点击显示 收起
Department of Chronic Diseases Epidemiology National Institute for Public Health and the Environment PO Box 1 3720 BA Bilthoven Netherlands E-mail: ejm.feskens{at}rivm.nl
Dear Sir:
Visioli and Galli offer 2 major comments on our article (1) on the intake of -linolenic acid and the incidence of coronary artery disease (CAD). First, they suggest that we should have controlled for plasma concentrations of -linolenic acid. However, to have done so would have resulted in an overadjustment of the results, thus yielding an invalid estimate. We agree that information on plasma -linolenic acid would have been useful, but not as a variable to be controlled for but instead as one to be studied in separate research on the association between plasma -linolenic acid and CAD risk. In our opinion, the association between dietary -linolenic acid and CAD risk is an interesting research question in itself (1).
In addition, note that the analysis of plasma fatty acids in relation to endpoints in the Lyon Diet Heart Study (2) is not clearly described in terms of adjustments for other dietary changes that were introduced in the trial. Therefore, as mentioned in our Discussion (1), we maintain that on the basis of this trial it cannot be concluded that the protective effect against CAD was solely due to -linolenic acid.
Visioli and Galli's second comment on our article refers to the quality of our dietary intake data. In reply to their criticism of our method of estimating intakes, we point out that we used the crosscheck dietary history method adapted to the Dutch population (1); this method is ackowledged to be valid in epidemiologic settings (1, 3). In addition, we carefully constructed an extensive table on the -linolenic acid content of the foods consumed (1, 4), and this is an exercise in which we have much experience (5). This table was also used by Bemelmans et al (6), who showed that -linolenic acid intake, as assessed by their food-frequency questionnaire and our food table, was clearly associated with the plasma cholesteryl ester concentration of -linolenic acid (r = 0.37). In addition, Visioli and Galli suggest that the intake of subjects in the Lyon Diet Heart Study was more validly assessed, but we respectfully disagree with this suggestion. Although the concentration of -linolenic acid in the experimental and control margarines in that study was known (7), information on the amount of margarine consumed daily, as well as on the consumption of other sources of -linolenic acid, was gained by methods similar to ours, as more extensively described by Renaud et al (8).
Therefore, we agree with the comment that no definite conclusions should be drawn from our data, but the same holds in general for all the results of other individual studies on this topic. We observed no beneficial association. We conclude that the methodologic limitations of our study and of other prospective studies, including trials, and the limited evidence on the mechanisms involved indicate that the protective effect of -linolenic acid has not yet been proven.
REFERENCES