Admission Phenotype Clarifies the Apparent Prognostic Contrast Between TOAST Etiologic Subtypes in Large-Core Anterior-Circulation Large-Vessel Occlusion.

OBJECTIVE

In large-core anterior-circulation large-vessel occlusion (LVO), differing admission clinical-imaging profiles complicate interpretation of prognostic differences between large-artery atherosclerosis (LAA) and cardioembolism (CE). We assessed whether these differences persisted after adjustment and whether endovascular thrombectomy (EVT)-outcome associations differed by etiology.

METHODS

We analyzed 631 patients with Alberta Stroke Program Early CT Score (ASPECTS) ≤ 5 from the 38-center MAGIC registry; 404 underwent EVT plus medical therapy and 227 received medical therapy alone. The primary outcome was 90-day modified Rankin Scale (mRS) 0-3. Mixed-effects models assessed etiology-outcome associations and treatment-by-etiology interaction. Among EVT-treated patients, Shapley decomposition quantified domain contributions to attenuation of the LAA-CE contrast.

RESULTS

Among EVT-treated patients, higher crude odds of mRS 0-3 with LAA (OR 1.59, 95% CI 1.05-2.42) were substantially attenuated after adjustment (aOR 0.92, 95% CI 0.53-1.59). No adjusted LAA-CE contrast was evident with medical therapy alone (aOR 1.12, 95% CI 0.47-2.65). In a separate fixed-effects decomposition model, the absolute log-odds contrast was attenuated by 96.2% after inclusion of measured pretreatment domains, with imaging/collateral features, demographics, and stroke severity contributing most. For the secondary mRS 0-2 outcome, adjusted odds favored CE (aOR 0.46, 95% CI 0.25-0.84). No treatment-by-etiology interaction was detected (p = 0.944).

INTERPRETATION: Admission clinical-imaging phenotype clarified the apparent LAA-CE prognostic contrast and may help reconcile discordant findings across EVT cohorts. These findings reinforce established clinical-imaging criteria as the basis for hyperacute EVT selection rather than LAA versus CE etiology alone.

TRIAL REGISTRATION: Chinese Clinical Trial Registry (ChiCTR.org.cn); ChiCTR2100051664; https://www.chictr.org.cn/.