点击显示 收起
Courser Road
Sutton, QC
Canada
E-mail: vos{at}health-heart.org
Department of Nutritional Sciences
University of Toronto
Toronto, ON
Canada
Research Center on Aging
University of Sherbrooke
Sherbrooke, QC
Canada
Dear Sir:
In evaluating the cardiovascular merits of n–3 fatty acids, Wang et al (1) appropriately sought the best available evidence and used randomized controlled human trials (RCTs) as their benchmark. As they showed, such trials designed to examine the cardiovascular effects of -linolenic acid (ALA) have not been done to everyone's satisfaction. Because they clearly defined at the outset what they meant by "high-quality evidence," we cannot disagree with their conclusion there is no "high-quality evidence to support a beneficial effect of ALA." What we reject however is that, once they make this initial point, Wang et al slide down a slippery slope toward broad enthusiasm for fish oils and outright dismissal of ALA. Neither position is supported by the existing literature.
First, it is incorrect to say that Dolecek's analysis of the MRFIT study (2) showed no association between ALA and cardiac death; Dolecek's analysis showed that, as a percentage of energy intake and in g/d, ALA was significantly negatively associated with cardiovascular, cancer, and all-cause mortality. Second, acute, short-term experiments showed that ALA has antiarrhythmic effects (3) and reduces platelet aggregation (4), and both effects could plausibly contribute significantly toward reduction of cardiovascular and all-cause mortality. Third, for all its possible confounders, the Lyon study (5) was a randomized controlled secondary prevention trial that, supported by a blood fatty acid analysis, clearly implicated ALA in risk reduction of cardiovascular disease and death. Hence, these diverse examples are consistent with cardiovascular benefits of ALA. They can in no way substitute for placebo-controlled RCTs, but they show that grounds exist for well-controlled trials to assess whether ALA reduces the risk of cardiovascular death.
Furthermore, Wang et al cite the concern with regard to ALA and prostate cancer in the absence of confirmatory RCT evidence but downplay some potentially equally important adverse cardiovascular effects of fish oils where RCTs exist. For instance, they cite the recent study by Raitt et al (6), which was conducted in subjects with an implantable cardioverter defibrillator, and mention that the risk of death did not change but downplayed the significantly increased risk of ventricular tachycardia or ventricular fibrillation when consuming 1.3 g fish oil n–3 fatty acids/d. At the time of publication of the article by Wang et al, they may not have been aware that Frost and Vestergaard (7) showed in a population study that Danes who consumed 1.29 g fish oil n–3 fatty acids/d (the top quintile) had a 34% higher rate of atrial fibrillation than did those who consumed 0.16 g fish oil n–3 fatty acids/d (the bottom quintile).
We are not saying that these reports of adverse cardiovascular outcomes with consumption of fish oils constitute sufficient evidence to dismiss the beneficial effects seen in controlled trials. We are saying that a systematic review purporting to give an "evidence-based review" of the cardiovascular effects of n–3 fatty acids should not conflate an absence of well-controlled trials examining cardiovascular effects of ALA with an absence of evidence that ALA has any benefits for the cardiovascular system. Furthermore, not all would agree that the arrhythmogenic effects of fish oils in certain cardiac patients are "minor;" the adverse effects of all n–3 fatty acids should be given appropriate and similar scrutiny. Clearly, additional ALA trials are overdue considering the strength of the existing evidence and the seriousness of the disease.
ACKNOWLEDGMENTS
None of the authors had any conflicts of interest.
REFERENCES