点击显示 收起
1 From the Department of Clinical Endocrinology and Metabolism C, Aarhus University Hospital, Aarhus, Denmark (BMR, EP, and KH); the Unit for Clinical Nutrition Research, Department of Public Health and Caring Sciences/Geriatrics, University of Uppsala, Uppsala, Sweden (BV); the Department of Clinical Nutrition, University of Kuopio, Kuopio, Finland (MU); the Uppsala Clinical Research Center, Uppsala University, Uppsala, Sweden (LB); the Department of Clinical and Experimental Medicine, School of Medicine, Federico II University, Naples, Italy (GR and AAR); and the Department of Biomedical Sciences and Smarts Food Centre, University of Wollongong, Wollongong, Australia (LT)
2 Supported by the Danish Medical Research Council, the Swedish Council for Forestry and Agricultural Research, Health Research Council Academy of Finland, Helga and Peter Kornings Foundation, and International Council of Olive Oil. The food for the study was generously supplied by MD Foods (ARLA), Denmark; Carlshamn Mejeri AB, Svenska Nestlé AB, and Van den Bergh Foods AB, Sweden; Eridania Beghin-Say, Belgium; and Meadow Lea Foods, Australia. The Pikasol capsules were supplied by Lube Ltd, Hadsund, Denmark. 3 Reprints not available. Address correspondence to K Hermansen, Department of Clinical Endocrinology and Metabolism C, Aarhus University Hospital, Aarhus Sygehus, Tage-Hansens Gade 2, DK 8000 Aarhus C, Denmark. E-mail: kjeld.hermansen{at}as.aaa.dk.
ABSTRACT
Background: The quantity and quality of fats consumed in the diet influence the risk of cardiovascular disease (CVD). Although the effect of diet on plasma lipids and lipoproteins is well documented, less information exists on the role of fats on blood pressure (BP).
Objective: The objective was to evaluate the effects of different types of dietary fat on BP in healthy subjects.
Design: Healthy subjects (n = 162) were randomly assigned for 3 mo to follow 1 of 2 isoenergetic diets: 1 rich in monounsaturated fatty acids (MUFA diet) and the other rich in saturated fatty acids (SFA diet). Each group was further randomly assigned to receive supplementation with fish oil (3.6 g n–3 fatty acids/d) or placebo.
Results: Systolic BP (SBP) and diastolic BP (DBP) decreased with the MUFA diet [–2.2% (P = 0.009) and –3.8% (P = 0.0001), respectively] but did not change with the SFA diet [–1.0% (P = 0.2084) and –1.1% (P = 0.2116)]. The MUFA diet caused a significantly lower DBP than did the SFA diet (P = 0.0475). Interestingly, the favorable effects of MUFA on DBP disappeared at a total fat intake above the median (>37% of energy). The addition of n–3 fatty acids influenced neither SBP nor DBP.
Conclusions: Changing the proportions of dietary fat by decreasing SFAs and increasing MUFAs decreased diastolic BP. Interestingly, the beneficial effect on BP induced by fat quality was negated by the consumption of a high total fat intake. The addition of n–3 fatty acids to the diet had no significant effect on BP.
Key Words: Diet • saturated fatty acids • monounsaturated fatty acids • n–3 fatty acids • blood pressure
INTRODUCTION
The quantity and quality of fats consumed in the diet are important features that influence the risk of cardiovascular disease (CVD) (1). The paradigm that dietary fats act exclusively via effects on serum lipids and lipoproteins has been challenged (2-6). Evidence suggesting the beneficial health effects of the Mediterranean diet has emerged from the classic studies of Keys (7), which indicate that the consumption of diets enriched in monounsaturated fat (MUFA) relates to a lower incidence of coronary heart disease. The Lyon Diet Heart Study (3) showed that a Mediterranean-type diet reduces the rate of recurrence after a first myocardial infarction. The replacement of saturated fat (SFA) with MUFA and -linolenic acids (n–3 fatty acids) seems to induce beneficial effects in persons with CVD without changing plasma lipid concentrations. The Diet and Reinfarction Trial (DART; 5) supported the ability of diets rich in eicosapentaenoic acid (EPA) to lower ischemic complications of arteriosclerosis. The study by Trichopoulou et al (8) corroborated the Lyon Diet Heart Study (3) and showed that strict adherence to a Mediterranean diet is associated with a significant reduction in total mortality.
Studies focusing on surrogate risk markers for CVD other than plasma lipids and lipoproteins are needed to clarify the effects of dietary fat on CVD. In this regard, it seems prudent to evaluate the effect of dietary fat on CVD risk factors such as blood pressure (BP), insulin sensitivity, endothelial function, hemostatic factors, and microalbuminuria.
Genetic factors seem to be responsible for as much as 20–40% of BP variations in the general population (9). However, epidemiologic data implicate that lifestyle factors (eg, dietary habits) are a major contributor to the high prevalence of hypertension (10). Nevertheless, our knowledge of the influence of macronutrients such as fat on BP is limited. Data suggest that both the fat quantity and quality of the diet could be important for the development of insulin resistance (11). In the KANWU study (12, 13) we showed that MUFA, in contrast with SFA, improved the insulin sensitivity in healthy subjects (12) and concomitantly reduced plasma cholesterol and triacylglycerol concentrations (13). More recently, it was shown that insulin sensitivity improved when dietary SFA was substituted for polyunsaturated fat (PUFA) (14). An important question is whether the quantity or quality of fat also affects BP and thereby the risk of hypertension. The aim of this study was to investigate whether dietary MUFA, compared with SFA, affects BP in healthy subjects. Second, we investigated whether the addition of long-chain n–3 fatty acids had a modifying effect.
SUBJECTS AND METHODS
Design
The study was a 3-mo controlled, parallel, multicenter study, performed at 5 different centers (Kuopio, Finland; Aarhus, Denmark; Naples, Italy; Wollongong, Australia; and Uppsala, Sweden). The design was reported in detail previously (12).
Healthy subjects were randomly assigned to a diet containing either a high proportion of SFAs (SFA diet) or a high proportion of MUFAs (MUFA diet). Within each of these 2 groups, the subjects were randomly assigned to receive supplementary capsules containing fish oil [3.6 g n–3 fatty acids/d providing 2.4 g EPA and docosahexaenoic fatty acids (DHA); Pikasol, Lube Ltd, Hadsund, Denmark] or placebo capsules containing the same amount of olive oil.
The test period was preceded by a 2-wk run-in period during which the subjects consumed their habitual diets supplemented with placebo capsules. Routine clinical tests, including an oral-glucose-tolerance test (15), were carried out during this period. The subjects kept a 3-d dietary record (2 weekdays and 1 weekend day) to document pretrial dietary habits. Two additional 3-d dietary records were kept at the beginning of the second and third months of the test period. Tests and laboratory analyses were carried out at baseline and at the end of the study. Blood pressure was measured at baseline and at the end of study.
Subjects
A total of 162 healthy white subjects (n = 95 men and 67 women) aged 30–65 y and with normal or moderately increased body weight [BMI (in kg/m2) = 22–32] were included. Health status was screened via medical history and routine laboratory examinations. Subjects with impaired glucose tolerance (15) but without diabetes were included. Other reasons for exclusion were specific eating habits due to cultural or religious beliefs, high habitual physical activity, high alcohol intake (ie, binge drinking or a regular alcohol intake >40 g/d), and hepatic, cardiac, thyroid, and disabling diseases. Body weight during the past 3 mo should not have changed >4 kg. Subjects taking acetyl salicylic acid, thiazide diuretics, ß-blockers, lipid-lowering drugs, and corticosteroids were also excluded. If a subject was taking any other medication, the dose had to remain constant during the test period. Alcohol consumption, weight, and the degree of physical activity were kept stable during the study. Smoking was allowed, but smoking habits had to remain unchanged during the study. Premenopausal women had all tests made during the same period of their menstrual cycle.
All subjects were fully informed of the experimental nature of the investigation, which had been approved by the local ethics committees. Informed consent was obtained before the study began, and all subjects complied fully with the protocol.
Diets
All subjects were instructed to eat isoenergetic diets containing the same amount of macronutrients: 37% of energy as fat with a high proportion of SFAs (SFA diet) or MUFAs (MUFA diet). The SFA diet contained 17% of energy as SFAs, 14% of energy as MUFAs, and 6% of energy as PUFAs, whereas the MUFA diet contained 8% of energy as SFAs, 23% of energy as MUFAs, and 6% of energy as PUFAs, respectively. Trained dietitians instructed all subjects on the preparation of their diets. To ensure good adherence to the diets, the subjects met the dietitians at least every second week until the end of the study. The participants were supplied with edible fats to be used as spreads on bread, for cooking, and in dressings. Core foods such as margarine, oils, and a range of other staple items were provided. Butter, margarines, and oils to be used in the diets were prepared centrally and distributed to the different European centers. The SFA diet included butter and table margarine containing a relatively high proportion of SFAs. The MUFA diet included spread and margarine with a high proportion of oleic acid, derived from high–oleic acid sunflower oil and negligible amounts of trans fatty acids, n–3 fatty acids, and olive oil. The study center in Australia obtained similar oil from local suppliers. The intake during the test period was calculated as the mean values from the dietary records provided during the second and third months of the study. The dietary records were estimated, not weighed. Local nutrient analysis software programs containing country-specific food databases were used in the analyses. Data on margarine and other specially prepared foods were entered into these databases for inclusion in the analyses. The adherence to the test fats was verified by analyses of the phospholipid fatty acid composition of serum (12).
Blood pressure and body weight
Blood pressure was measured at baseline and at the end of the study to the nearest 5 mm Hg with a sphygmomanometer. Systolic and diastolic BP was defined as phase I and V Korotkoff sounds, respectively. Blood pressure was measured from the same arm, with subjects in a sitting position, after 10 min of rest from the time the cuff had been placed on the arm. Measurements were made 3 times at 2-min intervals in each case. The data analyzed were the means of the 3 BP values. Body weight was measured at each visit to ensure that the participants were weight stable.
Statistical methods
Results are presented as mean ± SD. The study was an intention to treat study, ie all randomized subjects, with at least one measurement during treatment were included in the analyses. The treatment effects were estimated from a statistical model in which treatment categories (SFA or MUFA diet with or without n–3 fatty acids) and their interaction were analyzed factors, whereas center, age, sex, and baseline value of the outcome variable were covariates. The difference between groups for adjusted mean treatment effects are presented with P values and 95% CIs.
A post hoc subgroup analysis was made according to the relative intake of total fat during treatment (above or below the median of 37% of energy). The abovementioned model was used with the addition of an interaction term between treatment and relative fat intake (above or below the median).
Of the 83 subjects assigned to the SFA diet, 42 subjects were in the SFA + placebo group (1 dropout) and 41 subjects were in the SFA + n–3 fatty acid group (1 dropout). Of the 79 subjects assigned to the MUFA diet, 40 subjects were in the MUFA + placebo group (no dropouts) and 39 subjects were in the MUFA + n–3 fatty acid group (1 dropout).
RESULTS
The clinical characteristics of the 162 subjects (n = 76 women and 86 men) assigned randomly to the SFA and MUFA diets did not differ significantly (Table 1). Mean (±SD) body mass index (BMI; in kg/m2) and body weight remained unchanged during the study. BMIs at baseline and at end of the SFA diet period were 26.6 ± 2.9 and 26.7 ± 2.9 and at the end of the MUFA diet were 26.6 ± 3.1 and 26.3 ± 3.2, respectively.
View this table:
TABLE 1. Clinical characteristics of the healthy subjects at entry1
Dietary records
The average nutrient composition before the study, as calculated from the dietary records (Table 2), was not different between the SFA and MUFA diet groups, respectively. During the test period there was a slight increase in the proportion of dietary fat in both groups. The recorded mean intake of fat and fatty acids during the study was similar to the target values. During the study the amount of fiber was significantly higher (P = 0.0444), and the amount of cholesterol was significantly lower (P = 0.0006), in the MUFA than in the SFA diet. The estimated proportions of trans fatty acids were low and not different between the diets. Adherence to the diets was not different between the SFA and MUFA diet groups. These data were reported previously (12).
View this table:
TABLE 2. Dietary nutrient composition in the healthy subjects before and during the study1
Effects on blood pressure
A significant decrease in SBP (–2.2%; P = 0.009) and DBP (–3.8%; P = 0.0001) was observed during the MUFA diet, whereas no significant changes were seen during the SFA diet. The MUFA diet caused a significantly lower DBP than did the SFA diet (P = 0.0475) (data not shown). We also looked at the effects of a low compared with a high fat intake, ie, a fat intake below or above the median fat intake of 37% of energy (Table 3). In subjects with a total fat intake below the median, the MUFA diet reduced the SBP (P = 0.0408) and the DBP (P = 0.0023) significantly. Interestingly, the differences in SBP and in DBP disappeared with a total fat intake above the median. Addition of n–3 fatty acids did not influence SBP or DBP. Thus, the mean (±SD) changes in DBP in response to n–3 fatty acids and to placebo were –2.2 ± 0.7 and –1.6 ± 0.7 mm Hg (P = 0.5700), respectively, and the corresponding changes in SBP were –2.2 ± 1.1 and –1.8 ± 1.1 mm Hg (P = 0.7567), respectively (data not shown). As seen in Table 3, the addition of n–3 fatty acids had no influence on SBP or DBP. In the main model (without % of energy as fat), no interactions were found between SFA and MUFA and n–3 fatty acids. In the post hoc analysis, however, the 3-factor interaction between SFA or MUFA, n–3 fatty acids, and % of energy as fat (above or below the median) was significant for the change in DBP (P = 0.048). There were no significant interactions with SBP.
View this table:
TABLE 3. Effect of a 12-wk dietary intervention and n–3 fatty acids on diastolic (DBP) and systolic (SBP) blood pressure in the healthy subjects1
DISCUSSION
Comparison of the effects of MUFA and SFA on blood pressure
In the present study we found that a MUFA-rich diet, in contrast with an SFA-rich diet, reduced the DBP in healthy, normotensive subjects. Although the type of fat, rather than the amount of fat, in the diet may be more important in terms of determining health outcomes, it is noteworthy that a MUFA-rich diet causes a reduction in both SBP and DBP at a fat intake below 37% of energy, whereas the positive effect of MUFA on BP seems to be lost at a high total fat intake. Whether the slightly higher dietary fiber content and the lower cholesterol content may have contributed to the positive effect of MUFA on BP cannot be ruled out. The calculated dietary intakes of calcium, sodium, potassium, and alcohol did not differ significantly between the MUFA and SFA diet groups; this finding supports the suggestion that the difference in health outcomes was related to the quality of dietary fat. Body weight was stable in both diet groups.
In the Multiple Risk Factor Intervention Trial (MRFIT; 16, 17), the main findings in the multivariate analysis of the 6-y observational data were significant, independent, positive relations between dietary SFA and DBP and between dietary cholesterol and SBP and DBP and an inverse relation of the PUFA-SFA ratio to DBP. A BP-lowering effect of MUFA has been suggested in some epidemiologic studies among populations with a high intake of MUFA (7, 18-20). Although limited in number, intervention studies also suggest a BP-lowering effect when MUFA is substituted for SFA (21, 22). In a small group of hypertensive women, a diet rich in MUFA from olive oil showed beneficial effects on BP (23). However, other trials in normotensive subjects showed no evidence of a BP-reducing effect of MUFA (16). Interestingly, substituting dietary PUFA with MUFA lowered both systolic and diastolic BP in 16 type 2 diabetic subjects (24), whereas only a minor lowering in DBP could be detected in healthy subjects (25). A diet rich in olive oil lowered the SBP and DBP by 4–5 and 3 mm Hg, respectively, as compared with a carbohydrate-rich diet in normotensive type 2 diabetic subjects (26), whereas such an effect was not detected in a small group of insulin-treated type 2 diabetic subjects with microalbuminuria (27). Finally, a slight reduction in SFA intake along with a supplement of olive oil markedly lowered the daily dose of antihypertensive drugs needed by hypertensive subjects (28). We showed that the present change in the proportion of dietary fatty acids, ie, a decrease in SFAs and an increase in MUFAs, improves insulin sensitivity (12). Prolonged insulin resistance has been shown to be associated with structural alterations of arterial vascular smooth muscle, which may provide the anatomic substrate for the propagation of hypertension (29). Insulin is a growth factor that stimulates the synthesis of vascular smooth muscle cells and results in proliferative changes in the arteries. Elevated circulating concentrations of insulin are associated with activation of the sympathetic nervous system and increased peripheral resistance and sodium retention (10, 30). The beneficial effect on BP of MUFA, in contrast with SFA, may thus, at least in part, be mediated via an improvement in insulin sensitivity. Interestingly, the beneficial effect of the MUFA diet on insulin sensitivity was not seen when the absolute fat intake was high (>37% of energy) (12). In line with this, the effect of MUFA on BP faded away at a high fat intake (>37% of energy). Another possibility could be that the olive oil phenolics, which are powerful antioxidants, partially accounted for the BP-lowering effect (30).
Some of the discrepancies among studies investigating similar dietary changes may be due to differences in populations (eg, an effect of dietary factors may be easier to find in hypertensive than in normotensive individuals) or in methods of measuring BP. Thus, ambulatory BP monitoring with repeated measurements over 24 h much more accurately detects small changes in BP than do clinical BP measurements. Interestingly, we previously found that the diurnal BP was unaffected by a change in the quality of SFAs, ie, between the 2 most important SFAs, stearic and palmitic acids, in type 2 diabetic subjects (31). Mediterranean diets are associated with a reduced risk of CVD, and additional research on the effects of MUFA on BP is warranted to elucidate the potential of MUFA-rich diets to lower BP via diurnal BP measurements, sufficient population samples, and different populations.
n–3 Fatty acids and blood pressure
In the present study we found that the addition of 3.6 g n–3 fatty acids/d did not affect DBP or SBP in normotensive subjects, regardless of whether they were consuming a high-fat (>37% of energy) or a low-fat diet. In a meta-analysis (32), relatively high doses of n–3 PUFAs, typically >3 g/d, reduced BP but only in hypertensive subjects. Weighted pooled estimates of SBP and DBP changes (mm Hg) and 95% CIs were –1.0 (–2.0, 0.0) and –0.5 (–1.2, 0.2) in normotensive subjects and were –5.5 (–8.1, –2.9) and –3.5 (–5.0, –2.1) in the trials of untreated hypertensive subjects. The magnitude of BP reduction was greatest at a high BP but was not significantly associated with dose of n–3 fatty acids. In contrast, another meta-analysis showed a dose-response effect of fish oil on BP with an amount of n–3 fatty acids of >3.3 g/d needed to be associated with an effect on BP (33). However, the effect of n–3 fatty acids on BP occurred only in subjects with hypertension, hypercholesterolemia, and atherosclerosis and not in healthy normotensive subjects. In subjects with peripheral arterial disease, a recent Cochrane analysis (34) found that n–3 fatty acid supplementation reduced DBP but not SBP. Interestingly, the Lugalawa Study (35) found that a daily fish consumption of 300–600 g increased plasma n–3 fatty acids and decreased BP. The effect of n–3 fatty acids on BP in normotensive subjects is not convincing. Thus, in a 9-mo intervention study the addition of n–3 fatty acids lowered SBP and DBP in normotensive subjects (36). However, a 12-mo study of the addition of n–3 fatty acids in normotensive subjects found no effect on SBP or DBP (37).
Dietary fats may modulate BP through different mechanisms. Also, n–3 and n–6 PUFAs are converted to prostaglandins, which reduce BP by affecting arterial vasodilation, electrolyte balance, and renal release of renin or pressor hormones. The incorporation of unsaturated fatty acids into cell membranes increases membrane permeability, thus stimulating the transport of sodium and cations across the membrane. In summary, n–3 fatty acids seem to have a small dose-dependent, hypotensive effect, the extent of which seems to be dependent on the degree of hypertension. In view of the high dose required to lower BP, an increased intake of n–3 fatty acids has a limited role in the management of hypertension.
Conclusions
Changing the proportions of dietary fat by decreasing SFAs and increasing MUFAs decreased diastolic BP. Interestingly, the beneficial effect on BP induced by fat quality was negated by the consumption of a high total fat intake (>37% of energy). The addition of n–3 fatty acids did not alter the BP.
ACKNOWLEDGMENTS
BMR, BV, MU, GR, AAR, LT, and KH were the daily project leaders, were involved in the study scheme and data interpretation, and wrote the manuscript. LB was involved in the study scheme and data interpretation and provided significant advice. EP was involved in carrying out the study and provided significant advice. The KANWU Study Group consists of B Vessby, M Uusitupa, K Hermansen, G Riccardi, AA Rivellese, LC Tapsell, L Berglund, BM Rasmussen, and E Pedersen. None of the authors had a conflict of interest.
REFERENCES