Histopathological techniques for the diagnosis of combat-related invasive fungal wound infections

Sarah M. Heaton, Amy C. Weintrob, Kevin Downing, Bryan Keenan, Deepak Aggarwal, Faraz Shaikh, David R. Tribble, Justin Wells

Effective management of trauma-related invasive fungal wound infections (IFIs) depends on early diagnosis and timely initiation of treatment. We evaluated the utility of routine staining, histochemical stains and frozen section for fungal element identification. A total of 383 histopathological specimens collected from 66 combat-injured United States military personnel with IFIs were independently reviewed by two pathologists. Both periodic acid-Schiff (PAS) and Gomori methenamine silver (GMS) stains were used on 74 specimens. The performance of the two special stains was compared against the finding of fungal elements via any histopathological method (ie, special stains or hematoxylin and eosin). In addition, the findings from frozen sections were compared against permanent sections. The GMS and PAS results were 84 % concordant (95 % confidence interval: 70 to 97 %). The false negative rate of fungal detection was 15 % for GMS and 44 % for PAS, suggesting that GMS was more sensitive; however, neither stain was statistically significantly superior for identifying fungal elements (p = 0.38). Moreover, 147 specimens had frozen sections performed, of which there was 87 % correlation with permanent sections (60 % sensitivity and 98 % specificity). In 27 permanent sections, corresponding cultures were available for comparison and 85 % concordance in general species identification was reported. The use of both stains does not have an added benefit for identifying fungal elements. Furthermore, while the high specificity of frozen section may aid in timely IFI diagnoses, it should not be used as a stand-alone method to guide therapy due to its low sensitivity.