Supplemental Material for: Diagnostic Performance and Safety Profile of Shape-sensing Robotic-assisted Bronchoscopy in Peripheral Pulmonary Lesions: A Systematic Review and Meta-Analysis

PubMed, Embase, and Cochrane Library were searched from inception to December 2025. Study quality was assessed using the Quality Assessment of Diagnostic Accuracy Studies-3. Random-effects model was used to estimate the pooled diagnostic yield, sensitivity for malignancy, and complication, with rates (%) and 95% confidence intervals (CI) when the I2 index exceeded 50%.

Cumulative meta-analysis was performed to assess the temporal stability and evidence saturation of the pooled diagnostic yield. Publication bias was evaluated using funnel plots and Egger's test. Results: A total of 3678 lesions from 18 studies were included.

Notably, diagnostic yield varied significantly by outcome definition, with pooled diagnostic yield of 82.5% (95% CI: 78.3%-86.0%) under strict definition versus 88.1% (95% CI: 82.8%-92.0%) under intermediate definition (P=0.036). The pooled sensitivity for malignancy of 90.0% (95% CI: 86.4%-92.8%). Cumulative meta-analysis showed stabilized diagnostic performance after approximately 1,500 lesions for the strict definition and 500 lesions for intermediate definition.

Moreover, safety profile was favorable, with pooled rates of pneumothorax 1.8% (95% CI:1.3%-2.3%), chest tube placement 1.1% (95% CI: 0.7%-1.5%), and severe bleeding 0.0% (95% CI: 0.0%-0.3%). Subgroup analyses showed consistent performance across study designs, region, and nodule characteristics. Diagnostic yield was significantly influenced by publication year, and lesion size (P<0.05).

Conclusion: ssRAB was associated with favorable, definition-dependent diagnostic yield (82.5%-88.1%) with exceptional safety, supporting ssRAB as an attractive first-line minimally invasive diagnostic option in appropriately selected patients, especially for small nodules (<2cm) and patients at elevated procedural risk.</p>