Trans monounsaturated fatty acids and saturated fatty acids have ...

7 downloads 0 Views 95KB Size Report
fatty acids have similar effects on postprandial flow-mediated vasodilation. NM de Roos1, E Siebelink1, ML Bots2, A van Tol3, EG Schouten1 and MB Katan1,4*.
European Journal of Clinical Nutrition (2002) 56, 674–679 ß 2002 Nature Publishing Group All rights reserved 0954–3007/02 $25.00 www.nature.com/ejcn

ORIGINAL COMMUNICATION

Trans monounsaturated fatty acids and saturated fatty acids have similar effects on postprandial flow-mediated vasodilation NM de Roos1, E Siebelink1, ML Bots2, A van Tol3, EG Schouten1 and MB Katan1,4* 1 Division of Human Nutrition and Epidemiology, Wageningen University, Wageningen, The Netherlands; 2Julius Center for Patient Oriented Research, University Medical Center, Utrecht, The Netherlands; 3Department of Biochemistry, Cardiovascular Research Institute COEUR, Erasmus University Rotterdam, Rotterdam, The Netherlands; and 4The Wageningen Center for Food Sciences, Wageningen, The Netherlands

Objective: Several studies suggest that a fatty meal impairs flow-mediated vasodilation (FMD), a measure of endothelial function. We tested whether the impairment was greater for trans fats than for saturated fats. We did this because we previously showed that replacement of saturated fats by trans fats in a controlled diet decreased FMD after 4 weeks. Design: We fed 21 healthy men two different test meals with 0.9 – 1.0 g fat=kg body weight in random order: one rich in saturated fatty acids (Sat), mainly from palm kernel fat, and one rich in trans fatty acids (Trans) from partially hydrogenated soy bean oil. The study was performed in our metabolic ward. We had complete data for both diets of 21 men. Results: FMD increased from a fasting value of 2.3  2.0% of the baseline diameter to 3.0  1.7% after the Sat test meal (95% CI for change 7 0.33, 1.70) and from 2.7  2.3 to 3.1  2.0% after the Trans test meal (95% CI for change 7 0.57, 1.29). The increase after the Sat meal was 0.22 ( 7 1.18 – 1.61) FMD% higher than after the Trans meal. Serum triacylglycerols increased by 0.46  0.36 mmol=l after the Sat test meal and by 0.68  0.59 mmol=l after the Trans test meal; a difference of 0.23 (0.07, 0.39) mmol=l. Serum HDL-cholesterol was hardly affected by the test meals. The activity of serum paraoxonase, an esterase bound to HDL, increased slightly after the two test meals but the difference between meals was not significant. Conclusion: FMD was not impaired and not different after test meals with saturated or trans fatty acids. Thus, differences in long-term effects of these fats are not caused by differences in acute effects on the vascular wall. European Journal of Clinical Nutrition (2002) 56, 674 – 679. doi:10.1038=sj.ejcn.1601377 Keywords: endothelium-dependent vasodilation; diet, saturated fat; trans unsaturated fat; human

Introduction Consumption of a diet rich in trans fatty acids, found in fried fast foods, bakery goods, potato chips, french fries and shortenings, is associated with increased risk of coronary heart disease (CHD; Ascherio & Willett, 1997). Pooled data from prospective follow-up studies suggest that each 2% increase in energy intake from trans fatty acids is associated with a 25% increase in risk (Oomen et al, 2001). This increase *Correspondence: MB Katan, Division of Human Nutrition and Epidemiology, Wageningen University, Bomenweg 2, 6703 HD Wageningen, The Netherlands. Guarantor: NM de Roos. Contributors: E Siebelink, ML Bots, A van Tol, EG Schouten and MB Katan. Received 4 July 2001; revised 24 October 2001; accepted 29 October 2001

in risk is higher than that for a 2% increase in energy intake from saturated fats (Ascherio & Willett, 1997), although their effect on low-density lipoprotein (LDL)-cholesterol is similar. The difference in risk could be caused by effects on high density lipoprotien (HDL)-cholesterol, which is decreased when saturated fats are replaced by trans fats (Zock & Katan, 1992; Zock et al, 1995). We used metabolic studies with flow-mediated vasodilation as primary endpoint to study the causal role of HDL-cholesterol in CHD. Flowmediated vasodilation (FMD) is the percentage increase in arterial diameter after provoked increases in blood flow; impaired dilation appears to be an early stage of atherosclerosis (Celermajer et al, 1994). Moreover, an impaired FMD has been related to increased risk of recurrence of cardiovascular events in 73 patients with chest pain (Neunteufl et al, 2000).

Postprandial flow-mediated vasodilation NM de Roos et al

675 In the first study done at our laboratory we showed that replacement of  9% of energy (en%) from saturated fat by trans fat decreased serum HDL-cholesterol by 0.39 mmol=l and impaired FMD (De Roos et al, 2001a). This result was in line with the observation that trans fatty acids are more strongly related to cardiovascular disease than saturated fatty acids, and suggested that this was caused by the decrease in HDL-cholesterol. We verified the effect of a decrease in HDL-cholesterol in a subsequent study with a low-fat vs a high-oil diet, but this time FMD was not impaired by the decrease in HDL-cholesterol (De Roos et al, 2001b). This result appeared to weaken the evidence for a causal role of HDL-cholesterol but we may have missed an effect, because the decrease in HDL-cholesterol was half of that observed in our first study. We performed the present study to investigate whether differences in effect of trans and saturated fatty acids were independent of HDL-cholesterol. We postulated that the increased risk with consumption of trans fatty acids might be due to acute effects of circulating trans fatty acids in serum. Indeed, it has been shown by others that consumption of fat-enriched meals impair FMD (Table 1), but we found no data on effects of trans fatty acids. Therefore we compared the effects on FMD of a test meal with trans fatty acids with those of a test meal with saturated fatty acids. We also measured the activity of the HDL-associated enzyme paraoxonase, an esterase related to cardiovascular disease risk. We had found earlier that replacement of saturated fatty acids by trans fatty acids reduced the activity of this enzyme (unpublished results). A high serum paraoxonase activity prevents oxidation of lipids in LDL in vitro (Durrington et al, 2001); if this occurs in vivo paraoxonase might improve vascular function (Cox & Cohen, 1996). We hypothesized that the activity of serum paraoxonase would parallel changes in FMD.

Subjects and methods The study was approved by the Medical Ethics Committee of Wageningen University. All subjects signed an informed consent form. Table 1

Subjects We recruited 40 non-smoking men aged  35 y by advertising in local papers and by inviting subjects who had taken part in previous studies. We excluded nine men because of diabetes, thrombosis, or use of cholesterol-lowering drugs. All 31 remaining subjects had normal concentrations of urinary glucose and protein ( < 0.3 g=l). Five men were excluded because of high serum cholesterol ( > 8 mmol=l) and=or fasting triacylglycerols ( > 3 mmol=l), because we wanted to exclude men with an impaired lipid metabolism. Of the remaining 26 subjects 25 completed the study. Pre-study mean (  s.d.) serum cholesterol of these 25 men was 5.8  0.8 mmol=l, triacylglycerols 1.2  0.7 mmol=l, body weight 83.5  10.0 kg and body mass index (BMI) 25.4  2.6 kg=m2. Their habitual diet, as assessed by a food frequency questionnaire that was validated for energy and fat intake (Feunekes et al, 1993), contained 11 MJ=day, (2620 kcal=day) with 40.4% of energy (%en) from fat (saturated fat 14.9en%). Habitual daily intake of cholesterol was 274 mg=day. Four additional subjects had to be excluded from the analyses because of missing data, so data of 21 subjects are presented in the results.

Study design The two test meals were served in random order on separate days within 1 week; this was repeated in reverse order the next week to obtain duplicate measurements. Thus, we had a total of four test days, separated by at least 1 day. A test day was preceded by a controlled evening meal and an overnight fast. The evening meal was prepared by us and packaged for consumption at home. Subjects consumed on average 3047 kJ with this evening meal, with 20.5en% from fat (9.5en% saturated fat), 22.9en% from protein and 56.5en% from carbohydrates. They were asked to refrain from sports activities on the day preceeding the test day and in the morning before the measurements. In the morning, after an overnight fast, subjects came to our laboratory between 7:00 and 9:30 am; the appointment

Postprandial effects (3 – 4 h after a meal) of oral fat loads on flow-mediated vasodilation in men (m) and women (w)

Subjects

Test meals

Change from fasting

Reference

16 m 5 m, 5 w

Two isocaloric meals: low-fat (5 g fat) vs high-oleic acid (50 g fat) Two isocaloric meals: low-fat (0 g fat) vs high-fat (50 g fat, 14 g saturated). Vitamin C  90 mg higher in low-fat meal Two isocaloric high-fat fast-food meals (rich in saturated and trans fatty acids) One high-fat meal (65 g=m2 body surface area) (saturated fat) One fat tolerance test (1480 kcal, 80 g saturated fat) Four isocaloric meals: low-fat (0 g fat) and high-fat (50 g fat, 14 g saturated) with or without vitamins C and E Three meals (milkshakes): one control (low-fat) and two with 46 g extra fat (saturated), either unused or used for deep-fat frying Two isocaloric meals (1030 kcal, 61 g fat): saturated fatty acids vs monounsaturated fatty acids Five isocaloric meals (900 kcal, 50 g fat): olive oil (with or without vitamins C and E or vegetables), canola oil, salmon Three isocaloric meals high in fat (mainly saturated fat)

Decrease Decrease

Ong et al (1999) Vogel et al (1997)

Decrease Decrease Decrease Decrease, but not with added vitamins

Djousse´ et al (1999) Marchesi et al (2000) Anderson et al (2001) Plotnick et al (1997)

No decrease after unused fat; decrease after used fat No effect

Williams et al (1999) Raitakari et al (2000)

Only a decrease after olive oil

Vogel et al (2000)

Small increase

Katz et al, (2001)

13 10 m 7 w, 5 m 7 m, 13 w 10 m 12 10 m 25 m, 25 w

European Journal of Clinical Nutrition

Postprandial flow-mediated vasodilation NM de Roos et al

676 time was kept constant for each subject throughout the study. Some subjects came by bike, others by car; so we had everyone rest for at least 10 min before the start of the measurements. After subjects had rested, we measured FMD and took blood samples. Then, subjects were given the test meal which had to be consumed within 15 min. Postprandial measurements were done 3 h after the end of the test meal. Subjects were not allowed food or drink between test meal and postprandial measurements, and were asked to refrain from strenuous physical activities.

Test meals The two test meals differed only in source of fat. The fat used in the saturated fat test meal was derived mainly from palm kernel fat, and contained mainly lauric acid (C12:0) and myristic acid (C14:0); the fat used in the trans fat test meal was mainly partially hydrogenated soy bean oil rich in trans monoenes with 18 carbon atoms (Table 2). The test fats were offered in a breakfast that consisted of a milkshake (33 g fat=average portion), bread plus spread (45 g fat=average portion), and preserves. The fat dose that was given to the subjects was proportional to their body weight: 61 g (body weight < 69 kg), 70 g (70 – 79 kg), 78 g (80 – 89 kg), and 87 g ( < 90 kg); and ranged from 0.9 to 1.0 g=kg bodyweight. A cup of coffee or tea (125 ml) was given with the meals to prevent caffeine-withdrawal headache. Four subjects who wanted to take part in the study but who did not like milkshakes were fed all of the fat in a milk-based porridge. An average portion of a milkshake-breakfast provided 4946 kJ (1178 kcal), and an average portion of porridge 4103 kJ (Table 3). The type of fat was the only difference between the test meals.

Flow-mediated vasodilation of the brachial artery We measured FMD of the brachial artery, a non-invasive technique using ultrasound, as set up by Celermajer et al (1996). The ultrasound images were made with a linear array

Table 2 Fatty acid composition (g=100 g fatty acids) of the fat rich in saturated fatty acids and the fat rich in trans fatty acids

transducer (range 8 – 14 MHz) of an Esaote AU5 scanner (Pie Medical Benelux B.V., Maastricht, The Netherlands). All images were stored on super-VHS videotapes for off-line analysis. We inflated a cuff around the forearm to a pressure of 240 mmHg to induce ischemia for 5 min. Changes in brachial artery diameter at the site of the antecubital crease were recorded for 5 min after cuff release. The maximum diameter that was reached within 5 min was used to calculate the percentage FMD from the baseline diameter. Other details can be found elsewhere (De Roos et al, 2001a). The images were analyzed with special software developed by the Wallenberg Institute of Cardiovascular Research and the Chalmers University of Technology (Gothenburg, Sweden; Liang et al, 2000) at the Vascular Imaging Center of the University Medical Center in Utrecht. Of the 200 measurements (25 subjects measured eight times each) 31 (16% of total) had insufficient quality due to movement of the arm or in one case because we had recorded a vein instead of an artery. We eliminated these measurements from the analysis and only present data of 21 men for whom we had measurements for both test meals. To calculate within-subject variability we used the four pre-meal measurements of each subject’s resting diameter, maximum diameter and FMD. At a mean resting diameter of 4.763 mm, the s.d. within-subject was 0.311 mm and the corresponding CV 6.5%. The maximum diameter was 4.883 mm, s.d. within-subjects 0.330 mm and the corresponding CV 6.8%. Pre-meal FMD was 2.58% of the baseline diameter; the s.d. within-subject was 2.18 percentage points and the corresponding CV was 84% of the mean FMD. The within-subject variability of the FMD was larger than the 50 – 60% in our previous studies, but this was due to a smaller FMD, not to a larger s.d.

Measurement of serum triacylglycerols and lipoproteins Blood was collected in evacuated collection tubes (Venoject II, Terumo, Leuven, Belgium) from an antecubital vein and allowed to clot for 30 – 45 min at room temperature. Serum was obtained by centrifugation at 1187 g for 10 min at 4 C.

Table 3 Macronutrient composition and vitamin content of average portions of the two test meals

Fat rich in saturated fat Fat rich in trans fat a Fatty acid composition (%) C8:0 caprylic acid C10:0 capric acid C12:0 lauric acid C14:0 myristic acid C16:0 palmitic acid C18:0 stearic acid C18:1 cis b C18:1 trans C18:2 cis,cis linoleic acid a

1.9 2.2 39.6 16.3 13.3 4.2 12.0 0.4 8.6

Only those with > 1% of mass in one of the test fats. b Mainly (n-10), (n-9) and (n-11).

European Journal of Clinical Nutrition

0.0 0.0 0.0 0.1 9.5 16.8 25.5 33.8 11.9

a

Energy (kJ) Protein (g) Fat (g)b Carbohydrate (g) Mono- and disaccharides (g) Polysaccharides (g) Fibre (g) a

Milkshake þ bread (21 subjects)

Porridge (4 subjects)

4947 (1178 kcal) 16 79 102 59 43 7

4103 (977 kcal) 14 82 49 23 26 3

The average portion of porridge was smaller because subjects who consumed the porridge were lighter. b 1% dairy fat and 99% test fats (rich in either saturated or trans fat).

Postprandial flow-mediated vasodilation NM de Roos et al

677 To eliminate inter-assay variation, serum was stored at 7 75 C and all analyses were done in one assay at the end of the study. Total cholesterol and triacylglycerols (Cholesterol FlexTM and Triglycerides FlexTM reagent cartridge, Dade Behring, Newark, USA) and HDL cholesterol (liquid N-geneousTM HDL-C assay, Instruchemie BV, Hilversum, The Netherlands) were measured in one run, and LDL-cholesterol in fasting samples was calculated with the Friedewald formula. The analytical within-run coefficient of variation for all analyses was < 1.5%.

Measurement of serum paraoxonase activity Serum paraoxonase activity was measured in the serum samples of the first study week, so for each subject we had one pre-meal (fasting) and one postprandial sample per type of fat. PON1 activity was determined using paraoxon as a substrate at pH 8.0 (van der Gaag et al, 1999). Serum paraoxonase activity was expressed in U=l, which is equal to 1 mmol of hydrolyzed paraoxon=l=min.

Statistics We calculated for each person the mean of the two measurements of resting and maximum diameter and FMD per type of fat. We used SAS for statistical tests and calculations (SAS Institute Inc., Cary, NC, USA) and GraphPad Prism for graphing the data. The number of subjects (n ¼ 25) was calculated to detect a difference between changes of 1.8 percentage points with a power of 80% and a probability of 5%.

of the meal rich in saturated fat but not after the meal rich in trans fat; with a slight difference of 7 0.03 mmol=l between meals (95% CI, 7 0.06 to 0.008). The increase in serum total cholesterol after the meals was marginal (Table 4). The results for the brachial artery measurements are based on 21 subjects for whom we had paired fasting and postprandial measurements for both meals. Consumption of the test meals had little effect on the baseline diameter of the brachial artery, ie the diameter before occlusion; differences between fasting and postprandial diameters and between the two types of fat were small (Table 4). The maximum diameter that was reached after cuff release was larger after the meals than before, and the effect was similar for both types of fat. Thus, the absolute increase in mm was larger postprandially for both types of fat, as was the percentage FMD. The type of fat did not affect any of the changes in brachial artery dimensions. Changes in serum triacylglycerols were correlated with changes in FMD after the Sat meal (Pearson r 7 0.45, 95% CI 7 0.73 to 7 0.03), but not after the trans meal (Pearson r 7 0.06, 95% CI 7 0.45 – 0.35, Figure 1). Serum paraoxonase activity showed a large variability between subjects, with a range of 50.2 – 340.5 U=l. The variability within-subjects was small: at a mean fasting serum paraoxonase activity of 130.1 U=l the s.d. within-subject was 5.9 U=l, corresponding with a CV of 4.5% (25 subjects, two

Results Serum triacylglycerol concentrations were significantly increased from baseline at 3.5 h after consumption of the test meals (Table 4), with a 0.23 mmol=l stronger increase after trans fat than after saturated fat (95% confidence interval (CI) for difference 0.07 – 0.39). The concentration of serum HDL-cholesterol increased slightly after consumption

Figure 1 Correlation between postprandial changes in triacylglycerols and flow-mediated vasodilation (FMD) in 21 subjects after test meals rich in trans or saturated fatty acids.

Table 4 Serum lipoproteins, lipids, and paraoxonase activity for all 25 subjects, and brachial artery dimensions for 21 subjects for whom we had paired data on both test meals, before and after consumption of the test meals Testfmeal rich in trans fat

Total cholesterol (mmol=l) HDL-cholesterol (mmol=l) Triacylglycerols (mmol=l) Baseline diameter (mm) Maximum diameter (mm) Increase (mm) FMD (% of baseline) Paraoxonase activity (U=l)

Fasting

Postprandial

5.54 1.33 1.23 4.760 4.864 0.104 2.27 131.1

5.67 1.35 1.69 4.763 4.896 0.133 2.87 133.2

Difference (95% CI) 0.13 0.02 0.46 0.003 0.032 0.029 0.60 2.1

(0.08, 0.17) (0.01, 0.04) (0.31, 0.61) ( 7 0.14, 0.15) ( 7 0.10, 0.17) ( 7 0.02, 0.07) ( 7 0.45, 1.65) ( 7 0.8, 5.0)

Test meal rich in trans fat Fasting

Postprandial

Difference (95% CI)

5.60 1.34 1.14 4.795 4.919 0.125 2.62 129.2

5.68 1.33 1.82 4.830 4.976 0.145 3.00 132.5

0.08 ( 7 0.02, 0.18) 7 0.01 ( 7 0.04, 0.01) 0.68 (0.44, 0.92) 0.035 ( 7 0.11, 0.18) 0.056 ( 7 0.11, 0.22) 0.021 ( 7 0.03, 0.07) 0.39 ( 7 0.67, 1.45) 3.3 (0.3, 6.4)

95% CI for difference trans sat 7 0.03 7 0.03 0.23 0.032 0.025 7 0.008 7 0.217 1.2

( 7 0.15, 0.08) ( 7 0.06, 7 0.01) (0.07, 0.39) ( 7 0.14, 0.21) ( 7 0.17, 0.22) ( 7 0.08, 0.06) ( 7 1.62, 1.18) ( 7 3.1, 5.6)

a

3 h after the subjects finished their meal we measured flow-mediated vasodilation (FMD) and, after that, serum concentrations of lipoproteins and lipids.

European Journal of Clinical Nutrition

Postprandial flow-mediated vasodilation NM de Roos et al

678 fasting samples per subject). Serum paraoxonase activity was slightly higher 3.5 h after the meals than fasting, with similar effects for both types of fat (Table 4).

Discussion We found no difference in effect on postprandial FMD in 21 healthy men between high doses of trans fatty acids and saturated fatty acids. Thus, our previous observation in a separate study that replacement of saturated fatty acids by trans fatty acids impaired FMD after 4 weeks could apparently not be explained by differences in acute effects of the fatty acids.

Postprandial effects of fats on flow-mediated vasodilation FMD after the test meals was slightly higher than fasting. Although the confidence intervals for the changes are wide, they suggest improvements rather than reductions of FMD after the meals. Although previous studies yielded conflicting results, most of them showed marked reductions in FMD after an oral fat load (Ong et al, 1999; Vogel et al, 1997; Wilmink et al, 2000). Studies that showed an impairment of FMD by an oral fat load used similar doses of fat (0.9 – 1.0 g=kg bodyweight) and study designs (measurements fasting and 3 – 4 h postprandially) to ours (Table 1). We mixed the test fats with a milkshake, as was done by others (Sutherland et al, 1999; Vogel et al, 1997; Marchesi et al, 2000). The test meals were consumed under our supervision, and therefore non-compliance could not explain the unexpected outcome of the study. A difference with most other studies, however, is that our subjects were allowed to take either a 125 mL cup of coffee or tea (one choice throughout the whole study) with the test meals. If the corresponding dose of caffeine (60 mg for coffee and 40 mg for tea; Passmore & Eastwood, 1986) had strong vasodilatory effects, up to 3 h after consumption, this might have counteracted the effect of the test fats. However, we found no evidence for an effect of caffeine in the literature (Plotnick et al, 1997; Djousse´ et al, 1999). Thus, the fact that we did not find an impairment in FMD after a meal rich in either saturated or trans fat cannot be explained by differences in dose or study design. FMD was not impaired by the test fats in our study, but was already quite low in our subjects: the fasting values were 2.6% of the resting diameter. This value is about half of that observed in our previous studies, but those studies were done in younger subjects, and FMD appears to decline with age (Lyons et al, 1997). Ong et al found a similar value for lean, young, healthy men (Ong et al, 1999), but most investigators reported higher values. Part of the higher values can be explained by the position of the cuff; because higher values for FMD are found when the cuff is placed around the upper arm than around the lower arm, the position we used (Berry et al, 2000). European Journal of Clinical Nutrition

Correlation between changes in serum triacylglycerols and flow-mediated vasodilation In agreement with results of others (Marchesi et al, 2000; Vogel et al, 2000) we found that changes in serum triacylglycerols showed a negative correlation with changes in FMD, although statistically significant for saturated fatty acids only. This result appears to contradict our finding that FMD was slightly higher 3 h after the meals, when serum triacylglycerols were higher, than before the meals. Indeed, from Figure 1 it can be estimated that a zero change in serum triacylglycerols corresponded to an increase in FMD. This would suggest that FMD improves during the morning, because the first measurement was before 9:30 am and the second was 3 h later. Such a diurnal variation has been reported by others, with a change in FMD from 4.0% at 8:00 am to 5.3% at 12:00 noon in healthy young men (Etsuda et al, 1999), although a different diurnal variation was found in women (Ringqvist et al, 2000). Maybe the increase in serum triacylglycerols in our study was too small to counteract such a diurnal variation in FMD; the increase 3.5 h after the meals was moderate, with 0.46 mmol=l after the sat-meal and 0.68 mmol=l after the trans meal. However, many studies that did show an effect on FMD reported similar increases in serum triacylglycerols within the same time frame (Marchesi et al, 2000; Djousse´ et al, 1999; Plotnick et al, 1997; Williams et al, 1999).

Long-term vs postprandial effects of trans fatty acids and saturated fatty acids We performed the present study to investigate whether trans fatty acids had acute effects on FMD besides the long-term adverse effects we had seen in our previous study (De Roos et al, 2001a). Postprandial effects, however, were not different for saturated fatty acids and trans fatty acids. This is in agreement with the similar effects of trans fatty acids and saturated fatty acids on postprandial concentrations of serum triacylglycerols (Tholstrup et al, 2001; Sanders et al, 2000), chylomicron triacylglycerols, serum HDL-cholesterol, the activity of cholesteryl ester transfer protein (Tholstrup et al, 2001), and Factor VII coagulant activity and activated Factor VII (Sanders et al, 2000). Moreover, we showed that postprandial changes in serum paraoxonase activity were not different between the two fats despite a 6% difference in fasting values after 4 weeks consumption of meals enriched with either of the two fats. Our hypothesis was that changes in paraoxonase activity would parallel changes in FMD. This proved to be correct, but the changes in paraoxonase activity were probably too small to have caused the changes in FMD. Thus, postprandial differences between trans fatty acids and saturated fatty acids appear to be small or absent. In conclusion, we found no impairment of FMD after test meals rich in either saturated or trans fatty acids. It is more likely that the impairment of FMD after 4 weeks consumption of a diet rich in trans fatty acids was mediated by a decrease in HDL-cholesterol.

Postprandial flow-mediated vasodilation NM de Roos et al

Acknowledgements We wish to thank the volunteers who took part in this study; Loders Croklaan (Vlaardingen, The Netherlands) for donation of palm kernel fat; Professor Ronald Mensink and Dr Pierre Demacker for their advice on the study design; Vera Fierkens and Trudy Jansen for ultrasound measurements; and Irna Hertel for preparation of the test meals. This study was funded by the Dutch Dairy Foundation in Nutrition and Health.

References Anderson RA, Evans ML, Ellis GR, Graham J, Morris K, Jackson SK, Lewis MJ, Rees A & Frenneaux MP (2001): The relationships between post-prandial lipaemia, endothelial function and oxidative stress in healthy individuals and patients with type 2 diabetes. Atherosclerosis 154, 475 – 485. Ascherio A & Willett WC (1997): Health effects of trans fatty acids. Am. J. Clin. Nutr. 66(Suppl), 1006S – 1010S. Berry KL, Skyrme-Jones AP & Meredith IT (2000): Occlusion cuff position is an important determinant of the time course and magnitude of human brachial artery flow-mediated dilation. Clin. Sci. 99, 261 – 267. Celermajer DS, Sorensen KE, Bull C, Robinson J & Deanfield JE (1994): Endothelium-dependent dilation in the systemic arteries of asymptomatic subjects relates to coronary risk factors and their interaction. J. Am. Coll. Cardiol. 24, 1468 – 1474. Celermajer DS, Adams MR, Clarkson P, Robinson J, McCredie R, Donald A & Deanfield JE (1996): Passive smoking and impaired endothelium-dependent arterial dilatation in healthy young adults. New Engl. J. Med. 334, 150 – 154. Cox DA & Cohen MI (1996): Effects of oxidized low-density lipoprotein on vascular contraction and relaxation: clinical and pharmacological implications in atherosclerosis. Pharmac. Rev. 48, 3 – 19. De Roos NM, Bots ML & Katan MB (2001a): Replacement of dietary saturated fatty acids by trans fatty acids lowers serum HDL cholesterol and impairs endothelial function in healthy men and women. Arterioscler. Thromb. Vasc. Biol. 21, 1233 – 1237. De Roos NM, Bots ML, Siebelink E, Schouten E & Katan MB (2001b): Flow-mediated vasodilation is not impaired when HDL-cholesterol is lowered by substituting carbohydrates for monounsaturated fat. Br. J. Nutr. 86, 181 – 188. Djousse´ L, Ellison RC, McLennan CE, Cupples LA, Lipinska I, Tofler GH, Gokce N & Vita JA (1999): Acute effects of a high-fat meal with and without red wine on endothelial function in healthy subjects. Am. J. Cardiol. 84, 660 – 664. Durrington PN, Mackness B & Mackness MI (2001): Paraoxonase and atherosclerosis. Arterioscler. Thromb. Vasc. Biol. 21, 473 – 480. Etsuda H, Takase B, Uehata A, Kusano H, Hamabe A, Kuhara R, Akima T, Matsushima Y, Arakawa K, Satomura K, Kurita A & Ohsuzu F (1999): Morning attenuation of endothelium-dependent, flowmediated dilation in healthy young men: possible connection to morning peak of cardiac events? Clin. Cardiol. 22, 417 – 421. Feunekes GI, Van Staveren WA, De Vries JH, Burema J & Hautvast JG (1993): Relative and biomarker-based validity of a food-frequency questionnaire estimating intake of fats and cholesterol. Am. J. Clin. Nutr. 58, 489 – 496. Katz DL, Nawaz H, Boukhalil J, Giannamore V, Chan W, Ahmadi R & Sarrel PM (2001): Acute effects of oats and vitamin E on endothelial responses to ingested fat. Am. J. Prev. Med. 20, 124 – 129. Liang Q, Wendelhag I, Wikstrand J & Gustavsson T (2000): A multiscale dynamic programming procedure for boundary detection in ultrasonic artery images. IEEE Trans. Med. Imag. 19, 127 – 142.

Lyons D, Roy S, Patel M, Benjamin N & Swift CG (1997): Impaired nitric oxide-mediated vasodilatation and total body nitric oxide production in healthy old age. Clin. Sci. (Colch.) 93, 519 – 525. Marchesi S, Lupattelli G, Schillaci G, Pirro M, Siepi D, Roscini AR, Pasqualini L & Mannarino E (2000): Impaired flow-mediated vasoactivity during post-prandial phase in young healthy men. Atherosclerosis 153, 397 – 402. Neunteufl T, Heher S, Katzenschlager R, Wolfl G, Kostner K, Maurer G & Weidinger F (2000): Late prognostic value of flow-mediated dilation in the brachial artery of patients with chest pain. Am. J. Cardiol. 86, 207 – 210. Ong PJ, Dean TS, Hayward CS, Della MP, Sanders TA & Collins P (1999): Effect of fat and carbohydrate consumption on endothelial function. Lancet 354, 2134. Oomen CM, Ocke MC, Feskens EJ, van Erp-Baart MA, Kok FJ & Kromhout D (2001): Association between trans fatty acid intake and 10-year risk of coronary heart disease in the Zutphen Elderly Study: a prospective population-based study. Lancet 357, 746 – 751. Passmore R & Eastwood MA (1986) Beverages; herbs and spices. In Human Nutrition and Dietetics, ed. LSP Davidson & R Passmore, pp 218 – 225. Edinburgh: Churchill Livingstone. Plotnick GD, Corretti MC & Vogel RA (1997): Effect of antioxidant vitamins on the transient impairment of endothelium-dependent brachial artery vasoactivity following a single high-fat meal. J.A.M.A. 278, 1682 – 1686. Raitakari OT, Lai N, Griffiths K, McCredie R, Sullivan D & Celermajer DS (2000): Enhanced peripheral vasodilation in humans after a fatty meal. J. Am. Coll. Cardiol. 36, 417 – 422. Ringqvist A, Caidahl K, Petersson AS & Wennmalm A (2000): Diurnal variation of flow-mediated vasodilation in healthy premenopausal women. Am. J. Physiol. Heart Cirul. Physiol. 279, H2720 – H2725 Sanders TA, de Grassi T, Miller GJ & Morrissey JH (2000): Influence of fatty acid chain length and cis=trans isomerization on postprandial lipemia and factor VII in healthy subjects (postprandial lipids and factor VII). Atherosclerosis 149, 413 – 420. Sutherland WH, Walker RJ, de Jong SA, van Rij AM, Phillips V & Walker HL (1999): Reduced postprandial serum paraoxonase activity after a meal rich in used cooking fat. Arterioscler. Thromb. Vasc. Biol. 19, 1340 – 1347. Tholstrup T, Sandstrom B, Bysted A & Holmer G (2001): Effect of 6 dietary fatty acids on the postprandial lipid profile, plasma fatty acids, lipoprotein lipase, and cholesterol ester transfer activities in healthy young men. Am. J. Clin. Nutr. 73, 198 – 208. van der Gaag MS, van Tol A, Scheek LM, James RW, Urgert R, Schaafsma G & Hendriks HF (1999): Daily moderate alcohol consumption increases serum paraoxonase activity; a diet-controlled, randomised intervention study in middle-aged men. Atherosclerosis 147, 405 – 410. Vogel RA, Corretti MC & Plotnick GD (1997): Effect of a single highfat meal on endothelial function in healthy subjects. Am. J. Cardiol. 79, 350 – 354. Vogel RA, Corretti MC & Plotnick GD (2000): The postprandial effect of components of the Mediterranean diet on endothelial function. J. Am. Coll. Cardiol. 36, 1455 – 1460. Williams MJA, Sutherland WHF, McCormick MP, de Jong SA, Walker RJ & Wilkins GT (1999): Impaired endothelial function following a meal rich in used cooking fat. J. Am. Coll. Cardiol. 33, 1050 – 1055. Wilmink HW, Stroes ES, Erkelens WD, Gerritsen WB, Wever R, Banga JD & Rabelink TJ (2000): Influence of folic acid on postprandial endothelial dysfunction. Arterioscler. Thromb. Vasc. Biol. 20, 185 – 188. Zock PL & Katan MB (1992): Hydrogenation alternatives: effects of trans fatty acids and stearic acid versus linoleic acid on serum lipids and lipoproteins in humans. J Lipid Rese. 33, 399 – 410. Zock PL, Katan MB & Mensink RP (1995): Dietary trans fatty acids and lipoprotein cholesterol. Am. J. Clin. Nutr. 61, 617.

679

European Journal of Clinical Nutrition