Atrial fibrillation is commonly associated with stroke and thromboembolism. When stroke occurs in association with atrial fibrillation, patients have a greater mortality and morbidity, longer hospital stays, and greater disability than those without atrial fibrillation.
Pooled data from trials comparing antithrombotic treatment with placebo have shown that warfarin reduces the risk of stroke by 62% (95% confidence interval 48% to 72%) and that aspirin alone reduces the risk by 22% (2% to 38%). Overall, in high risk patients, warfarin was better than aspirin in preventing strokes, with a relative risk reduction of 36% (48% to 72%). The risk of major haemorrhage with warfarin was twice that with aspirin.w7
Anticoagulation treatment needs to be tailored individually for patients on the basis of age, comorbidities, and contraindications. We have incorporated risk factors for thromboembolism, derived from major clinical trials,w8-w15 and proposed a simple algorithm for selecting the most appropriate anticoagulant for patients in atrial fibrillation (). For patients with lone atrial fibrillation and no thromboembolic risk factors, the annual rate of stroke is 1%, and these patients seem to have little to gain from anticoagulation.w8 Such patients may be prescribed aspirin if no contraindications exist.
Fig 3 Algorithm for anticoagulating patients with atrial fibrillation. *Thyrotoxicosis is associated with a high thromboembolic risk in atrial fibrillation. Current guidelines recommend anticoagulation with warfarin, if no contraindications exist, at least (more ...)
Warfarin remains underprescribed in clinical practice. A review of more than 20 studies of patients with atrial fibrillation showed that only 15-44% of eligible patients were actually prescribed warfarin.w15 Elderly patients are often denied anticoagulation owing to a presumed increased risk of haemorrhagic complications. For such patients, who have a high thromboembolic risk, current data suggest that the benefits of anticoagulation may be greater than in other patients. Indeed, more than 50% of patients with atrial fibrillation in the community are aged over 75,w16 but existing trial data on the effectiveness of warfarin have been derived from selected secondary care populations that under-represent elderly people. The proposed BAFTA trial will assess the risks and benefits of aspirin versus warfarin exclusively in elderly patients with atrial fibrillation in a primary care setting.w17
Warfarin has a narrow therapeutic window and, with its known drug and food interactions, needs regular monitoring. This can be inconvenient for both patients and physicians. Community based studies have shown that patients receiving warfarin have international normalised ratio (INR) values at non-therapeutic levels more than half the time.w18 However, patients who have access to anticoagulation clinics have INR values within the therapeutic range more than 60% of the time.w18 w19 Similarly, monitoring of anticoagulation at clinics has been shown to maintain therapeutic INR values in both elderly patients and younger populations (71.5% and 66.1% respectively, with no statistical difference).w20 Anticoagulation clinics have improved the quality of anticoagulation within the community, and with further research a greater degree of anticoagulation control may be feasible. On the other hand, self monitoring of INR at home may improve the quality and convenience of anticoagulation monitoring. The HOME INR study is comparing this home self monitoring regimen with conventional anticoagulation monitoring at clinics.w21 However, this may be suitable for a only small proportion of patients, considering the age and associated comorbidities of patients with atrial fibrillation.
The limitations of warfarin treatment have prompted the development of new anticoagulants with predictable pharmacokinetics, such that monitoring is unnecessary (table B on bmj.com
). Ximelagatran, an oral direct thrombin inhibitor, may be a useful alternative to warfarin. The SPORTIF III and SPORTIF V trials comparing ximelagatran with warfarin found the two agents to be broadly similar.w22 w23
However, approximately 6% of patients on ximelagatran had abnormal liver function tests, so monitoring of liver function will be an important factor.
Idraparinux, a factor Xa inhibitor administered by once weekly subcutaneous injections, is being evaluated in patients with atrial fibrillation in the AMADEUS trial.w24 Oral factor Xa inhibitors are also in development as potential oral anticoagulants.w25 ACTIVE (atrial fibrillation clopidogrel trial with irbesartan for prevention of vascular events) is assessing the role of aspirin plus clopidogrel, compared with adjusted dose warfarin, in the prevention of vascular events in patients with atrial fibrillation.w26