Diagnostic accuracy of fungal identification in histopathology and cytopathology specimens.

Tools to diagnose fungal infection are microscopic examination, antigen or antibody-based detection tests, molecular diagnostics, and culture, with culture being the "gold standard" for species-level identification. Although these methods are commonly used in concert and yield concordant results, in some cases tissue is not available for culture, and/or different methodologies yield discrepant results. These discrepancies may be clinically significant, causing confusion and inappropriate or delayed initiation of antifungals. This study evaluates the correlation between microscopic examination and the results of laboratory studies, and identifies clinical scenarios and specimen characteristics associated with tissue sent for microscopic examination without concomitant laboratory studies. We performed an 18-year retrospective review at a tertiary-care, academic medical center in the Midwest United States of all fungal infection diagnoses made by microscopic examination. Only 16% of samples with fungal infection diagnosed by microscopic examination had a concomitant sample submitted for laboratory studies. Of these cases, 36% had no growth on culture and/or had a negative laboratory study. Among cases in which fungal infections were diagnosed and laboratory studies were positive, the accuracy of histopathologic identification was 95%. The most common cause for incorrect morphologic diagnoses was misidentification of Aspergillus spp. and Mucorales. Our results underscore the importance of educating pathologists with regard to appropriate terminology and increasing knowledge of mycology, particularly in relation to organisms forming hyphae in tissue. Species-level diagnosis of fungi cannot be made by microscopic examination of tissue alone. Anatomic pathology reports should recommend correlation with laboratory studies, and provide a differential diagnosis based on morphology.