缺血性脑卒中患者住院期间误吸的危险因素及预测模型构建

Risk factors of aspiration during hospitalization in patients with ischemic stroke and establishment of a predictive model

目的 探讨缺血性脑卒中(IS)患者住院期间误吸的危险因素并构建预测模型。方法 采用病例对照设计,回顾性收集2022年3月—2023年10月在苏州大学附属苏州九院神经内科治疗的316例IS患者的临床资料。依据患者住院期间误吸情况分为病例组(住院期间发生误吸)89例和对照组(住院期间未发生误吸)227例。对2组资料进行单因素和多因素Logistic回归分析,筛选出IS患者住院期间误吸的危险因素。采用R软件从2组中抽取70%数据作为训练集(建立列线图模型),剩余30%数据作为测试集。采用受试者工作特征(ROC)曲线的曲线下面积(AUC)、校准曲线、决策曲线评估预测模型的价值。结果 病例组与对照组在年龄、美国国立卫生研究院卒中量表(NIHSS)评分、病灶数、同型半胱氨酸(Hcy)水平、自主咳嗽、洼田饮水试验分级等方面比较,差异有统计学意义(P<0.05)。多因素Logistic回归分析结果表明,年龄大(OR=2.201, 95%CI:1.254~3.865)、NIHSS评分高(OR=4.816, 95%CI:1.652~14.041)、多个病灶(OR=2.649, 95%CI:1.249~5.613)、Hcy高水平(OR=1.501, 95%CI:1.044~2.158)、自主咳嗽减弱或消失(OR=3.384, 95%CI:1.639~6.987)、洼田饮水试验分级高(OR=2.878, 95%CI:1.422~5.783)是IS患者住院期间误吸的危险因素(P<0.05)。训练集中,列线图模型预测IS患者住院期间误吸的AUC为0.872(95%CI:0.827~0.919),在测试集中验证得到的AUC为0.859(95%CI:0.807~0.904)。校准曲线分析训练集的P=0.869,测试集的P=0.898。训练集和测试集的决策曲线分析均显示模型有良好的临床适用性。结论 IS患者住院期间误吸风险与年龄大、NIHSS评分高、多个病灶、Hcy高水平、自主咳嗽减弱或消失、洼田饮水试验分级高有关。基于此构建的列线图预测模型可有效评估IS患者住院期间的误吸风险。

Objective To explore the risk factors of aspiration during hospitalization in patients with ischemic stroke(IS)and establish a predictive model.Methods Based on the case-control design,clinical materials of 316 IS patients treated in the Department of Neurology of Suzhou Ninth Hospital Affiliated to Soochow University from March 2022 to October 2023 were retrospectively collected.According to incidence of aspiration during hospitalization,the patients were divided into case group with 89 cases(aspiration occurred during hospitalization)and control group with 227 cases(no aspiration occurred during hospitalization).Univariate and multivariate Logistic regression analyses were performed in both groups to screen out the risk factors of aspiration during hospitalization in IS patients.R software was used to extract 70%of the data from the two groups as the training set(establishing a Nomogram model),and the remaining 30%data was used as test set.Value of predictive model was evaluated by area under the curve(AUC)of the receiver operating characteristic(ROC)curve,calibration curve,and decision curve.Results There were significant differences in the terms of age,the National Institutes of Health Stroke Scale(NIHSS)score,number of lesions,homocysteine(Hcy)level,spontaneous cough,and grading of Wada drinking water test between the case group and the control group(P<0.05).Multivariate Logistic analysis showed that large age(OR=2.201,95%CI,1.254 to 3.865),high NIHSS score(OR=4.816,95%CI,1.652 to 14.041),multiple lesions(OR=2.649,95%CI,1.249 to 5.613),high level of Hcy(OR=1.501,95%CI,1.044 to 2.158),weakened or absent spontaneous cough(OR=3.384,95%CI,1.639 to 6.987),and high grading of Wada drinking water test(OR=2.878,95%CI,1.422 to 5.783)were the risk factors of aspiration during hospitalization in IS patients(P<0.05).In the training set,the AUC of the Nomogram model for predicting aspiration during hospitalization in IS patients was 0.872(95%CI,0.827 to 0.919),and was 0.859(95%CI,0.807 to 0.904)in the test set.The calibration curve analysis showed that P value was 0.869 in the training set and 0.898 in the test set.The decision curve analysis for both the training set and test set showed that the Nomogram model was of good clinical applicability.Conclusion The risk of aspiration during hospitalization in IS patients is related to large age,high NIHSS score,multiple lesions,high level of Hcy,weakened or absent spontaneous cough,and high grading of Wada drinking water test.The Nomogram predictive model established on these factors can effectively evaluate the risk of aspiration during hospitalization in IS patients.