Early empiric therapy for invasive fungal infection in the ICU

Abstract

The risk of infection increases with the length of stay in the intensive care unit (ICU). Invasive fungal infections (IFI) are now responsible for a sizeable proportion of ICU infections. Candida spp. are a particular problem in the ICU, where they are not only among the most common pathogens responsible for blood stream infections (BSI), but also associated with an unexpectedly high rate of mortality. Central venous catheters, total parenteral nutrition, and increasingly, surgery, are common sources of Candida BSI. While C. albicans remains the most common species implicated in candidemia, C. glabrata, C. parapsilosis, C. tropicalis, C. krusei, and other non-albicans species are often responsible for infection, especially among older patients (e.g., > 60 years). Risk scores and prediction rules for Candida BSI are helpful, but not straightforward to apply. There is also uncertainty around the appropriate length of therapy for deep-seated Candida infections (e. g., intra-abdominal infections). Despite predictive, diagnostic, and treatment uncertainties, correct assessment of patient risk factors for nosocomial Candida infection is crucial to avoiding the poor outcomes associated with delayed or inappropriate initial therapy. The reality is that many ICU patients with Candida IFI are treated late or with fluconazole despite guidelines that an echinocandin be used as initial therapy for critically ill patients. Furthermore, not all echinocandins are equally effective, particularly when it comes to non-albicans Candida infections. The problem of antifungal agents not being interchangeable extends to treatment of invasive aspergillosis, where selecting an agent such as voriconazole gives rise to concerns about drug interactions and the need for therapeutic drug monitoring. Given the increasing problem of candidemia in the ICU and the shift to non-albicans species, especially in older patients, early empiric therapy with an echinocandin is essential, as is consideration of surgical co-intervention when urgent measures are needed. Early empiric therapy with the appropriate echinocandin and source control are the cornerstones of effective treatment for Candida BSI.