Abstract 109: Thrombectomy is Safe for Dawn- and Defuse-3-Ineligible Patients Who Present Within the Extended Window: A Subgroup Analysis From the BEST Prospective Cohort Study

Abstract

Background: Recent data have demonstrated robust efficacy for endovascular thrombectomy in acute large vessel occlusion (LVO) up to 24h after symptom onset. Given the overwhelming benefit of this intervention in a highly selective, clinical trial population, it is plausible that a benefit may also be observed in some trial-ineligible patients. Methods: Using the BEST multi-center prospective cohort of consecutive patients who underwent thrombectomy for LVO, we compared DAWN and DEFUSE 3 trial-ineligible patients treated within 6-24 hours after last known normal (LKN) to trial-eligible patients, and to untreated controls from those trials. The co-primary outcomes were an early therapeutic response (decrease in NIHSS ≥10, or NIHSS 0-1 by 24h) and good functional outcome (90-day mRS 0-2). Secondary outcomes included change in NIHSS at 24h, intracerebral hemorrhage within 72h, and discharge to home or acute rehab. Results: Of 443 patients in the BEST cohort, 159 (36%) underwent endovascular therapy between 6-24 hours after LKN, of whom 71 (45%) were trial-ineligible. The most common trial exclusion criteria were M2 (53%) and multivessel occlusions (30%). Compared to trial-eligible patients, trial-ineligible patients were younger (median 63 vs. 72y, p=0.02) but had similar baseline NIHSS (median 14 vs. 15, p=0.28), ASPECTS scores (median 8 v. 8, p=0.69), times to recanalization (median 627 vs. 682 min, p=0.08), and rates of successful recanalization (TICI 2b/3, 83% v. 86%, p=0.57). Compared to DAWN untreated controls, BEST trial-ineligible patients were more likely to have an early therapeutic response (OR 2.02, 95%CI 1.00-4.06, p=0.049) and good functional outcome (OR 3.02, 95%CI 1.33-6.86, p=0.001). Compared to trial-eligible patients in BEST, trial-ineligible patients had similar improvements in 24h NIHSS (-2 vs. -4, p=0.33), rates of intracerebral hemorrhage (28% vs. 24%, p=0.54), and rates of discharge to home or rehab (77% vs. 73%, p=0.49). Conclusion: This multicenter prospective cohort demonstrates that thrombectomy for acute LVO within 24 hours in DAWN and DEFUSE 3 trial-ineligible patients is safe and may be effective. Treating these patients may be reasonable but our findings should be confirmed by additional randomized data trials.