DFUs healed or had a favorable outcome; the strain in 49 (86%) of them belonged to CC5/CC8. Conversely, 18 (24%) had a poor outcome but not a single strain belonged to CC5/CC8 clone. Moreover, lukDE was significantly associated with a favorable outcome of the wound.CONCLUSIONSAs suggested by our previous study, the use of DNA arrays appears to be a promising technique that might help distinguishing uninfected from infected wounds, predicting ulcer outcome and then contributing to a more adequate use of antibiotics.

The characteristics of the dose response of insulin on the glucose turnover rate and erythrocyte insulin binding parameters were determined in five normal men before and during experimentally induced hyperthyroidism [L-T4 (2 micrograms kg-1 day-1) for 4 weeks with additional L-T3 (1 microgram kg-1 day-1) for the following 3 weeks]. Hyperthyroidism was characterized by significant rises in T3 from 1.92 +/- 0.17 (+/- SEM) to 3.66 +/- 0.17 nmol/liter (P less than 0.01) and resting metabolic rate from 39 +/- 0.7 to 48 +/- 1 watt/m2 (P less than 0.001). While the subjects received a diet adapted to the metabolic rate, blood glucose rose from 3.8 +/- 0.07 to 4.46 +/- 0.11 mmol/liter (P less than 0.05) without a significant change in plasma insulin. During the insulin dose-response study, glucose infusion rates were unaltered by hyperthyroidism, and neither the maximum effect nor the sensitivity to insulin was altered. Glucose turnover rate, measured using [6,6-2H2]glucose as tracer, was determined in the basal state and during the 0.4 mU kg-1 min-1 insulin infusion. In the basal state, it was significantly increased by hyperthyroidism (control, 2.3 +/- 0.1; hyperthyroidism, 3.7 +/- 0.1 mg kg-1 min-1). During the insulin infusion, hepatic glucose production was totally suppressed before T4 and T3 treatment, but was 0.96 +/- 0.39 mg kg-1 min-1 during T4 and T3 treatment. A marked decrease in the insulin binding affinity to erythrocytes was found without a change in the insulin receptor number. In conclusion, glucose metabolism in experimental hyperthyroidism is characterized by 1) increases in basal glucose production and utilization; 2) antagonism between the effect of insulin and hyperthyroidism at the hepatic level; and 3) lack of peripheral insulin resistance in spite of marked alteration in erythrocyte insulin binding affinity.