O-001 Pre-treatment CTA ASPECTS as a predictor of clinical outcome in endovascular stroke therapy (EVT): results from the penumbra START trial. (4th July 2012)

• Record Type: Journal Article
• Authors: Frei, D; Yoo, A; Heck, D; Hellinger, F; McCollom, V; Fiorella, D; Turk, A; Malisch, T; Zaidat, O; Alexander, M; Devlin, T; Levy, E; Shah, Q; Hui, F; Goyal, M; Ghodke, B; Shaibani, A; Harrigan, M; Jovin, T; Madison, M; Chaudhry, Z; Gonzalez, R; Barraza, L; Sit, S; Bose, A
• Is Part Of: Journal of neurointerventional surgery. Volume 4(2012)Supplement 1
• Page Start: A1
• Page End: A1
• Publication Date: 2012-07-04
• Subjects: Nervous system -- Surgery -- Periodicals; Cerebrovascular disease -- Surgery -- Periodicals; 617.48
• DOI: 10.1136/neurintsurg-2012-010455a.1
• Languages: English
• ISSNs: 1759-8478

Abstract

Abstract : Introduction/purpose: There is no standard imaging approach for EVT patient selection. CT remains the most widely used modality for stroke evaluation. Ischemic change on non-contrast CT (NCCT) quantified using ASPECTS has been demonstrated to predict clinical response to EVT. To date, definitive studies evaluating the impact of CTA source image (CTA-SI) pre-treatment ASPECTS (pre-ASPECTS) on outcomes following EVT are lacking. START was a prospective, multicenter study to evaluate the influence of pre-treatment core infarct size in patients undergoing endovascular stroke therapy using the Penumbra System. Materials and Methods: The imaging method was at each center's discretion and included NCCT, CTA-SI, CT perfusion, or MRI diffusion imaging. This study focused on the preliminary CTA-SI results. Results are reported from an interim analysis of the START trial data as adjudicated by a central Core Laboratory. Graded in a blinded fashion, ASPECTS was analyzed according to the a priori classification (0–4, 5–7, 8–10), as well as using the entire scale. Clinical outcomes were dichotomized as 90-day modified Rankin Scale scores of 0–2 (good) vs 3–6. Univariate and multivariate analyses were performed to determine predictors of outcome. Results: Of the 147 patients enrolled, 77 met study criteria for this interim analysis. The mean age was 66.0±14.1 years; median NIHSS was 19 (14–24). Target vessel occlusions were in the ICA (22.1%), MCA (75.3%), and other (2.6%). The median pre-ASPECTS on CTA-SI was 6 (4–7). There were 20 (26%) patients with scores of 0–4, 43 (55.8%) with 5–7, 14 (18.2%) with 8–10. The rate of TIMI 2–3 revascularization was 85.3% (64/75). The median time from groin puncture to aspiration discontinuation was 71.5 (40–108) min. 37 (48.1%) patients achieved a good 90-day outcome. 22 (28.6%) died. Four (5.2%) patients suffered from symptomatic hemorrhage, and 11 (14.3%) suffered from asymptomatic hemorrhage. Higher pre-ASPECTS on CTA-SI was significantly associated with good outcomes (median 6 (IQR 5–7) vs 5 (IQR 3–7), p<0.05). The rate of good outcomes was 20.0% for ASPECTS 0–4, 55.8% for 5–7, and 64.3% for 8–10 (p=0.08). Adjusting for age and NIHSS and comparing ASPECTS 0–4 with 5–10, pre-ASPECTS 5–10 was an independent predictor of good outcome (OR 6.8, p=0.006). In ROC analysis, ASPECTS >4 was the optimal threshold for identifying good outcomes (89% sensitivity, 38% specificity). Other univariate predictors of good outcome were lower age (p=0.01), lower NIHSS (p=0.04), revascularization time (p<0.0001), and shorter time from groin puncture to aspiration cessation (p=0.0004). Conclusion: Higher pre-treatment ASPECTS on CTA source images are associated with better outcomes following EVT. Comparative studies with NCCT ASPECTS are required to evaluate relative accuracy for patient selection. Competing interests: D Frei: None. A Yoo: None. D Heck: None. F Hellinger II: None. V McCollom: None. D Fiorella: None. A Turk III: None. T Malisch: None. O Zaidat: None. M Alexander: None. T Devlin: None. E Levy: None. Q Shah: None. F Hui: None. M Goyal: None. B Ghodke: None. A Shaibani: None. M Harrigan: None. T Jovin: None. M Madison: None. Z Chaudhry: None. R Gonzalez: None. L Barraza: Penumbra, Inc. S Sit: Penumbra, Inc. A Bose: Penumbra, Inc.