| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research Communications |
1 From the Research Department of Human Nutrition and Centre for Advanced Food Studies, The Royal Veterinary and Agricultural University, Frederiksberg, Denmark, and the Department of Thrombosis Research, the University of Southern Denmark, and the Department of Clinical Biochemistry, Ribe County Hospital, Esbjerg.
2 Supported by the Danish Food Technology Research Programme (FØTEK 2) as part of the Danish Research Programme "Rapeseed Oil in Human Nutrition."
3 Address reprint requests to LF Larsen, Rolighedsvej 30, DK-1958 Frederiksberg C, Denmark. E-mail: LFL{at}KVL.DK.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Objective: We compared the effects of virgin olive oil with those of rapeseed and sunflower oils on blood coagulation factor VII (FVII), a key factor in thrombogenesis.
Design: In a randomized and strictly controlled crossover study, 18 healthy young men consumed diets enriched with 5 g/MJ (19% of total energy) olive oil, sunflower oil, or rapeseed oil for periods of 3 wk. On the final day of each period, participants consumed standardized high-fat meals (42% of energy as fat). Fasting and nonfasting blood samples were collected after each period.
Results: Mean (±SEM) nonfasting peak concentrations of activated FVII (FVIIa) were 11.3 ± 5.1 U/L lower after olive oil than after sunflower oil, an 18% reduction (P < 0.05). Olive oil also tended to cause lower FVIIa peak concentrations than did rapeseed oil (mean difference: 8.6 U/L, a 15% reduction; P = 0.09). There were no significant differences between diets with respect to nonfasting factor VII coagulant activity (FVII:c), prothrombin fragment 1+2 (F1+2), and tissue factor pathway inhibitor (TFPI) concentrations, or with respect to fasting plasma values of FVII protein, FVII:c, FVIIa, F1+2, or TFPI.
Conclusion: A background diet rich in olive oil may attenuate the acute procoagulant effects of fatty meals, which might contribute to the low incidence of IHD in Mediterranean areas.
Key Words: Monounsaturated fatty acids • olive oil • rapeseed oil • sunflower oil • postprandial values • blood coagulation factor VII • thrombogenesis • adults • humans
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
30% of all dietary energy came from olive oil (2). We hypothesized that olive oil could have beneficial effects on blood coagulation factor VII (FVII), a key protein in thrombosis and an IHD risk factor (3–7). Several dietary intervention studies have shown that FVII is indeed influenced by diet (8–11). We tested our hypothesis in a controlled dietary trial in which diets enriched with olive oil, sunflower oil, or rapeseed oil were compared. Sunflower oil has a high content of the n-6 polyunsaturated fatty acid (PUFA) linoleic acid. Similarly to olive oil, rapeseed oil has a high content of oleic acid, but it contains more linoleic acid than does olive oil. The n-3 PUFA 
-linolenic acid is also present in rapeseed oil. |
SUBJECTS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
: 24 y) participated in the study. Subjects had a weight of 62.2–98.6 kg (: 78.8 kg), a height of 1.72–1.99 m (: 1.82 m), and a body mass index (kg/m2) of 18.4–27.0 (: 22.9). All subjects were nonsmokers, had no history of IHD, and did not use medication regularly. Their mean fasting plasma lipid values were as follows: total cholesterol, 4.29 mmol/L (range: 2.89–5.96 mmol/L); HDL cholesterol, 1.22 mmol/L (range: 0.94–1.84 mmol/L); and triacylglycerols, 0.99 mmol/L (range: 0.41–1.67 mmol/L). The study protocol was carefully explained to the participants before they entered the study and all participants gave their written consent. The study was performed in accordance with Helsinki Declaration II and was approved by the Ethical Committee of Copenhagen and Frederiksberg.
Study design
In a double-blind crossover study, the participants were randomly assigned to consume olive oil (A), sunflower-oil (B), and rapeseed-oil (C) diets for periods of 21 d in 1 of 6 possible orders (ABC, ACB, BCA, BAC, CAB, and CBA). The randomization was balanced, ie, 3 participants were randomly assigned to each of the 6 possible diet combinations. The intervention periods were separated by washout periods lasting between 3 and 12 wk, during which time the participants resumed their habitual diets. On day 22 of each intervention period, postprandial responses to consumption of a standard fat load were studied. With standard fat loads, differences in the postprandial responses will originate only from differences in the experimental diets and not from differences in the fat loads or from interactions between the fat loads and the background diets.
During the intervention, the participants could eat and drink only the experimental diets, except for coffee (without sugar and cream), water, salt, and artificial sweeteners. Before the study, participants completed a 4-d weighed food record, which, in combination with body weights, heights, and physical activity levels, was used for planning individual energy intakes during the intervention. Body weights were recorded, with subjects in light clothing, 3 times/wk before lunch (nonfasting) during the intervention periods. If body weight changed by ±1 kg, energy intake was adjusted.
Experimental diets
The 3 experimental diets were identical except for their content of 5 g/MJ of either virgin olive oil (Navarino; Danton Trading, Århus, Denmark), chemically refined sunflower oil (Solex W; Århus Olie, Århus, Denmark), or rapeseed oil physically refined in a pilot plant (Department of Biotechnology, Technical University of Denmark, Lyngby). The oils constituted 19% of the total dietary energy intake and were incorporated into breads, cakes, dressings, and sauces. The experimental diet was composed of ordinary Danish food.
On weekdays, the participants consumed lunch under supervision in a special dining room at the Research Department of Human Nutrition. A ready-to-eat dinner, snack, and breakfast for the next morning were prepackaged to be eaten at home. On Fridays, subjects were provided with food and beverages for the weekend. Breakfast always consisted of bread, butter, cheese, and marmalade. Menus for lunch, dinner, and snacks were repeated every 7 d. Lunch typically consisted of bread with cheese, tuna, ham, or another kind of meat, and a salad with dressing. Dinner typically consisted of a hot meal of lean beef or chicken and mashed potatoes, pasta, or rice. Between- and after-meal snacks consisted of cake, sweets, yogurt, juice, or fruit. In addition, the participants were supplied with 0.5-MJ "hunger buns," which provided the same percentage of energy as fat and composition as did the experimental diet. After a few adjustments to energy intakes, mainly at the beginning of the first intervention period, the mean daily energy intake during the study was 15.4 MJ (range: 13–18 MJ). There were no statistically significant changes in body weight during the study. The participants were asked to record whether they felt sick, took any medication, consumed any hunger buns, or deviated from the prescribed diet or their usual behavior. Participants were told to bring any leftovers back to the institute, where they were recorded.
All experimental meals were prepared in our metabolic kitchen and coded with colored labels. Most of the food for each intervention period was prepared in one batch in our metabolic kitchen, weighed out in portions, and frozen until used. Duplicate portions (10 MJ) were collected and analyzed. The analyzed macronutrient composition of the 3 diets was virtually identical (Table 1
). The mean macronutrient composition of the habitual diet (baseline), assessed by a 4-d weighed food record, is also presented in Table 1. The fatty acid composition of the 3 test oils is presented in Table 2.
|
|
Blood sampling
To characterize the participants and to determine baseline values, fasting (12 h) blood samples were collected before the study began. On days 21 and 22 of each intervention period, fasting blood samples were collected in the morning. On day 22, nonfasting blood samples were collected at 1315 (4.25 h after the first meal), 1445 (5.75 h after the first meal), and 1745 (8.75 h after the first meal).
Blood was collected with minimal stasis by venipuncture after subjects rested 
10 min in a supine position. Factor VII (FVII) and tissue factor pathway inhibitor (TFPI) were measured in blood collected into tubes containing sodium citrate (final concentration: 0.0129 mol/L) at room temperature. Lipids were measured in blood collected into tubes containing EDTA-K3 (0.004 mmol/L). Blood sampling tubes without additives were used for analysis of C-reactive protein (CRP), an acute phase protein. Blood for prothrombin fragment 1+2 (F1+2) analysis was collected into 3-mL precooled tubes containing EDTA-K3, to which 6 µL D-phenylalanyl-1-propyl-1-arginine chloromethylketon (2.63 g/L) was added immediately after sampling to prevent further generation of thrombin. All tubes were centrifuged at 3000 x g for 15 min—the citrated tubes and the tubes without additives at room temperature and the tubes containing EDTA-K3 at 4°C. Within 1 h after blood sampling, platelet-poor plasma and serum were separated and frozen immediately at -50°C and then kept at -80°C until analyzed within 12 mo.
Blood analyses
All samples from each subject were rapidly thawed in a water bath at 37°C and analyzed in randomized order in one run, except for plasma triacylglycerols, which were analyzed during the study. FVII and F1+2 were measured only in samples collected during fasting and 5.75 and 8.75 h after the first meal. Methods for analyses of activated FVII (FVIIa) and FVII coagulant activity (FVII:c) were described previously (12). Briefly, FVII:c was analyzed with a one-stage clotting assay using human placenta tissue thromboplastin factor (Thromborel S; Behringwerke AG, Marburg, Germany) and human FVII-deficient plasma (Biopool, Umeå, Sweden) on an ACL 100 (Automated Coagulation Laboratory, Instrumentation Laboratory, Milan, Italy). FVIIa was analyzed with a clotting assay by using recombinant truncated tissue thromboplastin factor specific for FVIIa cofactor function (STACLOT VIIa-rTF; Diagnostica Stago, Asniers Sur Seine, France) on a coagulometer (type 410A 4B; Schnittger Gross, Amelung, Germany). Concentrations of FVII protein (FVII:Ag) were determined with an enzyme immunoassay (Diagnostica Stago). Concentrations of F1+2 were measured with a commercial enzyme-linked immunosorbent assay (Behringwerke AG). Plasma TFPI concentrations in samples collected during fasting and 5.75 h after the first meal were measured with an enzyme-linked immunosorbent assay as described previously (13). This assay specifically measures plasma TFPI not bound to lipoproteins. Serum CRP was measured with an immunoturbidimetric method (Roche, Basel, Switzerland). Plasma triacylglycerol, total cholesterol, and HDL cholesterol concentrations were determined with enzymatic methods on a Cobas Mira S analyzer (Boehringer Mannheim GmbH, Mannheim, Germany). Technical problems for one subject, elevated CRP values in one subject, and the inability of one subject to consume one of the fat loads necessitated the exclusion of a few observations.
Dietary compliance
To monitor dietary compliance, the fatty acid composition of plasma triacylglycerols and cholesterol esters in fasting blood samples was determined. Plasma triacylglycerols and cholesterol esters were separated by thin-layer chromatography and the fatty acid composition was determined by gas-liquid chromatography after methylation, as described previously (14).
Statistics
It was calculated that with 18 subjects, the study had 80% power to show a 10% difference in FVII:c and FVIIa at a significance level of 5%. All variables followed a Gaussian distribution. Fasting plasma concentrations after each diet were calculated as the mean of the fasting samples collected on days 21 and 22. Fasting concentrations were compared by analysis of variance (ANOVA). For analysis of the postprandial responses on day 22, summary statistics [nonfasting peak, nonfasting mean, and postprandial increases (difference between fasting and nonfasting peak values)] were used (15). The summary measures were compared by ANOVA. When the ANOVA showed statistically significant effects of diet, the 3 diets were compared pairwise with a Student's t test and Bonferroni corrected. Changes from fasting concentrations were analyzed with a Student's t test. We tested for carryover and period effects, but none were observed. The level of significance was set at P < 0.05. The SAS statistical package (version 6.08; SAS Institute Inc, Cary, NC) was used for all statistical analyses.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). -Linolenic acid concentrations were significantly elevated by rapeseed oil.
|
). Fasting triacylglycerol concentrations were 25% higher after the olive oil diet than after the sunflower-oil and rapeseed-oil diets. Fasting FVII:c and FVIIa with all 3 diets decreased significantly from baseline, 10% and 20%, respectively, as did the ratio of FVIIa to FVII:Ag. There were minor (NS) decreases from baseline in fasting FVII:Ag, F1+2, and TFPI. Fasting plasma triacylglycerol concentrations increased with the olive oil diet.
|
, respectively). Additionally, plasma TFPI concentrations 5.75 h after the first meal were 7% higher than fasting concentrations with all 3 diets. Nonfasting mean FVII:c was also slightly higher (3%) than fasting FVII:c with all 3 diets (P = 0.06, P = 0.07, and P = 0.05 for the olive oil, sunflower-oil, and rapeseed-oil diets, respectively). Plasma F1+2 concentrations did not change significantly postprandially.
|
|
). Nonfasting mean and peak concentrations of FVIIa also tended to be lower after the olive oil diet than after the rapeseed-oil diet (paired t test: P = 0.09). Nonfasting TFPI, FVII:c, and F1+2 did not differ significantly between the 3 diets. Nonfasting mean and peak concentrations of plasma triacylglycerols were significantly higher after the olive oil diet than after the sunflower-oil diet, but these differences were mainly due to differences in fasting concentrations (Figure 1
).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Several earlier studies compared the immediate effects of meals containing different edible fats and found no clear differences on postprandial FVII activation (12,16–20). The effect of the fatty acid composition of the background diet on nonfasting FVII was also investigated previously (21–26). In a recent study with a design comparable with ours, Roche et al (21) found that a background diet enriched with olive oil (18% of energy as MUFAs) caused similar fasting FVIIa concentrations, but less postprandial FVII activation than did a diet enriched with butter (17% of energy as SFAs). In agreement with our results, they concluded that this observation indicates the antithrombotic potential of olive oil diets, which may contribute to the lower rate of IHD observed in Mediterranean regions. A report by Mitropoulos et al (22) suggested less postprandial FVII activation with a background diet rich in n-6 PUFAs (37% of energy) than with a diet rich in SFAs (22% of energy) (22). Taken together, background diets rich in olive oil thus seem to be less prothrombotic than diets rich in n-6 PUFAs, which are less prothrombotic than SFA diets.
Two other studies reported contradictory findings however (23, 24). Miller et al (23) found that background diets rich in either n-6 PUFAs or SFAs had similar effects on postprandial FVII:c concentrations in 9 subjects. Sanders et al (24) reported that a background diet rich in olive oil was associated with a postprandial increase in FVII:c after a fat load, whereas no increase in FVII:c was observed after a background diet rich in SFAs. Note that background diets were fed for only 1 wk in the study by Miller et al (23). Furthermore, in the study by Sanders et al, diets were consumed in sequential order, which means that a period effect cannot be excluded (24). Note also that both studies assessed postprandial FVII activation from measurements of FVII:c, which is relatively insensitive to FVIIa, as also evidenced by our present observation of a weak postprandial increase in FVII:c despite an increase of up to 76% in FVIIa.
Finally, the divergent findings may have been because of the different characteristics of the experimental diets. Miller et al (23) used a diet with 14% of energy as SFAs, whereas Mitropoulos et al (22) used a diet with a much higher percentage of energy as SFAs (22% of energy). Similarly, Miller et al's diet had a much lower n-6 PUFA content (21% of energy) than did the diet used by Mitropoulos al (37% of energy). The fatty acid composition of the experimental diets used by Roche et al (21) and Sanders et al (24) were similar, however, and it is not likely that their divergent observations can be explained by differences in the fatty acid composition of their diets (21, 24). Fish-oil supplementation of the background diet did not affect postprandial FVII:c or FVIIa concentrations significantly in 2 recent studies (25, 26).
The attenuated postprandial FVII activation after the olive oil diet may have resulted from the MUFA oleic acid, which is present in a relatively high amount in olive oil (74 mol%). As does olive oil, rapeseed oil contains a relatively high amount of oleic acid (56 mol%). However, compared with the sunflower-oil diet, no attenuated postprandial FVII activation was observed after the rapeseed-oil diet. Additionally, we observed no significant difference in postprandial FVII activation between the olive oil and rapeseed-oil diets; however, a trend was seen, possibly because of the higher content of PUFAs in rapeseed oil than in olive oil. Another possibility might be that the postprandial activation of FVII is influenced by differences in minor non-fatty acid constituents of the oils.
The hypertriacylglycerolemia that occurs after the consumption of high-fat meals (23, 27, 28) has been suggested to activate FVII. In the present study, fasting and nonfasting plasma triacylglycerol concentrations were higher after the olive oil diet than after the sunflower-oil and rapeseed-oil diets. Thus, our present findings contradict the assumption that the plasma triacylglycerol concentration is the primary determinant of postprandial FVIIa concentrations and are in line with our earlier observations (12, 25).
The postprandial activation of FVII was accompanied by a moderate increase in TFPI concentrations with all 3 diets, whereas the plasma concentration of F1+2, a marker of thrombin generation, was unaffected. There was no significant difference between diets with respect to postprandial TFPI and F1+2. A lack of effect of dietary FVII activation on F1+2 was observed by others (28, 29). Our observations indicate that it may have been due to enhanced inhibition of FVIIa by TFPI. Two other studies, however, showed no postprandial effects of fat loads on the activity of TFPI in healthy subjects and in hypertriacylglycerolmic patients (30, 31). An alternative explanation for the lack of effect of FVII activation on F1+2 is that our healthy young subjects had limited vascular expression of tissue thromboplastin factor, which would prevent FVIIa from causing F1+2 formation. In line with this thinking, we would expect that individuals with atherosclerotic vessels and augmented tissue thromboplastin factor expression (32–34) would more readily react to FVII activation with an increased plasma concentration of F1+2. The findings of an epidemiologic study by Miller et al (35) support this explanation. Another possible explanation for the lack of effect of FVII activation on F1+2 is that F1+2 was efficiently cleared from the blood in our healthy young study subjects.
Fasting concentrations of FVII were not affected differently by the 3 diets enriched with different types of fat, which agrees with earlier observations (11). The decline in fasting FVII:c and FVIIa from baseline that was seen with all 3 experimental diets suggests that the experimental diets were less thrombogenic than were the habitual diets. The macronutrient composition of the habitual and experimental diets differed only slightly, however. Therefore, we believe that seasonal variations, the fixed diet patterns during the experimental periods, or both are more likely explanations of the changes from baseline in fasting FVII:c and FVIIa.
In conclusion, the study showed that incorporation of olive oil into the diet for 3 wk resulted in a lower postprandial FVIIa concentration than did the incorporation of sunflower oil. This finding suggests that a diet rich in olive oil reduces the thrombotic propensity associated with the consumption of fatty meals, which could partly explain the low incidence of IHD in populations with a high intake of olive oil.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. Ruano, J. Lopez-Miranda, R. de la Torre, J. Delgado-Lista, J. Fernandez, J. Caballero, M. I. Covas, Y. Jimenez, P. Perez-Martinez, C. Marin, et al. Intake of phenol-rich virgin olive oil improves the postprandial prothrombotic profile in hypercholesterolemic patients Am. J. Clinical Nutrition, August 1, 2007; 86(2): 341 - 346. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Weinbrenner, M. Fito, R. d. l. Torre, G. T. Saez, P. Rijken, C. Tormos, S. Coolen, M. F. Albaladejo, S. Abanades, H. Schroder, et al. Olive Oils High in Phenolic Compounds Modulate Oxidative/Antioxidative Status in Men J. Nutr., September 1, 2004; 134(9): 2314 - 2321. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Sagara, T. Kanda, M. NJelekera, T. Teramoto, L. Armitage, N. Birt, C. Birt, and Y. Yamori Effects of Dietary Intake of Soy Protein and Isoflavones on Cardiovascular Disease Risk Factors in High Risk, Middle-Aged Men in Scotland J. Am. Coll. Nutr., February 1, 2004; 23(1): 85 - 91. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Muller, A. S. Lindman, A. Blomfeldt, I. Seljeflot, and J. I. Pedersen A Diet Rich in Coconut Oil Reduces Diurnal Postprandial Variations in Circulating Tissue Plasminogen Activator Antigen and Fasting Lipoprotein (a) Compared with a Diet Rich in Unsaturated Fat in Women J. Nutr., November 1, 2003; 133(11): 3422 - 3427. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Rajaram The effect of vegetarian diet, plant foods, and phytochemicals on hemostasis and thrombosis Am. J. Clinical Nutrition, September 1, 2003; 78(3): 552S - 558. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Ros Dietary cis-monounsaturated fatty acids and metabolic control in type 2 diabetes Am. J. Clinical Nutrition, September 1, 2003; 78(3): 617S - 625. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. E. Finnegan, D. Howarth, A. M. Minihane, S. Kew, G. J. Miller, P. C. Calder, and C. M. Williams Plant and Marine Derived (n-3) Polyunsaturated Fatty Acids Do Not Affect Blood Coagulation and Fibrinolytic Factors in Moderately Hyperlipidemic Humans J. Nutr., July 1, 2003; 133(7): 2210 - 2213. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E Fernandez-Jarne, E Martinez-Losa, M Prado-Santamaria, C Brugarolas-Brufau, M Serrano-Martinez, and M. Martinez-Gonzalez Risk of first non-fatal myocardial infarction negatively associated with olive oil consumption: a case-control study in Spain Int. J. Epidemiol., April 1, 2002; 31(2): 474 - 480. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Pedersen, M. W. Baumstark, P. Marckmann, H. Gylling, and B. Sandström An olive oil-rich diet results in higher concentrations of LDL cholesterol and a higher number of LDL subfraction particles than rapeseed oil and sunflower oil diets J. Lipid Res., December 1, 2000; 41(12): 1901 - 1911. [Abstract] [Full Text]
|
|
|
|
|
|
F. Visioli and C. Galli Olive oil: more than just oleic acid Am. J. Clinical Nutrition, September 1, 2000; 72(3): 853 - 853. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
; n = 18), or rapeseed oil (
; n = 16) were consumed for 3 wk before consumption of the fatty meals. Responses after the sunflower-oil diet were significantly different from those after the olive oil diet, P < 0.05 (see Table 5