O-013 early ischemic core volion volumes predict infarct size in swift prime

Abstract

<h3>Background</h3> Previous studies suggest early imaging of stroke patients with CT perfusion or DWI can estimate the size of the irreversibly injured core and thresholded perfusion images (time to the maximum of the tissue residue function [Tmax] &gt;6 s) can approximate the volume of critically hypoperfused tissue. Furthermore, the union of baseline core and post-treatment hypoperfusion volumes predict infarct volume. <h3>Methods</h3> This is a pre-specified analysis of SWIFT-PRIME. Baseline ischemic core lesions (defined as regions with CBF reduced by &gt;70% on CTP or ADC &lt;620 on DWI) and Tmax &gt;6 s hypoperfusion volumes were assessed prior to randomized treatment with IV tPA alone vs. IV tPA + endovascular therapy (Solitaire stent-retriever) using RAPID automated software. Reperfusion was assessed with angiographic TICI scores at the end of the procedure (endovascular group) and 27 hr Tmax &gt;6 s volumes (both groups). Infarct volume was assessed at 27 hrs on CT or MRI (FLAIR). <h3>Results</h3> 140 patients with baseline penumbral imaging and no PH1/2 hematoma on the 27 hr scan were eligible for this substudy. Baseline imaging (&lt;4.5 h from stroke onset) was performed with CT perfusion in 124 (89%) and multimodal MRI in 16 (11%). Follow-up imaging (at 27 hrs) was performed with MRI in 74 (54%) and CT in 63 (46%). The median baseline core volume (N=140) was 4 ml (IQR 0–15). The median 27 hr infarct volume in patients who achieved &gt;90% reperfusion (based on Tmax &gt;6 s or TICI 2b-3, if Tmax not available) was 16 ml vs. 124 ml for patients with &lt;10%/TICI 0–1 reperfusion (P &lt; 0.0001, Wilcoxon-rank sum). 88% of the patients in the endovascular group achieved TICI 2b-3 reperfusion; in these patients there was a significant correlation between early core volume and final infarct volume (r = 0.47; P = 0.0002); median absolute difference 14 ml (10 ml for patients with Target mismatch profile [N&nbsp;=&nbsp;46]; 39 ml for patients with the Malignant profile [N&nbsp;=&nbsp;9]). 61 endovascular patients (85%) and 18 patients in the tPA alone group (42%) had &gt;90% reperfusion at 27 hrs (based on Tmax &gt;6 s/TICI 2b-3); these patients had a significant correlation (P &lt; 0.0001) between early core volumes and 27 hr infarct volume (endovascular group, median difference 13 ml; tPA alone group, median difference 11 ml). 12 patients (endovascular and tPA group combined) had &lt;10% reperfusion (or TICI 0/1); in these patients the correlation between baseline Tmax &gt;6 s perfusion volume and final infarct volume was r = 0.51; P = 0.09). In Target mismatch patients (N&nbsp;=&nbsp;86), the union of baseline core and early follow-up Tmax &gt;6 s volumes (predicted infarct volume) strongly correlated with the 27 hr infarct volume (r = 0.77; P &lt; 0.0001); the median absolute difference between the observed and predicted volume was 12 ml. <h3>Conclusions</h3> In SWIFT PRIME baseline ischemic core volume reasonably approximates 27 hr infarct volumes in reperfused patients (both endovascular and tPA alone groups) and baseline hypoperfusion volumes approximate 27 hr infarct volumes in patients who do not reperfuse. Among Target mismatch patients, the union of baseline core and early follow-up hypoperfusion volume predicts 27 hr infarct volume (median absolute difference 12 ml). <h3>Disclosures</h3> G. Albers: 2; C; Covidien, iSchemaView. 4; C; iSchemaView. A. Bonafe: None. H. Diener: None. E. Levy: None. V. Pereira: None. C. Cognard: None. D. Cohen: None. W. Hacke: None. O. Jansen: None. T. Jovin: None. H. Mattle: None. R. Nogueira: None. A. Siddiqui: None. D. Yavagal: None. B. Baxter: None. T. Devlin: None. D. Lopes: None. V. Reddy: None. R. du Mesnil de Rochemont: None. O. Singer: None. R. Jahan: None. M. Goyal: None. J. Saver: None.