The thrombotic risk of non-valvular AF

Atrial fibrillation

The most common sustained cardiac arrhythmia is atrial fibrillation

Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia that a clinical practioner will have to deal with. The prevalence of atrial fibrillation increases with increasing age. AF increases the risk of an embolic stroke five-fold, irrespective of whether the AF is chronic or paroxysmal and recently it was realised that up to 30% of cryptogenic strokes may be caused by AF.

AF also increases the risk of systemic embolism apart from stroke. Strokes caused by AF are associated with about a 50% increased risk of disability and a 60% increased risk of death as compared to strokes of other aetiologies. Strokes related to AF result from cardio-embolism of a large cerebral artery and therefore tend to be larger than other strokes. However, strokes related to AF are largely preventable because oral anticoagulants (OAC) are so effective, which includes vitamin K antagonists (VKA) such as warfarin and the newer non-VKA oral anticoagulants (NOACs) of which there are two available in SA – dabigatran and rivaroxaban. The most important risk factors for non-valvular stroke are: Prior stroke or transient ischaemic attack (TIA), increasing age, hypertension and diabetes mellitus. Women seem to have a higher stroke risk.

Effect of aspirin

A meta-analysis of aspirin to reduce the risk of stroke in AF did not show any benefit. The relative risk reduction (RRR) was 19% but the confidence interval (CI) went from -1 to 35, thus crossing the ‘1’ line, making the result non-significant. The bleeding risk of aspirin was considerable with major bleeding and intracranial bleeding not different between OAC and aspirin. Adverse events occurred more frequently among the very elderly who were taking aspirin.

Effect of warfarin

Stroke or systemic embolism is significantly reduced by VKA (warfarin): RRR 64% (CI: 49%-74%) as compared to placebo or no treatment. The absolute risk reduction is 2.7% per year which translates to a number-needed-to-treat (NNT) of 37 for embolism inpatients per year with no history of prior stroke (primary prevention). In patients with a prior history of stroke (secondary prevention) warfarin reduced stroke by absolute reduction of 8.4% with NNT of only 12, which is highly effective. Warfarin also significantly reduced mortality by a RRR of 26% (95% CI:3-43%) with absolute risk reduction of 1.6% per year, NNT 63.

The problem with warfarin is that it has a narrow therapeutic index that necessitates frequent monitoring with dose adjustments, slow onset of action and slow decline in effect when drug is stopped, poor patient adherence and many food and drug interactions.

Effect of NOACs

A meta-analysis of 4 phase 3 trials comparing the efficacy and safety of the new oral anticoagulants with warfarin for stroke prevention in patients with AF has been published. There were 42 411 patients receiving a new oral anticoagulant as compared to 29 272 patients with AF receiving warfarin. The underlying risk for stroke differed significantly between trials with CHADS2 scores varying from 3-6. Median time of follow-up ranged from 1.8 years to 2.8 years. Stroke or systemic embolism were significantly reduced by NOACs as compared to warfarin: Pooled effect with a random-effect model showed a RRR of 19% (95% CI: 9-27%) by NOACs as compared to warfarin. Haemorrhagic stroke was reduce by 51% (95% CI: 36-62%) in the NOAC group as compared to warfarin. Intracranial haemorrhage was equally significantly reduced by NOAC by RRR 52% (95% CI: 41%-61%). Major bleeding did not show any significant difference. The NOACs did not significantly reduce ischaemic strokes as compared to warfarin. All-cause mortality was also significantly reduced by NOACs with RRR 10% (95% CI: 5-15%) as compared to warfarin.

Since these randomised clinical trials there has been quite a number of observational real-world studies testing effectiveness in large numbers of patients as seen in clinical practice demonstrating the safety of NOAC drugs as the risk of major bleeding in the community was not significantly different from that of warfarin but with better efficacy profile, stroke reduction and mortality reduction.

An example of such a study is the XANTUS study, which was an international (311 centres in Europe, Israel and Canada) observational study evaluating the use of rivaroxaban 20mg daily in non-valvular atrial fibrillation and 15 mg daily in similar patients but with moderate to severe renal impairment. There were 6 784 patients observed with mean age 71.5 years of age and with CHADS2 score of 2. Safety data of rivaroxaban in this observational study showed a risk of major bleeding of 2.1 events per 100-patient-years as compared to 3.6 events per 100 patient-years in the ROCKET AF randomised trial of rivaroxaban.


  1. AF carries a high risk of stroke that is more devastating than stroke due to other causes.
  2. Warfarin significantly reduces the risk of stroke in AF.
  3. The new oral anticoagulants (NOACs) reduce stroke and mortality significantly more than warfarin.
  4. There are randomised clinical trials demonstrating the efficacy of NOACs, such as the ROCKET AF trial on rivaroxaban.
  5. Observational studies have also shown better adherence to NOACs compared to warfarin, particularly in atrial fibrillation patients with ≥ 2 stroke risk factors. Could this lead to better outcomes?
  6. NOACs are a viable alternative to warfarin in treating patients with non-valvular atrial fibrillation.



Author: Prof James Ker

References available on request.

, , , , , , , , , , , , , , ,

No comments yet.

Leave a Reply