Abstract
BACKGROUND AND AIM: Advantages of endoscopic ultrasonography-guided tissue acquisition (EUS-TA) using a Lancet and Franseen needles have been evaluated and compared. However, little is known about the performance of each needle in diagnostic methods such as cytology and histology or the amount of blood contamination associated with each needle. This study aimed to compare the diagnostic yield and amount of blood contamination between two needles in patients with solid pancreatic lesions. METHODS: We collected data of consecutive patients who underwent first time EUS-TA of solid pancreatic lesions at Osaka Metropolitan University between Jan 2006 and Jan 2021 from the electronic medical records. We compared the main outcomes (histological diagnostic accuracy) between the Lancet and Franseen needle groups. The amounts of core tissue and blood contamination were evaluated using a scoring system. This retrospective comparative study was conducted at a single center. RESULTS: A total of 315 patients were divided into the Lancet (n = 200) and Franseen needle group (n = 115). The histological sensitivity and histological diagnostic accuracy of the Franseen needle was higher than that of Lancet needle (82.4% vs. 59.8%, respectively; odds ratio [OR], 2.29; 95% confidence interval [CI], 1.21-4.35; p = 0.011). Multivariate analysis using inverse probability of treat weighting method revealed that the diagnostic performance of the Franseen needle was the significantly higher than the Lancet needle (OR, 2.74; 95% CI, 1.31-5.74; p = 0.007). The core tissue score of 4 was achieved in 53.3% of the Franseen group and 3.3% of the Lancet group (p < 0.001), while high blood contamination was observed in 53.3% and 40%, respectively (p = 0.089). CONCLUSION: The histological accuracy and the amount of tissue by the Franseen needle are higher than those of the Lancet needle. Franseen needle could achieve high histological diagnostic accuracy even with the same blood contamination rate as that in Lancet needle.