Association Of Atrial Fibrillation With Other Risk Factors And Ischemic Stroke

ObjectivesBased on the national screening data of high-risk populations of stroke,we investigated the association between atrial fibrillation（AF）coexist other factors and the risk of ischemic stroke（IS）.To provide a basis for the development of IS prevention strategies in different AF population.MethodsBase on the cross-sectional data from the 2014-2015“Screening and Intervention Program for Stroke High-Risk Groups in China”.According to the inclusion and exclusion criteria excluded patients with hemorrhagic stroke,subarachnoid hemorrhage and other cerebrovascular diseases,a total of 720,628 survey subjects were included.Eleven variables including IS,AF,age,gender,family history of stroke,hypertension,diabetes,dyslipidemia,overweight or obesity,smoking history and lack of exercise were taken as nodes.Based on the mountain climbing algorithm of Bayesian Information Criterion（BIC）score to construct Bayesian network of Directed Acyclic Graph（DAG）,to explore the association between IS and AF,and to analyze the association between AF and other IS risk factors and their possible roles in the occurrence of IS.Based on 720,628 subjects,continue to exclude individuals without AF who coexisted with any of the 8 risk factors（i.e.,age≥65 years,family history of stroke,hypertension,diabetes,dyslipidemia,overweight or obesity,smoking history,and physical inactivity）.A total of 251,241 research subjects were included to further construct the unconditional Logistic regression model to analyze:（1）To analyze the relationship between the number of AF coexistence and other risk factors and IS risk.The non-AF population without coexisting risk factors was used as a reference,and the P_trendtest was performed;The dose-response relationship between the number of coexisting risk factors for AF and the risk of IS was fitted by making restricted cubic spline plots;The correlation between the number of AF coexisting other risk factors and the risk of IS was analyzed by stratification according to the gender,ethnic group,urban and rural residence and social and economic level of the study subjects.Sensitivity analysis was conducted by excluding individuals with a family history of stroke to assess the robustness of the association between the number of other risk factors AF coexisting and IS.（2）On the basis of Bayesian network analysis,combined with the results of literature review and the opinions of clinical experts,the common conditions of AF coexistence with other risk factors were combined,and the association between the types of AF coexisting with other risk factors and IS risk was analyzed.Results1.A total of 720,628 subjects（336,518 males and 384,110 females）were included in the study based on Bayesian network,with 12,984 IS patients.There were statistically significant differences in age,socioeconomic status,hypertension,dyslipidemia,diabetes,AF,smoking history,physical inactivity,overweight or obesity,and family history of stroke in the general population,male population,and female population（P<0.001）.In the general population and male population,ethnic differences were statistically significant（P<0.001）,and there was no significant difference among ethnic groups in female population（P=0.082）.In the general population,male population and female population,there was no significant difference among urban and rural areas（P>0.05）.The results of Bayesian structure learning show that AF can directly influence IS,and the influence of AF on IS also affected by age,family history of stroke,hypertension,dyslipidemia and lack of exercise.When one or more of the above factors coexist in the AF population,these factors may affect the risk of IS by affecting certain characteristics of AF patients.2.A total of 251,241 subjects were included in the study based on the unconditional Logistic regression model.The results of the association between the number of AF coexisting with other risk factors and IS showed that when the non-AF population without other risk factors were taken as reference,AF coexisting 0,1,2,3,4 and 5 other risk factors after adjusting for gender,ethnic group,urban and rural area,and socioeconomic level,the OR and 95%CI were 7.17（4.40,11.69）,16.16（12.85,20.33）,30.55（25.69,36.33）,46.44（39.47,54.65）,73.98（62.72,87.28）,and112.75（93.32,136.21）,respectively.As a whole,after adjusting for gender,ethnic group,urban and rural area,and socioeconomic level,the risk of IS increased gradually with the increase of the number of coexisting risk factors for AF(OR:2.19,95%CI:2.15,2.24;P_trend<0.001).The risk of IS increased by 119%for each increase in the number of coexisting risk factors for AF.The results of subgroup analysis and sensitivity analysis were robust.3.Correlation analysis results of the AF coexisting types of other risk factors and IS risk showed that:For AF coexisting 1 risk factor,the top 3 in order from high to low risk of IS were AF coexisting family history of stroke,AF coexisting hypertension,and AF coexisting age≥65 years old.After adjusting for gender,ethnic group,urban and rural area,and socioeconomic level,the corresponding OR and 95%CI were42.36（22.13,81.06）,23.12（14.87,35.95）and 20.39（13.71,30.33）,respectively.When AF population coexisting two other risk factors,the top 3 in order from high to low risk of IS were AF coexisting age≥65 years and family history of stroke,AF coexisting diabetes and family history of stroke,and AF coexisting dyslipidemia and family history of stroke.After adjustment of gender,nationality,urban and rural areas,social economic level,the corresponding OR and 95%CI were 103.87（48.86,220.84）,80.91（18.52,353.56）and 71.74（32.60,157.88）,respectively.Conclusions1.Overall,the risk of IS in AF population increased significantly with the increase of the number of AF coexisting risk factors of IS.2.The type of other risk factors coexisting with AF was related to the risk intensity of IS.When a single risk factor coexisted,the IS risk was highest when AF coexisted with a family history of stroke.When AF coexisting two risk factors,the risk of IS was the highest when AF coexisted with a family history of stroke and age≥65 years old,followed by AF coexisting with diabetes and family history of stroke,AF coexisted with dyslipidemia and family history of stroke.3.In the prevention and treatment of IS,AF subpopulation can be divided according to the types of co-existing risk factors of AF,and the management and intervention of subpopulation can be carried out according to the degree of risk and the types of co-existing factors.