When is Anticoagulation Actually Not Indicated?

It is well recognized that atrial fibrillation (AF) is associated with an increased risk of stroke, and that oral anticoagulation (OAC) is an effective way to markedly reduce that risk. The main obvious limitation, however, is that blood thinners can cause bleeding. Therefore, patients at very low risk of stroke have historically not been anticoagulated. The approach to patients with AF has been to assess their individual risk of stroke and make a determination as to whether the risk of stroke is high enough to warrant the risks of OAC. Given that minor or even major bleeding usually does not have the same impact as a large embolic stroke, the accepted threshold for prescribing warfarin has been the point where the risk of stroke exceeds the risk of an intracranial hemorrhage with anticoagulation. That number has been estimated to be about 2% per year. 

The risk of stroke for individual patients has been based on clinical scoring systems. With the historical CHADS₂ scoring system, the threshold has been a score of  >1 to initiate OAC because the annual risk of stroke was at least 2%. Anticoagulation has been considered optional for patients with a CHADS₂ score of 1. The main impact on medical decision making of the more recently adopted CHA₂DS₂-VASc scoring system (congestive heart failure, hypertension, age >75, diabetes mellitus, prior stroke or transient ischemic attack, vascular disease, age 65 to 74, female) has been to parse patients who formerly were given a score of 0 or 1 using the CHADS₂ scoring system. For example, a 70-year-old woman with no other risk factors who had a CHADS₂ score of 1 now has a CHA₂DS₂-VASc score of 2, which gives her a solid indication for anticoagulation. The net effect is that more patients now have an indication for OAC. Furthermore, the relative risk of intracranial bleeding with the newer oral anticoagulants (NOACs) is 0.48 compared to warfarin.¹ Should the threshold risk of stroke to prescribe OAC be lowered to 1%? 

A major ingredient of the CHA₂DS₂-VASc scoring system is age, giving one point to a patient over age 65 years and two points to a patient over age 75. But does a person under age 65 have a low enough risk to withhold anticoagulation? A study by Shih-Ann Chen’s group in Taiwan recently looked at this issue.² His group used a large National Health Insurance Research Database in Taiwan to study almost 200,000 patients with AF and a CHA₂DS₂-VASc score of 0 or 1 who were not treated with OAC, and found that the annual risk of ischemic stroke was 1.78%/year for those >50 years of age compared to 0.53%/year for those <50 years of age. In a subgroup analysis, the annual risk of ischemic stroke for males and females aged 50 to 54 years was 1.47% and 1.07%, respectively. They concluded, “for Taiwanese patients 50 to 64 years of age, the annual stroke risk was 1.78%, which may exceed the threshold for OAC use for stroke prevention. The annual risk of ischemic stroke for AF patients <50 years of age was 0.53%, which was truly low-risk, and OACs could be omitted.” It is not clear if these data are unique to the Taiwanese, but it suggests that patients younger than age 65 should be considered for OAC.

There have been three developments in the last few years that should lead to a lower threshold to treat patients who have AF with OAC: replacement of the CHADS₂ stroke risk scoring system with the CHA₂DS₂-VASc scoring system, availability of NOACs that are half as likely to cause intracranial bleeding compared to warfarin, and increasing data to suggest that the age cutoff of 65 years to trigger a prescription for OAC may be too high.

References

1. Ruff CT, Giugliano RP, Braunwald E, et al. Comparison of the efficacy and safety of new oral anticoagulants with warfarin in patients with atrial fibrillation: a meta-analysis of randomised trials. Lancet. 2014;383(9921):955-962.
2. Chao TF, Wang KL, Liu CJ, et al. Age threshold for increased stroke risk among patients with atrial fibrillation: a nationwide cohort study from Taiwan. J Am Coll Cardiol. 2015;66:1339-1347.