Full robotic endovascular treatment for various head and neck hemorrhagic lesions

Abstract

Background

The use of robotics in neuro-interventional surgery is continuously expanding with promises to increase accuracy and safety. We have previously described the first human, full robotic intervention using the CorPath GRX Robotic System. Here we report a series of fully robotic interventions and outcomes using this robotic device for endovascular treatment of extracranial embolization.

Methods

The patient and disease characteristics, procedural details, and imaging outcomes of consecutive patients undergoing robotically assisted extracranial embolization between October 2021 and August 2022.

Results

Six patients underwent robotically assisted extracranial embolization of head and neck hemorrhagic lesions. Four patients presented with uncontrolled epistaxis or hemoptysis associated with nasopharyngeal tumors and one patient had epistaxis due to COVID-19 infection complications. All patients were treated with coil embolization and four patients were also treated with particles. All procedures were performed with robotic intervention, whereas four procedures required partial manual conversion. There was no morbidity or complications associated with the procedures and desired outcome was achieved in 100% of the procedures.

Conclusion

Our series support the feasibility of complete robotic intervention using the CorPath GRX System with accurate control of the guiding catheter, microcatheter, wire, and coil delivery with successful treatment of acute extracranial hemorrhagic lesions.

Introduction

Recent advances in robotics in medicine have expanded into the field of neuro-interventional surgery. Robot-assisted interventions are promising to increase procedural and technical accuracy, reduce radiation exposure to the interventionalists, and increase access to care by performing remote robotic procedures. The CorPath platform (Corindus, A Siemens Healthineers Company, Waltham, MA) is currently the only commercially available robotic device for endovascular surgery. ¹ The CorPath platform has been in use since 2012 for percutaneous coronary intervention and was also approved in the United States for peripheral vascular intervention.^2,3 In the past 2 years, there has been an increasing number of reported procedures including diagnostic cerebral and spinal procedures as well as interventional procedures such as carotid artery intervention, and the treatment of cerebral aneurysms.^4,5 However, no prior studies have evaluated the feasibility and efficacy of complete robotic intervention including guiding catheter manipulation, microcatheter and microwire manipulation, and delivery of coils to the target location using the robotic system. ⁴

We have previously described the first case of complete, robotic-assisted neuro-endovascular intervention for the treatment of epistaxis. ⁵ Here we report a series of complete robotic interventions and outcomes using this robotic device for endovascular treatment of extracranial lesions using coil embolization with or without particle embolization, as well as the degree of manual assistance required for the completion of robotically performed procedures.

Method

This was a retrospective single-institution, single-operator case series. A series of adult patients with extracranial pathologies were included and treated with robotic-assisted neuro-intervention at our institution between October 2021 and August 2022. We considered patients with external carotid artery branch supply to the head and neck hemorrhagic lesions eligible for coiling with or without particle embolization.

Diagnostic angiography and embolization of arterial source for hemorrhage were planned using the CorPath GRX Robotic System. The CorPath GRX Robotic System is currently cleared for percutaneous coronary and peripheral vascular interventions in the United States, and extracranial carotid artery selection is a part of the current CorPath indications. The robotic console is positioned outside of the angio-suite. It is equipped with a 26-inch monitor to view live biplane fluoroscopic views and controls consisting of 3 joysticks and a touchscreen. The robotic system is capable of advancing, retracting, and rotating the catheter and guidewire separately. It also includes a side port capable of deploying an additional device such as coils.

Procedures

The cerebral angiography procedure was performed under general anesthesia using a femoral artery approach. In an initial manual phase of the procedure, a 5-French or 6-French Pinnacle groin sheath was inserted into the groin. A 5-French or 6-French Envoy catheter (Cerenovus), 90 cm in length, was then placed in the descending aorta. The catheter was then connected to a co-pilot hemostatic valve and a microguidewire (V18 Control Wire, Boston Scientific, Marlborough, MA, USA) was inserted for subsequent robotic manipulation. The catheter was continuously flushed with heparinized saline and another side port was connected to an extended connection tubing that allowed for contrast injections manually or using a contrast media injector. The robotic arm was then brought into position and the catheter–guidewire combination was loaded into the single-use cassette and secured to the groin sheath.

The primary operator navigated the catheter to the target location in the proximal external carotid artery over the wire robotically while sitting at the console outside the angiography room. At the bedside, the supporting operator monitored flush lines and connection tubing as well as manual contrast injections from the side port as needed. Once the Envoy guiding catheter was placed in an optimal position, the wire was removed. The Envoy was then manually unloaded from the robotic cassette (while maintained in position in the external carotid artery) and the robot was loaded by a microcatheter (0.017–0.021) over a Synchro-14 Support microwire (Stryker Neurovascular), so that the microcatheter–microguidewire combination was loaded into the robotic cassette and secured to the Envoy guide sheath. Using the robotic console, the microcatheter was then navigated over the microwire into an ideal position for embolization using coils or polyvinyl alcohol (PVA) particles (Contour PVA 250-355, Boston Scientific) and embolization was performed. The coils were then delivered through the device port and deployed at the tip of the microcatheter using the robot.

Results

Six patients underwent robotically assisted extracranial embolization of head and neck hemorrhagic lesions between October 2021 and August 2022. The median age was 51 years (range 33–75 years). Embolization was performed in various branches of the external carotid artery (ECA) including the internal maxillary artery, facial artery, lingual artery, ascending pharyngeal artery, and the ascending palatine artery (Table 1). Four patients presented with uncontrolled epistaxis or hemoptysis associated with nasopharyngeal tumors and one patient had epistaxis due to COVID-19 infection complications.

Table 1.

Summary description of cases performed with complete robotic approach.

	Condition	Vessels selected	Catheters/devices	Embolization method	Manual conversion	Reason for conversion	Success	FT (min)	DAPcGy × cm2
Case 1	Hemoptysis/tumor	R CCA,R ECA,R AphA	5F Envoy/V18; Echelon-10/Synchro-14 Support	Coils	Partial	Need to control guide cath and MC together	Yes	19.7	111
Case 2	Hemoptysis/tumor	L CCA,L ECA,L facial a	Glidecath XP/V18; Echelon-10/Synchro-14 Support	PVA + nBCA	Partial	Second facial branch selected manually as MC contrast injection needed	Yes	27.7	122
Case 3	Hemoptysis/tongue SCC	R CCA,R ECA,R lingual a	5F Envoy/V18; Echelon-10/ Synchro-14 Support	PVA + coils	Partial	Particle remained in MC, coil delivery failure	Yes	38.6	105
Case 4	Epistaxis/COVID	b/l CCA, b/l ECA, b/l IMAX a	5F Envoy/V18; Rapidtransit/ Synchro-14 Support	PVA + coils	No	n/a	Yes	31.4	149
Case 5	Epistaxis/tumor	R CCA,R ECA,R AphA	5F Envoy/V18; Echelon-10/ Synchro-14 Support	coils	Partial	Size incompatibility	Yes	18	122
Case 6	Epistaxis	R CCA,R ECA,R IMAX/SPA	5F Envoy/V18; Echelon-10/ Synchro-14 Support	PVA + coils	No	n/a	Yes	26.7	177

AphA: ascending pharyngeal artery; CCA: common carotid artery; DAP: dose area product; ECA: external carotid artery; FT: fluoroscopy time; IMAX: internal maxillary artery; MC: microcatheter; nBCA: n-butyl cyanoacrylate; PVA: polyvinyl alcohol; SCC: squamous cell carcinoma; SPA: sphenopalatine artery.

Figure 1 demonstrates pre (Figure 1A) and post (Figure 1B) robotic-assisted embolization of intervention in a patient in 70s with a 12-pack-year smoking history and a recent diagnosis of squamous cell carcinoma of the right buccal mucosa status post right composite resection of the mandible, cheek, and intra-oral mucosa with right modified neck dissection, who developed acute large volume oral bleeding. Successful robot-assisted partial coil embolization was performed for the right ascending pharyngeal artery branch exhibiting dysplastic vessels supplying a region of abnormal hypervascularity consistent with the cause of hemorrhage (case 1).

Figure 1.

(A) Pre-embolization angiogram of the right ascending pharyngeal artery. (B) Postembolization angiogram of the right ascending pharyngeal artery.

Figure 2 represents a patient in 60s who presented with tumor-associated oral hemorrhage localized to the territory of facial artery on angiography. Successful catheterization and embolization of one of the facial artery feeders to the tumor using PVA (250-355) was performed using the robotic system; however, selection of a second facial artery tumor feeder and further by n-butyl cyanoacrylate (Cerenovous) embolization required manual conversion (case 2).

Figure 2.

(A) Pre-embolization angiogram of the left facial artery. (B) Postembolization angiogram of the left facial artery.

Figure 3 shows pre-embolization and postembolization images of a patient in 60s who presented with oral hemorrhage and a history of tongue squamous cell carcinoma, status postresection of the base of the tongue tumor and a graft arterial branch from the right external carotid artery. Robotic-assisted selection of the lingual branch associated with the tumor hypervascularity was performed using the robotic system; however, manual conversion was required for coil embolization of this feeder due to remnant particles in the microcatheter and coil delivery failure (case 3).

Figure 3.

(A). Pre-embolization angiogram of the right lingual artery. (B) Postembolization angiogram of the right lingual artery.

Table 1 summarizes clinical and procedural characteristics and technical details for the six cases undergoing robotic-assisted intervention. Transfemoral access was used in all cases. The average fluoroscopy time from groin puncture to closure was 25.2 min (range 18–35 min), which included diagnostic angiography and intervention.

Treatment efficacy and adverse events

All patients were treated with coil embolization and four patients were also treated with particles. ll procedures were initiated with robotic intervention, however, four procedures required partial manual conversion. The reasons for the conversions included the necessity to manipulate both guiding catheter and microcatheter simultaneously that was not feasible with the current robotic system, and to perform multiple intermittent super-selective angiographies via the microcatheter that would have required multiple unloading and loading of the microcatheter and microguidewire resulting in prolonged procedure time (Table 1).

There was no morbidity or complications associated with the procedures and desired outcome was achieved in 100% of the procedures. There were no cases of intraprocedural underlying vascular dissection or vasospasm or embolic events. There was no need for subsequent endovascular intervention. All six patients had at least one clinical follow up with no recurrence of hemorrhage and no procedure-related morbidity at discharge. There were no deaths, permanent neurological deficits, or other robotic-related complications.

Discussion

Robot-assisted procedures are growing for the treatment of patients in the field of neuro-intervention.^6,7 In this retrospective study of six patients, we demonstrated the feasibility of complete robotic intervention for embolization of extracranial head and neck hemorrhagic lesions.

This is the first and largest series of robot-assisted complete neuro-interventions. The CorPath GRX system requires manual placement of the catheter close to the target site for intervention. Thus, for neuro-interventional procedures, the use of the Corindus robot is often limited to the hybrid approach, where the extracranial placement of a guiding catheter and the navigation of an intermediate catheter is performed manually and a microcatheter is navigated intracranially to the point where the working length of the robotic system, which is a total of 20 cm, would be sufficient to perform the rest of the procedure. Only then, the intracranial manipulation of a microcatheter and implants is conducted robotically. We demonstrated the feasibility of this approach for complete robotic neuro-intervention where both the guiding catheter and microcatheter are manipulated by the robot. This is an important concept as complete robotic interventional procedures lay foundations for the future remote neuro-interventional procedures such as remote mechanical thrombectomy. In addition to the feasibility, we demonstrated successful procedural and discharge clinical outcomes in patients treated with robotic assistance. The assumption that the completeness of the procedures is compromised in robotic interventions was not shown in our study.

Appropriate selection of neuro-interventional equipment and consideration of the limited working length of the catheters with respect to the robotic arm position is critical for successful complete robotic interventions. The robot should be maximally pulled back before exchanging the guiding catheter with the microcatheter in order to allow for the full 20 cm of the forward working length of the robotic arm. For the interventional equipment, we recommend a 5-French long guide catheter (Envoy in our series) and a 6-French system in older patients for additional support in the aortic arch. Since there is a limit in the amount of unnecessary microcatheter length that the robotic system can hold, a longer guiding catheter is recommended or the use of a long microcatheter should be avoided.

Both the guiding catheter and microcatheter and coiling procedure were primarily manipulated using a push–pull and rotation joystick control solely based on the visual information. In our series, we used a Synchro Support microguidewire that provided proximal support for the proximal microcatheter (within the robotic cassette and not covered by the guiding catheter) that allowed for better pushability and navigation of the microsystem. Considering the limited working length of the robotic arm, the proximal support that prevents the formation of excessive microcatheter slack offers a tremendous benefit. Careful attention is needed to prevent buckling of the coils in the cassette as it is pushed into the microcatheter robotically. A variety of coils with appropriate sizing to the target vessel were successfully used for coil embolization. This was achieved with careful attention to the deflection of the coil loops and the microcatheter tip without the need for haptic feedback, resulting in precise and controllable delivery of the coil loops in the target location.

Reasons for manual conversion could be diverse depending on the clinical and anatomical considerations. These can include navigation of complex anatomical situations requiring placement of curves on wires or steam-shaping microcatheters, concomitant manipulation of guiding catheter with the microsystem, requirement for multiple concomitant super-selective angiograms, or robotic cassette failure requiring replacement. One of the limitations of interventional procedures using the CorPath is the lack of active robotic control of the guide catheter while manipulating the microcatheter and the microwire, making it one of the main reasons for the requirement of manual assistance during the robotic catheterizations and treatment in challenging anatomical conditions. Among our series of five interventions, this limitation resulted in partial manual conversion as microcatheterization of a small proximal ECA branch tumor feeder was not feasible without concomitant manipulation of the guide catheter and the microsystem.

The CorPath system has been used to perform carotid artery stenting or intracranial aneurysm embolization.^7,8 In all prior studies of carotid artery stenting (CAS), the interventionalist manually performed balloon angioplasty and deployment of stents for the carotid interventions, whereas the Corindus robot was used to navigate over the wire equipment to the ideal position for deployment. This is because it is not feasible to deploy the devices robotically using the peripheral Corindus device. One group compared robotic and manual CAS and found robotic CAS to have a significantly longer mean procedure duration, but there were no significant differences in fluoroscopy time, radiation exposure, or complications in their small sample of six robotically treated and seven manually treated patients. The upgraded CorPath GRX system has been first used for the treatment of a patient with a sidewall distal basilar aneurysm that was treated with a stent-assisted coiling procedure. This procedure was performed using a 6-French sheath that was manually placed in the right subclavian artery and placement of an intermidiate catheter in the right V4 vertebral artery which was then loaded to the robot, using which a 1.7-French microcatheter was advanced to the ideal position over a 0.014-inch microwire robotically. From this position, the microwire was replaced with a self-expanding stent, which was deployed across the aneurysm neck.

There are various advantages associated with the use of robotic interventions. These systems can reduce the occupational hazards to the operator, including radiation exposure and spinal injuries from long-term use of lead aprons, or the potential transmission of infectious pathogens such as COVID-19. In a study of robotic coronary intervention, radiation exposure for the primary operator was reduced by 95% compared with the usual position next to the operating table. ⁹ Additionally, it is hypothesized that errors during procedures could be reduced given less fatigue secondary to better ergonomic positioning and closer proximity to the screen. Most importantly, the evolution of robotic interventional systems could allow for the remote treatment of neuro-endovascular conditions such as acute large vessel occlusion strokes, therefore increasing access to mechanical thrombectomy in shorter times and faster reperfusion times in acute stroke. ¹⁰

There are several limitations to the widespread adoption of robotic systems in neuro-interventions. Current systems are primarily developed for peripheral and coronary interventions. There is a lack of clinical evidence with currently no trials for any robotic system to evaluate the efficacy or improved clinical outcomes as compared to manual procedures. In most cases, one of the staff is still recommended to stay at the bedside for loading and unloading the interventional equipment within the robotic arm, monitor the lines, adjust the fluoroscopic table and connections, and perform hand injections as needed.

Another limitation is the lack of haptic or tactile feedback including forces between the catheters and the wire/device/or blood, or forces between devices and the vessel. In our series, careful attention to visual feedback from forces exerted on the catheters and coils was sufficient for complete successful coil embolization. In the beginning, procedures may be prolonged due to the extra time required for adapting the set-up to the robot. Hybrid robotic procedures or in vitro practice sessions prior to performing complete robotic procedures are important steps to reduce procedural time and radiation and to increase the procedural success rate. Future robotic designs may be improved by lengthening the robotic cassette tubing to enhance microcatheter compatibility. Other current challenges include the cost of systems and the lack of cost-effectiveness studies with the use of these systems. Future engineering modifications to consider these limitations, along with clinical trials to investigate the efficacy of robotic interventions are needed to examine the reliability and safety of robotic systems for remote neuro-intervention without expert neuro-interventional manual assistance.

Conclusion

Our series support the feasibility of complete robotic intervention using the CorPath System with accurate control of the guiding catheter, microcatheter, wire, and coil delivery with successful treatment of acute extracranial hemorrhagic lesions. Successful outcomes results were achieved with no procedure-related morbidity at discharge.