Canadian Agency for Agence canadienne Drugs and Technologies des médicaments et des in Health technologies de la santé
CADTH Technology Report
Issue 128 Ablation Procedures for Rhythm Control in September 2010 Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses
Supporting Informed Decisions Until April 2006, the Canadian Agency for Drugs and Technologies in Health (CADTH) was known as the Canadian Coordinating Office for Health Technology Assessment (CCOHTA).
Publications can be requested from:
CADTH 600-865 Carling Avenue Ottawa ON Canada K1S 5S8 Tel.: 613-226-2553 Fax: 613-226-5392 Email: [email protected]
or downloaded from CADTH’s website: http://www.cadth.ca
Cite as: Assasi N, Blackhouse G, Xie F, Gaebel K, Robertson D, Hopkins R, Healey J, Roy D, Goeree R. Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses [Internet]. Ottawa: Canadian Agency for Drugs and Technologies in Health; 2010 (Technology report; no. 128). [cited 2010-09-17]. Available from: http://www.cadth.ca/index.php/en/hta/reports-publications/search?&type=16
Production of this report is made possible by financial contributions from Health Canada and the governments of Alberta, British Columbia, Manitoba, New Brunswick, Newfoundland and Labrador, Northwest Territories, Nova Scotia, Nunavut, Prince Edward Island, Saskatchewan, and Yukon. The Canadian Agency for Drugs and Technologies in Health takes sole responsibility for the final form and content of this report. The views expressed herein do not necessarily represent the views of Health Canada or any provincial or territorial government.
Reproduction of this document for non-commercial purposes is permitted provided appropriate credit is given to CADTH.
CADTH is funded by Canadian federal, provincial, and territorial governments.
Legal Deposit – 2010 Library and Archives Canada ISSN: 1922-6101 (print) ISSN: 1922-611X (online) H0491 – September 2010
PUBLICATIONS MAIL AGREEMENT NO. 40026386 RETURN UNDELIVERABLE CANADIAN ADDRESSES TO CANADIAN AGENCY FOR DRUGS AND TECHNOLOGIES IN HEALTH 600-865 CARLING AVENUE OTTAWA ON K1S 5S8
Canadian Agency for Drugs and Technologies in Health
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses
Nazila Assasi, MD, PhD1,2 Gord Blackhouse, MSc1,2 Feng Xie, PhD1,2,3 Kathryn Gaebel, MSc1,3 Diana Robertson, MLIS1,2 Rob Hopkins, MSc1,2 Jeff Healey, MD, MSc, FRCPC4 Denis Roy, MD5, 6 Ron Goeree, MA1,2,3
September 2010
1 Programs for Assessment of Technology in Health, McMaster University, Hamilton, Ontario 2 Department of Clinical Epidemiology & Biostatistics, McMaster University, Hamilton, Ontario 3 Centre for Evaluation of Medicines, St. Joseph’s Healthcare, Hamilton, Ontario 4 Population Health Research Institute, McMaster University, Hamilton, Ontario 5 Department of Medicine, University of Montreal, Montreal, Quebec 6 Montreal Heart Institute, Montreal, Quebec
Reviewers This document was externally reviewed by content experts, and the following individuals granted permission to be cited.
External Reviewers Jafna L. Cox, BA, MD Adrian Baranchuk, MD, FACC Professor of Medicine and of Community Associate Professor of Medicine and Health and Epidemiology Physiology Dalhousie University Queen’s University Halifax, Nova Scotia Kingston, Ontario
Craig Mitton, PhD Tanya Horsley, PhD Associate Professor Educational Research Scientist University of British Columbia Royal College of Physicians and Vancouver, British Columbia Surgeons of Canada Ottawa, Ontario
Rick Audas, BBA, MBA, MA(Econ), PhD Assistant Professor of Medicine Memorial University St. John’s, Newfoundland
CADTH Peer Review Group Reviewers Chris Skedgel, MDE James Brophy, MEng, MD, FRCPC, PhD Research Health Economist Professor of Medicine and Epidemiology Department of Medicine McGill University Dalhousie University Montreal, Quebec Halifax, Nova Scotia
Industry The following manufacturers were provided with an opportunity to comment on an earlier version of this report: Bard Canada Inc., Boston Scientific Ltd., Cardima Inc., Biosense Webster Inc., Medtronic Canada Inc., and St. Jude’s Medical Inc. All comments that were received were considered when preparing the final report.
This report is a review of existing public literature, studies, materials, and other information and documentation (collectively the “source documentation”) that are available to CADTH. The accuracy of the contents of the source documentation on which this report is based is not warranted, assured, or represented in any way by CADTH, and CADTH does not assume responsibility for the quality, propriety, inaccuracies, or reasonableness of any statements, information, or conclusions contained in the source documentation.
CADTH takes sole responsibility for the final form and content of this report. The statements and conclusions in this report are those of CADTH and not of its Panel members or reviewers. i Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses Authorship Nazila Assasi was the clinical research lead and a systematic reviewer. Nazila wrote the protocol, designed the screening questionnaires, conducted the clinical review and data abstraction, and wrote the clinical review portion of the report. She assisted with the creation of the Review Manager database and produced the clinical review results tables.
Gord Blackhouse was one of the health economists. Gord wrote the primary economic model section of the report and assisted in the review of economic evidence. Gord completed and wrote the budget impact analysis section of the report.
Feng Xie was one of the health economists. Feng reviewed the economic evidence and helped developed the primary economic evaluation.
Kathryn Gaebel was the systematic reviewer. Kathryn helped create the study protocol, screening questionnaires, and data abstraction forms. Kathy was responsible for the implementation issues section of the report.
Diana Robertson was the information specialist. She developed and conducted the literature search strategy and retrieved copies of relevant articles. Diana helped develop screening questionnaires and data abstraction forms and screened citations that were identified in the literature search. Diana wrote the methods for the systematic reviews.
Rob Hopkins served as a statistician. He conducted and advised on the statistical component of the meta-analyses.
Jeff Healey was a clinical content expert.
Denis Roy was a clinical content expert.
Ron Goeree was the project advisor and the project coordinator.
All authors reviewed drafts of the report and approved the final report.
Conflict of Interest Gord Blackhouse has been a consultant on economic evaluations for Eli Lilly Canada Inc., GlaxoSmithKline Inc., Pfizer Canada Inc., and Bristol-Myers Squibb Canada.
Ron Goeree has been an advisor or consultant for Actelion Pharmaceuticals Canada Inc., Sanofi- aventis Groupe (Global), and Amgen Canada. He has received research funding or grants from Eli Lilly Canada Inc., GlaxoSmithKline Inc., Pfizer Canada Inc., and Bristol-Myers Squibb Canada.
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: ii Clinical and Cost-Effectiveness Analyses Denis Roy has been an advisory board member for Sanofi-aventis, Merck, and Boehringer Ingelheim, and he has been an events committee member for Cryocath/Medtronic.
Jeff Healey has received research funding or grants from Boston Scientific, Boehringer Ingelheim, AstraZeneca, and Sanofi-aventis. He has been a speaker for Boehringer Ingelheim and Boston Scientific.
iii Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses EXECUTIVE SUMMARY
The Issue Atrial fibrillation (AF) leads to more disability than any other cardiac arrhythmia, because of its high incidence and the potential for adverse outcomes. It affects more than 200,000 Canadians. The prevalence increases with age (more than 5% of the population over 65 years of age have AF).
The first-line medical therapy for AF is antiarrhythmic drugs (AADs). Although medical treatment has the advantage of being non-invasive and available, chronic administration may be needed. On the other hand, ablation of AF may prevent long-term use of AADs. There remains uncertainty about the health impact of using ablation and its place in therapy.
Ablation procedures for AF are funded in at least five Canadian provinces. Some jurisdictions are interested in developing guidelines on ablation procedures for the management of patients with AF. In this technology assessment report, only the ablation procedures aimed at controlling cardiac rhythm in adults with AF are evaluated.
Objectives The aims of this health technology assessment were to: evaluate the comparative clinical effectiveness of minimally invasive ablation procedures (pulmonary vein isolation [PVI] alone and with other adjunctive atrial ablations) for AF and to compare these with other modalities for converting AF to normal sinus rhythm, including pharmacological or electrical cardioversion, and to more invasive surgical procedures evaluate the comparative cost-effectiveness of minimally invasive ablation procedures for AF and to compare these with pharmacological or electrical cardioversion and to more invasive surgical procedures evaluate the impact of using minimally invasive procedures on patients with paroxysmal AF, patients with persistent AF, patients who have not used drugs to treat AF (drug naive), patients who have used drugs to treat AF, and patients with congestive heart failure.
Methods Systematic literature searches were undertaken to identify relevant clinical and economic evaluations of ablation procedures for AF. One additional search was conducted to identify the latest Canadian and international guidelines on the use of minimally invasive ablation procedures for AF. Relevant controlled clinical trials (randomized and non-randomized) of any duration mainly designed to evaluate clinical efficacy, effectiveness, or safety of ablation procedures in adult patients with AF were identified. Decisions about eligibility and methodological quality of studies were made by two independent reviewers. Any discrepancies were solved by consensus. When two or more comparable studies were identified, a pooled estimate of effect was obtained in a meta-analysis. For the studies that were not comparable in population, interventions, or outcome measures, narrative descriptions are provided.
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: iv Clinical and Cost-Effectiveness Analyses Clinical Effectiveness Of the 2,648 potential citations that were identified during the systematic search, 2,362 and 256 citations were excluded during the title and abstract and the full text reviews respectively. Thirty citations reporting 23 randomized and six non-randomized controlled trials met the inclusion criteria of this review.
Catheter ablation versus medical treatment The systematic review of clinical evidence indicated that the use of catheter ablation was superior to treatment with AADs, in patients with AF, for the maintenance of sinus rhythm up to a year (relative risk [RR] 2.82, 95% CI 2.13 to 3.74). There was insufficient evidence comparing catheter ablation as a first-line treatment with medical therapy in patients for whom a rhythm control strategy was appropriate. Based on the subgroup analyses, the use of ablation techniques led to better results in patients with paroxysmal AF (RR 3.80, 95% CI 2.92 to 4.96), compared with the pooled results for all AF types (RR 2.82, 95% CI 2.13 to 3.74).
Catheter ablation versus electrical cardioversion The non-randomized controlled trial comparing the efficacy of using catheter ablation to that of using electrical cardioversion showed a higher success rate in patients undergoing ablation procedures (82%), compared with patients in the electrical cardioversion group (40%).
Catheter ablation versus surgical procedures No studies evaluated the effectiveness of using catheter ablation procedures compared with open heart surgery for the treatment of AF.
PVI versus PVI plus adjunctive atrial ablations The results of our meta-analyses showed that at 12 months, patients with AF who underwent PVI plus adjunctive atrial ablations (PVI+) had an 8% higher chance of maintaining in sinus rhythm compared with those who underwent PVI, (RR 0.92, 95% CI 0.86 to 0.99). The overall estimate of the effect size is interpreted with caution, because of between-study variations in patient populations (AF types) and heterogeneity of the ablation techniques that were used. Our meta- analysis suggested that PVI plus linear ablations of left atrial sites had a 15% higher success rate than PVI (RR 0.85, 95% CI 0.76 to 0.95). There was insufficient evidence to reliably estimate the effects of additional ablation lines in the right atrium, adjunctive ablation of ectopic triggers of AF, or other approaches such as stepwise and tailored ablation techniques. The results of subgroup analysis indicate that patients with persistent AF could benefit more from PVI+ strategies than from PVI (RR 0.59, 95% CI 0.39 to 0.91).
Our review failed to evaluate the long term consequences of AF ablation procedures. In the clinical review, few trials were found on the efficacy of catheter ablation as a first-line therapy. The studies did not address the effectiveness of AF ablation in patients with congestive heart failure, the comparative effectiveness of catheter ablation and surgical procedures, and the effectiveness of repeated ablations. There were insufficient data on adverse events after the use of ablation techniques and comparators.
v Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses Economic Analysis The primary economic analysis found the incremental cost effectiveness of AF ablation compared to anti-arrhythmic medication to be $59,194 per quality adjusted life year (QALY) in patients with a CHADS2 risk score of two, and for whom at least one anti-arrhythmic medication had failed. Therefore if society’s willingness to pay for each QALY is $59,194 or greater, AF ablation would be cost-effective in this population. Otherwise AADs would be the cost-effective strategy. The probability of AF ablation being cost-effective at willingness-to-pay thresholds for a QALY of $25,000, $50,000, $100,000, and $150,000 was estimated to be 0.03, 0.30, 0.89, and 0.98 respectively. When no difference in utility is assumed between normal sinus rhythm and AF health states, the cost per QALY of AF ablation becomes $221,839. If it is assumed that restoring normal sinus rhythm has no impact on stroke, the cost per QALY of AF ablation compared with AADs is $86,129. Such findings may be inconsistent with the clinical motivation for AF treatment, which may be stroke prevention instead of improving quality of life in the absence of stroke.
Health Services Impact In 2008, an estimated 2,030 minimally invasive AF ablation procedures were performed in Canada. The Quebec data are not included in the databases that are used to identify the number of procedures. Most ablations occurred in Ontario (910), British Colombia (851), Alberta (119), and Nova Scotia (98). The inpatient and physician costs are estimated to be $19,467,400. Based on trends over the past five years, the projected expenditures on these procedures are estimated to reach $40,888,821 by 2013.
Conclusions The evidence in this systematic review indicates that the use of catheter ablation increases the rate of maintenance of sinus rhythm compared with treatment with AADs in patients for whom the use of one or two drugs failed. The studies are of insufficient size and duration to evaluate the impact on stroke, heart failure, and mortality. Ablation techniques were shown to lead to better results in patients with paroxysmal AF. Limited data suggest that catheter ablation may be an effective first-line rhythm control strategy in patients with AF. More trials are needed to confirm these findings. Our review suggests that patients with persistent AF may benefit more from PVI+ strategies than from PVI.
The primary economic evaluation using a five-year time horizon found the incremental cost per QALY of AF ablation compared with AAD to be $59,194. These findings were similar to those of other published economic evaluations. The cost-effectiveness of AF ablation was found to be more favourable when longer time horizons were used.
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: vi Clinical and Cost-Effectiveness Analyses
vii Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses TABLE OF CONTENTS
EXECUTIVE SUMMARY ...... iv
ACRONYMS AND ABBREVIATIONS...... x
GLOSSARY ...... xi
1 INTRODUCTION ...... 1 1.1 Background and Setting in Canada ...... 1 1.2 Overview of Technology ...... 2
2 ISSUE ...... 2
3 OBJECTIVES ...... 3
4 RESEARCH QUESTIONS...... 3
5 CLINICAL REVIEW ...... 4 5.1 Methods ...... 4 5.1.1 Literature searches...... 4 5.1.2 Selection criteria ...... 5 5.1.3 Selection method...... 6 5.1.4 Data abstraction strategy...... 6 5.1.5 Strategy for validity assessment ...... 6 5.1.6 Data analysis methods ...... 6 5.2 Results...... 9 5.2.1 Quantity of research available ...... 9 5.2.2 Study characteristics...... 9 5.2.3 Data analyses and synthesis ...... 10
6 ECONOMIC ANALYSIS ...... 45 6.1 Review of Economic Studies: Methods...... 45 6.1.1 Literature searches...... 45 6.1.2 Selection criteria ...... 45 6.1.3 Selection method...... 46 6.1.4 Data abstraction ...... 46 6.1.5 Data analysis methods ...... 46 6.2 Review of Economic Studies: Results...... 46 6.2.1 Chan et al...... 46 6.2.2 Reynolds et al...... 47 6.2.3 McKenna et al...... 48 6.2.4 Eckard et al...... 49 6.3 Primary Economic Evaluation: Methods ...... 50 6.3.1 Type of economic evaluation ...... 50 6.3.2 Target population...... 50 6.3.3 Comparators...... 50 6.3.4 Perspective...... 50 6.3.5 Effectiveness ...... 51 6.3.6 Time horizon...... 51 6.3.7 Modelling ...... 51 6.3.8 Valuing outcomes ...... 53
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: viii Clinical and Cost-Effectiveness Analyses 6.3.9 Resource use and costs ...... 57 6.3.10 Discount rate ...... 59 6.3.11 Variability and uncertainty...... 59 6.4 Primary Economic Evaluation: Results ...... 60 6.4.1 Analysis and results...... 60 6.4.2 Results of variability analysis...... 60 6.4.3 Results of uncertainty analysis ...... 61
7 HEALTH SERVICES IMPACT...... 64 7.1 Population Impact ...... 64 7.2 Budget Impact...... 65 7.2.1 Methods...... 65 7.2.2 Results...... 65 7.3 Planning, Implementation, Utilization, and Legal or Regulatory Considerations ...... 66
8 DISCUSSION...... 67 8.1 Summary of Results...... 67 8.2 Strengths and Weaknesses of This Assessment...... 70 8.3 Generalizability of Findings...... 71 8.4 Knowledge Gaps ...... 71
9 CONCLUSIONS ...... 72
10 REFERENCES ...... 73
APPENDIX 1: Literature Search Strategy for Clinical Effectiveness and Economic Studies APPENDIX 2: Level 1 Clinical Screening Checklist APPENDIX 3: Level 1 Clinical Screening Checklist APPENDIX 4: Guideline Screening Checklist APPENDIX 5: Clinical Review Data Abstraction Form APPENDIX 6: Quality Score for Jadad Scale APPENDIX 7: Quality Assessment Tools for Guidelines and Recommendations APPENDIX 8: List of Excluded Studies from the Clinical Review and Reasons for Exclusion APPENDIX 9: Characteristics of the Included Studies APPENDIX 10: Baseline Characteristics of the Participants in the Included Studies APPENDIX 11: Forest Plots from Meta-Analyses of Clinical Data APPENDIX 12: Funnel Plots of the Studies Included in the Meta-Analyses APPENDIX 13: QUORUM Flowchart – Guideline Review APPENDIX 14: List of Guidelines Excluded After Full-Text Screening (Level 2) APPENDIX 15: QUORUM Flowchart – Economic Review APPENDIX 16: List of Articles Excluded From the Economic Review after Full-Text Screening (Level 2) APPENDIX 17: Probability of Reverting to Normal Sinus Rhythm at 12 months for Antiarrhythmic Medications APPENDIX 18: Calculation of Annual Probability of Major Bleed in Absence of Aspirin or Warfarin APPENDIX 19: General Population Utility Values APPENDIX 20: Utility Estimates for Ischemic and Hemorrhagic Stroke APPENDIX 21: Details on Professional Fees for AF ablation APPENDIX 22: Relative Risk of Stroke by Age using 65 years as a Reference APPENDIX 23: Distributions and Parameters used in Probabilistic Analysis APPENDIX 24: Implementation of AF Ablation Procedures Questionnaire
ix Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses ACRONYMS AND ABBREVIATIONS
AAD antiarrhythmic drug ACC American College of Cardiology AF atrial fibrillation AHA American Heart Association AT atrial tachycardia or tachyarrhythmia AV atrioventricular CADTH Canadian Agency for Drugs and Technologies in Health CCS Canadian Cardiovascular Society CFAEs complex fractionated atrial electrograms CHF congestive heart failure CI confidence interval DAD Discharge Abstract Database ECAS European Cardiac Arrhythmia Society EHRA European Heart Rhythm Association EP electrophysiology ESC European Society of Cardiology HTA health technology assessment ICH intracranial hemorrhage NACRS National Ambulatory Care Reporting System NICE National Institute for Health and Clinical Excellence NSR normal sinus rhythm PV pulmonary vein PVI pulmonary vein isolation QALY quality-adjusted life-year RCT randomized controlled trial RR relative risk SVC superior vena cava
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: x Clinical and Cost-Effectiveness Analyses GLOSSARY
Ablation: A non-surgical procedure that is used to destroy heart tissue through the directing of energy through a catheter. Different forms of energy (radiofrequency, laser, ultrasonic, cryogenic) may be used in ablation procedures (radiofrequency energy is the most commonly used).
CHADS2 Risk Score: A score use to predict the risk of stroke in patients with atrial fibrillation. Points are assigned for the presence of congestive heart failure, hypertension, age >75 years, diabetes mellitus and prior stroke or TIA. Two or more points indicates a higher risk of future stroke.
Forest plot: A graph in a meta-analysis that shows the point estimate and confidence interval of the effect that is observed in each study, with the overall pooled effect of all studies and its confidence interval. The point estimate of the effect in each study is shown as a square. A horizontal line runs through each square to show the study’s confidence interval. At the bottom, a diamond shows the overall estimate from the meta-analysis. The confidence interval around the diamond is shown.
Funnel plot: A simple scatter plot of estimates of effect size from studies in a meta-analysis against a measure of each study’s size or precision. An asymmetric appearance of a funnel plot suggests a relationship between treatment effect and study size (small study effect) or indicates the existence of publication bias.
Level of evidence: The type, quality, and trustworthiness of evidence that is used to support guideline recommendations. Various classification systems have been developed by practice guideline groups to assign levels of evidence.
Minimally invasive procedures: Procedures that avoid the use of open invasive surgery in favour of closed or local surgery.
Quality score: A value that is assigned to represent a study’s internal validity.
Random effect model: A statistical model that accounts for within-study sampling error and between-studies variation in assessing the results of a meta-analysis. In this model, it is assumed that different treatment effects are being estimated in studies. When there is heterogeneity among the results of the studies beyond chance, random effects models give wider confidence intervals than fixed effects models.
SF-36: A short form measure of perceived health status in the general population. The SF-36 survey includes 36 questions to measure eight health domains: physical function, social function, physical role, emotional role, mental health, vitality, pain, and general health perceptions. Each domain is scored from 0 to 100. The physical composite score and the mental composite score can be generated to provide an overall assessment of the respondent’s physical and mental health.
xi Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses 1 INTRODUCTION 1.1 Background and Setting in Canada
Atrial fibrillation (AF), which is the most common form of cardiac arrhythmia, is associated with high morbidity and mortality.1,2 This condition is characterized by disorganized, rapid, and irregular activity of the two upper chambers of the heart (atria), associated with irregular (and in untreated patients, rapid) response of the two lower chambers of the heart (ventricles). The heart rate typically varies between 120 and 160 beats per minute. In some patients it can be 200 beats per minute.3 Patients with AF are at a higher risk of clot formation and subsequent adverse hemodynamic events (such as stroke), because of the loss of atrial contractility, irregular ventricular rate of activity, and the loss of atrial appendage contractility and emptying.3 AF increases the risk of stroke four- to five-fold in all age groups and leads to 10% to 15% of all ischemic strokes.4 This arrhythmia, which is the most common cause of stroke among elderly people, causes approximately 25% of strokes in patients age 80 years or older.5
AF may be classified on the basis of electrocardiographic findings or frequency of episodes and the ability to convert back to sinus rhythm. According to the guidelines of the American College of Cardiology (ACC), the American Heart Association (AHA), and the European Society of Cardiology (ESC), AF is classified as a first-detected episode or a recurrent episode. Recurrent AF can be classified as paroxysmal (self-terminating and usually lasting less than 24 hours), persistent (sustained more than seven days), or permanent.6
The Heart and Stroke Foundation estimates that approximately 250,000 Canadians are affected by AF.7 The prevalence of AF increases with age (ranging from 0.1% of the population less than 55 years of age to 9% among individuals aged 80 years or older).2 It is influenced by gender, structural heart diseases, hypertension, obesity, diabetes, and other chronic conditions.8-11 In the United States, it is estimated that the number of AF cases will increase from 2.3 million in 2001 to 5.6 million in 2050, as a result of an aging population.2 AF is associated with higher morbidity and mortality, because it increases the risk of stroke and other thromboembolic events and congestive heart failure (CHF). As a risk factor, CHF can promote the development of AF.12 The rate of hospitalization for AF in Canada was approximately 583 per 100,000 people, between 1997 and 2000, with an average of 129,000 hospitalizations per year.13
The goals of AF treatments are to control the heart rate, prevent thromboembolism, and correct the rhythm disturbance.6 The two strategies in AF treatment are rhythm control (cardioversion and maintenance of sinus rhythm using antiarrhythmic drugs [AADs]) and rate control (atrioventricular [AV] nodal blockers and anticoagulation). The Canadian Cardiovascular Society (CCS) recommends both strategies as acceptable initial approaches. The exception is permanent AF, where rate control is recommended.8 Various treatment options are available for rhythm control including medication, electrical (direct-current) cardioversion, or surgical procedures.14 Most patients need more than one type of treatment. AAD therapy is recommended as a first choice for the restoration of normal sinus rhythm (NSR).14 AADs are commonly classified as sodium channel blockers (Class 1), beta-adrenergic or beta-blockers (Class II), potassium channel blockers (Class III), and calcium channel blockers (Class IV).15 Three Class I drugs (flecainide, quinidine, and propafenone) and two Class III drugs (sotalol and amiodarone)
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 1 Clinical and Cost-Effectiveness Analyses are commonly used in Canada for the treatment of AF.16 The limitations of these drugs in the maintenance of NSR include inconsistent efficacy and frequent side effects. As a result, non- pharmacological approaches, including catheter ablation and surgery have been considered in the treatment of AF.14,17,18 The standard surgical treatment is the Cox-Maze procedure, which requires open heart surgery19,20 and involves the creation of a maze of small incisions along the right and left atria. When these incisions heal, the scar tissue that is created inhibits the re-entry of the irregular electrical impulses that cause AF. Because the Cox-Maze procedure is invasive, minimally invasive catheter-based interventions have been developed.21
1.2 Overview of Technology
The goals of catheter ablation procedures are to eliminate the triggers of AF and to modify the atrial substrate(s) that maintain AF.22 Given that the pulmonary veins (PVs) are a critical anatomic site in the treatment of AF,23 some minimally invasive procedures involve the isolation of the source of abnormal impulses originating from these veins. In a minimally invasive catheter ablation procedure, a catheter is inserted through the femoral vein to access the heart and burn the abnormal foci of electrical activity by direct contact or by isolating them from the rest of the cardiac atrium. Radiofrequency energy is most commonly used for AF ablation.24
The following strategies are used for catheter ablation in patients with AF:25 Isolation of the triggers and perpetuating re-entrant circuits in the PVs Disruption of the substrate for perpetuating rotors in the antra of the PVs and the posterior left atrium Targeted ablation of ganglionated autonomic plexi in the epicardial fat pads Disruption of putative dominant rotors in the left and right atria, recognized by high- frequency complex fractionated atrial electrograms (CFAEs) during mapping of AF.
Other strategies are based on the electrical isolation of PVs plus adjuvant ablations to eliminate the substrate that may start AF (substrate modification), including the creation of ablation lines in one or both atria, ablation of CFAEs, or targeting AF nests (for example, superior vena cava [SVC]). A stepwise sequential ablation approach has been described.26,27 This strategy includes an initial isolation of PVs followed by sequential linear and CFAE ablations until AF ends. These approaches are mainly proposed for the treatment of patients with long-lasting persistent AF, who usually have multiple ectopic triggers. There is no consensus on what additional sites are ablated or if additional ablation is performed during the first ablation.
2 ISSUE
More disability is the result of AF than of any other cardiac arrhythmia, because of its high incidence and the potential for adverse outcomes.24 AF increases the risk of stroke and other thromboembolic events and CHF. The rate of hospitalization for AF in Canada is approximately 583 per 100,000 persons, with 3% readmission within a year due to stroke.24
2 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses The first-line medical therapy for AF is AADs.24,28 Some AADs may be used to abate the arrhythmia in AF of recent onset (cardioversion). Others may be used to control the heart rate when AF is persistent.
Drug therapy is a non-invasive and commonly available option, but chronic administration may be needed. On the other hand, ablation of AF may prevent the use of long-term cardiac antiarrhythmic therapy. There is, however, uncertainty about the health impact of this treatment (Is ablation enough to prevent AF from recurring or is ablation mainly improving symptoms?) and its place in therapy (Is it to be considered in minimally symptomatic patients?). Therefore, there is a need to compare the clinical and cost-effectiveness of the minimally invasive ablation procedures with those of surgical ablation procedures and those of other modalities for converting AF to NSR, including drug therapy and pharmacological or electrical cardioversion. There is also a need to identify the populations in which AF ablation interventions are clinically and cost-effective.
Ablation procedures for AF are funded in at least five Canadian provinces. There is interest among some jurisdictions in developing guidelines on the use of ablation procedures for the management of patients with AF. In this technology assessment report, only ablation procedures that are aimed at controlling cardiac rhythm in adults with AF are evaluated. Procedures that are mainly aimed at controlling the heart rate (AV node ablation) are excluded because these types of procedures are not subject to policy development.
3 OBJECTIVES
The aims of this health technology assessment (HTA) were to: Compare the clinical and cost-effectiveness of minimally invasive ablation procedures for AF with those of other modalities for converting AF to NSR, including pharmacological or electrical cardioversion, and with those of more invasive surgical procedures. Evaluate the impact of minimally invasive procedures on patients with paroxysmal AF, patients with persistent AF, patients who have never used drugs to treat AF (drug naive), patients who have used drugs to treat AF, and patients with CHF.
4 RESEARCH QUESTIONS
1. In adults with AF, what is the comparative clinical effectiveness of the minimally invasive ablation procedures for AF? 2. In adults with AF, what is the comparative cost-effectiveness of the minimally invasive ablation procedures for AF? 3. In adults with AF, what is the comparative clinical effectiveness of the minimally invasive ablation procedures versus alternative interventions (pharmacological or electrical cardioversion, or surgical procedures)? 4. In adults with AF, what is the comparative cost-effectiveness of minimally invasive ablation procedures versus alternative interventions (pharmacological or electrical cardioversion, or surgical procedures)?
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 3 Clinical and Cost-Effectiveness Analyses 5. What recommendations on the use of minimally invasive ablation procedures are included in Canadian and international guidelines for AF? 6. What is the level and strength of the evidence supporting the recommendations on the use of invasive ablation procedures that are included in Canadian and international guidelines for AF? 7. What is the expected budget impact on the Canadian provinces and territories with the provision of minimally invasive ablation procedures for AF to adults? 8. What are the expected planning issues (for example, quality measures on the volume of ablation procedures) in the Canadian provinces and territories with the provision of minimally invasive ablation procedures for AF to adults?
5 CLINICAL REVIEW 5.1 Methods
5.1.1 Literature searches
A literature search was conducted for the clinical review. All search strategies were developed by an information specialist with input from the project team and were peer reviewed. All search results were imported into a Reference Manager Version 11 database for de-duplication and title and abstract screening.
The following bibliographic databases were searched through the Ovid interface: Medline, MEDLINE In-Process & Other Non-Indexed Citations, and Embase. Parallel searches were run in PubMed (for non-Medline records only) and ISI’s BIOSIS Previews, which was searched through ISI’s Web of Knowledge. We ran searches in Wiley’s Cochrane Library (Issue 3, 2009) including Cochrane Database of Systematic Reviews, Database of Abstracts of Reviews of Effects, Cochrane Central Register of Controlled Trials (CENTRAL), Cochrane Methodology Register, and HTA Database. The search strategy comprised controlled vocabulary, such as the National Library of Medicine’s MeSH (Medical Subject Headings), and keywords. The main search concepts were “atrial fibrillation” and “catheter ablation.” These concepts were combined for the clinical search. The search concept of atrial fibrillation was repeated in the search for guidelines. Methodological filters were applied to limit retrieval for two searches. The first search was limited to randomized controlled trials (RCTs), meta-analyses, systematic reviews, and HTAs. The second search was limited to guidelines. RCTs, systematic reviews, and meta- analyses were managed separately from the guidelines. The search that included the RCTs was limited to the human population. No language or date restrictions were used. An attempt was made to translate all relevant studies, but this was limited to the capabilities of available staff to translate only French and Chinese. See Appendix 1 for the detailed search strategies.
OVID, PubMed, and BIOSIS Previews AutoAlerts were set up to send biweekly updates with new literature. Cochrane searches were updated when new database issues were released. All updates were continued until April 5, 2010.
4 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses Grey literature (literature that is not commercially published) was identified by searching the websites of HTA and related agencies and clinical trial registers. The websites of the following professional associations were searched for relevant evidence (including conference abstracts from 2008-2009): CCS, ACC, AHA, the European Cardiac Arrhythmia Society (ECAS), the European Heart Rhythm Association, and the Society of Thoracic Surgeons. Google and AlltheWeb Internet search engines were used to search for additional information. These searches were supplemented by handsearching the bibliographies and abstracts of key papers and conference proceedings and through contacts with appropriate experts. The Canadian Agency for Drugs and Technologies in Health (CADTH) supplied the documents that were forwarded after CADTH’s liaising with industry.
5.1.2 Selection criteria
a) Inclusion criteria Studies suitable for inclusion were selected from those identified during the literature search, using the following criteria.
Study design Randomized and non-randomized clinical trials of any duration designed to evaluate clinical efficacy, effectiveness, or safety of ablation procedures for AF. Population Adults (18 years of age or older) with AF, regardless of duration or severity of symptoms. Interventions Minimally invasive ablation procedures that convert AF to NSR: pulmonary vein isolation (PVI) ablation (catheter-based approach aiming at applying energy to cardiac electrical pathways [hot spots] originating from PVs to interrupt arrhythmogenic activity), PVI plus atrial ablation, minimally invasive surgical procedure and minimal access catheter Maze procedure (a probe is inserted through chest wall [thoracotomy] to create barriers [by cryotherapy] that interfere with electrical impulses that cause AF). Comparators Cardioversion to NSR, including pharmacological termination of AF, pharmacological maintenance of sinus rhythm, electrical shocks applied through paddles or patches on chest aiming at converting AF to NSR, and open-heart Cox-Maze procedure. Outcomes Freedom from AF (based on any definition), recurrence of AF (based on any definition or diagnostic method), number of tachycardia- flutter episodes (based on any definition or diagnostic method), hospitalization (number of hospital admissions or hospitalized patients), quality of life (measured using any scale), stroke (any cerebrovascular event), mortality.
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 5 Clinical and Cost-Effectiveness Analyses b) Exclusion criteria Studies were excluded if one of the following applied: the study evaluated ablation procedures that were aimed only at controlling the heart rate recruitment was unfinished (this did not include the reports of studies that had stopped early) the study published only baseline characteristics.
5.1.3 Selection method
Two reviewers (NA, KG) independently screened the titles and abstracts for relevance using a predefined checklist (Appendix 2). The kappa statistics for inter-rater agreement was 93%. Any discrepancies between reviewers were discussed until consensus was reached. The full texts of relevant studies were assessed. Two reviewers (NA, KG), using explicit predetermined criteria (Appendix 3), made inclusion and exclusion decisions independently. The kappa statistics for inter-rater agreement was 87%. Any discrepancies between reviewers were resolved by consensus and by consulting a third reviewer when necessary. The results of the guideline search were screened by two reviewers (NA, DR) using predefined criteria (Appendix 4) to identify practice (treatment) guidelines from cardiologic and thoracic professional organizations.
5.1.4 Data abstraction strategy
Data from all included studies were extracted using predefined data abstraction forms (Appendix 5). Relevant data were extracted from the texts, and in some cases, data were extracted from the figures. The data abstraction was performed by two independent reviewers. Any disagreements were discussed until consensus was reached.
5.1.5 Strategy for validity assessment
The methodological quality of all included clinical trials was assessed using the Jadad scale29 (Appendix 6). One reviewer assessed all the included studies for methodological quality, and a second reviewer verified the quality scores. The trials were considered to be of higher quality if their scores were three or greater on the 5-point Jadad scale. Studies were included in the meta- analyses irrespective of methodological quality. The methodological quality of the included guidelines and the strength of the recommendations were assessed using the AGREE instrument30 (Appendix 7). This instrument provides a framework for assessing the quality of clinical practice guidelines and confidence that the potential biases of guideline development were addressed adequately and that the recommendations are internally and externally valid. The instrument addresses the quality of the reporting, the quality of aspects of the recommendations, and the practical issues. The higher the AGREE score, the greater confidence that one has in the recommendations.
5.1.6 Data analysis methods
When two or more comparable studies were identified, a pooled estimate of effect was obtained in a meta-analysis. The comparability of the studies was assessed after reviewing the population, interventions, and outcomes. Review Manager 531 was used to synthesize the data. A random- effects model (inverse variance weighting method) was used for pooled estimates and the corresponding confidence intervals (CI), with the assumption that the treatment effects in the
6 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses studies are different but related. A pooled relative risk (RR) was used to summarize the effect on dichotomous data. RR, as an indicator of the treatment effect, is the ratio of the risks (incidence/mortality rates) in two treatment groups. It shows how much risk (or benefit, depending on the study outcome) is increased (RR greater than 1) or decreased (RR less than 1) in one group versus another group. No data were imputed because all the included studies provided event rates in summary tables, Kaplan-Meier curves, or text. The heterogeneity between studies was assessed using the χ2 test for heterogeneity and the I2 statistic,32 which describe the percentage of total variation across studies that is the result of heterogeneity instead of chance. The heterogeneity was considered to be statistically significant if the P value was less than 0.1. For I2 values greater than 70%, more investigation was undertaken to explore the potential sources of heterogeneity. The quality of studies was not weighted in the meta-analyses. A subgroup analysis was performed to investigate the effect of study quality on the effect size or direction by comparing RCTs with a Jadad score of three or greater (higher quality) to studies with a quality score of less than 3. To estimate the potential covariate effect that the blanking period (the period during which recurrences of AF or atrial tachyarrhythmias [ATs] are not counted as treatment failure) might have on the outcome, meta-regression using STATA 11.033 (metareg command) was performed. P values less than 0.05 were considered to be statistically significant for a covariate effect. Publication bias was assessed from the generation of funnel plots. The Egger test34 and the Harbord test35 for funnel plot asymmetry were performed using STATA’s metabias command. The Egger test detects the asymmetry of a funnel plot by determining whether the intercept deviates statistically significantly from zero in a regression of the standard effect estimates against precision. This test may result in false-positive results when all trials are the same size or if there are large treatment effects or few events per trial.
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 7 Clinical and Cost-Effectiveness Analyses Figure 1: QUOROM Flowchart — Clinical Review
Potentially relevant citations identified and screened for retrieval (Level 1) [2,648]*
Citations excluded after title and abstract screening [2,362]
Articles retrieved for full text screening (Level 2) [286]
Articles excluded after full-text screening [150] Patient education material [1] Study description only [3] Qualitative or academic review [28] Not study type of interest [8] Not primary report of study [4] No control group or inappropriate control group [50] No relevant outcomes [7] Meta-analysis, systematic review, or health technology assessment [33] Irrelevant study goal [6] Unable to translate [5] Inappropriate study population [3]
Articles screened for relevant comparisons (Level 3) [136]
Articles excluded [96], because they compared: ablation approaches [52] Catheter types [7] mapping techniques [19] Cox-Maze versus surgical ablation or other Maze procedures [13] Tricuspid-inferior vena cava ablation versus DC-cardioversion [2] Medication ± ablation (pulmonary vein isolation, or vena cava or isthmus ablation) [3]
Articles excluded during data abstraction: Duplicate publication of study already included [10]
Articles included [30] Pulmonary vein isolation versus pulmonary vein isolation plus atrial ablation [20] (17 publications reporting 17 randomized controlled trials and 3 publications reporting 3 non-randomized controlled trials) Pulmonary vein isolation versus medication [9] (7 publications reporting 6 randomized controlled trials and 2 articles reporting 2 non- randomized controlled trials) pulmonary vein isolation versus electrical cardioversion [1] (non- randomized controlled trial)
*Numbers in brackets indicate number of citations, not number of studies.
8 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses The Harbord test, which is a modified version of the Egger test, maintains its power but reduces the false-positive rate. For the studies that were not comparable in population, interventions, or outcomes, a narrative description of studies in text and tables is provided.
5.2 Results
5.2.1 Quantity of research available
Of the 2,648 potential citations that were identified during the systematic search, 2,362 were excluded during the title and abstract review because of irrelevance. The full texts of the remaining 286 articles were retrieved. Of these 286 articles, 150 did not meet the eligibility criteria and were excluded. Ninety-six articles were excluded because they did not address relevant comparisons. Ten citations with repeating overlapping data were excluded. Thirty articles meeting the inclusion criteria were maintained for this review. Figure 1 shows the QUOROM flowchart of the process that was used to select studies for the review and the main reasons for exclusion. The list of the excluded studies and the reasons for exclusion appears in Appendix 8.
5.2.2 Study characteristics
Thirty citations reporting on 23 RCTs and six non-RCTs (non-randomized controls trials) were included in the review.27,36-64 All the RCTs had a parallel group design and were published in English. All the included studies appeared in at least one peer-reviewed publication.
Seven publications reported on six RCTs55-61 and two non-RCTs62,63 comparing catheter ablation to medical treatment (rhythm control) in patients with AF. The sample sizes ranged from 15 to 106 per arm in RCTs and from 60 to 589 per arm in non-RCTs. One pilot RCT55 used ablation as first-line treatment; the remaining five RCTs and the non-RCTs included patients with AF who had used AADs that had failed. Another pilot trial included only patients with AF and type 2 diabetes. Three RCTs56,58,59 recruited only patients with paroxysmal AF. Two articles reported the results of the same trial;59,60 one abstract60 reported the longer term outcomes. All the full articles reported details of the predefined exclusion criteria. The treatment allocation was not concealed in any of the trials because of the study interventions. Two RCTs, which reported details about the randomization procedures, had a Jadad score of three or greater.55,56
Seventeen RCTs27,36-51 and three non-RCTs52-54 comparing PVI with PVI plus atrial ablations were identified. In the RCTs, the sample sizes varied from 30 to 280 per arm; and in the non- RCTs, from 30 to 100 per arm. All the studies included adult drug-refractory patients. One RCT41 was restricted to patients with persistent AF, and four trials42,46,49,50 included only patients with paroxysmal AF. Predefined exclusion criteria were reported in 11 RCTs and in none of the non-RCTs. Seven trials27,37,41,46-49 had a Jadad score of three or greater. Patients were blinded to treatments in three RCTs.37,39,46 In two RCTs,37,46 treatment success was determined by evaluators who were blinded to the interventions.
One non-RCT compared catheter ablation with electrical cardioversion.64 No studies compared minimally invasive catheter ablation with surgical procedures for the treatment of AF.
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 9 Clinical and Cost-Effectiveness Analyses The characteristics, participants’ eligibility criteria, methodological aspects, and the interventions that were examined appear in Appendix 9. The baseline characteristics of the participants in the studies appear in Appendix 10.
5.2.3 Data analyses and synthesis
a) Comparative clinical effectiveness of minimally invasive ablation procedures versus alternative interventions Catheter ablation versus pharmacological treatment Success rate: Six RCTs55-59,61 and two non-RCTs62,63 compared the clinical effects of catheter ablation with those of AADs in the treatment of AF. Freedom from AF recurrence at the end of the first year was reported as the primary outcome in all but one study,59 which used freedom from documented ATs as the definition of success. One RCT55 compared the effects of catheter ablation as a first-line therapy with those of medical treatment. The remaining RCTs and non- RCTs included the patients for whom at least one AAD had failed (second-line therapy). All but one61 of the RCTs used a blanking period ranging from two weeks to three months. Two RCTs allowed re-ablation procedures within the first three months after the first procedure.56,58 In another RCT59 re-ablations were permitted after a six-week blanking period. In the first-line therapy RCT,55 repeated ablations were performed after the end of the first year. Table 1 shows the intervention characteristics; definition of treatment success; and type, duration, and frequency of monitoring to determine treatment success across the studies. Details about the interventions appear in Appendix 9A.
Table 2 shows that all the studies reported statistically significant higher success rates in patients who underwent ablation procedures than those who received medical treatment. When the data from all the RCTs were pooled, the summary estimate of the effect size was RR = 2.82 (95% CI 2.13 to 3.74), which means that catheter ablation had an approximately three times higher success rate than treatment with AADs. A meta-analysis of the data from only second-line therapy studies showed similar results (pooled RR 2.93, 95% CI 2.09 to 4.11) (Table 3). The forest plots appear in Appendix 11.
Meta-regression analysis was done to investigate the effect of a blanking period on the overall effect. Including the length of the blanking period as a covariate in the model suggested a statistically non-significant increase of effect (favouring catheter ablation; P = 0.061) with increasing duration of the blanking period (months). The regression equation is: effect size = 1.87 + 0.59*blanking period.
A subgroup analysis was undertaken to explore the potential impact of study quality on the results. The pooled effect sizes were similar in the groups of higher quality (Jadad score of 3 or more) and lower quality (Jadad score less than 3) studies (Table 3).
There was an insufficient number of studies to assess publication bias. Ten is typically suggested as the minimum number of studies that is needed to assess publication bias with funnel plots.65
Effects on subpopulations of patients with AF: Of the six RCTs comparing catheter ablation with medication, three56,58,59included only patients with paroxysmal AF. The other three RCTs55,57,61 included patients with paroxysmal AF and patients with persistent AF. Separate
10 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses success rates were not reported for the different types of AF. The data from the RCTs including only patients with paroxysmal AF were pooled, resulting in a RR 3.80 (95% CI 2.92 to 4.96), which means that patients with paroxysmal AF who underwent ablation had an approximately four times higher success rate than patients who received medication. No heterogeneity was found between the studies.
One RCT55 included only patients who had never used AADs to treat AF (first-line therapy). In this study, the catheter ablation group had a statistically significantly greater success rate, compared with patients on medical treatment (RR 2.36, 95% CI 1.50 to 3.70). This was comparable to the overall effect size that was obtained from pooling the data from second-line therapy RCTs (pooled RR 2.93, 95% CI 2.09 to 4.11).
None of the studies included patients with AF and CHF.
Quality of life: Five of the six included RCTs and one non-RCT comparing catheter ablation with medical treatment in patients with AF measured health-related quality of life using the SF- 36. A pooled analysis could not be undertaken because of inconsistent reporting (Table 4).
Wazni et al.55 found that, at six months, catheter ablation as a first-line treatment of AF can lead to a statistically significant improvement in all four domains of physical health and social functioning, compared with medical treatment. Wilber et al.56 reported statistically significantly higher scores for the physical and mental components in the ablation group relative to the medication group, at the end of the third month. Similar results were reported in the non-RCT by Pappone et al.62 Jais et al.58 found a statistically significant difference between the two treatment groups in the physical component scores. Krittayaphong et al.61 showed patients who underwent catheter ablation had a statistically significantly higher score in the general health domain of the physical component at 12 months than did patients who received amiodarone.
All studies reported an improvement in SF-36 scores greater than what is considered to be minimal clinically important differences (3 to 5 points)66 in the general health domain55,57,61 or in mental health and physical health.56,58,62
Adverse events and safety: There was limited reporting of adverse events. Six RCTs55-59,61 and two non-RCTs62,63 reported adverse events that were associated with catheter ablation or AADs. The details and timing of adverse events were not reported in all these studies. Patients were followed up to 12 months in all but one of the studies that had a three-year follow-up.62 Table 5 shows the reported incidence of adverse events in each study.
One RCT61 reported one procedure-related stroke during the one-year follow-up. Three cases of mild or moderate PV stenosis were reported in 85 patients who underwent AF ablation in two RCTs.55,58 Cardiac tamponade after ablation was reported in one RCT (4%)58 and one non-RCT (0.7).62 Three RCTs56,57,61 and one non-RCT63 reported drug-related adverse events with the incidence ranging from 5.2% to 46.7%.
In a non-RCT, Pappone et al.62 reported longer term adverse events.62 The results of this study indicated a higher risk of CHF, MI, and cerebrovascular events in the medication group, compared with patients who underwent ablation. Patients in the ablation group also had a higher
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 11 Clinical and Cost-Effectiveness Analyses probability of adverse event-free survival than did medically treated patients. The authors reported a statistically significant decrease in the number of hospitalizations among ablated patients regardless of AF recurrence. A 54% reduction in death rate was observed in the ablation group, during the three years of follow-up (P < 0.001). Any of the following factors were reported to increase the mortality risk: coronary heart disease, ejection fraction less than 45%, left ventricular mass index greater than 25g/m2, or age greater than 65 years.
Catheter ablation versus electrical cardioversion Success rate: In one non-RCT comparing catheter ablation with electrical cardioversion in the treatment of AF,64 85 patients with drug-refractory AF who underwent PVI were compared with 85 age-, sex- and heart disease-matched patients with persistent AF who received electrical cardioversion. In the ablation group, the recurrence of AF was evaluated using 24-hour Holter monitoring at one, two, three, six, and 12 months after ablation, and every six months thereafter. Holter monitoring was also used for the evaluation of AF recurrence in the electrical cardioversion group, but the frequency of follow-up is not reported. An eight-week blanking period was used. At the end of follow-up (15 ± 7 months), 82% of the patients in the PVI group (70/85) maintained a sinus rhythm, compared with 40% of the patients in the electrical cardioversion group (34/85).
Quality of life: No data were reported on quality of life in this study.
Adverse events and safety: The results of the study comparing PVI with electrical cardioversion revealed a lower but not statistically significant incidence of cerebrovascular events in the ablation group (1/85) compared with electrical cardioversion (5/85, P = 0.059). The diagnostic method(s) that were used for the assessment of stroke was not reported. Of the patients in the PVI group, 72% had a high risk of stroke at baseline. Of those in the cardioversion group, 76% had a high risk of stroke at baseline. Moderate PV stenosis was observed in 7% of the patients who underwent PVI. No hematoma, pericardial effusion, or need for transfusion was reported in this group.
Minimally invasive catheter ablation versus surgical procedures No studies evaluated the effectiveness of catheter ablation procedures relative to open heart surgery for the treatment of AF.
12 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses
Table 1: Interventions, Definition of Success, and Outcome Monitoring in Studies Comparing Catheter Ablation with Medical Treatment in Patients with AF Study Comparison Definition of Treatment Type and Frequency of Monitoring Other Follow-Up Blanking Success to Decide Treatment Success Measurements Period RCTs First line Wazni et al.55 PVI versus Freedom from 1-month event-recorder 1st and 3rd CT scan after 3 2 months AADs symptomatic or months (recorded when patient months (only in asymptomatic AF experienced symptoms or 2 to 3 times ablation group), (> 15 seconds) during daily); after 3 months for patients repeated at 6 and 12 Holter or event monitoring with recurrence of symptoms months in case of in 1-year follow-up 24-hour Holter monitoring before evidence of PV discharge and at 3, 6, and 12 months stenosis Telephone interview monthly SF-36 questionnaire at 6 months Second line Wilber et al.56 PVI ± linear Freedom from documented ECG at 1, 3, 6, 9, and 12 months in CT scan or MRI 3 months for ablation versus symptomatic paroxysmal ablation group and at 5 to 10 days, within 30 days before ablation arm and new AAD (not AF between months 3 and 11 to 21 days, and 3, 6, and 9 months procedure and at 14 days for AAD including Amio) 12, absence of changes in in AAD group 3 and 12 months after arm AADs in ablation group or Transtelephonic monitoring between procedure (only in adverse events requiring months 3 and 12 in both groups ablation group) to drug discontinuation in (transmission of all symptomatic diagnose PV stenosis AAD group cardiac episodes, plus additional SF-36 questionnaire scheduled transmissions irrespective of symptoms — weekly for first 8 weeks, then monthly until last visit) Holter monitoring at baseline and last visit Forleo et al.57 CPVI ± linear Freedom from Patient-reported symptoms SF-36 questionnaire at 5 weeks for ablations (LA, electrocardiographically ECG Holter monitoring at 1 and 3 6 months ablation arm and RA, or both) confirmed episode of AF or and every 3 months thereafter or in 1 month for versus new atypical atrial flutter case of clinical symptoms AAD arm AAD(s) (> 30 seconds) at (including 12 months Amio) in
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 13 Clinical and Cost-Effectiveness Analyses Table 1: Interventions, Definition of Success, and Outcome Monitoring in Studies Comparing Catheter Ablation with Medical Treatment in Patients with AF Study Comparison Definition of Treatment Type and Frequency of Monitoring Other Follow-Up Blanking Success to Decide Treatment Success Measurements Period patients with diabetes Jaїs et al.58 CPVI ± linear Freedom from AF 12-lead ECG, AF symptom SF-36 questionnaire 3 months ablations (LA, (≥ 3 minutes) documented frequency and severity checklist, and (short form) at RA, or both) by ECG or reported by 24-hour Holter monitoring: at baseline and 3, 6, and versus new patient between months baseline and 3, 6, and 12 months 12 months AAD(s) 3 and 12, absence of transthoracic echocardiography at Telephone interview (including recurrent AF after baseline, after each ablation during 3-month Amio) ≤ 3 ablation procedures, or procedure, and at day 365 treatment stabilization changes in AADs during treadmill exercise test at baseline and first 3 months day 365 Pappone et al.59 CPVI + CTI Freedom from documented 12-lead ECG, 48-hour Holter Thyroid function 6 weeks [APAF]* versus new AT (> 30 seconds) during monitoring, and transthoracic tests, hepatic panel, AAD (including 12 months follow-up echocardiography at baseline and 3, serum chemical Amio) 6, and 12 months measurements at event recorder after month 3 3 months (recorded when patient experienced Chest x-ray and symptoms or 1 to 3 times daily) evaluation of potential corneal deposits in patients receiving Amio Krittayaphong CPVI + MIL + Freedom from AF or 12-lead ECG, 24-hour ambulatory SF-36 questionnaire none et al.61 CTI versus maintenance of NSR at ECG monitoring: at 1, 3, 6, and at 6 and 12 months Amio 12 months 12 months Non-RCTs Pappone et al.62 CPVI versus Freedom from 12-lead ECG, 24-hour Holter Patient-reported none AADs symptomatic episode of AF monitoring, and echocardiography adverse events and (> 10 minutes) confirmed on symptom recurrence or routinely hospitalization by ECG at 1, 3, 6, 9, and 12 months, and Review of hospital every 6 months thereafter records, death Patient-reported symptoms certificates, and Telephone interview to document autopsy reports cause of symptoms
14 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses Table 1: Interventions, Definition of Success, and Outcome Monitoring in Studies Comparing Catheter Ablation with Medical Treatment in Patients with AF Study Comparison Definition of Treatment Type and Frequency of Monitoring Other Follow-Up Blanking Success to Decide Treatment Success Measurements Period SF-36 questionnaire at baseline and every 3 months thereafter Lan et al.63 PVI or CPVI Freedom from AF Echocardiography and Holter Chest x-ray, thyroid, and 1 month versus Amio ± (> 30 seconds) documented monitoring at 1, 2, 3, 6, 9, and hepatic function test LO by 12-lead ECG or Holter 12 months every month for first 3 between months 1 and 12 ECG 2 times a month until end months and every 3 of study months thereafter (for Follow-up by telephone every 7 days patients receiving Amio) Physical examination every 15 days
AAD = antiarrhythmic drug; AF = atrial fibrillation; Amio = amiodarone; AT = atrial tachycardia/arrhythmia; CPVI = circumferential pulmonary vein isolation; CTI = cavotricuspid isthmus; CT scan = computed tomographic scan; ECG: electrocardiogram; LA = left atrium; LO = losartan; MIL = mitral isthmus line; MRI= magnetic resonance imaging; NSR: normal sinus rhythm; PV = pulmonary vein; PVI = pulmonary vein isolation; RA = right atrium; RCT = randomized controlled trial; SF-36 = 36-Item Short Form Health Survey. *Longer term results of study published in abstract by Santinelli et al.,60 details of outcome monitoring not provided.
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 15 Clinical and Cost-Effectiveness Analyses Table 2: Clinical Success Reported in Studies Comparing Catheter Ablation with Medical Treatment in Patients with AF Study Reported Success Outcome Success Rate P Value [test] Reported Repeat Ablation Medication Ablations Number of Number of Number of Events/Total (%) Events/Total (%) Events/Total (%) RCTs First line Wazni et al.55 Freedom from documented AF 28/32 (87.5) 13/35 (37.1) < 0.001 [χ2] Re-ablations after 1 year (> 15 seconds) at 12 months 4/32 (12.5) Second line Wilber et al.56 Freedom from documented 70*/106 (66) 10*/61 (16) 0.001 [log-rank] Up to 2 re-ablations/ symptomatic AF at 12 months patient allowed within 80 Freedom from symptomatic 74*/106 (70) 12*/61 (19) 0.001 [log-rank] days after first procedure recurrent AT at 12 months 13/106 (12.6%) Freedom from AT at 12 months 67*/106 (63) 10*/61 (17) 0.001 [log-rank] Forleo et al.57 Freedom from documented AF 28/35 (80) 15/35 (42.9) 0.001 [log-rank] Not reported (> 30 seconds) at 12 months Jaїs et al.58 Freedom from AF (≥ 3 minutes) 46/53 (89) 13/55 (23) < 0.0001 [log-rank] Re-ablations (1 to 3 / off AADs documented or reported patient) within 90 days by patient at 12 months from first procedure 23/53 (43.4) Pappone et al.,59 Freedom from documented AT 85/99 (86) 22/99 (22) < 0.001 [log-rank] Re-ablations allowed Santinelli et al. (> 30 seconds) at 12 months after 6 weeks from first abstract60 [APAF at 4 years 90/99 (91) 80/99 (81) 0.023 [not reported] procedure: 1 year 6/99 and APAF2 ] (6); 4 years 18/99(18) Krittayaphong et Freedom from AF at 12 months 12/15 (78.6) 6/15 (40) 0.018 [log-rank] Not reported al.61 Non-RCTs Pappone et al.62 Freedom from documented AF 479/589 (81) 354/582 (61) < 0.001 [log-rank] Not reported (> 10 minutes) at 1 year at 2 years 282/589 (48) 207/582 (35) at 3 years 135/589 (23) 97/582 (17) Lan et al.63 Freedom from documented AF CPVI 32†/60 (53.3) Amio 25†/60 (41.7) 0.003 (Amio + LO versus Not reported (> 30 seconds) at 12 months SPVI 46†/60 (76.6) Amio + LO 47†/60 (78.3) CPVA) [log-rank] AAD = antiarrhythmic drug; AF = atrial fibrillation; AT = atrial tachycardia or arrhythmia; RCT = randomized controlled trial. *Calculated using reported percentages and total number of participants in each treatment group. †Calculated by subtracting number of reported recurrence events from total number of participants in each treatment group.
16 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses Table 3: Pooled Analyses for Clinical Success After One Year Reported in Studies Comparing Catheter Ablation with Medical Treatment in Patients with AF Group Number of Number of Participants Pooled Risk Ratio Test for Overall Test for Studies AF/AT-Free / Total (95% CI) Effect Heterogeneity PVI PVI+ Random Effect Model RCTs All AF types First line 1 28/32 13/35 2.36 (1.50 to 3.70) Z = 3.73 (P < 0.001) Not applicable Second line 5 238/308 66/265 2.93 (2.09 to 4.11) Z = 6.25 (P < 0.001) χ2 = 9.47 (P = 0.05) I2 = 58% All subgroups 6 266/340 79/300 2.82 (2.13 to 3.74) Z = 7.20 (P < 0.001) χ2 = 10.29 (P = 0.07) I2 = 51% Paroxysmal AF Studies reporting success 3 198/258 45/215 3.80 (2.92 to 4.96) Z = 9.86 (P < 0.001) χ2 = 0.03 (P = 0.99) rate for paroxysmal AF I2 = 0% Methodological quality (Jadad score) ≥ 3 2 95/138 23/96 2.90 (1.80 to 4.68) Z = 4.38 (P < 0.001) χ2 = 1.71 (P = 0.19) I2 = 42% < 3 4 171/202 56/204 2.76 (1.85 to 4.12) Z = 4.99 (P < 0.001) χ2 = 8.58 (P = 0.04) I2 = 65% Non-RCTs All non-RCTs 2 557/709 426/582 1.23 (1.00 to 1.50) Z = 1.99 (P = 0.05) χ2 = 3.84 (P = 0.05) I2 = 74%
AF = atrial fibrillation; AT = atrial tachycardia or arrhythmia; CI = confidence interval; PVI = pulmonary vein isolation; PVI+ = PVI plus atrial ablations; RCT = randomized controlled trial. *Forest plots appear in Appendix 11.
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 17 Clinical and Cost-Effectiveness Analyses
Table 4: Quality-of-Life Measures (SF-36) Reported in Studies Comparing Catheter Ablation with Medical Treatment in Patients with AF Study Time of Treatment Quality-of-Life Measure* Assessment (sample B: Mean at Baseline, A: Mean after Intervention, C: Mean Change from Baseline (months) size) [mean difference between groups, P value] Physical Health Mental Health Physical Physical Pain General Vitality Social Emotional Mental Functioning Role Health Functioning Role Health Wazni et 6 Ablation (32) B 71 ± 3, B 73 ± 5, B 71 ± 3, B 57 ± 2, B 52 ± 4, B 87 ± 3, B 70 ± 1, B 65 ± 4, al.55 A 97 ± 3 A 71 ± 2 A 97 ± 1 A 79 ± 1 A 65 ±1 A 93 ± 3 A 76 ± 1 A 65 ± 2 AADs (35) B 69 ± 2, B 51 ± 5, B 70 ± 3, B 57 ± 2, B 51 ± 1, B 76 ± 3, B 70 ± 1, B 64 ± 2, A75 ± 7.5 [20 A 53 ± 3 [14.9 A 90 ± 3 [6 A 68 ± 2 [11 A 60 ± 2 A 82 ± 2 [9 A 75 ±1 [1 A 68 ± 3 (13.2 to 24.2); (9.9 to 19.9); (1.5 to 9.5); (8 to 14); [4 (1.7 to (7.5 to 11.5); (−4.0 to 4.3); [−4 (−7.5 P = 0.001] P = 0.047] P = 0.004] P < 0.001] 5.7); P = 0.004] P = 0.90] to −3.5); P = 0.21] P = 0.62] Wilber et 3 Ablation (90) C 6.9 (5.2 to 8.6) 8.5 (5.9 to 11.1) al.56 AADs (39) C 0.4 (−1.7 to 2.6) 1.6 (−1.1 to 4.3) [6.6 (3.6 to 9.4); P < 0.001] [6.9 (2.6 to 11.2); P < 0.001] Forleo et 12 Ablation (35) NR NR NR NR NR NR NR NR al.57 AADs (35) NR NR NR NR NR NR NR NR [8.4; P < 0.05] [8.9; [5.9; [7.7; [6.8; P < 0.05] P < 0.05] P < 0.05] P < 0.05] Jaїs et al.58 12 Ablation (53) B 44.8† ± NR, A 52 ± 7.6 B 46.1† ± NR, A 56.6 ± 7.8 AADs (59) B 43:0 ± NR, A 48.9 ± 7.2 [P = 0.015] B 44.0 ± NR, A 51.9 ± 9.7, [P = 0.09] Krittaya- 12 Ablation (15) B 62.7† ± NR, NR NR B 46 ± NR, NR NR NR NR phong et A 85.4† ± NR A 66 ± NR al.61 AADs (15) B 70.8† ± NR, NR NR B 42 ± NR, NR NR NR NR A 68.1† ± NR A 44 ± NR [P = 0.691] [P = 0.048] Pappone et 12 Ablation (109) B 38.7† ± NR, A 50† ± NR B 41.3† ± NR, A 49.5† ± NR al.62 AADs (102) B 39.5† ± NR, A 40.5† ± NR [P = 0.007] B 42.6† ± NR, A 43.9† ± NR [P = 0.004] AA = antiarrhythmic drug; AAD = antiarrhythmic drug; AF = atrial fibrillation; NR = not reported; SF-36 = 36-Item Short Form Health Survey. *Data presented as mean ± standard deviation or mean (95% confidence interval). †Numbers estimated from graphs.
18 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses
Table 5: Adverse Events and Safety Reported in Studies Comparing Catheter Ablation with Medical Treatment in Patients with AF Study Thromboembolic Bleeding or Drug Adverse Event Other Complications [follow up] Event Hematoma Number of Number of Events/Total (%) Number of Number of Events/Total (%) Events/Total (%) Events/Total (%) Ablation Medication Ablation Medication Ablation Medication Ablation Medication Wilber et al.56 0/106 (0) NR NR NR NR 3/57 (5.2) Total treatment-related AEs Total treatment- [12 months] (first 30 days) 5/103 (4.9); related AEs (first 30 pericardial effusion 1/103 days) 5/57 (8.8); life- (0.9); pulmonary edema threatening 1/103 (0.9); pneumonia arrhythmia 2/57 (3.5) 1/103 (0.9); vascular complication 1/103 (0.9); heart failure 1/103 (0.9); death (day 284) 1/103 (0.9) Forleo et al.57 0/35 (0) 0/35 (0) 2/35 (5.7) NR NR 6/35 (17.1) NR NR [12 months] Jaїs et al.58 NR NR Hematom 0/59 (0) NR NR Tamponade 2/53 (4); PV Death (not AAD- [12 months] a 2/53 (4) stenosis 1/53 (2) related) 1/59 (2) Pappone et NR NR NR NR NR NR 24 cardiovascular-related 167 cardiovascular- al.59 APAF- hospital admissions (99) related hospital study [12 admissions (99) months] Wazni et al.55 0/33 (0) 0/37 (0) 2/32 (6.3) 1/35 (2.9) NR NR Hospitalization 3/32 (9); Hospitalization 19/35 [12 months] bradycardia 0/32 (0); PV (54); bradycardia 3/35 stenosis (50% to 70%) 1/32 (8.6) (3); PV stenosis (< 50%) 1/32 (3) Krittayaphong 1/15 (6) 0/15 (0) NR NR Amio- Amio-related NR NR et al.61 related 7/15 (46.7) [12 months] 3/15 (21.4) Lan et al.63 NR NR NR NR See other complications Catheter-related side effects* Severe side effects* non-RCT CPVI 4/60 (6.6); SPVI 3/60 Amio 5/60 (8.3); [12 months] (5.0) Amio + LO 6/60 (10)
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 19 Clinical and Cost-Effectiveness Analyses Table 5: Adverse Events and Safety Reported in Studies Comparing Catheter Ablation with Medical Treatment in Patients with AF Study Thromboembolic Bleeding or Drug Adverse Event Other Complications [follow up] Event Hematoma Number of Number of Events/Total (%) Number of Number of Events/Total (%) Events/Total (%) Events/Total (%) Ablation Medication Ablation Medication Ablation Medication Ablation Medication Pappone62 15/589 52/582(8.9) NR NR NR NR CHF 32/589(5.4); MI 7/589 CHF 57/582(9.7); MI non-RCT [1, (2.5) TIA TIA (1.2); death 38/589 (6.5); 8/582 (1.4); death 2, and 3 8/589 27/582(4.6); tamponade 4/589 (0.7) 83/582 (14.3) years] (1.3); stroke stroke 22/582 (3.7) 6/589 (1)
AAD = antiarrhythmic drug; AE = adverse event; AF = atrial fibrillation; Amio = amiodarone; CHF = congestive heart failure; CPVI = circumferential pulmonary vein isolation; LO = losartan; MI = myocardial infarction; NR = not reported; PV = pulmonary vein; RCT = randomized controlled trials; SPVI = stepwise pulmonary vein isolation; TIA = transient ischemic attack. *Included pericardial tamponade that required pericardiocentesis, moderate to severe pulmonary vein stenosis, and cerebral embolism leading to transient retrograde amnesia. †Included sinus bradycardia, hypotension, clinically significant QT prolongation, hyperthyroidism, hypothyroidism, and hepatic deterioration.
20 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses b) Comparative clinical effectiveness of minimally invasive ablation The comparison between minimally invasive ablation procedures was limited to the comparative effectiveness of PVI versus PVI plus atrial ablation (PVI+) or minimally invasive surgical procedures to convert AF to NSR.
Success rate The clinical success rates of patients who received PVI and those who underwent PVI+ was compared in 17 RCTs27,36-51 and three non-RCTs.52-54 In most of the studies, the maintenance of sinus rhythm (freedom from AF and other ATs) with or without AADs at the end of the first year was used to define success. Four trials38,41,44,47 reported freedom from AF recurrence as the primary outcome. All but three36,38,42 of the RCTs used a blanking period ranging from one to three months. Repeated ablations were reported in all but four RCTs36,38,42,51 and one non-RCT.52 Table 6 shows the intervention characteristics; definition of treatment success; and the type, duration, and frequency of monitoring to determine treatment success. In Table 6, the RCTs are grouped based on the technique that is used for atrial ablation in addition to PVI. The details about ablation techniques appear in Appendix 9C.
Table 7 shows that six of the eight trials using linear ablations in the left atrium in addition to PVI reported statistically significantly higher success rates in the PVI+ arm.36,37,39,41-43 The two RCTs comparing PVI with PVI plus linear ablations in the right atrium did not find any statistically significant difference in success rate between the two study groups.44,45 Two RCTs reported no improvement in the clinical success rate when the SVC, as an ectopic origin of AF, was ablated in addition to PVI.46,47
One of the three RCTs that targeted CFAEs in the PVI+ arm reported a statistically significant difference when PVI was performed with versus without ablation of CFAEs.48 Two non- RCTs52,54 also used the ablation of CFAEs as an adjunctive approach to PVI and compared it with PVI. Lin et al.54 included only patients with non-paroxysmal AF, and the success rate was statistically significantly higher after one ablation in the PVI+ arm (P = 0.035). Verma et al.52 reported a higher but not statistically significant success rate in patients who underwent additional CFAEs ablation, compared with PVI.
Two RCTs compared PVI with tailored or extensive stepwise approaches, in which PVs were targeted first and then additional ablations at atrial sites showing continuous or fractionated electrical activity were performed only if AF persisted or was inducible. Liu et al.27 reported no statistically significant difference in freedom from AF or ATs between patients who had a circumferential PVI and those who underwent a stepwise technique. Khaykin et al.51 showed that the patients who underwent a tailored ablation had a higher rate of success in the long term (2 ± 1 years) than those who had PVI.
Table 8 shows the results of pooled analyses of clinical success data from the RCTs. Studies were included in the meta-analyses regardless of inclusion criteria and length of the blanking period. The data on success rates after one ablation were used, where available. In addition, the freedom from AF-ATs was used as an outcome in the meta-analyses. For the RCTs that reported the success rate separately for groups with different types of AF, the results from each group were added to calculate the total number of events and participants in the treatment arms.
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 21 Clinical and Cost-Effectiveness Analyses The results of the pooled analysis of all the RCTs indicate a small but statistically significant benefit of PVI+ over PVI, in terms of one-year success rate (pooled RR for PVI versus PVI+ comparison 0.92, 95% CI 0.86 to 0.99; P = 0.02). Subgroup analysis was done to explore the contribution of PVI+ to the overall effect (Table 8).When the data were pooled from the studies in each category of PVI+, only PVI plus linear ablations of the left atrium resulted in a statistically significantly higher rate of success (by 15%) than PVI (pooled RR for PVI versus PVI + 0.85, 95% CI 0.76 to 0.95; P = 0.004). There was statistically significant between-study heterogeneity (P =0.03; I2 = 56%). The analysis was repeated after exclusion of Willems et al.’s41 study, which, unlike the other studies in this category, included only patients with persistent AF. The pooled RR in this group remained statistically significant, but heterogeneity decreased (P = 0.34 for heterogeneity; I2 = 12%). The forest plots appear in Appendix 11.
A meta-regression analysis was done to investigate the effect of a blanking period on the overall effect. The inclusion of the length of the blanking period as a covariate in the model suggested a statistically non-significant increase of effect (favouring PVI; P = 0.82) with an increasing duration of the blanking period (months). The regression equation is effect size = 0.93 + 0.002*blanking period.
A subgroup analysis was undertaken to explore the potential impact of study quality on the effect size or direction. The pooled effect sizes were similar in the groups of higher quality (Jadad score ≥ 3) and lower quality (Jadad score < 3) studies (Table 8).
A funnel plot for all the studies that were included in the meta-analysis was constructed to assess the degree of publication bias. The plot revealed a symmetrical distribution around the mean effect size. The results of statistical tests for asymmetry of the funnel plot were not statistically significant (P = 0.67 for Egger test and P = 0.23 for Harbord test). Willems et al.’s41 study seemed to be an outlier, because of the different inclusion criteria (Appendix 12).
Effects on subpopulations of patients with AF All the studies enrolled patients for whom at least one AAD had failed. Of the 17 RCTs included in the comparison of PVI versus PVI+, four RCTs42,46,49,50 included only patients with paroxysmal AF. Of the three RCTs36,39,47 that reported the success rates of patients with paroxysmal and persistent AF, one also reported the results of patients with permanent AF separately.47 When the data from these seven RCTs were pooled, the summary estimate of effect was 0.88 (95% CI 0.81 to 0.95), favouring PVI+. No heterogeneity was found between the studies. In a subgroup analysis, only the success rate that was associated with PVI plus ablation of the left atrium was statistically significant (Table 8).
Two RCTs41,48 included patients with persistent or long-lasting permanent AF. When the results of these studies were pooled with the success rates reported for the patients with persistent or permanent AF in three other RCTs,36,39,47 the summary estimate of the effect was 0.59 (95% CI 0.39 to 0.91). There was a statistically significant heterogeneity between the studies (P = 0.002; I2 = 76%). A subgroup analysis, after the exclusion of the RCTs comparing PVI with PVI plus ablation of SVC47 and PVI plus ablation of CFAEs,48 suggested that patients with persistent AF who underwent PVI plus left atrial ablations had a 56% higher success rate than the patients who
22 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses were treated with PVI (pooled RR 0.44, 95% CI 0.30 to 0.64; P < 0.001). There was no heterogeneity between the studies (Table 8).
Of the three non-RCTs, one study54 enrolled only patients with non-paroxysmal AF. Another study52 reported the success rates of the patients with persistent and permanent AF. Both studies compared PVI with versus without ablation of CFAEs. The pooled estimate of effect size, using data from the two studies, was 0.72 (95% CI 0.35 to 1.47). Statistically significant heterogeneity (P = 0.02; I2 = 82%) may be due to the low number of studies that were included in the meta- analysis.
None of the studies included AF patients with CHF.
Quality of life One RCT comparing PVI with stepwise ablation reported quality of life as a secondary outcome.51 There was no statistically significant difference between the two ablation approaches in any quality-of-life domain (Table 9). Clinically important differences between the two groups (more than three to five points)66 were found in favour of PVI in the physical functioning, role physical, vitality, and social functioning domains. There was no clinically important difference in SF-36 scores in the general health domain.
Adverse events and Harm There was limited reporting of adverse events in the studies. Overall, 15 RCTs27,36-42,44-48,50,51 and three non-RCTs52-54 reported at least one adverse event that was associated with catheter ablation. However, the details and timing of adverse events were not reported in all these studies. In addition, the definition of an important adverse event, such as PV stenosis or a thromboembolic event, varied across the studies. Table 10 shows that the risk of access site hematoma or bleeding varied between 0% to 7.2% and 0% to 5.4% in the PVI and PVI+ groups respectively. Cardiac tamponade or pericardial effusion requiring clinical intervention were reported, at least in one ablation arm, in 12 RCTs36-42,45,47,48,50,51 and non-RCTs53,67 with a rate ranging from 0.6% to 5.5%. The incidence of mild or moderate PV stenosis after ablation was also reported in seven RCTs27,40,42,44,47,48,50 and one non-RCT.53 Overall, PV stenosis was rare in PVI and PVI+ groups (≤ 2%) and no severe stenosis of PV or SVC was reported in any trial. Transient ischemic attacks, stroke, and other thromboembolic complications were reported in eight RCTs.36,38-41,45-47 However, two major procedural strokes were reported in two RCTs (with a total sample size of 179).45,46 None of the studies reported ablation related mortality. One death due to other causes occurred at 12 months after the procedure in one RCT.39 Pooled analysis comparing ablation approaches was not possible because of scarce data and inconsistent reporting.
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 23 Clinical and Cost-Effectiveness Analyses Table 6: Interventions, Definition of Success, and Outcome Monitoring in Studies Comparing PVI with PVI plus Atrial Ablation Study Comparison Definition of Treatment Type and Frequency of Other Follow-Up Blanking Success Monitoring to Decide Measurements Period Treatment Success RCTs PVI versus PVI + ablation of linear lesions in left atrium Fassini et PVI versus PVI + Maintenance of stable SR 48-hour ECG monitoring: in Not stated None al.36 MIL (elimination of AF) hospital after procedure Transtelephonic ECG recording for 2 months, twice daily, or in presence of symptoms 24-hour Holter monitoring: at 1, 3, 6, 9, and 12 months or in presence of symptoms Pappone et CPVI versus CPVI + Freedom from symptomatic Patient-reported symptoms Telephone interviews 6 weeks 37 al. MIL and LA posterior incessant AT Transtelephonic ECG recording at 3-month intervals wall ablation every working day or in presence of symptoms Holter recording at 1 week and 1, 3, 6, and 12 months Sheikh et PVI versus PVI + Freedom from AF on or off ECG at 1, 3, and every 2 months Post-ablation None 38 al. MIL and LA roofline AADs after 1 ablation Holter monitoring if indicated echocardiography to based on symptoms or rhythm assess PV stenosis tracing Recording of clinical Patient-reported symptoms progress at 1, 3, and 9 months 39 Gaita et al. PVI versus PVI + Maintenance of SR (freedom 24-hour ECG monitoring in Clinical evaluations 2 months MIL and LA roofline from documented AF or atrial hospital after procedure at 1, 3, 6, 12, 18 and (CTI ablation in all flutter > 30 seconds) off AADs, 12-lead ECG, echocardiography, 24 months, and every patients) after 1 or 2 procedures and 24-hour Holter monitoring 6 months thereafter at 1, 3, 6, 12, 18, and 24 months and every 6 months thereafter ECG and event recorder in presence of symptoms between follow-up visits Tamborero CPVI + LA roofline Freedom from AF or LA flutter 48-hour ECG monitoring at 1, 4, Clinical evaluations 3 months et al.40 versus CPVI + LA after 1 ablation and 7 months, and every 6 at 1, 4, and 7 months, roofline + line months thereafter in presence of and every 6 months
24 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses Table 6: Interventions, Definition of Success, and Outcome Monitoring in Studies Comparing PVI with PVI plus Atrial Ablation Study Comparison Definition of Treatment Type and Frequency of Other Follow-Up Blanking Success Monitoring to Decide Measurements Period Treatment Success connecting 2 inferior symptoms thereafter in presence PVs (MIL ablation in Patient-reported symptoms of symptoms all patients) Willems et CPVI versus CPVI + Freedom from symptomatic AF 1-minute tele-ECG recording: Transesophageal 4 weeks al.41 MIL and LA roofline or AF episodes (or suspected ≥ 1 ECG per day irrespective of echocardiography and (CTI ablation in all atrial flutter) documented by symptoms, and 1 ECG in MRI: at 6 months patients) tele-ECG (> 30 seconds) presence of symptoms Hocini et CPVI versus CPVI + Freedom from symptomatic or Intra-hospital monitoring CT angiography to None al.42 LA roofline (CTI asymptomatic ATs, off AADs ≥ 5 days after procedure assess PV stenosis: at ablation in all Exercise testing and ambulatory 12 months patients) ECG monitoring: at 1, 3, 6, and 12 months Haїssaguerre PVI versus PVI + Freedom from arrhythmias (AF Intra-hospital monitoring ≥ 3 CT angiography to 1 month et al.43 MIL (CTI ablation in or atrial flutter), off AADs days after procedure assess PV stenosis at all patients) Transthoracic echocardiography, 12 months ambulatory monitoring, and stress testing at 1, 3, 6, and 12 months PVI versus PVI + ablation of linear lesions in the right atrium Wazni et PVI versus PVI + CTI Freedom from AF Loop event recording during Spiral CT-scan to 8 weeks al.44 first month, repeated 3 months assess PV stenosis later; additional recording (time and frequency beyond 3 months in presence of unspecified) symptoms Holter monitoring before discharge and at 3, 6, and 12 months Pontoppidan CPVI versus CPVI + Freedom from AF (> 60 ECG monitoring first 24 hours Clinical evaluations 3 months et al.45 CTI block; CPVI seconds) off AADs documented after procedure at 3, 6, and 12 months included CPVI + MIL by ECG or Holter recording 12-lead ECG and 7-day Holter and LA roofline monitoring: at 3, 6, and 12 ablation in all patients months with persistent AF and proportion of
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 25 Clinical and Cost-Effectiveness Analyses Table 6: Interventions, Definition of Success, and Outcome Monitoring in Studies Comparing PVI with PVI plus Atrial Ablation Study Comparison Definition of Treatment Type and Frequency of Other Follow-Up Blanking Success Monitoring to Decide Measurements Period Treatment Success patients with paroxysmal AF PVI versus PVI+ ablation SVC 46 Wang et al. CPVI versus CPVI + Freedom from symptomatic or Surface ECG at 1 day, 1 week, Telephone interviews 1 month SVC asymptomatic ATs (> 30 1, 2, 3, 6, 9, and 12 months monthly seconds) documented by ECG Event recording in presence of Spiral CT scan to or Holter monitoring symptoms assess PV or SVC Holter monitoring 24 hours after stenosis at 3 months procedure and every 2 months
Corrado et PVAI versus PVAI + Freedom from AF recurrence Intra-hospital monitoring Transthoracic 8 weeks 47 al. SVC (> 30 seconds) 48-hour Holter monitoring at 1, echocardiography at 3, 6, 9, and 12 months 3 months Transtelephonic rhythm Spiral CT scan to transmitters in presence of assess PV stenosis at symptoms, if symptoms not 3 months clarified during routine visits Outpatient visits at Device interrogation in patients 1,3, 6, 9, and 12 with implanted devices to months confirm recurrence of arrhythmia
PVI versus PVI + ablation of CFAEs 48 Elayi et al. PVAI versus CFAEs Freedom of AF/ATs Event recording: 4 times per Not stated 2 months + PVAI (> 60 seconds) week regardless of symptoms, and 1 recording in presence of symptoms, over ≥ 6 months 48-hour Holter monitoring at 3, 6, 9, 12, and 15 months 12-lead ECG outpatient visits (frequency of visits not stated) Device interrogation in patients with implanted devices to confirm recurrence of arrhythmia
26 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses Table 6: Interventions, Definition of Success, and Outcome Monitoring in Studies Comparing PVI with PVI plus Atrial Ablation Study Comparison Definition of Treatment Type and Frequency of Other Follow-Up Blanking Success Monitoring to Decide Measurements Period Treatment Success Di Biase et PVAI versus PVAI + Freedom from AF/ATs Event recording 4 times per Outpatient visits at 2 months al.49 CFAEs (> 60 seconds), on or off AADs week regardless of symptoms, 3 months and every and 1 recording in presence of 3 months thereafter symptoms, over ≥ 5 months 48-hour Holter monitoring at 3, 6, 9, 12, and 15 months Deisenhofer PVI versus PVI + Freedom of ATs (> 30 seconds) 24-hour Holter monitoring at 1 Outpatient visits at 1 1 month et al.50 CFAEs documented with 7-day Holter month and 3 months and ECG or symptomatic AT 7-day Holter monitoring at 3 every 3 months months thereafter Transesophageal echocardiography at 1 month MRI and CT scan to assess PV stenosis PVI versus tailored or stepwise approaches Khaykin et PVAI versus CPVI + Freedom from AF 24-hour Holter monitoring and Spiral CT scan to assess 2 months al.51 roofline and MIL + (> 30 seconds) or organized LA 12-lead ECG at 1, 3, 6, and 12 PV stenosis at 3 months CFAEs tachycardia months Loop event recording or Holter monitoring in presence of symptoms 27 Liu et al. CPVI versus stepwise Freedom from ATs 12-lead ECG and 24-hour Holter Telephone interviews 3 months SPVI (> 10 minutes) documented by monitoring at 2 weeks, and 1, 3, monthly Holter recording, off AADs 6, and 9 months Non-RCTs Verma et PVAI + SVC versus Freedom from AF or atypical ≥ 1 rhythm transmission per day CT scan to assess PV 2 months al.52 PVAI + SVC + atrial flutter off AADs (patient and 1 recording in presence of stenosis at 3 months CFAEs reported or documented) symptoms, over ≥ 3 months post Clinical visits at 3, 6, ablation, and for 3 more months and 12 months in patients with early recurrence 12 lead ECG and 48-hour Holter monitoring: at 3, 6, and 12 months
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 27 Clinical and Cost-Effectiveness Analyses Table 6: Interventions, Definition of Success, and Outcome Monitoring in Studies Comparing PVI with PVI plus Atrial Ablation Study Comparison Definition of Treatment Type and Frequency of Other Follow-Up Blanking Success Monitoring to Decide Measurements Period Treatment Success Device interrogation in patients with implanted devices to confirm recurrence of arrhythmia Patient-reported symptoms Matsuo et SPVI versus SPVI + Maintenance of SR off AADs Patient-reported symptoms Outpatient visits at None 53 al. ablation of ATP- ECG recordings and 24-hour least 1 a month induced dormant ambulatory monitoring at 1, 3, 6, conduction 9, and 12 months 5-day event recorder in presence of symptoms 54 Lin et al. CPVI versus CPVI + Freedom from ATs ECG at 2 weeks, and every 1 to Clinical visits at 2 months CFAEs (> 60 seconds), confirmed by 3 months thereafter 2 weeks, and every 1 ECG 24-hour to 3 months thereafter Holter monitoring or 1-week event recording or both in presence of symptoms
AAD = antiarrhythmic drug; AF = atrial fibrillation; AT = atrial tachycardia or arrhythmia; CFAE = complex fractionated atrial electrogram; CPVI = circumferential pulmonary vein isolation; CT = computed tomography; CTI = cavotricuspid isthmus; ECG= electrocardiogram; MIL = mitral isthmus line; MRI= magnetic resonance imaging; PV = pulmonary vein; PVAI = pulmonary vein antrum isolation; PVI = pulmonary vein isolation; RCT = randomized controlled trial; SPVI = segmental pulmonary vein isolation; SR = sinus rhythm.
28 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses
Table 7: Clinical Success Reported in Studies Comparing PVI with PVI plus Atrial Ablation Study Reported Success Outcome Success Rate P value, Test Reported Repeat PVI PVI+ Ablations Number of Number of Number of Events/Total Events/Total (%) Events/Total (%) (%) RCTs PVI versus PVI + ablation of linear lesions in left atrium Fassini et Maintenance of stable SR [12 months] All AF types 0.01 log-rank Not reported 36 al. 49/92 (53) 67/95 (71) Paroxysmal AF < 0.05 log- 39/63 (62) 48/63 (76) rank Persistent AF < 0.01 log- 10/29 (36) 24/32 (74) rank Pappone et Freedom from AF [12 months] 239/280 (85.7) 244/280 (87.1) 0.57 log-rank 4.5 ± 0.8 months after first 37 al. Freedom from symptomatic incessant AT 220/280 (81) 255/280 (91) < 0.001 log- procedure (in BP) [6 weeks] rank PVI group 28/280 (10); success rate 14% Freedom from symptomatic incessant AT 252/280 (90) 269/280 (96) 0.005 log-rank PVI+ group 11/280 (3.9); (beyond BP) [12 months] success rate 18% Sheikh et Freedom from AF (> 10 minutes) on or off 41/50 (82) 45/50 (90) NS Fisher’s None al.38 AADs after 1 ablation [9 months] exact test Gaita et al.39 Maintenance of SR, freedom from AF Paroxysmal AF 0.020 log-rank Paroxysmal AF (> 30 seconds) PVI group 27/41 (65.8) 19/41 (46) 48/84 (57) After 1 ablation [12 months] Persistent AF 0.017 log-rank PVI+ group 29/84 (34.5) 7/26 (27) 24/53 (45) After 1 or repeated ablations [3 years] Paroxysmal AF 0.003 log-rank Persistent AF PVI group 18/26 (69.2) 24/39 (62) 63/74 (85) Persistent AF 0.003 log-rank PVI+ group 25/53 (47.2) 9/23 (39) 35/47 (75) Tamborero et Freedom from AF or LA flutter after 1 27/60 (45) 27/60 (45) 0.943 log-rank Both groups al.40 ablation [10 ± 4 months] 25/120 (20.8); success 67.7%
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 29 Clinical and Cost-Effectiveness Analyses Table 7: Clinical Success Reported in Studies Comparing PVI with PVI plus Atrial Ablation Study Reported Success Outcome Success Rate P value, Test Reported Repeat PVI PVI+ Ablations Number of Number of Number of Events/Total Events/Total (%) Events/Total (%) (%) Willems et Freedom from asymptomatic or 6/30 (20) 22/32 (69) 0.0001 log- PVI+ group 2/32 (6) al.41 symptomatic AF (> 30 seconds) rank [12 months] Hocini et Freedom from symptomatic or 31/45 (69) 39/45 (87) 0.04 log-rank Not stated al.42 asymptomatic ATs (off AADs) [15 ± 4 months] Haїssaguerre Freedom from arrhythmias (off AADs) 26/35 (74) 29/35 (83) 0.03 log-rank Both groups 11/70 (15.7) et al.43 [7 ± 3 months] PVI versus PVI+ ablation of linear lesions in right atrium Wazni et Freedom from AF within BP 41/59 (70) 33/49 (67) NS log-rank Both groups 10/108 (9.2) 44 al. Freedom from AF beyond BP [12 months] 53/59 (90) 42/49 (86) NS log-rank Pontoppidan Freedom from documented AF 24*/76 (32) 25*/73 (34) 0.71 log-rank Both groups 40*/149 (27), et al.45 (> 60 seconds) off AADs [12 months] CPVI 27*/76 (36), CPVI+ * Maintenance of SR off AADs [12 months] 37/75 (49) 31/68 (46) 0.64 log-rank 13 /73 (18) Maintenance of SR on AADs [12 months] 14/75 (32) 6/68 (9) 0.07 log-rank PVI versus PVI + ablation SVC Wang et al.46 Freedom from symptomatic or 1 ablation 0.75 log-rank CPVI 9/54 (16.7), CPVI+ 8/52 asymptomatic ATs (> 30 seconds) 12/54 (22.2) 10/52 (19.2) (15.4), P = 0.86 [12 months] 1 or multiple ablations 0.73 log-rank 50/54 (92.6) 49/52 (94.2) Corrado et Freedom from AF recurrence All AF types 0.16 χ2 Paroxysmal AF PVI group 47 al. (> 30 seconds) off AADs [12 months] 118/160 (74) 108/134 (81) 11/73 (15) (PVAI + SVC 2 performed), other groups not Paroxysmal AF 0.04 χ reported 56/73 (77) 55/61 (90) Persistent AF 0.52 χ2 30/41 (74) 27/34 (80) Permanent AF 0.77 χ2
30 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses Table 7: Clinical Success Reported in Studies Comparing PVI with PVI plus Atrial Ablation Study Reported Success Outcome Success Rate P value, Test Reported Repeat PVI PVI+ Ablations Number of Number of Number of Events/Total Events/Total (%) Events/Total (%) (%) 32/46 (69) 26/39 (67) PVI versus PVI+ ablation of CFAEs Elayi et al.48 Freedom from AF-ATs (> 1 minute) 1 ablation (off AADs) 0.03 χ2 PVI group 12/48 (25) [16 ± 1 months] 19/48 (40) 30/49 (61) PVI+ group 10/49 (20) 2 ablations (off AADS) 0.01 χ2 27/48 (56) 39/49 (80) 1 or 2 ablations (on AADs) 0.17 χ2 41/48 (83) 46/49 (94) Di Biase et freedom from AF/ATs (> 1 min) On or off AADs NS χ2 Both groups 7/69 (10) 49 al. [12 months] 31/35 (89) 31/34 (91) Off AADs NS χ2 26/35 (74) 26/34 (76) Deisenhofer freedom from ATs (> 30 seconds) after 36/48 (75) 38/50 (76) NS log-rank Not allowed in first 3 months et al.50 1 ablation [3 months] after first ablation, PVI group 15/46 (33) freedom from symptomatic AF-AT 34/46 (74) 40/48 (83) 0.17 log-rank [1.3 procedures/patient) (beyond first 3 months after ablation) PVI+ group 17/48 (35) [19 ± 8 months] [1.4 procedures/patient) PVI versus tailored or stepwise approaches Khaykin et Freedom from AF (> 30 seconds) or 18/30 (60) 10/30 (33) 0.07 log-rank Not reported al.51 organized LA tachycardia in BP Freedom from AF (> 30 seconds) or off AADs 0.02 log-rank organized LA tachycardia beyond BP [2 ± 17/30 (57) 8/30 (27) 1 years] on AADs 0.06 log-rank 24/30 (80) 18/30 (60) Liu et al.27 Freedom from symptomatic ATs after 1 35/55 (63.6) 32/55 (58.2) 0.69 χ2 Not allowed in first 3 months ablation, in BP after first ablation, PVI group 5/55 (9); success 80%,
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 31 Clinical and Cost-Effectiveness Analyses Table 7: Clinical Success Reported in Studies Comparing PVI with PVI plus Atrial Ablation Study Reported Success Outcome Success Rate P value, Test Reported Repeat PVI PVI+ Ablations Number of Number of Number of Events/Total Events/Total (%) Events/Total (%) (%) Freedom from symptomatic ATs off 46/55 (83.6) 43/55 (78.2) 0.63 χ2 stepwise PVI group 7/55 (13); AADs, after 1 or multiple ablations beyond success 57% BP [9 months after last ablation]
Non-RCTs Verma et Freedom from AF or atypical atrial flutter All AF types 0.054 log-rank Not reported 52 al. off AADs [12 months] 80 /100 (80) 85/100 (85) Paroxysmal AF 0.39 log-rank 9/60 (15) 8/60 (13) Persistent or permanent AF 0.047 log-rank 11/40 (28) 7/40 (18) Matsuo et Freedom from AF off AADS 1 ablation < 0.05 log- PVI group 36/94 (38.3); 53 al. [19 ± 6 months] 56/94 (60) 43/54 (80) rank success rate 81% PVI+ group 9/54 (16.7); 1 or multiple ablations NR success rate 67% 85/94 (90) 49/54 (91) Lin et al.54 Freedom from documented ATs [> 1 year 1 ablation 0.035 log-rank PVI group 13/30 (43) to 2 years] 12/30 (40) 21/30 (70) PVI+ group 5/30 (17) Multiple ablations 0.13 log-rank 20/30 (67) 25/30 (83)
AAD = antiarrhythmic drug; AF = atrial fibrillation; AT = atrial tachycardia or arrhythmia; BP = blanking period; CFAE = complex fractionated atrial electrogram; NS = Not Significant; PVAI = pulmonary vein antrum isolation; PVI = pulmonary vein isolation; RCT = randomized controlled trial; SR = sinus rhythm; SVC = superior vena cava. *Numbers calculated using reported percentages and total number of participants in each treatment group.
32 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses
Table 8: Pooled Analyses of Clinical Success After One Year Reported in RCTs Comparing PVI with PVI plus Additional Atrial Ablation* Group Number Number of Participants Pooled Risk Test for Overall Test for of AF/AT-Free/Total Ratio (95% CI) Effect Heterogeneity Studies PVI PVI+ Random Effect Model RCTs All AF types PVI versus PVI + ablation of linear 8 458/659 570/734 0.85 (0.76 to 0.95) z = 2.85 (P = 0.004) χ2 = 15.80 (P = 0.03); lesions in left atrium I2 = 56% PVI versus PVI + ablation of linear 2 90/134 73/117 1.05 (0.92 to 1.20) z = 0.76 (P = 0.44) χ2 = 0.03(P = 0.87); lesions in right atrium I2 = 0% PVI versus PVI + ablation of SVC 2 130/214 118/186 0.92 (0.81 to 1.04) z = 1.32 (P = 0.19) χ2 = 0.36 (P = 0.55); I2 = 0% PVI versus PVI + ablation of CFAEs 3 86/131 99/133 0.92 (0.76 to 1.10) z = 0.93 (P = 0.35) χ2 = 3.48 (P = 0.18); I2 = 42% PVI versus tailored/stepwise 2 53/85 42/85 1.31 (0.82 to 2.10) z = 1.14 (P = 0.26) χ2 = 2.21 (P = 0.14); approaches I2 = 55% All subgroups 17 817/1,223 902/1,255 0.92 (0.86 to 0.99) z = 2.28 (P = 0.02) χ2 = 29.55 (P = 0.02); I2 = 46% Paroxysmal AF PVI versus PVI + ablation of linear 3 89/149 135/192 0.80 (0.69 to 0.94) z = 2.83 (P = 0.005) χ2 = 0.02 (P = 0.99); lesions in left atrium I2 = 0% PVI versus PVI + ablation of SVC 2 68/127 65/113 0.86 (0.74 to 1.00) z = 1.98 (P = 0.05) χ2 = 0.88 (P = 0.35); I2 = 0% PVI versus PVI + ablation of CFAEs 2 65/81 71/82 0.94 (0.83 to 1.07) z = 0.95 (P = 0.34) χ2 = 0.52 (P = 0.47); I2 = 0% All subgroups 7 222/357 271/387 0.88 (0.81 to 0.95) z = 3.22 (P = 0.001) χ2 = 3.69 (P = 0.72); I2 = 0% Persistent or permanent AF PVI versus PVI + ablation of linear 3 23/85 70/117 0.44 (0.30 to 0.64) z = 4.30 (P < 0.001) χ2 = 1.90 (P = 0.39); lesions in left atrium I2 = 0% PVI versus PVI + ablation of SVC 1 30/41 27/34 0.92 (0.72 to 1.19) z = 0.64 (P = 0.52) Not applicable
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 33 Clinical and Cost-Effectiveness Analyses Table 8: Pooled Analyses of Clinical Success After One Year Reported in RCTs Comparing PVI with PVI plus Additional Atrial Ablation* Group Number Number of Participants Pooled Risk Test for Overall Test for of AF/AT-Free/Total Ratio (95% CI) Effect Heterogeneity Studies PVI PVI+ Random Effect Model PVI versus PVI + ablation of CFAEs 1 19/48 30/49 0.65 (0.43 to 0.98) z = 2.06 (P = 0.04) Not applicable
All subgroups 4 53/126 97/151 0.59 (0.39 to 0.91) z = 2.42 (P = 0.02) χ2 = 12.16 (P = 0.002); I2 = 0.76% Methodological quality (Jadad score) ≥ 3 7 473/662 502/636 0.91 (0.81 to 1.02) z = 1.57 (P = 0.12) χ2 = 13.98 (P = 0.03); I2 = 57% < 3 10 344/561 400/619 0.93 (0.83 to 1.03) z = 1.46 (P = 0.14) χ2 = 15.57 (P = 0.08); I2 = 42% Non-RCTs All studies 3 148/224 149/184 0.80 (0.63 to 1.01) z = 1.86 P = 0.06) χ2 = 6.10 (P = 0.05); I2 = 67% Long-lasting AF 2 23/70 28/70 0.72 (0.35 to 1.47) z = 0.91 (P = 0.36) χ2 = 5.64 (P = 0.02); I2 = 82%
AF = atrial fibrillation; AT = atrial tachycardia or arrhythmia; CFAE = complex fractionated atrial electrogram; CI = confidence interval; PVI = pulmonary vein isolation; PVI+ = PVI plus additional atrial ablations; RCT = randomized controlled trial; SVC = superior vena cava. *Forest plots appear in Appendix 11.
34 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses
Table 9: Quality-of-Life Measures (SF-36) in Studies Comparing Catheter Ablation with Medical Treatment in Patients with AF Study Time of Treatment Quality of Life Measure,† B is Mean at Baseline, Assessment (sample A is mean after intervention [P value for mean difference between groups] (months) size) Physical Health Mental Health Physical Physical Pain General Vitality Social Emotional Mental Functioning Role Health Functioning Role Health Khaykin 6 PVAI (30) B 70 ± 25 B 37 ± 42 B 81 ± 20 B 70 ± 18 B 44 ± B 60 ± 24 B 55 ± 46 B 70 ± et al.51 A 79 ± 22 A 62 ± 43 A 82 ± 18 A 65 ± 18 15 A 75 ± 25 A 67 ± 43 15 A 59 ± A 70 ± 18 16 CPVI + B 66 ± 24 B 50 ± 41 B 71 ± 30 B 65 ± 17 B 52 ± B 72 ± 30 B 61 ± 46 B 66 ± roofline and A 67 ± 34 A 60 ± 44 A 74 ± 25 A 63 ± 21 23 A 68 ± 34 A 69 ± 45 20 MIL + [P = 0.29] [P = 0.73] [P = 0.48] [P = 0.79] A 53 ± [P = 0.32] [P = 0.95 A 68 ± CFAEs (30) 22 28 [P = [P = 0.07] 0.83]
AF = atrial fibrillation; CFAE = complex fractionated atrial electrogram; CPVI = circumferential pulmonary vein isolation; MIL = mitral isthmus line; PVAI = pulmonary vein antrum isolation; SF-36 = 36-Item Short Form Health Survey. *Numbers estimated from graphs. †Data presented as mean ± standard deviation or mean (95% confidence interval).
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 35 Clinical and Cost-Effectiveness Analyses
Table 10: Adverse Events and Safety Reported in Studies Comparing PVI with PVI plus Atrial Ablation Study [time of Thromboembolic Bleeding or Cardiac Tamponade PV Stenosis Other assessment] Events Hematoma or Pericardial Number of Events/Total Number of Number of Number of Effusion (%) Events/Total Events/Total (%) Events/Total (%) Number of (%) Events/Total (%) PVI PVI+ PVI PVI+ PVI PVI+ PVI PVI+ PVI PVI+ RCTs PVI versus PVI + ablation of linear lesions in left atrium Fassini et al.36 TIA 0/95 (0) NR NR 0/92 (0) 1/95 (1) NR NR NR NR [12 months] 1/92 (1) Pappone et al.37 0/280 0/280 (0) 3/280 (1) 2/280 2/280 (0.7) 2/280 (0.7) 0/280 (0) 0/280 (0) NR NR [12 months] (0) (0.7) Sheikh et al.38 [9 months] TIA 1/50 0/50 (0) NR NR Tamponade 0/50 (0) 0/50 (0) 0/50 (0) NR NR (2) 1/50 (2), pericardial effusion 1/50 (2) Gaita et al.39 [12 months TIA 2/[67 + 137] (1) NR NR 2/[67 + 137] (1) NR NR Death 1/[67 + for single ablation, ≥ 3 137] (0.5) (not years long term follow-up ablation- including single and related, month repeated procedures] 12) Tamborero et al.40 TIA TIA 1/60 NR NR Pericarditis Pericarditis 0/60 (0) 1/60 (1.6) Transient [10 ± 4 months] 2/60 (3) (1.6) 1/60 (1.6) 2/60 (3) [55% myocardial narrowing] ischemia 1/60 1/60 (1.6) (1.6) Willems et al.41 0/30 (0) 1/32 (3.1) 0/30 (0) 0/32 (0) 0/30 (0) 1/32 (3.1) 0/30 (0) 0/32 (0) NR NR [12 months] [minor]
Hocini et al.42 NR NR NR NR 1/[45 + 45] (1) 1/[45 + 45] (1) Right phrenic [15 ± 4months] injury 1/[45 + 45] (1)
36 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses Table 10: Adverse Events and Safety Reported in Studies Comparing PVI with PVI plus Atrial Ablation Study [time of Thromboembolic Bleeding or Cardiac Tamponade PV Stenosis Other assessment] Events Hematoma or Pericardial Number of Events/Total Number of Number of Number of Effusion (%) Events/Total Events/Total (%) Events/Total (%) Number of (%) Events/Total (%) PVI PVI+ PVI PVI+ PVI PVI+ PVI PVI+ PVI PVI+ PVI versus PVI + ablation of linear lesions in right atrium Wazni et al.44 [12 months] 0/59 (0) 0/49 (0) NR NR NR NR 1/59 (1.6) 1/49 (2) NR NR [asymptomatic [asymptoma- 50% to 75%] tic 50% to 75%] Pontopidan et al.45 0/76 (0) 1/73 (1.4) 1/76 (1.3) 0/73 (0) 2/76 (2.6) 4/73 (5.5) NR NR NR NR [12 months] PVI versus PVI + ablation of SVC Wang et al.46 [12 months] 1/54 + 52 (0.9) Femoral artery NR NR 0/54 (0) 0/52 (0) NR NR [major] pseudoaneurysm 1/54 (2) 1/52 (2) Corrado et al.47 DVT Coronary NR NR 1/160 (0.6) 0/134 (0) 0/160 (0) 1/134 (0.7) NR NR [12 months] 1/160 artery (0.6), embolism CVA 1/134 1/160 (0.7) (0.6) PVI versus PVI + ablation of CFAEs Elayi et al.48 [16.4 ± 0/48 (0) 0/49 (0) NR NR NR 2/49 (4) 1/48 (2) 1/49 (2) NR NR 1 months] Deisenhofer et al.50 [19 ± NR NR NR NR 0/46 (0) Pericardial 1/46 (2) 0/48 (0) NR NR 8 months] effusion [asymptomatic 1/48 (2) < 50%] PVI versus tailored/stepwise approaches Khaykin et al.51 [2 ± 0/30 (0) 0/30 (0) NR NR Pericardial 0/30 (0) 0/30 (0) 0/30 (0) NR NR 1 years] effusion 1/30 (3) Liu et al.27 [9 months] NR NR 4/55 (7.2) 3/55 (5.4) NR NR 1/55 (1.8) 1/55 (1.8) NR NR
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 37 Clinical and Cost-Effectiveness Analyses Table 10: Adverse Events and Safety Reported in Studies Comparing PVI with PVI plus Atrial Ablation Study [time of Thromboembolic Bleeding or Cardiac Tamponade PV Stenosis Other assessment] Events Hematoma or Pericardial Number of Events/Total Number of Number of Number of Effusion (%) Events/Total Events/Total (%) Events/Total (%) Number of (%) Events/Total (%) PVI PVI+ PVI PVI+ PVI PVI+ PVI PVI+ PVI PVI+ Non-RCTs Matsuo et al.53 NR NR NR NR Tamponade Pericardial 2/94 (2.1) 0/54 (0) NR NR [19 ± 6 months] 1/94 (1) effusion [asymptomatic 1/54 (2) 50% to 75%] Lin et al.54 [> 1 year to 0/30 (0) 0/30 (0) NR NR 1/30 (3) 0/30 (0) 0/30 (0) 0/30 (0) NR NR 2 years] Verma et al.52 [12 months] 0/100 (0) 0/100 (0) 1/100 (1) 2/100 (2) 0/100 (0) 0/100 (0) 0/100 (0) 0/100 (0) NR NR Jugular Femoral hematoma hematoma
CVA = cerebrovascular accident; DVT = deep vein thrombosis; NR = not reported; PV = pulmonary vein; PVI = pulmonary vein isolation; RCT = randomized controlled trial; SVC = superior vena cava; TIA = transient ischemic attack.
38 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses c) Treatment guidelines on minimally invasive ablation procedures in AF Of the 3,281 potentially relevant citations that were identified during a search for guidelines on the management of AF in adult patients, 3,185 citations were excluded after title and abstract screening. The full texts of the remaining 96 publications were retrieved. Seventy citations were excluded because they did not meet the eligibility criteria. Of the 26 remaining guidelines, 10 were excluded because of the absence of recommendations on minimally invasive ablation. Sixteen publications on 12 Canadian and international guidelines were kept for this review. Appendix 13 shows the QUOROM flowchart of the process that was used to select guidelines. The excluded guidelines and the reasons for exclusion appear in Appendix 14.
Different grading systems were used in different guidelines to grade the evidence on which recommendations were based (Table 11). The class of recommendation and the level of evidence, with a reference to the grading system, are provided with each recommendation, when available.
In its 2004 consensus conference, CCS recommended catheter ablation for patients with symptomatic paroxysmal AF refractory to one or more AADs.8 Catheter ablation, as a therapeutic option for second-line therapy in all types of AF, was included in the ACC-AHA- ESC revised clinical guideline for the management of AF in 2006.14,68,69 In 2007, a consensus statement was developed by the Heart Rhythm Society, EHRA, and ECAS, with ACC, AHA, and the Society of Thoracic surgeons.17,70 Their goal was to review the literature on catheter and surgical AF ablation and to provide recommendations on indications, techniques, and outcomes. In this consensus statement, the primary indication for catheter ablation was the treatment of symptomatic patients with AF and no response to one or more Class 1 or 3 AADs. The statement suggested that in rare situations catheter ablation could be performed as first-line therapy. The expert groups stated in this document that the isolation of the PVs was the cornerstone for most ablation procedures. They agreed that for patients with persistent AF, ablation of additional lesions might be needed with PVI. The document neither covered the adjunctive ablation sites nor specified the appropriate techniques for ablation of these lesions. In the HRS-EHRA-ECAS consensus statement, surgical ablation was mainly indicated for patients with symptomatic AF who were undergoing cardiac surgery. Other indications for surgical ablation were a patient’s choice or after the failure of catheter ablation procedures.
The clinical guideline on the management of AF, which was developed by the National Collaborating Centre for Chronic Conditions at the Royal College of Physicians,18 provided practice guidance on the diagnosis and treatment of AF. The recommendations were mainly related to medical treatment (rate and rhythm control). Referral for further specialist interventions, including catheter ablation, was recommended for patients with lone AF and those for whom pharmacological treatment had failed or those who had an underlying electrophysiological disorder. The National Institute for Health and Clinical Excellence (NICE) published a series of interventional guidance documents on AF ablation techniques.71-77 These guidelines state that ablation strategies would be generally performed as second-line treatment options. The quantity of evidence for the recommendations on percutaneous radiofrequency ablation,72 microwave ablation,75 and cryoablation76 was considered to be adequate by NICE. The recommendations on epicardial radiofrequency ablation (thoracoscopic and percutaneous) were based on inadequate or limited evidence.71,73
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 39 Clinical and Cost-Effectiveness Analyses The guideline that was developed by the New Zealand Guidelines Group77 recommended catheter ablation for drug refractory patients with paroxysmal AF. This document stated that symptomatic patients with AF who were undergoing open heart surgeries are considered to be potential candidates for undergoing surgical ablation.
The International Society of Minimally Invasive Cardiothoracic Surgery developed an evidence- based consensus statement to make recommendations on surgical ablation.78 In this document, concomitant surgical ablation was recommended for patients with persistent or permanent AF who were undergoing cardiac surgery for valve or coronary bypass grafting.
40 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses Table 11: Recommendations on Use of Ablation Procedures, Included in Canadian and International Guidelines on AF Guideline Year Recommendation [classification/level of evidence] Quality Score† ACC/AHA/ESC guidelines for 2006 Catheter ablation is a reasonable alternative to pharmacological therapy to prevent recurrent AF in 78% management of patients with atrial symptomatic patients with little or no LA enlargement. [Class IIa- Level of Evidence C]* fibrillation14,68,69 HRS-EHRA-ECAS expert 2007 Indications for catheter AF ablation [supporting ACC-AHA-ESC’s class IIa 74% consensus statement on catheter recommendations, level of evidence C]* 17,70 and surgical ablation Symptomatic AF refractory or intolerant to at least 1 Class 1 or 3 antiarrhythmic medication. In rare clinical situations, may be appropriate to perform AF ablation as first-line therapy. Selected symptomatic patients with heart failure or reduced ejection fraction or both. Presence of LA thrombus is contraindication to catheter ablation of AF. Ablation procedures Ablation strategies that target PVs or PV antrum or both are cornerstone for most AF ablation procedures. If PVs are targeted, complete electrical isolation should be the goal. Careful identification of PV ostia mandatory to avoid ablation in PVs. If focal trigger is identified outside PV at time of AF ablation procedure, it should be targeted, if possible. If additional linear lesions are applied, line completeness should be demonstrated by mapping or pacing maneuvers. Ablation of cavotricuspid isthmus recommended only in patients with a history of typical atrial flutter or inducible cavotricuspid isthmus-dependent atrial flutter. If patients with long-standing persistent AF are approached, ostial PV isolation alone may be insufficient. Indications for surgical AF ablation [evidence not provided] Symptomatic patients with AF undergoing other cardiac surgery. Selected asymptomatic patients with AF undergoing cardiac surgery in whom ablation can be performed with minimal risk. Stand-alone AF surgery should be considered for symptomatic AF. Patients who prefer surgical approach, have failed ≥ 1 attempts at catheter ablation, or are not candidates for catheter ablation. Canadian Cardiovascular Society 2004 Catheter ablation for rhythm control 76% consensus conference on atrial In patients with AF and pre-excitation, catheter ablation of accessory pathway is fibrillation8,79 recommended, particularly if associated with syncope, rapid ventricular rates, or if accessory pathway has short refractory period [Class I Level of Evidence B].* In young patients with lone, paroxysmal AF, electrophysiologic study should be considered to exclude re-entrant tachycardia as potential etiology for AF, and if present, curative ablation should be performed [Class IIa Level of Evidence B].*
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 41 Clinical and Cost-Effectiveness Analyses Table 11: Recommendations on Use of Ablation Procedures, Included in Canadian and International Guidelines on AF Guideline Year Recommendation [classification/level of evidence] Quality Score† Patients with highly symptomatic, paroxysmal AF refractory to medical therapy should be considered for ablation procedure aimed at maintaining sinus rhythm [Class IIa Level of Evidence B].† Surgical treatment of AF In patients undergoing mitral valve replacement or repair with history of symptomatic persistent or paroxysmal AF, concomitant intraoperative AF ablation should be considered to increase likelihood of restoration of sinus rhythm. [Class IIa Level of evidence B]† Patients with symptomatic persistent or paroxysmal AF undergoing other cardiac surgery (e.g., coronary artery bypass grafting, aortic valve replacement, or both) may be considered for intraoperative AF ablation. [Class IIb Level of Evidence C]† Patients with refractory, symptomatic AF not associated with organic heart disease and without comorbidities may be considered for surgical ablation when other non-pharmacological procedures have failed. [Class IIb Level of Evidence C]† Atrial fibrillation: national clinical 2006 Referral for further specialist intervention (for example, pulmonary vein isolation, pacemaker 72% guideline for management in therapy, arrhythmia surgery, atrioventricular junction catheter ablation, or use of atrial primary and secondary care, The defibrillators) should be considered in the following patients: ‡ National Collaborating Centre for those in whom pharmacological therapy has failed [B] 18 ‡ Chronic Conditions those with lone AF [B] those with ECG evidence of an underlying electrophysiological disorder (for example Wolff- Parkinson-White syndrome) [C]‡ Thoracoscopic epicardial 2009 There is evidence of efficacy for thoracoscopic epicardial radiofrequency ablation for AF in 62% radiofrequency ablation for atrial short term and in small numbers of patients. Assessment of cardiac rhythm during follow-up fibrillation, Interventional varied between studies, and some patients were concomitantly treated with antiarrhythmic Procedure Guidance 286, NICE73 medication. Evidence on safety shows low incidence of serious complications, but this is also based on limited number of patients. Therefore, the procedure should be used only with special arrangements for clinical governance, consent, and audit or research. Medication for AF may aim to maintain normal cardiac rhythm or control rate of ventricular response, and to reduce risk of thromboembolism (which may cause stroke). Ablation procedures are used when drug therapy is not tolerated or is ineffective. Percutaneous (non-thoracoscopic) 2009 Current evidence on safety and efficacy of percutaneous (non-thoracoscopic) epicardial 63% epicardial catheter radiofrequency catheter radiofrequency ablation for AF is inadequate in quantity. Therefore, this procedure ablation for atrial fibrillation, should be used only with special arrangements for clinical governance and consent. Interventional Procedure Guidance Antiarrhythmic medication is used to help maintain normal cardiac rhythm after successful 294, NICE71 cardioversion or to help reduce heart rate. Ablation procedures can be used when drug therapy is not tolerated or is ineffective.
42 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses Table 11: Recommendations on Use of Ablation Procedures, Included in Canadian and International Guidelines on AF Guideline Year Recommendation [classification/level of evidence] Quality Score† Percutaneous radiofrequency 2006 Current evidence on safety and efficacy of percutaneous radiofrequency ablation for AF 70% ablation for atrial fibrillation, appears adequate to support use of this procedure in appropriately selected patients, provided Interventional Procedure Guidance that normal arrangements are in place for audit and clinical governance. 72 168, NICE Percutaneous radiofrequency ablation is treatment option for symptomatic patients with AF refractory to antiarrhythmic drug therapy or where medical therapy is contraindicated because of comorbidity or intolerance. Radiofrequency ablation for atrial 2005 Current evidence on safety and efficacy of RFA for AF in association with other cardiac 62% fibrillation in association with other surgery appears adequate to support use of this procedure provided that normal arrangements cardiac surgery, Interventional are in place for consent, audit, and clinical governance. 74 Procedure Guidance 121, NICE RFA for AF typically carried out in patients undergoing concomitant open heart surgery (often mitral valve replacement or repair). Procedure involves thermal damage, instead of incisions, to block impulse conduction. Heat generated coagulates heart tissue, forming linear scars or lesions that disrupt transmission of abnormal electrical impulses. Procedure may be carried out on both atria or on left atrium only. It can be performed from within or outside atrium. Microwave ablation for atrial 2005 Current evidence on safety and efficacy of microwave ablation for AF in association with other 62% fibrillation in association with other cardiac surgery appears adequate to support use of this procedure, provided that normal cardiac surgery, Interventional arrangements are in place for consent, audit, and clinical governance. 75 Procedure Guidance 122, NICE Microwave ablation for AF typically carried out in patients undergoing concomitant open heart surgery (often mitral valve replacement or repair). Procedure involves thermal damage, instead of incisions, to block impulse conduction. Heat generated by flexible microwave probe coagulates heart tissue, forming linear scars or lesions that disrupt transmission of abnormal electrical impulses. Procedure may be carried out on both atria or on left atrium only. It can be performed from within or outside atrium. Cryoablation for atrial fibrillation 2005 Current evidence on safety and efficacy of cryoablation for AF in association with other 63% in association with other cardiac cardiac surgery appears adequate to support use of this procedure, provided that normal surgery, Interventional Procedure arrangements are in place for consent, audit, and clinical governance. 76 Guidance 123, NICE Cryoablation for AF typically carried out in patients undergoing concomitant open heart surgery (often mitral valve replacement or repair). Cryoprobe is used to freeze heart tissue. Damaged tissue forms linear scars or lesions that disrupt transmission of abnormal electrical impulses. Procedure may be carried out on both atria or on left atrium only. It can be performed from within or outside atrium. The management of people with 2005 Non-pharmacological approach to maintenance of sinus rhythm (e.g., surgical or catheter 70% atrial fibrillation and flutter, The ablation, or implantable pacemaker) may be justified in selected people with AF-AFL. [B]‡ 77 New Zealand Guidelines Group People with PAF in setting of paroxysmal supraventricular tachycardia should be considered for electrophysiology study and possible pathway ablation. [C]‡ Highly selected people with drug-refractory PAF and structurally normal hearts may be
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 43 Clinical and Cost-Effectiveness Analyses Table 11: Recommendations on Use of Ablation Procedures, Included in Canadian and International Guidelines on AF Guideline Year Recommendation [classification/level of evidence] Quality Score† considered candidates for attempt at catheter ablation. [C]‡ Selected people with problematic AF who are undergoing cardiac surgery for other reasons (e.g., mitral valve surgery) may be candidates for attempt at surgical ablation. [C]‡ Surgical ablation for atrial 2009 In patients with persistent and permanent AF undergoing cardiac surgery, concomitant surgical 64% fibrillation in cardiac surgery, A ablation recommended: † consensus statement of the to increase incidence of sinus rhythm at short- and long-term follow-up [class I, level A] † International Society of Minimally to reduce risk of stroke and thromboembolic events [class IIa, level B] † † Invasive Cardiothoracic Surgery to improve ejection fraction [class IIa, level A] and exercise tolerance [class 2a, level A] (ISMICS)78 and long-term survival [class 2a, level B].†
ACC = American College of Cardiology; AF = atrial fibrillation; AFL = atrial flutter; AHA = American Heart Association; ECAS = European Cardiac Arrhythmia Society; EHRA = European Heart Rhythm Association; ESC = European Society of Cardiology; HRS = Heart Rhythm Society; LA = left atrium; NICE = the National Institute for Health and Clinical Excellence; PAF = paroxysmal atrial fibrillation; PV = pulmonary vein; RFA = radiofrequency ablation. *Based on AGREE instrument for appraisal of guidelines for research and evaluation. †Based on ACC-AHA classification and grading of recommendations.68 ‡Based on SIGN format for grading of recommendation.80
44 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses 6 ECONOMIC ANALYSIS 6.1 Review of Economic Studies: Methods
6.1.1 Literature searches
A peer-reviewed literature search was performed to identify relevant economic analyses assessing the cost-effectiveness of catheter ablation for AF. All search results were imported into a Reference Manager Version 11 database for de-duplication, and title and abstract screening.
The following bibliographic databases were searched through the Ovid interface: Medline, MEDLINE In-Process & Other Non-Indexed Citations, and Embase. Parallel searches were run in PubMed (for non-Medline records only); Scholar’s Portal’s EconLit; Wiley’s Cochrane Library (including NHS Economic Evaluation Database and HTA Database); and Wiley’s Health Economic Evaluations Database (HEED). Social Sciences Citation Index and BIOSIS Previews were searched through the ISI’s Web of Knowledge interface.
Controlled vocabulary terms and keywords were used including terms for “atrial fibrillation” and “catheter ablation.” This terminology was combined for the economic search (Appendix 1).
A methodological filter was used to limit retrieval to primary economic studies or reviews of economic studies. No language or date restrictions were used.
OVID, PubMed, and BIOSIS Previews AutoAlerts were set up to send biweekly updates with new literature. Cochrane searches were updated when new database issues were released. All updates were continued until March 1, 2010.
Grey literature (literature that is not commercially published) was identified by searching the websites of HTA and related agencies and clinical trial registers. The websites of the following professional associations were searched for relevant evidence (including conference abstracts from 2008 to 2009): CCS, ACC, the American Heart Association, the Canadian Heart Association, ECAS, the European Heart Rhythm Association, and the Society of Thoracic Surgeons. Google and AlltheWeb Internet search engines were used to search for additional information. These searches were supplemented by handsearching the bibliographies and abstracts of key papers and conference proceedings and through contacts with appropriate experts. CADTH supplied the documents that were available after liaising with industry.
6.1.2 Selection criteria
The criteria that were used to select studies for the economic review of minimally invasive AF ablation appear in Table 12.
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 45 Clinical and Cost-Effectiveness Analyses Table 12: Selection Criteria for Economic Review Study Design Population Intervention Comparators Outcomes Patient- or Atrial Minimally Rhythm control medications, Costs and model-based fibrillation invasive atrial Open Heart Cox Maze, effectiveness economic fibrillation electrical cardioversion evaluation ablation
6.1.3 Selection method
The study selection was performed in two phases: title and abstract review, and full text review. Two independent reviewers (GB, FX) conducted the title and abstract review using the prespecified inclusion criteria. Based on the titles and abstracts (when available), studies were excluded if both reviewers agreed that they did not meet the inclusion criteria. Studies that were included were retrieved in full text and assessed by one reviewer (GB) to determine if they would be included for data abstraction.
6.1.4 Data abstraction
One reviewer (GB) abstracted data from the economic evaluations using standard forms that were created for this project. A second reviewer (FX) verified the abstracted data.
6.1.5 Data analysis methods
Different structural assumptions and different currencies are often used in economic analysis, even when the same problem is evaluated. This makes it difficult to formally pool the results of economic evaluations. As a result, the economic literature was analyzed using narrative descriptions only.
6.2 Review of Economic Studies: Results
Of the 709 citations that were initially identified during the electronic search (Appendix 15), 51 articles were retrieved for scrutiny. After 47 were excluded (Appendix 16), four economic studies were left for review.81-84
6.2.1 Chan et al.
Chan et al.81 conducted a cost-utility analysis on alternative treatments for 65-year-old patients with AF at a low or moderate risk of stroke. The treatment strategies that were compared included AF ablation, rhythm control medications (amiodarone), and rate control medications (digoxin and atenolol). The costs were based on 2004 US$. The analysis was taken from a societal perspective with a lifelong time horizon.
A decision tree was used to capture short-term treatment, and a Markov model was used to evaluate long-term outcomes. Ablation procedural complications and long-term health sequelae such as stroke, hemorrhage, and drug toxicity were taken into account in the model. It was assumed that patients at moderate risk of stroke were concomitantly treated with warfarin and that patients at low risk of stroke were on warfarin or aspirin therapy.
46 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses It was assumed that the probability of stroke depended on whether patients were in NSR. It was assumed, based on published literature, that the annual risk of stroke in the presence of AF and while the patient was on warfarin was 2.3% and 1.1% among low- and moderate-risk patients respectively. The annual risk of stroke while on aspirin was 1.3% and 0.7% among low- and moderate-risk patients respectively. The probability of stroke among patients in NSR was assumed to be 0.5% per year based on Framingham data.
Based on published literature, it was assumed in the model that 80% of patients undergoing ablation would be in NSR a year after initial treatment and that 2% of patients would revert to AF each year. It was assumed in the model that 85% of rhythm control patients would obtain NSR, 20% would revert to AF in the first six months after the start of treatment, and 5% would revert to AF annually thereafter. The cost per ablation was assumed to be US$16,500, and the annual cost of amiodarone was assumed to be US$1,200. The utility values for warfarin (0.987), aspirin (0.998), and amiodarone (0.987) were obtained from the literature. A disutility of 0.5 was applied to the duration of acute events. The utility weights for mild stroke and moderate or severe stroke were assumed to be 0.76 and 0.39 respectively. It was assumed that the utility weights did not differ between patients in NSR and patients reverting to AF.
For patients at low and moderate risk of stroke, Chan et al.81 found rhythm control medication (amiodarone) to be more costly and to produce fewer quality-adjusted life-years (QALYs) compared with rate control. Because rhythm control medication was dominated by rate control, the incremental cost-utility of AF ablation compared with rhythm control medications was not reported by the authors. The incremental cost-utility can be estimated based on data that were presented in the paper. For patients at moderate risk of stroke, the incremental costs and QALYs of ablation compared with rhythm control were $9,011 and 0.31 respectively. The resulting incremental cost per QALY for AF ablation is $29,068. For a patient at low risk of stroke, the incremental cost, incremental QALY, cost, and cost per QALY of AF ablation compared with rhythm control medications can be calculated as $4,611, 0.38, and $12,134 respectively. The source of funding for the study was not disclosed, but it is stated that two co-authors helped develop AF ablation techniques and had consulted with an AF ablation device manufacturer.
6.2.2 Reynolds et al.
In a cost-utility analysis, Reynolds et al.82 compared radiofrequency ablation with AADs in patients with AF for whom one or more AAD have failed. The time horizon of the model was five years. The costs were based in US dollars. A reference year for costs was not stated. The analysis was taken from a societal perspective.
The evaluation was based on a short-term decision tree representing the initial ablation and on a Markov model representing the long-term phase of the model. In the short-term phase of the model, patients undergoing ablation were at risk of operative death and other operative complications (cardiac tamponade, PV stenosis, and stroke). In the long-term phase, patients are at risk of a recurrence of AF and drug toxicity every one-month cycle. It was assumed that patients undergoing ablation were undergoing a specific sequence of treatments on the recurrence of AF: first-line AAD (Sotalol or Flecainide); re-ablation; second line AAD (Sotalol or Flecainide); and rate control treatment. It was assumed that patients on antiarrhythmic
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 47 Clinical and Cost-Effectiveness Analyses medication start on Sotalol or Flecainide. On the recurrence of AF or drug toxicity, it was assumed that patients switch first to Amiodarone, then to rate control treatment.
The risk of stroke after the initial ablation procedure is not considered in the model structure. The authors justify this assumption by stating that there is no evidence that AAD or ablation reduces the risk of stroke.
The authors state that model parameters were derived from clinical trials, a patient registry, Medicare claims data, and the primary data from the authors’ institution. The efficacy of single ablation was assumed to be 60%, with 25% re-ablation. The overall failure rate after ablation was 10%. The authors assume that 75% of patients on first-line AAD would experience a recurrence of AF within a year and that 65% of patients on second-line therapy would experience a recurrence of AF within a year. The utilities were calculated using the SF6D algorithm and SF36 data from a published registry, a published RCT, and primary unpublished data from the authors. A decrease in utility of 0.065 was applied to patients in model cycles where they had a recurrence of AF.
In a base-case analysis, Reynolds et al.82 estimated the incremental cost of AF ablation treatment compared with AAD to be $6,686. The incremental QALYs were estimated to be 0.13 for the AF ablation treatment. The resulting incremental cost per QALY for ablation was $51,431. One-way sensitivity analyses were conducted. The model was particularly sensitive to the cost of ablation, the time horizon, and utility values. At an ablation cost of $20,000, the authors state that the cost per QALY becomes approximately $100,000. If a three-year time horizon is used, the cost per QALY becomes $157,000. A lifelong time horizon resulted in a cost per QALY of < $1,000. The publication states that the lead author was supported by a grant from the National Institutes of Health. Two co-authors are reported to have consulted with an AF ablation device manufacturer.
6.2.3 McKenna et al.
McKenna et al.83 evaluated the cost-utility of radiofrequency ablation compared with AAD treatment (amiodarone) in patients with AF who are refractory to at least one AAD. The analysis was taken from the perspective of the UK National Health Service and Personal Services, and the costs were based on 2006 British pounds (£).
The model comprised a short-term decision tree representing the initial ablation and a Markov model representing long-term treatment and consequences. Patients undergoing ablation were at risk of operative complications (cardiac tamponade, PV stenosis, stroke, death) and patients on AAD were at risk of drug toxicity. At the end of the 12-month short-term models, patients were classified as being NSR or with AF. In the long-term Markov part of the model, patients in the AF health state had a higher probability of stroke compared with patients in the NSR health state. In both treatment arms, patients were at risk of reverting to AF. Patients who had a stroke entered a post-stroke health state.
The probability of NSR after 12 months was based on a Bayesian meta-analysis of RCTs, case series, and survey data. The 12-month probability of NSR was estimated to be 0.84 for ablation and 0.37 for AAD treatment. The annual reversion rates from NSR to AF were estimated to be 0.29 for AADs and 0.034 for ablation. The baseline risk of stroke was defined according to the
48 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses 85 CHADS2 index. The annual probability of stroke for patients with a CHADS2 score of 0, 1, 2, 3, and 4 was estimated to be 0.019, 0.028, 0.04, and 0.053 respectively. The relative reduction in stroke rate among patients in NSR compared with those with AF (0.60) was taken from an AFFIRM sub-study on the characteristics of stroke events.86
The utility decrements that were associated with AF by treatment and for AF-related health events (for example, stroke) were derived from published studies. The utility decrement for AF was 0.003 for ablation and 0.090 for AAD. A utility decrement of 0.02 was assigned to AAD patients in NSR. No decrement was assumed for patients in NSR after ablation.
The authors presented cost-utility results according to the different risks of stroke as defined using the CHADS2 index. Based on a lifetime time horizon, the incremental cost per QALY of ablation compared with AAD therapy varied from £7,763 (CHADS2 = 0) to £7,910 (CHADS2 = 3) depending on the baseline stroke risk. When a five-year time horizon was used, the cost-utility of catheter ablation ranged from £20,831(CHADS2 = 3) to £27,745 (CHADS2 = 0). The remaining sensitivity analyses were based on the stroke risk of patients with a CHADS2 equal to 1. The cost per QALY increased to £9,327 if it was assumed that there was no differential stroke risk among patients in NSR or AF. If the same utility decrement while in AF was applied to the ablation and AAD treatments, and no utility decrement was applied to AAD while in NSR, the cost per QALY became £12,840. The publication states that the project was supported by a grant from the National Institute for Health Research Technology Assessment Programme. The publication states that no competing interest existed.
6.2.4 Eckard et al.
Eckard et al.84 conducted a cost-utility analysis comparing radiofrequency ablation with AAD treatment in patients with paroxysmal or persistent AF for whom AAD treatment had failed. The analysis was taken from a Swedish societal perspective. A lifelong time horizon was taken. The authors reported the costs in Swedish kronor and in US dollars. The analysis included a one-year decision tree and a long-term Markov model. Patients in both treatment groups were at risk of stroke during the first year of the model. Patients in AF ablation treatment were at risk of procedural complications during the one-year model. At the end of the one year decision model, patients were classified as being with controlled AF or uncontrolled AF. In the long-term Markov model, patients in the controlled AF and non-controlled AF health states were at risk of stroke every annual cycle. Patients in the controlled AF health state could transition to the AF uncontrolled health state in each cycle. Patients could transition to a death health state from all other health states.
The probability of being free of AF (controlled AF) at 12 months was assumed to be 0.78 in the AF ablation group and 0.09 in the AAD treatment group. The authors state that these probabilities were based on data from RCTs comparing AF ablation with AAD treatment. The annual probability of stroke was assumed to be 1.5% for patients in the AF-controlled and AF- uncontrolled health states.
A utility decrement of 0.10 was used for patients with uncontrolled AF, and a 0.25 utility decrement was used for patients after stoke. These utility decrements do not seem to be based on
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 49 Clinical and Cost-Effectiveness Analyses any published data source. The cost of one AF ablation in the model was 90,000 SEK (US$9,860). An average of 1.47 procedures was assumed in the model.
In a base-case analysis, the authors estimated the expected QALYs to be 9.46 in the AF ablation group and 8.68 in the AAD treatment group. The expected costs of the AF ablation group were found to be 232,300 SEK (US$25,460) and 277,700 SEK (US$30,440) in the AAD group. Therefore, the AF ablation treatment group was found to dominate (less expected costs and more expected QALYs) compared with the AAD treatment group. In a sensitivity analysis, the authors ran the model assuming different annual rates of reversion from the AF-controlled health state to the AF-uncontrolled health state for the ablation treatment arm. The cost per QALY in US dollars was reported to be $8,828, $26,460, and $48,310 when the annual probability of reverting to uncontrolled AF is assumed to be 5%, 10% and 15% respectively.
6.3 Primary Economic Evaluation: Methods
6.3.1 Type of economic evaluation
None of the four published economic evaluations of AF ablation compared with AAD was Canada-based. Because of differences in resource costs and, possibly, clinical practice, it is difficult to generalize the cost-effectiveness results of other countries. Therefore a primary Canada-specific cost-utility analysis was conducted using a Markov model for patients with AF. A cost-utility analysis (cost per QALY) allows for the incorporation of the quality-of-life impact of the clinical effects of AF treatment strategies. The use of the cost per QALY outcome allows for comparison with evaluations of other diseases that use this outcome.
6.3.2 Target population
The starting population of the model is 65-year-old males with paroxysmal AF who are unsuccessfully treated with an AAD. Most of the RCTs comparing AF ablation to AADs enrolled patients for whom AAD therapy had failed. It is assumed that the starting population has 85 a CHADS2 stroke risk score of 2. The CHADS2 index is used to predict the annual probability of stroke among individuals based on a risk score ranging from 0 to 6. A point each is assigned to a patient for CHF, hypertension, or diabetes. An additional point is assigned if a patient is 75 years of age or older. Patients with a history of stroke or transient ischemic attack are assigned 2 points. This assumption of a CHADS2 risk score of 2 is based on the mean CHADS2 score of 2.1 among participants in the National Registry of Atrial Fibrillation.85
6.3.3 Comparators
The treatment comparators in this analysis are AF ablation and AAD (amiodarone 200 mg per day).
6.3.4 Perspective
The analysis was taken from the perspective of a publicly funded health care system. The costs include those of drugs that are covered by the provincial formularies for eligible patients, inpatient costs, and physician fees for services that are covered by provincial fee schedules. Indirect costs, such as productivity losses, were not considered in the analysis.
50 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses 6.3.5 Effectiveness
The primary effectiveness measure in the model is the expected QALYs for each treatment comparator. The secondary effectiveness measure is the expected number of strokes.
6.3.6 Time horizon
In the base-case analysis, the time horizon of the model was set to five years with a cycle length of three months. Although the clinical impact of AF treatment may last beyond five years, this time horizon was chosen because of the short-term nature (12 months) of the randomized clinical trials comparing AF ablation with AAD. Alternative time horizons were tested in a sensitivity analysis.
6.3.7 Modelling
The model is similar in structure to that in McKenna et al.’s83 paper. It comprised a one-year decision tree and a longer-term Markov model. Figures 2 and 3 show the short-term and long- term models respectively. A proportion of patients in AF ablation treatment are at risk of operative complications (Figure 2). These include cardiac tamponade, PV stenosis, ischemic stroke, and transient ischemic attack. It is assumed that patients having a stroke incur a permanent disability and start the long-term Markov model in the post-stroke health state. Patients without an operative complication or with a non-stroke complication end the short model in NSR or with AF. Patients in the AAD treatment group follow similar pathways as AF ablation patients during the one-year model. AAD patients are not at risk of operative complications but are at risk of pulmonary toxicity, which can be fatal, reversible, or irreversible. The proportion of patients in NSR after a year is based on freedom from AF outcomes from RCTs comparing AF ablation with AADs.
In Figure 3, patients who are alive and not having a stroke after the one-year model enter the Markov model in the NSR health state or the AF health state. In each three-month model cycle, patients are at risk of ischemic stroke and major bleeding events due to concomitant anti- coagulants. The risk of stroke differs between patients in the NSR and AF health states. Based on treatment algorithms in the RCTs, it is assumed that AF ablation patients discontinue warfarin three months after their procedure, resulting in different bleeding risks between AF ablation patients and AAD-treated patients. It is assumed that the proportion of patients taking warfarin in both treatment groups entering the model was 0.44.87 A proportion of major bleeds are intracranial hemorrhages (ICHs). The remainder of the major bleeds are assumed to be gastrointestinal.
Patients in the AF ablation treatment arm who do not achieve NSR after 1 year or have a subsequent recurrence of AF are assumed to switch to AAD treatment. Patients in the AAD treatment arm who do not achieve NSR after one year or have a subsequent recurrence of AF are assumed to remain on AAD treatment. There are separate health states to distinguish between the first and subsequent years after ischemic stroke and ICH. In every three month cycle, patients in the NSR health state can revert to the AF health state. Each health state is associated with different costs, utilities, and mortality rates.
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 51 Clinical and Cost-Effectiveness Analyses Figure 2: Structure of the Short-Term Model
AAD = antiarrhythmic drug; AF = atrial fibrillation; NSR = normal sinus rhythm.
52 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses Figure 3: Structure of the Long-Term Markov Model
AF = atrial fibrillation; ICH = intracranial hemorrhage; NSR = normal sinus rhythm.
6.3.8 Valuing outcomes
Several clinical input parameters were used to populate the model. These parameters were used to estimate the expected QALYs and the expected number of strokes in each treatment group. These parameters include the probability of achieving NSR in each treatment group, the probability of stroke, the probability of major bleeding, the probability of reverting to AF after achieving NSR, the probability of AF ablation procedural complications, the probability of pulmonary toxicity while on an AAD, utility values for the health states, and mortality associated with the health states. Whenever possible, data from the clinical review were used.
a) Probability of NSR at one year The probability of NSR at one year for AAD treatment was calculated by pooling data from the AAD arms of studies comparing AF ablation with AAD in patients for whom an AAD had failed. Using random effects meta-analysis88 the pooled probability of AAD patients remaining in NSR at 12 months was estimated be 0.26 (95% CI 0.17 to 0.34). Appendix 17 describes the studies that were used in this calculation.
In this report’s clinical review, the pooled RR of being AF free for ablation patients compared with AAD patients was found to be 2.93 (2.09 to 4.11) in RCTs of previous AAD failures. This RR was used to estimate the probability of those in the ablation treatment arm being in NSR at one year. Based on this RR and the probability for the AAD treatment arm, the probability of AF ablation patients being in NSR at one year was estimated to be 0.756 (2.93 0.258).
b) Probability of ischemic stroke The probability of ischemic stroke among patients with AF was based on the CHADS2 85 classification system that was published by Gage et al. The probability of stroke by CHADS2 score85 was estimated using a US registry of patients with AF. Table 13 shows the annual risk of
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 53 Clinical and Cost-Effectiveness Analyses stroke by CHADS2 score. The CHADS2 stroke probabilities were applied for patients in the AF health state of the model. In the base-case analysis, a CHADS2 risk score of 2 was assumed. In a sensitivity analysis, the model is run assuming different risk scores.
In an analysis of stroke from the AFFIRM study, the presence of AF was found to have a hazard ratio for stroke of 1.6 (95% CI 1.11 to 2.30).86 Therefore, patients in the NSR health state of the model were assumed to have a risk of stroke equivalent to that of patients with AF multiplied by a factor of the inverse of 1.6 (0.625).
Table 13: Annual Risk of Stroke by CHADS2 Score
CHADS2 Annual Probability of Stroke (95% CI) 0 0.019 (0.012 to 0.030) 1 0.028 (0.020 to 0.038) 2 0.040 (0.031 to 0.051) 3 0.059 (0.046 to 0.073) 4 0.085 (0.063 to 0.111) 5 0.125 (0.082 to 0.175) 6 0.182 (0.105 to 0.274) CI = confidence interval.
c) Probability of major bleed The annual risk of bleeding while on warfarin and aspirin therapy was based on data in Lip et al.’s89 systematic review and meta-analysis. First, the annual risk of major bleed while not on warfarin or aspirin therapy was estimated based on details that were reported for the placebo arms of the studies in the systematic review. These details included the number of patients and number of major bleeds in the placebo arm of each study with the mean duration of each study. Based on these data, the annual rate of a major bleed in the absence of warfarin or aspirin therapy was estimated to be 0.0058. The calculations appear in Appendix 18.
Lip et al.89 reported the RR of a major bleed when placebo was compared with warfarin to be 0.45 (95% CI 0.25 to 0.82). These RRs were applied to the probability of major bleed in the absence of aspirin and warfarin. The resulting annual probability of a major bleed with or without warfarin appears in Table 14.
Table 14: Annual Risk of Major Bleed by Treatment Therapy Annual Risk of Major Bleed Without warfarin 0.0058 With warfarin 0.0128
Data from the stroke prevention in Atrial Fibrillation II study90,91 were used to estimate the proportion of major bleeds that are ICHs. The annual rate of major hemorrhages and the annual rate of ICH for patients on aspirin and warfarin therapy were presented. Based on these data, the proportion of major bleeds that were ICHs was estimated to be 0.332.
54 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses d) Probability of AF recurrence The probability of AF recurrence after achieving NSR was derived from Pappone et al.’s62 long- term observational study of AF ablation and AAD treatment. Using Kaplan-Meier survival curves, Pappone et al. reported the probability of being free of AF at one, two, and three years for AAD and ablation-treated patients. Based on the one- and three-year data, the annual probability of AF recurrence among patients achieving NSR was estimated to be 0.036 and 0.221 for AF ablation- and AAD-treated patients respectively.
e) AF ablation procedural complications The probability of AF ablation procedural complications was taken from a systematic review of RCT and non-RCT studies evaluating catheter based AF ablation procedures.92 Table 15 shows the probability of AF ablation procedural complications.
Table 15: Probability of AF Ablation Procedural Complications Procedural Complication Probability of Complication Ischemic stroke 0.003 Transient ischemic attack 0.002 Cardiac tamponade 0.008 Pulmonary vein stenosis 0.016
AF = atrial fibrillation.
f) Probability of pulmonary toxicity The probability of pulmonary toxicity while on AAD treatment was based on Vorperian et al.’s93 data. Vorperian et al. performed a meta-analysis on adverse events for patients on low-dose amiodarone. The authors report 1.9 % of amiodarone patients in four studies had pulmonary toxicity. Based on the weighted mean follow-up in the studies (27.54 months), the annual probability of pulmonary toxicity is estimated to be 0.00832. The proportion of irreversible cases of pulmonary toxicity was assumed to be 0.25.94 Dusman et al.95 reported the probability of death after pulmonary toxicity to be 0.091. These values were assumed in the model.
g) Mortality General population age and gender-specific mortality rates based on Canadian life tables96,97 were applied for patients in the model in the absence of ischemic stroke and major bleeds. Several sources were used to derive the mortality rates of patients who have an ischemic stroke. Johansen et al.98 reported the outcomes for 34,448 patients who were hospitalized in Canada for a first stroke. Among the reported outcomes were 28-day mortality rates by age group and gender according to stroke type. The 28-day mortality rates of patients with strokes classified as cerebral infarction (ICD-9 434,436) were applied to patients having an ischemic stroke. To account for the increased risk of death after the remainder of the first year after stroke, data from another Canada-based publication were used. Tu et al.99 reported the mortality of patients who were hospitalized for acute stroke in Canada at 30 days (18.9%) and at one year (32.0%). Data were not presented by stroke type, age, or gender. The ratio of one-year mortality to 30-day mortality in this study is estimated to be 1.78 (0.32/0.189). In the model, this factor was applied to the 28 mortality rates that were reported by Johansen et al.98 to estimate one-year age- and gender-specific mortality rates of ischemic stroke. For post-stroke mortality after one year, the general population mortality was increased by a factor of 2.3, based on a long-term population-
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 55 Clinical and Cost-Effectiveness Analyses based stroke study.100 Table 16 shows the ischemic stroke mortality rates that were used in the model.
Table 16: Ischemic Stroke Mortality Age (years) 28 Days98 1 Year Male Female Male female 20 to 44 0.059 0.050 0.105 0.089 45 to 64 0.062 0.069 0.110 0.123 65 to 79 0.116 0.120 0.206 0.214 80+ 0.222 0.218 0.395 0.388 Ratio of 1-year mortality to 30-day mortality99 1.78 Relative risk of death compared with general population years 2+ post stroke 2.3
Mortality post ICH was estimated using 30-day age- and gender-specific mortality after ICH [ICD9-431], which was taken from Johansen et al.’s98 Canada-based study. To estimate one year mortality, the 30-day mortality rates were increased by a factor of 1.2. This adjustment factor was based on a study looking at long-term mortality post-ICH.101 Flaherty et al.101 reported mortality to be 0.48 and 0.59 at one month and one year respectively. Table 17 shows the mortality rates after ICH that were used in the model.
Table 17: Hemorrhagic Stroke Mortality Age (years) 28 Days98 1 Year Male Female Male Female 20 to 44 0.207 0.226 0.248 0.271 45 to 64 0.293 0.365 0.352 0.438 65 to 79 0.419 0.411 0.503 0.493 80+ 0.511 0.495 0.613 0.594 Ratio of 1-year mortality to 30-day mortality99 1.2 Relative risk of death compared with general population years 2+ post stroke100 2.3
h) Utilities Patients with NSR were assigned age- and gender-specific general population utility values102 (Appendix 19).
No published studies primarily reported utilities for AF. Reynolds et al.103 describe utility estimates that were derived as part of a cost-effectiveness analysis of radio frequency AF ablation. Reynolds et al. transformed SF-12 responses from patients who were enrolled in the FRACTAL registry to utility scores using the Brazier algorithm.67 The FRACTAL registry included more than 1,000 patients with a first-time diagnosis of AF. Reynolds et al.103 reported the average change in utility among patients with no documented recurrences of AF over 12 months to be 0.046. Based on these data, a disutility of 0.046 was applied to patients in the AF health state.
Reynolds et al. estimated utilities by transforming SF-36 responses from patients before and after AF ablation in two other patient populations. We thought that the data from the FRACTAL trial
56 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses best represented the change in utility moving from an AF to a NSR health state, because the other studies evaluated intervention-specific changes in utility values.
Utility weights were estimated for ischemic and hemorrhagic stroke. Data from two studies were used to derive stroke health state utility weights. Rivero-Arias et al.104 provided estimates of post-stroke utility scores according to a modified Rankin Score. In a Canadian cohort study, Goeree et al.105 reported the distribution of discharge modified ranking score according to type of stroke (ischemic, hemorrhagic, transient ischemic attack). The modified Rankin Score specific utility values that were reported by Rivero-Arias et al.104 were applied to the distribution of hospital discharge modified Rankin Score that was reported by Goeree et al. to derive a weighted average utility weight for ischemic and hemorrhagic stroke. Based on these data, the utility weight that was applied to patients post ischemic and hemorrhagic stroke was 0.46 and 0.28 respectively. Appendix 20 shows the calculations.
A disutility of 1.0 for seven days was applied to the AF ablation complications. For pulmonary toxicity, a disutility of 1.0 for the duration of a related hospitalization was used. The mean length of stay for a pulmonary toxicity-related hospitalization was estimated to be 13 days. This was based on hospitalizations that were identified from the Ontario Case Costing Initiative database106 with a primary diagnosis of J70.2 (acute drug-induced interstitial lung disorders), J70.3 (chronic drug-induced interstitial lung disorders), J70.4 (respiratory conditions due to drug- induced interstitial lung disorders, unspecified), or J84.1(idiopathic pulmonary fibrosis).
For irreversible pulmonary toxicity, a utility weight of 0.6 was applied in each cycle.107
6.3.9 Resource use and costs
a) Cost of radiofrequency ablation The cost of an inpatient stay for a radiofrequency ablation procedure was estimated using data from the Ontario Case Costing project.106 We used the average cost of hospitalizations with a primary procedure code indicating catheter ablation using a percutaneous transluminal approach (1.HH.59.GP-AW) and a most responsible diagnosis of AF (I480). Based on these criteria, the mean cost per AF ablation hospitalization was estimated to $7,056.
The physician fees for an AF ablation procedure was estimated using expert opinion on physician fee codes and fees listed in the Ontario schedule of benefit for physician services.108 The total physician fees for an AF ablation were estimated to be $2,534. The physician fees and codes that were used in the calculation appear in Appendix 21. The total cost per AF ablation in the model was $9,590.
The number of ablation procedures per patient was derived from Cappato et al.’s109 survey of electrophysiology (EP) laboratories. Cappato et al. reported that among 8,745 AF ablation patients, 2,389 (27%) required one or more ablation procedures. Therefore, in the model, it was assumed that each patient would require 1.27 ablation procedures. The costs appear in Table 18.
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 57 Clinical and Cost-Effectiveness Analyses Table 18: AF Ablation Costs Inpatient or day surgery cost $7,056 Physician fees $2,534 Total cost per procedure $9,590 Number of procedures per patient 1.27 Total cost per patient $12,179
AF = atrial fibrillation.
During the follow-up, it was assumed that patients had three cardiologist consultations and underwent a CT scan in the first year post ablation. The cumulative cost of follow up in the first year was $666. No follow up costs were applied after the first year post ablation.
b) Cost of ablation procedural complications The cost of ablation procedural complications were estimated from Khaykin et al.’s110 Canadian costing study. The unit cost of cardiac tamponade, PV stenosis, stroke, and transient ischemic attack was $5,842, $8,487, $14,1872, and $4.297 respectively. These costs included inflation to 2010 $C using the health care component of the consumer price index.
c) Cost of pulmonary toxicity The acute cost of pulmonary toxicity was estimated from the Ontario Case Costing Initiative. The mean cost per hospitalization with a primary diagnosis that is potentially related to pulmonary111,112 toxicity was found to be $20,436. These were hospitalizations with a primary diagnosis code of J70.2 (acute drug induced interstitial lung disorders), J70.3 (chronic drug- induced interstitial lung disorders), J70.4 (respiratory conditions due to drug induced interstitial lung disorders, unspecified), or J84.1 (idiopathic pulmonary fibrosis).
For irreversible pulmonary toxicity, an annual cost of $3,799 was applied.81 These costs include inflation to 2010 $C using the health care component of the consumer price index.111,112
d) Cost of AAD treatment The unit costs of AAD treatment were based on reimbursement prices from the Ontario Drug Benefit Formulary.113 The assumed daily dosage of amiodarone was 200 mg per day. An 8% pharmacy markup was applied to the drug costs, with a $7.00 dispensing fee. It was assumed that a 90-day supply of the drug would be dispensed each time. The total annual cost of amiodarone was estimated to be $433.29. Patients with NSR and those in the AF health state were assigned amiodarone costs in each cycle.
e) Cost of warfarin treatment The unit costs of warfarin treatment were based on reimbursement prices from the Ontario Drug Benefit Formulary.113 The assumed dosage of warfarin was 5 mg per day. An 8% pharmacy markup was applied to the drug costs with a $7.00 dispensing fee. It was assumed that a 90-day supply of the drug would be dispensed each time. The total annual warfarin costs that were used in the model were $75.30. In addition an annual monitoring cost of $387.54 was assumed.114 This cost included inflation to 2010 $C using the health care component of the consumer price index.111,112
58 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses f) Cost of stroke The cost of stroke was estimated from Goeree et al.’s105 Canadian costing study. In this study, the total one year health care costs of patients having an ischemic stroke was found to be $53,576. The one-year cost after a hemorrhagic stroke was estimated to be $56,573. These costs were applied to the first year post stroke in the current model. Goeree et al. presented stroke costs by resource category. Costs for the second year and beyond post stroke was assumed to be equal to the sum of the costs for long-term care, home care, prescription medications, outpatient visits, doctor visits, and assistive devices. These costs amounted to $6,265 and $4,841 for ischemic and hemorrhagic stroke respectively. After inflating to 2010 $C111,112 the first-year costs of ischemic and hemorrhagic stroke were $61,413 and $58,159 respectively. In subsequent years, the costs of ischemic and hemorrhagic stroke were $6,801 and $5,843 respectively.
g) Cost of major gastrointestinal bleed The cost of a major gastrointestinal bleed that was used in the model was $6,023. It was based on the average cost of a hospitalization with a case mix group for a gastrointestinal bleed (CMG 281).
6.3.10 Discount rate
In accordance with CADTH guidelines,115 a 5% discount was applied to costs and QALYs. The discount rate was varied in a sensitivity analysis.
6.3.11 Variability and uncertainty
The variability of patient groups was assessed using one-way sensitivity analyses. The model results according to age and gender were tested using model results by baseline CHADS2 stroke risk score. A sensitivity analysis on age and gender was conducted. In the first scenario, the same probability of stroke was assumed for all age groups. In the second scenario, the risk of stroke is varied according to age using Framingham data.116 Wolf et al.116 presented the 10-year risk of stroke according to age and gender. Using these data the, RR for stroke at different ages relative to 65-year-olds was applied to the risk of stroke based on a CHADS2 score of 2. Appendix 22 shows the RR of stroke by age and gender that was applied in this scenario.
Model structural uncertainty was assessed using one-way sensitivity analyses. Cost-effectiveness results were estimated varying discount rates, model time horizon, and the disutility of being in the AF health state. In addition, the model was run under the assumption that the restoration of NSR does not affect the stroke risk and under the assumption that amiodarone treatment is discontinued for patients not in NSR. Parameter uncertainty is evaluated using probabilistic sensitivity analysis and presented in a cost-effectiveness acceptability curve for AF ablation. One thousand Monte Carlo simulations were used in the PSA. Appendix 23 shows the distributions, parameters, and 95% confidence intervals associated with the parameters.
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 59 Clinical and Cost-Effectiveness Analyses 6.4 Primary Economic Evaluation: Results
6.4.1 Analysis and results
Table 19 shows that the ablation treatment group incurs an estimated $8,539 incremental expected costs compared with the AAD treatment group. The model estimated that the ablation group had more expected QALYs and fewer expected strokes compared with the AAD treatment group. It was estimated that the ablation treatment arm results in 0.033 fewer expected strokes and 0.144 more QALYs compared with the AAD treatment arm. Based on the expected costs and QALYs that were estimated in the model, the incremental cost per QALY of the ablation arm compared with the AAD arm is $59,194. Therefore using the base-case analysis, ablation would be considered to be cost-effective if society’s willingness to pay for a QALY is $59,194 or higher. Otherwise, AAD therapy is the cost-effective treatment.
Table 19: Base-Case Cost-Effectiveness Results Treatment Expected Cost Expected Stroke Expected QALY $/QALY Ablation $21,150 0.122 3.416 AAD $12,611 0.155 3.272 Incremental $8,539 (0.033) 0.144 $59,194 (ablation − AAD)
AAD = antiarrhythmic drugs; QALY = quality-adjusted life-year.
6.4.2 Results of variability analysis
Table 20 shows that the cost-effectiveness results varied according to age and gender. In scenario 1, the same risk of ischemic stroke was assumed for each starting age. The incremental cost per QALY is estimated to be $57,088 for a 55-year-old female and $65,147 for a 75-year-old female. The incremental cost per QALY was estimated to be $57,167 for a 55-year-old male and $65,129 for a 75-year-old male. In scenario 2, we adjusted baseline stroke risk by age using Wolf et al.’s116 data. In this scenario, the incremental cost per QALY is estimated to be $67,918 for a 55- year-old female and $49,363 for a 75-year-old female. The incremental cost per QALY was $65,672 for a 55-year-old male and $55,275 for a 75-year-old male.
The cost-effectiveness results varied according to the CHADS2 index score. When the CHADS2 index score for the model cohort is 0, the incremental cost per QALY of ablation compared with AAD is $68,822. When the CHADS2 index score was 4, the incremental cost per QALY was $44,652.
60 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses Table 20: Cost-Effectiveness Results by Age, Gender, and CHADS2 Index Score Age (years); Gender Incremental Cost Incremental QALY Incremental $/QALY Scenario 1: Sensitivity analysis by age and gender assuming same risk of stroke for all starting ages 55; male $8,330 0.146 $57,167 60; male $8,365 0.145 $57,846 65; male $8,539 0.144 $59,194 70; male $8,630 0.141 $61,120 75; male $8,787 0.135 $65,129 55; female $8,330 0.146 $57,088 60; female $8,365 0.145 $57,765 65; female $8,548 0.144 $59,219 70; female $8,637 0.141 $61,142 75; female $8,793 0.135 $65,147 Scenario 2: Sensitivity analysis by age and gender adjusting stroke risk according to starting age 55; male $9,028 0.137 $65,672 60; male $8,796 0.140 $63,034 65; male $8,539 0.144 $59,194 70; male $8,310 0.146 $57,112 75; male $8,005 0.145 $55,275 55; female $9,201 0.135 $67,918 60; female $8,870 0.139 $63,923 65; female $8,548 0.144 $59,219 70; female $7,999 0.150 $53,371 75; female $7,481 0.152 $49,363
CHADS2 risk score 0 $9,259 0.135 $68,822 1 $8,941 0.139 $64,412 2 $8,539 0.144 $59,194 3 $7,952 0.152 $52,214 4 $7,242 0.162 $44,652
QALY = quality-adjusted life-year.
6.4.3 Results of uncertainty analysis
Table 21 shows the time horizon that was chosen had a large impact on results. Using a 10-year time horizon, the incremental cost per QALY was $14,273 when AF ablation was compared with AAD. With a 20-year time horizon, ablation becomes less costly and more effective than AAD. Discount rates had little impact on results. If a zero per cent discount is used, the incremental cost per QALY of ablation became $49,308 per QALY. In an economic evaluation of ablation compared with AAD, Reynolds et al.82 assumed that the restoration of NSR had no impact on the risk of stroke. If this assumption is applied in the current model, the cost per QALY of AF ablation was $86,129. Even with the assumption that the restoration of NSR does not affect the
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 61 Clinical and Cost-Effectiveness Analyses stroke risk, the incremental QALYs from the AF ablation were reduced only from 0.144 in the base case to 0.116. This indicates that much of the difference in QALYs that is predicated by the model was due to the disutility that is applied to patients in the AF health state. If the disutility of being in AF compared with NSR is 0.08 instead of the base-case value of 0.043, the cost per QALY of AF ablation becomes $38,390. If the disutility of being in the AF health state is 0.02 or 0.00, the incremental cost per QALY becomes $101,083 and $221,839 respectively.
Table 21: Cost-Effectiveness Results by Model Structure Assumptions Incremental Incremental Incremental $/QALY Costs QALYs Time horizon (years) 3 $10,670 0.082 $130,711 5 $8,539 0.144 $59,194 10 $4,299 0.301 $14,273 20 ($71) 0.505 AF ablation dominates Discount 0% $7,995 0.162 $49,308 3% $8,335 0.151 $55,211 5% $8,539 0.144 $59,194 Amiodarone treatment stopped if not in NSR Restoring NSR has no impact on $10,019 0.116 $86,129 stroke Disutility of AF health state 0 $8,539 0.038 $221,839 0.01 $8,539 0.061 $138,883 0.02 $8,539 0.084 $101,083 0.03 $8,539 0.107 $79,457 0.04 $8,539 0.130 $65,454 0.05 $8,539 0.153 $55,647 0.06 $8,539 0.176 $48,396 0.07 $8,539 0.199 $42,816 0.08 $8,539 0.222 $38,390
AF = atrial fibrillation; NSR = normal sinus rhythm; QALY = quality-adjusted life-year.
Figure 4 shows that AF ablation, compared with AAD, has the highest probability of being cost- effective at a willingness to pay of $62,000 per QALY and higher. The probability of AF ablation being cost-effective at a willingness to pay for QALY thresholds of $25,000, $50,000, $100,000, and $150,000 was estimated to be 0.03, 0.30, 0.89, and 0.98 respectively.
62 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses Figure 4: Cost-Effectiveness Acceptability Curve
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
Probability AF ablation is cost-effective 0.2 Probability AF ablation us cost effective
0.1
0 $0 $50,000 $100,000 $150,000 $200,000 Willingness to pay for a QALY
AF = atrial fibrillation; QALY = quality-adjusted life-year.
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 63 Clinical and Cost-Effectiveness Analyses 7 HEALTH SERVICES IMPACT 7.1 Population Impact
To estimate the number of Canadians with AF, prevalence studies were identified in a targeted literature search. Prevalence rates were then applied to age- and gender-specific Canadian population estimates.96,97
No Canadian prevalence studies were identified. Therefore, the prevalence rates were based on Go et al.’s2 US study. Table 22 shows Go et al.’s estimates of the percentage of the Canadian population with AF. The prevalence rates increase with the age of males and females.
Table 22: Canadian Population and AF Prevalence Rates by Age and Gender Age Male Population Prevalence of AF Female Population Prevalence of AF (years) in Canada among Males (%) in Canada among Females (%) < 55 11,782,340 0.2 11,820,815 0.1 55 to 59 1,026,395 0.9 1,058,230 0.4 60 to 64 780,140 1.7 809,730 1 65 to 69 593,805 3 640,770 1.7 70 to 74 493,465 5 560,320 3.4 75 to 79 386,485 7.3 493,090 5 80 to 84 251,420 10.3 395,285 7.2 > 85 161,925 11.1 358,685 9.1
AF = atrial fibrillation.
Based on Canadian population and prevalence estimates, Table 23 shows that the number of Canadians with AF is estimated to be 300,486. The number of males with AF is estimated to be 160,635 (Table 23). The number of females with AF is estimated to be 139,851 (Table 23).
Table 23: Estimate of Number of People with AF in Canada by Age and Gender Age (years) Male Population Female Population Total with AF with AF < 55 23,565 11,821 35,386 55 to 59 9,238 4,233 13,471 60 to 64 13,262 8,097 21,359 65 to 69 17,814 10,893 28,707 70 to 74 24,673 19,051 43,724 75 to 79 28,213 24,655 52,868 80 to 84 25,896 28,461 54,357 > 85 17,974 32,640 50,616 Total 160,635 139,851 300,486
AF = atrial fibrillation.
64 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses 7.2 Budget Impact
The budget impact of minimally invasive AF ablation procedures was assessed by estimating the current expenditures for these procedures. The projections of future expenditures are based on trends in expenditures over the last five years.
7.2.1 Methods
Estimates of Canadian expenditures on minimally invasive AF ablation procedures were based on the number of minimally invasive procedures by province using the DAD and NACRS databases. Minimally invasive AF ablation procedure-related encounters were those including a procedure code for percutaneous catheter ablation (1.HH.59.GP-AW) and a diagnosis code of AF (I48.0).
To estimate total expenditures, the number of identified procedures was multiplied by the cost per AF ablation procedure. The derivation of the cost per AF ablation procedure ($9,590) is described in section 5.3.9. This cost includes the inpatient costs and physician fees.
7.2.2 Results
Table 24 shows the data that were taken from the DAD and NACRS databases. The number of annual procedures increased every year from 2004 to 2008. In 2008, there were 2,030 minimally invasive AF ablations in the reporting provinces. Most of the procedures occurred in British Columbia (851) and Ontario (910). Quebec is not included in the DAD and NACRS databases.
Table 24: Number of Minimally Invasive AF Procedures by Province and Year Province 2004 2005 2006 2007 2008 Alberta 55 63 52 97 119 British Columbia 449 544 518 548 851 Saskatchewan 0 0 0 1 11 Manitoba 3 1 0 7 14 Ontario 398 460 684 866 910 New Brunswick 0 0 0 13 20 Nova Scotia 45 41 49 78 98 Newfoundland 7 19 13 11 7 All 957 1,128 1,316 1,621 2,030
AF = atrial fibrillation. Source: CIHI-DAD: fiscal years ending March 31, 2005-2009; CIHI-NACRS: fiscal years ending March 31, 2005-2009; Canadian Institute for Health Information, Toronto.
Table 25 shows that in 2008 the total expenditures on minimally invasive AF ablation procedures are estimated to be $19,467,700.
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 65 Clinical and Cost-Effectiveness Analyses Table 25: Estimates of Expenditures on Minimally Invasive AF Ablation Procedures Year Province 2004 2005 2006 2007 2008 Alberta $527,450 $604,170 $498,680 $930,230 $1,141,210 British Columbia $4,305,910 $5,216,960 $4,967,620 $5,255,320 $8,161,090 Saskatchewan $0 $0 $0 $9,590 $105,490 Manitoba $28,770 $9,590 $0 $67,130 $134,260 Ontario $3,816,820 $4,411,400 $6,559,560 $8,304,940 $8,726,900 New Brunswick $0 $0 $0 $124,670 $191,800 Nova Scotia $431,550 $393,190 $469,910 $748,020 $939,820 Newfoundland $67,130 $182,210 $124,670 $105,490 $67,130 Total $9,177,630 $10,817,520 $12,620,440 $15,545,390 $19,467,700
AF = atrial fibrillation.
The total expenditures in the reporting provinces have increased from $9,117,630 in 2004 to $19,467,700 in 2008. This is equivalent to an annual increase in expenditures of 16.2%. Assuming that this trend continues, the expenditures on minimally invasive AF ablation procedures will increase to more than $40,000,000 by 2013 (Table 26).
Table 26: Projected Expenditures for the Next Five Years Year Projected Expenditures 2009 $22,582,532 2010 $26,195,737 2011 $30,387,055 2012 $35,248,984 2013 $40,888,821
7.3 Planning, Implementation, Utilization, and Legal or Regulatory Considerations
The articles that were identified during the literature searches did not discuss the issues in the implementation of ablation procedures for AF among Canadian adults. After discussions with clinical experts (JH, DR), it was decided that the directors of the EP training programs in Canada (or a physician who is designated by the program) be invited to answer general questions about AF ablation procedures (Appendix 24) in hospitals. The data from five physicians in Ontario, Quebec, and Nova Scotia are summarized here.
The minimum number of personnel who are needed during an ablation procedure would be one EP physician, two to three nurses cross-trained as EP technicians, or one nurse and one technician. For complex ablation procedures, a second or third EP physician is needed. The number of AF ablation procedures varied between 150 and 370 per year at the five institutions. Approximately 75% of the ablations are done as outpatient procedures. Complex procedures or those that occur at the end of the day require an overnight stay as a minimum. Usually two procedures per day are performed, but this varied between one and four procedures. The number
66 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses of procedures booked depends on the complexity of the procedure, the expertise of the EP physician, and the logistics (transportation, operating room availability, bed availability).
The waiting times for consultations and ablation procedures varied across institutions. For an urgent consultation, the waiting time was reported to be two to seven days; for a semi-urgent consultation, approximately two to three months. Overall, the consultation waiting times for non- urgent procedures varied between six weeks and four months. Urgent procedures could be done within seven days, but the wait time from consultation to ablation procedures generally varied between two and 21 months.
The CCS Access to Care Working Group and the Canadian Heart Rhythm Society117 published waiting time benchmarks for EP consultations and ablation procedures. The benchmark consultation waiting times for a patient with structural heart disease were 30 days and 90 days for patients with supraventricular tachycardia. The waiting times for catheter ablation procedures were two weeks for high-risk patients and three months for low-risk patients. The waiting times that are reported by the five institutions are within the benchmarks for consultations but not for ablation procedures. Two institutions reported waiting times of greater than six months. One institution reported that the maximum capacity for operating room allotment had been reached.
If all adults with AF were to undergo ablation procedures, then more EP training spots for residents and fellows could be funded by the federal government. Also, EP laboratories could be expanded and upgraded. The costs would vary depending on the extent of the upgrade and the existing infrastructure. Questions that could be asked before implementing an expansion include: “Are the floors able to handle the weight of extra equipment, including lead for shielding if expansion is close to patient areas? Is the ventilation system adequate? Is the electrical system adequate?” New equipment costs are then added to the infrastructure costs. If the existing structure cannot be expanded, then new EP laboratories would be built. New facilities have a greater ability for infrastructure cost sharing (ventilation, electrical) if more than one specialty needs new space.
8 DISCUSSION 8.1 Summary of Results
This HTA is aimed at evaluating the clinical, cost-effectiveness and health services impact of minimally invasive ablation procedures for the treatment of AF in adult patients and comparing these with other modalities for converting AF to NSR, including pharmacological or electrical cardioversion, and more invasive surgical procedures.
The findings of the systematic review of clinical evidence indicate that catheter ablation was superior to treatment with AADs, in patients with AF, for the maintenance of sinus rhythm up to a year. A meta-analysis of RCT data showed that catheter ablation had an approximately three- times-higher success rate than medical therapy. There was insufficient evidence (one RCT) comparing catheter ablation as a first-line treatment with medical therapy in patients for whom rhythm control was appropriate. The results of this RCT were in favour of ablation. Based on the
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 67 Clinical and Cost-Effectiveness Analyses results of the subgroup analyses, ablation techniques had better results in patients with paroxysmal AF (RR 3.80, 95% CI 2.92 to 4.96), compared with the pooled results for all AF types (RR 2.82, 95% CI 2.13 to 3.74).
Based on the results of our meta-analyses, at 12 months, patients with AF who underwent PVI+ had an 8% higher chance of maintaining NSR compared with those who underwent PVI (pooled RR 0.92, 95% CI 0.86 to 0.99). The overall estimate of the effect size is interpreted with caution, because of between-study variations in patient populations (AF types) and heterogeneity of the ablation techniques that were used. Subgroup analysis revealed that PVI plus left-sided linear ablations was superior to PVI (RR 0.85, 95% CI 0.76 to 0.95). Although the evidence was limited, the results of the review suggest that patients with non-paroxysmal AF may benefit more from adjunctive ablation of CFAEs. There was insufficient evidence to make a definitive conclusion about the effects of additional ablation lines in the right atrium, adjunctive ablation of ectopic triggers of AF, or other approaches such as stepwise and tailored ablation. The results of subgroup analyses indicate that patients with persistent AF could benefit more from PVI+ than from PVI (RR 0.59, 95% CI 0.39 to 0.91).
Most of the studies that were included in this review had scores of low methodological quality, mainly due to lack of blinding and failure to report the method of randomization. Although subgroup analysis showed that the quality of the studies did not have an impact on the results, the possibility of inflated effect sizes cannot be ruled out because of generally small sample sizes and an inadequate number of studies with appropriate allocation concealment.118-120
Most of the studies, regardless of the comparison groups, used a blanking period ranging from two weeks to three months. AF or AT recurrences were not counted as treatment failures during this period. With the hypothesis that the blanking period might have had an impact on the overall effect sizes, a meta-regression analysis including the length of the blanking period as a covariate was done. The statistically insignificant results suggested that effect size could be overestimated in the studies with longer blanking periods.
The adverse effects of ablation were poorly reported in the studies, mostly because of the limited duration of the trials and the fact that most studies were designed to evaluate the efficacy of ablation, not the adverse events. Most studies reported only on procedural complications such as access site hematomas, cardiac tamponade, or PV stenosis. The complication data were too scarce to statistically compare the procedural risks between ablation approaches. There were insufficient data to evaluate the impact of interventions on stroke, heart failure, and mortality. Because of limited data, it was impossible in some cases to separate the presence of stroke as an adverse effect of ablation from its incidence as a result of treatment failure (uncontrolled AF). Overall, the studies reported fewer adverse events in patients who underwent ablation than in medically treated patients. Procedural complications were comparable in PVI and PVI+. Ablation-related mortality was rare in RCTs comparing ablation with medical treatment of AF, and no ablation-related deaths were reported in the studies comparing PVI with PVI+. The results of a long-term non-RCT62 showed a statistically significant decrease (54%) in mortality rate three years after ablation.
68 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses The findings of the review suggested that ablation can improve the quality of life measures, particularly the physical health component, compared with medical therapy. A meta-analysis was not performed because of the inconsistency between studies in reporting the quality of life measures. There was insufficient evidence to compare the effects of PVI versus PVI+ on quality of life.
The review found no studies comparing catheter ablations with surgical approaches in the treatment of AF. The only non-RCT comparing the efficacy of catheter ablation with electrical cardioversion suggested that the patients undergoing ablation procedures have a relatively higher success rate. This difference was not statistically significant.
Our review identified 12 guidelines on ablation for the treatment of AF, one of which was exclusively on surgical ablations. In general, catheter ablation was recommended for patients with symptomatic AF refractory or intolerant to antiarrhythmic medical treatment, in whom restoration and maintenance of sinus rhythm is desired. Surgical ablation was advised in patients with AF undergoing cardiac surgeries. Some guidelines made their recommendations for selected patient populations. For example, CCS and the New Zealand Guidelines Group recommended catheter ablation only in patients with refractory paroxysmal AF.8,77 The International Society of Minimally Invasive Cardiothoracic Surgery restricted the indications for surgical ablation to persistent and permanent AF.78 One expert consensus document70 stated that AF ablation might be an appropriate first-line treatment in rare clinical situations. No details were given about the situations in which ablation could be considered as a first-line therapy.
The primary economic analysis found the incremental cost-effectiveness of AF ablation compared with AAD therapy to be $59,194 in patients with a CHADS2 risk score of 2 for whom at least one AAD had previously failed. Therefore, if society’s willingness to pay for a QALY is $59,194 or greater, AF ablation would be cost-effective in this population. Otherwise AAD would be the cost-effective strategy. The cost-effectiveness results were more sensitive to the utility values that are associated with achieving NSR than to the assumption of NSR having an impact on stroke risk. When no difference in utility is assumed between NSR and AF health states, the cost per QALY of AF ablation is $221,839. If it is assumed that restoring NSR has no impact on stroke, the cost per QALY of AF ablation compared with AAD is $86,129. Such findings may be inconsistent with the clinical motivation for AF treatment, which may emphasize stroke prevention more than the improvement of quality of life in the absence of stroke events. Because of the short-term nature of the clinical trials, a time horizon of five years was taken in the base-case analysis. If the model is based on a 10-year time horizon, the cost per QALY of AF ablation is $14,273. If a 20-year time horizon is taken, AF ablation becomes less costly and more effective compared with AAD.
Our findings are similar to those of other published economic evaluations of AF ablation compared with AAD therapy. In a base-case analysis, McKenna et al.83 found the incremental cost-utility ratio of AF ablation compared with AAD to be US$51,431 per QALY over a five- year time horizon. Based on a five-year time horizon analysis, Reynolds et al.82 found the incremental cost per QALY of AF ablation to be £20,831 in patients with a CHADS2 score of 3. Chan et al.81 estimated the incremental cost per QALY of AF ablation compared with AAD to be US$29,068 among low stroke risk patients and US$12,134 among low and moderate risk
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 69 Clinical and Cost-Effectiveness Analyses patients over a lifelong time horizon. This is based on the costs and QALYs reported by the authors.
In 2008, there were 2,030 minimally invasive AF ablation procedures performed in Canada, not including Quebec. Most of these ablations occurred in Ontario (910), British Colombia (851), Alberta (119), and Nova Scotia (98). The hospital or outpatient clinic costs of these procedures are estimated to be $12,231,634. Based on trends over the past five years, the projected expenditures on these procedures are estimated to reach $22,742,677 by 2013.
8.2 Strengths and Weaknesses of This Assessment
The clinical and economic reviews and the preliminary economic analyses followed CADTH guidelines. The review has several strengths. A comprehensive search was performed to identify randomized and non-randomized trials as well as economic evaluations. There was a supplemental review of the websites of professional associations, conference proceedings, and bibliographies. For an unbiased selection of studies, all judgments about the eligibility and methodological quality of studies were made in duplicate and independently. Appropriate methods were used for statistical pooling and for investigation of the sources of heterogeneity among studies.
The following factors that could result in between-study variability in effect size estimation and have an impact on the overall estimate of the effect would be taken into account in interpreting the results of our meta-analysis. Most of the studies enrolled patients with paroxysmal AF; some studies included patients with other types of AF. Ablation sites, energy sources, catheter types, and mapping techniques varied across the studies. The definition of treatment success included freedom from symptomatic AF, symptomatic and asymptomatic AF, or maintenance of NSR, each with or without using AADs. The duration of AF or atrial arrhythmia that was counted as AF recurrence varied among studies, ranging from 30 seconds to 10 minutes. In addition, different studies used various methods and protocols for surveillance and outcome measurement. Success rates at a follow-up of 12 months, or the longest reported follow-up for studies with shorter duration, were included in the meta-analyses. Variable follow-up periods may also have had an impact on the results.
The median quality score (Jadad score) of the studies in our meta-analysis was two out of five. This led to a hypothesis that the effect sizes could be overestimated in low-quality studies. In subgroup analyses, comparable results were found in the high- and low-quality studies. One reason for a lack of a clear association between effect size and study quality is that most of the studies had the same methodological flaws, although they received different quality scores. For example, because of the nature of the interventions, especially in the comparison of ablation versus medication, blinding was often impossible, and a lack of blinding was the most common reason for a Jadad quality score of less than three. Two RCTs with higher quality scores blinded patients and assessors who reported the incidence of AF recurrence. Appropriate allocation concealment, another methodological factor, could be influencing effect sizes. This feature is not included in the Jadad scale that was used in this review. The interpretation of the results based on study quality would have been more precise if other features that protect against bias, such as adequacy of concealment, statistical power to find significant treatment effect, intention-to-treat
70 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses analysis, and consistency in outcome measurement, had been taken into account in the assessment of study quality.
The review found limited data on the risk of stroke and mortality associated with AF ablation. This could be partly due to a limitation of the review to randomized or non-randomized controlled trials, which generally are of short duration and have small sample sizes. This may contribute to an underestimation of rare adverse events and an overestimation of short-term benefit. There was also an inability to pool the quality-of-life data, because of inconsistency in reporting such data between the studies.
The primary economic evaluation had several limitations. The indirect costs were not considered in the analysis. The clinical data inputs on NSR for AF ablation and AAD treatment were based on short-term data. The model assumes a lower risk of ischemic stroke risk associated with AF ablation compared with AAD because of the greater proportion of patients in NSR. However, there is no RCT evidence that AF ablation reduces the risk of stroke compared with AAD. The disutility of not being in NSR had a large impact on the results. However, there were limited published data on this variable. In addition, other AF treatment alternatives such as rate control medications were not considered in the analysis.
Despite its limitations, this systematic review provides a comprehensive summary of the evidence on catheter ablation for rhythm control in patients with AF. The report includes a primary economic analysis taken from a Canadian perspective. A fully probabilistic economic analysis based on RCT evidence was conducted to evaluate the cost-effectiveness of ablation.
8.3 Generalizability of Findings
The results of this systematic review are limited by the evidence. Recent RCTs met the inclusion criteria of this review, because of an evolution in techniques, and the follow-up in most of the included trials lasted 12 months or less. A proportion of the patients who were enrolled in the studies had paroxysmal AF, which may not reflect a real-world situation. Nearly all the trials were done in patients for whom at least one AAD had failed and who had no history of undergoing ablation. Most of the studies excluded elderly patients or those with concomitant conditions, such as heart failure. Therefore, the generalizability of results is uncertain.
Most of the trials occurred in high-volume centres. Specialized care performed by experienced staff and the large volume of procedures are determining factors that may result in higher success and lower adverse event rates of ablation. In Canada, this does not seem to be a limit to the generalizability of findings to patients who are treated in existing EP laboratories, which are generally located at large referral centers.
8.4 Knowledge Gaps
Our review failed to evaluate the long-term clinical and economic consequences of AF ablation. The clinical review found a paucity of trials on the efficacy of catheter ablation as a first-line therapy. The studies did not address the study questions on the effectiveness of AF ablation in patients with CHF, comparative effectiveness of catheter ablation and surgical procedures, and
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 71 Clinical and Cost-Effectiveness Analyses the effectiveness of repeated ablations. In addition, there were insufficient data on adverse events.
The following clinical areas can be explored in the future: Long-term benefits of ablation Adverse events and long-term safety of ablation Efficacy of adjunctive ablation Need for anticoagulation to prevent thromboembolic events in patients undergoing ablation Effectiveness of ablation versus medication in patients with AF and heart failure and in elderly patients.
9 CONCLUSIONS
The evidence in this systematic review indicates that catheter ablation increases the rate of maintenance of sinus rhythm compared with treatment with AADs in patients for whom one or two drugs had failed. The studies are of insufficient size and duration to evaluate the impact on stroke, heart failure, and mortality. Ablation has better results in patients with paroxysmal AF. Limited data suggest that catheter ablation may be an effective first-line rhythm control strategy in patients with AF. More trials are needed to confirm these findings. Our review suggests that patients with persistent AF may benefit more from PVI+ than from PVI.
The primary economic evaluation using a five-year time horizon found the incremental cost per QALY of AF ablation compared with AAD to be $59,194. These findings were similar to those of other published economic evaluations. The cost-effectiveness of AF ablation was more favourable when longer time horizons were used.
72 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses 10 REFERENCES
1. Krahn AD, Manfreda J, Tate RB, Mathewson FA, Cuddy TE. The natural history of atrial fibrillation: incidence, risk factors, and prognosis in the Manitoba Follow-Up Study. Am J Med 1995;98(5):476-84. 2. Go AS, Hylek EM, Phillips KA, Chang Y, Henault LE, Selby JV, et al. Prevalence of diagnosed atrial fibrillation in adults: national implications for rhythm management and stroke prevention: the AnTicoagulation and Risk Factors in Atrial Fibrillation (ATRIA) Study. JAMA 2001;285(18):2370-5. 3. Marchlinski F. The tachyarrhythmias. In: Fauci A, Braunwald E, Kasper D, Hauser S, Longo D, Jameson J, editors. Harrison's Principles of Internal Medicine. 17th ed. New York: McGraw-Hill Medical; 2008. 4. Lip GY, Boos CJ. Antithrombotic treatment in atrial fibrillation. Heart 2006;92(2):155-61. 5. Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: the Framingham Study. Stroke 1991;22(8):983-8. 6. Fuster V, Ryden LE, Cannom DS, Crijns HJ, Curtis AB, Ellenbogen KA, et al. ACC/AHA/ESC 2006 guidelines for the management of patients with atrial fibrillation: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the European Society of Cardiology Committee for Practice Guidelines (Writing Committee to Revise the 2001 Guidelines for the Management of Patients With Atrial Fibrillation): developed in collaboration with the European Heart Rhythm Association and the Heart Rhythm Society. Circulation 2006;114(7):e257-354. 7. Heart and Stroke Foundation of Canada. Atrial fibrillation. Ottawa: The Foundation; 2010. Available: http://www.heartandstroke.com/site/c.ikIQLcMWJtE/b.5052135/k.2C86/Heart_disease__Atrial_fibrillat ion.htm (accessed 2010 May 11). 8. 2004 Canadian Cardiovascular Society Consensus Conference: atrial fibrillation. Can J Cardiol 2005;21(Suppl B):9-73B. 9. Vaziri SM, Larson MG, Benjamin EJ, Levy D. Echocardiographic predictors of nonrheumatic atrial fibrillation. The Framingham Heart Study. Circulation 1994;89(2):724-30. 10. Wang TJ, Parise H, Levy D, D'Agostino RB, Wolf PA, Vasan RS, et al. Obesity and the risk of new- onset atrial fibrillation. JAMA 2004;292(20):2471-7. 11. Benjamin EJ, Chen PS, Bild DE, Mascette AM, Albert CM, Alonso A, et al. Prevention of atrial fibrillation: report from a national heart, lung, and blood institute workshop. Circulation 2009;119(4):606-18. 12. Ehrlich JR, Nattel S, Hohnloser SH. Atrial fibrillation and congestive heart failure: specific considerations at the intersection of two common and important cardiac disease sets. J Cardiovasc Electrophysiol 2002;13(4):399-405. 13. Humphries KH, Jackevicius C, Gong Y, Svensen L, Cox J, Tu JV, et al. Population rates of hospitalization for atrial fibrillation/flutter in Canada. Can J Cardiol 2004;20(9):869-76. 14. Fuster V, Ryden LE, Cannom DS, Crijns HJ, Curtis AB, Ellenbogen KA, et al. ACC/AHA/ESC 2006 guidelines for the management of patients with atrial fibrillation. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the European Society of Cardiology Committee for Practice Guidelines (Writing Committee to Revise the 2001 Guidelines for the Management of Patients With Atrial Fibrillation). J Am Coll Cardiol 2006;48(4):e149-246. 15. Vaughan Williams EM. A classification of antiarrhythmic actions reassessed after a decade of new drugs. J Clin Pharmacol 1984;24(4):129-47. 16. Humphries KH, Kerr CR, Steinbuch M, Dorian P. Limitations to antiarrhythmic drug use in patients with atrial fibrillation. CMAJ 2004;171(7):741-5.
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 73 Clinical and Cost-Effectiveness Analyses 17. Calkins H, Brugada J, Packer DL, Cappato R, Chen S-A, Crijns HJG, et al. HRS/EHRA/ECAS expert consensus statement on catheter and surgical ablation of atrial fibrillation: recommendations for personnel, policy, procedures and follow-up. A report of the Heart Rhythm Society (HRS) Task Force on Catheter and Surgical Ablation of Atrial Fibrillation. Developed in partnership with the European Heart Rhythm Association (EHRA). Heart Rhythm 2007;4(6):816-61. 18. National Collaborating Centre for Chronic Conditions. Atrial fibrillation: national clinical guideline for management in primary and secondary care. London: Royal College of Physicians; 2006. Available: http://www.nice.org.uk/nicemedia/live/10982/30055/30055.pdf (accessed 2010 Apr 22). 19. Cox JL, Sundt TM. The surgical management of atrial fibrillation. Annu Rev Med 1997;48:511-23. 20. Cox JL, Jaquiss RD, Schuessler RB, Boineau JP. Modification of the maze procedure for atrial flutter and atrial fibrillation. II. Surgical technique of the maze III procedure. J Thorac Cardiovasc Surg 1995;110(2):485-95. 21. Dewire J, Calkins H. State-of-the-art and emerging technologies for atrial fibrillation ablation. Nat Rev Cardiol 2010;7(3):129-38. 22. Cheng J, Arnsdorf M.F. Radiofrequency catheter ablation to precent recurrent atrial fibrillation. In: UpToDate [database online]. Waltham (MA): UpToDate; 2009. 23. Australian Safety and Efficacy Register of New Interventional Procedures - Surgical (ASERNIP-S). Horizon scanning technology prioritising summary: minimally invasive surgical treatment of atrial fibrillation. East Melbourne (Australia): Royal Australasian College of Surgeons; 2008. Available: http://www.health.gov.au/internet/horizon/publishing.nsf/Content/E76BDEECDE7BD1A8CA2575AD0 080F341/$File/PRIORITISING%20SUMMARY- Minimally%20invasive%20treatment%20atrial%20fibrillation.pdf (accessed 2010 Apr 15). 24. Medical Advisory Secretariat, Ministry of Health and Long-Term Care (Ontario). Ablation for atrial fibrillation: an evidence-based analysis [Ontario health technology assessment series vol 6, no 7]. Toronto: The Ministry; 2006. Available: http://www.health.gov.on.ca/english/providers/program/mas/tech/reviews/sum_af_030106.html (accessed 2010 Apr 15). 25. Keane D, Reddy V, Ruskin J. Emerging concepts on catheter ablation of atrial fibrillation from the tenth annual Boston Atrial Fibrillation Symposium. J Cardiovasc Electrophysiol 2005;16(9):1025-8. 26. Haissaguerre M, Sanders P, Hocini M, Takahashi Y, Rotter M, Sacher F, et al. Catheter ablation of long- lasting persistent atrial fibrillation: critical structures for termination. J Cardiovasc Electrophysiol 2005;16(11):1125-37. 27. Liu X, Long D, Dong J, Hu F, Yu R, Tang R, et al. Is circumferential pulmonary vein isolation preferable to stepwise segmental pulmonary vein isolation for patients with paroxysmal atrial fibrillation? Circ J 2006;70(11):1392-7. 28. Gray J, editor. Therapeutic choices. 4th ed. Ottawa: Canadian Pharmacists Association; 2003. 29. Jadad AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJ, Gavaghan DJ, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials 1996;17(1):1-12. 30. AGREE Collaboration. Appraisal of Guidelines for Research & Evaluation (AGREE) instrument. London: The Collaboration; 2001. Available: http://www.agreecollaboration.org (accessed 2010 Apr 15). 31. Review Manager (RevMan). [computer program]. Version 5.0. Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration; 2008. 32. Higgins JPT, Green S, editors. Principles of meta-analysis. In: Cochrane handbook for systematic reviews of interventions 5.0.2 [updated September 2009]. Cambridge (UK): The Cochrane Collaboration; 2009. Available: http://www.cochrane-handbook.org (accessed 2010 Apr 2). 33. StataCorp. Stata statistical software: release 11. College Station (TX): StataCorp LP; 2009.
74 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses 34. Egger M, Davey SG, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ 1997;315(7109):629-34. 35. Harbord RM, Egger M, Sterne JA. A modified test for small-study effects in meta-analyses of controlled trials with binary endpoints. Stat Med 2006;25(20):3443-57. 36. Fassini G, Riva S, Chiodelli R, Trevisi N, Berti M, Carbucicchio C, et al. Left mitral isthmus ablation associated with PV isolation: long-term results of a prospective randomized study. J Cardiovasc Electrophysiol 2005;16(11):1150-6. 37. Pappone C, Manguso F, Vicedomini G, Gugliotta F, Santinelli O, Ferro A, et al. Prevention of iatrogenic atrial tachycardia after ablation of atrial fibrillation: a prospective randomized study comparing circumferential pulmonary vein ablation with a modified approach. Circulation 2004;110(19):3036-42. 38. Sheikh I, Krum D, Cooley R, Dhala A, Blanck Z, Bhatia A, et al. Pulmonary vein isolation and linear lesions in atrial fibrillation ablation. J Interv Card Electrophysiol 2006;17(2):103-9. 39. Gaita F, Caponi D, Scaglione M, Montefusco A, Corleto A, Di MF, et al. Long-term clinical results of 2 different ablation strategies in patients with paroxysmal and persistent atrial fibrillation. Circ Arrhythm Electrophysiol 2008;1(4):269-75. 40. Tamborero D, Mont L, Berruezo A, Matiello M, Benito B, Sitges M, et al. Left atrial posterior wall isolation does not improve the outcome of circumferential pulmonary vein ablation for atrial fibrillation: a prospective randomized study. Circ Arrhythm Electrophysiol 2009;2(1):35-40. 41. Willems S, Klemm H, Rostock T, Brandstrup B, Ventura R, Steven D, et al. Substrate modification combined with pulmonary vein isolation improves outcome of catheter ablation in patients with persistent atrial fibrillation: a prospective randomized comparison. Eur Heart J 2006;27(23):2871-8. 42. Hocini M, Jais P, Sanders P, Takahashi Y, Rotter M, Rostock T, et al. Techniques, evaluation, and consequences of linear block at the left atrial roof in paroxysmal atrial fibrillation: a prospective randomized study. Circulation 2005;112(24):3688-96. 43. Haïssaguerre M, Sanders P, Hocini M, Hsu LF, Shah DC, Scavée C, et al. Changes in atrial fibrillation cycle length and inducibility during catheter ablation and their relation to outcome. Circulation 2004;109(24):3007-13. 44. Wazni O, Marrouche NF, Martin DO, Gillinov AM, Saliba W, Saad E, et al. Randomized study comparing combined pulmonary vein-left atrial junction disconnection and cavotricuspid isthmus ablation versus pulmonary vein-left atrial junction junction disconnection alone in patients presenting with typical atrial flutter and atrial fibrillation. Circulation 2003;108(20):2479-83. 45. Pontoppidan J, Nielsen JC, Poulsen SH, Jensen HK, Walfridsson H, Pedersen AK, et al. Prophylactic cavotricuspid isthmus block during atrial fibrillation ablation in patients without atrial flutter: a randomised controlled trial. Heart 2009;95(12):994-9. 46. Wang X-H, Liu X, Sun Y-M, Shi H-F, Zhou L, Gu J-N. Pulmonary vein isolation combined with superior vena cava isolation for atrial fibrillation ablation: a prospective randomized study. Europace 2008;10(5):600-5. 47. Corrado A, Bonso A, Madalosso M, Rossillo A, Themistoclakis S, Di BL, et al. Impact of systematic isolation of superior vena cava in addition to pulmonary vein antrum isolation on the outcome of paroxysmal, persistent, and permanent atrial fibrillation ablation: results from a randomized study. J Cardiovasc Electrophysiol 2010;21(1):1-5. 48. Elayi CS, Verma A, Di BL, Ching CK, Patel D, Barrett C, et al. Ablation for longstanding permanent atrial fibrillation: results from a randomized study comparing three different strategies. Heart Rhythm 2008;5(12):1658-64. 49. Di Biase L, Elayi CS, Fahmy TS, Martin DO, Ching CK, Barrett C, et al. Atrial fibrillation ablation strategies for paroxysmal patients: randomized comparison between different techniques. Circ Arrhythm Electrophysiol 2009;2(2):113-9.
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 75 Clinical and Cost-Effectiveness Analyses 50. Deisenhofer I, Estner H, Reents T, Fichtner S, Bauer A, Wu J, et al. Does electrogram guided substrate ablation add to the success of pulmonary vein isolation in patients with paroxysmal atrial fibrillation? A prospective, randomized study. J Cardiovasc Electrophysiol 2009;20(5):514-21. 51. Khaykin Y, Skanes A, Champagne J, Themistoclakis S, Gula L, Rossillo A, et al. A randomized controlled trial of the efficacy and safety of electroanatomic circumferential pulmonary vein ablation supplemented by ablation of complex fractionated atrial electrograms versus potential-guided pulmonary vein antrum isolation guided by intracardiac ultrasound. Circ Arrhythm Electrophysiol 2009;2(5):481-7. 52. Verma A, Patel D, Famy T, Martin DO, Burkhardt JD, Elayi SC, et al. Efficacy of adjuvant anterior left atrial ablation during intracardiac echocardiography-guided pulmonary vein antrum isolation for atrial fibrillation. J Cardiovasc Electrophysiol 2007;18(2):151-6. 53. Matsuo S, Yamane T, Date T, Inada K, Kanzaki Y, Tokuda M, et al. Reduction of AF recurrence after pulmonary vein isolation by eliminating ATP-induced transient venous re-conduction. J Cardiovasc Electrophysiol 2007;18(7):704-8. 54. Lin Y-J, Tai C-T, Chang S-L, Lo L-W, Tuan T-C, Wongcharoen W, et al. Efficacy of additional ablation of complex fractionated atrial electrograms for catheter ablation of nonparoxysmal atrial fibrillation. J Cardiovasc Electrophysiol 2009;20(6):607-15. 55. Wazni OM, Marrouche NF, Martin DO, Verma A, Bhargava M, Saliba W, et al. Radiofrequency ablation vs antiarrhythmic drugs as first-line treatment of symptomatic atrial fibrillation: a randomized trial. JAMA 2005;293(21):2634-40. 56. Wilber DJ, Pappone C, Neuzil P, De Paola A, Marchlinski F, Natale A, et al. Comparison of antiarrhythmic drug therapy and radiofrequency catheter ablation in patients with paroxysmal atrial fibrillation: a randomized controlled trial. JAMA 2010;303(4):333-40. 57. Forleo GB, Mantica M, De LL, Leo R, Santini L, Panigada S, et al. Catheter ablation of atrial fibrillation in patients with diabetes mellitus type 2: results from a randomized study comparing pulmonary vein isolation versus antiarrhythmic drug therapy. J Cardiovasc Electrophysiol 2009;20(1):22-8. 58. Jaïs P, Cauchemez B, Macle L, Daoud E, Khairy P, Subbiah R, et al. Catheter ablation versus antiarrhythmic drugs for atrial fibrillation: the A4 study. Circulation 2008;118(24):2498-505. 59. Pappone C, Augello G, Sala S, Gugliotta F, Vicedomini G, Gulletta S, et al. A randomized trial of circumferential pulmonary vein ablation versus antiarrhythmic drug therapy in paroxysmal atrial fibrillation. The APAF study. J Am Coll Cardiol 2006;48(11):2340-7. 60. Santinelli V, Radinovic A, Vicedomini G, Ciconte G, Augello G, Sala S, et al. Early rhythm control strategy by catheter Ablation in patients with Paroxysmal Atrial Fibrillation: the 4-year results of a randomized follow-up study comparing catheter ablation and antiarrhythmic drug therapy: the APAF 2 study. Circulation 2009;120(18 Suppl 2):S657. 61. Krittayaphong R, Raungrattanaamporn O, Bhuripanyo K, Sriratanasathavorn C, Pooranawattanakul S, Punlee K, et al. A randomized clinical trial of the efficacy of radiofrequency catheter ablation and amiodarone in the treatment of symptomatic atrial fibrillation. J Med Assoc Thai 2003;86(Suppl 1):S8- 16. 62. Pappone C, Rosanio S, Augello G, Gallus G, Vicedomini G, Mazzone P, et al. Mortality, morbidity, and quality of life after circumferential pulmonary vein ablation for atrial fibrillation: outcomes from a controlled nonrandomized long-term study. J Am Coll Cardiol 2003;42(2):185-97. 63. Lan X, Su L, Ling Z, Liu Z, Wu J, Yang X, et al. Catheter ablation vs. amiodarone plus losartan for prevention of atrial fibrillation recurrence in patients with paroxysmal atrial fibrillation. Eur J Clin Invest 2009;39(8):657-63. 64. Rossillo A, Bonso A, Themistoclakis S, Riccio G, Madalosso M, Corrado A, et al. Role of anticoagulation therapy after pulmonary vein antrum isolation for atrial fibrillation treatment. J Cardiovasc Med 2008;9(1):51-5.
76 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses 65. Macaskill P, Walter SD, Irwig L. A comparison of methods to detect publication bias in meta-analysis. Stat Med 2001;20(4):641-54. 66. Samsa G, Edelman D, Rothman ML, Williams GR, Lipscomb J, Matchar D. Determining clinically important differences in health status measures: a general approach with illustration to the Health Utilities Index Mark II. Pharmacoeconomics 1999;15(2):141-55. 67. Brazier JE, Roberts J. The estimation of a preference-based measure of health from the SF-12. Med Care 2004;42(9):851-9. 68. ACC/AHA/ESC 2006 guidelines for the management of patients with atrial fibrillation - executive summary. Circulation 2006;114(7):700-52. 69. Fuster V, Ryden LE, Cannom DS, Crijns HJ, Curtis AB, Ellenbogen KA, et al. ACC/AHA/ESC 2006 guidelines for the management of patients with atrial fibrillation: full text: a report of the American College of Cardiology/American Heart Association Task Force on practice guidelines and the European Society of Cardiology Committee for Practice Guidelines (Writing Committee to Revise the 2001 guidelines for the management of patients with atrial fibrillation) developed in collaboration with the European Heart Rhythm Association and the Heart Rhythm Society. Europace 2006;8(9):651-745. 70. Calkins H, Brugada J, Packer DL, Cappato R, Chen S-A, Crijns HJG, et al. HRS/EHRA/ECAS expert consensus statement on catheter and surgical ablation of atrial fibrillation: Recommendations for personnel, policy, procedures and follow-up. A report of the Heart Rhythm Society (HRS) Task Force on catheter and surgical ablation of atrial fibrillation. Europace 2007;9(6):335-79. 71. National Institute for Health and Clinical Excellence. Percutaneous (non-thoracoscopic) epicardial catheter radiofrequency ablation for atrial fibrillation [Interventional procedure guidance 294]. London: The Institute; 2009. Available: http://guidance.nice.org.uk/IPG294 (accessed 2010 Apr 22). 72. National Institute for Health and Clinical Excellence. Percutaneous radiofrequency ablation for atrial fibrillation [Interventional procedure guidance 168]. London: The Institute; 2006. Available: http://guidance.nice.org.uk/IPG168 (accessed 2010 Apr 22). 73. National Institute for Health and Clinical Excellence. Thoracoscopic epicardial radiofrequency ablation for atrial fibrillation [Interventional procedure guidance 286]. London: The Institute; 2009. Available: http://guidance.nice.org.uk/IPG286 (accessed 2010 Apr 22). 74. National Institute for Health and Clinical Excellence. Radiofrequency ablation for atrial fibrillation in association with other cardiac surgery [Interventional procedure guidance 121]. London: The Institute; 2005. Available: http://guidance.nice.org.uk/IPG121 (accessed 2010 Apr 22). 75. National Institute for Health and Clinical Excellence. Microwave ablation for atrial fibrillation in association with other cardiac surgery [Interventional procedure guidance 122]. London: The Institute; 2005. Available: http://guidance.nice.org.uk/IPG122 (accessed 2010 Apr 22). 76. National Institute for Health and Clinical Excellence. Cryoablation for atrial fibrillation in association with other cardiac surgery [Interventional procedure guidance 123]. London: The Institute; 2005. Available: http://guidance.nice.org.uk/IPG123 (accessed 2010 Apr 22). 77. New Zealand Guidelines Group. The management of people with atrial fibrillation and flutter. Wellington (New Zealand): The Group; 2005. Available: http://www.nzgg.org.nz/guidelines/0085/AF_Full_Guide_(final).pdf (accessed 2010 Apr 22). 78. Ad N, Cheng DCH, Martin J, Berglin EE, Chang B-C, Doukas G, et al. Surgical ablation for atrial fibrillation in cardiac surgery: a Consensus Statement of the International Society of Minimally Invasive Cardiothoracic Surgery (ISMICS) 2009. Innovations 2010;5(2):74-83. Available: http://journals.lww.com/innovjournal/Fulltext/2010/03000/Surgical_Ablation_for_Atrial_Fibrillation_in .3.aspx (accessed 2010 Apr 28). 79. Guerra PG, Skanes AC. Catheter ablation therapy for atrial fibrillation. Can J Cardiol 2005;21(Suppl B):31-34B.
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 77 Clinical and Cost-Effectiveness Analyses 80. Forming guideline recommendations. In: A guideline developer's handbook [SIGN 50]. Edinburgh: Scottish Intercollegiate Guidelines Network; 2008. Available: http://www.sign.ac.uk/pdf/sign50.pdf (accessed 2010 Apr 27). 81. Chan PS, Vijan S, Morady F, Oral H. Cost-effectiveness of radiofrequency catheter ablation for atrial fibrillation. J Am Coll Cardiol 2006;47(12):2513-20. 82. Reynolds MR, Ellis E, Danilov T, Zimetbaum P, Josephson ME, Cohen DJ. Cost effectiveness of radiofrequency catheter ablation vs. anti-arrhythmic drugs for paroxysmal atrial fibrillation. Circulation 2008;118(18 Suppl 2):S1164-5. 83. McKenna C, Palmer S, Rodgers M, Chambers D, Hawkins N, Golder S, et al. Cost-effectiveness of radiofrequency catheter ablation for the treatment of atrial fibrillation in the United Kingdom. Heart 2009;95(7):542-9. 84. Eckard N, Davidson T, Walfridsson H, Levin L-A. Cost-effectiveness of catheter ablation treatment for patients with symptomatic atrial fibrillation. J Atr Fibrillation 2009;1(8):461-70. 85. Gage BF, Waterman AD, Shannon W, Boechler M, Rich MW, Radford MJ. Validation of clinical classification schemes for predicting stroke: results from the National Registry of Atrial Fibrillation. JAMA 2001;285(22):2864-70. 86. Sherman DG, Kim SG, Boop BS, Corley SD, Dimarco JP, Hart RG, et al. Occurrence and characteristics of stroke events in the Atrial Fibrillation Follow-up Investigation of Sinus Rhythm Management (AFFIRM) study. Arch Intern Med 2005;165(10):1185-91. 87. Reynolds MR, Shah J, Essebag V, Olshansky B, Friedman PA, Hadjis T, et al. Patterns and predictors of warfarin use in patients with new-onset atrial fibrillation from the FRACTAL Registry. Am J Cardiol 2006;97(4):538-43. 88. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials 1986;7(3):177-88. 89. Lip GY, Edwards SJ. Stroke prevention with aspirin, warfarin and ximelagatran in patients with non- valvular atrial fibrillation: a systematic review and meta-analysis. Thromb Res 2006;118(3):321-33. 90. Bleeding during antithrombotic therapy in patients with atrial fibrillation. The Stroke Prevention in Atrial Fibrillation investigators. Arch Intern Med 1996;156(4):409-16. 91. Warfarin versus aspirin for prevention of thromboembolism in atrial fibrillation: Stroke Prevention in Atrial Fibrillation II study. Lancet 1994;343(8899):687-91. 92. Calkins H, Reynolds MR, Spector P, Sondhi M, Xu Y, Martin A, et al. Treatment of atrial fibrillation with antiarrhythmic drugs or radiofrequency ablation: two systematic literature reviews and meta- analyses. Circ Arrhythm Electrophysiol 2009;2(4):349-61. 93. Vorperian VR, Havighurst TC, Miller S, January CT. Adverse effects of low dose amiodarone: a meta- analysis. J Am Coll Cardiol 1997;30(3):791-8. 94. Owens DK, Sanders GD, Harris RA, McDonald KM, Heidenreich PA, Dembitzer AD, et al. Cost- effectiveness of implantable cardioverter defibrillators relative to amiodarone for prevention of sudden cardiac death. Ann Intern Med 1997;126(1):1-12. 95. Dusman RE, Stanton MS, Miles WM, Klein LS, Zipes DP, Fineberg NS, et al. Clinical features of amiodarone-induced pulmonary toxicity. Circulation 1990;82(1):51-9. 96. Statistics Canada. Life Tables, Canada, Provinces and Territories, 2000-2002, Table 2b Complete life table, Canada, 2000 to 2002: females. Ottawa: Statistics Canada; 2006. Available: http://www.statcan.gc.ca/pub/84-537-x/t/pdf/4198611-eng.pdf (accessed 2009 Oct 1). 97. Statistics Canada. Life Tables, Canada, Provinces and Territories, 2000-2002, Table 2b Complete life table, Canada, 2000 to 2002: males. Ottawa: Statistics Canada; 2006. Available: http://www.statcan.gc.ca/pub/84-537-x/t/pdf/4198612-eng.pdf.
78 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses 98. Johansen HL, Wielgosz AT, Nguyen K, Fry RN. Incidence, comorbidity, case fatality and readmission of hospitalized stroke patients in Canada. Can J Cardiol 2006;22(1):65-71. 99. Tu JV, Gong Y. Trends in treatment and outcomes for acute stroke patients in Ontario, 1992-1998. Arch Intern Med 2003;163(3):293-7. 100. Hardie K, Hankey GJ, Jamrozik K, Broadhurst RJ, Anderson C. Ten-year survival after first-ever stroke in the perth community stroke study. Stroke 2003;34(8):1842-6. 101. Flaherty ML, Haverbusch M, Sekar P, Kissela B, Kleindorfer D, Moomaw CJ, et al. Long-term mortality after intracerebral hemorrhage. Neurology 2006;66(8):1182-6. 102. Kind P, Hardman G, Macran S. UK population norms for EQ-5D. York (UK): Center for Health Economics, University of York; 1999. Available: http://www.york.ac.uk/inst/che/pdf/DP172.pdf (accessed 2009 Jan 30). 103. Reynolds MR, Zimetbaum P, Josephson ME, Ellis E, Danilov T, Cohen DJ. Cost-effectiveness of radiofrequency catheter ablation compared with antiarrhythmic drug therapy for paroxysmal atrial fibrillation. Circ Arrhythm Electrophysiol 2009;2(4):362-9. 104. Rivero-Arias O, Ouellet M, Gray A, Wolstenholme J, Rothwell PM, Luengo-Fernandez R. Mapping the modified Rankin Scale (mRS) measurement into the generic EuroQol (EQ-5D) health outcome. Med Decis Making 2009;Epub 2009 Oct 26. 105. Goeree R, Blackhouse G, Petrovic R, Salama S. Cost of stroke in Canada: a one-year prospective study. J Med Econ 2005;8:147-67. 106. OCCI costing analysis tool. In: Ontario Case Costing Initiative (OCCI). Toronto: Ontario Case Costing Initiative; 2010. Available: http://www.occp.com/ (accessed 2010 Jun 22). 107. Catherwood E, Fitzpatrick WD, Greenberg ML, Holzberger PT, Malenka DJ, Gerling BR, et al. Cost- effectiveness of cardioversion and antiarrhythmic therapy in nonvalvular atrial fibrillation. Ann Intern Med 1999;130(8):625-36. 108. Ministry of Health and Long-Term Care (Ontario). Schedule of benefits for physician services under the Health Insurance Act: effective October 1, 2009. Toronto: The Ministry; 2009. Available: http://www.health.gov.on.ca/english/providers/program/ohip/sob/physserv/physserv_mn.html (accessed 2010 Apr 30). 109. Cappato R, Calkins H, Chen SA, Davies W, Iesaka Y, Kalman J, et al. Worldwide survey on the methods, efficacy, and safety of catheter ablation for human atrial fibrillation. Circulation 2005;111(9):1100-5. 110. Khaykin Y, Morillo CA, Skanes AC, McCracken A, Humphries K, Kerr CR. Cost comparison of catheter ablation and medical therapy in atrial fibrillation. J Cardiovasc Electrophysiol 2007;18(9):907- 13. 111. Statistics Canada. Inflation rate calculations. Ottawa: Statistics Canada; 2010. 112. Statistics Canada. Consumer Price Index, health and personal care, by province (monthly) (Newfoundland and Labrador). Ottawa: Statistics Canada; 2010. Available: http://www40.statcan.ca/l01/cst01/cpis13a-eng.htm (accessed 2010 Jun 22). 113. Ministry of Health and Long-Term Care (Ontario). Drugs funded by Ontario Drug Benefit (ODB) Program: e-formulary [Last updated: 13/04/2009, version 2.0]. Toronto: The Ministry; 2010. Available: http://www.health.gov.on.ca/english/providers/program/drugs/odbf_eformulary.html (accessed 2010 Apr 30). 114. Regier DA, Sunderji R, Lynd LD, Gin K, Marra CA. Cost-effectiveness of self-managed versus physician-managed oral anticoagulation therapy. CMAJ 2006;174(13):1847-52. 115. Guidelines for the economic evaluation of health technologies: Canada. 3rd ed. Ottawa: Canadian Agency for Drugs and Technologies in Health (CADTH); 2006. Available: http://www.cadth.ca/media/pdf/186_EconomicGuidelines_e.pdf (accessed 2010 Apr 30).
Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: 79 Clinical and Cost-Effectiveness Analyses 116. Wolf PA, D'Agostino RB, Belanger AJ, Kannel WB. Probability of stroke: a risk profile from the Framingham Study. Stroke 1991;22(3):312-8. 117. Simpson CS, Healey JS, Philippon F, Dorian P, Mitchell LB, Sapp JL, Jr., et al. Universal access -- but when? Treating the right patient at the right time: access to electrophysiology services in Canada. Can J Cardiol 2006;22(9):741-6. 118. Moher D, Pham B, Jones A, Cook DJ, Jadad AR, Moher M, et al. Does quality of reports of randomised trials affect estimates of intervention efficacy reported in meta-analyses? Lancet 1998;352(9128):609- 13. 119. Schulz KF, Chalmers I, Hayes RJ, Altman DG. Empirical evidence of bias. Dimensions of methodological quality associated with estimates of treatment effects in controlled trials. JAMA 1995;273(5):408-12. 120. Kjaergard LL, Villumsen J, Gluud C. Reported methodologic quality and discrepancies between large and small randomized trials in meta-analyses. Ann Intern Med 2001;135(11):982-9. Erratum in: Ann Intern Med 2008;149(3):219. 121. Petersen P, Boysen G, Godtfredsen J, Andersen ED, Andersen B. Placebo-controlled, randomised trial of warfarin and aspirin for prevention of thromboembolic complications in chronic atrial fibrillation. The Copenhagen AFASAK study. Lancet 1989;1(8631):175-9. 122. The effect of low-dose warfarin on the risk of stroke in patients with nonrheumatic atrial fibrillation. The Boston Area Anticoagulation Trial for Atrial Fibrillation Investigators. N Engl J Med 1990;323(22):1505-11. 123. Connolly SJ, Laupacis A, Gent M, Roberts RS, Cairns JA, Joyner C. Canadian Atrial Fibrillation Anticoagulation (CAFA) Study. J Am Coll Cardiol 1991;18(2):349-55. 124. Stroke Prevention in Atrial Fibrillation Study. Final results. Circulation 1991;84(2):527-39. 125. Ezekowitz MD, Bridgers SL, James KE, Carliner NH, Colling CL, Gornick CC, et al. Warfarin in the prevention of stroke associated with nonrheumatic atrial fibrillation. Veterans Affairs Stroke Prevention in Nonrheumatic Atrial Fibrillation Investigators. N Engl J Med 1992;327(20):1406-12. 126. Secondary prevention in non-rheumatic atrial fibrillation after transient ischaemic attack or minor stroke. EAFT (European Atrial Fibrillation Trial) Study Group. Lancet 1993;342(8882):1255-62.
80 Ablation Procedures for Rhythm Control in Patients with Atrial Fibrillation: Clinical and Cost-Effectiveness Analyses APPENDIX 1: LITERATURE SEARCH STRATEGY FOR CLINICAL EFFECTIVENESS AND ECONOMIC STUDIES
OVERVIEW Interface: Ovid Databases: EMBASE <1980 to present> Ovid MEDLINE In-Process & Other Non-Indexed Citations Ovid Medline <1950 to present> Note: Subject headings have been customized for each database. Duplicates between databases were removed in Ovid. Date of July 28, 2009 Search: Alerts: Monthly search updates will begin August 10, 2009 and run to April 5, 2010 Study Types: RCT’s, MAs, SRs; and Guidelines; also costs and cost analysis studies and economic literature. Limits: Humans for RCTs and guidelines
SYNTAX GUIDE / At the end of a phrase, searches the phrase as a subject heading .sh At the end of a phrase, searches the phrase as a subject heading MeSH Medical Subject Heading fs Floating subheading exp Explode a subject heading * Before a word, indicates that the marked subject heading is a primary topic; or, after a word, a truncation symbol (wildcard) to retrieve plurals or varying endings # Truncation symbol for one character ? Truncation symbol for one or no characters only ADJ Requires words are adjacent to each other (in any order) ADJ# Adjacency within # number of words (in any order) .ti Title .ab Abstract .hw Heading Word; usually includes subject headings and controlled vocabulary .pt Publication type .rn CAS registry number .tw Title, Abstract and Drug Trade Name .nm Name of Substance Word
A-1
EMBASE, Ovid MEDLINE(R) Multi-database Clinical Strategy
# Searches Results
Atrial fibrillation concept
1 (afib or a-fib).ti,ab. 212
2 exp Atrial Fibrillation/ 51814
3 exp Tachycardia/ 83724
4 exp Arrhythmias, Cardiac/ 137956
5 exp Heart Arrhythmia/ 158624
6 *Heart Fibrillation/ 306
7 AF.ti,ab. 24269
((atrial or atrium or atrio* or cardiac or heart or supraventicular or auricular) adj2 (node or 8 76659 arrhythmia* or tachyarrhythmia or fibrillation or tachycardia)).ti,ab.
9 or/1-8 325783
Catheter Ablation concept
10 exp catheter ablation/ 22697
11 exp ablation techniques/ 69598
12 (surg* adj10 atrial fibrillation).ti,ab. 2937
13 (catheter* and (ablat* or isolat*)).ti,ab. 21707
14 (transcatheter and (ablat* or isolat*)).ti,ab. 1234
15 *cryosurgery/ 7913
16 cryosurg*.ti,ab. 5266
17 ((intraoperative* or intra-operative*) adj5 ablat*).ti,ab. 412
18 (maze and surg*).ti,ab. 1891
19 Heart Atria/su [Surgery] 3242
20 pulmonary vein isolation.ti,ab. 941
21 (PVI and pulmonary).ti,ab. 211
22 or/10-21 101383
A-2 EMBASE, Ovid MEDLINE(R) Multi-database Clinical Strategy
RCTs
23 Randomized Controlled Trials as Topic/ 60840
24 Randomized Controlled Trial/ 442598
25 Randomized Controlled Trial.pt. 273098
26 Randomization/ 91474
27 Double Blind Procedure/ 72638
28 Double-Blind Studies/ 174456
29 Single Blind Procedure/ 8203
30 Single-Blind Studies/ 21168
31 Placebos/ or Placebo/ 155320
32 (random* or sham or placebo*).ti,ab,hw. 1260580
33 ((singl* or doubl*) adj (blind* or dumm* or mask*)).ti,ab,hw. 258098
34 ((tripl* or trebl*) adj (blind* or dumm* or mask*)).ti,ab,hw. 397
35 25 or 34 or 26 or 31 or 29 or 30 or 32 or 28 or 23 or 24 or 27 or 33 1293691
Human limit
36 exp animals/ 14176418
37 exp animal experimentation/ 1302297
38 exp models animal/ 804525
39 exp animal experiment/ 1302297
40 nonhuman/ 3232877
41 exp vertebrate/ 22235531
42 or/36-41 23694119
43 exp humans/ 17337143
44 exp human experiment/ 257542
45 43 or 44 17338011
46 42 not 45 6356540
47 9 and 22 and 35 1378
A-3 EMBASE, Ovid MEDLINE(R) Multi-database Clinical Strategy
48 47 not 46 1295
Systematic Reviews and Meta-Analyses
49 Meta-Analysis.pt. 21436
((systematic$ adj3 (review$ or overview$)) or (methodologic$ adj3 (review$ or 50 43179 overview$))).ti,ab.
((quantitative adj3 (review$ or overview$ or synthes$)) or (research adj3 (integrati$ or 51 5891 overview$))).ti,ab.
((integrative adj3 (review$ or overview$)) or (collaborative adj3 (review$ or overview$)) 52 9623 or (pool$ adj3 analy$)).ti,ab.
53 (data synthes$ or data extraction$ or data abstraction$).ti,ab. 15338
54 (handsearch$ or hand search$).ti,ab. 4739
(meta analy$ or metaanaly$ or met analy$ or metanaly$ or health technology assessment$ 55 or HTA or HTAs or biomedical technology assessment$ or bio-medical technology 53159 assessment$).ti,ab.
(meta-analy* or metaanaly* or systematic review* or biomedical technology assessment* 56 83286 or bio-medical tehcnology assessment).hw.
57 (meta regression$ or metaregression$ or mega regression$).ti,ab. 1478
(mantel haenszel or peto or der simonian or dersimonian of fixed effect* or latin 58 8238 square*).ti,ab.
meta-analysis/ or systematic review/ or meta-analysis as topic/ or exp technology 59 90963 assessment, biomedical/
60 (medline or Cochrane or pubmed or medlars).ti,ab,hw. 78908
61 (cochrane or health technology assessment or evidence report).jw. 9567
62 or/49-61 206715
63 22 and 62 and 9 265
64 63 or 48 1418
65 remove duplicates from 64 979
A-4
EMBASE, Ovid MEDLINE(R) Multi-database Guideline Strategy
# Searches Results
Atrial Fibrillation Concept
1 (afib or a-fib).ti,ab. 212
2 exp Atrial Fibrillation/ 51814
3 *Heart Fibrillation/ 306
4 AF.ti,ab. 24262
5 ((atrial or atrium or atrio*) adj2 fibrillation).ti,ab. 46312
Guidelines
6 Guidelines as Topic.sh. 22657
7 Health Planning Guidelines.sh. 2899
8 Practice Guidelines as Topic.sh. 50548
9 Clinical Protocols.sh. 15292
10 Guideline.pt. 14694
11 Practice Guideline.sh. 118375
12 Practice Guideline.pt. 13446
13 Consensus Development Conference.pt. 6440
14 (guideline* or standards).ti. 80641
(expert consensus or consensus statement or consensus conference* or practice parameter* 15 45149 or position statement* or policy statement* or cpg or cpgs or best practice*).ti,ab.
16 or/6-15 290908
17 or/1-5 75932
18 16 and 17 1872
19 remove duplicates from 18 1638
20 exp animals/ 14176418
21 exp animal experimentation/ 1302297
22 exp models animal/ 804525
A-5 EMBASE, Ovid MEDLINE(R) Multi-database Guideline Strategy
23 exp animal experiment/ 1302297
24 nonhuman/ 3232877
25 exp vertebrate/ 22235531
26 or/20-25 23694119
27 exp humans/ 17337143
28 exp human experiment/ 257542
29 27 or 28 17338011
30 26 not 29 6356540
31 19 not 30 1637
A-6
EMBASE, Ovid MEDLINE(R) Multi-database Economic Strategy
# Searches Results
Atrial Fibrillation Concept
1 (afib or a-fib).ti,ab. 212
2 exp Atrial Fibrillation/ 51814
3 exp Tachycardia/ 83724
4 exp Arrhythmias, Cardiac/ 137956
5 exp Heart Arrhythmia/ 158624
6 *Heart Fibrillation/ 306
7 AF.ti,ab. 24269
((atrial or atrium or atrio* or cardiac or heart or supraventicular or auricular) adj2 (node or 8 76659 arrhythmia* or tachyarrhythmia or fibrillation or tachycardia)).ti,ab.
9 or/1-8 325783
Catheter Ablation concept
10 exp catheter ablation/ 22697
11 exp ablation techniques/ 69598
12 (surg* adj10 atrial fibrillation).ti,ab. 2937
13 (catheter* and (ablat* or isolat*)).ti,ab. 21707
14 (transcatheter and (ablat* or isolat*)).ti,ab. 1234
15 *cryosurgery/ 7913
16 cryosurg*.ti,ab. 5266
17 ((intraoperative* or intra-operative*) adj5 ablat*).ti,ab. 412
18 (maze and surg*).ti,ab. 1891
19 PVI.mp. and pulmonary.ti,ab. [mp=ti, ab, sh, hw, tn, ot, dm, mf, nm] 211
20 Heart Atria/su [Surgery] 3242
21 pulmonary vein isolation.ti,ab. 941
22 or/10-21 101383
A-7 EMBASE, Ovid MEDLINE(R) Multi-database Economic Strategy
Primary Economic Studies
23 (Economics or Economics, Medical or Economics, Pharmaceutical or "Value of Life").sh. 44979
24 exp "Costs and Cost Analysis"/ or exp Models, Economic/ 293186
25 economics.fs. 254889
26 (econom$ or cost$ or budget$ or pharmacoeconomic$ or pharmaco-economic$ or valu$).ti. 296100
((cost$ adj benefit$) or costbenefit$ or (cost adj effective$) or costeffective$ or econometric$ or life value or quality-adjusted life year$ or quality adjusted life year$ or 27 115137 quality-adjusted life expectanc$ or quality adjusted life expectanc$ or sensitivity analys$ or "value of life" or "willingness to pay").ti,ab.
28 exp "Health Care Cost"/ or exp Health Economics/ or exp Resource Management/ 277075
(Economic Aspect or Economics or Quality Adjusted Life Year or Socioeconomics or 29 155444 Statistical Model).sh.
30 or/23-29 907828
31 22 and 30 and 9 628
32 remove duplicates from 31 488
OTHER DATABASES PubMed -
A-8 Grey Literature and Hand Searches
Dates for Search: August 2009 – March 2010 Keywords: Included terms for atrial fibrillation and catheter ablation. Limits: Publication years 2009 to 2010
NOTE: This section lists the main agencies, organizations, and websites searched; it is not a complete list. For a complete list of sources searched, contact CADTH (http://www.cadth.ca).
Health Technology Assessment Agencies
Alberta Heritage Foundation for Medical Research (AHFMR) http://www.ahfmr.ab.ca
Agence d’Evaluation des Technologies et des Modes d’Intervention en Santé (AETMIS). Québec http://www.aetmis.gouv.qc.ca
Canadian Agency for Drugs and Technologies in Health (CADTH) http://www.cadth.ca
Centre for Evaluation of Medicines. Father Sean O'Sullivan Research Centre, St.Joseph's Healthcare,Hamilton, and McMaster University, Faculty of Health Sciences. Hamilton, Ontario http://www.thecem.net/
Centre for Health Services and Policy Research, University of British Columbia http://www.chspr.ubc.ca/cgi-bin/pub
Health Quality Council of Alberta (HQCA) http://www.hqca.ca
Health Quality Council. Saskatchewan. http://www.hqc.sk.ca/
Institute for Clinical Evaluative Sciences (ICES). Ontario http://www.ices.on.ca/
Institute of Health Economics (IHE). Alberta http://www.ihe.ca/
Manitoba Centre for Health Policy (MCHP) http://www.umanitoba.ca/centres/mchp/
Ontario Ministry of Health and Long Term Care. Health Technology Analyses and Recommendations http://www.health.gov.on.ca/english/providers/program/ohtac/tech/techlist_mn.html
The Technology Assessment Unit of the McGill University Health Centre http://www.mcgill.ca/tau/
Therapeutics Initiative. Evidence-Based Drug Therapy. University of British Columbia http://www.ti.ubc.ca
Health Technology Assessment International (HTAi) http://www.htai.org
A-9 International Network for Agencies for Health Technology Assessment (INAHTA) http://www.inahta.org
WHO Health Evidence Network http://www.euro.who.int/HEN
Australian Safety and Efficacy Register of New Interventional Procedures – Surgical (ASERNIP-S) http://www.surgeons.org/Content/NavigationMenu/Research/ASERNIPS/default.htm
Centre for Clinical Effectiveness, Monash University http://www.med.monash.edu.au/healthservices/cce/
Medicare Services Advisory Committee, Department of Health and Aging http://www.msac.gov.au/
NPS RADAR (National Prescribing Service Ltd.) http://www.npsradar.org.au/site.php?page=1&content=/npsradar%2Fcontent%2Farchive_alpha.html
Institute of Technology Assessment (ITA) http://www.oeaw.ac.at/ita/index.htm
Federal Kenniscentrum voor de Gezendheidszorg http://www.kenniscentrum.fgov.be
Danish Centre for Evaluation and Health Technology Assessment (DCEHTA). National Board of Health http://www.dihta.dk/
DSI Danish Institute for Health Services Research and Development http://www.dsi.dk/engelsk.html
Finnish Office for Health Care Technology and Assessment (FinOHTA). National Research and Development Centre for Welfare and Health http://finohta.stakes.fi/EN/index.htm
L’Agence Nationale d’Accréditation et d’Evaluation en Santé (ANAES). Ministere de la Santé, de la Famille, et des Personnes handicappés http://www.anaes.fr/anaes/anaesparametrage.nsf/HomePage?ReadForm
Committee for Evaluation and Diffusion of Innovative Technologies (CEDIT) http://cedit.aphp.fr/english/index_present.html
German Institute for Medical Documentation and Information (DIMDI). Federal Ministry of Health http://www.dimdi.de/static/de/hta/db/index.htm
College voor Zorgverzekeringen/Health Care Insurance Board (CVZ) http://www.cvz.nl
Health Council of the Netherlands http://www.gr.nl
New Zealand Health Technology Assessment Clearing House for Health Outcomes and Health Technology Assessment (NZHTA) http://nzhta.chmeds.ac.nz/
Norwegian Centre for Health Technology Assessment (SMM) http://www.kunnskapssenteret.no/
A-10 Agencia de Evaluación de Tecnologias Sanitarias (AETS), Instituto de Salud “Carlos III”/ Health Technology Assessment Agency http://www.isciii.es/htdocs/investigacion/Agencia_quees.jsp
Basque Office for Health Technology Assessment (OSTEBA). Departemento de Sanidad http://www.osasun.ejgv.euskadi.net/r52-2536/es/
Catalan Agency for Health Technology Assessment and Research (CAHTA) http://www.gencat.net/salut/depsan/units/aatrm/html/en/Du8/index.html
CMT - Centre for Medical Technology Assessment http://www.cmt.liu.se/pub/jsp/polopoly.jsp?d=6199&l=en
Swedish Council on Technology Assessment in Health Care (SBU) http://www.sbu.se/
Swiss Network for Health Technology Assessment http://www.snhta.ch/about/index.php
European Information Network on New and Changing Health Technologies (EUROSCAN). University of Birmingham. National Horizon Scanning Centre http://www.euroscan.bham.ac.uk
National Horizon Scanning Centre (NHSC) http://www.pcpoh.bham.ac.uk/publichealth/horizon
NIHR Health Technology Assessment programme, Coordinating Centre for Health Technology Assessment (NCCHTA) http://www.hta.ac.uk/
NHS National Institute for Clinical Excellence (NICE) http://www.nice.org.uk
NHS Quality Improvement Scotland http://www.nhshealthquality.org
University of York NHS Centre for Reviews and Dissemination (NHS CRD) http://www.york.ac.uk/inst/crd
The Wessex Institute for Health Research and Development. Succinct and Timely Evaluated Evidence Review (STEER) http://www.wihrd.soton.ac.uk/
West Midlands Health Technology Assessment Collaboration (WMHTAC) http://www.wmhtac.bham.ac.uk/
Agency for Healthcare Research and Quality (AHRQ) http://www.ahrq.gov/
Dept. of Veterans Affairs Research & Development, general publications http://www1.va.gov/resdev/prt/pubs_individual.cfm?webpage=pubs_ta_reports.htm
VA Technology Assessment Program (VATAP) http://www.va.gov/vatap/
A-11 ECRI http://www.ecri.org/
Institute for Clinical Systems Improvement http://www.icsi.org/index.asp
Blue Cross and Blue Shield Association's Technology Evaluation Center (TEC) http://www.bcbs.com/blueresources/tec/
University HealthSystem Consortium (UHC) http://www.uhc.edu/
Health Economic
Bases Codecs. CODECS (COnnaissances et Décision en EConomie de la Santé) Collège des Economistes de la Santé/INSERM http://infodoc.inserm.fr/codecs/codecs.nsf
Centre for Health Economics and Policy Analysis (CHEPA). Dept. of Clinical Epidemiology and Biostatistics. Faculty of Health Sciences. McMaster University, Canada http://www.chepa.org
Health Economics Research Group (HERG). Brunel University, U.K. http://www.brunel.ac.uk/about/acad/herg
Health Economics Research Unit (HERU). University of Aberdeen http://www.abdn.ac.uk/heru/
The Hospital for Sick Children (Toronto). PEDE Database http://pede.bioinfo.sickkids.on.ca/pede/index.jsp
University of Connecticut. Department of Economics. RePEc database http://ideas.repec.org
Conferences and Meetings
Canadian Cardiovascular Congress http://www.ccs.ca/congress/index_e.aspx
American College of Cardiology Annual Scientific Session http://acc09.acc.org/Pages/default.aspx
European Heart Rhythm Association http://www.escardio.org/communities/EHRA/courses_meetings/Pages/europace-congresses.aspx
Society of Thoracic Surgeons Annual Meeting http://www.sts.org/sections/annualmeeting/
American Heart Association, Scientific Sessions http://www.americanheart.org/presenter.jhtml?identifier=3004167
Organizations
Canadian Cardiovascular Society http://www.ccs.ca/home/index_e.aspx
A-12 American College of Cardiology http://www.acc.org/
American Heart Association http://www.americanheart.org/presenter.jhtml?identifier=3052043
Society of Thoracic Surgeons http://www.sts.org/
European Society of Cardiology http://www.escardio.org/communities/Pages/welcome.aspx
European Heart Rhythm Association http://www.escardio.org/communities/EHRA/Pages/welcome.aspx
Search Engines
Google http://www.google.ca/
AlltheWeb http://www.alltheweb.com/
A-13 APPENDIX 2: LEVEL 1 CLINICAL SCREENING CHECKLIST
Ablation Procedures for Atrial Fibrillation (AF) Level 1 (Title and Abstract) Screening Ref #: ------Author (year):------Reviewer:------1. Live HUMAN subjects or study participants Yes (include) No (exclude) Can’t decide (include)
2. AGES of subjects or study participants Adults 18 years and over (include) Children / Adolescents (exclude) Can’t decide (include)
3. What is the PATIENT GROUP in this article? AF with any duration or severity (paroxysmal, persistent, or permanent) (include) Other cardiac arrhythmias (exclude) Can’t decide (include)
4. What is the INTERVENTION? A minimally invasive ablation procedure which convert AF to normal sinus rhythm (include) A minimally invasive ablation procedure which controls the heart rate (AV node ablation) (exclude) Neither (exclude) Can’t decide (include)
5. TYPE OF STUDY reported in this article Report of a controlled clinical trial (randomized/non-randomized) (include) Meta-analyses/systematic reviews/HTAs (include) Practice/treatment guideline (include) Report of a cohort/registry/phase IV (post-marketing) (exclude) Academic/Narrative Review, Comment, Editorial, Letter, Note, Patient Handout, Study Design Description (exclude) Other observational studies (e.g. Case Control, Cross-Sectional, Case Report/Series, Survey) (exclude) Can’t decide (include)
A-14 APPENDIX 3: LEVEL 2 CLINICAL SCREENING CHECKLIST
Ablation Procedures for Atrial Fibrillation (AF) Level 2 (full text) Screening
Ref #: ------Author (year):------Reviewer:------
1. Is the study PRIMARILY designed to evaluate clinical EFFICACY, EFFECTIVENESS, OR SAFETY of RHYTHM CONTROL ablation procedures for AF including PVI± atrial ablation, or minimally invasive surgical procedure/minimal access catheter Maze procedure? Yes (include) No (exclude) Maybe (include)
2. Is the patient group in this article included patients with AF (paroxysmal, persistent, or permanent)? Yes (include) No (exclude) Maybe (include)
3. Are the study participants adults 18 YEARS AND OVER? Yes (include) No (Exclude)
4. Is the article the PRIMARY REPORT of the FINAL results from: Randomized Controlled Trial (RCT) (include) A non-randomized controlled trial (include) An uncontrolled (before-after) clinical trial (exclude) A prospective or retrospective observational study (exclude) A post-marketing (phase IV) study (exclude) Registry/surveillance data (exclude) Other study design (exclude) Not original report (exclude) Can’t tell (include)
A-15 5. What COMARATOR is used in the study? cardio-version to normal sinus rhythm (either pharmacological or electrical) (include) surgical ablation technique (open-heart Cox-Maze procedure) (include) rate control strategies (exclude) other comparators (exclude) no comparator/placebo (exclude)
6. Include if the OUTCOME of interest in the study is one of the following: Freedom from AF recurrence of AF number of tachycardia/flutter episodes hospitalization quality of life stroke mortality