To determine the safety, clinical effectiveness and cost-effectiveness of radio frequency catheter ablation (RCFA) for the curative treatment of atrial fibrillation (AF) and typical atrial flutter.
For the systematic reviews of clinical studies 25 bibliographic databases and internet sources were searched in July 2006, with subsequent update searches for controlled trials conducted in April 2007. For the review of cost-effectiveness a broad range of studies was considered, including economic evaluations conducted alongside trials, modelling studies and analyses of administrative databases.
Systematic reviews of clinical studies and economic evaluations of catheter ablation for AF and typical atrial flutter were conducted. The quality of the included studies was assessed using standard methods. A decision model was developed to evaluate a strategy of RFCA compared with long-term antiarrhythmic drug (AAD) treatment alone in adults with paroxysmal AF. This was used to estimate the cost-effectiveness of RFCA in terms of cost per quality-adjusted life-year (QALY) under a range of assumptions. Decision uncertainty associated with this analysis was presented and used to inform future research priorities using the value of information analysis.
A total of 4858 studies were retrieved for the review of clinical effectiveness. Of these, eight controlled studies and 53 case series of AF were included. Two controlled studies and 23 case series of typical atrial flutter were included. For atrial fibrillation, freedom from arrhythmia at 12 months in case series ranged from 28% to 85.3% with a weighted mean of 76%. Three RCTs suggested that RFCA is more effective than long-term AAD therapy in patients with drug-refractory paroxysmal AF. Single RCTs also suggested superiority of RFCA over electrical cardioversion followed by long-term AAD therapy and of RFCA plus AAD therapy over AAD maintenance therapy alone in drug-refractory patients. The available RCTs provided insufficient evidence to determine the effectiveness of RFCA beyond 12 months or in patients with persistent or permanent AF. Adverse events and complications were generally rare. Mortality rates were low in both RCTs and case series. Cardiac tamponade and pulmonary vein stenosis were the most frequently recorded complications. For atrial flutter, freedom from arrhythmia at 12 months in case series ranged from 85% to 92% with a weighted mean of 88%. Neither of the atrial flutter RCTs reported freedom from arrhythmia at 12 months. One RCT found a statistically significant benefit favouring ablation over AADs in terms of freedom from arrhythmia at a mean follow-up of 22 months. A second RCT reported a more modest effect favouring ablation in terms of freedom from atrial flutter at follow-up in older patients (mean age 78 years) after their first episode of flutter. In the atrial flutter case series, mortality was rare and the most frequent complications were atrioventricular block and haematomas. Complications in the RCTs were similar, except for those events likely to have been caused by AAD therapy (e.g. thyroid dysfunction). The review of cost-effectiveness evidence found one relevant study, which from a UK NHS perspective had a number of important limitations. The base-case analysis in the decision model demonstrated that if the quality of life benefits of RFCA are maintained over the remaining lifetime of the patient then the cost-effectiveness of RFCA appears clear. These findings were robust over a wide range of alternative assumptions, being between 7763 and 7910 pounds per additional QALY with very little uncertainty. If the quality of life benefits of RFCA are assumed to be maintained for no more than 5 years, cost-effectiveness of RFCA is dependent on a number of factors. Estimates of cost-effectiveness that explored the influence of these factors ranged from 23,000 to 38,000 pounds per QALY.
RFCA is a relatively safe and efficacious procedure for the therapeutic treatment of AF and typical atrial flutter. There is some randomised evidence to suggest that RFCA is superior to AADs in patients with drug-refractory paroxysmal AF in terms of freedom from arrhythmia at 12 months. RFCA appears to be cost-effective if the observed quality of life benefits are assumed to continue over a patient's lifetime. However, there remain uncertainties around longer-term effects of the intervention and the extent to which published effectiveness findings can be generalised to 'typical' UK practice. All catheter ablation procedures for the treatment of AF or atrial flutter undertaken in the UK should be recorded prospectively and centrally and measures to increase compliance in recording RFCA procedures may be needed. This would be of particular value in establishing the long-term benefits of RFCA and the true incidence and impact of any complications. Collection of appropriate quality of life data within any such registry would also be of value to future clinical and cost-effectiveness research in this area. Any planned multicentre RCTs comparing RFCA against best medical therapy for the treatment of AF and/or atrial flutter should be conducted among 'non-pioneering' centres using the techniques and equipment typically employed in UK practice and should measure relevant outcomes.