Effect of Dietary Fish Oil on Atrial Fibrillation After Cardiac Surgery

Transcription

1 Effect of Dietary Fish Oil on Atrial Fibrillation After Cardiac Surgery Aaron L. Farquharson, MNursSc a, Robert G. Metcalf, PhD b,e, *, Prashanthan Sanders, MBBS, PhD b,f, Robert Stuklis, MBBS g, James R.M. Edwards, MBBS g, Robert A. Gibson, PhD c, Leslie G. Cleland, MBBS, MD a,c, Thomas R. Sullivan, BMaCompSc d, Michael J. James, PhD a,b,e, and Glenn D. Young, MBBS b,f An open-label study reported that ingestion of a fish oil concentrate decreased the incidence of atrial fibrillation (AF) after coronary artery bypass grafting (CABG) surgery. However, a general cardiac surgery population involves valve and CABG surgeries. We undertook a double-blinded randomized controlled trial to examine the effectiveness of fish oil supplementation on the incidence of postsurgical AF after CABG and valve procedures. The primary end point was incidence of AF in the first 6 days after surgery. Two hundred patients were randomized to receive fish oil (providing 4.6 g/day of long-chain -3 fatty acids) or a control oil starting 3 weeks before surgery; 194 subjects completed the study, with 47 of 97 subjects in the control group and 36 of 97 subjects in the fish oil group developing AF (odds ratio 0.63, 95% confidence interval [CI] 0.35 to 1.11). There was a nonstatistically significant delay in time to onset of AF in the fish oil group (hazard ratio 0.66, 95% CI 0.43 to 1.01). There was a significant decrease in mean length of stay in the intensive care unit in the fish oil group (ratio of means 0.71, 95% CI 0.56 to 0.90). In conclusion, in a mixed cardiac surgery population, supplementation with dietary fish oil did not result in a significant decrease in the incidence of postsurgical AF. However, there was a significant decrease in time spent in the intensive care unit Elsevier Inc. All rights reserved. (Am J Cardiol 2011;108: ) Atrial fibrillation (AF) is a common postoperative complication of cardiac surgery and is associated with increased length of intensive care unit (ICU) and hospital stays and an increased mortality. 1 3 A previous open-label study reported that preoperative intervention with -3 polyunsaturated fatty acids (PUFAs) markedly decreased the incidence of postoperative AF in patients undergoing coronary artery bypass grafting (CABG) surgery by 65%. 4 Although this demonstration of an atrial antiarrhythmic effect of -3 PUFAs complements reports of its ventricular antiarrhythmic action in animals 5 7 and humans, 8,9 the magnitude of decrease suggests a profound effect that needs to be confirmed in a randomized controlled study. Therefore, in this prospective double-blinded randomized controlled study, we examined the effect of perioperative fish oil supplementation on occurrence of AF in a general a Discipline of Medicine, b Centre for Heart Rhythm Disorders, c FoodPlus Research Centre, d Data Management and Analysis Centre, Discipline of Public Health, University of Adelaide, Adelaide, Australia; e Rheumatology Unit and f Departments of Cardiology and g Cardiothoracic Surgery, Royal Adelaide Hospital, Adelaide, Australia. Manuscript received February 22, 2011; revised manuscript received and accepted April 29, This work was supported by Grant from the National Health and Medical Research Council of Australia, Canberra, Australia. Dr. Sanders is supported by the National Heart Foundation of Australia, Melbourne, Australia. Dr. Gibson is supported by the National Health and Medical Research Council of Australia. *Corresponding author: Tel: ; fax: address: robert.metcalf@health.sa.gov.au (R.G. Metcalf). cardiac surgery population that included CABG and/or valve surgical patients. Methods This study was a single-center, randomized, doubleblinded, placebo-controlled trial investigating the effect of -3 PUFA supplementation on incidence of AF after cardiac surgery. The study protocol was approved by the Royal Adelaide Hospital (Adelaide, Australia) research ethics committee and was registered with the Australian New Zealand Clinical Trials Register ( identifier ACTRN ). All patients 18 years of age who were accepted for cardiac surgery involving CABG and/or valve repair or replacement at the Royal Adelaide Hospital were eligible for inclusion in the study. The following exclusion criteria were used: previous diagnosis of AF or atrial flutter, antiarrhythmic drug use (class 1 or 3) within the previous 3 months, urgent surgery ( 3 weeks), New York Heart Association class IV heart failure, myocardial infarction within previous 2 weeks, or any condition that might affect the ability to ingest or absorb dietary fat. To maximize differences in tissue -3 PUFA levels between the treatment and control groups, we also excluded patients who consumed dietary supplements rich in -3 oils, e.g., fish oil or flaxseed oil, or self-reported habitual consumption of 1 fish meal per week. All patients provided written informed consent to the study protocol. Study enrollment commenced in April 2006 and concluded in December Patients were randomly allocated to fish oil or placebo /11/$ see front matter 2011 Elsevier Inc. All rights reserved. doi: /j.amjcard

2 852 The American Journal of Cardiology ( Table 1 Fatty acid composition of oils used (percent total fatty acids) Fatty Acid Fish Oil Sunola Total saturated fatty acids Oleic acid (18:1-9) Total monounsaturated fatty acids Linoleic acid (18:2-6) Arachidonic acid (20:4-6) Total -6 fatty acids Linolenic acid (18:3-3) Stearidonic acid (18:4-3) Eicosapentaenoic acid (20:5-3) Docosapentaenoic acid (22:5-3) Docosahexaenoic acid (22:6-3) Total -3 fatty acids oil (high monounsaturated sunflower oil). Group assignment was based on a computer-generated randomization list using blocks of 20 and individual allocation was by sealed envelope. The 2 oils were supplied in liquid form in 500-ml bottles (Melrose Laboratories, Pty. Ltd., Mitcham, Victoria, Australia) and were citrus flavored to increase masking. Compositions of the oils used are presented in Table 1. Subjects were instructed to ingest oil 15 ml/day, providing eicosapentaenoic acid 2.7 g/day and docosahexaenoic acid 1.9 g/day in the fish oil group, commencing 3 weeks before their scheduled surgery date. This dose of eicosapentaenoic acid plus docosahexaenoic acid 4.5 g/day was chosen to increase rapid short-term incorporation into tissue phospholipids and maximize differences in tissue -3 PUFA content between the treatment and control groups. Patients were instructed to maintain their normal dietary patterns and to not consume any additional oil supplements. In the event of surgery being delayed, participants were instructed to continue with the allocated oil until surgery. Subjects continued to ingest the allocated oil for 6 days after surgery or until discharge, whichever came first. Apart from this intervention, monitoring and treatment were standard practice as determined by the attending physician. Blood was sampled at baseline and before surgery for fatty acid analysis. A sample of atrial appendage was obtained at the time of surgery. Collection and evaluation procedures were as described previously. 10 All patients underwent surgical intervention as clinically indicated. All but 1 of the procedures used cardiopulmonary on-pump intervention. No patient underwent concurrent atrial ablation. Postoperative management was in accordance with routine care within the institution. A minimum of 72 hours of continuous electrocardiographic monitoring was performed. After this patients underwent daily 12-lead electrocardiographic recording until discharge from the hospital with further continuous monitoring if there were any symptoms or signs to suggest AF. To determine a clinically relevant period of atrial arrhythmia, we prospectively established an event as 1 that lasted for 10 minutes or that required intervention. AF was defined as an irregular rhythm with no discernable discrete atrial activation. Atrial flutter or atrial tachycardia was defined as an atrial rate 100 beats/min with discrete atrial activation with P-wave structure distinct from that of sinus rhythm. In addition, onset and termination of these latter tachycardias were scrutinized to exclude sinus tachycardia. All episodes were reviewed by 2 investigators blinded to patient treatment arm, with disagreement being resolved by consensus. The primary outcome measurement was occurrence of sustained AF/atrial flutter (duration 10 minutes or requiring intervention) during the first 6 postoperative days or until discharge if this occurred first. Secondary outcomes were time to first occurrence of AF and length of time in cardiothoracic ICU and total hospital length of stay. Logistic regression was used to compare the odds of in-hospital AF between treatment groups. A multivariate logistic regression model adjusted for potential confounding variables that were unbalanced by the randomization process (gender and surgery type) was also performed. Analysis of time to first episode of AF was performed by the Kaplan Meier method. Cox proportional hazards regression was used to compare treatment groups and control for potential confounding effects of gender and surgery type. Length of time in the ICU and total length of stay were assessed using negative binomial regression in unadjusted models and models adjusted for gender and surgery type. Results are expressed as mean SD unless otherwise specified. A probability value 0.05 (2-tailed) was considered statistically significant. Analysis was performed using SAS 9.2 (SAS Institute, Cary, North Carolina) using intention-to-treat principles. A previous open-label trial reported a 54% relative risk decrease in AF after CABG using a modified fish oil in which the fatty acids were present as ethyl esters. 4 It was hypothesized that our study using unmodified commonly available fish oil would have a similar effect size but with a different control event rate because our study included valve procedures and CABG. Based on the ratio of CABG/aortic valve/mitral valve/combination procedures at this hospital, an overall event rate of 42% was estimated using published figures for postoperative AF rates for each procedure. 11 In this scenario, 200 participants would provide 90% power to detect a 53% relative risk decrease with a p value Results Two hundred subjects were enrolled into the study, 100 into each group. Six subjects did not have surgery (3 in each group), leaving 194 subjects (97 in each group) included in the intention-to-treat analysis (Figure 1). Demographic, clinical, and surgical variables were similar in the control and fish oil groups, with the exception of gender, where more men were enrolled into the fish oil group, and surgery type, where more patients undergoing a valve procedure were enrolled into the control group (Table 2). The study aimed for ingestion of fish oil or control oil for 3 weeks before surgery. The surgery date varied according to availability of staff and theaters. Five subjects (1 in the control group and 4 in the fish oil group) had their surgery

3 Preventive Cardiology/Fish Oil and Postoperative Atrial Fibrillation 853 Figure 1. Screening, randomization, and follow-up. Table 2 Baseline characteristics of patients according to study group Variable Control (n 97) Fish Oil (n 97) Age (years) Men 62 (64%) 80 (82%) Body mass index (kg/m 2 ) Hypertension 75 (77%) 76 (78%) Myocardial infarction 34 (35%) 34 (35%) Stroke 4 (4%) 5 (5%) Diabetes mellitus 35 (36%) 26 (27%) Chronic obstructive pulmonary disease 12 (12%) 9 (9%) Left ventricular ejection fraction (%) Smoker 67 (69%) 80 (71%) Medications Angiotensin-converting enzyme inhibitor/ 57 (59%) 53 (55%) angiotensin receptor blocker Blockers 38 (39%) 42 (43%) Calcium channel blocker 39 (40%) 34 (35%) Statin 71 (73%) 71 (73%) Aspirin 68 (70%) 76 (78%) Clopidogrel 17 (18%) 17 (18%) Type of surgery Coronary artery bypass grafting only 53 (55%) 69 (71%) Valve coronary artery bypass grafting 44 (45%) 28 (29%) Time in theater (minutes) Bypass time (minutes) Aortic crossclamp time (minutes) Data are presented as mean SD or number (percentage). brought forward and did not consume any oil before surgery. When surgery was delayed, participants continued to ingest the allocated oil until surgery. Median (interquartile range) times on treatment were 22 days (18 to 28) and 21 days (13 to 35) for the control and fish oil groups, respectively (p 0.8, Mann Whitney U test). There were no differences in red blood cell fatty acid levels between the intervention and control groups at baseline (Table 3). At the time of surgery the fish oil group had significant increases from baseline in red blood cell eicosapentaenoic acid and docosahexaenoic acid and a significant decrease in -6 PUFAs, whereas there were no significant changes in the control group (Table 3). The fish oil supplemented group had significantly larger proportions of eicosapentaenoic acid and docosahexaenoic acid and a significantly lower proportion of arachidonic acid in atrial tissue than the control group (Table 4). Overall incidence of in-hospital AF was 43%. AF occurred in 47 of 97 (48%) in the control group and 36 of 97 (37%) in the fish oil group (unadjusted odds ratio 0.63, 95% confidence interval [CI] 0.35 to 1.11, p 0.11). After adjustment for baseline covariates that were imbalanced by randomization, gender and type of surgery (CABG only vs valve), the adjusted odds ratio was 0.70 (95% CI 0.39 to 1.28, p 0.25). Kaplan Meier estimates of time to first occurrence of AF showed a consistently delayed time to AF in the fish oil group compared to the control group, but this was not statistically significant (p 0.06, log-rank test; Figure 2). After Cox proportional hazards regression analysis, the unadjusted hazard ratio for time to first episode of AF associated with fish oil was 0.66 (95% CI 0.43 to 1.01, p 0.06) and the hazard ratio was 0.71 (95% CI 0.46 to 1.12, p 0.14) after adjustment for gender and type of surgery (CABG only vs valve). Length of postoperative stay in the cardiac ICU was significantly decreased in the fish oil group compared to the control group (67 52 and hours, respectively,

4 854 The American Journal of Cardiology ( Table 3 Red blood cell phospholipid fatty acids (percent total fatty acids) Control Fish Oil Baseline At Surgery Baseline At Surgery (n 79) (n 80) (n 78) (n 77) Total saturated fatty acids * Oleic acid (18:1-9) * Total monounsaturated fatty acids * Linoleic acid (18:2-6) * Arachidonic acid (20:4-6) * Total -6 fatty acids * -Linolenic acid (18:3-3) Eicosapentaenoic acid (20:5-3) * Docosapentaenoic acid (22:5-3) * Docosahexaenoic acid (22:6-3) * Eicosapentaenoic acid docosahexaenoic acid * Total long-chain -3 fatty acids * Data are presented as mean SD. *p 0.05 compared to control (analysis of variance, Tukey post hoc analysis). p 0.05 compared to baseline (analysis of variance, Tukey post hoc analysis). Table 4 Atrial phospholipid fatty acids (percent total fatty acids) Control Fish Oil p Value (n 59) (n 64) Total saturated fatty acids Total monounsaturated fatty acids Linoleic acid (18:2-6) Arachidonic acid (20:4-6) Total -6 fatty acids Linolenic acid (18:3-3) Eicosapentaenoic acid (20:5-3) Docosapentaenoic acid (22:5-3) Docosahexaenoic acid (22:6-3) Eicosapentaenoic acid docosahexaenoic acid Total long-chain -3 fatty acids Data are presented as mean SD. unadjusted ratio of means 0.71, 95% CI 0.56 to 0.90, p 0.005). After adjustment for gender and surgery type, the difference between groups remained unchanged (ratio of means 0.71, 95% CI 0.56 to 0.90, p 0.006). Mean lengths of hospital stay were days in the fish oil group and days in the control group (unadjusted ratio of means 0.87, 95% CI 0.73 to 1.04, p 0.12). After adjustment for gender and surgery type, the ratio of length of stay means was 0.90 (95% CI 0.75 to 1.08, p 0.24). There were no differences in any adverse events between groups and there was no difference in blood loss through mediastinal chest drains (Table 5). A larger proportion of subjects in the control group received transfusion of red blood cells after surgery. Figure 2. Kaplan Meier curve for time to onset of first episode of atrial fibrillation after surgery. Discussion In this prospective, randomized, double-blinded, placebo controlled study in patients undergoing cardiac surgery involving CABG and/or valve procedures, perioperative treatment with high-dose fish oil failed to significantly decrease the incidence of AF. However, in prespecified secondary analyses, fish oil supplementation was associated with a statistically significant decrease in time spent in the cardiac ICU. Fish oil was associated with a nonstatistically significant decrease in time to a first AF event. In contrast with our results, Calò et al 4 reported that -3 PUFAs as ethyl esters decreased the incidence of postsurgical AF by 65% in patients undergoing isolated CABG surgery. Incidences of AF in the control groups were 33% in the study by Calò et al and 48% in the present study, with a major difference between studies being the patient population enrolled. Our study included patients undergoing valve surgery and CABG, with the increased event rate in our study being due primarily to inclusion of patients undergoing valve surgery. In our

5 Preventive Cardiology/Fish Oil and Postoperative Atrial Fibrillation 855 Table 5 Postsurgical adverse events Control Fish Oil p Value (n 97) (n 97) Death (in hospital) 1 (1%) 2 (2%) 1.00 Ventricular arrhythmia 6 (6%) 6 (6%) 1.00 Respiratory failure 7 (7%) 8 (8%) 1.00 Cardiogenic shock 2 (2%) 2 (2%) 1.00 Myocardial infarction 3 (3%) 0 (0%) 0.25 Stroke 2 (2%) 1 (1%) 1.00 Infections 22 (23%) 27 (28%) 0.51 Blood loss (ml)* 1, , Major bleeding episode 8 (8%) 3 (3%) 0.21 Subjects receiving blood products Red blood cells 42 (43%) 25 (26%) 0.02 Platelets 9 (10%) 5 (5%) 0.41 Fresh frozen plasma 12 (13%) 6 (6%) 0.21 Data are presented as mean SD or number (percentage). * Total blood loss through mediastinal chest drains. Total blood loss 3 L through chest tube drains. control group, incidence of AF in subjects undergoing CABG or valve surgery was 38% or 61%, respectively. Two other studies of fish oil intervention for postoperative AF are not comparable to the studies discussed earlier. A United Kingdom study with 108 participants undergoing CABG used an end-point definition of AF 30 seconds. This is expected to include a substantial number of nonclinically significant events that might terminate spontaneously. 12 End points in the study by Calò et al 4 and our study were AF 5 and 10 minutes, respectively, or requiring intervention. An Icelandic study used an end point of AF 5 minutes. However, approximately 80% of participants were taking cod liver oil or other fish oils at enrollment and there were only very small changes in plasma -3 PUFAs because of the intervention, including a decrease in the control group. 13 Thus, it is unlikely that this study adequately tested the hypothesis that fish oil could affect incidence of AF. Relative differences between the effect of surgery type on event rate in our study are similar to those reported in a multicenter Veterans Affairs study with 4,000 patients in which rates of AF were 49% after mitral valve replacement, 33% after aortic valve replacement, and 28% after CABG. 11 Corresponding event rates in our control group were 75%, 54%, and 38%, respectively. This suggests different causal conditions for AF after CABG compared to valve procedures. Although not significant, we found a trend toward a greater decrease in AF with fish oil in subjects undergoing valve surgery compared to those undergoing CABG only (data not shown), which may warrant further investigation. Although there was no difference overall in incidence of AF between the 2 groups, fish oil significantly decreased residence time in the cardiac ICU by approximately 30%. Visual inspection of the Kaplan Meier plot (Figure 1) indicates the greatest deviation of curves occurred in the first 24 hours, i.e., during the period in the cardiac ICU. Thus, the decreased time in the ICU in the fish oil group is likely to be due to the decreased incidence of AF in the first 24 hours after surgery, although this study was not able to examine that. Regardless of the reasons for decreased time in the ICU, this outcome is likely to be significant for a health economic analysis because of the expenses inherent in intensive care. Our study has several limitations. First, there was only 1 published study on which to base an estimated effect size and our study used a different surgical population. Although we saw a trend toward a decrease in incidence of AF, at rates observed in this study we would have required a sample size of about 300 per group for the result to be statistically significant. Second, we had no information on required target tissue -3 PUFA levels. Therefore, we based our dosing regimen on our previous experience with supplementing fish oil to cardiac surgical patients and selected a dose that would allow rapid increases in tissue -3 PUFAs in a relatively short period. 10,14 Third, although the treatment and placebo oils were flavored with citrus oil to increase masking, 1 inherent difficulty with fish oil supplementation trials is blinding of treatment and potential self-supplementation by patients. Fourth, we specifically enrolled patients with low baseline -3 PUFA intakes to maximize the differences in tissue -3 PUFAs between groups at time of surgery. However, despite statistically significant differences in -3 PUFA levels between groups, there was still considerable overlap in the range of -3 PUFAs between groups. This is an ongoing issue for most, if not all, dietary fish oil supplementation studies. A more appropriate design for future studies may be to treat to a target -3 PUFA level to overcome this problem. 1. Hill LL, Kattapuram M, Hogue CW Jr. Management of atrial fibrillation after cardiac surgery part I: Pathophysiology and risks. J Cardiothorac Vasc Anesth 2002;16: Siebert J, Anisimowicz L, Lango R, Rogowski J, Pawlaczyk R, Brzezinski M, Beta S, Narkiewicz M. Atrial fibrillation after coronary artery bypass grafting: does the type of procedure influence the early postoperative incidence? Eur J Cardiothorac Surg 2001;19: Villareal RP, Hariharan R, Liu BC, Kar B, Lee VV, Elayda M, Lopez JA, Rasekh A, Wilson JM, Massumi A. Postoperative atrial fibrillation and mortality after coronary artery bypass surgery. J Am Coll Cardiol 2004;43: Calò L, Bianconi L, Colivicchi F, Lamberti F, Loricchio ML, de Ruvo E, Meo A, Pandozi C, Staibano M, Santini M. N-3 fatty acids for the prevention of atrial fibrillation after coronary artery bypass surgery: a randomized, controlled trial. J Am Coll Cardiol 2005;45: Billman GE, Kang JX, Leaf A. Prevention of sudden cardiac death by dietary pure omega-3 polyunsaturated fatty acids in dogs. Circulation 1999;99: McLennan PL, Abeywardena MY, Charnock JS. Dietary fish oil prevents ventricular fibrillation following coronary artery occlusion and reperfusion. Am Heart J 1988;116: McLennan PL, Bridle TM, Abeywardena MY, Charnock JS. Dietary lipid modulation of ventricular fibrillation threshold in the marmoset monkey. Am Heart J 1992;123: Metcalf RG, Sanders P, James MJ, Cleland LG, Young GD. Effect of dietary n-3 polyunsaturated fatty acids on the inducibility of ventricular tachycardia in patients with ischemic cardiomyopathy. Am J Cardiol 2008;101: Schrepf R, Limmert T, Claus Weber P, Theisen K, Sellmayer A. Immediate effects of n-3 fatty acid infusion on the induction of sustained ventricular tachycardia. Lancet 2004;363: Metcalf RG, James MJ, Gibson RA, Edwards JR, Stubberfield J, Stuklis R, Roberts-Thomson K, Young GD, Cleland LG. Effects of fish-oil supplementation on myocardial fatty acids in humans. Am J Clin Nutr 2007;85:

6 856 The American Journal of Cardiology ( 11. Almassi GH, Schowalter T, Nicolosi AC, Aggarwal A, Moritz TE, Henderson WG, Tarazi R, Shroyer AL, Sethi GK, Grover FL, Hammermeister KE. Atrial fibrillation after cardiac surgery: a major morbid event? Ann Surg 1997;226: Saravanan P, Bridgewater B, West AL, O Neill SC, Calder PC, Davidson NC. Omega-3 fatty acid supplementation does not reduce risk of atrial fibrillation after coronary artery bypass surgery: a randomized, double-blind, placebo-controlled clinical trial. Circ Arrhythmia Electrophysiol 2010;3: Heidarsdottir R, Arnar DO, Skuladottir GV, Torfason B, Edvardsson V, Gottskalksson G, Palsson R, Indridason OS. Does treatment with n-3 polyunsaturated fatty acids prevent atrial fibrillation after open heart surgery? Europace 2010;12: Metcalf RG, Cleland LG, Gibson RA, Roberts-Thomson KC, Edwards JR, Sanders P, Stuklis R, James MJ, Young GD. Relation between blood and atrial fatty acids in patients undergoing cardiac bypass surgery. Am J Clin Nutr 2010;91: