Purpose To evaluate the performance of a new microcatheter based device (ReStore, Reverse Medical, Irvine, California, USA) for endovascular mechanical thrombectomy/flow restoration in a swine thromboembolic model and to compare its efficacy over other available devices for mechanical thromboembolectomy.

Method We investigated the use of ReStore for thrombectomy treatment against similar control devices. The cerebral vasculature swine model was utilized in which 12 of each carotid artery branches were prepared with prophylactic papaverine due to the vasospastic nature of the swine model and occluded with radio-opacified blood thrombi. The thrombi were prepared in such a way to represent characteristics of those found in acute stroke patients. The thrombi were injected into target arteries via a 6 Fr guiding catheter. The thrombi were allowed to set up in situ for at least 15 min until complete localized occlusion of the target vessels were confirmed by angiography. Once the occlusion was confirmed, ReStore or a control device was advanced via a 6 Fr guide catheter and precisely positioned across the target thrombi over a microguide wire. No flow arresting means was utilized during retrieval or recanalization attempts. Only localized aspiration was conducted via the 6 Fr guide catheter, once the ReStore or control device holding retrieved thrombi was pulled down to the tip of the guide catheter. Parameters collected for this assessment study were TIMI score, number of retrieval attempts, per cent clot retrieved, time to recanalization, thrombus fragmentation and loss of thrombus at the tip of the guide catheter.

Results We found higher TIMI scores (OR 9; 95% CI 0.28 to 285.3), recanalization on the first attempt (9/10 vs 0/2; p<0.01) and per cent clot retrieved (74% vs 20%; p=0.06) with ReStore over the control devices while procedure times for ReStore appeared to be the same as that of the control device. Despite the use of prophylactic papaverine, vasospasm was still prevalent in all cases, even during guidewire and catheter manipulations. No vessel dissections or device failures were encountered during this study. Study recanalization rates for ReStore (9/10, 90%) in the swine model, which is much less tortuous than humans, and the radio-opacified thrombi allowed for more accurate target positioning and retrieval.

Conclusion ReStore is effective in removal of thrombi and recanalization of blood flow within the swine cranial arteries.

Abstract O-006 Figure 1