Olson VA, Turcotte EL, Batjer HH, Nelson BJ, Bendok BR. Operative Neurosurgery. 2026;31(2):355–360. August 2026 issue; published online October 10, 2025. DOI: 10.1227/ons.0000000000001790
This preclinical study explores whether robotic neuroendovascular procedures can be performed safely across intercontinental distances. A neurosurgeon in Arizona remotely navigated aspiration catheters through a vascular phantom located in Switzerland. The system reached every simulated intracranial target and demonstrated a rapid learning curve. Although still far from clinical application, the experiment represents an important step toward remotely delivered mechanical thrombectomy.
Objective
To test whether a neurointerventionalist could remotely perform magnetic robotic navigation and simulated aspiration thrombectomy across an intercontinental distance.
Methods
A magnetic-field generator, mechanical catheter advancer and vascular phantom were installed in an interventional suite in Switzerland. Through a wireless connection, a neurosurgeon operating from Arizona navigated a guidewire and aspiration catheter to simulated middle cerebral artery and basilar artery occlusions.
Twenty consecutive navigation procedures were performed: 10 targeting the M1 segment and 10 the basilar artery. Aspiration was attempted in four trials.
Main results
All 20 vascular targets were reached successfully from more than 5,700 miles away. Mean navigation time was 114 seconds for the M1 segment (SD 44.4) and 134 seconds for the basilar artery (SD 65.8).
All four attempted clot aspirations were successful. Navigation time decreased by an average of 77% between the initial and final trials, suggesting a rapid operator learning curve. The absence of haptic feedback did not prevent completion of the simulated procedures.
Interpretation
This preclinical experiment demonstrates the technical feasibility of intercontinental robotic neuroendovascular navigation. If validated in progressively more realistic models and ultimately in patients, this technology could help extend thrombectomy expertise to hospitals without on-site neurointerventional coverage.
The study establishes feasibility—not clinical safety, effectiveness or readiness for routine use.
Limitations
Procedures were performed in a silicone phantom rather than living vascular anatomy. The model could not reproduce arterial tortuosity, vasospasm, vessel-wall injury, thrombus variability or unexpected patient movement. Only four aspiration attempts were performed, and the study did not test the consequences of network interruption or equipment failure. Full-text verification was access-limited, but the complete indexed abstract and numerical results were available.
Clinical takeaway
Remote robotic thrombectomy is technically possible over intercontinental distances, but substantial preclinical and clinical validation, redundant communication systems and clearly defined emergency-conversion protocols will be required before human application.