轻度急性缺血性卒中早期神经功能恶化的风险因素及列线图预测模型的构建

A nomogram of early neurological deterioration in patients with acute minor ischemic stroke

目的:评估轻度急性缺血性卒中患者起病后早期神经功能恶化(early neurological deterioration,END)的风险因素,构建起病后END的列线图预测模型。方法:回顾性收集南京医科大学附属南京医院(南京市第一医院)及南通市第三人民医院2015年4月—2018年6月收治的轻度急性缺血性卒中患者临床资料,比较END组与非END组患者的基线临床资料,采用多因素Logistic回归分析确定END的独立相关因素,基于独立相关因素构建轻度急性缺血性卒中END列线图预测模型。结果:共纳入507例患者,END组99例(19.5%),非END组408例(80.5%)。END组患者的年龄(P=0.001)、心房纤颤病史构成比(P=0.001)、缺血性心脏病病史构成比(P=0.010)、基线美国国立卫生研究院卒中量表(National Institute of Health stroke scale,NIHSS)评分(P=0.023)、快速血糖水平(P=0.001)、超敏C反应蛋白水平(P=0.006)显著高于非END组,饮酒史构成比(P=0.042)、白蛋白水平(P=0.001)显著低于非END组。多因素Logistic回归分析提示,年龄(OR=1.031,95%CI:1.008~1.054;P=0.007)、心房纤颤病史(OR=4.349,95%CI:1.932~9.792;P=0.001)、基线NIHSS评分(OR=1.219,95%CI:1.021~1.455;P=0.029)、快速血糖水平(OR=1.199,95%CI:1.083~1.328;P=0.001)、超敏C反应蛋白水平(OR=1.069,95%CI:1.027~1.113;P=0.001)、白蛋白水平(OR=0.826,95%CI:0.733~0.930;P=0.002)是轻型急性缺血性卒中END的独立相关因素。基于多因素Logistic回归发现的独立相关因素,构建列线图模型,一致性指数为0.736(95%CI:0.677~0.796,P <0.001)。结论:本列线图对轻型急性缺血性卒中后END的发生具有一定的预测价值。

Objective:To evaluate the risk factors of early neurological deterioration(END)in patients with acute minor ischemic stroke,and to construct a nomogram model of END. Methods:From April 2015 to June 2018,the clinical data of the patients with acute minor ischemic stroke in Nanjing First Hospital and Nantong Third People’s Hospital were prospectively collected.Demographics and baseline clinical data were compared between the END group and the non-END group. We used the multivariate logistic regression analysis to determine the independent risk factors for END. Based on these independent factors,we constructed the nomogram of END in patients with acute minor ischemic stroke. Results:A total of 507 patients were enrolled in the study. The age(P=0.001),atrial fibrillation(P=0.001),ischemic heart disease history(P=0.010),baseline NIHSS(P=0.023),fasting blood glucose levels(P=0.001)and hypersensitivity C-reactive protein levels(P=0.006)in the END group were significantly higher than those in the nonEND group,and drinking history(P=0.042)as well as the albumin levels(P=0.001)were significantly lower than those in the non-END group. Multivariate logistic regression analysis showed that age[odds ratio(OR)1.031,95% confidence interval(95% CI)1.008~1.054;P=0.007],atrial fibrillation(OR=4.349,95% CI:1.932 ~9.792;P=0.001),baseline NIHSS(OR=1.219,95% CI:1.021~1.455;P=0.029),fasting blood glucose(OR=1.199,95%CI:1.083~1.328;P=0.001),high sensitivity C-reactive protein(OR=1.069,95%CI:1.027~1.113;P=0.001),albumin(OR=0.826,95%CI:0.733~0.930;P=0.002)were the independent risk factors for END. Based on the independent risk factors,a nomogram model was constructed,and the consistency index was 0.736(95%CI:0.677~0.796,P < 0.001).Conclusion:This nomogram has a certain predictive value for END in patients with mild acute ischemic stroke.