Objective To systematically review the value of rapid on-site evaluation (ROSE) for diagnosing pulmonary and mediastinal lesions with endobronchial ultrasound (EBUS). MethodsPubMed, EMbase, The Cochrane Library, Web of Science, CNKI, Wanfang, and VIP databases were searched by computer to collect the studies of ROSE and EBUS in the diagnosis of pulmonary and mediastinal lesions from inception to August 2022. Two researchers independently screened the literature, extracted the data, and evaluated the risk of bias in the included studies. Meta-analysis was implemented by RevMan 5.4 and Stata 12.0 software. ResultsA total of 15 studies (9 retrospective studies and 6 prospective studies) with 3 577 patients were included. The meta-analysis results of main outcomes showed that the adequacy of the sample (RD=0.10, 95%CI 0.05 to 0.15, P<0.000 1), overall diagnosis rate (RD=0.07, 95%CI 0.04 to 0.10, P<0.000 1) and the diagnosis rate of the malignant lesion (RD=0.06, 95%CI 0.02 to 0.09, P=0.004) of the ROSE combined with EBUS group were significantly higher than those of the EBUS group. Subgroup analysis showed that the diagnosis rates of pulmonary lesions (RD=0.12, 95%CI 0.08 to 0.17, P<0.000 01) and mediastinal lesions (RD=0.06, 95%CI 0.01 to 0.12, P=0.02) in the ROSE group was significantly higher than those in the EBUS group. The overall diagnosis rate and malignant diagnosis rate of ROSE combined with EBUS were 90% and 92%. The meta-analysis results of secondary outcomes showed that the number of lesions punctures (MD=–1.16, 95%CI –1.89 to –0.43, P=0.002) in the ROSE combined with EBUS group were significantly less than that in the EBUS group; there was no statistical difference in operation time (MD=0.09, 95%CI –5.22 to 5.39, P=0.97) or incidence of complications (RD=–0.06, 95%CI –0.13 to 0.01, P=0.1) between the two groups. Conclusion ROSE can improve the diagnostic efficiency of EBUS in pulmonary and mediastinal lesions, and has the value of the clinical application.