Clinical Implications of Stroke Progressor Phenotypes defined based on ASPECTS decay and perfusion estimated infarct growth rate: insights from a large national thrombectomy registry

Introduction Infarct growth rate (IGR) is highly heterogeneous among ischemic stroke patients, reflecting a spectrum of progressor phenotypes with clinical implications. We aim to compare different imaging approaches to investigate stroke progressors phenotypes and their clinical implications in patients undergoing thrombectomy. Methods Data are from the prospective Italian Registry of Endovascular Treatment in Acute Stroke (IRETAS). Patients with M1/M2 occlusion and known symptom onset were included. Progressor phenotypes were defined using (1) NCCT-based definitions (ASPECTS points decay per hour < 0.25 pts/h=slow progressor, 0.25–0.50 pts/h=intermediate, and > 0.50 pts/h=fast); and (2) CTP-based definitions (CTP-estimated core divided by time of onset < 5 mL/h=slow progressors, 5–10mL/h=intermediate, and > 10 mL/h=fast). The primary outcome was 90-day good functional outcome (modified Rankin Scale [mRS] = 0–2). Associations were assessed with logistic regression analyses adjusted for age, sex, NIHSS, TICI score, thrombolysis, and imaging-to-recanalization time. Results Of 26799 patients screened, 8322 (31.1%) were included (NCCT group: 8076; CTP group: 897 patients). NCCT-based progressor phenotype was associated with lower odds of good outcome (aOR 0.82 [95%CI = 0.72–0.92] per each progressor phenotype increase). ASPECTS decay per hour was associated with lower odds of good outcome (acOR 0.94 [95%CI = 0.89–0.99]). No significant association was observed for either CTP-based progressor phenotype or CTP-based IGR (mL/h). Similar findings were observed for secondary outcomes. Conclusions In this large, real-world cohort of stroke patients, NCCT-based IGR was associated with functional outcomes, whereas CTP-based IGR was not. This highlights the need to refine and identify more accurate markers of infarct growth within perfusion imaging.