Background
Empiric antimicrobial regimes for diabetic foot infections (DFI) vary according to local epidemiology and prior colonisation/infection with resistant pathogens. The utility of using results from culture of superficial swabs to guide empiric antibiotic therapy is not well established in Australia, particularly in the context of rising prevalence of drug resistant organisms. Antibiotic resistance is an emerging threat to DFI management.
Aims
To assess the utility of DFI microbiology results and their relationship to empiric antibiotic prescribing.
Methods
Retrospective observational study of 151 admissions in 128 patients admitted to Fiona Stanley Hospital MDFU from 1st February 2015 to January 31st 2016.
Results
The mean age was 60.4±14.9 years, 11.7% were Indigenous Australians. Most admissions (92%) were for moderate or severe DFI. Empiric antibiotic prescribing was consistent with published guidelines. Of those with positive cultures, 85% were polymicrobial. Dominant pathogens included: S. aureus (48%), beta-hemolytic streptococci (24%), Enterobacteriaceae (15%), enterococci (8%) and Pseudomonas (6%). Excluding cultures of known skin commensals, the proportion of superficial cultures concordant with deep tissue or blood cultures ranged from 0% (Pseudomonas) to 50% (Enterobacteriaceae). Prior colonization with methicillin resistant S. aureus (MRSA) or vancomycin resistant enterococci (VRE) occurred in 12% and 8% patients respectively. Prior MRSA colonization had a positive predictive value (PPV) of 52% and negative predictive value (NPV) of 97%, previous VRE colonization had a PPV of 0% and NPV of 100% for inpatient infection. Overprescribing with anti-MRSA and/or anti-pseudomonal agents occurred commonly (76%) given subsequent culture results, however empiric prescribing provided insufficient antimicrobial activity in 12%.
Conclusion
Cultures of superficial swabs correlated poorly with deep tissue or blood cultures. Knowledge of absence of prior MRSA colonization provides justification to withhold empiric anti-MRSA therapy. Broad spectrum empiric therapy may not be necessary in this setting, highlighting opportunities for rationalization of antimicrobial therapy.