Background: This study sought to evaluate a single institution's experience using shape-sensing robotic-assisted bronchoscopy (ssRAB) for the diagnosis and management of pulmonary nodules.

Methods: All patients underwent ssRAB between September 2020 and May 2024. A procedure was defined as diagnostic if a benign or malignant result was obtained. A benign diagnosis required confirmation with repeat biopsy, resection, resolution, or 1-year follow-up. Logistic regression models were used to determine factors associated with a positive result of a diagnostic procedure. Surgical resection data were collected.

Results: A total of 1121 nodules were biopsied. The median procedure time was 67 minutes (interquartile range, 45-100 minutes), and the median nodule size was 2.1 cm (interquartile range, 1.3-3.3 cm). The diagnostic yield was 86.1%. The sensitivity and negative predictive value for malignant disease were 90.8% and 90.0%, respectively, whereas the specificity and positive predictive value for benign disease were 89.4% and 91.7%, respectively. In univariate analysis, the presence of a bronchus sign and radial endobronchial ultrasound were associated with an increased diagnostic yield (P = .05). In multivariate analysis, an increase by 1 cm in mean lesion size was associated with an increased diagnostic yield (odds ratio, 1.46; 95% CI, 1.85-1.15; P = .002). The pneumothorax rate was 3.3%, with 1.9% of patients requiring chest tube insertion. A total of 26.6% of patients underwent future surgical resection, and 104 patients gave consent for robotic bronchoscopy at the time of surgical resection. There were no mortalities and minimal complications (4.79%). The benign resection rate was 10.6%.

Conclusions: This report describes a large series of ssRAB procedures with long-term follow-up, demonstrating excellent diagnostic yield and safety profile. In select cases, ssRAB can effectively be combined with surgical resection.