【摘要】 目的探究CT血管造影(CTA)和CT灌注(CTP)对急性缺血性卒中患者梗死体积的预测效果。方法选择该院就诊的急性前循环缺血性卒中患者,获取患者一般资料和随访影像检查结果,构建分别包含患者临床特征、平扫CT、CTA、CTP预测因子的4种不同预测模型,利用多因素Logistic回归分析其对脑梗死存在的预测效果,并利用受试者工作特征曲线进行比较,采用Tobit回归分析模型对脑梗死体积的预测效果,分析梗死体积与临床预后之间的关系。结果共纳入934例患者,632例患者(67.7%)证实发生脑梗死。包含全部预测因子的多因素Logistic回归模型4具有最好的梗死存在预测效果,对应的AUC值为0.902。包含全部预测因子的Tobit模型4对脑梗死体积的预测效果最佳,R~2为0.58。脑梗死体积较大的患者预后越差,差异具有统计学意义(P<0.05)。结论基于临床特征和NCCT的CTA与CTP联合预测模型对卒中患者梗死的发生及梗死体积的预测具有较好的效果,有利于及时制定诊疗方案。更多还原

【Abstract】 Objective To evaluate the predictive effect of CT angiography( CTA) and CT perfusion( CTP) on infarct volume among patients with acute ischemic stroke. Methods The clinical features and imaging findings were obtained from patients with an acute anterior circulation of ischemic stroke admitted into our hospital. Four different predictive models including clinical features,CT,CTA and CTP predictors were built. Multivariate logistic regression was used to analyze the predictive effect on cerebral infarction,and the results were compared by receiver operating characteristic curve. Tobit regression analysis model was used to predict the infarct volume and analyze,the relationship between infarct volume and clinical prognosis. Results A total of 934 patients were enrolled in this study,and 632 patients( 67. 7%) were diagnosed with cerebral infarction at follow-up. Logistic regression Model 4 including all the predictors had the best predictive effect on infarction presence,with a corresponding AUC value of 0. 902. Tobit model analysis showed that model 4 including all the predictors had the best predictive effect on cerebral infarct volume,whose R2 was 0. 58. The prognosis of the patients with larger cerebral infarction volume in the follow-up images was worse and the difference was statistically significant( P < 0. 05). Conclusion The CTA and CTP combined prediction model based on clinical features and noncontrast CT has a good predictive effect on infarction and infarct volume in patients with acute ischemic stroke,which is helpful to develop the treatment plan in time.更多还原