ObjectiveTo analyze the application value of rapid on-site evaluation by endoscopic physicians themselves （self-ROSE）of endoscopic ultrasound-guided fine needle aspiration (EUS-FNA) for solid pancreatic lesions. MethodsData of 124 consecutive patients who underwent EUS-FNA for solid pancreatic lesions from January 2017 to December 2017 in Nanjing Drum Tower Hospital were retrospectively analyzed. The patients were divided into self-ROSE group (n=60, patients who received self-ROSE from July to December in 2017) and non-self-ROSE group (n=64, patients who didn’t receive self-ROSE from January to June in 2017). Diagnostic efficiency including sensitivity, specificity, positive predictive value, negative predictive value and accuracy, the number of punctures and complication incidence of EUS-FNA in self-ROSE group and non-self-ROSE group was compared. ResultsThe sensitivity, specificity, positive predictive value, negative predictive value and accuracy of EUS-FNA were 98.21% (55/56), 4/4, 100.00% (55/55), 4/5 and 98.33% (59/60), respectively in self-ROSE group. While in non-self-ROSE group, the sensitivity, specificity, positive predictive value, negative predictive value and accuracy were 81.82% (45/55), 9/9, 100.00% (45/45), 47.37% (9/19) and 84.38% (54/64), respectively. The sensitivity (P=0.004) and accuracy (P=0.009) were both significantly higher during EUS-FNA in self-ROSE group than those in non-self-ROSE group. There was a high degree of consistency between endoscopists and pathologists in the evaluation of sample satisfaction (Kappa=1.000, P＜0.05) and cytopathological diagnosis (Kappa＝0.815, P＜0.05). There was no significant difference in the number of punctures between the two groups （2.78±0.83 VS 2.61±0.75, P＞0.05). No complication occurred in either group. ConclusionSelf-ROSE is valuable for EUS-FNA in the diagnosis of solid pancreatic lesions, which could be an important strategy to increase the accuracy of EUS-FNA.