ABC of Atrial Fibrillation: ATRIAL FIBRILLATION IN GENERAL AND HOSPITAL PRACTICE

General practice

Despite the considerable interest in atrial fibrillation in epidemiological and hospital studies, little information exists on the prevalence and management of atrial fibrillation in general practice.

In a recent audit of a general practice list of about 10000 patients, we found 67 patients who were currently in atrial fibrillation or who were known to have had past episodes of the arrhythmia.

This prevalence increased with age, rising from 1.5% in people in their 60s to 8% in those aged over 90. A third of the 67 patients had paroxysmal atrial fibrillation. Only two thirds were currently receiving anticoagulant treatment, and an echocardiogram had been obtained in only a third. Of the 30 patients under 75 years, 10 had mitral valve disease, while only one patient aged over 75 had mitral valve disease.

From the stroke prevention in atrial fibrillation study, major clinical risk factors for stroke and thromboembolism were congestive heart failure, hypertension, and previous thromboembolism. We found that at least one of these risk factors was present in three quarters of patients seen in general practice with atrial fibrillation (and in 84% of those aged over 75).

Prevalence of atrial fibrillation by age in a general practice population.

• Download figure
• Open in new tab
• Download powerpoint

The general population is aging—that is, the proportion of elderly people is increasing—and by the year 2000 the proportion of people over 70 in Britain will constitute 20-25% of the total population. Atrial fibrillation will therefore be an increasingly common cause of stroke, thromboembolism, and heart failure, emphasising that atrial fibrillation is an important public health problem. Many patients will also be taking antiarrhythmic drugs and anticoagulants.

As many elderly patients have other medical conditions, the problem of side effects and drug interactions will become increasingly common. For example, hypokalaemia secondary to the use of high doses of loop diuretics may result in dangerous arrhythmias in patients taking digoxin. The use of antiarrhythmics such as amiodarone may result in drug interactions with warfarin and digoxin with consequent toxicity.

Such possibilities emphasise the need for concurrent treatments to be fully recorded and for a high index of suspicion for drug interactions. Non-compliance by patients because of inadequate understanding or supervision may also be dangerous. Careful monitoring of requests for repeat prescriptions is therefore essential.

The increasing prevalence of atrial fibrillation will result in more patients taking anticoagulant drugs. A high proportion of patients in general practice have risk factors for thromboembolism, which necessitate long term treatment with anticoagulants. This presents two problems: inconvenience and safety. Patients taking warfarin need regular monitoring of anticoagulation intensity and adjustment of dosage. Although this is often done at specialist anticoagulation clinics either in hospitals or in the community, this task is now increasingly being undertaken by some general practitioners. Many general practitioners, however, may be reluctant to take on this responsibility.

Increasing numbers of patients needing warfarin would cause considerable strain on the current provision of anticoagulation monitoring services, and new initiatives to provide anticoagulation in the community are urgently needed. Frail elderly people should not have to attend hospital clinics, which may be some distance from their homes. The safety considerations are related to the risk of haemorrhage and of drug interactions.

With the potentially increasing importance of atrial fibrillation as a cause of mortality or morbidity, the question arises of whether screening programmes would be cost effective? All general practitioners should at least have easy access to a reliable electrocardiogram machine to confirm atrial fibrillation; electrocardiography is easy to perform, although careful placement of the leads and interpretation is needed. Some difficulties, however, may arise in distinguishing atrial fibrillation from other supraventricular arrhythmias and in identifying pre-excitation syndromes.

Open access to general practitioners for echocardiography in local hospitals for patients with heart failure has been tried successfully in some centres, and there is a good case for extending this facility to the management of atrial fibrillation.

Standard positions for chest leads (top) and limb leads (bottom) for electrocardiography.

• Download figure
• Open in new tab
• Download powerpoint

The adequate treatment of atrial fibrillation, as of other chronic diseases, needs an effective channel of communication between hospital and general practice. Decisions on the advantages and disadvantages of treatment with anticoagulants need to be made jointly by the physician, the general practitioner, and the individual patient. A realistic assessment should be made of the burden placed on older patients by medical intervention. Some may prefer to be left alone, and this wish should be respected when possible, after discussion of the risks and benefits of treatment.

Hospital practice

As with general practice, the future increasing prevalence of atrial fibrillation will mean there will be more patients at increased risk of stroke and heart failure needing hospital care. In a Scottish hospital based study atrial fibrillation was present in 6.3% of emergency medical admissions. In our survey of admissions to a city centre district hospital in Birmingham covering a multiethnic population, atrial fibrillation was equally common in white, black (Afro-Caribbean), and Asian people. In both surveys this arrhythmia was associated with a pronounced morbidity and mortality from heart failure, stroke, and syncope.

Atrial fibrillation has important implications for the provision of investigations, such as echocardiography. Most cardiologists use this investigation for their patients with atrial fibrillation. In patients presenting with syncope, 24 hour Holter monitoring is advisable. Syncope is common in patients with atrial fibrillation, and the question therefore arises whether we can afford to investigate all patients with atrial fibrillation with so many detailed and expensive procedures.

Implications for more anticoagulation clinics

Patients taking warfarin as thromboprophylaxis need to attend anticoagulation clinics regularly so that the international normalised ratio can be monitored. Warfarin has potentially serious haemorrhagic side effects, particularly in elderly people, in whom peptic ulcers are common. Furthermore, many drugs have potentially serious drug interactions with warfarin. Cimetidine, for example, may inhibit the metabolism of warfarin, resulting in overanticoagulation. People taking warfarin would also need well designed literature, written in non-technical language, about what to do in the event of unusual symptoms or signs of bleeding. If such patients rely on ambulances this adds to the overall costs of treating such patients. Our hospital runs domiciliary anticoagulation checks, which minimise this expense.

Recent studies suggest that outpatient anticoagulation clinics may be less than satisfactory, with fewer than 50% of the results falling within the therapeutic range and nearly one third of patients being classed as “poorly controlled.” This may be due to the high workload and inadequate time available to communicate with patients. Patients may also have to travel a long way to the nearest clinic and so may not keep all their appointments.

One solution may be more monitoring of anticoagulation treatment by general practitioners. A recent report suggests that the degree of anticoagulation control among patients in the care of general practitioners was better than that achieved by hospitals. If the benefits of anticoagulation treatment are to be extended to more patients with atrial fibrillation, particularly to older patients at higher risk, more attention will have to be given to the degree of control achieved. Only if this approaches that achieved in the large clinical trials will there be the substantial reduction of stroke reported in these trials.

The future

The future management of patients with atrial fibrillation will need to address two important considerations. Firstly, new developments in thromboprophylaxis are needed so that patients with atrial fibrillation are inconvenienced as little as possible and receive prophylaxis with maximal efficacy in preventing strokes and thromboembolism. Large scale randomised studies exploring the use of very low intensity warfarin and aspirin and combinations of these are currently in progress.

Secondly, the ability to identify patients with atrial fibrillation who are at high risk of thromboembolic complications would be an important advance. Although clinical risk factors for stroke have been identified for patients with atrial fibrillation, transthoracic echocardiography may further refine risk stratification. Recently, transoesophageal echocardiography has been similarly used—for example, in the risk stratification of such patients before cardioversion. Patients with atrial fibrillation have also been found to have a hypercoagulable or prothrombotic state, with abnormalities of indices of thrombogenesis. Clearly, further study of ways to stratify the risks of patients with atrial fibrillation is needed.