Author + information
- Received October 14, 2012
- Revision received November 15, 2012
- Accepted November 17, 2012
- Published online January 29, 2013.
- Alejandro Macchia, MD⁎,⁎ (, )
- Hugo Grancelli, MD⁎,
- Sergio Varini, MD⁎,
- Daniel Nul, MD⁎,
- Nicolás Laffaye, MD⁎,
- Javier Mariani, MD⁎,
- Daniel Ferrante, MD⁎,
- Raúl Badra, MD†,
- Julio Figal, MD‡,
- Silvina Ramos, MD§,
- Gianni Tognoni, MD∥,
- Hernán C. Doval, MD⁎,
- GESICA Investigators
- ↵⁎Reprint requests and correspondence:
Dr. Alejandro Macchia, Fundación GESICA. Av. Rivadavia 2358, Piso 1, Departamento 4, Ciudad de Buenos Aires, Argentina
Objectives The aim of this study was to evaluate the efficacy of polyunsaturated fatty acids (n-3 PUFA) for the prevention of recurrent atrial fibrillation (AF) in patients with normal sinus rhythm.
Background Current pharmacological treatments to limit recurrent AF in patients with previous AF have limited efficacy and high rates of adverse events. Results of trials that tested the efficacy of n-3 PUFA provided heterogeneous results.
Methods This was a prospective, randomized, double-blind, placebo-controlled, multicenter trial involving 586 outpatient participants with confirmed symptomatic paroxysmal AF that required cardioversion (n = 428), at least 2 episodes of AF in the 6 months before randomization (n = 55), or both (103). Patients were randomly allocated to n-3 PUFA (1 g/day) or placebo for 12 months. The primary endpoint was symptomatic recurrence of AF.
Results There were no significant differences between patients allocated to placebo and those who received n-3 PUFA for the main outcome. At 12 months, 56 of 297 participants (18.9%) in the placebo group and 69 of 289 participants (24.0%) in the n-3 PUFA group had a recurrent symptomatic AF (hazard ratio: 1.28, 95% confidence interval: 0.90 to 1.83, p = 0.17). There was no difference between treatment with placebo and n-3 PUFA for any of the other pre-specified endpoints, including the composite of all-cause mortality, nonfatal stroke, nonfatal acute myocardial infarction, systemic embolism, heart failure development, or severe bleeding that occurred in 20 (6.7%) and 16 (5.5%) of patients randomized to placebo or n-3 PUFA, respectively (hazard ratio: 0.86, 95% confidence interval: 0.44 to 1.66, p = 0.65).
Conclusions Pharmacological supplementation with 1 g of n-3 PUFA for 1 year did not reduce recurrent AF. (Randomized Trial to Assess Efficacy of PUFA for the Maintenance of Sinus Rhythm in Persistent Atrial Fibrillation [FORWARD]; NCT00597220)
Atrial fibrillation (AF) is the most common type of arrhythmia in adults, accounting for approximately one-third of hospital stays for arrhythmia (1). The arrhythmia confers an increased risk of death (2), thromboembolism, and impaired quality of life, even if asymptomatic (3).
Current treatment strategies are less than satisfactory, and a number of trials and a meta-analysis (4) inform us that we should be satisfied with ventricular rate control and anticoagulation. Although many factors might contribute to these unsatisfactory results, the limited efficacy and the many and dangerous side effects of antiarrhythmic drugs might be particularly responsible for this failure (5). In this context, polyunsaturated fatty acids (n-3 PUFA) emerge as attractive candidates to manage AF, because these agents have demonstrated benefit by reducing overall mortality (6), mostly through a decrease in ventricular arrhythmias (7). It has been speculated that some of these antiarrhythmic properties might be translated also to a decrease of atrial arrhythmias (8). However, results of randomized clinical trials were heterogeneous, with some showing promising results and others yielding neutral conclusions (9–14).
We present the results of the FORWARD (Randomized Trial to Assess Efficacy of PUFA for the Maintenance of Sinus Rhythm in Persistent Atrial Fibrillation Fish Oil Research with omega-3 for Atrial fibrillation Recurrence Delaying), a randomized, double-blind, placebo-controlled trial assessing the efficacy of n-3 PUFA supplementation for the maintenance of sinus rhythm in patients with previous AF.
Details of the trial have been previously published (15). Briefly, FORWARD was an independent, double-blind, randomized, controlled trial testing the efficacy of pharmacological supplementation with 1 g/day of n-3 PUFA for 1 year for the maintenance of normal sinus rhythm in patients with previous persistent AF. The study population included men and women ≥21 years diagnosed in an outpatient setting with previous symptomatic AF who had recovered normal sinus rhythm. For inclusion in the trial, patients must have had either: 1) ≥2 symptomatic episodes of documented AF in the 6 months before randomization, with the last episode occurring within 3 to 90 days before randomization (paroxysmal AF); or 2) successful electrical or pharmacological cardioversion for persistent AF performed within 3 to 28 days before randomization.
To avoid the inclusion of patients with lone AF, all subjects <65 years of age must have presented ≥1 characteristic of moderate-to-high risk of stroke. After providing informed consent, patients were centrally assigned in a ratio of 1:1 to receive either 1 g n-3 PUFA (provided by SPA and Sigma-Tau, Italy), which provide 850 to 882 mg eicosapentaenoic acid/docosahexaenoic acid ethyl esters, or placebo (olive oil) in a double-blind fashion. Study treatments were in addition to any other treatments prescribed for AF.
The study was conducted according to the Declaration of Helsinki at 42 Argentinean centers. All centers obtained institutional review board approval for the study before participant screening. All participants provided written informed consent before enrollment and randomization. Clinical outcomes, adherence, and adverse events were assessed 2, 4, 8, and 12 months after randomization.
The primary efficacy point was the time to first recurrence of an AF episode of symptomatic or asymptomatic AF (documented by a 12-lead electrocardiogram). Secondary outcomes included: 1) the hierarchical composite of all-cause mortality, nonfatal stroke, nonfatal acute myocardial infarction, systemic embolism, heart failure development, severe bleeding; 2) all-cause hospital stays; 3) survival free of thromboembolic events; and 4) hospital stays for cardiovascular reasons.
Statistical analysis and sample size calculation
The event rate for a similar endpoint in patients with AF who restored normal sinus rhythm was 40% recurrence at 1 year (16). We originally planned to recruit 1,398 participants. This would have provide 85% power with a 2-sided alpha error of 5% to detect a 20% relative reduction at 1 year—equating to event rates of 40% and 32% in those randomized to placebo and n-3 PUFA, respectively.
A slower-than-expected recruitment rate and lower event rates led to ongoing consideration of recruitment and termination dates by the trial steering committee. Although the trial had not included any efficacy interim analysis, the data and safety monitoring board was formally called to make a formal recommendation about the appropriateness of continuing recruitment. This committee reviewed the data and reported to the Steering Committee that a full complement of planned number of patients would not likely have led to a positive finding and led to stopping the recruitment process.
Efficacy comparisons were performed on the basis of time to the first event, according to the intention-to-treat principle, including all patients in the group to which they were randomized, with patients lost to follow-up censored at the day of the last visit. Safety analyses were performed on data from all enrolled patients. We used the Cox proportional hazards model to estimate hazard ratios (HRs) along with 95% confidence intervals (CIs). All analyses were performed with SAS (version 9.1, SAS Institute, Inc., Cary, North Carolina) and with SPSS (version 16.0 for Windows, SPSS, Inc., Chicago, Illinois).
A total of 586 participants with confirmed symptomatic AF were recruited from January 2008 to March 2011. Of these, 428 required cardioversion, 55 had ≥2 episodes of AF within 6 months before randomization, and 103 met both criteria.
Of the 586 patients randomized, 541 (92.3%) completed the study. Premature withdrawals occurred in 45 patients and were comparable across treatment groups, with the most frequent reason being consent withdrawal (n = 31), followed by gastrointestinal intolerance (n = 14) (Fig. 1). Demographic, clinical, and treatment characteristics were similar between both treatment groups (Table 1).
There were no statistical differences between patients randomized to placebo or to n-3 PUFA for the primary endpoint. At the end of follow-up, 56 of 297 participants (18.9%) in the placebo group and 69 of 289 participants (24.0%) in the n-3 PUFA group had recurrent symptomatic AF (HR: 1.28, 95% CI: 0.90 to 1.83, p = 0.17) (Fig. 2). Additionally, 5 patients (1.7%) in the placebo group and 4 (1.4%) in the n-3 PUFA group died during the study (HR: 0.80, 95% CI: 0.21 to 3.00, p = 0.74).
The composite of all-cause mortality, nonfatal stroke, nonfatal acute myocardial infarction, systemic embolism, heart failure development, or severe bleeding occurred in 20 (6.7%) and 16 (5.5%) of patients randomized to placebo or n-3 PUFA, respectively (HR: 0.86, 95% CI: 0.44 to 1.66, p = 0.65).
The lack of statistically significant differences between patients randomized to placebo and n-3 PUFA were observed in all study outcomes (Table 2). Additionally, all pre-specified subgroup analyses yielded similar results for all tested outcomes. Particularly, the mean effect of n-3 PUFA for prevention of AF recurrence was similar in patients who previously received antiarrhythmic agents (HR: 1.22, 95% CI: 0.78 to 1.91) and those without previous treatment (HR: 1.41, 95% CI: 0.79 to 2.52). Other pre-specified subgroup analyses, including elderly patients, women, and patients with different duration of previous AF, yielded similar results (Fig. 3). Among the 428 patients that were recruited because of a single episode of AF that required cardioversion, the recurrence rate was 20.6% (88 of 428). By contrast, in the remaining 158 patients, the recurrence rate was 23.4% (37 of the 158 patients). This difference was not statistically significant (p = 0.496). If we considered, not proportion, but time to event, the differences remains statistically not significant (p = 0.515).
Few patients discontinued experimental treatment due to intolerance. Specifically, 2.7% of those randomized to placebo and 2.0% of those allocated to n-3 PUFA early abandoned treatment for gastrointestinal intolerance. Additionally, 5.7% in the placebo arm and 4.8% in the n-3 PUFA group withdrew consent during the study. None of these patients reported clinical events.
Nine patients died during the study; no deaths were considered related to treatment.
The results reported here could be interpreted from at least 2 perspectives. The first and more direct interpretation would be that the early stoppage of the FORWARD trial resulted in an underpowered clinical trial unable to verify its hypothesis. However, in absolute terms, the trial captured the second largest number of events among all trials ever conducted in patients with AF (9–14) with n-3 PUFA. When the study was terminated, few patients had experienced clinical events, and the estimation of the effect of n-3 PUFA resulted in no statistically significant reductions of any of the clinical outcomes tested, including AF recurrence, death, hospital stay for any reason, and the composite endpoints.
The low frequency of events could have been the result of several factors. First, the population recruited had a low risk of AF recurrence at recruitment, mainly because most patients (73%) had only experienced a single episode of AF that required cardioversion; the remaining patients had experienced at least 2 episodes of AF within 6 months preceding randomization. Second, a high proportion of patients (63%) received prophylaxis with amiodarone at the time of inclusion. Although the FORWARD trial was a pragmatic clinical trial that allowed the use of any other antiarrhythmic agent, this high proportion of patients who received amiodarone possibly explains the lower-than-expected rate of events. Finally, the identification of patients with recurrent AF was done with clinical criteria and not trans-telephonic monitoring, which might have resulted in a difference in detection of events.
The second interpretation of the results of the FORWARD trial is in the perspective of all randomized clinical trials evaluating n-3 PUFA supplementation for the prevention of recurrent AF. Although both the FORWARD and Kowey et al. (10) trials stopped before fully enrolling their intended patient populations, together they contributed >70% of patients and events in clinical research of this question and might constitute a refutation of the hypothesis of a protective role of these agents for this indication.
The dose used in this trial was quantitatively and qualitatively similar to that effective in the GISSI Prevenzione trial (6) and was deemed appropriate, because of the balance between proved therapeutic action and good tolerance and adherence. It could be that higher doses would have resulted in more efficient prevention of recurrent AF; however, trials that used larger doses provided results materially similar to those reported here (10). It should be acknowledged as a limitation that we did not track fish consumption among participants. It could be that high fish intake might have mitigated the effect of oral supplementation. Although this is possible, it should also be noted that fish intake is low-to-very-low in Argentina.
The FORWARD trial found no significant differences between n-3 PUFA supplementation and placebo in prevention of AF recurrence.
This was an independent clinical trial. Funding was through unrestricted grants provided by the companies that supplied the study drugs: SPA Società Prodotti Antibiotici, Milan, Italy; and Sigma Tau, Rome, Italy; but these companies did not have representatives on the Steering Committee. The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- atrial fibrillation
- n-3 PUFA
- polyunsaturated fatty acids
- Received October 14, 2012.
- Revision received November 15, 2012.
- Accepted November 17, 2012.
- American College of Cardiology Foundation
- Bialy D.,
- Lehmann M.H.,
- Schumacher D.N.,
- Steinman R.T.,
- Meissne M.D.
- Benjamin E.J.,
- Wolf P.A.,
- D'Agostino R.B.,
- Silbershatz H.,
- Kannel W.B.,
- Levy D.
- Fuster V.,
- Rydén L.E.,
- Cannom D.S.,
- et al.
- Marchioli R.,
- Barzi F.,
- Bomba E.,
- et al.,
- GISSI-Prevenzione Investigators
- Bianconi L.,
- Calò L.,
- Mennuni M.,
- et al.
- Erdogan A.,
- Bayer M.,
- Kollath D.,
- et al.
- Nodari S.,
- Triggiani M.,
- Campia U.,
- et al.
- Margos P.,
- Leftheriotis D.,
- Katsouras G.,
- Livanis E.G.,
- Kremastinos D.T.
- Macchia A.,
- Varini S.,
- Grancelli H.,
- et al.,
- FORomegaARD investigators