急性缺血性脑卒中静脉溶栓患者预后不良的影响因素及其预测模型的构建

Influencing factors of poor prognosis in patients with acute ischemic stroke undergoing intravenous thrombolysis and the construction of the prognostic prediction model

目的筛选与临床紧密相关的独立影响因素,并构建急性缺血性脑卒中(AIS)静脉溶栓患者预后不良的预测模型。方法回顾性分析2020年10月至2022年12月于东莞市滨海湾中心医院行静脉溶栓治疗的227例AIS患者的临床资料,根据治疗90 d改良Rankin量表(mRS)评分结果分为预后良好组150例(mRS 0~1分)和预后不良组77例(mRS 2~6分)。比较两组患者的临床资料,采用单因素和二元Logistic回归分析筛选影响因素,构建AIS静脉溶栓患者预后不良的列线图预测模型,绘制受试者工作特征(ROC)曲线、校准曲线,应用决策曲线评价其预测效能、校准度和临床效能。结果预后不良组患者的年龄、溶栓前NIHSS评分、溶栓后NIHSS评分、中性粒细胞与淋巴细胞比值(NLR)、国际标准化比值(INR)分别为58(52,72.50)岁、11(7,17)分、8(5,13)分、2.34(1.58,4.06)、0.92(0.87,0.96),明显高于预后良好组的55(47,62.25)岁、5(3,8)分、2(1,5)分、2.24(1.59,3.83)、0.88(0.84,0.93),淋巴细胞绝对值为1.90(1.28,2.60)×10~9,明显低于预后良好组的2.06(1.48,2.67)×10~9,差异均有统计学意义(P<0.05)。预后不良组患者的机械取栓率、早期神经功能恶化(END)发生率、症状性颅内出血(sICH)发生率、既往冠心病史比率、房颤病史比率、大动脉粥样硬化型(LAA)比率分别为20.8%、51.9%、6.5%、16.9%、24.7%、50.6%,明显高于预后良好组的9.3%、14.0%、0.7%、4.7%、8.7%、34.0%,差异均有统计学意义(P<0.05)。Logistic回归分析结果显示,溶栓前NIHSS评分、溶栓后NIHSS评分、机械取栓、END、LAA均是AIS静脉溶栓患者预后不良的独立影响因素(P<0.05),依其构建预测模型列线图,其曲线下面积为0.877,校准曲线图显示患者实际预后和预测预后具有良好的一致性,绝对误差为0.064;应用决策曲线显示在阈值5.2%~93.0%之内有较高的临床适用性,可带来临床净收益。结论基于溶栓前NIHSS评分、溶栓后NIHSS评分、机械取栓、END、LAA绘制的预测模型列线图能较好预测AIS静脉溶栓患者的预后,可为患者带来临床效益。

Objective To screen independent influencing factors closely related to clinical practice and construct a prediction model for poor prognosis in patients with acute ischemic stroke(AIS)undergoing intravenous thrombolysis.Methods The clinical data of 227 AIS patients undergoing intravenous thrombolytic therapy from October 2020 to December 2022 at Binhaiwan Central Hospital of Dongguan was collected for a retrospective study.The patients were divided into a good prognosis group(Modified Rankin Scale[mRS]score 0-1,n=150)and a poor prognosis group(mRS score 2-6,n=77)according to the mRS of 90 days after treatments.The clinical data of the two groups were compared.Univariate and binary logistic regression analysis was used to screen the influencing factors,and the nomogram prediction model for the poor prognosis of AIS patients undergoing intravenous thrombolysis was constructed.The receiver operating characteristic(ROC)curve,calibration curve,and decision curve were drawn to evaluate their predictive efficacy,consistency and clinical efficacy of the model.Results The age,NIHSS score before thrombolysis,NIHSS score after thrombolysis,neutrophil-to-lymphocyte ratio(NLR),and international normalized ratio(INR)of the patients in the poor prognosis group were 58(52,72.50)years,11(7,17)points,8(5,13)points,2.34(1.58,4.06),and 0.92(0.87,0.96),respectively,which were all significantly higher than 55(47,62.25)years,5(3,8)points,2(1,5)points,2.24(1.59,3.83),and 0.88(0.84,0.93)of the good prognosis group(P<0.05).The absolute value of lymphocytes in the poor prognosis group was 1.90(1.28,2.60)×109,which was significantly lower than 2.06(1.48,2.67)×109 in the good prognosis group(P<0.05).The rates of mechanical thrombectomy,early neurological deterioration(END),symptomatic intracranial hemorrhage(sICH),patients with history of coronary heart disease,patients with history of atrial fibrillation,and large artery atherosclerosis(LAA)in the poor prognosis group were 20.8%,51.9%,6.5%,16.9%,24.7%,and 50.6%,respectively,which were significantly higher than 9.3%,14.0%,0.7%,4.7%,8.7%,and 34.0%in the good prognosis group(P<0.05).Logistic regression analysis showed that NIHSS score before thrombolysis,NIHSS score after thrombolysis,mechanical thrombectomy,END,and LAA were all independent risk factors for poor prognosis in AIS patients after intravenous thrombolysis(P<0.05).Based on the logistic regression results,a nomogram prediction model was constructed with the area under the curve of 0.877(95%CI,0.832-0.922).The calibration curve showed excellent agreement between the actual and predicted prognosis of the patients,with an absolute error of 0.064.The decision curve showed high clinical applicability within a threshold of 5.2%to 93.0%,resulting in a net clinical benefit.Conclusion The nomogram prediction model based on NIHSS score before thrombolysis,NIHSS score after thrombolysis,mechanical thrombectomy,END,and LAA can better predict the prognosis of AIS patients undergoing intravenous thrombolysis and bring clinical benefit.