| Objective: In order to early identify and intervene malignant brain edema in patients with large cerebral hemisphere infarction,we analyzed the general data,clinical features,laboratory indicators,imaging features and vascular recanalization of the patients within 24 hours of onset,searched for independent risk factors for malignant brain edema,and established a risk prediction model for malignant brain edema based on the screened risk factors,so as to provide reference basis for guide the clinical work.Methods: A total of 220 patients with cerebral hemispheric massive infarction admitted to Lanzhou University Second Hospital within 24 hours of onset from January 2015 to May 2020 were retrospectively collected as the study group.1.Analysis of risk factors for malignant cerebral edema in patients with massive cerebral hemisphere infarction: according to the occurrence of malignant brain edema,the two groups were divided into malignant brain edema group and non-malignant brain edema group.The general data,clinical characteristics and laboratory indicators,imaging characteristics and vascular recanalization were compared between the two groups.Through multivariate Logistic regression analysis,independent risk factors of malignant brain edema were screened out from the variables with statistical difference.2.Short-term prognosis analysis of different treatment methods in the malignant brain edema group: The mortality rate within 14 days of different treatment methods in the malignant brain edema group was analyzed.3.Establishment of risk prediction model for malignant cerebral edema caused by massive cerebral infarction in cerebral hemisphere: according to the independent risk factors screened by the multi-factor Logstic regression model,the regression coefficients of each risk factor were divided by the minimum regression coefficient and rounded to an integer to obtain the risk prediction scoring system for the occurrence of malignant brain edema,and the risk prediction model for malignant brain edema was established;differentiation degree and calibration degree were used to evaluate the efficiency of the risk prediction model;the prediction model was stratified by risk;4.Validation of risk prediction model for malignant brain edema: By the same method,we prospectively collected patients with cerebral hemispheric massive infarction who were admitted to the Second Hospital of Lanzhou University within 24 hours of onset from June 2020 to January 2021 as the validation group to verify the established model and evaluate the predictive efficacy of this model in other populations.Results: 1.The analysis of risk factors of malignant cerebral edema in cerebral hemispheric massive infarction:(1)Results of clinical data between the two groups was showed: baseline NIHSS score,age,GCS score,white blood cell count,neutral lymphocyte ratio,random blood glucose,CK-MB,D-dimer,MCA hyperdense,involving the ACA and(or)PCA blood supply area,the early middle shift difference had a statistical difference(P < 0.05)between the two groups.(2)After the continuous variables in the above results were converted into dichotomous variables,all variables with statistical differences were performed dichotomous multivariate Logistic regression analysis,and the results were as follows: age ≤60 years old,NIHSS score≥15 points,GCS score ≤8 points,neutral lymphatic ratio(NLR)≥6.83,MCA high density sign,early midline displacement,and involvement of ACA and/or PCA blood supply area were independent risk factors for the occurrence of malignant cerebral edema with large cerebral infarction.2.Short-term prognosis analysis of different treatment methods in the malignant brain edema group: the mortality within 14 days in the conservative treatment group was higher than that in the decompression group(P < 0.05).3.Establishment of risk prediction model for malignant brain edema:according to the regression coefficient of each risk factor in the regression model divided by the minimum regression coefficient,rounded to an integer.The risk prediction score system of malignant brain edema was obtained.It includes age ≤60years(1 point),NIHSS score ≥15 points(1 point),GCS score ≤8 points(2 points),neutral lymphocyte ratio ≥6.83 points(1 point),MCA high density sign(1 point),early midline displacement(2 points),and involvement of ACA and/or PCA blood supply area(2 points).The total score of this scoring system is 0-10 points.With the gradual increase of the score,the prediction probability of the model also increases correspondingly.The area under the ROC curve of the model was 0.907(95%CI:0.868～0.946,P < 0.001).The result of Hosmer-Lemeshow goodness of fit test wasX 2=6.586,P=0.582 > 0.05.The above of two points indicated that the model had good differentiation and calibration,and had good performance in predicting disease.According to the ROC curve,the best cutoff value of the scoring model was 3.5,which was divided into low risk(0-3 points)and high risk(4-10 points).The accuracy,sensitivity and specificity of the stratified model were 83.6%,77.9% and 86.7%,respectively.4.Validation of risk prediction model for malignant brain edema: The model was used in the validation group.The area under the ROC curve in the validation group was 0.891(95%CI : 0.817～0.966,P < 0.001),the result of Hosmer-Lemeshow goodness of fit test was X2=6.020,P= 0.304> 0.05,indicating that the model also showed good clinical predictive ability in the validation group.Conclusion: 1.Age ≤60 years,NIHSS score ≥15,GCS score ≤8,neutral lymphatic ratio(NLR)≥6.83,MCA high density sign,early midline displacement,and involvement of ACA and/or PCA blood supply area is independent risk factors for malignant cerebral edema in massive cerebral infarction of the cerebral hemisphere.2.The risk prediction model established in this study can accurately predict the occurrence of malignant cerebral edema in acute cerebral hemispheric infarction,which is helpful for clinicians to conduct disease assessment and clinical decision-making,and has certain guiding significance for clinical work. |
PDF Full Text Request