Abstract 3820: Penumbra Imaging Collaborative Study (PICS): Utilization of Imaging for Patient Selection and Its Impact on Outcomes Following Penumbra System ® Treatment

Abstract

Purpose: Various imaging approaches are utilized for evaluating acute stroke patients. Despite numerous studies, there is no clear evidence that one approach is superior to another for identifying which patients will benefit from mechanical thrombectomy (MT). The PICS registry was established to determine the imaging modalities being employed to triage acute stroke patients for MT and to assess their impact on patient functional outcome. Methods: Patients were enrolled across 35 centers. For each patient, the baseline imaging modality was recorded to determine the utilization rates of noncontrast CT (NCCT), CT perfusion (CT-P), and MRI diffusion weighted imaging (DWI). All patients subsequently treated by the Penumbra System per standard of care were followed for 3 months after the procedure to assess functional outcome using the modified Rankin Scale (mRS). Results: A total of 305 patients were enrolled in this registry, of which 267 had the requisite imaging modalities used for this analysis. Of these, 49.4% were females. Mean age was 66.6 ± 15.9 years; median NIHSS score was 17.0 (IQR 12-21). The median time from stroke onset to presentation was 2.3 hours, from stroke symptom onset to arterial puncture was 4.8 hours, and from arterial puncture to end of thromboaspiration was 73.0 minutes. Post-treatment evaluation revealed that 83.5% of patients were successfully recanalized to TIMI 2/3 (from TIMI 0/1) with 43.5% achieving a 90-day mRS of 0-2. All cause mortality was 19.9% with symptomatic intracranial hemorrhage reported in 4.5% of the patients. The principal imaging modalities used for patient triage were NCCT in 61.8%, CT-P in 27.3% and DWI in 10.9% of the study population. Of the patients selected for NCCT, 42.5% were functionally independent at 90 days, whereas for CT-P and DWI, the rates were 45.9% and 42.9%, respectively. There was no significant difference between groups for patient age, baseline NIHSS, time from onset to presentation, time from arterial puncture to end of thromboaspiration. The time from stroke symptom onset to groin puncture was significantly different in the DWI cohort (p<0.0001) with a median of 4.4 hrs for NCCT, 4.8 hrs for CTP, and 6.5 hrs for DWI. There was a significant difference in % of female patients with CTP 37.0%, DWI 48.3%, and NCCT 55.2% (p=0.034). Conclusion: The type of imaging modality utilized to triage acute ischemic stroke patients for MT varies widely across the US. Based on the present findings, the imaging approach appears to have little impact on patient functional outcome after MT.