Accuracy and inter-observer agreement of the Procore™ 25 gauge needle for endoscopic ultrasound-guided tissue core biopsy

BACKGROUND: Scanty data on the performance of the new 25-gauge Procore™ biopsy needle are available. METHODS: Consecutive patients who underwent endoscopic ultrasound-guided fine needle biopsy (EUS-FNB) using the 25G Procore™ were retrospectively retrieved. All samples were independently reviewed by 3 pathologists for the following: histological, cytological or no specimen, neoplasia, diagnostic or non-diagnostic. Diagnostic accuracy and inter-rater concordance among pathologists were calculated. RESULTS: 94 patients underwent EUS-FNB of 101 sites (69 solid masses, 25 lymph nodes, 5 wall thickening). Forty-one biopsies (40.5%) were classified as histological samples by at least two pathologists, 29 as cytological (28.7%), 31 had no sample (30.7%). Good and almost perfect agreements among pathologists in defining cytological vs. histological samples (k 0.82; 95% CI: 0.74-0.90), diagnostic vs. non-diagnostic (k 0.95; 95% CI: 0.85-1.00) and neoplastic vs. non-neoplastic (k 0.94; 95% CI: 0.83-1.00). According to consensus rating, 61 cases were diagnostic samples (60.4%). Histological samples were more likely to lead to a correct diagnosis (OR, 4.1; 95% P=0.027), while neoplastic lesions were less likely to be correctly classified than benign (OR, 0.11; P=0.04). CONCLUSIONS: EUS-FNB with the Procore™ 25G needle provided samples for histological examination in only 40% of the cases, with 31% of inadequate specimens, despite excellent results in term of inter-observer variability.