Cone Beam CT Imaging-Guided Bronchoscopy vs CT Imaging-Guided Transthoracic Needle Biopsy for Peripheral Pulmonary Lesion Diagnosis

BACKGROUND: The identification of peripheral pulmonary lesions (PPLs) has increased significantly, either incidentally or through lung cancer screening, necessitating more biopsies to differentiate between malignant and benign causes. Cone beam CT imaging-guided bronchoscopic biopsy (CBCT-GB) and CT imaging-guided transthoracic needle biopsy (CT-TTNB) are used commonly for these biopsies, but their comparative diagnostic abilities have not been studied. RESEARCH QUESTION: How does CBCT-GB compare with CT-TTNB for diagnosing PPLs? STUDY DESIGN AND METHODS: This single-center retrospective comparative cohort study analyzed PPLs biopsied at an academic center via either CBCT-GB or CT-TTNB. The primary outcome was diagnostic accuracy at the 24-month follow-up, defined as the proportion of cases yielding a specific diagnosis (malignant or nonmalignant) or a nonspecific diagnosis that remained accurate through 24 months of clinical follow-up. Secondary outcomes included complication rates, procedure duration, radiation exposure, and the need for additional diagnostic procedures. RESULTS: Of 895 patients analyzed, 340 of 375 patients (90.7%) in the CBCT-GB group and 440 of 475 patients (92.6%) in the CT-TTNB group had a diagnostic result (P = .301; OR, 0.979; 95% CI, 0.939-1.020). Complications occurred in 4.3% of patients undergoing CBCT-GB and 41.6% of patients undergoing CT-TTNB (P < .001). Pneumothorax rates were 1.8% for CBCT-GB and 31.4% for CT-TTNB (P < .001), and severe bleeding or cardiorespiratory failure occurred in 3.3% and 6.0% of patients, respectively (P < .001). Among patients meeting criteria for upfront invasive mediastinal staging, 86.5% of patients undergoing CBCT-GB received it at the time of PPL biopsy, compared with 14.0% of patients after biopsy in the CT-TTNB group (P < .001). Median effective radiation dose was 8.6 mSv in the robotic bronchoscopy CBCT-GB group and 7.5 mSv in the CT-TTNB group (P = .074). INTERPRETATION: Our results show that CBCT-GB demonstrated a 24-month diagnostic accuracy comparable with that of CT-TTNB while offering improved safety and concurrent mediastinal lymph node staging. These data support CBCT-GB as the optimal initial procedure for PPL diagnosis.