Related Experiment Video
Updated: Jun 6, 2026
Robotic-assisted Bronchoscopy Combined with Multimodal Imaging for Targeted Lung Cryobiopsies
Published on: July 19, 2024
Diagnostic Outcomes of Robotic Bronchoscopy in a Multidisciplinary, Multi-hospital Lung Nodule Program
Rachel L Deitz1, John P Ryan1, Thomas Rubino1
1Department of Cardiothoracic Surgery, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania.
Introduction:
In many centers, robotic bronchoscopy has become the standard approach for diagnosis of lung lesions. Smaller series have demonstrated improvements in diagnostic accuracy. The aim of our study was to assess the diagnostic accuracy of robotic bronchoscopy across multiple users in multiple hospitals within a single health system with the largest patient cohort in the literature to date.
Methods:
All patients with a suspicious intraparenchymal lung lesion that was biopsied using the Monarch robotic bronchoscope by a pulmonologist or thoracic surgeon at a multi-site, academic institution from August 2018 to April 2023 were included. Data were collected prospectively, and patients were followed postprocedure from at least 1 and up to 5 y. Diagnostic accuracy was determined by comparing Monarch-acquired samples to final known pathology. True negative (benign) diagnoses were confirmed with either resolution or stability on computed tomography scan at one-year post-procedure.
Results:
Over the five-year period, 1084 lesions were biopsied via robotic bronchoscopy (Monarch) by 16 practitioners. Lesions were primarily solid (84%) and peripherally located (62.0%), with a median size at 25 mm (17-37). At median follow-up of 1.07 (interquartile range 0.58-1.67) years, diagnostic accuracy was 82.7% (80.1-85.2). In univariate analysis, larger lesion size (P < 0.001), peripheral location (P < 0.001), bronchus sign on computed tomography scan (P < 0.001), intraoperative direct endobronchial visualization (P < 0.001), and use of radial endobronchial ultrasound (P < 0.001) were all associated with successful diagnosis. Clinically significant pneumothoraces (requiring chest tube) were low at 3% (n = 33).
Conclusions:
Diagnostic accuracy from this large cohort, cross-specialty analysis of robotic bronchoscopy among multiple operators was 82.7%. Robotic bronchoscopy can be successfully employed as the initial and preferred sampling methodology for solid lung lesions given acceptable diagnostic performance and low complication rates in this long-term assessment of outcomes.
