Construction and evaluation of a 1-year risk prediction model for ischemic stroke in patients with acute myocardial infarction and atrial fibrillation

Abstract:

Objective: To investigate the independent risk factors of ischemic stroke in patients with acute myocardial infarction and atrial fibrillation in 1 year, and to construct a risk prediction model. Methods: We retrospectively enrolled 383 patients diagnosed with acute myocardial infarction and atrial fibrillation in the Department of Cardiology of Xinqiao Hospital of Army Medical University. Telephone follow-up for ischemic stroke within 1 year. According to the sequence of follow-up events, they were divided into groups by 7∶3. There were 268 cases in the modeling group and 115 cases in the verification group. According to whether patients in the modeling group had an ischemic stroke, they were divided into a stroke group and a non-stroke group. The risk factors of the two groups were compared and a risk prediction model was established. A total of 115 patients in the verification group verified the efficacy of the ischemic stroke risk prediction model. Results: Advanced age (age>65 years), smoking, drinking, hypertension, D-dimer, low-density lipoprotein, and left atrial diameter, neutrophils-lymphocytes ratio (NLR), and C-reactive protein-albumin ratio (CAR) increased the risk of ischemic stroke in patients with acute myocardial infarction and atrial fibrillation. The final model established in this study was P=e^x/ (1+e^x), where e was the natural logarithm, X=-22.547+0.044×Age+0.653×Smoking+1.277×Drinking+1.134×Drinking hypertension+0.867×D-dimer+0.702×LDL+0.297×Left atrium ID+0.421×NLR+0.665×CAR. The ROC curve area of the modeling group was 0.866 (95% CI=0.820-0.911, P<0.001), the sensitivity was 87.50%, and the specificity was 73.11%. The predictive power of this model was at a moderate level. The area under the ROC curve of the verification group was 0.913 (95% CI=0.854-0.973, P<0.001), the sensitivity was 85.71%, and the specificity was 87.36%. Conclusion: This model has good predictive ability and can be used for the screening of ischemic stroke in patients with acute myocardial infarction and atrial fibrillation within 1 year, and can provide a reference for timely identification and intervention of ischemic stroke.