ABSTRACT
BACKGROUND: Toxic epidermal necrolysis (TEN) is a severe life-threatening drug eruption with rapid evolution. A fast histologic differentiation between TEN and clinically similarly looking staphylococcal scalded skin syndrome is of vital importance for relevant treatment decision. The recently developed ex vivo confocal laser scanning microscopy (CLSM) offers innovative and extremely fast histological visualization of fresh tissue specimens. OBJECTIVE: To assess the diagnostic efficacy of ex vivo CLSM in comparison with standard histopathology for TEN. METHODS: We performed side-by-side comparison of TEN specimens analysed with ex vivo CLSM and haematoxylin and eosin staining. Analysis focused on typical histopathological features of TEN, including epidermal cleavage in the basal layer and confluent epidermal necrosis. We retrospectively assessed the diagnostic performance of ex vivo CLSM for TEN in clinically confirmed cases. RESULTS: We report substantial agreement between ex vivo CLSM and classical histology for the detection of subepidermal cleavage and confluent epidermal necrosis. When considering full-thickness epidermal loss, epidermal cleavage in the basal layer showed the highest diagnostic performance, reaching 87.5% sensitivity and 100% specificity. CONCLUSION: Based on our data, ex vivo CSLM appears as a rapid, resource-optimizing, and reliable approach for morphological TEN emergency screening on fresh skin samples.
Subject(s)
Stevens-Johnson Syndrome , Humans , Stevens-Johnson Syndrome/diagnosis , Stevens-Johnson Syndrome/pathology , Retrospective Studies , Skin/pathology , Microscopy, Confocal , NecrosisABSTRACT
Background: The assessment of surgical margins is mandatory to prevent local recurrence or distant dissemination of skin cancers. Histological examination of haematoxylin and eosin (H&E)-stained slides from paraffin-embedded or frozen samples is the gold standard for margin assessment, but is a time-consuming procedure. Ex vivo confocal laser scanning microscopy (CLSM) is an upcoming technique that scans unfixed fresh tissue rapidly, allowing fast per-operative margin assessment. Objective: Here, we propose to assess the efficiency of a new ex vivo confocal microscope for the per-operative assessment of surgical margins. Methods: We analyzed 16 biopsies and 93 surgical specimens of basal cell and squamous cell carcinomas by ex vivo CLSM using Histolog® Scanner V2. Surgical specimens included fusiform excisions, slow-Mohs peripheral and deep compartments, and Mohs excisions. The time required from surgical excision to image analysis was recorded and the quality of the images obtained for each specimen assessed. The presence or absence of tumour was estimated based on ex vivo CLSM images and compared with conventional H&E-stained sections from paraffin-embedded or frozen (Mohs) specimens. Results: Mean time for specimen processing using Histolog Scanner was 5.1 ± 3.4 min. We obtained 89% of high quality images. Mean time for confocal image analysis was 1 ± 0.76 min. The diagnostic sensitivity and specificity for ex vivo CLSM compared to classical H&E procedures were respectively 93% and 100% when performed on tumour biopsies. The overall sensitivity and specificity for ex vivo CLSM for margin assessment compared to classical H&E procedures were respectively 61.5% and 95%, with variations depending on the type of tumour or surgical specimen analyzed. In particular, we obtained 80% sensitivity and 100% specificity for the assessment of BCC surgical margins. Conclusion: Our data suggest that ex vivo CLSM using Histolog® Scanner V2 could be a valid help for surgeons for a fast and accurate per-operative margin analysis.
