Invasive candidiasis is associated with high mortality, particularly in adults. Retrospective studies show that shorter times to treatment are correlated with a lower risk of death. A number of factors can be used to predict which patients would benefit from antifungal prophylaxis or early (pre-emptive or empirical) therapy. Detection of the fungal cell wall component (1→3)-β-D-glucan (BDG) shows promise as an early biomarker of invasive fungal infection and may be useful in identifying patients who would benefit from early antifungal treatment. To date, no consistent early treatment strategy has evolved. Proof-of-concept studies are needed to assess the role of pre-emptive and empirical therapy in ICU patients and the relevance of BDG as an early marker of infection.