Prospective real time margin assessment in laser cordectomy using Histolog® Scanner

Background and Objectives: Accurate intraoperaive assessment of surgical margins remains a critical challange in transoral laser microsurgery (TLM) for glottic squamous cell carcinoma. Conventional hematoxylin-eosin (H&E) histopathology, the current reference standard, is time-consuming and subject to laser-induced thermal artifacts and tissue shrinkage, which may generate false-positive results in up to 50% of cases. The Histolog® Scanner is an innovative digital confocal microscopy device capable of real-time imaging of fresh surgical specimens directly in the operating room. This prospective pilot diagnostic accuracy study aimed to perform a preliminary validation of the Histolog® Scanner for intraoperative margin assessment in patients undergoing transoral laser cordectomy, using conventional H&E histopathology as the reference standard. Materials and Methods: This study was conducted at the University Hospital of Modena between October2025 and March 2026, following STARD 2015 reporting guidelines. Adult patients undergoing TLM cordectomy (types I-VI) fot glottic lesions were prospectively enrolled. Five surgical margins per specimens (anterior, posterior, superior, inferior, and deep) were sistematically assessed using both the Histolog® Scanner and conventional histopathology. Margins were classified as positive (Rclose/R1) or negative (R0). Diagnostic performance was quantified by sensititvity, specificity, positive predictive value (PPV), negative predictive value (NPV), and overall accuracy, with 95% confidence intervals calculated using the Clopper–Pearson method. Agreement beyond chance was assessed by Cohen's kappa coefficient. A secondary per-patient analysis was also performed. Results: Of 24 consecutive patients screened, 19 were included in the final analysis after exclusion of 5 patients enrolled during the initial run-in phase. The cohort comprised 18 men and 1 woman (mean age 70 years). All 95 predefined margins were evaluable by both methods. Overall concordance between Histolog® Scanner and conventional histopathology was 92.6% (88/95 margins), with a Cohen's kappa of 0.73, indicating substantial agreement. The per-margin diagnostic performance showed a sensitivity of 85.7%, a specificity of 93.8%, a PPV of 70.6%, an NPV of 97.4%, and an overall accuracy of 92.6%. In the secondary per-patient analysis (n=19), sensitivity was 88.9%, specificity 90.0%, and accuracy 89.5%. Discussion and Conclusions: This study provides the first preliminary evidence supporting the feasibility and diagnostic potential of the Histolog® Scanner for real-time intraoperative margin assessment in TLM cordectomy. The device demonstrated high specificity and an outstanding NPV, confirming its reliability as a "rule-out" tool to minimize false-positive margin reports and unnecessary re-interventions. While it does not replace definitive histopathology , particularly for deep margin evaluation, it represents a promising adjunctive instrument to optimize intraoperative decision-making and surgeon-pathologist communication in laryngeal oncological surgery. Larger multicenter studies are warranted to validate these findings.