Background: The diagnostic value of rapid on-site evaluation (ROSE) of cytology during endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) remains controversial. The purpose of this study was to validate the value of EUBS-TBNA combined with ROSE in the diagnosis of peripheral pulmonary lesions (PPLs).Methods: Enrolled in this study were 260 patients with nodules, masses, cavities or inflammatory lesions on pulmonary CT images, and age ranged from 23 to 83 years old. They were assigned to undergo EBUS-TBNA with ROSE (n=134) or without ROSE (n=126). The diagnostic results of ROSE during EBUS-TBNA and the final pathologic reports were analyzed and compared by utilizing SPSS21.0 software to see whether the ROSE method during EBUS-TBNA would increase the risk of procedure-related complications, improve the diagnostic yield, and evaluate the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV). Results: The number of punctures in ROSE group was significantly lower than that in non-ROSE group (P<0.001). There was no significant difference in hemoptysis between the two groups (P=0.402). The overall diagnostic yield of EBUS-TBNA for malignant diseases in the ROSE-group and the non-ROSE group was 29.9% and 11.1%, respectively. The sensitivity, specificity, PPV and NPV of ROSE during EBUS-TBNA were 97.4%, 96.9%, 92.5% and 98.90% respectively, showing a high agreement with the pathological results. Youden index reached 94.3%. Conclusions: ROSE examination during EBUS-TBNA could effectively improve the diagnostic yield and reduce the number of intraoperative punctures with the diagnostic results in cytology well consistent with those of traditional pathology, and therefore is worth promoting and applying in clinical practice.