Author + information
- Yang Guo,
- Dong Zhao,
- Jiayi Sun,
- Miao Wang,
- Wuxiang Xie and
- Jing Liu
The assessment of thromboembolic risk is important in atrial fibrillation (AF) management. The latest AHA/ACC/HRS and ESC guidelines recommended CHA2DS2-VASc score be superior to CHADS2 in assessing thromboembolic risk for patients with nonvalvular AF (NVAF), while which one is more appropriate for Chinese patients is unknown. Our objective was to compare the performance of CHADS2 and CHA2DS2-VASc score in a real world cohort of Chinese patients with NVAF.
We used data from Beijing Hospital Discharge Information System and Vital Registration Monitoring System. A total of 28871 patients with NVAF between January 2007 and December 2009 were enrolled and followed up one year. The outcome was thromboembolic events (TE) including ischemic stroke, transient ischemic attack (TIA), systemic embolism, pulmonary embolism, or acute myocardial infarction. Risk factors for TE were investigated using multivariate Cox regression analyses. The C-statistics (area under the receiver-operating-characteristic curve) were calculated to assess the discrimination of the model for the outcome. The net reclassification improvement (NRI) was calculated to quantify the improvement by a new model as compared to an old model for low-, intermediate-, and high-risk patients.
The annual incidence of TE was 11.1 per 100 person-years and risk factors for TE were advanced age, hypertension, diabetes, and prior stroke/TIA/TE (P<0.001). C-statistics in CHADS2 and CHA2DS2-VASc score were 0.566 (95% CI: 0.555-0.577) and 0.570 (95% CI: 0.559-0.580) respectively (P<0.001); NRI was 0.059 (P<0.001) when using CHADS2 versus CHA2DS2-VASc score. Then, HA2DS-A score based on risk factors was proposed. The new score was calculated by adding 1 point each for any of the following-hypertension (H), age 65-74 years (A), diabetes (D) and a history of stroke or TIA or TE (S)-and 2 points for age 75 years or older (A2). C-statistic was 0.583 (95% CI: 0.573-0.594, P<0.001) improving significantly and a significant NRI (0.060, P<0.001) was found between HA2DS-A and CHA2DS2-VASc score.
In Chinese NVAF patients, CHADS2 score performed better than CHA2DS2-VASc score although both had only modest predictive capacity. The HA2DS-A score may be a simpler and useful risk score for Chinese patients but further external validation is needed.