Objective: In this retrospective analysis, we compared fungal disease outcomes in the pre-intervention (1/02-8/05) and post-intervention (9/05-6/09) periods.
Methods: All deaths and fungal culture results were evaluated for the study periods. Deaths were evaluated for presence of skin or tissue evidence of disseminated fungal disease. Disseminated fungal infection were defined as patients with positive cultures from sterile sites or multiple positive cultures from distant non-sterile sites in the presence of systemic illness and source. Incidence rates are calculated per 1000 JHBU patient-days (pds). Odds ratios (OR) are presented to reflect intervention impact.
Results: Patient demographics of the two study periods were similar and included average age 58.9 vs 58.5, length of stay (15.5 vs 16.3), average BSA burn (51% vs 52%). Pre and post-intervention overall mortality rates were 8.7 and 5.5 per 1000pds (Intervention OR=0.63, 95%CI=0.44-0.90). Death rates with definitive DFI were 0.79 pre-intervention and 0.27 post-intervention per 1000pds (OR=0.34; 95%CI:0.07-1.52). Incident rates of patients with ≥1 cultures positive for Candida fell from 2.2 pre-intervention to 0.4 post-intervention per 1000pds (OR= 0.24; 95%CI:0.08-0.65). Non-yeast infections (i.e molds) also decreased in the latter period but the difference was not significant
Conclusions: Fluconazole prophylaxis and antibiotic stewardship significantly decreases incidence of candidiasis among critically ill patients in the burn ICU setting . There was a trend towards decreased mortality.