Robotic Assisted Bronchoscopy: The Ultimate Solution for Peripheral Pulmonary Nodules?

Since Fielding's [1] publication of the first human experience of a robotic assisted bronchoscopy platform in this journal in 2019, the scientific and commercial interest for robotic assisted bronchoscopy systems has grown rapidly. Two robotic systems are currently FDA approved and commercially available in the USA: the Auris Monarch® and the Ion® system of Intuitive (CE-approval is still awaited). In this issue, two new studies are presented that study the ability of diagnosing pulmonary lesions with the help of a robotic assisted platform and concurrent imaging technology.

Doctors Thiboutot and Argento together with their coworkers investigated which variables of the shape sensing based robot assisted navigation bronchoscopy affect diagnostic outcome in a ventilated cadaver model. In contrast to non-robotized platforms such as electromagnetic navigation technology (EMN), the distal tip of both robotic platforms provide a rigid exit point for needle sampling that can be actively angulated. Interestingly, they show that reducing the distance between the distal catheter and lesion to a minimum was the most important factor affecting diagnostic outcome when having an actively steerable tip. If the distance was kept minimal and the virtual image was manually aligned with the real endoscopic image, the addition of c-arm fluoroscopy and radial EBUS then seemed to result only in a limited increase in accuracy [2]. Also in this issue, in a case study using the Monarch robotic platform, doctor Manley and coworkers showed the feasibility and promising result of using the Monarch robotic platform and needle-based confocal laser endomicroscopy for obtaining real-time sampling guidance and the potential of an image-based diagnosis [3].

With an ever-increasing amount of incidentally and screening detected nodules, the interventional pulmonologist is challenged to accurately reach and diagnose even the smallest of lesions. Furthermore, it is likely that image guided transbronchial treatments for malignant nodules will also be added to our inventory in short-term as well [4]. The diagnostic accuracy of available combinations of technologies such as c-arm fluoroscopy, thin bronchoscopes, radial EBUS miniprobes, EMN, and virtual navigation bronchoscopy therein all seem to approach the 70% diagnostic yield range [5, 6, 7]. A recent meta-analysis by Folch et al. [8], for example, showed that EMN as often used in combination with other tools (c-arm fluoroscopy, rEBUS) had an overall sensitivity of 77%, but may be less accurate when the nodule size is smaller than 2 cm [5, 6, 9, 10].

The two robotic assisted bronchoscopy platforms that are now commercially available in the USA use virtual navigation bronchoscopy in combination with shape sensing (Ion®, Intuitive) or EMN (Monarch®, Auris/J&J) for navigation. Both systems uniquely allow complete robotized control of the distal tip for navigation and sampling in an intubated patient. An actively steerable distal tip that can maintain its shape and that simultaneously functions as a rigid exit point for sampling is a property of both platforms that was not available in any other commercially available technology to date. The PRECISION-1 study by Yarmus et al. [11] showed in a cadaveric model that this distal tip control could significantly more often reach artificial pulmonary lesions when compared to previously available technology such as EMN or ultrathin bronchoscopy combined with rEBUS. The first in vivo studies show the diagnostic yields of the robotic platforms already have good accuracy upon initial use, ranging from 69% to 82% [1, 12, 13, 14].

Yet, as with all other technologies evaluated to date, the reported diagnostic yields of robotic bronchoscopy are lower than the reported navigation success. Chaddha et al. [12] observed in a multicenter study of the robotic Auris platform that navigation was successful in 88.6% of cases, but that lesion positioning relative to the rEBUS imaging probe had a profound effect on diagnostic accuracy. An absent-, eccentric-, and concentric-rEBUS image of the target nodules translated into diagnostic yields of 26.9%, 71.7% and 81.5% [12]. Another study by Kalchiem-Dekel et al. [14] in a cohort of 159 lesions with an airway-lesion relationship judged as favorable for sampling showed an impressive navigation success of 98.7%, but eventually the diagnostic accuracy was found to be lower (but still high overall), at 81.7%.

The knowledge provided by the studies in this issue of Respiration illustrates how making effective use of the actively steerable distal tip properties and additional imaging technology could help reduce the mismatch between navigation success and diagnostic accuracy [2, 3]. Both reports seem to underline the importance of exact millimeter positioning accuracy for facilitating a correct diagnosis. While Thiboutot and Argento et al. report the addition of imaging was of minimal value in their cadaveric study, currently available in vivo outcomes suggest it does remain of high importance in removing the mismatch between navigation success and diagnostic accuracy. In that respect the observation of using needle based confocal endomicroscopy, as illustrated by the case report in this issue [3], has the potential to improve minute positioning and diagnostic yield provided that the exact same position can be maintained by the robotic tip when the CLE catheter is withdrawn from the TBNA needle. Both observations may prove essential to facilitate immediate minimal invasive treatment.

Almost at the same time the first robotic navigation bronchoscopy study results became available, we and several others also explored the feasibility of using novel imaging modalities that provide detailed position information as an adjunct tool for navigation and sampling guidance. The feasibility of using a CBCT system as readily available in the hybrid OR was assessed in combination with several other navigation modalities [15, 16, 17, 18, 19, 20] or as a standalone system for navigation and diagnosis of small pulmonary nodules [21, 22, 23, 24, 25]. This hybrid OR system allows for lesion and pathway augmentation on fluoroscopy and repeated cone beam CT-scans for 3D confirmation of nodule access [21, 22, 23, 24, 25]. The addition of this imaging technology therein not only showed to significantly improve the outcome of the EMN-guided procedure. After having become experienced with the image system, a 98% navigation success and a diagnostic accuracy of 90% in a consecutive series of 64 unselected patients with nodules of a median size of 16 mm using only pre-angulated catheters could be obtained [23, 24]. In experienced hands, the ability of real-time detailed 2D and 3D imaging with a CBCT system thus seems to compete with both robotic systems in terms of accuracy [1, 12, 13, 14, 15, 16, 23, 24]. However, to reach this level of accuracy and diagnostic yield with the currently reported CBCT-guided navigation bronchoscopy, it is acknowledged an important learning curve has to be overcome [23]. Furthermore, CBCT navigation requires access to a hybrid OR and is dependent on radiation imaging (even though staff dose is minimal if effective protection measures are taken [23]).

The big potential advantage of robotic platforms is likely to be their ease of use and intuitiveness along with a high initial accuracy. These will make procedures feasible and safe, with predictable outcomes even in less experienced centers. The most important hurdle in becoming adapted at a CBCT-based navigation bronchoscopy that is dependent on pre-curved catheters as means of navigation is the ability to translate the radiology perspective and tactile catheter feedback to an endoscopic position and angulation, without any endoscopic image feedback. Furthermore, the current versions of pre-angulated catheters do not retain angulation upon instrument insertion, making precise biopsy difficult. In contrast, the robotic systems bring “normal” bronchoscopy to the subsegmental level. Their steerable catheters also make repeated tissue sampling an action that can be done with exact and continuous control, which will also prove useful for future transbronchial treatment such as microwave ablation [4]. From an accuracy perspective, future transbronchial treatments will likely be performed using a combination of robotic control and 3D imaging capabilities such as by using a CBCT imaging system.

In awaiting the approval and implementation of the robotic systems for worldwide use to meet an ever-growing demand of diagnostics for pulmonary lesions, not only the physicians and patients need to be convinced on safety and procedural efficacy. Healthcare ministries, healthcare organizations, and healthcare insurance companies will also need convincing in order to allow their use and obtain reimbursement for robotic bronchoscopy procedures. Several platforms seem to be suitable to target the peripheral pulmonary lesion with relatively high accuracy, and combining them might even result in a synergistic outcome. The cost of adding technology upon technology into the already fully packed examination room for these high-end procedures is, however, to be justified. Health technology assessments that model the (prevented) burden and costs of new diagnostic and therapeutic workflows are needed to evaluate the added value of finding and treating these lesions at an earlier rather than at a later phase of disease, and how much this procedure may cost.

Conflict of Interest Statement

Both authors state no conflicts of interest in direct relation to this editorial comment. Our department has received unrestricted research grants from Astra Zeneca Oncology, Pentax Medical, and Philips Medical for past, and ongoing research projects in the area of interventional and navigation bronchoscopy. Detailed COI statements are supplied in all our prior scientific publications in this area.

Funding Sources

No funding was received.

Author Contributions

Both authors contributed equally and approved the final version of this manuscript. Both authors gave their substantial contribution to the conception and design and review of the relevant literature and interpretation thereof. Both authors wrote and revised this manuscript critically for intellectual content and gave final approval of the version to be published and agree to be 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.