Novel oral anticoagulants (NOACs) are increasingly being used for the prevention of embolic events in patients with atrial fibrillation. These drugs inhibit the activation of thrombin (factor IIa) or factor Xa; thus, they prevent the propagation of the coagulation cascade and the formation of thrombus. The Randomized Evaluation of Long-Term Anticoagulation (RE-LY) trial was the first mega-trial evaluating the efficacy and safety of these drugs in patients with atrial fibrillation.¹ This trial randomized patients with non-valvular atrial fibrillation to therapy with either warfarin, dabigatran 110 mg twice-daily, or dabigatran 150 mg twice-daily, and showed that treatment with dabigatran 110 mg twice-daily was non-inferior to warfarin in reducing the incidence of stroke or systemic embolization (hazard ratio [HR] 0.91, 95% confidence interval [CI] 0.74-1.11, p-value (non-inferiority) <0.001) and the 150 mg twice-daily reduced the incidence of stroke or systemic embolization (HR 0.66, 95% CI 0.53-0.82, p-value (superiority) <0.001). In addition, patients treated with dabigatran 150 mg had lower incidence of death due to vascular etiology (2.3% vs. 2.7%, HR 0.85, 95% CI 0.72-0.99, p=0.04). Similar trends were seen in the patients treated with dabigatran 110 mg although this difference was not statistically significant (2.4% vs. 2.7%, HR 0.90, 95% CI 0.77-1.06, p=0.21).

Interestingly, in the main publication of the RE-LY trial, patients treated with dabigatran 150 mg had an increased incidence of myocardial infarction (n=89) as compared with warfarin (n=63) (HR 1.38 95% CI 1.00-1.91 P=0.048). While not statistically significant, there was a similar trend seen in the 110 mg dabigatran arm (n=86) when compared with warfarin (HR 1.35, 95% CI 0.98-1.87; p=0.07) During the drug approval process, additional events were discovered which reduced the relative risk of myocardial infarction seen with dabigatran such that it was no longer statistically significant in either dabigatran treatment group, although numerically there were fewer cases of myocardial infarction with warfarin: warfarin 0.64% per year, dabigatran 110 mg 0.82% per year (HR 1.29 with dabigatran 110 mg vs. warfarin, 95% CI 0.96-1.75; p=0.07), and dabigatran 150 mg 0.81% per year (HR 1.27 with dabigatran 150 mg vs. warfarin; 95% CI 0.94-1.71, p=0.12). While the annual incidence of myocardial infarction was very low, this raised questions regarding the cardiovascular safety of the NOAC.² Similar studies of other NOACs, such as the ROCKET AF trial, which randomized patients with non-valvular atrial fibrillation to rivaroxaban 20 mg once-daily or matching placebo, and the ARISTOTLE trial, which randomized patients with non-valvular atrial fibrillation to apixaban 5 mg twice-daily or matching placebo, did not see similar trends toward an increase in myocardial infarctions (Table 1).^{3, 4}

Following these findings, Uchino and Hernandez published a meta-analysis of all dabigatran trials (n=7) that reported myocardial ischemic events.⁵ This meta-analysis included trials studying the drug for a variety of conditions and thus the comparator arm differed across many trials (e.g., placebo, low dose enoxaparin, warfarin). With the exception of the RE-LY trial, none of the individual 7 studies included in this meta-analysis observed a statistically significant excess of myocardial infarction with dabigatran. After incorporating the results of all 7 studies, dabigatran was associated with increased risk of myocardial infarction (odds ratio [OR] 1.33, 95% CI 1.03-1.71; p=0.03). It should be noted that the RE-LY trial randomized more patients than all of the other trials combined; therefore, the results of the meta-analysis were largely driven by the results seen in RE-LY. However, an additional meta-analysis excluding the RE-LY trial continued to find an association between dabigatran and myocardial infarction.⁶

These findings have led to considerable debate in the cardiology and regulatory communities as to whether or not dabigatran increases the risk of myocardial infarction. The RE-LY trial was a large (N=18,113), well-conducted study using a PROBE (Prospective Randomized Open, Blinded End-point) design in which an independent clinical events committee adjudicated all myocardial infarctions without knowledge of treatment assignment. Furthermore, a small (N=1,861), placebo-controlled, phase 2 dose escalation trial that evaluated dabigatran in patients after an acute coronary syndrome (ACS) found a numerical increase in the number of patients treated with dabigatran who had a non-fatal myocardial infarction, raising the possibility that dabigatran itself may increase the risk of myocardial infarction.⁷

There are multiple reasons to question these findings. First, there is no clear biological plausibility that would explain why treatment with an anticoagulant would increase the risk of myocardial infarction. To the contrary, patients with ACS benefit from treatment with heparin, and trials in patients stabilized after an ACS such as ATLAS ACS 2-TIMI 51 and WARIS II have found that treatment with an oral anticoagulant reduces the rate of future myocardial infarction. Similarly, other NOACs studied in patients with atrial fibrillation and post myocardial infarction (including the direct thrombin inhibitor ximelegatran) have shown either a neutral or even a protective effect with respect to the risk of subsequent myocardial infarction.⁸ In addition, similar increases in vascular ischemic events were not seen with other broader endpoints analyzed in the RE-LY trial. For example, the composite endpoint of cardiac death, cardiac arrest, myocardial infarction, or unstable angina was similar in dabigatran and warfarin (dabigatran 110 mg compared with warfarin, HR 0.93, 95% CI 0.80-1.06; dabigatran 150 mg compared with warfarin, HR 0.98, 95% CI 0.85-1.12).² Furthermore, there was no evidence with a dose response relationship seen between dabigatran and the risk of myocardial infarction. Finally, these findings have not been supported by observation studies of the use of dabigatran in clinical practice. Larsen et. al studied patients in Denmark and found that treatment with dabigatran was associated with lower incidence of myocardial infarction as compared with warfarin.⁹

If dabigatran is not thought to increase the risk of myocardial infarction, how does one explain the findings seen in RE-LY and subsequent meta-analyses? One cannot discount the possibility that this simply reflects a chance finding. This finding occurred in the setting of a well-conducted clinical trial with rigorous follow-up; however, there were multiple endpoint comparisons performed. Without adjustment for these multiple comparisons, the likelihood of a Type I (false positive) finding is increased. These findings may also be due to the relative efficacy of warfarin as compared with dabigatran in the prevention of myocardial infarction. It is well established that anticoagulation with warfarin reduces ischemic events following ACS.¹⁰ The increased incidence of bleeding as opposed to a lack of efficacy has prevented warfarin from becoming an established therapy following ACS. Thus, it is possible that the findings seen in RE-LY simply reflect that warfarin is relatively more effective than dabigatran in reducing myocardial infarctions, and does not indicate that dabigatran actually increases the risk of myocardial infarction.

There are several implications of these findings for clinicians as they consider the use of NOACs in patients at risk for cardiovascular events. While it is possible that dabigatran may be associated with a higher incidence of myocardial infarction when compared with warfarin, observational studies of dabigatran in a clinical setting have not found an association. There are no indications that NOACs, as a class, increase the risk of myocardial infarction. In patients at high risk of coronary events, preference may be given to the use of a NOAC other than dabigatran; however, it appears that the risk of myocardial infarction with dabigatran, if present at all, is likely very small.

Table 1: Myocardial Infarction in Mega Trials of Novel Anticoagulants
	NOAC		Warfarin
	Total Events	Event Rate (%/yr)	Total	Event Rate (%/yr)	Hazard Ratio	P-Value
RE-LY
Dabigatran 110 mg	98/6015	0.82	75/6022	0.64	1.29 (0.96-1.75)	0.09
Dabigatran 150 mg	97/6076	0.81	75/6022	0.64	1.27 (0.94-1.71)	0.12
ROCKET AF	101/7131	0.91	126/7133	1.12	0.81 (0.63-1.06)	0.12
ARISTOTLE	90/9120	0.53	102/9081	0.61	0.88 (0.66-1.17)	0.37

References

1. Connolly SJ, Ezekowitz MD, Yusuf S, et al.: Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med 2009;361:1139-1151.
2. Hohnloser SH, Oldgren J, Yang S, et al.: Myocardial ischemic events in patients with atrial fibrillation treated with dabigatran or warfarin in the RE-LY (Randomized Evaluation of Long-Term Anticoagulation Therapy) trial. Circulation 2012;125:669-676.
3. Patel MR, Mahaffey KW, Garg J, et al.: Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med 2011;365:883-891.
4. Granger CB, Alexander JH, McMurray JJ, et al.: Apixaban versus warfarin in patients with atrial fibrillation. N Engl J Med 2011;365:981-992.
5. Uchino K, Hernandez AV: Dabigatran association with higher risk of acute coronary events: meta-analysis of noninferiority randomized controlled trials. Arch Intern Med 2012;172:397-402.
6. Sipahi I, Celik S, Akyol A: Dabigatran's 'Real-World' Data About Risk of Myocardial Infarction and Gastrointestinal Bleeding Contradicts With Randomized Trials. J Am Coll Cardiol 2013;62:945-946.
7. Oldgren J, Budaj A, Granger CB, et al.: Dabigatran vs. placebo in patients with acute coronary syndromes on dual antiplatelet therapy: a randomized, double-blind, phase II trial. Eur Heart J 2011;32:2781-2789.
8. Wallentin L, Wilcox RG, Weaver WD, et al.: Oral ximelagatran for secondary prophylaxis after myocardial infarction: the ESTEEM randomised controlled trial. Lancet 2003;362:789-797.
9. Larsen TB, Rasmussen LH, Skjoth F, et al.: Reply: Dabigatran's 'Real-World' Data About Risk of Myocardial Infarction and Gastrointestinal Bleeding Contradicts With Randomized Trials. J Am Coll Cardiol 2013;62:946-947.
10. Hurlen M, Abdelnoor M, Smith P, et al.: Warfarin, aspirin, or both after myocardial infarction. N Engl J Med 2002;347:969-974.

Keywords: Anticoagulants, Atrial Fibrillation, Benzimidazoles, Blood Coagulation, Confidence Intervals, Factor Xa, Incidence, Prothrombin, Stroke, Thrombin, Thrombosis, Warfarin

< Back to Listings