【作者】 王琳；

【导师】 满玉红；

【作者基本信息】 吉林大学 ， 临床医学硕士（专业学位）， 2023， 硕士

【摘要】 背景:急性缺血性卒中是脑血管狭窄、闭塞和脑灌注减少导致的神经功能障碍疾病,大约占我国脑卒中的83%,是导致死亡和严重残疾的主要原因之一。出血转化是急性缺血性卒中后缺氧、炎症导致血脑屏障通透性增加,缺血中心区再灌注后血液外渗发生的继发性出血,是急性缺血性卒中患者的常见并发症之一。出血转化与急性缺血性卒中较高的死亡率、早期神经功能恶化和较差的预后相关,目前多数研究集中在溶栓后出血转化,本研究总结发生出血转化的非溶栓患者临床特征,筛查非溶栓患者出血转化的危险因素及可能的预测因子。目的:回顾性总结非溶栓治疗的急性缺血性卒中患者发生出血转化的可能危险因素,从临床特征、影像特点、血液指标三个角度进行分析和筛选具有高灵敏度和可靠性的特异性临床特征,以及影像和生物学标志物,为急性缺血性卒中出血转化的早期预警和早期防治提供依据。方法:回顾性分析2021年1月至2022年10月期间就诊于吉林大学第二医院急性缺血性卒中患者,按照已制定的纳入标准及排除标准确定入组病例,根据是否发生出血转化分为出血转化组和非出血转化组。收集每例患者一般临床资料(年龄、性别、高血压、糖尿病、房颤、吸烟史、饮酒史),实验室指标(血小板计数、中性粒细胞计数、淋巴细胞计数、NLR、D-2聚体、尿酸、肌酐、谷丙转氨酶、谷草转氨酶、血同型半胱氨酸、低密度脂蛋白、空腹血糖、FT3、FT4),影像学特点(大面积脑梗死、脑白质疏松程度),TOAST分型,出血转化组患者ECASS分型及出血部位。采用SPSS26.0软件进行数据分析,正态分布的计量资料用均数±标准差表示,组间比较使用t检验;不符合正态分布的计量资料采用中位数及百分位数表示,组间比较使用秩和检验;计数资料以频数、百分比(%)表示,组间比较使用卡方检验。对统计分析结果存在差异的变量进行多因素Logistic回归分析明确危险因素,根据回归结果建立预测模型,使用ROC曲线分析预测模型价值。结果:(1)出血转化组与非出血转化组在房颤病史、尿酸水平、低密度脂蛋白水平、FT3、FT4、大面积脑梗死及脑白质疏松症组间差异显著(P<0.05)。两组TOAST分型具有统计学差异,出血转化组在LAA型、PAD型、CE型、UE型均有分布;非出血转化组在LAA型和PAD型有分布,两组均为LAA型占比最多,两组OE型例数为0。(2)多因素Logistic回归分析显示高尿酸水平、低LDL、低FT3、低FT4、大面积脑梗死、脑白质疏松症为自发性出血转化危险因素(P<0.05),纳入上述因子进行预测模型验证,选约登指数最大时切点,曲线下面积为0.800。(3)自发性出血转化发生情况:本研究共纳入198例患者。出血转化组87例,其中HI1型29例,HI2型33例,PH1型13例,PH2型12例。HI型出血部位在额叶、颞叶、枕叶、顶叶、基底节、小脑均有分布,PH型在单独的额叶、颞叶、顶叶中例数为0。HI型主要分布部位为基底节区(27.4%)和枕叶(24.2%),PH型主要分布部位为基底节区(58.3%)。结论:(1)TOAST分型心源性脑梗死更易出现自发性出血转化。(2)高尿酸水平、低LDL、低FT3、低FT4、大面积脑梗死、脑白质疏松是自发性出血转化的独立危险因素。(3)自发性出血转化的ECASS分型中,HI型出血部位最常见于基底节区、枕叶,PH型出血部位最常见于基底节区,且HI型发生率高于PH型。更多还原

【Abstract】 Background:Acute ischemic stroke,a neurological dysfunction disease resulting from cerebral vessel stenosis,occlusion and reduced cerebral perfusion,accounts for approximately83% of strokes in China and is one of the leading causes of death and severe disability.Hemorrhagic transformation is one of the common complications in patients with acute ischemic stroke,which is caused by hypoxia and inflammation after acute ischemic stroke,leading to increased permeability of the blood-brain barrier,and blood exosmosis after reperfusion in the ischemic centre.Hemorrhagic transformation is associated with high mortality,deterioration of neurological function and poor prognosis of acute ischemic stroke.At present,most studies focus on the transformation of hemorrhage after thrombolysis.This study summarized the clinical characteristics of non-thrombolytic patients with hemorrhagic transformation,screened the risk factors and possible predictors of hemorrhagic transformation in non-thrombolytic patients.Purposes:This study analyzed and screened specific clinical features with high sensitivity and reliability from clinical features,imaging features,and blood indicators to provide basis for early warning and early prevention of hemorrhagic transformation in acute ischemic stroke.Methods:A retrospective analysis was made on the patients with acute ischemic stroke who hospitalized at the Second Hospital of Jilin University from January 2021 to October2022.According to the established inclusion criteria and exclusion criteria,we determined the enrolled cases and divided them into hemorrhagic transformation group and non-hemorrhagic transformation group.This study collected the general clinical data(age,gender,hypertension,diabetes,atrial fibrillation,smoking history,drinking history),laboratory indicators(platelet count,neutrophil count,lymphocyte count,NLR,D-2 polymer,uric acid,creatinine,alanine aminotransferase,glutamic oxaloacetic transaminase,blood homocysteine,low-density lipoprotein,fasting blood glucose,FT3,FT4)of each patient,imaging characteristics(massive cerebral infarction,leukoaraiosis),TOAST classification,ECASS classification and bleeding site of patients in hemorrhagic transformation group.We used SPSS26.0 software to analyze the data.The measurement data of normal distribution were expressed by mean± standard deviation,and the inter-group comparison was performed by t-test;The measurement data that do not conform to the normal distribution are expressed by the median and percentile,and the rank sum test is used for inter-group comparison;The counting data are expressed in frequency and percentage(%),and the chi-square test is used for comparison between groups.Multi-factor logistic regression analysis shall be carried out for the variables with different statistical analysis results to identify the risk factors,the prediction model shall be established according to the regression results,and the ROC curve shall be used to analyze the value of the prediction model.Results:(1)There were significant differences between the hemorrhagic transformation group and the non-hemorrhagic transformation group in the history of atrial fibrillation,uric acid level,low density lipoprotein level,FT3,FT4,massive cerebral infarction and leukoaraiosis(P<0.05).The TOAST classification of the two groups was statistically different,and the hemorrhagic transformation group showed distribution in LAA type,pad type,CE type,and UE type;The non-hemorrhagic transformation group had a distribution in LAA type and pad type,with the highest proportion of LAA type in both groups,and the number of OE type cases was 0 in both groups.(2)Multiple logistic regression analysis showed that high uric acid level,low LDL,low FT3,low FT4,massive cerebral infarction,and leukoaraiosis were risk factors for spontaneous hemorrhagic transformation(P < 0.05),and the inclusion of the above factors for predictive model validation selected the maximum time cut point for the Youden index with an area under the curve of 0.800.(3)A total of 198 patients were included in this study.The hemorrhagic transformation group included 87 patients,including 29 with HI1,33 with HI2,13 with PH1,and 12 with PH2.The location of HI type was in the frontal,temporal,occipital,parietal,basal ganglia,and cerebellum,and the number of PH type was 0 in separate frontal,temporal,and parietal lobes.The main distribution sites of HI type were basal ganglia(27.4%)and occipital lobe(24.2%),and that of PH type was basal ganglia(58.3%).Conclusions:(1)In the TOAST classification,cardiocerebral infarction was more likely to occur with spontaneous HT.(2)High uric acid level,low LDL,low FT3,low FT4,massive cerebral infarction,and leukoaraiosis were independent risk factors for spontaneous HT.(3)In the ECASS classification of spontaneous HT,the location of bleeding in the HI type was most commonly in the basal ganglia and occipital lobes,while the location of bleeding in the PH type was most commonly in the basal ganglia,and the incidence of HI type was higher than that in the PH type.更多还原