Physicians' Academy for Cardiovascular Education

First 30 days after warfarin initiation come with higher ischaemic stroke risk

Literature - Azoulay et al. Eur Heart J. 2013 - Eur Heart J. 2013 Dec 18


Initiation of warfarin in patients with atrial fibrillation: early effects on ischaemic strokes

Azoulay L, Dell'aniello S, Simon TA et al.
Eur Heart J. 2013 Dec 18. [Epub ahead of print]


An increased risk of thromboembolic events was observed in the first 30 days after atrial fibrillation (AF) patients were transitioned from blinded rivaroxaban to open label warfarin at the end of the ROCKET-AF study [1]. Patients who switched from rivaroxaban, a factor Xa inhibitor, to warfarin had an over threefold risk (HR: 3.71, 95%CI: 1.51-9.16) of ischaemic stroke or systemic embolism  as compared to patients who transitioned from blinded warfarin to open label warfarin [2]. It was suspected that discontinuation of rivaroxaban may increase the risk of ischaemic stroke [3], but a post hoc analysis did not reveal an increased risk in patients who temporarily or permanently discontinued rivaroxaban during the ROCKET-AF trial [2].
Similarly, in the ARISTOTLE trial, an increased risk of ischaemic stroke was seen in the first 30 days after patients randomised to apixaban were transitioned to open label warfarin (4.02% per year vs. 0.99% per year for switch from blinded to open label warfarin) [4].
Thus, there may be an increased risk of ischaemic stroke early after warfarin initiation. Due to differential depletion of certain vitamin K-dependent clotting factors, it has indeed been appreciated that warfarin can theoretically lead to a transient hypercoagulable state at the start of treatment [5-7].
Benefitting from the Clinical Practice Research Datalink (CPRD), a UK primary care database, this study aimed to determine whether the initiation of warfarin is associated with an increased risk of ischaemic stroke in patients with AF. Data of 70766 patients newly diagnosed with AF were included in a nested case-control analysis, amounting to 275 987 person-years of follow-up. Exposure to warfarin was assessed using an algorithm that simultaneously estimates warfarin exposure and therapeutic range and patients were assigned to one of eight mutually exclusive groups based on their exposure at index date.

Main results

  • Cases (those experiencing an ischaemic stroke during follow-up) were more likely to have higher CHADS2-scores, peripheral artery disease, and more often used antidepressants, antipsychotics and NSAIDs than controls. Controls on the other hand were more likely to be obese, and to have used ACE inhibitors and ARBs than cases. These differences were similar across all warfarin exposure groups.
  • In the first 30 days of warfarin initiation, AF patients had a 71% higher risk of ischaemic stroke than patients who did not use antithrombotic therapy (adjRR: 1,71, 95%CI: 1.39-2.12).
    Risk was highest in the first week of use, with a peak at day 3 (RR: 2.33, 95%CI: 1.50-3.61).
  • After 30 days, the risk of ischaemic stroke was significantly lower in patients who initiated warfarin, as compared to patients not on antithrombotic agents (31-90 days: adjRR: 0.50, 95%CI: 0.34-0.75, >90 days: adjRR: 0.55, 95%CI: 0.49-0.61).
  • History of stroke prior to cohort entry modified the studied association, in that previous stroke increased the risk of ischaemic stroke in the first 30 days after warfarin initiation (adjRR: 2.45, 95% CI: 1.72–3.49), but reduced that risk in patients who had an event  >90 days after start of warfarin treatment (adjRR: 0.54,95%CI: 0.43–0.68).


This population-based study shows that initiation of warfarin use in AF patients is associated with a nearly two-fold increased risk of ischaemic stroke in the first 30 days, as compared to not taking antithrombotic therapy. The highest risk was seen in the first week after warfarin initiation. After 30 days, warfarin use was associated with a decreased risk of ischaemic stroke. This paradoxical procoagulant effect of warfarin is biologically plausible, as it may transiently selectively deplete the andogenous anticoagulant protein C. This hypothesis needs to be studied. Also, it should be determined whether this initial risk can be reduced by a heparin bridging strategy.
Find this article on Pubmed


1. Patel MR, Mahaffey KW, Garg J, et al. Rivaroxaban vs. warfarin in nonvalvular atrial fibrillation. N Engl J Med 2011;365:883–891.
2. Patel MR, Hellkamp A, Lokhnygina Y, et al. Outcomes of discontinuing rivaroxaban compared with warfarin in patients with nonvalvular atrial fibrillation: analysis from the ROCKET AF Trial. Circulation 2012;125:2445–2447.
3. FDA approves Xarelto to prevent stroke in people with common type of abnormal heart rhythm. U S Food and Drug Administration. (31 July 2012).
4. Granger CB, Alexander JH, Hanna M, et al. Events after discontinuation of randomized treatment at the end of the ARISTOTLE trial. 33 ed. 2012. p. 685.
5. Freedman MD. Oral anticoagulants: pharmacodynamics, clinical indications and adverse effects. J Clin Pharmacol 1992;32:196–209.
6. Harrison L, Johnston M, Massicotte MP, et al. Comparison of 5-mg and 10-mg loading doses in initiation of warfarin therapy. Ann Intern Med 1997;126:133–136.
7. Crowther MA, Ginsberg JB, Kearon C, et al. A randomized trial comparing 5-mg and 10-mg warfarin loading doses. Arch Intern Med 1999;159:46–48.