Cerebrospinal fluid galactomannan detection for the diagnosis of central nervous system aspergillosis: a diagnostic test accuracy systematic review and meta-analysis

BACKGROUND: Cerebrospinal fluid (CSF) galactomannan is an adjunctive test for central nervous system (CNS) aspergillosis diagnosis with unclear diagnostic test characteristics. OBJECTIVES: To evaluate the diagnostic test characteristics of CSF galactomannan in CNS aspergillosis. METHODS: Systematic review and meta-analysis. DATA SOURCES: MEDLINE, Embase, Web of Science, and Scopus, from inception to 24 February 2023. STUDY ELIGIBILITY CRITERIA: Prospective and retrospective studies with 1-group and 2-group designs using any galactomannan assay on CSF to diagnose CNS aspergillosis. PARTICIPANTS: Adult and/or paediatric patients with CNS aspergillosis. TEST(S): Galactomannan testing on CSF specimens. REFERENCE STANDARD: European Organization for Research and Treatment of Cancer and the Mycoses Study Group Education and Research Consortium (EORTC/MSGERC) diagnostic criteria, or equivalent. ASSESSMENT OF RISK OF BIAS: QUADAS-2 assessment in duplicate. METHODS OF DATA SYNTHESIS: Bivariate restricted maximum likelihood estimation random-effects meta-analysis, summarized using forest and summary receiver operating characteristic plots; bivariate meta-regression models to investigate heterogeneity; and subgroup and sensitivity analyses to explore subgroup effects and methodologic choices (PROSPERO registration: CRD42022296331; funding: none). RESULTS: We included eight studies (n = 342 participants). The summary estimates of CSF galactomannan sensitivity and specificity were 69.0% (95% CI, 57.2-78.7%) and 94.4% (95% CI, 82.8-98.3%), respectively. Using meta-regression, galactomannan cut-off (p = 0.38), EORTC/MSGERC criteria version (p = 0.48), or whether the reference standard was defined as both proven and probable or only proven aspergillosis (p = 0.48) did not explain observed heterogeneity. No subgroup effects were demonstrated by analysing the EORTC/MSGERC criteria reference standard used (e.g. 2002 vs. 2008 definitions) or whether paediatric patients were included. Diagnostic sensitivity was improved using a galactomannan cut-off of 1.0, and by excluding high risk of bias and 1-group design studies. DISCUSSION: CSF galactomannan is a highly specific but insensitive test for use as a component of CNS aspergillosis diagnosis. Few included studies, no prospective studies, and a high risk of bias are study limitations.