Training and Supervision of Thrombectomy by Remote Live Streaming Support (RESS) : Randomized Comparison Using Simulated Stroke Interventions.

Abstract

Stroke patients are excluded from expeditious thrombectomy in regions lacking neurointerventional specialists. An audiovisual online streaming system was tested, allowing aneurointerventional specialist located at aneurovascular center to supervise and instruct athrombectomy performed at adistant hospital without being physically present (remote streaming support [RESS]). In total, 36thrombectomy procedures were performed on aMentice endovascular simulator by six radiologists not specialized in neurointerventions. Each radiologist was challenged with six different endovascular simulation scenarios under alternating conventional local support (specialist inside the room [LOS]) and RESS, which was performed using an advanced live streaming platform. Both support modes led to amedian of 2attempts (interquartile range [IQR] 2.0-2.0 each) until successful recanalization. There was no statistically significant difference in time from first catheter insertion to recanalization between LOS (median24.9 min, IQR 21.0-31.5 min) and RESS (23.9 min, IQR 21.7-28.7 min, p = 0.89). The percentage of thrombi covered by the stent-retriever and average speed when retrieving the stent-retriever (3.7 mm/s, IQR 3.25-5.35 mm/s vs. 3.6 mm/sec, IQR 2.5-4.7) were similar in both groups. Fluoroscopy time did not differ (19.0 min, IQR 16.9-23.5 min vs. 19.9 min, IQR 15.9-23.5 min) with atrend towards increased median amounts of contrast medium used under RESS (62.9 ml vs. 43.1 ml; p = 0.055). This study confirmed the feasibility of RESS for thrombectomy procedures in asimulated environment. This serves as basis for future studies planned to analyze the effectiveness of RESS in areal-world environment and to test if it improves the learning curve of interventionalists with limited thrombectomy experience in remote areas.