前循环动脉瘤夹闭术后迟发性脑缺血情况及影响因素分析  

Analysis of delayed cerebral ischemia and influencing factors after clipping surgery in anterior circulation aneurysm patients

作　　者：吕光淘 孙凌云[1] 黄小龙 黄浩 张晓强[1] LÜGuangtao;SUN Lingyun;HUANG Xiaolong(Department of Neurosurgery,Guang'an People's Hospital,Guang'an 638000,China)

机构地区：[1]广安市人民医院神经外科,638000

出　　处：《临床神经病学杂志》2024年第5期366-371,共6页Journal of Clinical Neurology

摘　　要：目的探究前循环动脉瘤患者行夹闭术后发生迟发性脑缺血(DCI)情况及其影响因素。方法回顾性分析2018年5月至2023年5月本院采用夹闭术方式救治的前循环动脉瘤患者248例,根据术后DCI的发生与否分为DCI组和非DCI组,采用1∶1倾向性评分方法匹配后DCI和非DCI组各54例。采用随机森林算法对DCI影响因素进行重要性排序,多因素Logistic回归模型分析DCI的独立危险因素,E值法分析模型参数的敏感性。构建DCI风险预测模型,采用ROC曲线、Hosmer-Lemeshow检验评价模型的区分度和准确性。结果匹配后两组患者在性别、年龄、身体质量指数(BMI)、吸烟史、饮酒史、脑梗死病史、蛛网膜下腔出血史、动脉瘤位置、动脉瘤最大直径方面差异均无统计学意义(均P>0.05)。与非DCI组患者相比,DCI患者改良Fisher 3~4级、Hunt-HessⅣ级患者比率更高,D二聚体(D-D)、总胆红素(TBIL)、TNF-α、超敏C反应蛋白(hs-CRP)、IL-6、谷氨酸水平更高,免疫球蛋白(Ig)G、IgM、IgA水平更低(均P<0.05)。随机森林算法分析结果显示将Hunt-Hess分级、改良Fisher分级、D-D、谷氨酸、TBIL、IgA、IL-6纳入多因素回归分析获得的袋外数据错误率最低。多因素分析发现,Hunt-HessⅣ级、改良Fisher 3~4级、D-D、谷氨酸、TBIL、IL-6为DCI的独立危险因素,IgA为保护因素(均P<0.05)。风险预测方程:P=1/[1+e^((3.142+1.307 Hunt-Hess分级+1.086改良Fisher分级+0.976 D-D+0.748)谷氨酸+0.552 TBIL-0.524 IgA+0.487 IL-6)],当P>0.65时认为该患者术后发生DCI。ROC曲线结果显示,模型经内部验证后,AUC为0.869(95%CI:0.836~0.897),敏感度和特异度分别为81.08%和97.34%,模型具有良好的区分度。Hosmer-Lemeshow检验结果显示,模型经内部验证后拟合优度为χ^(2)=4.219,P=0.561,模型具有良好的拟合优度。结论Hunt-HessⅣ级、改良Fisher 3~4级、D-D、谷氨酸、TBIL、IL-6为DCI的独立危险因素,IgA为保护因素。由以上因素构建的Objective To explore the occurrence and influencing factors of delayed cerebral ischemia(DCI)after clipping surgery in patients with anterior circulation aneurysms.Methods A retrospective analysis was conducted on 248 patients with anterior circulation aneurysms who underwent clipping surgery in our hospital from May 2018 to May 2023.The patients were divided into the DCI group and the non-DCI group based on the occurrence of DCI after surgery.A 1∶1 propensity score matching method was used to match 54 patients in the DCI group with 54 patients in the non-DCI group.The random forest algorithm was employed to rank the importance of factors influencing DCI.Multivariable Logistic regression analysis was performed to identify independent risk factors for DCI,and the E-value method was used to assess the sensitivity of model parameters.A DCI risk prediction model was constructed,and the discriminative ability and accuracy of the model were evaluated using ROC curve and Hosmer-Lemeshow test.Results After matching,no statistically significant differences were observed between the two groups in terms of gender,age,body mass index(BMI),smoking history,alcohol consumption,history of cerebral infarction,history of subarachnoid hemorrhage,aneurysm location,and maximum aneurysm diameter(all P>0.05).Compared to the non-DCI group,the DCI group had a higher proportion of modified Fisher grade 3-4 and Hunt-Hess grade IV patients,higher levels of D-dimer(D-D),total bilirubin(TBIL),TNF-α,high-sensitivity C-reactive protein(hs-CRP),IL-6,and glutamate,and lower levels of immunoglobulin(Ig)G,IgM,and IgA(all P<0.05).The random forest analysis showed that including Hunt-Hess grade,modified Fisher grade,D-D,glutamate,TBIL,IgA,and IL-6 in the multivariable regression analysis achieved the lowest out-of-bag error rate.Multivariable analysis revealed that Hunt-Hess grade IV,modified Fisher grade 3-4,D-D,glutamate,TBIL were independent risk factors for DCI,while IgA was a protective factor(all P<0.05).The risk prediction equation was as f

关 键 词：前循环动脉瘤 夹闭术 迟发性脑缺血 蛛网膜下腔出血 免疫 

分 类 号：R743.3[医药卫生—神经病学与精神病学]