Good News for Marine Polyunsaturated Fatty Acids, but not so Good for Vitamin E

Abstract & Commentary

Synopsis: Long-term polyunsaturated fatty acids but not vitamin E were beneficial for death and combined death, nonfatal myocardial infarction, and stroke due to the decrease in risk for overall cardiovascular death.

Source: GISSI. Lancet 1999;354:447-455.

The latest contribution from the gissi investigators is a study of more than 11,000 individuals with a recent myocardial infarction (MI) (less than 3 months, mean time to study entry 12 days) who were randomized to fish oil or vitamin E, or both in 2 ´ 2 factorial design study. There were approximately 2800 patients in each cell; supplements included N-3 polyunsaturated fatty acids (PUFA) 1 g daily; vitamin E 200 mg; the combination of the above; and neither. The study was open label and was carried out for an average of 3.5 years. The two primary end points were all-cause mortality, nonfatal MI (NFMI), and nonfatal stroke; and cardiovascular death, NFMI, and nonfatal stroke. A secondary analysis was performed for each individual event class. The trial was carried out from 1993-1995. The results were favorable for the fish oil supplement and neutral for vitamin E. Plasma lipids at six months demonstrated a decline in tryglicerides from baseline in individuals taking PUFA, and an increase in LDL cholesterol in all groups, greater in the PUFA cohort. HDL and total cholesterol increased in all groups. The primary outcome demonstrated a decrease with PUFA for all-cause death, NFMI, and stroke (P = 0.053) of 10% (P = 0.048), with a similar decrease of 11% when cardiovascular death was included. Vitamin E resulted in no difference from control, and no further reduction of events when combined with PUFA. A four-way analysis of PUFA indicated a relative decrease of 15% in the combined end point, and 20% in the secondary combined end points. Furthermore, individual event end points demonstrated a decrease in total mortality by 20%, cardiovascular death at 30%, and sudden death at 45% with N-3 PUFA. These represent the major benefits of this trial. Vitamin E demonstrated no differences from control, except for a decrease in cardiovascular death, but not for any of the combined end points. N-3 PUFA plus vitamin E was no more beneficial than N-3 PUFA alone. Adverse effects were relatively minor. Approximately 27% of subjects had discontinued either study drug by the end of the trial. The GISSI investigators noted that the regimen of N-3 PUFA corresponds to a large fatty fish meal every day of the week.

The data are concordant with the DART Trial reported a decade ago that analyzed fish intake on cardiovascular death and reinfarction in post-MI patients. The GISSI investigators emphasize that the study population was relatively low risk, as most of them consumed a Mediterranean diet, and many were treated with aspirin, ACE inhibitors, beta-blockers, and statins. Therefore, this Italian post-MI population represents a model approach to therapy of MI. The GISSI investigators conclude that "long term N-3 PUFA, but not vitamin E…was beneficial for death and for combined death, non-fatal MI, and stroke. All the benefit… was attributable to the decrease in risk for overall cardiovascular death."

Comment by Jonathan Abrams, MD

This is certainly a "good news" story with respect to dietary supplementation with marine fish oils. There are considerable epidemiologic and research data in the literature, including fish oil and fatty fish consumption, that predicted this beneficial outcome. The vitamin E results are disappointing but are concordant with all reported large trials of vitamin E supplementation available today. The mechanisms of PUFA benefits are unclear, and these benefits include antifibrinitytic and lipid modification effects. Decreased oxidation of LDL cholesterol has been suggested for vitamin E. The GISSI investigators believe without good evidence that the major effect of fatty acids was on arrhythmogenesis—not on "atherosclerotic-thrombolic events." They further suggest that the relatively ideal profile of the study cohort with respect to the Mediterranean diet and high rates of use of proven post-MI therapies would make it difficult to demonstrate a major effect of either fatty acids or vitamin E. Nevertheless, the PUFA groups did demonstrate a major benefit.

In an accompanying editorial, Brown suggests that it may take a much larger study to demonstrate a favorable effect of vitamin E. He is less sanguine about the magnitude of PUFA benefit and points out that vitamin E did reduce risk by 11% when compared to no vitamin E, but this did not result in statistical significance. One can therefore conclude that benefit will accrue to post-MI patients who ingest marine fish oils, but not necessarily in the mega-amounts used in prior trials. The benefits of vitamin E remain unproven. Unfortunately, the recently released HOPE Trial of subjects with vascular disease and diabetes also did not show a benefit for vitamin E, although the ACE inhibitor ramipril was shown to be beneficial in reducing cardiovascular risk. Another effective agent, heretofore not proven to be effective, is the fibrate gemfibrozil, which was recently reported to reduce death, recurrent MI, and revascularization rates by 22-23% in a cohort of U.S. veterans with isolated low HDL cholesterol treated with this agent for a period of five years (Rubins HB, et al. N Engl J Med 1999;341:410-418). Mean total and LDL cholesterol were low. There was a decrease in triglyceride levels and a modest increase in HDL throughout the study.

In conclusion, recommendations for secondary prevention now include a statin if LDL cholesterol is elevated (more than 130-135 mg/dL). In individuals with aggressive or premature coronary disease, the use of N-3 polyunsaturated fatty acids should be considered, in addition to a diet high in fatty fish consumption. It remains to be convincingly demonstrated that vitamin E is of any benefit for primary or secondary prevention. In patients with established coronary disease who have a low HDL and otherwise normal lipids, a fibrate is clearly indicated.

Which of the following supplements was shown in GISSI to reduce cardiovascular events?

a. Vitamin E

b. N-3 polyunsaturated fatty acids (fish oil)

c. Vitamin C

d. All of the above