Shape-Sensing Robotic Bronchoscopy Matched ENB on Yield but Reduced Repeat Procedures
A single-operator comparison of shape sensing robotic assisted bronchoscopy and electromagnetic navigation bronchoscopy found nearly identical diagnostic yield from the first procedure, yet the robotic platform reached a malignant diagnosis with fewer repeat bronchoscopies. In a large urban hospital setting, both systems were used without cone beam CT, digital tomosynthesis, or rapid on-site evaluation, creating a pragmatic view of how these technologies perform when advanced adjuncts are not available.
The cohort included 359 patients, with 226 evaluated by ENB and 133 by ssRAB. The strict diagnostic yield from the index procedure was 64% for both approaches. Final diagnostic accuracy was 71.2% with ENB and 76% with ssRAB, a difference that did not reach statistical significance. Even without a measurable separation in overall accuracy, the procedural path to diagnosis differed in a clinically meaningful way.
Fewer malignant lesions needed a second look when ssRAB was used.
Diagnostic Performance in a Real-World Bronchoscopy Lab
The comparison matters because peripheral pulmonary lesions remain difficult targets for bronchoscopic sampling. Their small size, distal airway location, and variable bronchial access limit the chance of landing the biopsy tools directly in the lesion. Navigation systems are designed to solve that problem, but the value of each platform is shaped not only by the technology itself, but also by what is available in the room at the time of the procedure. Here, both bronchoscopy methods were paired with radial endobronchial ultrasound and two-dimensional fluoroscopy, while more advanced localization tools were absent.
That setting may help explain why the overall yield was modest and nearly indistinguishable between the two groups. When cone beam CT or rapid on-site evaluation is unavailable, bronchoscopic success depends heavily on airway anatomy, lesion size, tool stability, and the operator’s ability to interpret fluoroscopic and ultrasound cues in real time. A single experienced bronchoscopist removes one source of variability, but it also means the comparison reflects one operator’s technical style rather than a broader multi-physician practice pattern.
The lesions sampled with ssRAB were smaller than those sampled with ENB, with median sizes of 1.7 cm and 2.4 cm, respectively. Smaller nodules are inherently harder to sample, so maintaining the same strict diagnostic yield in a smaller-lesion group is clinically relevant even if the study did not show a statistical difference in final accuracy. In practice, lesion size often tracks with sampling difficulty, especially when tool-in-lesion confirmation is limited to two-dimensional imaging and radial ultrasound.
Fewer Procedures to Reach a Cancer Diagnosis With ssRAB
Where the robotic platform separated itself was in the number of procedures needed to establish malignancy. Among patients ultimately diagnosed with cancer, 84.6% in the ssRAB group were diagnosed after a single procedure, compared with 60.8% in the ENB group. The difference was statistically significant, with P=0.007.
This is the kind of result that changes bedside logistics even when headline yield numbers look similar. A single completed diagnostic bronchoscopy can accelerate staging, avoid additional anesthesia exposure, and reduce the interval between first suspicion and definitive cancer planning. In a system already strained by backlogs and repeat procedural demand, shaving off a second bronchoscopy has real consequences for workflow and patient burden.
The biological explanation is straightforward enough: peripheral lesions are missed when the sampling tool edges past the tumor margin or stops short of the most viable tissue. Robotic control may improve catheter stability and allow more precise positioning inside tortuous distal airways, which can translate into a better chance of securing diagnostic tissue on the first pass. Even without cone beam CT confirmation, a platform that stabilizes access to small targets may reduce the need for a return procedure when malignancy is the concern.
What the Numbers Do and Do Not Prove
The lack of a significant difference in final diagnostic accuracy is important. With final accuracy at 71.2% for ENB and 76% for ssRAB, the study suggests comparable real-world performance rather than a clear superiority claim. The sample size, retrospective design, and single-center structure all limit how far the findings can be generalized. Differences in case mix also matter, especially because the ssRAB cohort contained smaller lesions than the ENB cohort.
The absence of cone beam CT, digital tomosynthesis, and rapid on-site evaluation is another defining feature of the study. Those tools often amplify diagnostic yield, so this comparison is best read as a test of what these platforms can achieve in a more restrained procedural environment. Many hospitals operate exactly this way, without advanced imaging support in every bronchoscopy suite. For them, the study offers a practical benchmark rather than an idealized performance ceiling.
A single operator can also cut both ways. Technical consistency strengthens internal comparison because variation in biopsy style is minimized, but it narrows the lens to one clinician’s skillset and one institutional workflow. The results may not map cleanly onto centers with different bronchoscopy volumes, lesion referral patterns, or anesthesia support.
Peripheral Lesion Sampling Still Hinges on Imaging Constraints
Peripheral pulmonary lesions continue to challenge bronchoscopists because they sit at the edge of what flexible instruments can reliably access. When the distal airway path is uncertain, the operator must rely on navigation software, ultrasound signal, fluoroscopic landmarks, and procedural judgment. Shape sensing robotic systems are built to improve that distal reach by offering more controlled movement and better catheter stability, while ENB has long served as a familiar navigation backbone for many pulmonologists.
The study suggests that, in the absence of advanced three-dimensional imaging, ssRAB may offer a procedural advantage in efficiency for malignant lesions even if overall diagnostic accuracy remains similar. For a patient, that often means fewer repeat procedures and less uncertainty after the first bronchoscopy. For a practice, it may mean better use of procedure time and fewer delayed cancer diagnoses.
The unanswered clinical question is not whether one platform can biopsy a peripheral lesion at all; both can. The harder question is which system delivers the best balance of access, tissue acquisition, and procedural economy across lesion sizes and institutional settings that do not have every imaging adjunct at hand. This study does not settle that question, but it does show that robotic bronchoscopy can reduce the need for repeat sampling while operating in a relatively austere diagnostic environment.