The Effectiveness Of Rapid On-Site Cytological Evaluation (Rose) On The Diagnostic Yield Of Bronchoscopy In Peripheral Pulmonary Lesions: A Systematic Review And Meta-Analysis

Background: Peripheral pulmonary lesions (PPLs) cannot be directly visualized during bronchoscopy sampling, and the quality of specimens affects the diagnostic yield. The rapid on-site cytological evaluation (ROSE) system provides immediate feedback regarding the adequacy of specimens and guides the operators to modify the bronchoscopy technique and the site and depth sampled. However whether ROSE can increase the diagnostic yield of bronchoscopy in PPLs sampling has not been systematically examined. Methods: We comprehensively searched PubMed, EMBASE and relevant reviews up through April 2018 and screened for studies investigating the effectiveness of ROSE on the diagnostic yield of bronchoscopy in PPLs. Results: 15 out of 125 studies (4035 patients from 7 countries) were eligible for qualitative analysis. The pooled diagnostic yield of all included studies was 0.84 (95% CI 0.77-0.90). ROSE significantly increased the diagnostic yield versus diagnosis without ROSE (RD 0.15, 95% CI, 0.12-0.18). The pooled yield was 0.80 (95% CI 0.62-0.91) when bronchoscopy was guided by fluoroscopy, 0.85 (95% CI 0.78-0.90) when EBUS was used and 0.85 (95% CI 0.78-0.90) when ENB was used. When the lesions > 2 cm, the pooled diagnostic yield was 0.90 (95% CI 0.87-0.93), while the yield was 0.79 (95% CI 0.72-0.84) when the lesions <= 2 cm. Conclusion: The use of ROSE increased the diagnostic yield of bronchoscopy in PPLs diagnosis, particularly in the context of lesions <= 2 cm, fluoroscopy-guided and EBUS-guided bronchoscopy, especially when the probe was adjacent to the lesions in EBUS-guided bronchoscopy.