论文部分内容阅读
目的:探讨血浆低密度脂蛋白胆固醇/高密度脂蛋白胆固醇比值（LHR）和同型半胱氨酸（Hcy）对饮酒所致脑梗死的诊断价值。方法:选取2020年1月至2021年1月在阜阳市人民医院神经内科就诊的420例脑梗死患者，根据有无饮酒史，分为饮酒组（236例）和非饮酒组（184例）。收集患者一般基线资料，发病24 h内血浆高密度脂蛋白胆固醇（HDL-C）、低密度脂蛋白胆固醇（LDL-C）和Hcy等，并比较两组脑梗死患者上述资料的差异。利用二元logistic回归分析饮酒导致脑梗死发生的危险因素，用受试者工作特征曲线（ROC）评估血浆LHR和Hcy对饮酒脑梗死的诊断准确度。结果:饮酒组的男性、吸烟、高血压、糖尿病比例、入院时的美国国立卫生研究院卒中量表（NIHSS）评分、LDL-C、LHR和Hcy水平均高于非饮酒组，差异均有统计学意义（均n P＜0.05）；多因素logistic回归模型分析显示，性别、吸烟、NIHSS评分、LHR以及高同型半胱氨酸血症均是饮酒导致脑梗死发生的危险因素（均n P＜0.05）；ROC分析示，LHR判断饮酒所致脑梗死患者的曲线下面积为0.700（95%n CI 0.649～0.750），取LHR临界值2.34时，灵敏度72.9%、特异度65.2%；高同型半胱氨酸血症判断饮酒导致脑梗死患者的曲线下面积为0.794（95%n CI 0.751～0.837），取Hcy临界值14.46 μmol/L时，灵敏度72.9%、特异度80.4%；与血浆LHR和高同型半胱氨酸单独检测相比，LHR和Hcy联合检测诊断饮酒合并脑梗死准确度更高（AUC=0.835，95%n CI 0.796～0.874，n P＜0.001），灵敏度为92.1%，特异度为89.6%。n 结论:血浆LHR及Hcy检测诊断饮酒所致脑梗死有较高准确度，二者联合检测准确性更高。“,”Objective:To explore the value of plasma low-density lipoprotein cholesterol/high-density lipoprotein cholesterol ratio (LDL-C/HDL-C ratio, LHR) and homocysteine in the diagnosis of cerebral infarction due to alcohol consumption.Methods:A total of 420 patients with cerebral infarction treated at Department of Neurology, Fuyang People\'s Hospital from January 2020 to January 2021 were divided into a drinking group (236 cases) and a non-drinking group (184 cases) according to whether they had a history of drinking. The patients\' general baseline data and plasma HDL-C, LDL-C, and homocysteine within 24 hours of onset were collected and compared between the two groups. Binary logistic regression was used to analyze the risk factors of alcohol-drinking cerebral infarction, and receiver operating characteristic curve (ROC) was used to evaluate the accuracy of plasma LHR and homocysteine in the diagnosis of alcohol-drinking cerebral infarction.Results:The proportions of males, smoking, hypertension, and diabetes, score of National Institutes of Health Stroke Scale, LDL-C level, LHR, and Hcy level in the drinking group at admission were higher than those in the non-drinking group, with statistical differences (all n P＜0.05). The multivariate logistic regression analysis showed that gender, smoking, NIHSS score, LHR, and hyperhomocysteinemia were the risk factors for the occurrence of alcohol-drinking cerebral infarction. The ROC analysis showed that the area under the curve for LHR to judge alcohol-drinking cerebral infarction was 0.700 [95% confidence interval (n CI): 0.649-0.750), and the LHR cut-off value was taken at 2.34, with a sensitivity of 72.9% and a specificity of 65.2%. The area under the curve for hyperhomocysteinemia to judge alcohol-drinking cerebral infarction was 0.794 (95%n CI: 0.751-0.837), and the Hcy cut-off value was 14.46 μmol/L, with a sensitivity of 72.9% and a specificity of 80.4%. Compared with the single detection of plasma LHR and high homocysteine, the combined detection of LHR and homocysteine had a higher accuracy in the diagnosis of alcohol-drinking cerebral infarction (AUC 0.835, 95%n CI: 0.796-0.874, n P＜0.001), with a sensitivity of 92.1% and a specificity of 89.6%.n Conclusion:Plasma LHR and Hcy have high accuracies in diagnosing alcohol-drinking cerebral infarction, and the combined detection of the two has a higher accuracy.