ABSTRACT
BACKGROUND: The serum (1 â 3)-ß-D-glucan (BG) assay has been approved for diagnosing invasive fungal diseases (IFDs). However, the performance of (1 â 3)-ß-D-glucan assay in bronchoalveolar lavage (BAL) fluid is various among studies. The present study aimed to assess the accuracy of (1 â 3)-ß-D-glucan assay in bronchoalveolar lavage fluid for the diagnosis of invasive fungal diseases by means of meta-analysis and systematic review of relevant studies. METHOD: The sensitivity, specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR), diagnostic odds ratio (OR) and a summary receiver-operating characteristic curve of BAL-BG for diagnosing invasive fungal diseases were pooled using meta-analysis. We also performed meta-regression analysis. RESULTS: A total of 838 patients (138 with proven or probable invasive fungal diseases), included in 6 studies, were analyzed. The pooled sensitivity, specificity, PLR, NLR and diagnostic odds ratio were 0.52 (95%CI, 0.38-0.53), 0.58 (95%CI, 0.55-0.61), 1.34 (95%CI, 1.08-1.66), 0.82 (95% CI, 0.63-1.07) and 1.71 (95%CI, 1.01-2.92) respectively. The area under the summary receiver operating characteristic curve, with 95% confidence intervals was 0.61 (95%CI, 0.67-0.55). CONCLUSION: The accuracy of (1 â 3)-ß-D-glucan test in bronchoalveolar lavage fluid is marginal, so that the results should not be interpreted alone but can be used as a part of full assessment with clinical features, image findings and other laboratory results for the diagnosis of invasive fungal diseases.
