| Purposes:To evaluate the predictive value of CHADS2, CHA2DS2-VASc and R2-CHADS2 scores for risk stratification of ischemic stroke in chinese patients with nonvalvular atrial fibrillation.Methods:Patient with nonvalvular atrial fibrillation, who had complete clinical data, were included in this retrospective study in October 2012-March 2014. We collected all patients’ basic clinical data, such as age, sex, BMI, blood pressure, medical history, and so on. Then, we calculated all patients’scores according to the standard for CHADS2, CHA2DS2-VASc, R2-CHADS2 score, we retrospectively investigated the incidence of ischemic stroke in lyear.Comparativing the predictive value of CHADS2, CHA2DS2-VASc and R2-CHADS2 scores for risk stratification of ischemic stroke in patients with nonvalvular atrial fibrillation is on the basis of the receiver-operating characteristic curve (ROC curve).Result:1. A total of 474 patients were selected, of which 255 cases of female, male 219 cases. Average age 68+10; Medical history of hypertension and coronary heart disease is the most common,63.92% of patients with hypertension,43.67% with coronary heart disease. In the regular oral drugs, antiplatelet drug utilization rate is highest(81.65%). Follow-up of 1 year, a total of 38 cases of patients with is chemic stroke.2.CHADS2 score scored an average of 2.06±1.54, CHADS2 score= 0(low risk) accounted for 17.10%, CHADS2= 1 (moderate) accounted for 25.30%, CHADS22?2 (high risk group) accounted for 57.59%; CHA2DS2-VASc score scored an average of 3.53±2.04, CHA2DS2-VASc score= 0(low risk) accounted for 6.90%, CHA2DS2-VASc= 1 (moderate) accounted for 9.49%, CHA2DS2-VASc≥2(high risk group) accounted for 83.54%; R2-CHADS2 score scored an average of 2.38±1.91, R2-CHADS2 score= 0(low risk) accounted for 17.10%, R2-CHADS2=1 (moderate) accounted for 24.10%, R2-CHADS2≥2 (high risk group) accounted for 58.86%.With the increase of three score, the incidence of stroke is rising.3.The C-statistics of CHADS2 score is 0.657 (95% CI,0.567 to 0.748, p= 0.001); The C-statistics of CHA2DS2-VASc score is 0.718 (95% CI, 0.633 to 0.804, p= 0.001). The C-statistics of R2-CHADS2 score is 0.668 (95% CI,0.576 to 0.761, p= 0.001). The overall predictive value of CHA2DS2-VASc scores are better than other two.4.1t is demonstrated that stroke/TIA, renal failure, hypertension, diabetes, age≥75, vascular disease, and congestive heart failure are the independent risk factors for ischemic stroke in atrial fibrillation and prior stroke/TIA and renal failure OR value is significantly higher than other factors, both the OR, respectively, (3.064,95% CI 1.5606.017, p= 0.001), (2.733,95% CI 1.3115.695, p= 0.006). Age 65 to 75 years, female seem to have nothing to do with ischemic stroke in nonvalvular atrial fibrillation.5. The rate of CHA2DS2-VASc score reclassirying CHADS2 score 0 and 1 is 59.25% and 85% respectively, and after reclassification the incidence of stroke significantly become lower, by 3.7% and 4.16% reduced to 0 and 3.7% respectively. The rate of R2-CHADS2 score reclassirying CHADS2 score 0 and 1 is 0 and 7.5% respectively, and after reclassification the incidence of stroke did not fall significantly. The ability of CHA2DS2-VASc score identifying the "real" low-risk patients is better than other two scores.6.CHADS2≥2 points accounted for 57.59%, the incidence of stroke is 10.98%; CHA2DS2-VASc≥2 points accounted for 83.85%, the incidence of stroke is 9.34%; R2-CHADS2≥ 2 points accounted for 58.86%, the incidence of stroke is 11.11%. The ability of R2-CHADS2 score identifying the "real" high-risk patients is better than other two scores.Conclusions:CHADS2, CHA2DS2-VASc and R2-CHADS2 all can make predictions on risk of ischemic stroke in nonvalvular atrial fibrillation. The overall predictive and identifying the "real" low-risk patients value of CHA2DS2-VASc score are better than other two score, but its identifying the "real" high-risk patients ability is wost in three scores;The ability of R2-CHADS2 score identifying the "real" high-risk patients is better than other two scores. |
PDF Full Text Request