The CHA₂DS₂-VASc score appears to be better than the CHADS2 score at predicting patients with atrial fibrillation (AF) who will have a stroke over the next 10 years. It is also more accurate in identifying those at very low risk who do not need anticoagulation, a registry-based study has found.

Many GPs will be familiar with the CHADS₂ score which assigns one point to each of: cardiac failure, hypertension, age≥75 and diabetes and two points to a history of stroke/TIA. A total score ≥2 indicates a high risk of thromboembolism and should prompt the initiation of anticoagulation. A score of zero indicates a low risk which can be managed with aspirin or preferably no thromboprophylaxis. However, patients with an intermediate score are more tricky and could be managed with either aspirin or anticoagulants, although guidelines express a preference for the latter.

CHA₂DS₂-VASc divides age into two categories, 65-74 and ≥75, and adds female sex and a history of vascular disease as risk factors. Each component scores one point, except for stroke and age ≥75 which score two points. 

The study used a Danish registry of patients admitted to hospital with AF between 1997 and 2006 who were not currently receiving anticoagulation. A total of 73,538 patients were included and the main outcome measure was stroke and thromboembolism.

The results showed that in patients at low-risk (score = 0) the rate of thromboembolism per 100 person-years was 1.67 (95% CI: 1.47-1.89) with CHADS2 and 0.78 (95% CI: 0.58-1.04) with CHA₂DS₂-VASc at one year of follow-up. In those at intermediate risk (score = 1) the rate was 4.75 (95% CI: 4.45-5.07) and 2.01 (95% CI: 1.70-2.36) respectively. In those at high risk (score ≥2) the rate was 12.27 and 8.82 respectively.

Therefore the more detailed scoring system is much more accurate at identifying truly low-risk patients. However, in the population studied it would classify 80% of patients as high risk compared with less than half using CHADS₂.

Estimating the risk of thrombo-embolism in AF is one of the most important aspects of the management of this common arrhythmia.

To avoid a stroke at all costs using CHA₂DS₂-VASc thresholds for anticoagulation is preferable, but applying this at a population level may lead to an appreciable and significant increase in anticoagulant-related adverse events in patients who may still have a low risk of stroke.

I think ultimately CHADS2 is the easiest to apply in everyday practice as a first-line tool with the more detailed scoring system applied to those scoring 1 with CHADS₂ and who clinically appear to be at higher risk.

Dr Peter Savill