← Back to Archives

Comparison of Procedural Safety and Diagnostic Yield Between Robotic-Assisted Bronchoscopy Platforms.

IPBronch Review

🎯 Background & Rationale

Robotic-assisted bronchoscopy (RAB) platforms (specifically the Ion and Monarch systems) have become ubiquitous for sampling peripheral pulmonary nodules. Despite their widespread adoption, a head-to-head comparison evaluating performance differences in procedural safety and diagnostic yield between these two major platforms had not been investigated, as most prior data stem from platform-specific studies with inconsistent definitions of diagnostic yield.

👥 Study Design & Population

  • Study Type: Retrospective, 2-site cohort study.
  • Population: 217 adult patients (122 in the Ion group at UPMC; 95 in the Monarch group at VA Pittsburgh) undergoing outpatient or inpatient RAB for peripheral pulmonary nodules between August 2020 and August 2023.
  • Intervention: Robotic-assisted bronchoscopy performed using either the Ion or Monarch platform, assisted by standard 2D fluoroscopy (cone-beam CT and cryobiopsy were not available).

📈 Methodology & Rigor

The authors utilized an unadjusted logistic regression model, a parsimonious adjusted model (controlling for nodule size, bronchus sign, and the Brock University cancer prediction score), and an exploratory model with additional covariates (age, smoking status, spiculations). Rigorous sensitivity analyses were performed, including excluding the initial 20 cases for each platform, restricting the cohort to patients with a completed PET scan, and nearest-neighbor matching via Mahalanobis distance. A strict, consensus-based definition of diagnostic yield was applied, and all nondiagnostic cases underwent 24-month clinical/radiographic follow-up.

🔬 Key Findings [or Planned Endpoints]

  • Diagnostic Yield: The overall cohort diagnostic yield was 56% (Ion group: 53%; Monarch group: 46%). Sub-centimeter nodules (20% of the cohort) had a yield of 35%.
  • Platform Comparison: Neither the unadjusted, parsimonious, nor exploratory models showed a statistically significant difference in diagnostic yield between the Ion and Monarch platforms (Parsimonious OR: 1.5 [0.9–2.7], $P=0.15$). A small advantage for Ion was noted only in the nearest-neighbor matched sensitivity analysis.
  • Predictors of Success: Nodule size [OR: 1.42 (1.09–1.86)] and the presence of a bronchus sign [OR: 2.14 (1.14–4.01)] correlated positively with diagnostic yield.
  • Safety: Complications were exceedingly rare across both platforms, with only 2 direct pneumothoraces (1%) and grade 4 bleeding occurring in 1% of patients.

⚖️ Critical Appraisal

The study's strengths include its head-to-head design utilizing operators credentialed on both platforms, a conservative consensus-based definition of diagnostic yield, and a 24-month follow-up. Limitations include its retrospective nature, reliance on medical record documentation (which lacked granular data on biopsy tool order, precise pass counts, and radial EBUS use), and baseline demographic/clinical differences between the two hospital sites (VA vs. university hospital). Furthermore, the lack of advanced imaging modalities like cone-beam CT limits generalizability to centers utilizing modern imaging-integrated workflows.

💡 The Clinical Bottom Line

In the absence of advanced imaging adjuncts (like cone-beam CT), there is no significant difference in diagnostic yield or safety profile between the Ion and Monarch robotic bronchoscopy platforms. Proceduralists should recognize that nodule size and the presence of a bronchus sign remain the primary drivers of procedural success, and low-complexity fluoroscopic RAB achieves safe, acceptable real-world yields when sampling small peripheral pulmonary lesions.


RATIONALE: Robotic-assisted bronchoscopy platforms are becoming ubiquitous to aid in sampling peripheral pulmonary nodules. A head-to-head performance between the 2 widely available platforms (Ion and Monarch) has not been investigated. OBJECTIVES: To explore differences in diagnostic yield and safety parameters of the Ion and Monarch platforms. A secondary objective was to elucidate radiographic features of nodules that predicted procedural success. METHODS: A retrospective, 2-site cohort study of adult patients scheduled for robotic-assisted bronchoscopy of pulmonary nodules between August 2020 and August 2023. The procedure was performed with either the Ion or Monarch robotic-assisted bronchoscopy platform, depending on the site, with the assistance of standard 2D fluoroscopy. A series of logistic regression models was constructed to explore the effect of platform type on diagnostic yield, after correction of prespecified confounders. We utilized an inferential lasso regression model to explore the predictive properties of nodule radiographic characteristics on yield. RESULTS: A total of 217 cases were analyzed. The median nodule size was 1.4 cm (IQR=1.1 to 2.1), which was similar across both groups. Our parsimonious model suggested no difference in obtaining a positive diagnosis for the Ion platform [OR: 1.5 (0.9-2.7)]. Both nodule size [OR: 1.42 (1.09-1.86)] and presence of a bronchus sign [OR: 2.14 (1.14-4.01)] correlated with diagnostic yield in our inferential model, both in systems and in aggregate. CONCLUSION: Both platforms offer promise in sampling small peripheral nodules. Neither platform appears to improve procedural yield, although caution is advised regarding the generalizability of this finding. Both nodule size and the presence of the bronchus sign may be helpful in preprocedural patient selection.
Read Full Text at Publisher ↗