| Objective Studying the risk factors associated with cognitive impairment in patients with Leukoaraiosis(LA)and establishing a risk prediction model and scoring system for cognitive impairment,in order to identify high-risk individuals early and intervene timely to delay the progress of dementia.Methods 126 patients who were admitted to the Department of Neurology of the Second Affiliated Hospital of Guangxi Medical University from October 2018 to October 2019 and completed a skull magnetic resonance examination and met the diagnostic criteria for leukoaraiosis were collected,using cross-sectional study.People were divided into cognitive impairment group(MCI)and non-cognitive impairment group(NC)according to Mo CA scale,whose blood biochemical indicators,neuropsychological scale,Doppler ultrasound examination,cardiovascular screening,and magnetic resonance imaging data were collected and analyzed by statistical method.Univariate and multivariate logistic regression analysis were performed on the relationship between the above parameters and cognitive impairment in patients with LA,and then a risk prediction model and scoring system were established.The diagnostic value of the model and scoring system were evaluated using the ROC curve.Results 70 of the 126 patients with leukoaraiosis were suffered from cognitive impairment.Age,total cholesterol,homocysteine,endogenous creatinine clearance rate,PI value,ABI value and Fazekas scale in the cognitive impairment group were distinguished from those in the non-cognitive impairment group and there was statistical significance(P <0.05),while others had no significant difference between two groups.In univariate logistic regression analysis,advanced age,hyperhomocysteinemia,hyperlipidemia,higher PI value,and Fazekas scale were risk factors for cognitive impairment in white matter lesion,while the endogenous creatinine clearance rate and ABI value were not significantly associated with mild cognitive impairment.Multivariate logistic regression analysis indicated that advanced age(OR=4.865;95%CI:1.922-12.318,P=0.001),hyperlipidemia(OR=3.340;95%CI:1.134-9.835,P=0.029),hyperhomocysteinemia(OR=3.954;95% CI: 1.500-10.419,P=0.005),higher PI value(OR= 3.322;95%CI:1.294-8.532,P=0.013)and Fazekas scale(OR=2.776;95%CI:1.570-4.909,P<0.001)were independent risk factors for cognitive impairment in white matter lesion after adjusting for the effects of confounders.We established a risk prediction model based on the results of multivariate logistic regression analysis,and ROC curve of which was 0.867(95% CI: 0.800-0.934,P <0.001),indicating that the model had good discrimination validity;Hosmer-Lemeshow analysis results showed that P = 0.298> 0.05,(?(17)=8.403,ν = 7),suggesting that the model fit well.A predictive scoring system for cognitive impairment in patients with leukoaraiosis was established using the five indicators obtained from multivariate logistic regression analysis.The total score of scoring system was 7 points,and the area under the ROC curve was0.856(95% CI: 0.787-0.925,P <0.001).The sensitivity was 82.9% and the specificity was 78.6% when the threshold was 3.5 points.Conclusion Advanced age,hyperlipidemia,hyperhomocysteinemia,high PI value,and Fazekas score were independent risk factors for cognitive impairment in patients with white matter lesion.The risk prediction model and scoring system which included 5 risk factors had good discrimination and calibration,as well as high sensitivity and specificity in distinguishing the patients with cognitive impairment.By establishing a simple predictive model and scoring system for the risk of cognitive impairment in patients with leukoaraiosis,early diagnosis and intervention of high-risk individuals were important in the clinical work.The clinical application value was worth further expanding the sample and follow-up study. |
PDF Full Text Request