Interobserver agreement among pathologists regarding core tissue specimens obtained with a new endoscopic ultrasound histology needle; a prospective multicentre study in 50 cases

Aim: To evaluate the interobserver agreement among pathologists in grading the quality of specimens obtained with a new 19-gauge endoscopic ultrasound histology needle. Methods and results: This multicentre prospective study involved 50 slides prepared using material obtained with the new needle. Five experienced pathologists independently reviewed all of the samples, and made assessments of the following features: the presence of a core, the adequacy of the specimen, the interpretability of the specimen, and the possibility of performing additional analyses using the material. Interobserver agreement, determined by Fleiss' kappa statistic and 95% confidence intervals (CIs), was used as the primary outcome measure. Overall, the presence of a core was reported in 88% of cases with good agreement among the pathologists (κ = 0.61; 95% CI 0.52-0.70). The specimens were adequate in 91.2% of cases, and Fleiss' κ was 0.73 (95% CI 0.61-0.81). The interpretation of the specimens was reported to be 'easy' in approximately 87% of cases, with moderate agreement among the pathologists (κ = 0.44; 95% CI 0.35-0.53). The possibility of performing additional analyses from the same sample was rated as positive in approximately 91%, with good agreement (κ = 0.66; 95% CI 0.58-0.75). Conclusions: There was excellent interobserver agreement among pathologists in the assessment of the histological material, especially with regard to sample adequacy.