|
|
||||||||
Original Research Communications |
1 From the Oregon Health Sciences University, Division of Endocrinology, Diabetes, and Clinical Nutrition, Portland, and Oregon State University, Department of Nutrition and Food Management, Corvallis, OR.
2 Supported by US Department of Agriculture grant 93-37200-8963; the General Clinical Research Center Program RR59 of the Bureau of Research Resources, National Institutes of Health; and National Institutes of Health grant RR00163-67.
3 Address reprint requests to GT Gerhard, Department of Medicine, L465 Oregon Health Sciences University, 3181 SW Sam Jackson Park Road, Portland, OR 97201-3098. E-mail: gerhardg{at}ohsu.edu.
| ABSTRACT |
|---|
|
|
|---|
Objective: The objective of our study was to compare the effect of diet on fasting plasma lipids and lipoproteins and postprandial lipemia in premenopausal African American and white women.
Design: Thirteen African American and 9 white healthy premenopausal women were fed a low-fat, high-fiber diet and a high-fat, low-fiber diet for 4 wk each in a randomized crossover design. Fasting plasma lipid and lipoprotein concentrations and the 24-h plasma triacylglycerol response to a standard fatty test meal were measured at the end of each dietary period.
Results: Plasma total and LDL-cholesterol concentrations were higher after the high-fat, low-fiber diet in both white and African American women (P < 0.0001). The 24-h area under the plasma triacylglycerol curve after the test meal was lower after the low-fat diet than after the high-fat diet (P < 0.04).
Conclusions: African American and white women had lower fasting plasma total and LDL-cholesterol concentrations and less postprandial lipemia after the low-fat than the high-fat diet. Diets low in total and saturated fat and cholesterol and high in fiber may reduce the risk of CHD by lowering fasting plasma total and LDL-cholesterol concentrations and by reducing the lipemic response to fatty meals.
Key Words: Lipids and lipoproteins postprandial lipemia premenopausal women African American women white women dietary fat
| INTRODUCTION |
|---|
|
|
|---|
Compared with white women, premenopausal African American women have greater rates of obesity (8) and hypertension (9) and higher plasma concentrations of lipoprotein(a) (10) and higher plasma total homocysteine (11), all of which may contribute to their higher rate of CHD. The increased risk of CHD in African American women conferred by the above risk factors may be reduced by their higher HDL-cholesterol concentrations (12). One important risk factor for CHD that has not been adequately studied in African American women is diet. In particular, diets high in saturated fat and cholesterol raise the LDL-cholesterol concentration by suppressing LDL receptor expression in the liver (1315) and promote atherogenesis (15). Furthermore, there is evidence that individuals may differ in the degree of responsiveness of their fasting plasma LDL-cholesterol concentrations to changes in the saturated fat and cholesterol contents of the diet; hyperresponders have an exaggerated LDL response to diet, which may increase their risk of CHD (16). In addition, hyperresponsiveness of the plasma triacylglycerol concentration to a single high-fat meal (ie, an enhanced postprandial lipemic response) may also increase the risk of CHD (17).
There are limited or no data on plasma lipid responses to dietary fat in premenopausal African American and white women. Therefore, we conducted a metabolic study in which premenopausal African American and white women were fed low-fat and high-fat diets for 4-wk periods and the response of the plasma lipids and lipoproteins measured. In addition, the plasma triacylglycerol response to an acute fat load was determined at the end of each of the 2 dietary periods. Our hypothesis was that, compared with white women, premenopausal African American women would be hyperresponsive to diets and single meals rich in saturated fat and cholesterol. If African American women are indeed hyperresponsive, this could contribute to their higher rate of CHD.
| SUBJECTS AND METHODS |
|---|
|
|
|---|
3 alcoholic drinks/wk or who were known or suspected current drug abusers were excluded. The study was approved by the Institutional Review Board of the Oregon Health Sciences University. Written, informed consent was obtained from all the participants.
Experimental protocol
The study was conducted in the Clinical Research Center (CRC) at our institution. African American and white women who were matched for age, BMI, educational attainment, and LDL-cholesterol concentration were randomly assigned to receive 1 of 2 diets, 1 low and 1 high in total and saturated fat and cholesterol. Each diet, was fed for 4 wk. After a washout period of
4 wk, each subject then received the diet that was not fed in the first 4 wk in a crossover design.
The subjects visited the CRC on a daily basis Monday through Friday to be weighed, to be questioned about their compliance or problems with the experimental diets, and to receive their meals. All meals were prepared by the CRC kitchen. The subjects typically consumed one meal daily at the CRC and packed the rest to eat elsewhere. They were allowed to pack their meals on the weekends but were encouraged to come in to be weighed on Saturday. The energy intake for weight maintenance was computed for each woman by using the Mayo Clinic Nomogram (19), and energy intake was adjusted as needed to maintain body weight. The subjects were additionally instructed to maintain the same level of physical activity throughout the course of the study.
Fasting plasma lipid and lipoprotein concentrations were measured 3 times during the final week of each of the 2 dietary periods. In addition, the 24-h plasma triacylglycerol response to a standard fatty test meal was measured in all subjects at the end of each of the 2 dietary periods.
Diets
The compositions of the 2 experimental diets are shown in Table 1
. The combined effect of cholesterol and saturated fat was expressed by using an index termed the cholesterol-saturated fat index (CSI), which was developed to rank foodstuffs on the basis of their ability to increase plasma LDL-cholesterol concentrations. A high-CSI diet raises LDL cholesterol and a low-CSI diet lowers it (20). The high-CSI diet mimicked but somewhat exaggerated the current American diet, providing 40% of the total energy as fat, 20% of energy as saturated fat, and 80 mg cholesterol/MJ (333 mg/1000 kcal). The cholesterol and saturated fat in the high-CSI diet were provided by egg yolks, cheese, whole milk, butter, sour cream, beef, ham, bacon and sausage, and palm oil. The CSI of this diet was 9/MJ. The high-CSI diet provided 45% of the daily energy as carbohydrate and 2.2 g fiber/MJ (9 g/1000 kcal) daily. The low-CSI diet used nonfat or low-fat dairy products in place of milk and cheese, and chicken or turkey breast and water-packed tuna in place of beef. This diet provided 20% of the total energy as fat, including 6% as saturated fat, and 7 mg cholesterol/MJ (29 mg/1000 kcal), with a CSI of 2/MJ, about one-fifth of the CSI of the high-fat diet. The low-CSI diet supplied 65% of the total daily energy as carbohydrate and 3.6 g fiber/MJ (15 g/1000 kcal) daily. The low-CSI diet was bulkier than was the high-CSI diet, as indicated by its higher fiber content and greater weight (202 g/MJ) compared with the high-CSI diet (150 g/MJ). Both diets supplied 15% of the total daily energy as protein.
|
|
|
Lipids and lipoproteins
Lipids and lipoproteins were measured in our lipid laboratory by using standard procedures in compliance with the standardization and surveillance programs of the Centers for Disease Control and Prevention Laboratory in Atlanta, according to procedures established by the Lipid Research Clinics Program (21).
Statistical analysis
For all analyses, if the statistical assumptions of normality and equal variance were not met, variables were log transformed before analysis. The results are presented as the original (untransformed) values for ease of interpretation. Baseline characteristics of the African American and white women were compared by using an unpaired t test (22). A multiple linear regression analysis of baseline plasma triacylglycerol concentrations (dependent variable) on race and BMI (independent variables) was performed to determine predictors of baseline triacylglycerol concentrations in the women. A similar regression analysis was performed with VLDL cholesterol as the dependent variable. Weight changes associated with the low-fat and high-fat diets and between the races were compared by using a two-way repeated-measures analysis of variance. Main effects of diet and race and the interactive effect of race and diet (race x diet) were tested. Fasting plasma lipid and lipoprotein concentrations and the postprandial plasma triacylglycerol response to the fatty test meal were compared at the end of the 2 dietary periods in African American and white women by using a two-way analysis of variance procedure with repeated measures of diet (23). To minimize the effect of day-to-day intraindividual variation in fasting plasma lipid and lipoprotein concentrations, 3 values were obtained in the final week of each dietary period and the average was used in the statistical analysis. The 24-h plasma triacylglycerol response to the fatty test meal was quantitated in 3 ways (24):
![]() |
where C0 is the 0-h (baseline) plasma triacylglycerol concentration.
Pearson correlation coefficients (r) (22) were computed to assess the relation between the baseline plasma triacylglycerol concentration and the 24-h IAUC, Cmax, %CI, and time to peak lipemia. Because the baseline (0 h) plasma triacylglycerol concentrations differed between the diets and races (as determined by a two-way repeated-measures analysis of variance procedure; data not shown), statistical adjustment for the baseline triacylglycerol concentration was performed for the comparison of the 24-h area under the curve and the maximal triacylglycerol concentration attained (22). Statistical adjustment for baseline triacylglycerol concentrations was not performed for the comparison of the percentage increase in plasma triacylglycerols because use of the percentage rather than the absolute increase in effect corrects for differences in the baseline plasma triacylglycerol concentrations. Triacylglycerol values at individual time points on the 24-h plasma triacylglycerol curves were compared between the low-fat and high-fat diets by using paired t tests with the Bonferroni correction (22). Statistical analyses were performed by using SIGMA STAT (version 1.0; Jandel Scientific, San Rafael, CA). The graphic display was created with SIGMA PLOT (version 2.0; Jandel Scientific).
| RESULTS |
|---|
|
|
|---|
|
|
Postprandial triacylglycerol response
The postprandial plasma triacylglycerol responses to the fatty test meal in the white and African American women after the low-fat, high-fiber and high-fat, low-fiber dietary periods are shown in Table 6
and Figure 1
. Baseline (0 h) plasma triacylglycerol concentrations were highly correlated in both groups with the 24-h IAUC (r = 0.659, P < 0.0001, all subjects combined) and with the maximal triacylglycerol concentration attained over 24 h (Cmax) (r = 0.823, P < 0.0001). The %CI was not correlated with the baseline triacylglycerol concentration in African American or white women. Furthermore, the baseline plasma triacylglycerol concentrations differed significantly between the groups and between the 2 diets (data not shown). Therefore, to isolate the effects of diet and race on the degree of postprandial lipemia independent of their effects on the baseline triacylglycerol concentrations, the 24-h IAUC and Cmax analyses were statistically adjusted for differences in the baseline (0 h) plasma triacylglycerol concentrations. Use of the percentage increase in the plasma triacylglycerol concentration after the fatty test meal corrects, in essence, for differences in the baseline triacylglycerol concentrations, so statistical adjustment was not necessary for this analysis.
|
|
4 h and was not significantly correlated with baseline plasma triacylglycerol concentrations (data not shown). After reaching peak lipemia, the plasma triacylglycerol concentrations progressively declined until the 12-h time point and then rose at 24 h (Figure 1
Weight stability
When the white and African American women were grouped together, the overall mean weight changes were -1.1 kg with the low-fat, high-fiber diet and -0.3 kg with the high-fat, low-fiber diet, with an overall weight difference of -0.8 kg with the low-fat, high-fiber diet relative to the high-fat, low-fiber diet (P < 0.03). The subjects consumed >99% of the energy offered for both diets. The white women lost significantly more weight with both diets than did the African American women (P < 0.02). The weight changes for the white women were -1.8 and -1.3 kg with the low-fat, high-fiber and high-fat, low-fiber diets, respectively. For African American women the weight changes were -0.6 kg with the low-fat, high-fiber diet but 0.4 kg with the high-fat, low-fiber diet. There were no significant weight changes over the final 2 wk of the dietary periods (data not shown).
| DISCUSSION |
|---|
|
|
|---|
In our study, plasma concentrations of HDL cholesterol were lower and triacylglycerols higher after the low-fat, high-fiber diet than after the high-fat, low-fiber diet. Similar findings were reported in other studies (37). Lower HDL-cholesterol and higher triacylglycerol concentrations are associated with an increased risk of CHD (38, 39). Does this lowering of HDL-cholesterol and elevation of plasma triacylglycerol concentrations with low-fat diets, then, negate the protective effects of the decline in total and LDL-cholesterol concentrations? There are several reasons that this is probably not the case. Lowering the HDL-cholesterol concentration by dietary measures does not confer the same CHD risk as do low HDL-cholesterol concentrations in Americans eating a high-fat diet (40). The Lifestyle Heart Trial by Ornish et al (41) showed this point nicely. In that study, patients with coronary artery disease who consumed a vegetarian diet (with 10% of the total energy as fat) had less angina, less coronary artery stenosis, better myocardial perfusion, and fewer cardiac events than did a control group; these improvements occurred despite a reduction in HDL- and LDL-cholesterol concentrations with the low-fat diet. Furthermore, populations consuming a low-fat diet that have low HDL-cholesterol concentrations do not have high CHD rates (42, 43). An example of such a population is the Tarahumara Indians of Mexico (42), who have a very low rate of CHD despite an HDL-cholesterol concentration [0.65 mmol/L (25 mg/dL)] that would certainly be considered atherogenic in the US population.
The higher fasting plasma triacylglycerol concentration with the low-fat, high-fiber diet represented a physiologic response to a high-carbohydrate diet termed carbohydrate induction (44). Carbohydrate induction occurs when the amount of dietary carbohydrate is suddenly increased; it is the result of enhanced synthesis of triacylglycerol and VLDL by the liver (45). However, if the carbohydrate is phased in gradually, carbohydrate-induced hypertriglyceridemia may not occur (46).
Questions have been raised regarding the slightly greater weight loss with the low-fat, high-fiber diet than the high-fat, low-fiber diet. The difference in energy adjustment between the 2 diets was not due to differences in food intake between the diets, which was not significant, and was not secondary to the enhanced thermogenesis associated with carbohydrate compared with fat intake that was observed in other studies (47, 48). More specifically, if one assumes an increase in heat production of 20% for protein, 8% for carbohydrate, and 2% for fat (49), the predicted thermogenic effect of the 2 diets differed by only
146 kJ (35 kcal)/d, given a total daily intake of 11715 kJ (2800 kcal). This is clearly not enough to account for the 0.8 kg greater weight loss with the low-fat diet. The difference in energy adjustment between the 2 diets remains completely unexplained.
In our study, the low-fat background diet resulted in a lower 24-h postprandial lipemic response to a fatty test meal than did the high-fat diet in African American and white women. We cannot rule out a racial difference in response that was not detectable statistically because of the relatively small sample in our study. In a study by Duell et al (50), as in our study, administration of a background diet low in saturated fat and cholesterol to healthy subjects resulted in lower postprandial lipemia than did a diet high in saturated fat and cholesterol. Friday et al (51) showed that African American men had a lower lipemic response to a fatty test meal associated with higher postheparin lipoprotein lipase activity than did white men. However, the racial difference in postprandial lipemia was attenuated after statistical adjustment for several covariates, including fasting plasma triacylglycerol concentration and alcohol intake (
5 times higher in the white men). We are aware of no comparative studies of postprandial lipemia in premenopausal African American and white women.
Other recent studies (52, 53) with designs different from our own examined the effect of meals with differing fatty acid compositions on postprandial lipemia, hormonal status, and hemostatic function, which may relate to cardiovascular risk. In a study by Thomsen et al (52), postprandial lipemia was greater, and plasma gastric inhibitory peptide lower, after a saturated fatty acid meal than after a monounsaturated fatty acid meal; postprandial plasma glucose and insulin concentrations did not differ. Hunter et al (53) used a more physiologic fat load (44 g) and found that the postprandial hemostatic response of healthy young subjects was minimal irrespective of the fatty acid composition of the test meal (stearic, oleic, or linoleic acid).
Interestingly, in our study, the 2 fat-tolerance-test curves did not begin to diverge until 8 h after administration of the test meal; the greatest difference occurred at 24 h. In addition to the increase in plasma chylomicrons that occurs after a fatty meal, the concentration of VLDL particles and their remnants increases as a result of competition for hydrolysis between intestinally derived chylomicron particles and hepatically derived VLDL. The increase in VLDL particles and their remnants tends to persist longer (54, 55). Our data are thus consistent with a more prolonged postprandial increase in VLDL and VLDL remnant concentrations with the high-fat than with the low-fat background diet. The postprandial increase in VLDL and remnants may be the result of down-regulation of the hepatic LDL receptor that occurs during consumption of diets high in saturated fat and cholesterol, because the LDL receptor clears VLDL remnants as well as LDL (56, 57). Delayed clearance of VLDL and remnant particles is associated with an increased risk of atherogenesis (58). Thus, diets low in saturated fat and cholesterol may protect against CHD not only by decreasing fasting plasma LDL-cholesterol concentrations but also by attenuating the delayed increase in VLDL and VLDL remnant concentrations that may occur after a high-fat meal.
In conclusion, fasting plasma total and LDL-cholesterol concentrations were lower after the low-fat than after the high-fat background diet in premenopausal African American and white women. The total group of African American and white women had less lipemia over the 24-h period after a high-fat meal after consumption of the low-fat, high-fiber background diet; this should be regarded as a preliminary finding until confirmed by future studies. We did not find racial differences in the response to diets and meals rich in fat, although our ability to detect a differential racial response was limited by the small sample in our study. Diets low in total and saturated fat and cholesterol and high in fiber may decrease the risk of CHD not only by lowering fasting plasma total and LDL-cholesterol concentrations but possibly also by decreasing postprandial lipemia.
| ACKNOWLEDGMENTS |
|---|
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
S. G. Aldana, R. L. Greenlaw, H. A. Diehl, R. M. Merrill, A. Salberg, and H. Englert A video-based lifestyle intervention and changes in coronary risk Health Educ. Res., February 1, 2008; 23(1): 115 - 124. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Dewell, P. L. W. Hollenbeck, and C. B. Hollenbeck A Critical Evaluation of the Role of Soy Protein and Isoflavone Supplementation in the Control of Plasma Cholesterol Concentrations J. Clin. Endocrinol. Metab., March 1, 2006; 91(3): 772 - 780. [Abstract] [Full Text] [PDF] |
||||
![]() |
Z. Li, S. Lamon-Fava, J. Otvos, A. H. Lichtenstein, W. Velez-Carrasco, J. R. McNamara, J. M. Ordovas, and E. J. Schaefer Fish Consumption Shifts Lipoprotein Subfractions to a Less Atherogenic Pattern in Humans J. Nutr., July 1, 2004; 134(7): 1724 - 1728. [Abstract] [Full Text] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |