Abstract

Introduction: Endovascular thrombectomy (EVT) reduces disability in patients with large ischemic core stroke. Increasing core volume is prognostic but does not modify treatment effect. CT hypodensity can vary within an area of core. We assessed the relationship between the severity of CT core hypodensity and clinical outcome in the SELECT2 trial. Methods: Non-contrast CT ischemic core was drawn manually by two blinded investigators with consensus review by a third investigator. An absolute Hounsfield unit (HU) threshold for severe hypodensity was set at the lower 99% confidence interval [CI] for normal contralateral thalamic grey matter. Outcomes assessed included ordinal mRS, functional independence (mRS 0-2), independent ambulation (mRS 0-3), and mortality at 90 days. The association between the volume of severe hypodensity and outcome was adjusted for age, baseline NIHSS, and total baseline core volume, via logistic and proportional odds models with treatment*hypodensity interaction. Results: Of 317 assessable patients, the median hypodensity of ischemic core was 31HU (IQR 28-33). Median density of normal thalamus was 38HU. The severe hypodensity threshold (lower 99% CI) was 26HU with median volume affected 14mL (IQR 5.6-29.2) or median 19% (IQR 9-35%) of the total hypodensity volume. The volume of core ≤26HU (per 1 mL increase) was associated with a 90-day mRS shift towards worse outcomes in EVT patients (n=161, adjusted common odds ratio, acOR 1.03, 95%CI 1.01-1.05) but not medical management ([MM], n=156, acOR 1.00, 95%CI 0.98-1.02, p-interaction 0.005); lower functional independence in EVT (aOR 0.96 95%CI 0.93-0.99) but not MM (aOR 1.02, 95%CI 0.99-1.06, p-interaction 0.004); lower independent ambulation in EVT (aOR 0.97, 95%CI 0.94-0.99) but not MM (aOR 1.01, 95%CI 0.98-1.04, p-interaction 0.002) and higher mortality in EVT (aOR 1.03, 95%CI 1.01-1.05) but not MM (aOR 1.00, 95%CI 0.98-1.02, p-interaction 0.02). Conclusion: In patients with large ischemic core, areas of significant CT hypodensity are associated with worse clinical outcomes after EVT but not MM. This is independent of total core volume and modifies the treatment effect of EVT. This may improve prognostic decision-making and drive developments in automated image processing.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call