Performance of existing clinical scores and laboratory tests for the diagnosis of invasive candidiasis in critically ill, nonneutropenic, adult patients: a systematic review with qualitative evidence synthesis

Abstract

Background: The Fungal Infections Definitions in Intensive Care Unit (ICU) patients (FUNDICU) project aims to provide standard sets of definitions for invasive fungal diseases in critically ill, adult patients. Objectives: To summarise the available evidence on the diagnostic performance of clinical scores and laboratory tests for invasive candidiasis (IC) in nonneutropenic, adult critically ill patients. Methods: A systematic review was performed to evaluate studies assessing the diagnostic performance for IC of clinical scores and/or laboratory tests vs. a reference standard or a reference definition in critically ill, nonneutropenic, adult patients in ICU. Results: Clinical scores, despite the heterogeneity of study populations and IC prevalences, constantly showed a high negative predictive value (NPV) and a low positive predictive value (PPV) for the diagnosis of IC in the target population. Fungal antigen-based biomarkers (with most studies assessing serum beta-D-glucan) retained a high NPV similar to that of clinical scores, with a higher PPV, although the latter showed important heterogeneity across studies, possibly reflecting the targeted or untargeted use of these tests in patients with a consistent clinical picture and risk factors for IC. Conclusions: Both clinical scores and laboratory tests showed high NPV for the diagnosis of IC in nonneutropenic critically ill patients. The PPV of laboratory tests varies significantly according to the baseline patients' risk of IC. This qualitative synthesis will provide the FUNDICU panel with baseline evidence to be considered during the development of definitions of IC in critically ill, nonneutropenic adult patients in ICU.