Clinicaland Imaging Features Of Cancer-Related Ischemic Cerebral Infarction And Construction Of Risk Prediction Models

Background: To analyzed the clinical characteristics and imaging findings of patients with cancer-related ischemic cerebral infarction(CAIS),and a risk prediction model was constructed.Methods: A total of 28 patients with acute ischemic stroke with malignant tumor between January 2015 and November 2020 were collected and assigned to the cancer group.In addition,33 patients hospitalized for routine acute ischemic stroke during the same period were randomly selected as the control group.Demographic data and common risk factors of cerebrovascular disease,laboratory data,and imaging characteristics in these two groups were compared.Single-factor correlation analysis and step-wise Logistic regression were used to analyze multiple risk factors of CAIS,and a comprehensive prediction model was obtained.The ROC curve was used to analyze the diagnostic value of the risk factors and the risk prediction model to CAIS.Results: Cancer Patients with CAIS have higher levels of plasma D-dimer and FDP,and are more likely to have cerebral infarction in ≥ 3arterial territories,multiple and even bilateral anterior and posterior circulation.Multivariate logistic regression analysis showed that D-dimer(OR=1.504,P=0.052)and FDP(OR=3.680,P=0.002)levels increased,and AMBI(OR=13.099,P=0.018)were independent risk factors for CAIS.ROC analysis showed that the cut-off value of DD was 0.77 μ g/ml,the AUC was 0.885,the sensitivity was 82.1%,and the specificity was 90.9%;and the cut-off value of FDP was 3.165 μ g/ml,the AUC was 0.857,the sensitivity was 75.0%,and the specificity was 90.9%;while the AUC of the prediction model was 0.925,which was significantly higher than the AUC area predicted by other risk factors alone,The predicted cut-off value was0.636,and the sensitivity was 78.6 %,the specificity was 100.0%.Conclusions: AMBI that occurs in≥3 arterial territories and multiple cerebral circulations is the imaging features of CAIS.The increased levels of DD,FDP,and AMBI are independent risk factors for CAIS.D-dimer≥0.77μg/ml,FDP≥3.165μg/ml may be effective cut-off values and sensitive indicators for identifying CAIS patients.The risk prediction model established based on the relevant risk factors has improved the prediction efficiency to a certain extent.