Evaluation of Risk Stratification Schemes for Ischaemic Stroke and Bleeding in 182,678 Patients With Atrial Fibrillation: The Swedish Atrial Fibrillation Cohort Study
Study Questions:
What are the most accurate risk stratification systems for predicting strokes and bleeding complications in patients with atrial fibrillation (AF)?
Methods:
This cohort study enrolled 170,291 patients with AF, including 90,490 patients (mean age 76 years) who never received anticoagulant therapy. The mean duration of follow-up was 1.5 years. The thromboembolism endpoint consisted of ischemic stroke, transient ischemic attack, or systemic embolism. The bleeding endpoint was intracranial hemorrhage (ICH).
Results:
The annual thromboembolism rate was 0.3% in patients with a CHA2DS2-VASc score of 0, compared to 0.9% in patients with a CHADS2 score of 0. The most accurate risk stratifier for thromboembolism was the CHA2DS2-VASc scheme (c-statistic 0.67). The rate of ICH was 0.6%/year. The HAS-BLED and HEMORR2HAGES scoring systems were equally accurate in predicting ICHs (c-statistic 0.6).
Conclusions:
The authors concluded that the CHA2DS2-VASc scoring system is a more accurate predictor of thromboembolism than other scoring systems, including CHADS2, in patients with AF. The HAS-BLED and HEMORR2HAGES schemes have similar ability to predict ICHs.
Perspective:
The CHA2DS2-VASc provides a greater degree of risk stratification among patients who are ‘low risk’ based on the CHADS2 system. For example, a 67-year-old woman with carotid artery disease would have a CHADS2 score of 0, but a CHA2DS2-VASc score of 3. Anticoagulation would be recommended based on the latter, but not the former. Risk stratification for ICH is not as accurate as for stroke. The simpler of the two schemes evaluated in this study is HAS-BLED, based on hypertension, abnormal renal/liver function, stroke, bleeding, labile international normalized ratio, age >65 years, and drug/alcohol use.
Clinical Topics: Prevention, Vascular Medicine, Hypertension
Keywords: International Normalized Ratio, Thromboembolism, Stroke, Intracranial Hemorrhages, Follow-Up Studies, Ischemic Attack, Transient, Carotid Artery Diseases, Risk Factors, Liver Diseases, Hypertension
< Back to Listings