Abstract
BACKGROUND: Among combat injured, invasive fungal infections (IFIs) result in significant morbidity. Cultures and histopathology are the primary diagnostic methods for IFIs, but they have limitations. We previously evaluated a panfungal polymerase chain reaction assay, which was 83% sensitive and 99% specific for angioinvasive IFIs. Here, we evaluated 3 less resource-intensive seminested assays targeting clinically relevant fungi in the order Mucorales and genera Aspergillus and Fusarium. METHODS: Formalin-fixed paraffin-embedded tissue specimens from a multicenter trauma IFI cohort (2009-2014) were used. Cases were US military personnel injured in Afghanistan with histopathologic IFI evidence. Controls were patients with similar injury patterns and no laboratory IFI evidence (negative culture and histopathology). Seminested assays specific to Mucorales (V4/V5 regions of 18S rDNA), Aspergillus (mitochondrial tRNA), and Fusarium (internal transcribed spacer [ITS]/28A regions of DNA) were compared with a panfungal assay amplifying the internal transcribed spacer 2 region of rDNA and to histopathology. RESULTS: Specimens from 92 injury sites (62 subjects) were compared with control specimens from 117 injuries (101 subjects). We observed substantial agreement between the seminested and panfungal assays overall, especially for the order Mucorales. Moderate agreement was observed at the genus level for Aspergillus and Fusarium. When compared with histopathology, sensitivity and specificity of seminested assays were 67.4% and 96.6%, respectively (sensitivity increased to 91.7% when restricted to sites with angioinvasion). CONCLUSIONS: Prior studies of seminested molecular diagnostics have focused on culture-negative samples from immunocompromised patients. Our findings underscore the utility of the seminested approach in diagnosing soft-tissue IFIs using formalin-fixed paraffin-embedded tissue samples, especially with angioinvasion.