Effect of empirical antifungal treatment on mortality in non-neutropenic critically ill patients: a propensity-matched retrospective cohort study.

To evaluate the effect of empirical antifungal treatment (EAFT) on mortality in critically ill patients without invasive fungal infections (IFIs). This was a single-center propensity score-matched retrospective cohort study involving non-transplanted, non-neutropenic critically ill patients with risk factors for invasive candidiasis (IC) in the absence of IFIs. We compared all-cause hospital mortality and infection-attributable hospital mortality in patients who was given EAFT for suspected IC as the cohort group and those without any systemic antifungal agents as the control group. Among 640 eligible patients, 177 patients given EAFT and 177 control patients were included in the analyses. As compared with controls, EAFT was not associated with the lower risks of all-cause hospital mortality [odds ratio (OR), 0.911; 95% CI, 0.541-1.531; P = 0.724] or infection-attributable hospital mortality (OR, 1.149; 95% CI, 0.632-2.092; P = 0.648). EAFT showed no benefit of improvement of infection at discharge, duration of mechanical ventilation, and antibiotic-free days. However, the later initiation of EAFT was associated with higher risks of all-cause hospital mortality (OR, 1.039; 95% CI, 1.003 to 1.076; P = 0.034) and infection-attributable hospital mortality (OR, 1.046; 95% CI, 1.009 to 1.085; P = 0.015) in patients with suspected IC. This effect was also found in infection-attributable hospital mortality (OR, 1.042; 95% CI, 1.005 to 1.081; P = 0.027) in septic patients with suspected IC. EAFT failed to decrease hospital mortality in non-neutropenic critically ill patients without IFIs. The timing may be critical for EAFT to improve mortality in these patients with suspected IC. ChiCTR2000038811, registered on Oct 3, 2020.