Establishment and validation of a predictive model for early cognitive dysfunction in ischaemic stroke
10.3969/j.issn.1004-1648.2025.03.002
- VernacularTitle:缺血性脑卒中早期认知功能障碍预测模型的建立与验证
- Author:
Min CHEN
1
;
Bing HUANG
;
Youjin CUI
Author Information
1. 230000 合肥,安徽医科大学第二附属医院神经重症监护室
- Publication Type:Journal Article
- Keywords:
ischaemic stroke;
early cognitive dysfunction;
multifactorial analysis;
nomograms
- From:
Journal of Clinical Neurology
2025;38(3):167-173
- CountryChina
- Language:Chinese
-
Abstract:
Objective To develop and validate a predictive model for early cognitive dysfunction in ischaemic stroke.Methods A total of 327 ischaemic stroke patients seen in our hospital from March 2021 to September 2023 were selected for the study.The patients were divided into 229 cases in the model group and 98 cases in the validation group in a ratio of 7∶3.The general conditions,disease-related factors and laboratory indicators of the two groups were compared,and a multifactorial Logistic regression was performed after screening potential variables by LASSO regression,and a nomograms model was established and validated.Results In the model group,229 patients had a Montreal Cognitive Assessment(MoCA)score of 19-30,and the mean score was(26.08±3.11)at 3 months after discharge,of which a total of 62(27.07％)had cognitive dysfunction.The differences between the cognitive dysfunction group and the normal group of patients in terms of age,smoking history,hypertension disease,diabetes,atrial fibrillation,coronary artery disease,NIHSS,infarct area,admission MoCA score,cerebral leukoaraiosis,cerebral atrophy,plasma cystatin C,C-reactive protein,and serum amyloid A were all statistically significant(all P＜0.05).The results of multifactorial Logistic regression analysis performed on the basis of LASSO regression showed that hypertension disease(OR=2.79,95％CI:1.03-7.59),diabetes mellitus(OR=3.51,95％CI:1.39-8.84),atrial fibrillation(OR=4.68,95％CI:1.41-15.50),NIHSS(OR=1.07,95％CI:1.03-1.11),cerebral leukoaraiosis(OR=5.82,95％CI:2.46-13.78),plasma cystatin(OR=1.12,95％CI:1.05-1.19)and serum amyloid A(OR=1.11,95％CI:1.07-1.16)were independent influences on cognitive dysfunction in early stage of ischemic stroke(all P＜0.05).The results of model validation showed the following:differentiation:the area under the ROC curve in the model group was 0.884(95％CI:0.832-0.935);the area under the ROC curve in the validation group was 0.822(95％CI:0.761-0.884).Accuracy:the slopes of the calibration curves of the model group and the validation group were 1,the intercept was 0.000,the model curve was basically fitted to the ideal model as a diagonal line,and the H-L fit goodness-of-fit test was performed(P＞0.05).The results of the clinical validity analysis showed the highest net benefit in predicting early cognitive dysfunction in ischaemic stroke using this study's model when the predictive probability threshold was 0.15-1.00.Conclusion Early cognitive dysfunction in ischaemic stroke is influenced by hypertension disease,diabetes mellitus,atrial fibrillation,and the column-line diagram model developed in this study can be used to predict early cognitive dysfunction in ischaemic stroke.
