Supplementary Material for: Infarct Pattern and Clinical Outcome in Acute Ischemic Stroke Following Middle Cerebral Artery Occlusion

<b><i>Background:</i></b> Cerebral arterial occlusion develops via two distinct mechanisms: thrombosis and embolism. Discrimination between thrombosis and embolism is an important aspect needed for further determining the etiology of stroke in a patient. This study evaluated infarct patterns and outcomes in acute stroke patients with relevant artery occlusions, focusing on features specific to each occlusion mechanism. <b><i>Methods:</i></b> Acute ischemic stroke patients who were consecutively registered in a tertiary hospital between 2002 and 2010 with infarctions in the middle cerebral artery territory and a corresponding M1 occlusion confirmed by magnetic resonance angiography, computed tomography angiography, or conventional angiography were enrolled. Patients with a high-risk cardioembolic source, clear recanalization, concurrent infarct in an arterial territory other than the occlusion site, or no prior occlusion in a previous imaging within 1 month were assigned to the embolic occlusion group, and the remaining patients were assigned to the thrombotic occlusion group. The infarct pattern was categorized into seven groups: scattered, territorial, lenticulostriatal, scattered-territorial, scattered-lenticulostriatal, territorial-lenticulostriatal, and scattered-territorial-lenticulostriatal. Data of stroke recurrence and mortality were collected through electronic medical record and the National Vital Statistics System. <b><i>Results:</i></b> Of 114 patients, 54 (47.4%) were classified as having an embolic occlusion. When infarct patterns were compared between the groups, any-scattered infarct pattern was more common in the thrombotic occlusion group (71.2% vs. 40.7%, p = 0.002), and any-territorial infarct pattern was more prevalent in the embolic occlusion group (55.6% vs. 28.8%, p = 0.005). In addition, scattered-without-territorial pattern was higher in the thrombotic occlusion group (OR: 0.25; CI: 0.11-0.57; p = 0.001). Any-territorial infarct pattern was also related to initial stroke severity (NIHSS on admission, OR: 400.98; CI: 2.94-54,741.32; p = 0.017) and poor functional outcome (modified Rankin Scale score ≥4) at discharge (OR: 14.40; CI: 1.37-152.00; p = 0.027) independent of other parameters. However, no association was found between stroke recurrence, mortality and occlusion mechanism. <b><i>Conclusion:</i></b> This study shows that specific infarct patterns are related to cerebral arterial occlusion mechanisms and are correlated with functional outcome. Otherwise, the results of our study indicates that infarct patterns on DWI might be a clue for determining ischemic stroke etiology on patients with major cerebral artery occlusion