2023 年11 期 第31 卷
脑卒中专题研究![查看全文 查看全文](/Areas/201911120101/Content/img/icon_html.png)
![PDF下载 PDF下载](/Areas/201911120101/Content/img/icon_pdf.png)
基于 CT 灌注成像参数构建急性脑梗死患者溶栓治疗后发生出血转化风险预测列线图模型
Construction of Nomograph Model for Predicting the Risk of Hemorrhagic Transformation after ThrombolyticTherapy in Patients with Acute Cerebral Infarction Based on Parameters of CT Perfusion Imaging
作者:雷爱春,储燕,杨雷,李其元
- 单位:
- 1.223200江苏省淮安市淮安医院影像科 2.223200江苏省淮安市淮安医院神经内科 3.223200江苏省淮安市淮安医院超声科
- 单位(英文):
- 1.Department of Imaging, Huai'an Huai'an Hospital, Huaian 223200, China2.Department of Neurology, Huai'an Huai'an Hospital, Huaian 223200, China3.Department of Ultrasound, Huai'an Huai'an Hospital, Huaian 223200, China
- 关键词:
- 脑梗死;出血转化;溶栓治疗;CT灌注成像;列线图
- 关键词(英文):
- Brain infarction; Hemorrhagic transformation; Thrombolytic therapy; CT perfusion imaging; Nomograms
- 中图分类号:
- R 743.33
- DOI:
- 10.12114/j.issn.1008-5971.2023.00.294
- 基金项目:
- 江苏省自然科学基金资助项目(20KJB520008)
摘要:
目的 基于CT灌注成像参数构建急性脑梗死(ACI)患者溶栓治疗后发生出血转化风险预测列线图模型。方法 选取2019年10月至2023年1月淮安市淮安医院收治的263例ACI患者为研究对象,根据溶栓治疗后是否发生出血转化将其分为无出血转化组(n=207)和出血转化组(n=56)。收集患者临床资料和CT灌注成像参数,采用单因素分析和多因素Logistic回归分析探讨ACI患者溶栓治疗后发生出血转化的影响因素;采用R语言中的“rms”程序包构建ACI患者溶栓治疗后发生出血转化风险预测列线图模型,并绘制ROC曲线以评估该列线图模型预测ACI患者溶栓治疗后发生出血转化的区分度,采用Bootstrap法及校准曲线验证该列线图模型预测ACI患者溶栓治疗后发生出血转化的准确性。结果 无出血转化组与出血转化组脑梗死面积、达峰时间(TTP)、血流峰值时间(Tmax)、表面渗透性(PS)、相对表面渗透性(rPS)比较,差异有统计学意义(P<0.05)。多因素Logistic回归分析结果显示,脑梗死面积、TTP、Tmax、PS、rPS是ACI患者溶栓治疗后发生出血转化的独立影响因素(P<0.05)。基于多因素Logistic回归分析结果,构建ACI患者溶栓治疗后发生出血转化风险预测列线图模型。ROC曲线分析结果显示,该列线图模型预测ACI患者溶栓治疗后发生出血转化的AUC为0.958〔95%CI(0.929,0.987)〕,最佳截断值为0.275,灵敏度为92.97%,特异度为89.98%;采用Bootstrap法将原始数据重复抽样100次进行内部验证,C指数为0.829;校准曲线分析结果显示,该列线图模型预测ACI患者溶栓治疗后出血转化发生率与实际发生率接近。结论 大面积脑梗死、TTP、Tmax、PS、rPS是ACI患者溶栓治疗后发生出血转化的独立影响因素,而基于上述影响因素构建的ACI患者溶栓治疗后发生出血转化风险预测列线图模型具有较高的区分度和准确性。
英文摘要:
Objective To construct a nomograph model for predicting the risk of hemorrhagic transformation afterthrombolytic therapy in patients with acute cerebral infarction (ACI) based on parameters of CT perfusion imaging. MethodsA total of 263 patients with ACI admitted to Huai'an Huai'an Hospital from October 2019 to January 2023 were included asthe study subjects. According to whether there was hemorrhagic transformation after thrombolysis, 207 patients were includedinto non-hemorrhagic transformation group and 56 patients were included into hemorrhagic transformation group. The clinicaldata and CT perfusion imaging parameters of patients were collected. Univariate analysis and multivariate Logistic regressionanalysis were used to analyze the influencing factors of hemorrhagic transformation after thrombolytic therapy in ACI patients.The "ms" package in R language was used to construct a nomogram model for predicting the risk of hemorrhagic transformationafter thrombolytic therapy in ACI patients, and the ROC curve was drawn to evaluate the discrimination of the nomogram modelin predicting hemorrhagic transformation after thrombolytic therapy in ACI patients. Bootstrap method and calibration curve wereused to verify the accuracy of the nomogram model in predicting hemorrhagic transformation after thrombolytic therapy in ACI patients. Results There were significant differences in cerebral infarction area, time to peak (TTP) , time to maximum (Tmax) ,surface permeability (PS) and relative surface permeability (rPS) between the non-hemorrhagic transformation group and thehemorrhagic transformation group (P < 0.05) . Multivariate Logistic regression analysis showed that cerebral infarction area,TTP, Tmax, PS and rPS were independent influencing factors of hemorrhagic transformation after thrombolytic therapy in ACIpatients (P < 0.05) . Based on the results of multivariate Logistic regression analysis, a nomogram model for predicting the risk ofhemorrhagic transformation after thrombolytic therapy in ACI patients was constructed. The results of ROC curve analysis showedthat the AUC of the nomogram model for predicting hemorrhagic transformation after thrombolytic therapy in ACI patients was0.958 [95%CI (0.929, 0.987) ] , the optimal cut-off value was 0.275, the sensitivity was 92.97%, and the specificity was 89.98%.Bootstrap method was used to repeatedly sample the original data for 100 times for internal verification, and the C index was 0.829.The results of calibration curve analysis showed that the incidence of hemorrhagic transformation after thrombolytic therapy inACI patients predicted by the nomogram model was close to the actual incidence. Conclusion The cerebral infarction area, TTP,Tmax, PS and rPS are independent influencing factors of hemorrhagic transformation after thrombolytic therapy in ACI patients.The nomogram model for predicting the risk of hemorrhagic transformation after thrombolytic therapy in ACI patients constructedbased on the above influencing factors has high discrimination and accuracy.
参考文献: