Again in a population with gastrointestinal leakages but additionally including patients with acute necrotising pancreatitis, the same group recently showed in a pilot non-comparative trial that caspofungin was successful in the prevention of intra-abdominal check details IC in 18/19 patients (95%, one breakthrough IC 5 days after inclusion).80 This finding will have to await confirmation in a randomised trial. Moreover, it may not be the most prudent choice to use echinocandins for prophylaxis as these agents have evolved as an important option for therapy of established Candida infections. In a double-blind

placebo-controlled trial, Garbino et al. [81] investigated low-dose fluconazole (100 mg Temozolomide day−1) in mechanically ventilated ICU patients and found a significant reduction of the rate of candidaemia episodes but with no mortality benefit. Pelz et al. [82] used a predicted ICU stay of >3 days on admission as an inclusion criterion in their placebo-controlled trial using 400 mg day−1 fluconazole for prophylaxis. In a time-to-event analysis, the risk of IC in the fluconazole arm was significantly reduced vs. placebo. However, the proven infection rate in

the placebo arm was 61% after 21 days, which is a quite unusual finding precluding general conclusions from the results. Even at this unacceptably high background fungal infection rate, no survival benefit was observed in the prophylaxis arm. Nonetheless, in their current guidelines the IDSA recommends the prophylactic

use of fluconazole in for high-risk patients in adult ICUs with a high incidence of IC (>10%). mafosfamide However, apart from the patients with intra-abdominal leakage, it remains largely unclear as to which specific risk factor profiles are associated with a benefit from antifungal prophylaxis. Objections to the prophylactic use of antifungal drugs in ICU patients include the potential to select for species or strains with reduced azole susceptibility. However, in none of the prophylaxis trials in the ICU setting this kind of pathogen shift was observed.42 While invasive Candida species clearly are the leading cause of invasive fungal infections in the ICU, invasive aspergillosis (IA) has been described in ICU patients at varying incidence rates. In a retrospective autopsy-controlled study, as many as 6.9% of patients had histopathological or microbiological evidence of IA, 70% of these patients did not suffer from underlying haematological malignancies, one of the classical risk factors for IA.83 The mortality was 80%, much higher than predicted from Simplified Acute Physiology Score values. Other studies suggest that IA is among the frequently undiagnosed conditions in ICU patients.84 While in most ICU the incidence rate may be much lower than that described above, awareness of Aspergillus spp.