Omega-3 fatty acids have multiple cardiovascular benefits but may also inhibit platelet aggregation and increase bleeding risk. If this platelet inhibition is clinically meaningful, patients with the highest omega-3 indexes (red blood cell eicosapentaenoic acid plus docosahexaenoic acid), which reflect long-term omega-3 fatty acid intake, should be at the risk for bleeding. In this study, 1,523 patients from 24 United States centers who had their omega-3 indexes assessed at the time of acute myocardial infarction were studied. The rates of serious bleeding (Thrombolysis In Myocardial Infarction [TIMI] major or minor) and mild to moderate bleeding (TIMI minimal) were identified in patients with low (<4%), intermediate (4% to 8%), and high (>8%) omega-3 indices. There were no differences in bleeding across omega-3 index categories. After multivariate adjustment, there remained no association between the omega-3 index and either serious (per 2% increase, relative risk 1.03, 95% confidence interval 0.90 to 1.19) or mild to moderate bleeding (per 2% increase, relative risk 1.02, 95% confidence interval 0.85 to 1.23). In conclusion, no relation was found between the omega-3 index and bleeding in this large, multicenter cohort of patients with acute myocardial infarction, suggesting that concerns about bleeding should not preclude the use of omega-3 supplements or increased fish consumption when clinically indicated.
Several previous studies have found no relation between omega-3 supplementation and bleeding in patients with cardiovascular disease, yet most of these studies focused on long-term bleeding risk, and few studied patients with acute myocardial infarction (AMI). Patients with AMI are at particularly high risk for in-hospital bleeding because of intensive treatment with multiple potent antithrombotic agents (such as intravenous heparin, glycoprotein IIb/IIIa inhibitors, and dual-antiplatelet therapy) and the use of invasive diagnostic and therapeutic procedures. Accordingly, if guideline-recommended omega-3 fatty acid intake is associated with increased bleeding risk in the setting of modern antithrombotic regimens, patients hospitalized with AMI who have high red blood cell omega-3 indexes should be most likely to develop bleeding complications. We studied the relation between the omega-3 index and bleeding in 1,523 patients from 24 United States hospitals who had their omega-3 indexes assessed at the time of AMI.
Methods
The design and methods of the Translational Research Investigating Underlying Disparities in Acute Myocardial Infarction Patients’ Health Status (TRIUMPH) study have been previously reported. Patients were ≥18 years of age, with elevated cardiac biomarkers (troponin or creatine kinase-MB fraction assessed <24 hours after admission) and had supporting evidence of AMI (electrocardiographic ST-segment changes or prolonged ischemic signs or symptoms). Participants were required either to present to the enrolling hospital or to have been transferred within 24 hours of presentation, so the primary clinical decision making occurred at the enrolling center. Patients with elevated cardiac biomarkers caused by elective coronary revascularization were excluded. Trained data collectors performed detailed baseline chart abstractions to document patients’ medical histories, the processes of inpatient care, laboratory results, and treatments. Each patient underwent a standardized interview by a research staff member to document sociodemographic and clinical data. Patients were enrolled in TRIUMPH from April 11, 2005, to December 31, 2008. All 1,523 who were enrolled before September 28, 2007, had red blood cells collected for omega-3 analyses and formed the analytic cohort for the present study. All patients provided written informed consent approved by the participating institution, and institutional review board approval was obtained at each participating center.
Red blood cells were obtained from ethylenediaminetetraacetic acid blood samples after the plasma and buffy coat were removed. Briefly, red blood cell aliquots were heated at 100°C for 10 minutes with methanol containing 14% boron trifluoride. The fatty acid methyl esters generated were extracted with hexane and water and were analyzed with a GC2010 gas chromatograph (Shimadzu Corporation, Columbia, Maryland) equipped with a 30-m capillary column (Omegawax 250; Supelco, Bellefonte, Pennsylvania). Fatty acids were identified through comparison with a standard fatty acid methyl ester mixture (GLC-727; Nu-Chek Prep, Inc., Elysian, Minnesota). The coefficient of variation for the omega-3 index is <5%. Values are expressed as eicosapentaenoic acid plus docosahexaenoic acid, as a percentage of total red blood cell fatty acids. Congruent with previous work, patients were classified as having a low omega-3 index (unfavorable, associated with higher mortality risk) if their omega-3 index was <4%, as having intermediate omega-3 indices when values were 4% to 8%, and as having a high omega-3 index (favorable, associated with lower mortality risk) when their omega-3 values were >8%.
Trained data collectors prospectively recorded all in-hospital bleeding events, the site of bleeding (cardiac catheterization site, gastrointestinal, intracranial, retroperitoneal, or other), and the severity of bleeding using the Thrombolysis In Myocardial Infarction (TIMI) classification. TIMI major bleeding was defined as intracranial hemorrhage or a hemoglobin decrease >5 g/dl. TIMI minor bleeding was assigned if the decrease in hemoglobin was 3 to 5 g/dl in the setting of observed bleeding. Any bleeding episode with a decrease in hemoglobin <3 g/dl was classified as TIMI minimal bleeding. All TIMI categories accounted for blood transfusion, with adjustment of hemoglobin values by 1 g/dl per unit transfused. Because TIMI major and minor bleeding represent clinically meaningful bleeding, the composite of these events was considered together as serious bleeding. TIMI minimal bleeding, which has not been linked to poor outcomes but could influence recommendations for antiplatelet or anticoagulant therapy, was considered mild to moderate bleeding.
The patient characteristics, treatments and bleeding rates of patients with low, intermediate, and high omega-3 indexes are presented as mean ± SD for continuous variables and were compared using 1-way analysis of variance. Categorical variables are presented as proportions and were compared using chi-square tests. To identify the independent association between the omega-3 index and bleeding, we fit separate hierarchical modified Poisson regression models for each bleeding outcome (serious and mild to moderate). These models accounted for clustering within hospitals by including enrolling hospital as a random variable. To maximize power, the omega-3 index was modeled as a continuous variable. We adjusted for potentially important confounders that we identified a priori on the basis of previous research and clinical experience. These covariates included age and gender, history of chronic heart failure, peripheral arterial disease, initial creatinine, initial hemoglobin, body weight, myocardial infarction type (ST-segment elevation myocardial infarction vs non–ST-segment elevation myocardial infarction), prehospital warfarin use, in-hospital cardiac catheterization or percutaneous coronary intervention, and use of bivalirudin, intravenous heparin, glycoprotein IIb/IIa inhibitors, and thienopyridines. We also tested for clinically and statistically significant interactions (p<0.05) between omega-3 index and key AMI treatments. These included interactions of omega-3 index with heparin, thienopyridine, bivalirudin, glycoprotein IIb/IIIa inhibitor use, and cardiac catheterization or percutaneous coronary intervention. Missing data for model covariates were minimal (2 patients were missing initial creatinine values).
We calculated the power to detect clinically meaningful differences in the rate of bleeding, defined as a 25% relative difference in bleeding rates between groups, for major bleeding and mild to moderate bleeding. We then calculated the power to detect these differences for the comparison of patients with omega-3 indices of <4% versus ≥4% and for omega-3 indexes of ≤8% versus >8%. All analyses were conducted using SAS version 9.2 (SAS Institute Inc., Cary, North Carolina).
Results
At the time of AMI, 408 patients (26.8%) had low omega-3 indices (<4%), 1,036 (68.0%) had intermediate values (4% to 8%), and 79 (5.2%) had high omega-3 indices (>8%). The mean omega-3 index was 3.3 ± 0.5%, 5.4 ± 1.0%, and 9.3 ± 1.0% among those in low, intermediate, and high groups, respectively. Patients with higher omega-3 indices were older, more frequently had histories of previous myocardial infarction and coronary revascularization, and had higher discharge to 6-month Global Registry of Acute Coronary Events (GRACE) scores ( Table 1 ). They also had lower admission hemoglobin values, were more likely to be taking omega-3 supplements, and were less frequently treated with fibrinolytic therapy.
| Variable | Omega-3 Index | p Value | ||
|---|---|---|---|---|
| 0% to <4% (n = 408) | 4% to 8% (n = 1,036) | >8% (n = 79) | ||
| Omega-3 index (%) | 3.3 ± 0.5 | 5.4 ± 1.0 | 9.3 ± 1.0 | <0.001 |
| Age (years) | 54.6 ± 11.0 | 60.6 ± 12.6 | 64.0 ± 10.5 | <0.001 |
| Men | 287 (70.3%) | 682 (65.8%) | 60 (75.9%) | 0.067 |
| Caucasian | 336 (82.4%) | 732 (70.8%) | 55 (69.6%) | <0.001 |
| Body mass index (kg/m 2 ) | 29.4 ± 6.5 | 29.7 ± 6.5 | 29.1 ± 5.6 | 0.506 |
| Ejection fraction (%) | 48.3 ± 13.7 | 49.3 ± 13.0 | 49.4 ± 13.3 | 0.476 |
| GRACE 6-month mortality risk score | 90.8 ± 28.0 | 102.5 ± 30.5 | 108.8 ± 28.0 | <0.001 |
| Initial hemoglobin (g/dl) | 14.5 ± 2.1 | 14.0 ± 2.1 | 13.9 ± 2.0 | 0.001 |
| Creatinine (mg/dl) | 1.0 (0.9–1.2) | 1.0 (0.9–1.2) | 1.1 (0.9–1.3) | 0.029 |
| Chronic heart failure | 21 (5.1%) | 96 (9.3%) | 8 (10.1%) | 0.030 |
| Dyslipidemia | 190 (46.6%) | 543 (52.4%) | 53 (67.1%) | 0.002 |
| Hypertension | 245 (60.0%) | 674 (65.1%) | 57 (72.2%) | 0.062 |
| Peripheral vascular disease | 14 (3.4%) | 57 (5.5%) | 12 (15.2%) | <0.001 |
| Previous myocardial infarction | 72 (17.6%) | 191 (18.4%) | 25 (31.6%) | 0.011 |
| Previous percutaneous coronary intervention | 71 (17.4%) | 204 (19.7%) | 29 (36.7%) | <0.001 |
| Previous coronary artery bypass grafting | 31 (7.6%) | 129 (12.5%) | 17 (21.5%) | <0.001 |
| Diabetes mellitus | 123 (30.1%) | 318 (30.7%) | 16 (20.3%) | 0.148 |
| Chronic kidney disease | 16 (3.9%) | 85 (8.2%) | 2 (2.5%) | 0.004 |
| ST-segment elevation myocardial infarction | 210 (51.5%) | 457 (44.1%) | 35 (44.3%) | 0.039 |
| Admission medications | ||||
| Aspirin | 129 (31.6%) | 438 (42.3%) | 47 (59.5%) | <0.001 |
| Thienopyridine | 27 (6.6%) | 128 (12.4%) | 13 (16.5%) | 0.002 |
| Warfarin | 17 (4.2%) | 39 (3.8%) | 5 (6.3%) | 0.445 |
| Omega-3 supplements | 26 (6.4%) | 180 (17.5%) | 45 (58.4%) | <0.001 |
| In-hospital treatments | ||||
| Aspirin | 391 (95.8%) | 983 (94.9%) | 79 (100.0%) | 0.069 |
| Intravenous heparin | 325 (79.7%) | 811 (78.3%) | 60 (75.9%) | 0.720 |
| Low–molecular weight heparin | 65 (15.9%) | 158 (15.3%) | 13 (16.5%) | 0.922 |
| Fibrinolytic | 35 (8.6%) | 67 (6.5%) | 0 (0.0%) | 0.018 |
| Glycoprotein IIb/IIIa antagonist | 270 (66.2%) | 637 (61.5%) | 45 (57.0%) | 0.147 |
| Thienopyridine | 303 (74.3%) | 723 (69.8%) | 57 (72.2%) | 0.235 |
| Bivalirudin | 18 (4.4%) | 46 (4.4%) | 5 (6.3%) | 0.656 |
| In-hospital coronary artery bypass | 37 (9.1%) | 90 (8.7%) | 9 (11.4%) | 0.714 |
| In-hospital cardiac catheterization | 390 (95.6%) | 966 (93.2%) | 73 (92.4%) | 0.195 |
| In-hospital percutaneous coronary intervention | 289 (70.8%) | 706 (68.1%) | 53 (67.1%) | 0.577 |
There was no crude association between the omega-3 index and either serious bleeding or mild to moderate bleeding ( Table 2 ), and there were no significant differences in the site of bleeding across the omega-3 index categories. There were also no significant differences in bleeding rates after stratifying the population by use of omega-3 supplements at the time of arrival at the hospital (serious bleeding: 14 of 251 [5.5%] using omega-3 supplements vs 103 of 1,258 [8.2%] not using omega-3 supplements, p = 0.18; mild to moderate bleeding: 17 of 251 [6.8%] using omega-3 supplements vs 67 of 1,258 [5.3%] not using omega-3 supplements, p = 0.43).
