Impact of Intraoperative C-arm 3D Imaging on the Diagnostic Yield of Robotic Shape-Sensing Navigation Transbronchial Biopsy

Background: Robotic navigational bronchoscopy using shape-sensing technology aids in diagnosing suspicious nodules. It can be used with different imaging modalities to improve accuracy. We aimed to determine the impact of C-arm 3-dimensional (3D) imaging on the diagnostic yield. Methods: In this single-center retrospective study, we assessed the accuracy of robotic navigation bronchoscopy-guided lung nodule biopsy by using C-arm 3D imaging. We collected demographic data, nodule imaging characteristics, procedural details, and pathology reports, analyzed patients with definitive diagnoses and those without, and followed up these patients for at least 1 year. Results: The study included 95 patients (median age, 69 years; 52% female; 67% current or former smokers) who underwent robotic bronchoscopy with C-arm 3D imaging. The median nodule size was 1.70 cm (interquartile range, 1.18-2.40 cm). A total of 55 nodules (58%) were described as spiculated, with most located in the right upper lobe (34%) and right lower lobe (23%). One patient (1%) experienced pneumothorax on postprocedure chest radiography, and it was managed with serial chest roentgenograms without a chest tube. Diagnoses included malignant disease (n = 52; 55%), benign conditions (n = 25; 26%), and no definitive diagnosis (n = 18; 19%). Among those without a definitive diagnosis, 6 patients (6%) were later confirmed to have cancer after surgical resection, whereas 12 patients (13%) showed no malignancy at the 12-month follow-up, resulting in 94% diagnostic accuracy at 12 months. Conclusions: Robotic shape-sensing navigation bronchoscopy combined with advanced C-arm 3D imaging yielded a high diagnostic accuracy with minimal complications. This approach is recommended for patients with suspicious lung nodules to enhance the diagnostic yield.