Risk Factors Of Early Neurological Deterioration In Acute Ischemic Stroke And Construction Of Nomogram Prediction Model

Background:Acute ischemic stroke(AIS)is a common disease in the Department of Neurology.Early neurological deterioration(END)is not uncommon in patients with acute ischemic stroke during hospitalization.How to identify END as early as possible and carry out effective intervention has become an urgent problem to be solved in clinical work.At present,the known predictive factors include laboratory test,imaging examination,neuroelectrophysiology and other related indicators,which provide us with a variety of predictive tools for timely judgment of early neurological deterioration.The value of a single predictor is limited,and a series of prediction models integrating multiple risk factors came into being.Different models include different risk factors and have different predictive ability.The ideal model can accurately predict the probability of disease progression of individual patients,which not only has guiding significance for clinicians’ decision-making,but also provides an objective and valuable reference for doctors to predict the clinical outcome of patients.In addition,the prediction model can also classify the risk level of subjects,guide clinical research,and has clinical application value.In addition to common risk factors,whether laboratory indicators such as blood urea nitrogen / creatinine ratio(BUN/Cr)and urine specific gravity(USG)can objectively reflect the dehydration status of the body,whether they are independent risk factors for END,and whether they can be combined with these indicators to predict end and give targeted rehydration treatment to prevent the progression of infarction are still unclear,which need to be further explored.In this paper,BUN/Cr and USG are included to explore the prediction model for predicting END.Objective:To explore the risk factors of early neurological deterioration in acute ischemic stroke,and to construct a nomograph prediction model according to independent risk factors.Methods:A total of 284 patients with AIS diagnosed by cranial magnetic resonance imaging(MRI)in the Department of Neurology,Huaihe Hospital of Henan University from March 2020 to October 2021 were retrospectively collected.According to the NIHSS score,the patients were divided into END group and nonEND group.Data were collected from patients,baseline data:gender,age,history of hypertension,history of diabetes,smoking history,clinical and laboratory data: admission to National Institute of Health Stroke Scale(NIHSS),fasting blood glucose(FBG),Bun/Cr,Total cholesterol(TC),triglyceride(TG),high-density lipoprotein cholesterol(HDL-C),low-density lipoprotein cholesterol(LDL-C),homocysteine(HCY),DDimer(D-D),USG,Imaging data: whether there is intracranial responsible artery stenosis and whether there is extracranial responsible artery stenosis.The above data of the two groups were analyzed by univariate analysis,and the statistically significant indexes of univariate analysis were included in multivariate logistic regression analysis to explore the independent risk factors of END in patients with primary infarct expansion of acute ischemic stroke.The nomogram prediction model was established and evaluated based on the independent risk factors..Results:(1)A total of 284 patients meeting the experimental requirements were included,including 70patients(24.65%)in the END group and 214 patients(75.35%)in the non-END group.Univariate analysis showed that in END group and non-END group,hypertension(P=0.038),diabetes(P=0.032),BUN/Cr(P=0.000),TC(P=0.037),LDL-C(P=0.038),fasting blood glucose(P=0.001),D-Dimer(P=0.005),NIHSS score(P=0.000),intracranial responsible artery stenosis(P=0.000),fever(P=0.000),lobar infarction(P=0.010),watershed infarction(P=0.008),and paraventricular infarction(P=0.000)the difference was statistically significant.The scores of BUN/Cr,TC,LDL-C,fasting blood glucose,D-dimer and NIHSS in the END group were higher than those in the non-END group.(2)Multivariate logistic regression analysis showed that BUN/Cr(P=0.013 OR=1.069 95%CI1.014-1.126),fasting blood glucose(P=0.016 OR=1.241 95%CI 1.042-1.479),NIHSS score(P=0.000OR=1.272 95%CI 1.120-1.445),intracranial responsible artery stenosis(P=0.031 OR=2.120 95%CI 1.073-4.189),fever(P=0.015 OR=2.731 95%CI 1.220-6.113),watershed infarction(P=0.008 OR=3.637 95%CI1.409-9.383),paraventricular infarction(P=0.002 OR=4.231 95%CI 1.707-10.486)were independently correlated with early neurological deterioration(P<0.05).(3)ROC curve analysis showed that the best critical value of BUN/Cr for predicting END in patients with acute ischemic stroke was 17.115,the area under the curve was 0.702(95%CI 0.633-0.771),the sensitivity was 82.9%,and the specificity was 43.5%;The best critical value of fasting blood glucose to predict END is 5.465,the area under the curve was 0.603(95%CI 0.523-0.683),the sensitivity was 57.1%,and the specificity was 58.4%;The best critical value of END predicted by NIHSS score is 3.5,the area under the curve was 0.709(95%CI 0.642-0.776),the sensitivity was 61.4%,and the specificity was 72.9%.(4)According to the independent risk factors of early neurological deterioration of acute ischemic stroke screened by multivariate logistic regression analysis,a nomograph prediction model was constructed.The area under the curve of the model was 0.828(95%CI 0.769-0.888)using ROC curve,suggesting that the model has good differentiation,and the calibration curve shows that the model has good consistency.The decision curve analysis(DCA)indicates that the net benefit of this model is high.Conclusion:(1)Bun/Cr,fasting blood glucose,NIHSS score,intracranial responsible artery stenosis,fever,watershed infarction and paraventricular infarction are independent risk factors for END in patients with acute ischemic stroke,and have certain predictive value.(2)The nomogram prediction model based on the above independent risk factors has good prediction ability.