Middle cerebral artery embolism can be predicted by CT perfusion

Abstract Background Identifying the etiology of acute middle cerebral artery occlusion (MCAO) is important in endovascular treatment. This study aimed to explore the predictive value of computed tomography perfusion (CTP) for determining the etiology of acute MCAO. Methods Clinical data and CTP images of patients with acute M1 segment MCAO who had undergone endovascular treatment at Linyi People’s Hospital from January 2020 to December 2023 were retrospectively analyzed. Patients were categorized based on the location of core infarction on CTP as cortical, subcortical, or mixed infarction. The etiology of MCAO was determined as either intracranial atherosclerotic stenosis (ICAS) or embolism via digital subtraction angiography (DSA). Results A total of 360 patients were included (median age 66 years, 63.9% male), of whom 241 had experienced embolism and 119 ICAS. Compared with the embolism group, the ICAS group was younger (p = 0.009), had a higher percentage of hyperlipidemia (p < 0.001), lower rates of coronary heart disease (p = 0.007) and atrial fibrillation (p < 0.001), a longer duration of endovascular treatment (p< 0.001), lower NIHSS scores (p< 0.001), smaller core volumes (p< 0.001), and better collateral compensation (p < 0.001). The two groups also differed in infarction location, with the embolism group primarily showing cortical infarctions (88%) and the ICAS group primarily exhibiting subcortical infarctions (69.7%). Multivariate analysis revealed that atrial fibrillation and cortical infarction were independent predictors of middle cerebral artery embolism (odds ratio = 36.496, 95% confidence interval: 10.732–124.106). Cortical infarction predicted middle cerebral artery embolism with a sensitivity of 88%, specificity of 86.6%, and accuracy of 87.5%. Conclusion The location of core infarction on baseline CTP can aid in identifying the etiology of acute MCAO. Cortical infarction is an independent predictor for middle cerebral artery embolism.