Comparative study of frozen and permanent section for diagnosis of toxic epidermal necrolysis

Abstract

Abstract Background Toxic epidermal necrolysis (TEN) is a severe, life-threatening mucocutaneous reaction, causing widespread sloughing of skin and mucosal surfaces. Accurate and prompt diagnosis is essential for optimal management and subsequent outcome. In this study, frozen sections were used as a rapid examination for initial diagnosis of TEN, and the frozen section diagnoses were assessed compared with permanent sections. Methods One hundred patients of suspected TEN were referred to our burn unit, and 67 had sufficient clinical findings for frozen and permanent biopsies. The accuracy of frozen section relative to permanent section was evaluated by calculating diagnostic accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). And McNemar’s tests were used to analyze the difference between the two methods. Results Fifty-two specimens were classified as TEN by frozen section, 51 of which were confirmed by permanent biopsy. The exception was diagnosed as bullous pemphigoid on permanent section. Fifteen specimens were read as negative for TEN on frozen slides but 4 were changed to positive by permanent biopsy. Overall, the diagnostic accuracy of frozen section was 92.5%, with sensitivity and specificity 92.7% and 91.7% respectively. The positive predictive value, or coherence of positive diagnosis between the two methods, was as high as 98.1%, and the negative predictive value was 73.3%. The p-value of McNemar’s tests was 0.375, indicating there was no significant difference between the two biopsy methods. Conclusion The data suggest that as a rapid histological assessment, frozen section is a reliable tool in the early diagnosis of TEN.