Construction And Validation Of Risk Prediction Model For Malignant Middle Cerebral Artery Infarction

Objective:1.This study focused on the prediction of early non-enhanced CT imaging signs and clinically relevant information by comparing the NEMMI score with the development of malignant middle cerebral artery infarction(MMI)in large cerebral infarction in the hemisphere(LHI).Artery Infarction(MMI)to assess the predictive power of the NEMMI score by briefly comparing it with other common predictors of MMI.2.The purpose of this study was to explore the risk factors that can predict the development of MMI in patients with LHI,and to develop and validate a clinical prediction model through the screening of independent predictors to provide early clinical identification of people at risk for MMI and early intervention.The purpose of this study is to investigate the risk factors that can predict the occurrence of MMI in patients with LHI.Methods:Clinical data were collected retrospectively from January 2019 to December 2022 from inpatients who had been diagnosed with LHI in the intensive care unit of the Department of Neurology at the Third Clinical Hospital of Shanxi Medical University.Demographic data were collected: gender,age,history of smoking,history of alcohol consumption;clinical characteristics: baseline blood pressure(systolic and diastolic),NIH Stroke Scale at admission(National Institutes of Health stroke scale(NIHSS)score at admission,history of hypertension,history of diabetes,history of coronary artery disease,history of TIA(Transient Ischemic Attack)/cerebral infarction,history of atrial fibrillation,and whether acute interventions(including thrombolysis and embolization)were performed at admission;Laboratory data: including baseline glucose,neutrophil lymphocyte ratio NLR,lymphomonocyte ratio LMR,total cholesterol,triglycerides,LDL,albumin,D-dimer,fibrinogen,urea creatinine ratio,proteinuria,etc.;imaging data:including infarct lateralization,hemorrhagic transformation(HT)Imaging data: including infarct side,Hemorrhagic transformation(HT),Middle cerebral artery hyperdensity sign(HMCAS),involvement of ACA and PCA blood supply area,NEMMI score,Alberta Stroke Program Early CT score(ASPECTS).The MMI group and non-MMI group were divided according to the progression of MMI.The general data,clinical characteristics and laboratory indices,and imaging features were analyzed for intergroup variability between the two groups.The NEMMI score was briefly compared with the ASPECTS score and NIHSS score to assess the predictive ability of the NEMMI score.A single-factor binary logistic regression method was used to screen out all potential risk indicators,and then the independent influential indicators of MMI were determined by the forward-step algorithm of the multi-factor binary logistic regression method.The MMI risk prediction model was constructed using R software,and column line plots(Nomogram plots)were used for model presentation,and the predictive efficacy and clinical utility of the constructed model were assessed by plotting subject operating characteristic curves(ROC),calibration curves and decision analysis curves(DCA).The validation of the model was tested internally by cycling 1000 Bootstrap’s self-sampling method.Results:1.The analysis of this study indicated that the risk of MMI in patients with LHI was33.22%(96/289).2.The NEMMI score(AUC: 0.84)showed good discrimination compared to the ASPECTS score(AUC: 0.57)and NHISS score(AUC: 0.60).3.Multifactorial binary logistic regression analysis showed that age [Odds ratio(OR): 1.06,95% confidence interval(CI): 1.03-1.10,P<0.001],baseline blood glucose(OR=1.17,95%CI: 1.03-1.33,P=0.014),NEMMI score(OR= 1.13,95% CI: 1.09-1.17,P< 0.001),NIHSS score(OR=0.95,95% CI: 0.91-0.99,P=0.024),proteinuria(OR=2.84,95% CI: 1.24-6.24,P=0.015),HT(OR=3.56,95% CI: 1.53-8.51,P=0.003),and HMCAS(OR=4.49,95% CI:2.06-10.33,P<0.001)were independent risk factors for the progression of LHI into MMI.4.The sensitivity of the risk column line graph model constructed based on the above seven indicators was 0.89,the specificity was 0.86,and the area under the curve(Area under the The root mean square error(RMSE)was 0.339,the absolute mean error(MAE)was 0.260,and the R2 was 0.492,suggesting a good compliance.The decision analysis curve(DCA)indicated that patients could benefit as much as possible by applying the model for clinical decision making and intervention.Conclusion:1.NEMMI score showed good predictive potential in MMI prediction;2.higher age at admission,blood glucose,NIHSS score,NEMMI score and the presence of proteinuria,hemorrhagic transformation and MCA high-density sign are seven factors that can be used as independent risk factors for the occurrence of MMI;3.the column line graph model established in this study has good accuracy and differentiation,which can effectively improve the diagnosis of conversion to MMI in patients with massive cerebral infarction and help physicians in disease assessment and clinical decision making,and has important clinical application value.