Sensitivity and Safety of Electromagnetic Navigation Bronchoscopy for Lung Cancer Diagnosis: Systematic Review and Meta-analysis.

ABSTRACT

BACKGROUND: Bronchoscopy is a useful tool for the diagnosis of lesions near central airways; however, the diagnostic accuracy of these procedures for peripheral pulmonary lesions (PPLs) is a matter of ongoing debate. In this setting, electromagnetic navigation bronchoscopy (ENB) is a technique used to navigate and obtain samples from these lesions. This systematic review and meta-analysis aims to explore the sensitivity of ENB in patients with PPLs suspected of lung cancer. RESEARCH QUESTION In patients with peripheral pulmonary lesion suspected of lung cancer, what is the sensitivity and safety of electromagnetic navigation bronchoscopy compared to surgery or longitudinal follow up? STUDY DESIGN AND METHODS: A comprehensive search of several databases was performed. Extracted data included sensitivity of ENB for malignancy, adequacy of the tissue sample, and complications. The study quality was assessed using the QUADAS-2 tool, and the combined data were meta-analyzed using a bivariate method model. A summary receiver operatic characteristic curve (sROC) was created. Finally, the quality of evidence was rated using the Grading of Recommendations Assessment, Development and Evaluation approach. RESULTS: Forty studies with a total of 3,342 participants were included in our analysis. ENB reported a pooled sensitivity of 77% (95% CI, 72%-82%; I2 = 80.6%) and a specificity of 100% (95% CI, 99%-100%; I2 = 0%) for malignancy. The sROC showed an area under the curve of 0.955 (P = .03). ENB achieved a sufficient sample for ancillary tests in 90.9% (95% CI, 84.8%-96.9%; I2 = 80.7%). Risk of pneumothorax was 2.0% (95% CI, 1.0-3.0; I2 = 45.2%). We found subgroup differences according to the risk of bias and the number of sampling techniques. Meta-regression showed an association between sensitivity and the mean distance of the sensor tip to the center of the nodule, the number of tissue sampling techniques, and the cancer prevalence in the study. INTERPRETATION: ENB is very safe with good sensitivity for diagnosing malignancy in patients with PPLs. The applicability of our findings is limited because most studies were done with the superDimension navigation system and heterogeneity was high. TRIAL REGISTRY PROSPERO; No. CRD42019109449; URL https//www.crd.york.ac.uk/prospero/.