点击显示 收起
INSERM Unit 330 University Bordeaux 2 33076 Bordeaux France E-mail: serge.renaud{at}bordeaux.inserm.fr
Dear Sir:
In a recent prospective study of 667 men in Zutphen (Netherlands) and of 98 cases of coronary artery disease (CAD), Oomen et al (1) concluded that the protective effect of dietary -linolenic acid (ALA) against CAD is questionable. However, recently confirmed positive effects of ALA (2) were reported in several large studies.
The first prospective study showing a beneficial effect of ALA on CAD was conducted in 6250 middle-aged men of the usual care group of the Multiple Risk Factor Intervention Trial (3). After 10.5 y of follow-up, 175 deaths from CAD occurred in that group. ALA intake, as evaluated by dietary recall interviews at 5 different periods, was significantly inversely related to mortality from CAD (P < 0.04) and from all causes (P < 0.02). The intake of ALA in the highest quintile was 3.2-fold that in the lowest quintile.
More recently, 2 large prospective studies in 76 283 nurses (4) and 43 757 health professionals (5) showed that ALA was the only fatty acid that protected against cardiac death (4) and against nonfatal myocardial infarction (5), independently of other dietary or nondietary factors. In both studies, the intake of ALA in the highest quintile was 1.9-fold that in the lowest quintile.
In the Lyon intervention trial in 600 patients with coronary heart disease (6, 7), both fatal and nonfatal myocardial infarctions were lowered by > 70%. Statistical analysis has indicated that most of the beneficial effects are attributable to plasma ALA concentrations. The experimental group had an ALA intake 2.9-fold that of the control group.
In a double-blind, placebo-controlled study in India, the effects of ALA (supplied by mustard oil) in 120 patients with suspected acute myocardial infarction were compared with those of a placebo in 98 control subjects (8). After 1 y of follow-up, both cardiac death and nonfatal myocardial infarction were significantly lower in the group treated with mustard oil. ALA intake in the treated patients was 3.6-fold that in the placebo group.
Finally, the most recent results of the effects of linolenic acid (mostly ALA) on CAD are from a cross-sectional study in 4406 participants of the National Heart, Lung, and Blood Institute Family Heart Study (2). The intake of ALA was significantly inversely related to the prevalence (485 cases) of CAD, in both women and men.
Concordant with the results mentioned above are those of a dietary intervention study conducted in the entire country of Finland over the past 25 y (9). During that period, CAD mortality was reduced overall by > 65% and by 80% in 40–50-y-old men. Canola oil rich in ALA is now the main oil used for cooking and to make margarines in Finland.
Thus, 7 human studies (3 prospective, 1 cross-sectional, and 3 intervention studies) have reported significant protective effects of a diet enriched in ALA on CAD morbidity, mortality, or both, whereas negative results have only been reported in the Zutphen Elderly Study (1). Because of the suspected key role of ALA in the prevention of CAD, it may be important to unravel the possible explanation (other than the small sample size) for the discrepant results of the Zutphen Elderly Study.
Possible confounding factors in the Zutphen Elderly Study include the following. First, the intake of ALA was strongly associated with that of trans fatty acids, which are known for their positive association with CAD (10). When the statistical analysis was performed only with ALA sources without trans fatty acids, the positive association between the intake of ALA and CAD was no longer observed. In the Nurses Health Study (4), the intake of trans fatty acids also inhibited the inverse relation between ALA and fatal CAD, but not to the extent of the Zutphen Elderly Study. The ratio of trans fatty acids to ALA in the group with the highest intake of ALA in the Nurses Health Study (4) was 2.86 and in the Zutphen Elderly Study was 8.65. Thus, the high intake of trans fatty acids may be the main reason for the discrepant results of the Zutphen Elderly Study. Even when the relation of food without trans fatty acids was evaluated, a residual confounding was probably not totally excluded. Second, an additional factor may be the difference in the intake of ALA between the experimental group or the highest tertile (or quintile) and the control group or the lowest tertile (or quintile). In the Indian intervention trial (8), the intake of ALA in the experimental group was 3.6-fold that in the control group; it was 2.9-fold that in the Lyon study (6). The intake of ALA in the highest quintile or tertile was 3.2-fold that in the lowest quintile or tertile in the prospective Multiple Risk Factor Intervention Trial (3), 2.15-fold that in the men and 2.05-fold that in the women in the Family Heart Study (2), 1.9-fold that in the Nurses Health Study (4) and the Health Professionals Follow-up Study (5), and 1.68-fold that in the Zutphen Elderly Study (1). Thus, it seems that an intake 1.9-fold that of the control group may be required to observe a positive effect of ALA.
REFERENCES