Endobronchial Ultrasound for the Diagnosis of Centrally Located Lung Tumors: A Systematic Review and Meta-Analysis

Introduction: Obtaining a tissue diagnosis of centrally located lung tumors in patients presenting without endobronchial abnormalities is challenging, and therefore a considerable diagnostic problem. Objective: The objective of this study was to evaluate the performance of linear endobronchial ultrasound guided-transbronchial-needle aspiration (EBUS-TBNA) for the diagnosis of centrally located lung tumors. Methods: We performed a systematic review (PROSPERO, CRD42017080968) and searched MEDLINE, Embase, BIOSIS Previews, and Web of Science till November 18, 2018 for studies that evaluated the yield and/or sensitivity of EBUS-TBNA for diagnosing centrally located lung tumors. We assessed the study quality using QUADAS-2 and performed random-effects meta-analysis. Results: A total of 5,657 manuscripts were identified; of these 14 were considered for the study, including 1,175 patients who underwent EBUS-TBNA for diagnosing an intrapulmonary tumor. All studies had a high risk of bias or applicability concerns, predominately regarding patient selection. The average yield of EBUS-TBNA for diagnosing centrally located lung tumors was 0.89 (95% CI 0.84-0.92) and average sensitivity for diagnosing malignant tumors was 0.91 (95% CI 0.88-0.94). Among studies reporting this information, EBUS-related complications occurred in 5.4% of patients (42/721). Conclusion: EBUS-TBNA has a high yield and sensitivity for diagnosing centrally located lung tumors and is safe in selected patients. Prospective studies are recommended to evaluate the routine use of this procedure for diagnosing intrapulmonary tumors.