12 When Speed Counts: Molecular Diagnosis of Invasive Fungal Sinusitis. (11th January 2018)

• Record Type: Journal Article
• Authors: Lieberman, Joshua; Mathias, Patrick; Sengupta, Dhruba; Cookson, Brad; Schmidt, Rodney; Bryan, Andrew
• Is Part Of: American journal of clinical pathology. Volume 149(2018)Supplement 1
• Page Start: S169
• Page End: S170
• Publication Date: 2018-01-11
• Subjects: Diagnosis, Laboratory -- Periodicals; Pathology -- Periodicals; 616.07
• DOI: 10.1093/ajcp/aqx149.381
• Languages: English
• ISSNs: 0002-9173

Abstract

Abstract: Invasive fungal sinusitis (IFS) is a highly morbid infection. Diagnostic speed is essential as rapid initiation of antifungal therapy decreases mortality and optimal therapy can minimize exposure to nephrotoxic drugs. Diagnosis relies on both microbiologic and histopathologic evaluation. The first phase of this study tested three hypotheses: (1) PCR-based fungal diagnostic assays have shorter turnaround-time than culture; (2) PCR-based assays are as sensitive as culture; and (3) PCR assays performed on pathologist-selected blocks of formalin-fixed paraffin-embedded tissue (FFPE) are as sensitive as fresh tissue submitted without histologic review due to increased pretest probability. Fungal assays performed on sinus/paranasal sinus specimens January 1, 2011 through June 30, 2016 at the University of Washington Clinical Microbiology Laboratory using clinically validated CLIA protocols were identified and analyzed using the R software environment (RStudio v 0.99.902). Assays included broad-range fungal and Zygomycete group-specific identification by PCR and Sanger sequencing, as well as Aspergillus fumigatus -specific real-time PCR. We identified 1035 unique specimens collected from 669 patients. Cultures were performed on 916 specimens; 125 had PCR-based studies, including broad-range 28S/ITS PCR/sequencing assays in 107. Median turnaround-times were 104.1 hours for positive cultures and 48.9 hours for all PCR assays ( P < .01; Mann-Whitney). Among the PCR studies, median turnaround times for fresh vs FFPE tissue were 48.9 (75 samples) vs 69.6 hours (50 samples). Raw positivity rates were 11.9% for all cultures, 82% for PCR on fresh tissue and 58% for PCR on FFPE samples ( P < .05 for fresh vs FFPE PCR). These data confirm that time to organism identification by any molecular method, including multistep broad-spectrum fungal identification was significantly faster than traditional culture. In addition, molecular assays were positive ~6x more frequently than culture alone. However, PCR-based assays performed on FFPE yielded a result less often than fresh tissue in unmatched specimens possibly as a result of formalin fixation. Major issues that still need to be addressed include relative assay performance on matched fresh vs FFPE samples and the degree to which overall IFS diagnosis rates can be improved by optimum histopathologic assessment (eg, of invasion) and guidance of FFPE sample selection. To this end, we have initiated a second phase of this study based on 371 cases of histopathologic fungal sinusitis (67 definitively invasive) to identify histologic features and fungal characteristics that may alter the pretest probability of fungal identification by PCR. Together, we expect these data to form the basis for a standardized, efficient, pathologist-driven diagnostic algorithm for suspected invasive fungal infections.