Increased risk of acute MI on DOAC vs. VKA in real world AF patients

Risk of myocardial infarction in patients with atrial ﬁbrillation using vitamin K ant agonists, aspirin or direct acting oral anticoagulants

Literature - Stolk LM, De Vries F, Ebbelaar C et al., - Br J Clin P harmacol 2017. DOI: 10.1111/bcp.13264

Background

Multiple studies have shown that direct-acting oral anticoagulants (DOACs) are either non-inferior, or possibly superior for some outcomes, to warfarin in the prevention of stroke and thromboembolic events in patients with atrium fibrillation (AF). In the RE-LY trial, the risk of acute myocardial infarction (AMI) was elevated in AF patients randomised to dabigatran, as compared to those on warfarin [1]. A posthoc analysis of revised data from the RE-LY trial did not confirm this finding [2]. Meta-analyses of randomised non-inferiority trials have also resulted in conflicting findings on the risk of AMI with DOAC use [3-7]. Other analyses have indicated a higher risk of AMI with the factor IIa inhibitor dabigatran, as compared with factor Xa inhibitors rivaroxaban, apixaban and edoxaban [8, 9]. Observational cohort studies have also yielded conflicting results. In a recent phase IV study in patients using rivaroxaban, no increased risk of MI was seen in 1 year.

This study aimed to determine the risk of AMI in real-world patients with AF, using three different classes of antithrombotic agents: DOACS (both IIA and Xa inhibitors), VKAs and aspirin. Data of the UK primary care Clinical Practice Research Datalink (CPRD) were used.

Data of 30146 new users were analysed, among which 1266 users of DOACs (71.6% on rivaroxaban and 28.4% on dabigatran), 13098 of VKAs and 15400 of low-dose aspirin, and 382 mixed users at the index date. Follow-up time was about 1 year for DOAC and about 3 years for VKA or aspirin use. Mean age was approximately 72 years.

Main results

The division of CHA2DS2-VASc score categories among the three treatment groups was balanced, with 25% having a low score (≤1), 25-30% having a high score (≥4) and the remainder having a medium score between 1 and 4. The distribution of antithrombotic use did not follow guidelines for stroke prevention in AF.
The risk of AMI was increased when comparing current use of DOACs with current use of VKA (adj HR: 2.11, 95%CI: 1.08-4.12).
A similarly increased risk was seen when comparing current use of aspirin with VKA (adjHR: 1.91, 95%CI: 1.45-2.51).
When stratifying by gender, among aspirin users, men showed a 1.60 times increased risk of AMI (95%CI: 1.10-2.33) and women a 2.33-fold higher risk (95%CI: 1.55-3.50).
When stratifying current users by CHA2DS2-VASc score at index date, among those with a high score, aspirin users showed a higher risk than VKA users (adjHR: 2.21, 95%CI: 1.37-3.55), while among those with a medium score, as compared with VKA users, an increased risk was seen in current DOAC users (adj HR: 2.67, 95%CI: 1.11-6.40) and in aspirin users (adjHR: 1.82, 95%CI: 1.23-2.68).

Conclusion

This analysis shows that in real-world patients with AF who are currently treated with DOACs or aspirin, the risk of AMI was doubled as compared with current users of VKA. After stratifying for gender, a significantly increased risk was only seen for users of aspirin. When stratifying for stroke risk based on CHA2DS2-VASc score, the risk of AMI remained significantly increased in high and medium risk patients on DOAC and aspirin, as compared with current users of VKAs.

References

Show references

Find this article online at Br J Clin Pharmacol