Macroscopic Tissue Integrity and Histological Diagnostic Yield During Endoscopic Ultrasound-Guided Tissue Acquisition for Solid Pancreatic Lesions

Background and Aim: Macroscopic on-site evaluation (MOSE) is being increasingly used during endoscopic ultrasound (EUS)-guided tissue acquisition when rapid on-site evaluation (ROSE) is unavailable. However, current MOSE criteria are primarily based on the presence or length of visible tissue fragments. The potential diagnostic significance of a macroscopically intact tissue specimen with preserved gross integrity has not been adequately investigated. This study evaluated the relationship between a macroscopically intact cord-like biopsy specimen (cord sign) and pathological diagnostic outcomes in solid pancreatic lesions. Methods: This retrospective cohort study included 380 patients who underwent EUS-guided biopsy for solid pancreatic lesions between January 2021 and December 2025. Cord sign positivity was defined as retrieval of a single intact tissue fragment measuring at least 1 cm with preserved macroscopic integrity and was assessed independently in formalin-fixed block and smear specimens. The primary outcome was histological diagnostic yield. Secondary outcomes included specimen adequacy, histological malignant or suspicious diagnosis, distribution of histological diagnoses according to overall cord sign status, and overall diagnostic success. Multivariable logistic regression analyses were performed to evaluate factors associated with histological diagnostic yield and, in an additional exploratory analysis, histological malignant or suspicious diagnosis. Results: A total of 380 patients were included in the final analysis, including 338 block cord sign-positive and 42 block cord sign-negative cases. Histological diagnostic yield was numerically higher in the block cord sign-positive group (82.2% vs. 71.4%, p = 0.098), although the difference was not statistically significant. Histological malignant or suspicious diagnosis was significantly more frequent among block cord sign-positive specimens (54.7% vs. 28.6%, p = 0.002), whereas specimen adequacy and combined diagnostic success were comparable between the groups. In the primary multivariable analysis, block cord sign positivity was not independently associated with histological diagnostic yield (OR 1.63, 95% CI 0.75–3.55, p = 0.218). However, an additional exploratory multivariable analysis demonstrated an association between block cord sign positivity and malignant or suspicious histological diagnosis (OR 3.44, 95% CI 1.56–7.59, p = 0.002). Conclusions: Block cord sign positivity was associated with a numerically higher histological diagnostic yield in patients undergoing endoscopic ultrasound-guided tissue acquisition for solid pancreatic lesions, although this association did not reach statistical significance in either the unadjusted or adjusted analyses. In contrast, malignant or suspicious histological findings were more frequently observed among block cord sign-positive specimens, and this association remained significant in an additional exploratory multivariable analysis. However, the potential influence of underlying lesion characteristics and the absence of an independently established final diagnosis prevent these findings from being interpreted as evidence that cord sign independently reflects specimen quality or predicts malignancy. Cord sign represents a practical macroscopic observation based on specimen length and preserved tissue integrity rather than a previously validated assessment criterion. Although its evaluation in both formalin-fixed block and smear specimens may provide a useful framework for further investigation, the present findings do not establish its role in guiding additional needle passes or improving diagnostic performance. Prospective multicenter studies using standardized and reproducible criteria, including formal interobserver agreement assessment, are needed to clarify its clinical relevance.