This recently published meta-analysis evaluated the diagnostic accuracy of Fungitell BDG assay in diagnosis of IFI in patients with haematologic malignancies and solid organ transplant recipients. After screening 12,426 articles (only full-text articles) from EMBASE and MEDLINE database from 1995 (when BDG was first used as a diagnostic assay) to 2017, they ultimately included 19 studies (3040 patients) in their statistical analysis. The included studies had been published between 2008 to 2017. All these studies used serum Fungitell BDG test at a threshold of 80 pg/mL (85 pg/mL in one study) to identify any type of IFI among patients. IFI was defined as definite, probable and possible using EORTC criteria/ study-specific diagnostic methods similar to EORTC.
Of the 19 studies, 9 studies were prospective, 6 were case control and 4 were retrospective cohort studies. The majority of studies (58%) included all fungal infections as part of their study design, followed by the inclusion of only Aspergillus (n = 5), PJP (n = 2), or Candida (n = 1).
The overall sensitivity and specificity estimates for the 19 studies were 83% (95% CI, 74%-89%) and 79% (95% CI, 64%-88%), respectively. There was substantial heterogeneity between studies. After excluding 9 studies with a high risk of bias the pooled sensitivity and specificity was 80% and 63% respectively. There continued to be substantial heterogeneity in regards to reported specificities among high-quality studies. The NPV ranged between 90-100% barring one study (68%). PPVs varied widely (6-100%). On average, the PPV increased by 1% for every 1% increase in prevalence.
The authors concluded that the Fungitell BDG assay is fairly sensitive for diagnosis IFI but has lower specificity among cancer patients. Therefore, it is important that BDG be used in conjunction with other clinical findings when deciding if antifungal therapy should be administered.