Establishment And Validation Of Stroke-Heart Syndrome Prediction Model Based On Brain Imaging

Objective:1.To analyze the clinical correlation between insular lesions and stroke–heart syndrome(SHS)after acute cerebral infarction in the anterior circulation;2.Through the analysis of SHS risk factors,including past history,personal history,laboratory and imaging indicators,the risk prediction model of SHS high-risk groups in ischemic stroke patients was established,and it was visualized and instrumentalized to provide a basis for early identification and prevention of cardiac damage,improve the prognosis of acute cerebral infarction,and reduce cardiogenic mortality after stroke.Methods:This study retrospectively collected 258 patients with acute cerebral infarction who met the inclusion and exclusion criteria in the First Affiliated Hospital of Dalian Medical University from June 2020 to December 2022 to understand the incidence of SHS in the enrolled patients.The 124 patients in the SHS group were divided into insula group and insular leaf group according to whether they were involved,and the differences in ECG,myocardial markers and BNP between the two groups were compared.The acute cerebral infarction site involving the insular lobe was divided into the left insula infarction group and the right insular infarction group,and the differences in electrocardiogram,myocardial markers and BNP between the two groups were compared.According to whether SHS occurred during hospitalization,the enrolled patients were divided into SHS group and non-SHS group,and the final selected risk factors were taken as the independent variable and whether SHS occurred as the dependent variable,univariate analysis was carried out to obtain statistically significant independent variables(P<0.05),and multivariate analysis was entered to obtain independent risk factors for SHS.According to the SHS independent risk factors,combined with the partial regression coefficients of each risk factor,and fitting the logistic regression equation,the mathematical calculation formula of SHS risk prediction model was established.R language was used to establish the Nomogram for predicting the risk of SHS in patients with acute cerebral infarction,the C index was calculated,and the model verification was repeated for 1000 times by internal validation bootstrap method,and the calibration curve was drawn to evaluate the effectiveness of the Nomogram model in predicting SHS in patients with acute cerebral infarction.The area under the ROC curve(AUC)was calculated to evaluate the discrimination validity of the model itself,and high-risk patients were distinguished by cut-off value.Result:A total of 258 patients with acute cerebral infarction were included in this study,and they were divided into non-SHS group and SHS group according to whether SHS occurred,of which 124 cases in SHS group,accounting for 48.06%.A total of 65 patients in the SHS group and 21 patients in the non-SHS group affected the insular leaf,χ~2=39.136,P<0.001,the difference was statistically significant.The incidence of SHS in infarction involving the insular leaf is higher than that without insular leaf.SHS group was divided into insular leaf group(65 cases)and non-insular leaf group(59 cases),and the incidence of myocardial marker elevation in infarction and insular leaf group was significantly higher than that in non-insular infarction group,P<0.05,the difference was statistically significant;the incidence of BNP abnormality in infarction and insular leaf group was significantly higher than that in non-insular infarction group,P<0.05,the difference was statistically significant,indicating that insular infarction was closely related to myocardial markers and BNP elevation.The incidence of ECG abnormalities in the insular group was lower than that in the noninsular infarction group,P<0.05,and the difference was statistically significant.The acute cerebral infarction site involving the insular lobe in the stroke center syndrome group was divided into left insular infarction group(32 cases)and right insular infarction group(33 cases),and the differences in electrocardiogram,myocardial markers and BNP between the two groups were compared,and the P > 0.05,and the difference was not statistically significant.In the modeling cohort,124 patients with stroke-heart syndrome occurred in 258 patients,and the incidence of stroke-heart syndrome in this study was 48.06%.There were four independent risk factors for acute cerebral infarction complicated by strokeheart syndrome: advanced age,hyperlipidemia,abnormal elevation of glutamate aminotransferase,and infarction involving the insular leaf.The risk prediction model predicted the probability of SHS in acute cerebral infarction,and repeated sampling was carried out 1000 times by Bootstrap method.The Hosmer-Lemeshow goodness-of-fit test showed that the difference was not statistically significant(χ~2 = 5.501,P = 0.703);The calibration curve was plotted,the mean absolute error was 0.024,the mean square error was 0.0008,and the observation probability was generally consistent with the theoretical probability;the area under the ROC curve was 80.32%,and the truncated value of the predicted probability was 60%.Nomogram: In this study,a visual prediction tool was established based on logistic regression analysis model of SHS-related risk factors.The variables included in the Nomogram were independent risk factors for SHS in this study,including age,previous hyperlipidemia,abnormally elevated glutamate aminotransferases,and spatial distribution of acute cerebral infarction involving the insular lobe.Conclusion:1.The incidence of stroke-heart syndrome after acute anterior circulation cerebral infarction was 48.06%;Insula infarction is closely related to the occurrence of strokeheart syndrome;Advanced age,previous hyperlipidemia,abnormal elevation of glutamate aminotransferases,and spatial distribution of acute cerebral infarction involving the insular lobe are independent risk factors for the development of strokeheart syndrome.2.Insular infarction is closely related to abnormal elevation of myocardial markers and BNP.Acute cerebral infarction involving the insular side is not significantly related to the occurrence of stroke-heart syndrome.3.The risk prediction model of stroke-heart syndrome established by this institute has a good prediction effect,which can provide support for clinical staff to identify high-risk groups of stroke-heart syndrome early.4.The variables included in the SHS prediction of the Nomograme were age,previous hyperlipidemia,abnormal elevation of glutamate aminotransferases,and spatial distribution of acute cerebral infarction involving the insular leaf.The risk of stroke-heart syndrome after acute cerebral infarction can be visualized by Nomogram,When the predicted probability ≥60% is determined to be high-risk.