Abstract 161: Addition of MRI for Patient Selection in Intra-arterial Stroke Therapy Leads to Better Clinical Outcomes, a Pre-Post Study

Background: Intra-arterial (IA) therapies are becoming prevalent in the acute ischemic stroke management. Despite increasingly rates of vessel recanalization, outcomes have not shown substantial improvement. This discordance is may be due problems with patient selection. Diffusion weighted imaging (DWI) lesion volume >70 cm 3 was recently shown to correlate with poor outcome despite vessel recanalization. On April 30, 2010, our institution implemented a protocol to perform hyperacute MRI scanning on patients eligible for IA therapy to improve patient selection. Following CT/CTA, to assess for established infarct and large vessel occlusion, patients were immediately taken to MRI for imaging including DWI, FLAIR and PWI. Based on the results of the MRI, determination for IA therapy was made. Objective: To assess if enhanced patient selection utilizing pre-treatment MRI improves patient outcomes following IA acute stroke therapy. Methods: Single, tertiary center retrospective study of patients eligible for IA therapy based on CT/CTA imaging. Patients prior to April 30, 2010 were selected based upon the results of the CT/CTA alone, whereas patients on or after April 30, 2010 were selected based upon the results of CT/CTA and MRI. Demographic, radiological, and clinical data was collected. Change in modified Rankin Scale (mRS) distribution pre- vs. MRI protocol was analyzed by the Wilcoxon Rank Sum test and t-test. Results: 267 patients with acute large vessel occlusion presented to our institution from July 2006 to June 2011. 171 patients (95 males, 76 females, mean age 67 years +/- 15 years) during the pre-MRI period, 158 (92%) of whom were treated with IA therapy; and 96 (43 males, 53 females, mean age 68 +/- 15 years) during the MRI period, 43 (45%) of whom were treated with IA therapy [Chi-square p Conclusions: Following the implementation of the MRI based selection for patients with acute large vessel occlusion for IA therapy, we demonstrated a significant one-category improvement in the mRS. Our data strongly suggest that this approach significantly reduces resource-intense IA therapy utilization while significantly improving overall outcomes, presumably by targeting patients more likely to benefit and removing patients unlikely to benefit or even be harmed by IA therapy. Prospective randomized studies are needed to further assess the clinical benefit of using MRI as a selection tool for IA therapy.