Diagnostic Value of Concurrent Endoscopic Ultrasound-Guided Fine-Needle Aspiration and Fine-Needle Core Biopsy in Cytology Categories of Pancreatic Solid Lesions: An Institutional Experience

Abstract

Abstract Introduction/Objective Endoscopic ultrasound is the current standard of care in the evaluation of pancreatic solid lesions. Two methods have been employed in obtaining EUS-guided pancreatic specimens; fine needle aspiration (EUS-FNA) and fine needle core biopsy (EUS-FNB). EUS-FNA may have certain limitations, including the inability to assess the histological architecture and a limited sample size for immunohistochemical staining on cell blocks. EUS- FNB, which employs a larger needle for core biopsy, was developed to address these limitations. However, it is still uncertain which method is better suited for specific lesions. In this study, we evaluated the diagnostic efficacy of concurrent EUS-FNA and EUS-FNB from pancreatic solid lesions. Methods/Case Report We reviewed our institutional database for cases that had EUS-FNA from pancreatic solid lesions with a subsequent definitive diagnosis on histology. Similarly, the concurrent EUS-FNB findings were recorded. The cases were categorized based on the World Health Organization pancreaticobiliary cytology reporting system. Furthermore, the absolute risk of malignancy (ARM) of each diagnostic category (defined as the presence of high-grade dysplasia or worse on the histologic follow-up) was calculated. Results (if a Case Study enter NA) A total of 108 cases were reviewed for the study from 1/1/2019 to 6/1/2022. The patients were 47% male and 53% female with an average age of 68. No case was categorized as non-diagnostic or neoplastic high-grade. There were 20 benign, 7 atypical, 3 neoplastic low-grade, 6 suspicious for malignancy, and 72 malignant cases, and the associated ARM related to each category was 40%, 86%, 33%, 33%, and 96% respectively. Concurrent EUS-FNB was performed in 23 cases in total (12 in malignant and 11 in other categories). EUS-FNB exhibited 95% positive and 100% negative predictive values (PPV and NPV) in all categories combined and 89% PPV and 100% NPV in non-malignant and indeterminate categories (categories II, III, IV, VI) for correlating to the histologic follow-up. Conclusion Cases diagnosed in all categories except the malignant category demonstrated ARM rates higher than expected. When approaching a solid pancreatic lesion, EUS-FNB may increase the diagnostic value, especially in cases of non-malignant and indeterminate categories. Moreover, categorizing the cases in the benign category may require additional supporting data, and placing them in the nondiagnostic category may be more appropriate warranting re-sampling if there is a clinical vs pathologic discrepancy.