Comparison of ex-vivo confocal micrographic surgery of basal cell carcinoma and conventional frozen section Mohs surgery (Munich method)

Introduction A new, reliable way to confidently diagnose basal cell carcinoma (BCC) out of fresh or frozen tissue is ex vivo confocal microscopy (CLSM). BCC belongs to the most frequently occurring epithelial skin tumours in humans and its incidence is increasing year by year. The most efficient treatment of BCC is micrographic surgery on frozen sections with Mohs surgery. Using a confocal microscope based on two different lasers with varying wavelengths, images in reflectance (RM) and fluorescence mode (FM) can be generated. In addition to that, the newest generation allows creating an overlap between the earlier mentioned modes combined with a staining software, making direct comparison to conventional histologic haematoxylin & eosin (HE) staining possible. Images produced by RM are based on the natural contrast of subcellular structures, whereas FM images have an increased cell-to-stroma contrast due to fluorochromes. With a reported sensitivity of 88% and a specificity of 99% for the diagnose of BCC on fresh tissue, CLSM has shown to be an attractive alternative to conventional frozen histology for detecting margins in freshly excised tissues. Furthermore CLSM can be applied during surgery as a so-called “bedside histology” and since there is no tissue damaged during the ex-vivo CLSM examination the analyzed samples can still undergo the process of standard histopathological evaluation. Material and Methods In this ongoing study, fifty-seven BCCs 35 microns thick horizontal frozen sections of primary BCCs stained with acridine orange were analysed with ex vivo CLSM. Complete or partial resection, histological subtype, as well as presence of adnexal structures and artefacts were evaluated. Features were compared with conventional H&E of the same sample. Results Ex-vivo CLSM reached a sensitivity of 90% and a specificity of 95% in the diagnosis of BCC on frozen sections, being able to recognize the presence of the tumour and its margins. Following features were correctly recognized in CLSM on frozen sections: palisading, clefting, stromal reaction, adnexal structures, fat tissue, muscle, scars. The tumour subtype was correctly defined. The most common artefact was a blurred margin, probably due to acridine orange accumulation. Discussion Our preliminary results confirm the high potential of CLSM for perioperative BCC diagnosis as well as margin assessment on frozen sections. The new generation device adds overlap images and digital staining, for a more accurate comparison to conventional histopathological H&E staining. This brings to a better definition not only of BCCs but also of other structures such as adnexa. Time sparing in comparison to standard frozen sections, CLSM allows bedside surgery, immediate reoperations or one-step wound closures. As a consequence, the economic burden of multiple operations can be decreased. We believe in CLSM will gain importance in dermato-surgery in the next future, with its use not only narrowed to micrographic surgery of BCCs but also of other skin tumours. The new digital staining tool can improve the diagnostic accuracy thanks to the intuitive comparison with H&E staining in conventional histopathology.