Detection of (1, 3)-β-D-glucan in bronchoalveolar lavage and serum samples collected from immunocompromised hosts.

The incidence of invasive fungal infections (IFI) has increased in recent years, especially among immunocompromised hosts (ICH). In 2003, the Fungitell(®) assay received FDA clearance for the presumptive diagnosis of IFI using serum and detects (1-3)-β-D-glucan, which is a major cell wall component of certain fungi (e.g., Candida, Aspergillus, and Pneumocystis). The goal of the current study was to assess the performance of the assay on bronchoalveolar lavage (BAL) fluid and serum to identify IFI in ICH. Patients were classified as having proven, probable, possible, or no IFI according to the European Organization for Research and Treatment of Cancer/Mycoses Study Group (EORTC/MSG) guidelines. Among 109 patients for whom the results of Fungitell were compared to the EORTC/MSG criteria, Fungitell showed a low positive predictive value for the identification of IFI from both BAL (20.0%) and serum (26.7%). However, the negative predictive value of Fungitell was significantly higher for both sample types (BAL, 83.0%; serum, 84.8%). Interestingly, the results of Fungitell were positive in BAL and serum in 7/8 (87.5%) patients diagnosed with Pneumocystis pneumonia (PcP) by real-time, non-nested PCR. These data indicate that the Fungitell assay has a low positive predictive value for the diagnosis of IFI in ICH, regardless of the specimen type that is tested. However, testing of serum samples by Fungitell may permit a rapid and noninvasive initial screening approach in patients with presumed PcP.