The Predictive Factor And Its Value In Predicting Ischemic Stroke/TIA Of Patients With Non-valvular Atrial Fibrillation

Part 1: The Value of Urine Albumin in Predicting Ischemic Stroke/TIA Events for Patients with Non-valvular Atrial FibrillationBackground: In the management of patients with non-valvular atrial fibrillation?NVAF?,accurate risk stratification is important.However,one cohort study demonstrated that the annual stroke rate of the patients classified in “the true low risk” according to CHA₂DS₂-VASc score system is 1.61 %.As a result,we aimed to discover more indicators and further evaluate their abilities in predicting ischemic stroke/transient ischemic attack?TIA?.Methods: This study enrolled 400 NVAF patients,including 134 patients with a history of stroke or TIA in the stroke/TIA group,and 266 patients without stroke/TIA in the non-stroke/TIA group.The clinical parameters of these patients were analyzed by univariate analysis and unconditioned logistic regression analysis.And the parameter with statistical significance would be further analyzed by the receiver operating characteristic?ROC?curve for determining the optimal cut-off point,before the combination with CHA₂DS₂-VASc score.Results: Urine albumin?UA?of the stroke/TIA group was significantly higher than that of the non-stroke/TIA group?0.1g/L vs 0.0g/L,P<0.01?.The OR of UA,which was analyzed by unconditional logistic regression analysis,was 3.198?95% CI:1.149—8.897,P <0.05?.The result of the ROC curve analysis showed the Area Under the Curve?AUC?of UA was 0.630 with an optimal cut-off point of 0.03g/L.And the AUC of CHA₂DS₂-VASc-UA2 score was lager than that of CHA₂DS₂-VASc score?0.889 vs 0.781,P <0.01?.Conclusion: UA?0.03g/L is the independent predictive factor of stroke/TIA events for NVAF patients.And compared with the CHA₂DS₂-VASc score,the CHA₂DS₂-VASc-UA2 score might perform better in predicting stroke/TIA.Part 2: The Value of Glycosylated Hemoglobin A1 c in Predicting Ischemic Stroke/Transient Ischemic Attack Events for Non-valvular Atrial Fibrillation Patients with DiabetesBackground: CHA₂DS₂-VASc score system is the common stroke risk stratification of atrial fibrillation.However,an annual stroke rate of 1.61% has been reported among the group of NVAF patients with a CHA₂DS₂-VASc of 0 and 1 score.Diabetes is a well-known risk factor of stroke in patients with atrial fibrillation,while HbA1c reflects the severity of diabetes.Our study aimed to evaluate the value of HbA1c in predicting stroke/TIA events in NVAF patients with diabetes.Methods: Totally 101 hospitalized diabetic patients with NVAF were included and divided in the stroke/TIA group?n=37?and the non-stroke/TIA group?n=64?in this study.The clinical parameters of each patient were subjected to univariate analysis and unconditioned logistic regression analysis for evaluating the predictive value of HbA1c.The optimal cut-off point of HbA1c was determined by receiver operating characteristic curve analysis,and compared with the goal of HbA1c suggested by American Diabetes Association,<6.5% for diabetic patients.Results: HbA1 c of patients in the stroke/TIA group was higher than that of patients in the non-stroke/TIA group?7.61 ± 1.69% vs 6.44 ± 1.30%,P<0.01?.The results of unconditioned logistic regression analysis indicated the OR of HbA1c is 1.620?95%CI:1.138—2.306,P <0.01?.The optimal cut-off point of HbA1c was 6.15% with an AUC of 0.737.The percentages of patients with high HbA1c levels between the thrombosis group and the non-thrombosis group had significant difference under the HbA1c goal of 6.15%?85.7% vs 55.4%,P<0.05?,but no statistical differences between groups under the HbA1c goal of 6.5%?67.9% vs 47.7%,P=0.074?.Conclusion: HbA1c is closely related to the risk of stroke/TIA events occurrence in NVAF patients with diabetes,and it might be the predictive factor in stroke/TIA events.