🎯 Background & Rationale
Augmented fluoroscopy (AF) overlays CT-derived virtual targets onto live fluoroscopic images, offering a cost-effective, radiation-sparing alternative for real-time guidance during navigational bronchoscopy of peripheral pulmonary lesions (PPLs) in centers without cone-beam CT (CBCT) availability. However, the diagnostic utility of AF remains poorly defined. Specifically, the predictive value of confirming the tool-in-virtual-lesion (TIVL) across multiple fluoroscopic projection planes has not been systematically investigated, particularly in challenging scenarios such as when radial endobronchial ultrasound (R-EBUS) signals are absent.
👥 Study Design & Population
- Publication Type: Retrospective Cohort Study.
- Population: 226 AF-assisted bronchoscopic procedures for PPLs performed between December 2019 and March 2024 utilizing virtual bronchoscopic navigation (VBN) and R-EBUS at participating centers (University Medicine Essen-Ruhrlandklinik, Lungenklinik Hemer, University Hospital Essen, and Johns Hopkins University).
- Intervention: Augmented fluoroscopy-guided navigational bronchoscopy with TIVL assessed across five standard fluoroscopic planes (postero-anterior, 30° towards, 30° away, 30° clockwise, 30° counterclockwise).
- Comparison: Lesions stratified by R-EBUS signal presence (present vs. absent), number of confirmed TIVL planes, and lesion proximity to the entry point.
- Outcomes: Diagnostic yield, determined using strict diagnostic criteria, and identification of predictors for diagnostic success via multivariable logistic regression.
📈 Methodology & Rigor
The authors performed a retrospective analysis of prospectively maintained or retrospectively reviewed bronchoscopy databases. TIVL was rigorously evaluated across five distinct fluoroscopic angles. Statistical analysis utilized multivariable logistic regression with backward selection based on the Akaike information criterion (AIC) to identify independent predictors of diagnostic success, accounting for both patient-level variables (e.g., BMI) and procedural factors (e.g., tool selection, biopsy method). Given the retrospective observational design, the study is susceptible to selection bias, though the evaluation of multiplane alignment introduces a novel, quantifiable metric for intra-procedural confirmation.
🔬 Key Findings [or Planned Endpoints]
- R-EBUS Yield Stratification: Out of 226 lesions, 125 (55.3%) exhibited eccentric or concentric R-EBUS signals, while 101 (44.7%) showed no R-EBUS signal.
- Impact of Multiplane TIVL in R-EBUS Negative Lesions: For lesions without an R-EBUS signal, diagnostic yield rose stepwise from 0% with $\le 1$ TIVL plane to 42% with $\ge 4$ planes.
- Odds Ratios for TIVL Confirmation: TIVL confirmation in 3 planes was associated with a >5-fold higher likelihood of diagnostic success (OR = 5.57; $p < 0.05$), and $\ge 4$ planes yielded a nearly 14-fold higher likelihood (OR = 13.6; $p = 0.01$).
- Proximity Effect: Lesions located within 16 mm of the entry point showed stepwise yield improvements up to 89% with $\ge 4$ confirmed planes, whereas more distant lesions showed no significant multiplane benefit.
- Predictors of Success:
- Positive predictors: Cryobiopsy (OR = 3.93; $p = 0.002$).
- Negative predictors: Higher BMI (OR = 0.66; $p = 0.046$), absence of a bronchus sign, and the use of an ultrathin bronchoscope (OR = 0.28; $p < 0.001$).
⚖️ Critical Appraisal
While this study introduces a valuable, practical paradigm (multiplane TIVL confirmation) for centers relying on augmented fluoroscopy instead of costly CBCT systems, several limitations must be noted. As a retrospective cohort, operator-dependent bias in confirming spatial overlay and subjective interpretation of multiplane alignment could influence results. Furthermore, CT-to-body divergence remains a fundamental vulnerability; even flawless multiplane TIVL can result in non-diagnostic outcomes if respiratory motion or atelectasis shifts the true target tissue away from the virtual overlay. Generalizability may also be limited by institutional expertise in AF integration.
💡 The Clinical Bottom Line
For interventional pulmonologists performing navigational bronchoscopy in facilities without CBCT, Augmented Fluoroscopy is not a binary "hit-or-miss" tool. Confirming tool-in-virtual-lesion alignment across 3 to 4 distinct fluoroscopic planes dramatically rescues diagnostic yield—particularly in the frustrating subset of peripheral lesions lacking an R-EBUS signal or located close to the pleural entry point. Clinicians should incorporate multiplane fluoroscopic checks into their procedural workflow, keeping in mind that factors like high patient BMI and lack of a bronchus sign continue to degrade success rates, while cryobiopsy enhances tissue acquisition.