Sir,—We welcomed the excellent editorial by Waktare and Camm on the deleterious effects of atrial fibrillation and agree wholeheartedly with their conclusion that “the important principle is to investigate and treat appropriately from the outset”.1 There is little doubt that without proper investigation and treatment, atrial fibrillation begets trouble.

While it is well recognised that there are cardiovascular complications associated with atrial fibrillation such as heart failure, thromboembolism, and stroke, and that there are well validated treatment strategies to reduce the occurrence of stroke and thromboembolism by appropriate use of antithrombotic therapy, the message has failed to get through to many clinicians in hospital and general practitioners.

Much of the clinical epidemiology of atrial fibrillation in the United Kingdom has been criticised for being based on small and elderly populations that are unrepresentative. While there is need for more information on the prevalence of atrial fibrillation in Britain and the treatment and investigations of such patients, there have been some studies in this area. In a recent study on the use of anticoagulants among patients with atrial fibrillation in the community from Newcastle,2 only 44% of patients with atrial fibrillation aged 65 to 74 years, and 11% of patients over 75 years were treated with warfarin. In the same study, only 33% of the subjects in the 65–74 year age group without contraindications to treatment were actually treated with warfarin, and 14% of the over 75 age group. Our survey of atrial fibrillation in two general practices in West Birmingham3 broadly agrees with these findings. For example warfarin was prescribed to only 36% of the 111 patients with atrial fibrillation and of those not anticoagulated (n = 71), only 12 patients (17%) had significant contraindications to warfarin treatment. While aspirin is often considered as an alternative to warfarin, it was prescribed in only 19% of patients, primarily for established vascular disease. Similar low rates of antithrombotic treatment have been found among patients with atrial fibrillation in hospital,4-6 but our general practice survey suggested that less than a third of the patients had ever been admitted to hospital.3 Such information from hospital and general practice would have major implications for health care resources and service provision for this common problem.

The low use of anticoagulation in patients with atrial fibrillation may be related to the perceived absence of suitable guidelines; however, many such guidelines for the treatment of atrial fibrillation do exist.7 8 Despite this there remains considerable variation among physicians in the management of patients with atrial fibrillation, especially between cardiologists and non-cardiologists, in the use of antithrombotic and antiarrhythmic therapy, and consideration for cardioversion.9 The existence of many different guidelines would probably result in a very wide range on the actual rates of anticoagulation if applied to the same population of patients with atrial fibrillation.

If we improve screening, detection, and anticoagulation for atrial fibrillation, these have considerable implications especially with an aging population. A question often raised is who should be responsible for monitoring anticoagulant therapy? There is clear evidence that general practitioners can monitor anticoagulation intensity more efficiently than hospital anticoagulant clinics, however, many general practitioners are reluctant to undertake this role.10 In our general practice survey, anticoagulation was monitored in hospital in the majority of cases (75%), by both general practitioner and in hospital in 17.5%, and by general practitioner alone in only 7.5%.9 11 This issue requires further clarification and possible solutions, such as decision support for dosing and self-monitoring, need further evaluation.12

While many hospital clinicians and general practitioners are aware of atrial fibrillation, its associated problems, and the need for treatment, the message from many studies is that the management of atrial fibrillation remains suboptimal. Many of us would welcome any suggestions for implementation of proper investigations and treatment for patients with atrial fibrillation. Do we need more guidelines? Probably not; however, we do need a consensus plan involving our general practitioner and hospital physician colleagues in the detection and management of this common problem. We, as cardiologists, keep emphasising the need for managing this problem appropriately, but the evidence of much variation in management, even among cardiologists, suggests that much more work is needed before we can deliver appropriate care to all patients with atrial fibrillation.