ABC | Volume 113, Nº4, October 2019

Updated Updated Cardiovascular Prevention Guideline of the Brazilian Society Of Cardiology – 2019 Arq Bras Cardiol. 2019; 113(4):787-891 aspects that can influence the chance of developing CVD. 225,226 Despite the general agreement that regular consumption of fish rich in omega-3 fatty acids is part of a healthy diet, recommending dietary supplementation with fish oil capsules is controversial, fostered by conflicting results of clinical studies. 32,227-229 6.9. Effects of Omega-3 on the Lipid Profile Clinical studies show that 2 to 4 grams of EPA/DHA supplementation per day can reduce TG levels by up to 25 to 30%, slightly increase HDL-c (1 to 3%), and raise LDL-c by 5-10%. 32 The ability to decrease TG levels depends on the dose, with a reduction of approximately 5 to 10% for each 1 g of EPA/DHA consumed per day, which can be higher in individuals with greater baseline TG concentrations. These data show that high doses of omega-3 supplementation can be used to treat hypertriglyceridemia. 6.10. Omega-3 and Cardiovascular Outcomes In a meta-analysis of 36 RCT, supplementation with fish oil (median dose of 3.7 g/day) reduced SBP by 3.5 mmHg and DBP by 2.4 mmHg. 230 The decrease in adrenergic tonus and systemic vascular resistance is a proposed mechanism. Although several old pieces of evidence suggest a protective effect of fish and omega-3 fatty acids of marine origin on CV events, 230 particularly in individuals with established CVD, more recent studies, showed no benefit of omega-3 supplementation for subjects who had or had not presented manifestations of atherosclerotic disease. 227,228 In fact, a meta- analysis of 10 studies involving 77,917 individuals both in secondary (64% with prior coronary disease, 28% with prior CVA) and primary prevention (37% of diabetic individuals) failed to show any benefit of omega-3 supplementation (EPA doses ranging from 226 to 1,800 mg/day and DHA from 0 to 1,700 mg/d) after a mean follow-up of 4.4 years, presenting 6,273 coronary events (2,695 coronary deaths). 231 These results were confirmed in an extensive systematic review and meta-analysis by the Cochrane group, with more than 119,000 individuals from 79 randomized studies. 232 Also, the same meta-analysis did not find the benefit of supplementation with alpha-linolenic acid (ALA), the plant omega-3. Possible reasons for the divergent results between old and contemporary studies concern the profile of the population studied, mainly regarding the more frequent use of medicines known as protectors (e.g., statins, beta-blockers, ACEI), the more aggressive control of traditional risk factors, and the higher number of myocardial revascularization procedures in more recent studies. Another difficulty in analyzing studies with omega-3 supplementation is the diversity in its composition and the lack of control regarding the intake of omega-3 in the diet. More recently, two published clinical trials used low doses (up to 1 g/day of EPA + DHA) of omega-3 in primary prevention of CVD. One of them has assessed the role of omega-3 in the primary prevention of CVD and cancer among men over 50 years of age and women over 55 years of age (VITAL study). 233 Using a formulation containing 460 mg of EPA and 380 mg of DHA, the study included 25,871 patients with a median follow-up of 5.3 years and found no benefit of omega-3 in reducing major CV event or invasive cancer. 233 Another study on primary prevention, but in diabetic patients, also examined the combination of EPA/DHA in the same composition of the VITAL study. It included 15,480 diabetic patients followed for an average of 7.4 years and found no benefit of omega-3 in reducing major vascular event. 234 Thus, the role of omega-3 fatty acids in the doses used with respect to the primary prevention of CV events is questionable. The Reduction of Cardiovascular Events with Icosapent Ethyl-Intervention Trial (REDUCE-IT) 33 tested omega-3 in the reduction of CV outcomes among patients with hypertriglyceridemia and established CVD or diabetic patients with an additional risk factor. The patients received highly purified EPA (icosapent ethyl) at a dose of 4 g/day. The study included 8,179 patients who used statins and had TG ranging from 135 to 499 mg/dL (median of 216 mg/dL) and a median LDL-c of 74 mg/dL. The median reduction was 18% for TG and 6.6% for LDL-c in the EPA group. REDUCE-IT showed a relative decrease of 25% in composite CV outcomes among patients who received EPA and an absolute risk reduction of 4.8%, NNT of 22 patients to prevent an event. The hierarchical analysis demonstrated a significant decrease of 20% in CV mortality. On the other hand, the risk of hospitalization for atrial flutter or fibrillation presented a relative increase of 67% (1% absolute) with the treatment. The reduction of events in the REDUCE-IT study is similar to the results of the Japan EPA Lipid Intervention Study (JELIS), in which 1.8 g/day of EPA also led to a significant decrease in CV events among individuals who already used low doses of statins. 227 However, the results of this last study are limited by their open design and lack of a placebo group. Data from these studies suggest that high doses of EPA (4 g) can be used in patients with prior CVD and who remain with elevated TG levels, despite taking statins to prevent CVD. However, there is no evidence for the use of lower doses and other formulations of omega-3 for CV prevention, both primary and secondary. We emphasize, however, that several studies are still testing moderate to high doses of EPA and EPA + DHA in individuals at high risk for CVD who present persistent moderately high TG. 6.11. Omega-3 in Heart Failure The GISSI-Heart Failure (GISSI-HF) trial evaluated the role of omega-3 in HF. 235 This study randomized patients with chronic HF functional class II-IV of different etiologies to receive 1 g of omega-3 (EPA + DHA) (n = 3,494) or placebo (n = 3,481) per day. 235 The primary outcome was time to death and time to death or hospitalization for CV causes. During a median follow-up of 3.9 years, the omega-3 group presented lower mortality rate (27 versus 29%, HR 0.91, 95% confidence interval 0.83–0.99, p = 0.041, with NNT = 56) and lower incidence of primary outcome (57 versus 59%, HR 0.92, 95% confidence interval 0.84–0.99, p = 0.009, with NNT = 44). 235 Data from this study, however, need to be confirmed. 820