Robotic <i>versus</i> CBCT navigation bronchoscopy: propensity matched analysis of diagnostic yield

Objective Small peripheral pulmonary lesions are nowadays preferably diagnosed by navigational bronchoscopy, yet reported diagnostic yields vary across different techniques. Shape sensing robotic assisted bronchoscopy (ssRAB), now also available in Europe, combines real-time shape sensing with an actively steerable catheter, potentially improving diagnostic yield. We aimed to compare ssRAB combined with CBCT imaging (ssRAB+CBCT) against our current standard, CBCT-based navigation bronchoscopy alone (CBCT-NB). Methods We conducted a single-center, propensity score–matched analysis comparing patients undergoing ssRAB+CBCT versus CBCT-NB for the diagnosis of small peripheral pulmonary lesions. Matching was performed on known lesion characteristics influencing yield. Primary outcome was strict diagnostic yield. Secondary outcomes included diagnostic accuracy at follow-up, safety, and procedure-related metrics. Results A total of 131 patients with 183 biopsied lesions were included in the ssRAB+CBCT arm. Median lesion size was 12 millimeter (8–18). Propensity score matching with lesions from our reference CBCT-NB-cohort was successful in 150 out of 183 lesions. The diagnostic yield on lesion level was 73% for ssRAB+CBCT and 70% for CBCT-NB, the mean difference of 3.3% (95% CI: −6.9 to 13.5) was not statistically significant (p=0.521). The diagnostic yield on patient level for ssRAB+CBCT was 82%. Conclusion ssRAB+CBCT has similar diagnostic yield and below the study's powering assumption of 15% increase when compared to our highly optimized CBCT-NB program, with a signal toward benefit in small bronchus-sign–negative nodules. The unique features of ssRAB hold promise but larger studies are warranted to clarify its position and optimal case-selection compared to other navigation bronchoscopy technologies, clinical impact and cost-effectiveness.