Association of Ischemic Core Hypodensity With Thrombectomy Treatment Effect in Large Core Stroke: A Secondary Analysis of the SELECT2 Randomized Controlled Trial

BACKGROUND: We aimed to determine whether extensive severe computed tomography (CT) hypodensity, representing blood-brain barrier injury, would be associated with a reduced benefit of endovascular therapy (EVT) in patients presenting with large core stroke. METHODS: This study is an exploratory analysis of SELECT2 (Randomized Controlled Trial to Optimize Patient’s Selection for Endovascular Treatment in Acute Ischemic Stroke), a randomized controlled trial of EVT versus medical management in patients with large ischemic core who presented to 31 comprehensive stroke centers across the United States, Canada, Europe, Australia, and New Zealand. Visible CT hypodensity was outlined, and a threshold of severe CT hypodensity was defined as the lower 99% CI of contralateral thalamic gray matter in Hounsfield units (HU). The association between the volume of severe CT hypodensity and modified Rankin Scale (mRS) score of 0 to 3 was evaluated using logistic regression models, with adjustment for age, National Institutes of Health Stroke Scale, total noncontrast CT core volume, and a volume-by-treatment interaction. The relationship between severe CT hypodensity volume and the probability of an mRS score of 0 to 3 was used to select clinically relevant volume cut points for further evaluation. The treatment effect of EVT versus medical management on independent ambulation and hemicraniectomy was assessed in 2 subgroups based on these volume cut points. RESULTS: In 322 patients, the median CT density was 31 HU (interquartile range, 28–34). The selected threshold of severe CT hypodensity was 26 HU. The volume of ischemic core ≤26 HU (per 1 mL increase) was associated with lower odds of mRS score of 0 to 3 after EVT (adjusted odds ratio [aOR], 0.96 [95% CI, 0.94–0.99]), but not medical management (aOR, 1.01 [95% CI, 0.98–1.03]; P interaction<0.01). In 101 patients with ≥26 mL of severe CT hypodensity, EVT, compared with medical management, was not associated with mRS score of 0 to 3 (aOR, 0.98 [95% CI, 0.33–2.88]) and was associated with hemicraniectomy (≥26 mL: aOR, 3.45 [95% CI, 1.09–10.86] versus <26 mL: aOR, 0.74 [95% CI, 0.31–1.75]; P interaction=0.03), whereas among 221 patients with <26 mL of severe hypodensity EVT was associated with mRS score of 0 to 3 (aOR, 7.20 [95% CI, 3.55–15.47]; P interaction<0.01). CONCLUSIONS: Severe hypodensity within large ischemic regions modifies the thrombectomy treatment effect and increases the likelihood of hemicraniectomy, independent of lesion volume. REGISTRATION: URL: https://www.clinicaltrials.gov ; Unique identifier: NCT03876457.