Effects of Dietary Fat Quality and Quantity on Postprandial Activation of Blood Coagulation Factor VII
Abstract Acute elevation of the coagulant activity of blood coagulation factor VII (FVIIc) is observed after consumption of high-fat meals. This elevation is caused by an increase in the concentration of activated FVII (FVIIa). In a randomized crossover study, we investigated whether saturated, monounsaturated, or polyunsaturated fats differed regarding postprandial activation of FVII. Eighteen healthy young men participated in the study. On 6 separate days each participant consumed two meals (times, 0 and 13/4 hours) enriched with 70 g (15 and 55 g) of either rapeseed oil, olive oil, sunflower oil, palm oil, or butter (42% of energy from fat) or isoenergetic low-fat meals (6% of energy from fat). Fasting and series of nonfasting blood samples (the last at time 81/2 hours) were collected. Plasma triglycerides, FVIIc, FVIIa, and free fatty acids were analyzed. There were marked effects of the fat quantity on postprandial responses of plasma triglycerides, FVII, and free fatty acids. The high-fat meals caused, in contrast to the low-fat meals, considerable increases in plasma triglycerides. Plasma levels of FVIIc and FVIIa peaks were 7% and 60% higher after consumption of high-fat meals than after consumption of low-fat meals. The five different fat qualities caused similar postprandial increases in plasma triglycerides, FVIIc, and FVIIa. These findings indicate that high-fat meals may be prothrombotic, irrespective of their fatty acid composition. The postprandial FVII activation was not associated with the plasma triglyceride or free fatty acid responses.
- Received July 3, 1997.
- Accepted July 31, 1997.
Thrombus formation plays an essential role in manifestation of the ischemic heart syndrome, and results from the Northwick Park Heart study and the PROCAM study have indicated that FVIIc is independently associated with the risk of ischemic heart death.1,2 Humans exist mostly in the nonfasting state, and nonfasting blood levels of risk markers may be particular important for the risk of developing ischemic heart disease.3 Acutely elevated FVIIc may be observed in the hours after consumption of high-fat meals, and the elevation seems to be dose dependent.4–8 The postprandial elevation of FVIIc is caused by an increase in the level of FVIIa and not by an increase in the plasma protein concentration of FVII.4,6–12
Only a few earlier studies have compared the acute effects of different dietary fat qualities on FVII.5,8,13–18 In general, their statistical power was low, and only three of them included fats rich in MUFAs.8,15,18 In the present study we tested whether consumption of dietary fats high in MUFAs, ie, rapeseed oil and olive oil, causes less postprandial FVII activation than other fats rich in either saturated or n-6 polyunsaturated fatty acids. This might be suggested because the incidence of ischemic heart disease is low in Mediterranean populations where olive oil is the predominant dietary fat. Our study included comparison of high- and low-fat meals and, finally, we investigated whether postprandial FVII levels are associated with plasma triglyceride and/or FFA responses to high-fat meals.
Eighteen healthy men, mean age 27 (range, 22 to 33) participated in the study. None of the participants were smokers or were taking any medication, including acetylsalicylic acid, during the study. Their mean body mass index was 22.3 kg/m2 (range, 19.4 to 24.8 kg/m2), mean fasting plasma triglycerides 0.91 mmol/L (range, 0.35 to 1.99 mmol/L), mean total cholesterol 4.11 mmol/L (range 3.04 to 6.12 mmol/L) and mean HDL-cholesterol 1.21 mmol/L (range, 0.76 to 2.07 mmol/L). The study was performed in accordance with the Second Helsinki Declaration and was approved by the Ethical Committee of Copenhagen and Frederiksberg (journal No. 01- 272/95).
The study was a randomized double-blinded crossover study. Each individual participated in 6 different meal tests, with at least 3 weeks between each test. At each test, they consumed two test meals enriched with either rapeseed oil, olive oil, sunflower oil, palm oil, or butter. Isoenergetic low-fat meals were used as control. Because of a very different food composition it was not possible to blind researchers and participants to the fat content of the test foods. All other elements of the study were blinded. We used two separate meals in each test to achieve reasonably marked postprandial triglyceride responses. In our experience, this may be difficult to achieve when single meals are fed to fasting young men. The participants were told not to change their dietary habits nor their levels of physical activity during the study, which started in October 1995 and ended in May 1996. To minimize the preexperimental variation, the participants were told not to drink alcohol or to perform any heavy physical activity 48 hours before each meal test. The evening before they consumed a low-fat pasta meal (total energy: 4760 kJ, 21% of energy from fat) supplied by us. A fasting blood sample (after 10 hours of fasting) was taken at 8:30 am on the test day. The two meals were consumed under supervision at 09:00 am (time 0 hours) and at 10:45 am (time 13/4 hours). The first meal had to be consumed within 10 minutes, the second within 20 minutes. Nonfasting blood samples were collected 8 times during the day (times, 11/2, 21/2, 31/2, 41/2 51/2, 61/2, 71/2, and 81/2 hours). The participants were allowed to drink tap water but could not leave the institute or perform any heavy physical activity between blood samplings.
As described, low-fat test meals (6% of energy from fat) and high-fat test meals (42% of energy from fat) enriched with different types of fat were served in randomized order. The high-fat meals were low-fat rice dishes (rice, beef, onion, red pepper, corn, small amounts of spices, and bread) to which 15 g (first meal) and 55 g (second meal) test fat was added. When butter was added we took into account that it contains water (17 wt/wt%). The meals were heated briefly in a microwave oven before serving. The low-fat test meals were also based on low-fat rice dishes, but bread, bananas, and raisins were added in replacement for the test fats. The low-fat test meals were isoenergetic with the high-fat test meals. All meals were prepared in one batch in a metabolic kitchen, and all ingredients were precisely weighed out. Duplicate portions of the high-fat (without test-fat added) and the low-fat test meals were chemically analyzed. The calculated and analyzed nutrient contents of the test meals are presented in Table 1⇓. The analyzed fatty acid compositions of the applied test fats are presented in Table 2⇓. The analyzed energy contents of the two meals were 1600 and 5800 kJ, respectively.
Blood samples were taken with minimal stasis by venipuncture after 10 minutes of supine rest. The first 3 mL blood was collected in tubes containing EDTA-K3 (final concentration, 0.004 mol/L) and used for determination of total plasma triglycerides, total cholesterol, HDL-cholesterol, and plasma FFAs. Blood sampling tubes without additives were used for analysis of C-reactive protein (CRP), an acute phase protein. Analysis of FVII was performed on blood collected in tubes containing sodium citrate (final concentration, 0.0129 mol/L) at room temperature to avoid cold activation of FVII. All the tubes were centrifuged at 3000g for 15 minutes. The tubes without additives and the citrated tubes were spun at room temperature, the EDTA tubes at 4°C. Within 1 hour after blood sampling, platelet-poor plasma was separated and samples were snap-frozen at −50°C and then kept at −80°C until they were analyzed within less than 12 months.
All samples from each subject were analyzed in randomized order in one run. Plasma triglycerides were analyzed in all samples; plasma FVII (FVIIc and FVIIa) and FFAs were determined only in blood samples collected during fasting and at times 31/2, 51/2, and 81/2 hours. CRP, total cholesterol, and HDL-cholesterol were measured in fasting samples only.
CRP was measured by ELISA with antibodies from DAKO (Glostrup, Denmark). Plasma triglycerides, total cholesterol, and HDL-cholesterol were determined by commercial enzymatic methods on a Cobas Mira S (Boehringer Mannheim GmbH, Mannheim, Germany). Plasma FFAs were determined with a commercial enzymatic colorimetric method on a Cobas Mira S (Wako Chemicals GmbH, Germany). Plasma coagulant activity of FVII (FVIIc) was measured in an one-stage clotting assay using human placenta thromboplastin (Thromborel S, Behringwerke AG, Marburg, Germany). The analyses were performed on an ACL 100 (Automated Coagulation Laboratory, Instrumentation Laboratory, Italy). The 40-μL diluted test sample (diluted 1+9 in Tris HCl buffer: 50 mmol/L Tris, 100 mmol/L NaCl, pH 7.4) reacted with 40 μL FVII deficient plasma (Biopool, Umeå, Sweden). Then 80 μL of human placenta thromboplastin, which contains CaCl2 (17 mmol/L), was added and the clotting time was registered. Results were expressed in percentage by relating the clotting time to a standard calibration curve obtained from lyophilized normal human plasma (Biopool, Umeå, Sweden) calibrated against an international calibrator. For the FVIIc analysis it was not possible to analyze all samples from each person in one run, but they were analyzed within two successive runs. Determination of FVIIa was performed with STACLOT® VIIa-rTF (Diagnostica Stago, France), a clotting assay with recombinant truncated tissue factor (rTF) specific for FVIIa cofactor function. The analyses were performed on a coagulometer (Type 410A 4B, Schnittger Gross, Amelung, Germany). The FVIIa levels (U/L) of the test samples were deduced from a standard curve given by a lyophilized preparation of human recombinant FVIIa.
The within-run imprecisions (CV%) of the different determinations were triglycerides, 2.0%; FFAs, 2.1%; FVIIc, 1.4%; and FVIIa, 3.0%. The between run-imprecisions (CV%) were triglycerides, 2.3%; FFAs, 2.9%; FVIIc, 5.1%; and FVIIa, 9.0%.
The simple, but statistically valid, method of summary statistics19 was applied for analysis of the repeated measurements. ANOVA with the fasting value as covariate was applied to two summary measures: postprandial peak and postprandial mean values. These summary measures were found to give the best description of the postprandial responses. For assessment of correlations Pearson’s correlation coefficient (r) was calculated. Only a single value from each participant was used in the correlation analyses. Values derived from the palm oil test were selected because the postprandial plasma triglyceride response tended to be larger after palm oil.
The level of statistical significance was set at P<.05. The SAS statistical package (SAS Institute Inc, NC, USA) was used for all the statistical analyses.
One participant had raised CRP (16 mg/L) on one occasion, and his results from this particular test day were excluded. All other subjects had CRP serum concentrations <5 mg/L, indicating that they were without an acute phase response.
Fasting and postprandial values for plasma triglyceride, FFA, FVIIc, and FVIIa responses after consumption of test meals enriched with rapeseed oil, olive oil, sunflower oil, palm oil, and butter and isoenergetic low-fat meals are presented in Tables 3⇓ and 4⇓. There were no statistically significant differences between the fasting samples. The five different test fats caused similar and statistically significant increases in plasma triglycerides, FVIIc and FVIIa (P<.0001). In contrast, there was a marked decrease in FFAs. At time 31/2 hours, the plasma level of FFAs after consumption of butter was significantly higher than after consumption of rapeseed oil– and olive oil–enriched test meals (P=.025 and P=.006, respectively) (Table 3⇓).
Because the five fat qualities did not result in statistically significant different postprandial plasma triglyceride and FVII responses and because the FFA differences were very small, results from all high-fat meal tests were pooled. Fig 1⇓ illustrates the mean postprandial responses of plasma triglycerides, FFAs, FVIIc, and FVIIa after consumption of the high-fat meals (n=18x5) and the low-fat meals (n=18). The postprandial profiles of plasma triglycerides, FFAs, FVIIc, and FVIIa differed markedly after consumption of the high-fat meals and the low-fat meals (P<.0001). Consumption of the low-fat meals did not result in significantly different nonfasting plasma triglyceride and FVIIa levels compared with the fasting values. After consumption of the high-fat meals, plasma triglycerides increased from a mean fasting concentration of 0.93 mmol/L (95% CI, 0.85 to 1.01 mmol/L) to a peak concentration of 1.96 mmol/L (95% CI, 1.78 to 2.14 mmol/L) (note that the mean peak value may differ from the peak of the average response curves [Fig 1⇓] because the individuals peaked at different time points). Plasma FVIIa increased from 48.4 U/L (95% CI, 45.3 to 51.5 U/L) to 81.4 U/L (95% CI, 76.5 to 86.3 U/L). Plasma FVIIc increased significantly from a mean fasting value of 0.81 IU (95% CI, 0.75 to 0.87 IU) to 0.84 IU (95% CI, 0.81 to 0.87 IU) after consumption of the high-fat meals and decreased to 0.72 IU (95% CI, 0.66 to 0.78 IU) after consumption of the low-fat meals (P<.0001). Earlier studies have shown that the protein concentration of FVII is decreasing postprandially, irrespective of the fat content of meals,5,6 and this is probably the reason for the decrease in FVIIc after consumption of the low-fat meals.
Plasma FFAs decreased initially from a fasting mean value of 0.51 mmol/L (CI: 0.42 to 0.60 mmol/L) to 0.02 mmol/L (CI: 0.01 to 0.03 mmol/L) after consumption of the low-fat meals and to 0.19 mmol/L (CI: 0.18 to 0.21 mmol/L) after consumption of the high-fat meals. At time 81/2 hours, the level of FFAs after consumption of the low-fat and all the high-fat meals were identical: 0.25 mmol/L (CI: 0.15 to 0.35 mmol/L).
As illustrated in Fig 2⇓, there were considerable within- and between-subject variations in nonfasting plasma triglycerides and FVIIa peak concentrations. Correlation analyses did not show any statistically significant associations between nonfasting plasma peak concentrations of triglycerides and FVIIa (r2<.06, P < .37). This can also be deduced from Fig 2⇓. Both absolute values, absolute changes from baseline, and percentage changes from baseline were applied in the analyses of correlations. Subsequently, we investigated whether the postprandial triglyceride and FVIIa responses were correlated after consumption of the other four applied test fats. This was not the case. We also did not observe a correlation between fasting plasma triglycerides and FVIIa (r2=.04, P=.93). As expected, correlations between fasting plasma triglycerides and nonfasting triglyceride peak concentrations (r2=.62, P=.0001) and between fasting FVIIa and nonfasting FVIIa peak concentrations (r2=.48, P=.01) were observed.
To assess possible relations between the plasma lipolytic activity and FVII activation, we used two different estimates for the lipolytic activity. The first estimate was the maximum triglyceride clearance rate (the steepest slope of the downward plasma triglyceride curve) (Fig 1⇑). The second was the FFA concentration at time of maximum plasma insulin concentration (time 31/2 hours, results not shown). At this time the release of FFAs from the adipose tissue is believed to be negligible, and accordingly plasma FFAs are primarily originating from lipolysis of triglyceride-rich lipoproteins (TRLP).20 No correlations between our two estimates of the lipolytic activity and FVII activation were observed (r2<.12, P>.17). In addition, the two estimates for the lipolytic activity were not mutually correlated (r2=.05, P=.65).
The fat quantity had a marked impact on the postprandial responses of plasma triglycerides, FFAs and FVII (Fig 1⇑). Plasma FVIIc and FVIIa peaks were approximately 7% and 60% higher after consumption of high-fat meals than after consumption of low-fat meals, indicating an immediate prothrombotic effect of the high-fat meals (Table 4⇑).
The five different test fats that were used had similar postprandial effects on plasma triglycerides, FFAs, FVIIc, and FVIIa. We cannot therefore reject our hypothesis about MUFAs and acute FVII activation. The low incidence of ischemic heart disease in the Mediterranean countries is not, based on our experiments, explained by an acute, favorable postprandial effect of MUFAs on FVII.
How dietary fat promotes the acute activation of FVII is not clear, but a relation to the plasma triglyceride concentration10,11,13,22 or to lipolytic degradation of TRLP have been suggested.21 We did not find any statistically significant associations between plasma triglycerides and postprandial FVII activation. As shown in Fig 2⇑, individuals with high postprandial plasma triglycerides may have low postprandial FVII activation, and vice versa. In addition, we did not observe correlations between estimates of TRLP lipolysis and postprandial FVII activation. However, our estimates may not be the best markers of lipoprotein lipase activity, and, therefore, the present results do not exclude an effect of lipolysis on the acute FVII activation. Earlier studies reached conflicting conclusions regarding associations between plasma triglycerides and postprandial FVII activation4–6,10–13,22. This may partly be explained by differences in study populations (eg, young, healthy, or hyperlipemia), but also by application of different methods for analysis of the FVII activation, ie, specific or nonspecific FVIIa assays or because triglycerides were measured in different lipoprotein fractions. Our study population was homogeneous and the absence of any relation between postprandial plasma triglyceride concentration and FVII levels was so clear cut that we find it justified to conclude that postprandial triglycerides do not predict the extent of postprandial FVII activation in healthy young individuals.
Only few earlier studies compared the acute effects of different fats on FVII, and of these only three studies included monounsaturated fats. In a study of 12 females, Freese and Mutanen found no differences between rapeseed oil, sunflower oil, and butter oil (1 g/kg body weight) with regard to postprandial FVIIc and serum triglyceride responses.15 In another study of 4 subjects, olive oil (90 g) was shown to cause postprandial increases in plasma triglycerides and FVII, whereas no increases were observed with a similar amount of medium-chain triglycerides (C8:0+C10:0).8 In a recent study, Roche and Gibney18 did not find any influence in altering the saturated fatty acid:MUFA ratio of test meals on the magnitude of the postprandial FVIIc response, a low MUFA content did however, prolong the increase in FVIIc. Miller and coworkers13 found no difference between saturated and n-6 polyunsaturated fats in their study of 9 participants. In a study of 10 subjects, Salomaa et al5 found the same effect on postprandial lipemia and FVIIc of saturated and n-6 polyunsaturated fatty acids (1 g/kg body weight). Mitropoulos et al14 have reported an effect of fat quality on postprandial FVII in a study of 5 subjects. They found postprandial FVIIc to be higher after saturated fats than after n-6 polyunsaturated fats. A dietary change of 4 weeks duration preceded the meal tests of this study, however. It is possible that their findings rather reflect differences in the background diet than different acute effects of the meal tests. Finally, Tholstrup and coworkers16 found no significant differences between fats rich in stearic acid (C18:0) and myristic acid (C14:0) regarding postprandial effects on FVII.
In conclusion, the present study confirmed that the consumption of high-fat meals, in contrast to low-fat meals, cause an acute FVII activation, which indicates that high-fat meals may be prothrombotic. The small differences in the fiber content of the high- and low-fat meals could also have contributed to their distinct effects. However, this possibility does not seem likely.4 Our results also demonstrate that fats rich in monounsaturated fatty acids do not differ from fats rich in polyunsaturated or saturated fatty acids with regard to acute effects on plasma triglycerides and FVII, when consumed in realistic amounts as part of natural meals. Finally, the postprandial activation of FVII was not associated with plasma triglyceride or FFA responses. Further studies are needed to evaluate the mechanism(s) linking high-fat meals with FVII activation. In addition, the long-term effects of different edible fats, including monounsaturated fats, still needs to be investigated.
Selected Abbreviations and Acronyms
|FFA||=||free fatty acid|
|FVII||=||blood coagulation factor VII|
|FVIIc||=||FVII coagulant activity|
|MUFA||=||monounsaturated fatty acid|
|rTF||=||recombinant truncated tissue factor|
We gratefully acknowledge technicians Klara Jørgensen and Bente Hansen and dietitian Lena T. Andersen for their assistance with this study. Thanks are due Joan Bentzen for her assistance with the FVII analyses and technicians Kirsten Nielsen and Bente Pedersen from the Department of Biochemistry and Nutrition at the Technical University of Denmark for analyses of the fatty acid composition of the test fats and the fat content of the test meals. The study is part of the research program “Rapeseed Oil in Human Nutrition” financed by The Danish Food Technology Research Program (FØTEK 2).
Heinrich J, Balleisen L, Schulte H, Assmann G, van de Loo J. Fibrinogen and factor VII in the prediction of coronary risk. Results from the PROCAM study in healthy men. Arterioscler Thromb. 1994;14:54–59.
Zilversmit DB. Atherogenesis: a postprandial phenomenon. Circulation. 1979;60:473–485.
Bladbjerg EM, Marckmann P, Sandström B, Jespersen J. Non-fasting factor VII coagulant activity (FVII:C) increased by high-fat diet. Thromb Haemost. 1994;71(6):755–758.
Miller GJ, Martin JC, Mitropoulos KA, Esnouf MP, Cooper JA, Morrisey JH, Howarth DJ, Tuddenham EGD. Activation of factor VII during alimentary lipemia occurs in healthy adults and patient with congenital factor XII or factor XI deficiency, but not in patients with factor IX deficiency. Blood. 1996;87(10):4187–4196.
Sanders TAB, Miller GJ, de Grassi T, Yahia N. Postprandial activation of coagulant factor VII by long chain dietary fatty acids. Thromb Haemost. 1996;76(3):369–371.
Silveira A, Karpe F, Blombäck M, Steiner G, Walldius G, Hamsten A. Activation of coagulation factor VII during alimentary lipemia. Arterioscler Thromb. 1994a;14:60–69.
Silveira A, Green F, Karpe F, Blombäck M, Humphries S, Hamsten A. Elevated levels of factor VII activity in the postprandial state: effect of the factor VII Arg-Gln polymorphism. Thromb Haemost. 1994b;72(5):734–739.
Kapur R, Hoffman CJ, Bhushan V, Hultin MB. Postprandial elevation of activated factor VII in young adults. Arterioscler Thromb Vasc Biol. 1996;16:1327–1332.
Mitropoulos KA, Miller GJ, Martin JC, Reeves BE, Cooper J. Dietary fat induces changes in factor VII coagulant activity through effects on plasma free steric acid concentration. Arterioscler Thromb. 1994;14:214–222.
Tholstrup T, Andreasen K, Sandström B. Acute effects of high-fat meals rich in either stearic or myristic acid on haemostatic factors in healthy young men. Am J Clin Nutr. 1996;64:168–176.
Matthews JNS, Altman DG, Campbell MJ, Royson P. Analysis of serial measurements in medical research. BMJ. 1990;300:230–235.
Fielding BA, Frayn KN, Halliday D, Bannister PA, Callow J, Venkatesan S. Rapid entry of dietary fatty acids into plasma non-esterified fatty acid pool. Proc Nutr Soc. 1996;55(2):162A.Abstract.
Silveira A, Karpe F, Johnsson H, Bauer KA, Hamsten A. In vivo demonstration in humans that large postprandial triglyceride-rich lipoproteins activate coagulation factor VII through the intrinsic coagulation pathway. Arterioscler Thromb Vasc Biol. 1996;16:1333–1339.