Detecting Invasive Fungal Disease in Surgical Patients: Utility of the (1→3)- β-D-Glucan Assay

Background: The specificity and sensitivity of the (1 -> 3)-beta-D-glucan (BDG) assay in surgical patients needs further investigation. We hypothesized that the BDG assay would have lower sensitivity/specificity compared with that of medical patients. Methods: We reviewed patients who had undergone laparotomy, gastrectomy, hepatectomy, or colectomy and had a BDG assay post-operatively. Results: A total of 71 patients met study criteria. There were 29 (40.8%) who had proven/probable invasive fungal infection. Sensitivity for BDG level >= 80 diagnosed within one week of the assay draw was 77.3% (95% confidence interval [CI], 54.6-92.2%), and specificity was 44.9% (95% CI, 30.7-59.8). The positive predictive value was 38.6% (95% CI, 31.0-46.9%), and negative predictive value was 82.5% (95% CI, 65.7-91.0%). A BDG assay result of 149 pg/mL had a classification rate of 63.4%. Therefore, a BDG assay result >= 150 pg/mL has a sensitivity of 78.6% and a specificity of 41.4%. Conclusion: A BDG assay can be useful for ruling out invasive fungemia in post-operative patients.