Need to know: AtriaFibrillation
With AF worth 30 QOF points from April, Dr Timothy Watson and Professor Gregory Lip answer GP Dr Neil Statham's questions
With AF worth 30 QOF points from April, Dr Timothy Watson and Professor Gregory Lip answer GP Dr Neil Statham's questions
1. What are the causes of AF?
It used to be considered that IHD was the principal cause but now it appears hypertension is a major factor. It is true that ischaemic heart disease is one of the leading causes of AF. But studies (especially Framingham) have shown numerous factors including diabetes, hypertension, valvular disease (in fact all structural heart disease) may enhance the risk for AF.
A recent British survey of acute hospital admissions for AF suggested ischaemic heart disease was implicated in a third of patients, while hypertension contributed to around 26 per cent of AF cases. One study in primary care found that one-third of patients with AF have not required hospital care, so hospital-based studies may give a clinical epidemiology picture that relates more to patients who are more ill or have more complications related to heart disease. The mechanism of why the hypertension link is so strong is still unclear.
Hypertension also contributes to coronary artery disease, and hence, AF. It should not be forgotten that alcohol is an important contributor to hypertension and new onset AF.It is important to remember there is a good association between AF and non-cardiac disease, including lung pathology, excess alcohol, thyrotoxicosis and perioperatively. These should be screened for and treated. Overall, so many conditions are implicated in the pathogenesis of AF, it is perhaps safer to describe the patient's cumulative risk factors than look for a single cause.
2. Do patients with paroxysmal AF need anticoagulation?
Paroxysmal AF carries much the same risk of stroke as permanent AF. It has been postulated that this risk is dependent on the frequency and duration of these paroxysms, but there is little evidence to support this.
Indeed, not all patients get symptomatic paroxysms and it may be difficult to judge how frequently and long paroxysms last without clear documentary evidence of this, as provided by Holter monitoring. Furthermore, patients with paroxysmal AF tend to be younger and are less likely to have associated risk factors for AF so it is less likely an easily treatable precipitant will be found.
There is limited data on thromboprophylaxis in this subgroup of patients. However, evidence suggests anticoagulation is equally effective in preventing ischaemic stroke in the presence of risk factors that put such patients at moderate-high risk. Indeed, given that there remains a significant risk of thromboembolic events, it would be advisable to follow the same stroke risk stratification protocol as that for permanent AF.
3. Are there any patients we should not consider referring for anticoagulation?
The risk of stroke and thromboembolism in AF is high. The association continues to strengthen with age and in those with valvular or other cardiovascular pathology. Furthermore, those who have a stroke caused by AF have a higher morbidity, mortality, longer hospital stays and greater disability.
Adjusted-dose warfarin has been shown to significantly reduce the risk of stroke and thromboembolism in AF. The use of warfarin results in a relative risk reduction of two-thirds compared with placebo. However, it must be recognised that anticoagulation has pitfalls: an increased risk of major haemorrhage, interaction with various drugs, lifestyle implications (restriction of alcohol and the necessity to attend for therapeutic monitoring).
Because of this, each patient must be individually assessed. Most clinicians agree it would be unwise to offer anticoagulation to those with frequent falls, those who are unlikely to attend for monitoring or those with a tendency to bleed significantly.
The following risk factors increase the bleeding risk in AF:
- age over 75
- concomitant use of antiplatelet agents (aspirin, clopidogrel) or NSAIDs
- uncontrolled BP
- history of bleeding/intracranial haemorrhage
- history of poor INR control.
4. If a patient can't be anticoagulated, how effective is aspirin?
Often aspirin is offered in the hope that at least some stroke prevention will be offered. But what is the evidence? In one meta-analysis of the six main randomised trials of aspirin versus placebo, aspirin significantly reduced the risk of stroke by 22 per cent. Unfortunately, this figure is probably skewed by results from one trial, the SPAF-I (Stroke Prevention in AF) trial, although aspirin was less useful in patients aged under 75 and did not prevent severe strokes.
Other population-based studies have shown strokes that occur in patients while on aspirin tend to be more severe, with greater inpatient mortality and disability. If the above prescribing guidelines dictate that adjusted-dose warfarin would be unsafe, or if the patient prefers it, aspirin 75-300mg daily should be prescribed.
5. Is digoxin still the most appropriate therapy for rate limitation?
Traditionally, one of the cardiac glycosides (typically digoxin) has been used for limiting ventricular rate in AF. However, these drugs are of limited efficacy in thyrotoxicosis, fever, perioperatively and, most importantly, during exertion. Furthermore, rate control with mono-therapy is often inadequate and combination therapy with either a ?-blocker or non-dihydropyridine calcium-channel blocker may be required.
Current recommendations for rate control in permanent AF therefore advise the use of ?-blocker or non-dihydropyridine calcium-channel blocker (verapamil, diltiazem) for initial rate control, although digoxin may be adequate in the elderly, sedentary patient. Where monotherapy does not adequately control rate, addition of digoxin to the ?-blocker or calcium channel blocker may be useful.
Where drug therapy fails, one option would be referral to a specialist for consideration of amiodarone or atrioventricular nodal ablation and pacemaker insertion.
6. When are ?-blockers indicated??-blockers, very effective at controlling ventricular rate in permanent AF, are increasingly being used first-line, both acutely for rapid rate control, and also in the outpatient setting for optimising ventricular rate in those who remain tachycardic at rest.Various intravenous ?-blockers (eg esmol-ol) are also reasonably effective in terminating some episodes of AF and additionally offer protection against recurrence in particular after pharmacological or electrical cardioversion of persistent AF, as prophylaxis in paroxysmal AF, perioperatively or when AF is associated with thyrotoxicosis.As AF commonly coexists with hypertension or heart failure due to systolic function, ?-blockers may be part of the therapeutic management in such patients.7. Are there any other drugs we could use to limit the heart rate?In some patients there may still be suboptimal control or first-line drugs may be contraindicated. It is reasonable to consider amiodarone, but specialist referral is advis-ed because of its broad spectrum of toxicity. Patients should be warned of side-effects in particular photosensitivity and monitored for thyroid dysfunction.Amiodarone has a long half-life and takes a while to achieve a steady-state plasma concentration, particularly if loaded orally: it increases the bioavailability of digoxin so it is important to halve the dose of digoxin.8. At what heart rate should we consider rate-limiting therapy? This is still the subject of significant debate. The onset of AF reduces cardiac output by as much as 10-20 per cent regardless of ventricular rate. However, the addition of an uncontrolled ventricular response reduces cardiac output further still, occasionally predisposing to heart failure. Clearly optimal cardiovascular status is best achieved by restoration of sinus rhythm, but this is not always feasible or successful and in these circumstances the ventricular rate should be controlled.It is generally accepted that a resting heart rate below 90bpm is optimal, while on exertion the heart rate should not exceed 110 in the sedentary patient or 200 minus age in the ambulatory patient. This often requires combination rate-limiting drugs. 9. How effective is DC cardioversion? Electrical cardioversion is not always successful. In some cases, there is complete shock failure, while in others the patient reverts back to AF within a few minutes. It is estimated that this occurs in around 25 per cent of patients.In other patients, although cardioversion is initially successful, there is early recurrence and around another 25 per cent will revert to AF in the first two weeks after the cardioversion attempt.It is routine clinical practice to review patients at around four-weeks post-cardioversion. For those who remain in AF, many are offered a further attempt at cardioversion, sometimes with concomitant administration of an antiarrhythmic to enhance the chances of successful cardioversion and long-term maintenance of SR.10. Who should we refer for DC cardioversion? Cardioversion should not normally be attempted (unless the patient is compromised) until any underlying precipitant has been treated. This includes infection, thyrotoxicosis and also electrolyte disturbances. Many young patients with an acute underlying precipitant will spontaneously revert to sinus rhythm.Many other patients with recent-onset AF could be offered cardioversion, except those in whom sinus rhythm is unlikely to be maintained or cardioversion is contraindicated. This includes patients who have a contraindication to anticoagulation, structural heart disease (left atrium >5.5cm or mitral stenosis which precludes maintenance of sinus rhythm), or a long duration of AF (conventionally >12 months).Symptomatic patients with persistent AF may benefit, with evidence of improvement in functional capacity post-cardioversion.11. Do patients need to be anticoagulated prior to cardioversion? Conventional clinical practice suggests that formation of intracardiac thrombi takes at least 48 hours from the onset of AF and as such the risk of stroke is minimal if cardioversion is to be attempted in this time period. However, most clinicians would avoid cardioversion at this early stage, unless there was haemodynamic instability or a life-threatening situation. In any case, it is often difficult to be confident of clear identifiable event at the onset of AF.Therapeutic anticoagulation (with an INR of 2-3) should be established and maintained for at least three weeks before cardioversion and at least four weeks afterwards. In patients with stroke risk factors or in those at high risk of recurrence, long-term anticoagulation is recommended, even with successful cardioversion.In elective patients, transoesophageal echocardiography can be used to exclude thrombus in the left atrial appendage before cardioversion and negates the requirement for three weeks of anticoagulation beforehand, but not the requirement for anticoagulation at cardioversion and for at least four weeks after.Antiplatelet agents such as aspirin do not provide sufficient protection against stroke before, during or after cardioversion in any patient with AF for more than 48 hours.Timothy Watson is research fellow and Gregory Lip is professor of cardiovascular medicine at the University Department of Medicine, City Hospital, Birmingham Competing interestsProfessor Lip has received funding for research, educational symposia, consultancy and lecturing from different manufacturers of drugs used for the treatment of AF and thrombosis. He is clinical adviser to the Guideline Development Group writing the National Institute for Clinical Excellence (NICE) Guidelines on AF management (www.nice.org.uk).Timothy Watson None declared.
What I will do now
Dr Neil Statham responds to the answers to his questions· I had not appreciated the high risk of thromboembolism that patients with paroxysmal AF have and I will certainly be more aware of this when I counsel patients in the future· The use of ?-blockers or non-dihydropyridine calcium antagonists as a first line rate limiter is new information to me and this will result in a change to my prescribing patterns· Some guidance as to when to introduce rate limiting therapy is also useful