Combination of Candida biomarkers in patients receiving empirical antifungal therapy in a Spanish tertiary hospital: a potential role in reducing the duration of treatment.

Abstract

Initiation of empirical antifungal therapy for invasive candidiasis (IC) is usually based on clinical suspicion. Serological biomarkers have not yet been studied as a means of ruling out IC. We evaluated the potential role of two combined biomarkers in stopping unnecessary antifungals in patients at risk of IC in the ICU and in other wards. This was a prospective observational study including adults starting empirical antifungal treatment for suspected IC, at Gregorio Marañón Hospital, Madrid (Spain). Patients were stratified according to admission department (ICU or other wards) and final diagnosis (no IC or proven or probable IC). Type of candidiasis (candidaemia or deep-seated candidiasis) was also considered. The Candida albicans germ tube antibody (CAGTA) test and the β-d-glucan (BDG) test were performed on serum samples collected by venepuncture on days 0, 3 and 5 after starting empirical antifungal therapy. Sixty-three ICU patients and 37 non-ICU patients were included. High-risk gastrointestinal surgery and sepsis in non-surgical patients were the main indications for empirical treatment (30% each). Patients had no IC (58%), proven IC (30%) or probable IC (12%). Overall, sensitivity and negative predictive value of the combination of both the CAGTA test and the BDG test were 97% for the entire population. The best performance was observed in ICU patients (sensitivity and negative predictive value of 100%). Among patients without IC, all biomarkers were negative in 31 patients. Serial determination of CAGTA/BDG during empirical antifungal therapy has a high sensitivity and negative predictive value. If properly confirmed, this strategy could be used to discontinue antifungal treatment in at least 31% of patients as a complementary tool in antifungal stewardship programmes.