Shape-sensing robotic-assisted versus electromagnetic navigation bronchoscopy for peripheral pulmonary lesions: a single-operator experience

Introduction

Lung cancer is the leading cause of cancer related death worldwide and accounts for nearly one fifth of cancer mortality (1). In Tarrant County, the third most populous county within Texas, USA, lung cancer deaths exceed those from breast, liver, prostate, and cervical cancers combined (2). Earlier diagnosis is essential to improving patient outcomes, as more than half of lung cancers are diagnosed at a distant metastatic stage, while fewer than one quarter are identified at a localized stage (3). Five-year relative survival of patients diagnosed with localized lung cancers is 65% while those presenting with distant metastases is only 10% (3). The increased use of computed tomography (CT), and the expansion of lung cancer screening programs have led to a rise in the detection of lung nodules (4,5).

Over the past two decades, technological advancements in peripheral bronchoscopy have expanded the role of bronchoscopic biopsy for peripheral pulmonary lesions (PPLs). Electromagnetic navigation bronchoscopy (ENB) improves access to PPLs by combining CT based virtual airway mapping with real-time electromagnetic tracking of a locatable guide advanced through the bronchial tree (6). CT-guided transthoracic biopsy showed a diagnostic accuracy of 92.1% among pooled data of over 10,000 lesions but a pneumothorax rate of 20.5% (7). In the VERITAS study, ENB achieved non-inferior diagnostic accuracy compared to CT guided biopsy (79.0% vs. 73.6%) while reducing pneumothorax rates from 28.3% to 3.3% (8). VERITAS used image guidance (digital tomosynthesis) in the ENB arm. A large meta-analysis including over 10,000 lung lesions reported a pooled diagnostic yield of 70.9% for navigational bronchoscopy, with an overall pneumothorax rate of 2.5% (6). An additional advantage of navigational bronchoscopy is the ability to perform mediastinal staging with linear EBUS bronchoscopy during the same procedure (9).

Shape sensing robotic assisted bronchoscopy (ssRAB) is a newer technology designed to further improve access to PPLs. This platform uses a flexible catheter with real-time shape-sensing technology to guide navigation and provide increased stability with controlled tip articulation. Improved catheter stability may help maintain position adjacent to the target PPL during sampling and reduce lesion’s mobility due to respiratory motion, a known limitation of navigational bronchoscopy techniques. CT-to-body divergence (CTBD) is the difference between the PPL location in the pre-procedure CT scan and the actual location during the procedure (10). Initial observational studies have reported high diagnostic yield with low complication rates using ssRAB (11-14). A recent meta-analysis including 25 studies demonstrated a pooled diagnostic yield of 84.3% across 1779 lesions, with pneumothorax and significant hemorrhage rates of 2.3% and 0.5%, respectively (15).

Despite the rapid adoption of these technologies, comparative data between ENB and ssRAB remain limited, especially in settings without access to advanced three-dimensional imaging. Most published comparisons originate from large academic centers with multiple operators, introducing variability in procedural technique and experience. A single center prospective randomized study from Vanderbilt University reported comparable outcomes between ENB and ssRAB across procedures performed by multiple interventional pulmonologists and fellows (16). However, data from safety-net hospitals and single operator experiences are limited.

Safety-net hospitals care for socioeconomically diverse populations and may often face resource limitations that can impact access to advanced technologies and procedural support (17-20). These systems may have fewer subspecialty physicians, and reduced availability of adjunct imaging, requiring efficient workflows and consistent procedural performance across a wide range of clinical presentations. There is limited data regarding how ENB or ssRAB perform in these environments, and existing studies rarely include patients who reflect the racial, ethnic, and economic diversity seen in safety-net populations. To our knowledge, no study has compared ENB and ssRAB using two-dimensional fluoroscopy alone in a safety-net hospital staffed by a single interventional pulmonologist. The objective of this study is to describe the diagnostic yield, final diagnostic accuracy, number of procedures required to reach a final diagnosis, and complication rates for ENB and ssRAB in this setting. This analysis provides real-world data on the performance of both technologies in a diverse and resource limited population. We present this article in accordance with the STROBE reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1-2623/rc).

Methods

Ethical approval

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by North Texas Institutional Review Board (No. 2199873-1). The IRB approved the waiver of informed consent because of the retrospective nature of the study and the research could not have been practicably conducted without the waiver.

Study design and setting

This was a single center retrospective study conducted at JPS Health Network (JPS), a safety-net healthcare system with a 582-bed hospital, serving Tarrant County, Texas, USA. ENB using the SPiN Thoracic Navigation System (Veran Medical Technologies, St. Louis, Missouri, USA) was performed between December 2020 until June 2023. SsRAB using the Ion endoluminal platform (Intuitive Surgical, Inc., Sunnyvale, California, USA) was performed between September 2023 and August 2024. All procedures were performed by a single interventional pulmonologist.

Interventional pulmonologist started using ENB in 2013 as a pulmonary fellow, then finished interventional pulmonology fellowship in 2016. He started as a staff physician at JPS in 2018 and performed about 100 navigation bronchoscopy cases per year, on average, prior to the timeline of this study.

Radial endobronchial ultrasound (r-EBUS) and two-dimensional fluoroscopy were available for all ENB and ssRAB procedures. No procedures using either technology used cone-beam CT, digital tomosynthesis, or rapid on-site cytologic evaluation (ROSE). Endotracheal intubation with general anesthesia was used for all patients. Standard biopsy forceps (Boston Scientific, Marlborough, MA, USA) was used, and 8–10 biopsies were done for both ENB and ssRAB groups. Our standard protocol to prevent atelectasis is tidal volume of 6–8 mL/kg ideal body weight and a positive end expiratory pressure of 8–10.

The sample size was determined by the availability of eligible participants within the study timeframe.

Patient identification and data collection

Patients who underwent ENB between December 2020 to June 2023 were identified through Current Procedure Terminologies (CPT) based queries by the JPS Information Technology Department. Patients who underwent ssRAB between September 2023 and August 2024 were identified prospectively within the interventional pulmonology program. For both cohorts, electronic health records (EHRs) were reviewed to obtain demographic data, imaging findings, lesion characteristics, pathology results, and clinical course. Chart abstraction was performed by interventional pulmonology research coordinator and nurse practitioner.

Outcomes and definitions

The primary outcome was strict diagnostic yield from the index navigational bronchoscopy. A procedure was considered diagnostic when pathology or cytology results identified malignancy or benign diagnosis that explained the PPL. Results reported as normal lung or airway, nonspecific inflammation, or atypical cells not diagnostic of malignancy were classified as non-diagnostic. For nondiagnostic index procedures, follow-up pathology or imaging was reviewed to determine final diagnostic status (21). Secondary outcomes included the total number of procedures required to establish a final diagnosis, lesion size, lung cancer staging when malignancy was diagnosed, and procedure-related complications. For diagnostic yield and diagnostic accuracy assessments, identifying a malignancy was considered ‘malignant’. For the secondary outcome of ‘final diagnosis’, cumulative of all procedures that were performed within 12 months to arrive at this conclusion was taken. These procedures could be additional navigational bronchoscopy (ENB or ssRAB), interventional radiology (IR) biopsy or surgical biopsy or resection. For example, if the initial ENB was non-diagnostic, but surgical resection showed a benign diagnosis, then this was noted as ‘benign’ in diagnostic accuracy and final diagnosis was ‘benign’ and counted as two procedures. If ENB showed malignancy on index procedure, for diagnostic yield and diagnostic accuracy, it was included as ‘malignant’. For treatment decision, if additional IR procedure was needed to get adequate tissue to assess adenocarcinoma or squamous and then further additional material was obtained by another procedure for next generation sequencing or mutational testing, then final diagnosis needed three procedures.

Statistical analysis

Descriptive statistics were used to summarize patient demographics, including age, race and sex. No demographic variables were missing. Continuous variables were reported as medians with interquartile ranges (IQRs), and categorical variables were reported as counts and percentages. To assess the association between the diagnostic yield and the navigational technology used for biopsy, a Chi-squared test was performed with an alpha =0.05 confidence level.

Results

A total of 359 patients underwent navigational bronchoscopy PPL biopsy during the study period, including 226 ENB procedures and 133 ssRAB procedures. All procedures were performed by a single interventional pulmonologist. Demographic characteristics were similar between groups (Table 1).

Lesion characteristics and procedural timing

Lesions sampled with ssRAB were significantly smaller than those sampled with ENB (P<0.001). The median nodule size was 2.4 cm (IQR, 1.6–3.6 cm) in the ENB group and 1.7 cm (IQR, 1.4–2.6 cm) in the ssRAB group. The median time from initial chest CT to bronchoscopy was 21 days for both cohorts (P=0.84). Lesion size was missing for 29 ssRAB procedures and 30 ENB procedures. Larger lesions (>2 cm) are significantly associated with a higher diagnostic yield (P=0.008), but ssRAB had a much lower percentage of lesions larger than 2 cm than ENB. The rate of malignancy was very similar between medium (1–2 cm) and large lesions. Small (<1 cm) lesions had a higher rate of being non-diagnostic compared to large lesions (P<0.001). These are shown in Table 2.

Diagnostic yield and diagnostic accuracy

Strict diagnostic yield based on the index procedure was 64% for both ENB and ssRAB. Final diagnostic accuracy, defined as confirmation of malignant or specific benign pathology after all necessary procedures, was 72.1% for ENB and 75% for ssRAB. This difference was not statistically significant (P=0.39).

Number of procedures required to establish diagnosis

A greater proportion of patients undergoing ssRAB achieved a final diagnosis after a single procedure compared with those undergoing ENB (85.1% vs. 67.1%). Malignancy was ultimately diagnosed in 120 patients in the ENB group (53%) and 65 patients in the ssRAB group (49%).

Among patients with a malignant diagnosis, ssRAB required fewer total procedures to establish the diagnosis. With ENB, 60.8% of malignant cases were diagnosed after one procedure, 23.3% after two procedures, 13.3% after three procedures, and 2.5% after four procedures. In contrast, 84.8% of malignant diagnoses in the ssRAB group were achieved after one procedure, 12.3% after two procedures, and 3.1% after three procedures. The difference in number of procedures required to establish a malignant diagnosis was statistically significant (P=0.007). The total number of procedures included the index bronchoscopy and any subsequent bronchoscopy, CT-guided bronchoscopy, or surgical procedures required to reach a final diagnosis (Figures 1,2).

Figure 1 Comparison of number of total procedures needed for a definite benign or an adequate final diagnosis of malignancy between ENB and ssRAB. ENB, electromagnetic navigation bronchoscopy; ssRAB, shape sensing robotic assisted bronchoscopy.

Figure 2 Comparison of number of total procedures needed for an adequate final diagnosis of malignancy between ENB and ssRAB. ENB, electromagnetic navigation bronchoscopy; ssRAB, shape sensing robotic assisted bronchoscopy.

Malignancy type and staging

The distribution of malignancy types did not differ significantly between the two groups (Table 3). Among patients diagnosed with lung adenocarcinoma or lung squamous cell carcinoma, early-stage disease (stage 0–1) was identified in 23 ENB patients (29%) and 16 ssRAB patients (41%). Stage IV disease was diagnosed in 25 ENB patients (31.6%) and 10 ssRAB patients (25.6%).

Complications

Procedure related complications were uncommon. Pneumothorax occurred in two ENB procedures and one ssRAB procedure, resulting in an overall complicate rate of less than 1%.

Discussion

In a single-operator study conducted within a safety-net hospital, we compared the performance of ENB and ssRAB using two-dimensional fluoroscopy. The primary finding was that ssRAB achieved a similar diagnostic yield and diagnostic accuracy compared with ENB while sampling significantly smaller PPLs and requiring fewer procedures to establish a malignant diagnosis. These findings highlight differences in procedural diagnostic efficiency that may be particularly relevant in resource limited settings.

Our strict diagnostic yield based on the index bronchoscopy procedure was lower than that reported in many ENB and RAB series (11,14,16,22-24). Our diagnostic yield may be lower because we opted to follow American Thoracic Society/American College of Chest Physicians endorsed strict definition that came out in the year 2024 (21). Some of these previously reported studies used different definitions that may have led to higher diagnostic yield. In addition, regional factors may have influenced results. Texas has a higher prevalence of granulomatous disease, which can result in benign or non-specific histopathologic findings and reduce index diagnostic yield (25). The lack of formal preprocedural malignancy risk stratification may also have contributed to diagnostic outcomes (26).

There was no difference in diagnostic yield between ENB vs. ssRAB, both are at 64%. A notable difference between the two technologies was first procedure diagnostic success (84.6% in ssRAB vs. 60.8% in ENB group, P=0.007). Patients undergoing ssRAB were significantly more likely to achieve a final malignant diagnosis required to make necessary treatment decisions after a single procedure compared with those undergoing ENB. This reduction in repeat procedures, be it bronchoscopy or IR biopsy, is clinically meaningful. Fewer procedures decrease patient burden, reduce procedural risk, and conserve limited resources in publicly funded health systems. At JPS, IR has generally moved away from doing biopsy of PPLs due to 20% risk of pneumothorax and high risk of bleeding. Referring physicians consult interventional pulmonology.

Although overall diagnostic yield and accuracy was similar between group, ssRAB was used to biopsy significantly smaller PPLs. Larger lesions are associated with a higher diagnostic yield but ssRAB had a much lower percentage of large lesions. Considering ENB and ssRAB showed similar diagnostic yield, this size related finding further reinforces that ssRAB was a better technique. In addition, a higher proportion of early-stage lung cancers was identified in the ssRAB cohort. While this observation was not statistically significant, it is clinically relevant given the importance of diagnosing lung cancer at earlier stages. These findings should be interpreted cautiously and warrant further evaluation in larger studies.

Importantly, the integration of ssRAB as compared to ENB into a safety-net hospital did not delay care. The median time from diagnostic chest CT to biopsy was similar between groups. The team that oversees workflow from CT to biopsy has stayed the same through the course of this study period. This suggests that adoption of ssRAB can be accomplished without disrupting workflow and may facilitate more efficient diagnostic pathways by improving first-procedure diagnostic success. In resource-constrained systems, reducing the need for repeat procedures may also increase access for other patients and improve overall procedural capacity. Overall workflow and the number of days per week procedures are performed have not changed but more ssRAB procedures are done monthly compared to EMB because of the ability to target smaller PPLs, some of these lesions are too small for PET scan. So, PPLs that in the past would have been monitored are taken for ssRAB right away.

This study demonstrates the feasibility of performing navigational bronchoscopy in a safety-net setting; however, our findings are broadly applicable for all hospital settings and geographic regions. Safety-net hospitals serve racially, ethnically, and socioeconomically diverse populations that are often underrepresented in procedural outcomes research. The performance of both ENB and ssRAB in this environment provides real-world insight into how these technologies function outside of highly resourced academic centers.

There are several limitations. Having a single interventional pulmonologist could bring challenges of coverage during vacation or unexpected prolonged leave. Though adding another proceduralist could be helpful, there are issues related to space in the endoscopy suite that is shared with gastroenterologists. The ENB cohort was retrospective, while the ssRAB cohort was collected prospectively, which may introduce a slight positive bias in favor of the ssRAB cohort. Neither technology was used with advanced three-dimensional imaging, likely affecting overall diagnostic performance (15). ROSE was not available for either cohort, which may have contributed to nondiagnostic sampling and the need for additional procedures (8). Lesion size was missing in both cohorts, more so in ssRAB due to either not being reported on chest CT or not being able to identify reports in EHR if imaging was done at an outside facility, this study reflects the experience of a single interventional pulmonologist, limiting generalizability. Operator learning over time may also have contributed to bias in favor of ssRAB outcomes. Finally, cost analysis and long-term clinical outcomes were not evaluated.

Conclusions

In this single center, safety net hospital experience, ssRAB achieved diagnostic yield and final diagnostic accuracy comparable to ENB while sampling smaller PPLs and requiring fewer procedures to establish a malignant diagnosis. These findings suggest that ssRAB can improve procedural efficiency without delaying care and can be integrated into resource limited settings that serve diverse patient populations. Larger multicenter studies incorporating cost analysis and long-term outcomes are needed to confirm these results.

Acknowledgments

We thank Dr. Jerry D. Henderson of JPS Health Network for his role in providing us the EMB dataset.

The abstract and some of the figures and tables were presented at the 8^th AABIP Annual Conference held at Austin, TX from August 14^th to 16^th, 2025 and for online publication at the 2025 ASCO Annual Meeting held at Chicago, IL from May 30^th to June 3^rd, 2025.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1-2623/rc

Data Sharing Statement: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1-2623/dss

Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1-2623/prf

Funding: This work was supported by JPS Health Network’s Internal Medicine Discretionary Fund.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1-2623/coif). K.S. reports honoraria from Stanford University for faculty participation in an educational bronchoscopy course, and travel support from Intitutive Surgical, Inc. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by North Texas Institutional Review Board (No. 2199873-1). The IRB approved the waiver of informed consent because of the retrospective nature of the study and the research could not have been practicably conducted without the waiver.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Bray F, Laversanne M, Sung H, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2024;74:229-63. [Crossref] [PubMed]
2. Tarrant County Public Health. Data Brief. Available online: https://www.tarrantcountytx.gov/content/dam/main/public-health/PH%20DOCUMENTS/Epi/Data%20Briefs/2025/2025-11_Data_Brief_Lung_Cancer.pdf
3. National Cancer Institute. Cancer Stat Facts: Lung and Bronchus Cancer. Available online: https://seer.cancer.gov/statfacts/html/lungb.html
4. Gould MK, Tang T, Liu IL, et al. Recent Trends in the Identification of Incidental Pulmonary Nodules. Am J Respir Crit Care Med 2015;192:1208-14. [Crossref] [PubMed]
5. Wolf AMD, Oeffinger KC, Shih TY, et al. Screening for lung cancer: 2023 guideline update from the American Cancer Society. CA Cancer J Clin 2024;74:50-81. [Crossref] [PubMed]
6. Kops SEP, Heus P, Korevaar DA, et al. Diagnostic yield and safety of navigation bronchoscopy: A systematic review and meta-analysis. Lung Cancer 2023;180:107196. [Crossref] [PubMed]
7. DiBardino DM, Yarmus LB, Semaan RW. Transthoracic needle biopsy of the lung. J Thorac Dis 2015;7:S304-16. [PubMed]
8. Lentz RJ, Frederick-Dyer K, Planz VB, et al. Navigational Bronchoscopy or Transthoracic Needle Biopsy for Lung Nodules. N Engl J Med 2025;392:2100-12. [Crossref] [PubMed]
9. Miller RJ, Chrissian AA, Kheir F, et al. American Association for Bronchology and Interventional Pulmonology (AABIP) Evidence-Based Guidelines on Bronchoscopic Diagnosis and Staging of Lung Cancer. J Bronchology Interv Pulmonol 2025;32:e1034. [PubMed]
10. Pritchett MA, Bhadra K, Calcutt M, et al. Virtual or reality: divergence between preprocedural computed tomography scans and lung anatomy during guided bronchoscopy. J Thorac Dis 2020;12:1595-611. [Crossref] [PubMed]
11. Kalchiem-Dekel O, Connolly JG, Lin IH, et al. Shape-Sensing Robotic-Assisted Bronchoscopy in the Diagnosis of Pulmonary Parenchymal Lesions. Chest 2022;161:572-82. [Crossref] [PubMed]
12. Reisenauer J, Simoff MJ, Pritchett MA, et al. Ion: Technology and Techniques for Shape-sensing Robotic-assisted Bronchoscopy. Ann Thorac Surg 2022;113:308-15. [Crossref] [PubMed]
13. Simoff MJ, Pritchett MA, Reisenauer JS, et al. Shape-sensing robotic-assisted bronchoscopy for pulmonary nodules: initial multicenter experience using the Ion™ Endoluminal System. BMC Pulm Med 2021;21:322. [Crossref] [PubMed]
14. Styrvoky K, Schwalk A, Pham D, et al. Shape-Sensing Robotic-Assisted Bronchoscopy with Concurrent use of Radial Endobronchial Ultrasound and Cone Beam Computed Tomography in the Evaluation of Pulmonary Lesions. Lung 2022;200:755-61. [Crossref] [PubMed]
15. Ali MS, Ghori UK, Wayne MT, et al. Diagnostic Performance and Safety Profile of Robotic-assisted Bronchoscopy: A Systematic Review and Meta-Analysis. Ann Am Thorac Soc 2023;20:1801-12. [Crossref] [PubMed]
16. Low SW, Lentz RJ, Chen H, et al. Shape-Sensing Robotic-Assisted Bronchoscopy vs Digital Tomosynthesis-Corrected Electromagnetic Navigation Bronchoscopy: A Comparative Cohort Study of Diagnostic Performance. Chest 2023;163:977-84. [Crossref] [PubMed]
17. Institute of Medicine (US) Committee on the Changing Market, Managed Care, and the Future Viability of Safety Net Providers, Ein Lewin M, Altman S. editors. Americas’s Health Care Safety Net: Intact but Endangered. Washington, DC, USA: National Academies Press; 2000.
18. Paradise J, Garfield R. What is Medicaid’s Impact on Access to Care, Health Outcomes, and Quality of Care? Setting the Record Straight on the Evidence. San Francisco, CA, USA: The Kaiser Commision on Medicaid and the Uninsured; 2013.
19. Sutton JP, Washington RE, Fingar KR, et al. Characteristics of Safety-Net Hospitals, 2014. In: Healthcare Cost and Utilization Project (HCUP) Statistical Briefs. Rockville, MD, USA: Agency for Healthcare Research and Quality; 2016.
20. Ibrahim S. Identifying Safety Net Hospitals Consistently. JAMA Health Forum 2021;2:e211240. [Crossref] [PubMed]
21. Gonzalez AV, Silvestri GA, Korevaar DA, et al. Assessment of Advanced Diagnostic Bronchoscopy Outcomes for Peripheral Lung Lesions: A Delphi Consensus Definition of Diagnostic Yield and Recommendations for Patient-centered Study Designs. An Official American Thoracic Society/American College of Chest Physicians Research Statement. Am J Respir Crit Care Med 2024;209:634-46. [Crossref] [PubMed]
22. Bellinger C, Poon R, Dotson T, et al. Lesion characteristics affecting yield of electromagnetic navigational bronchoscopy. Respir Med 2021;180:106357. [Crossref] [PubMed]
23. Katsis J, Roller L, Aboudara M, et al. Diagnostic Yield of Digital Tomosynthesis-assisted Navigational Bronchoscopy for Indeterminate Lung Nodules. J Bronchology Interv Pulmonol 2021;28:255-61. [Crossref] [PubMed]
24. Avasarala SK, Roller L, Katsis J, et al. Sight Unseen: Diagnostic Yield and Safety Outcomes of a Novel Multimodality Navigation Bronchoscopy Platform with Real-Time Target Acquisition. Respiration 2022;101:166-73. [Crossref] [PubMed]
25. Nazarullah A, Nilson R, Maselli DJ, et al. Incidence and aetiologies of pulmonary granulomatous inflammation: a decade of experience. Respirology 2015;20:115-21. [Crossref] [PubMed]
26. Silvestri GA, Bevill BT, Huang J, et al. An Evaluation of Diagnostic Yield From Bronchoscopy: The Impact of Clinical/Radiographic Factors, Procedure Type, and Degree of Suspicion for Cancer. Chest 2020;157:1656-64. [Crossref] [PubMed]