Background: Cone-beam computed tomography (CBCT)-guided bronchoscopy is increasingly utilized for diagnosing peripheral pulmonary lesions (PPLs). We carried out the meta-analysis for assessing the efficacy and safety of CBCT-guided bronchoscopy for PPLs.

Methods: An extensive search in several databases was conducted to identify relevant articles. We evaluated the quality of studies with the Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) tool. The pooled diagnostic yield (DY) and adverse event rate with the 95% confidence interval (CI) were computed. Subgroup analyses were performed according to additional use of navigation, use of radial endobronchial ultrasound (rEBUS), use of fixed or mobile CBCT, whether computed tomography (CT) spin was performed before biopsy to affirm tool-in-lesion, use of rapid onsite cytologic examination (ROSE), strictness of the definition of DY, and study design. Further analysis was performed to explore the association between odds of diagnosis with CBCT guided bronchoscopy and PPLs characteristics (>20 vs. ≤20 mm, non-upper lobe vs. upper lobe, with bronchus sign vs. without bronchus sign, and solid vs. non-solid) as well as sampling methods (forceps vs. fine needle aspiration, forceps vs. cryoprobe sampling). The pooled odds ratio (OR) and 95% CI were calculated. The significance level was set at 0.05. All analyses were performed by using meta package in R version 4.3.2.

Results: We included 23 studies involving 1,769 patients and 1,863 PPLs in the meta-analysis. The overall pooled DY of CBCT-guided bronchoscopy was 80.2% (95% CI: 76.0–84.1%). Subgroup analysis showed that the DY was highest when CBCT was used with robotic-assisted navigation bronchoscopy (pooled DY 87.5%; 95% CI: 81.5–92.4%), the DY was 78.9% (95% CI: 70.8–85.9%) when CBCT was used alone without other navigation techniques. Lesion size >20 mm, presence of bronchus sign and solid lesions were associated with significant increase in the odds of diagnosis with CBCT-guided bronchoscopy. Pooled adverse event rate was 2.3% (95% CI: 1.2–3.6%).

Conclusions: CBCT-guided bronchoscopy is a safe technique with high DY in diagnosing PPLs.

Keywords: Cone-beam computed tomography (CBCT); bronchoscopy; peripheral pulmonary lesions (PPLs); diagnostic yield (DY)