Family Practice Vol. 18, No. 5, 555-557
© Oxford University Press 2001
Selections from Current Literature |
Re-thinking the management of atrial fibrillation: a new approach
Department of Family Medicine, Health Sciences Center L-4, 050, SUNY at Stony Brook, Stony Brook, NY 11794, USA.
Kopes-Kerr CP. Re-thinking the management of atrial fibrillation: a new approach. Family Practice 2001; 18: 555–557.
Introduction
The management of atrial fibrillation has become too confusing. Do you cardiovert everyone? Do you anticoagulate everyone? Do you use diltiazem, a ß-blocker or digoxin for rate control? Which is more important—rhythm or rate control? These are just a few of the questions that have vexed us. While the literature on these subjects is conflicting and still in evolution, I would like to take a snap-shot of a portion of the current literature that really promises to lean towards a much simpler approach.
The issue of rhythm control
Hohnloser SH, Kuck K, Lilienthal J for the PIAF Investigators. Rhythm or rate control in atrial fibrillation: pharmacological intervention in atrial fibrillation (PIAF): a randomised trial. Lancet 2000; 356: 1789–1794.
Many physicians believe that restoration and maintenance of sinus rhythm is superior to rate control only. There are, however, no prospective data that compare both strategies. The recent PIAF study (pharmacological intervention in atrial fibrillation) addresses this issue. This was a randomized trial among 252 patients with atrial fibrillation of between 7 and 360 days duration which compared a rate control group with a rhythm control group. In the rate control group, diltiazem was used as first-line therapy and, in the rhythm control group, amiodarone was used. The primary endpoint measured was improvement in symptoms related to atrial fibrillation. Over a 1-year period a similar proportion of patients reported improvement in symptoms in both groups. Amiodarone was successful in pharmacological restoration of sinus rhythm in 23% of patients. Walking distance in a 6-minute walk test was better in the amiodarone group, but overall assessment of quality of life showed no differences between the two groups. The incidence of hospital admission was higher in the amiodarone group (69% versus 24%). Adverse drug effects more frequently led to a change in therapy in the amiodarone group (25% versus 14%).
Comment
What I take home from this is that a riskier drug associated with significantly more adverse reactions and hospitalizations achieved only a 23% restoration of rhythm and no advantage in overall assessment of quality of life. This would make me want to go with the rate control strategy for primary management.
But what about cardioversion?
Catherwood E, Fitzpatrick WD, Greenberg ML, Holzberger PT, Malenka DJ, Gerling BR, Birkmeyer JD. Cost-effectiveness of cardioversion and antiarrhythmic therapy in nonvalvular atrial fibrillation. Ann Intern Med 1999; 130: 625–636.
This study is really a decision analysis rather than a clinical trial. In the decision analysis, the authors concluded that strategies involving cardioversion alone were more effective and less costly than those not involving this option. For patients at high risk of ischaemic stroke (5.3% per year), cardioversion alone followed by repeated cardioversion plus amiodarone therapy on relapse was most cost-effective [$9300 per quality of life year (QALY)] compared with cardioversion alone followed by warfarin therapy on relapse. This strategy was also preferred for the moderate-risk cohort (3.6% per year), but the benefit was more expensive ($18 900 per QALY). For the low-risk group (1.6% per year), cardioversion alone followed by aspirin therapy on relapse was optimal. Strategies using cardioversion and quinidine prophylaxis offer no advantage in quality-adjusted survival unless the rate of ischaemic stroke is very high and amiodarone is contraindicated.
Comment
The limitations of this study are that it addresses only a hypothetical cohort of 70-year-old patients with different baseline risks for stroke and a time horizon of only 3 months. Can one argue that the above cardioversion-based strategy is still primary even if you have to repeat it regularly at intervals of 3 months or more? The attractiveness of cardioversion is that it might be a once and done cure for the abnormal rhythm, but reality does not bear this out. The relapse rate is high. These authors have considered strategies with options to include failure within a period of only 3 months. Once you consider longer periods (which you must for the younger population groups), the high failure rate of cardioversion make it unacceptable unless it offers a benefit that is several orders of magnitude better than alternative strategies. There are, in fact, no data to suggest this. What data I have tend to be negative on this issue; they come from the literature on rhythm control after cardioversion.
How to make up for the failures of cardioversion
Kuhlkamp V, Schirdewan A, Stangl K, Homberg M, Ploch M, Beack OA. Use of metoprolol CR/XL to maintain sinus rhythm after conversion from persistent atrial fibrillation: a randomized, double-blind, placebo-controlled study. J Am Coll Cardiol 2000; 36: 139–146.
Berry C, McMurray J. Warfarin should be given for up to one year after successful cardioversion (letter). Br Med J 2000; 321: 639.
In this study of 394 patients with atrial fibrillation who had already been cardioverted successfully, after only 6 months the authors found a 48.7% relapse rate in the metoprolol group compared with a 59.9% relapse rate in the placebo group (P = 0.005).
Comment
What this tells us is that cardioversion does not offer an adequate long-term solution at all. The failure rate is almost 60% in just 6 months, and it is only marginally improved by adding a relatively safe rate and rhythm agent such as metoprolol. The idea of using a more effective rhythm agent such as amiodarone is precluded by the findings of the first study above. In a comment appearing with this article, an editor asks the disturbing question, if the rhythm relapse rate is so high after cardioversion, do we need to consider prolonging the period of anticoagulation with coumadin. The authors of a letter written to the editor in follow-up to this article describe their experience in a prospective observation study of elective cardioversion in 111 consecutive patients over one 12-month period. Sinus rhythm was restored immediately in 86%, but only 61% of the patients discharged in sinus rhythm were still in sinus rhythm at 1 month and, of these, another 39% had relapsed into atrial fibrillation by 12 months. They conclude that anticoagulation should be continued for at least a full year, but they have absolutely no data to suggest that it would be safe to discontinue after even a year.
Is it really necessary to stratify patients with atrial fibrillation to decide on therapy?
Lip GYH. Thromboprophylaxis for atrial fibrillation. Lancet 1999; 353: 4–6.
Pearce LA, Hart RNG, Halperine JL. Assessment of three schemes for stratifying stroke risk in patients with nonvalvular atrial fibrillation. Am J Med 2000; 109: 45–51.
The SPAF III Writing Committee for the Stroke Prevention in Atrial Fibrillation Investigators. Patients with nonvalvular atrial fibrillation at low risk of stroke during treatment with aspirin: Stroke Prevention in Atrial Fibrillation III Study. J Am Med Assoc 1998; 279: 1273–1277.
There are two basic reasons for stratifying patients with atrial fibrillation into subgroups with risk factors: (i) to identify a high-risk group in whom the restoration of sinus rhythm might be particularly desirable, if it could be maintained; and (ii) to treat a high-risk group more aggressively, in this case with coumadin instead of aspirin. As the above articles tend to show, given the very high failure rate of all strategies to maintain sinus rhythm, this does not appear to be a valid rationale for risk stratification. If, however, coumadin were shown absolutely clearly to be adequately safe and demonstrably more effective than aspirin therapy, this would provide sufficient rationale for risk stratification and for treatment of at least the high-risk group with coumadin along with an agent such as diltiazem for rate control.
Patients whose annual stroke risk is <2% when taking aspirin do not benefit substantially from treatment with warfarin. More than 100 patients must be treated with warfarin for 1 year to prevent one stroke. For high-risk patients with an annual stroke risk >6%, the number needed to treat (NNT) is <25. There are several different schemata for assigning stroke risk in atrial fibrillation. The British guideline uses these assignments: high-risk (stroke risk 8–12%/year): patients with a previous TIA or stroke, over the age of 75 with either diabetes or hypertension, or clinical evidence of valve disease, congestive heart failure, thyroid disease or impaired left ventricular function on echocardiography. The moderate risk (annual stroke risk, 4%) consisted of those patients under the age of 65 with risk factors such as diabetes, hypertension, peripheral vascular disease or coronary artery disease; or persons over the age of 65 without such risk factors. All other patients are in the low-risk group (<1% stroke risk/year).
A more recent look at risk stratification compared the recommendations of various other groups—the Atrial Fibrillation Investigators (1994), the American College of Chest Physicians (1998) and the Stroke Prevention in Atrial Fibrillation (SPAF) investigators (1995). Each of these schemata effectively identified patients at low risk (stroke rate 0.3–1.1 per 100 000 patient-years), and the size of this group varied from 14% (ACCP) to 45% (SPAF). The authors comment that all of the schemata are far less effective at identifying high-risk patients.
The SPAF schema offers the advantage of considerable simplicity and also defines the largest group at low risk (45%). In their study, such low-risk patients were treated with 325 mg of aspirin daily. Among 892 patients, the stroke rate was 2% per year and the rate of disabling ischaemic stroke was 0.8% per year. This approximated the rate of ischaemic stroke in the general population of the same age range (~1% per year). High-risk patients had a stroke rate of 8% per year. Aspirin therapy reduced this risk by ~20%, with major haemorrhage rates similar to that with placebo. If these patients were treated with coumadin, ~60 strokes per 1000 would be prevented, and 10–12 major haemorrhages would occur.
Is coumadin really better than aspirin for patients with chronic non-valvular atrial fibrillation?
Green CJ, Hadorn DC, Kazanjian A. Anticoagulation in chronic nonvalvular atrial fibrillation: a critical appraisal and meta-analysis. Can J Cardiol 1997; 13: 811–815.
This meta-analysis of five published randomized controlled trials concerning primary stroke prevention in chronic non-valvular atrial fibrillation with warfarin showed that warfarin did not reduce the incidence of fatal strokes to a statistically significant extent, nor was the incidence of fatal bleeding increased significantly. Warfarin did reduce the absolute annual incidence of major strokes in patients with atrial fibrillation by 0.89%, but at the same time increased the absolute annual risk of major bleeding incidents by 1.8%. The authors conclude, "On balance, the margin between expected benefit and harm for warfarin prophylaxis in patients with chronic nonvalvular atrial fibrillation is uncomfortably thin." These results and conclusions differ from many previously published recommendations.
Conclusion
The data on the management of atrial fibrillation are complex and confusing. From the handful of studies outlined above, however, I think we can identify a consistent simple approach that is at least reasonable, if not definitive.
All patients with atrial fibrillation should be treated with diltiazem for rate control and with aspirin (325 mg/ day) for stroke prevention. For those comfortable with and preferring a less invasive and lower cost strategy, no cardioversion and no anticoagulation is necessary. This is a reasonable basic strategy and is consistent with a substantial body of the current divided literature. The question of whether a patient is really stroke-adverse is a subjective one and should be determined by the patient. The option of anticoagulation with coumadin need be offered only to those patients who have both risk factors present and a distinct preference for a higher risk treatment strategy based on their personal desire to avoid any risk of stroke. For the purposes of selecting a schema for risk stratification of patients with a trial fibrillation, the SPAF schema is probably preferable since it results in the lowest number of high-risk interventions.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||