Background and Aims: Several studies have compared EUS fine-needle aspiration (FNA) with biopsy (FNB) needles, but none has proven superiority. We performed a multicenter randomized controlled trial to compare the performance of a commonly used 25-gauge FNA needle with a newly designed 20-gauge FNB needle.
Methods: Consecutive patients with a solid lesion were randomized in this international multicenter study between a 25-gauge FNA (EchoTip Ultra) or a 20-gauge FNB needle (ProCore). Primary endpoint was diagnostic accuracy for malignancy and the Bethesda classification (non-diagnostic, benign, atypical, malignant). Technical success, safety, and sample quality were also assessed. Multivariable and supplementary analyses were performed to adjust for confounders.
Results: A total of 608 patients were allocated to FNA (n=306) or FNB (n=302); 312 pancreatic lesions (51%), 147 lymph nodes (24%), and 149 other lesions (25%). Technical success rate was 100% for the 25-gauge FNA and 99% for the 20-gauge FNB needle (p=0.043), without differences in adverse events. The 20-gauge FNB needle outperformed 25-gauge FNA in terms of histological yield (77% vs 44%, p<0.001), accuracy for malignancy (87% vs 78%, p=0.002) and Bethesda classification (82% vs 72%, p=0.002). This was robust when corrected for indication, lesion size, number of passes, and presence of an on-site pathologist (OR, 3.53; 95% CI, 1.55-8.56; p=0.004), and did not differ between centers (p=0.836).
Conclusion: The 20-gauge FNB needle outperformed the 25-gauge FNA needle in terms of histological yield and diagnostic accuracy. This benefit was irrespective of the indication and consistent among participating centers, supporting the general applicability of our findings.