Associative Risk Factors For Stroke In Patients With Non-valvular Atrial Fibrillation

BackgroundAtrial fibrillation(AF)is the most common arrhythmia treated in clinical practice,the association of this arrhythmia with stroke has long been recognized.There are clinical schemas such as CHA2DS2-VASc score,incompletely account to predict thromboembolic risk(stroke).This score combines common clinical risk factors for stroke such as congestive heart failure,hypertension,age.diabetes,previous stroke/TIA,vascular disease and sex.AF,CHA2DS2-VASc score,cardiac biomarkers,and echocardiography parameters have been separately shown to be associated with stroke risk.The combination of these factors to identify stroke risk is still unclear,as some studies have realized the association with stroke risk for each factor individually.Recent studies have found that extra-cellular matrix protein measurable in the plasma correlate with AF risk suggesting potential value as a biomarker,this later is involved in different pathways such like coagulation,myocardial damage,inflammation also fibrosis substrate production and cause arrhythmogenic changes sustaining AF,thromboembolism event.Our study will investigate three biomarkers and will seek to assess additional risk factor independently associated alone and in combination with risk factors in nonvalvular atrial fibrillation and strokes.ObjectiveWe aimed to further determine the association between additional risk factors for non-valvular atrial fibrillation and first onset stroke in a sample patients group.We will investigate the predictive ability of additional risks factors such as cardiac biomarkers and AF status for odds of incident stroke in the sample patient group.MethodsThis retrospective study(cross-sectional design)enrolled more than 2000 cases of atrial fibrillation continuously hospitalized in different internal medicine services of the Second Affiliated Hospital of***Medical University from 01/2015 to 12/2017.We excluded patients with prior stroke,valvular heart diseases and included 455 patients suffered from non-valvular atrial fibrillation divided into two distinct groups,with informed consent.These two groups were based on whether the patients were having a first onset stroke(stroke group)or they were not having a stroke(non-stroke group).We made the baseline characteristic of our sample group from the substantial evidence of risk factors(CHA2DS2-VASc),lifestyle(cigarette smoking,alcohol intake),type of atrial fibrillation;to controversial risk(cardiac biomarkers:high sensitive troponin,BNP,and D-dimer)described in the new international guideline of atrial fibrillation.(ESC,ACC/AHA).We used the statistical package for social science(SPSS 23 for Windows)for statistical analysis of study data at 95%CI,p<0.05 was considered statistically significant.We described non-normally distributed continuous variables as median±SD,and categorical data as a percentage.Chi-square and Kruskall-Wallis H test was performed to compare categorical data and continuous variables between two groups.Multivariate(binary)logistic regression model was performed to identify risk factors independently associated with non-valvular atrial fibrillation and first onset stroke.The receiver operating characteristic(ROC)curve was drawn to evaluate the predictive value of these risk factors(Biomarkers).ResultsThis study enrolled 455 patients aged 74± 10 years,comprised of 213 women and 242 men.The first group included 312 patients without a stroke and the second involved 143 patients with a stroke.1.Chi-square and Kruskall-Wallis H test revealed us respectively that:heart failure,coronary disease,cigarette smoking,type of atrial fibrillation,warfarin treatment.high sensitive troponin and BNP are statistically associated with stroke in non-valvular atrial fibrillation patients(p<0.05).Otherwise,age,sex,hypertension,kidney disease,diabetes,thyroid disease,alcohol intake,season,aspirin and D-dimer medication were not associated with stroke(p>0.05).2.Multivariate analysis:age[OR=1.465,95%CI(1.168-1.900),p=0.006],smokers[OR=3.200,95%CI(1.080-5.000),p=0.006];coronary heart disease[OR=0.394,95%CI(0.220-0.706),p=0.002];heart failure[OR=0.477,95%CI(0.371-0.673),p=0.000],high sensitive troponin level[OR=1.000,95%CI(0.999-1.001),p=0.010];BNP[OR=0.999,95%Cl(0.999-1.000),p=0.001],permanent atrial fibrillation[OR=0.453,95%CI(0.192-1.073),p=0.042]warfarin[OR=0.429 95%CI(0.238-0.773),p=0.005],were independently associated with a high stroke risk outcome in non-valvular atrial fibrillation patients,after adjustment of various factors.However,sex,hypertension,kidney disease,diabetes,thyroid disease,alcohol intake,season,aspirin medication,paroxysmal atrial fibrillation and D-dimer level were not independently risk factors of stroke.3.Receiver Operator Characteristics(ROC)analysis.High sensitive troponin and BNP level displayed an excellent area under ROC curve respectively[(AUROC=(0568±0.029),p=0.010];[(AUROC=(0.288±0.0.026),p=0.001]for predicting prevalence of stroke.The combination of biomarkers with the clinical features can be useful(CHA2DS0-VASc/BS)for the prediction of stroke than CHA2DS0-VASc alone,respectively[AuROC=(0.378±0.028),p=0.000;AUROC=(0.410±0.028),p=0.002].The sensitivity and specificity of CHA2DS0-VASc/BS score were 88.1%and 94.2%.The sensitivity and specificity of CHA2DS0-VASc score were 74.I%and 81.7%.Conclusion1.Coronary heart disease,heart failure,cigarette smoking,type of atrial fibrillation,warfarin medication,high sensitive troponin and BNP level are associated with elevated stroke risk in patients with non-valvular atrial fibrillation.2.Age,coronary heart disease,heart failure,cigarette smoking,permanent atrial fibrillation warfarin medication,high sensitive troponin,BNP level are independent risk factors in stroke patients with non-valvular atrial fibrillation,even after adjustment for various risk factors.3.Cardiac biomarker with clinical risk factor(CHA2-DS2-VASc)independently associated with high stroke risk(outcome)in non-valvular atrial fibrillation patients.Impact of biological markers as an additional risk factor in this life-threatening event may hold a clinical role which may help to redefine and predict stroke event better than the current CHA2DS2-VASc.