Plasma phospholipid omega-3 fatty acids and incidence of postoperative atrial fibrillation in the OPERA trial

Abstract

Background: Long-chain polyunsaturated omega-3 fatty acids (n-3 PUFA) demonstrated antiarrhythmic potential in experimental studies. In a large multinational randomized trial (OPERA), perioperative fish oil supplementation did not reduce the risk of postoperative atrial fibrillation (PoAF) in cardiac surgery patients. However, whether presupplementation habitual plasma phospholipid n-3 PUFA, or achieved or change in n-3 PUFA level postsupplementation are associated with lower risk of PoAF is unknown.

Methods and results: In 564 subjects undergoing cardiac surgery between August 2010 and June 2012 in 28 centers across 3 countries, plasma phospholipid levels of eicosapentaenoic acid (EPA), docosapentaenoic acid (DPA), and docosahexaenoic acid (DHA) were measured at enrollment and again on the morning of cardiac surgery following fish oil or placebo supplementation (10 g over 3 to 5 days, or 8 g over 2 days). The primary endpoint was incident PoAF lasting ≥ 30 seconds, centrally adjudicated, and confirmed by rhythm strip or ECG. Secondary endpoints included sustained (≥ 1 hour), symptomatic, or treated PoAF; the time to first PoAF; and the number of PoAF episodes per patient. PoAF outcomes were assessed until hospital discharge or postoperative day 10, whichever occurred first. Relative to the baseline, fish oil supplementation increased phospholipid concentrations of EPA (+142%), DPA (+13%), and DHA (+22%) (P < 0.001 each). Substantial interindividual variability was observed for change in total n-3 PUFA (range = -0.7% to 7.5% after 5 days of supplementation). Neither individual nor total circulating n-3 PUFA levels at enrollment, morning of surgery, or change between these time points were associated with risk of PoAF. The multivariable-adjusted OR (95% CI) across increasing quartiles of total n-3 PUFA at enrollment were 1.0, 1.06 (0.60 to 1.90), 1.35 (0.76 to 2.38), and 1.19 (0.64 to 2.20); and for changes in n-3 PUFA between enrollment and the morning of surgery were 1.0, 0.78 (0.44 to 1.39), 0.89 (0.51 to 1.55), and 1.01 (0.58 to 1.75). In stratified analysis, demographic, medication, and cardiac parameters did not significantly modify these associations. Findings were similar for secondary PoAF endpoints.

Conclusions: Among patients undergoing cardiac surgery, neither higher habitual circulating n-3 PUFA levels, nor achieved levels or changes following short-term fish oil supplementation are associated with risk of PoAF.

Trial registration: ClinicalTrials.gov NCT00970489.

Keywords: biomarker; cardiac surgery; omega‐3 fatty acids; postoperative atrial fibrillation; randomized controlled trial.

Figures

Figure 1.

A, Plasma phospholipid concentrations of individual and total n‐3 PUFA in the OPERA trial. Subjects (n=523) received 1 to 5 days of fish oil (8 to 10 g) (white bars) or placebo (black bars). n‐3 PUFA are expressed as percentage of total plasma phospholipid fatty acids analyzed and results are presented as mean±SE of each group. *P<0.001 comparing the fish oil to placebo groups, with adjustment for baseline n‐3 PUFA concentrations. B, Change in plasma phospholipid total n‐3 PUFA concentrations in subjects who received fish oil (n=267). Data were stratified by the number of fish oil loading days presurgery. The top, middle, and lower lines of each box represent the 75th, median, and 25th percentile, respectively. The lines extending above and below the box indicate the most extreme value within the 75th percentile+1.5×(interquartile range) and 25th percentile−1.5×(interquartile range), respectively. The coefficient of variation of the change in total n‐3 PUFA were 49.1%, 66.0%, 53.8%, 43.2%, and 64.3% for subjects who received 1 to 5 days of fish oil supplementation, respectively. DHA indicates docosahexaenoic acid; DPA, docosapentaenoic acid; EPA, eicosapentaenoic acid; n‐3 PUFA, n‐3 polyunsaturated fatty acids; OPERA, omega‐3 fatty acids for prevention of postoperative atrial fibrillation; SE, standard error.

Figure 2.

Multivariable‐adjusted association of change in plasma phospholipid total long‐chain n‐3 PUFA with risk of postoperative atrial fibrillation, evaluated by restricted cubic splines. Adjusted for age (years), gender (male/female), country (US, Italy, Argentina), body mass index (kg/m2), prevalent hypertension (yes/no), prevalent diabetes (yes/no), prevalent coronary heart disease (yes/no), prevalent chronic renal failure (yes/no), prevalent heart failure (yes/no), smoking (never or former/current), dyslipidemia (yes/no), statin medication use (yes/no), ejection fraction (%), LA diameter (tertiles), and logistic Euroscore (continuous). The solid red line and dashed black lines represent the odds ratio and 95% CIs, respectively, in comparison to the reference level representing the median value of the lowest quartile (12.5th percentile). The values shown on the x‐axis correspond to the 10th, 25th, 50th, 75th, and 90th percentiles for changes in total n‐3 PUFA. There was little evidence of either overall association (Wald‐test P=0.70) or nonlinearity (P=0.50). CI indicates confidence interval; n‐3 PUFA, n‐3 polyunsaturated fatty acids.