Intern Med. 2026 Jul 11. doi: 10.2169/internalmedicine.7638-26. Online ahead of print.
ABSTRACT
Objective To evaluate the needle selection process for endoscopic ultrasound-guided tissue acquisition (EUS-TA) using 22-gauge needles in pancreatic solid lesions ≤20 mm. Methods Consecutive patients with pancreatic solid lesions ≤20 mm who underwent EUS-TA were retrospectively reviewed at a referral center. Only procedures that used 22-gauge needles were included. The overall diagnostic accuracy was compared between the fine-needle aspiration (FNA) and fine-needle biopsy (FNB) groups. Multivariable logistic regression and propensity score-matched analyses were performed. Patients or Materials A total of 348 patients with pancreatic solid lesions ≤20 mm were included (FNA, n=193; FNB, n=155). Results No statistically significant difference in the overall diagnostic accuracy was observed between the FNA and FNB groups (92.7% vs. 95.5%, p=0.27). The median number of needle passes was lower in the FNB group (2 vs. 3, p<0.001), and the histological yield was higher (92.9% vs. 84.5%, p=0.019). The needle type was not independently associated with diagnostic accuracy (OR 1.76, 95% CI 0.69-4.52; p=0.24). In lesions measuring ≤10 mm, subgroup analysis was limited by the small sample size of the study. The rarly adverse event rates were low in both groups. Conclusion In pancreatic solid lesions ≤20 mm, FNB showed a higher histological yield and required fewer needle passes, whereas no significant difference in the diagnostic accuracy was observed between FNA and FNB. Therefore, the routine replacement of FNA with FNB may not be justified when benign-malignant differentiation is sufficient. FNB may be preferable when histological architecture or immunohistochemical evaluation is required and is considered technically feasible.
PMID:42438026 | DOI:10.2169/internalmedicine.7638-26
