Soluble fiber intake at a dose approved by the US Food and Drug Administration for a claim of health benefits: serum lipid risk factors for cardiovascular dis

¹ From the Clinical Nutrition and Risk Factor Modification Center, St Michael'sHospital, Toronto (DJAJ, CWCK, VV, EV, TP, DF, and CCM); the Departments of Nutritional Sciences (DJAJ, CWCK, VV, EV, SCC, and MAR) and Preventive Medicine and Biostatistics (PNC), Faculty of Medicine, University of Toronto; and the Department of Food Science Technology and Microbiology, Division of Human Nutrition, University of Milan, Milan, Italy (MG and GT).

² Supported by The University-Industry Research Partnership Program of the Natural Sciences and Engineering Research Council of Canada, Ottawa, and The Kellogg Company, Battle Creek, MI. DJAJ is funded as a Canada Research Chair in Metabolism and Nutrition at the University of Toronto by the Federal Government of Canada, Ottawa.

³ Address reprint requests to DJA Jenkins, Department of Nutritional Sciences, Faculty of Medicine, University of Toronto, Toronto, Ontario, Canada, M5S 3E2. E-mail: cyril.kendall{at}utoronto.ca.

ABSTRACT  
Background: The US Food and Drug Administration (FDA) approved health claims for 2 dietary fibers, ß-glucan (0.75 g/serving) and psyllium (1.78 g/serving), on the assumption that 4 servings/d would reduce cardiovascular disease risk.

Objective: We assessed the efficacy of this dose of fibers in reducing serum lipid risk factors for cardiovascular disease.

Design: Sixty-eight hyperlipidemic adults consumed a test (high-fiber) and a control low-fat (25% of energy), low-cholesterol (<150 mg/d) diet for 1 mo each in a randomized crossover study. The high-fiber diet included 4 servings/d of foods containing ß-glucan or psyllium that delivered 8 g/d more soluble fiber than did similar, unsupplemented foods in the control diet. Fasting blood samples and blood pressure readings were obtained at baseline and weeks 2 and 4, and the subjects' weight was monitored weekly.

Results: Compared with the control diet, the high-fiber diet reduced total cholesterol (2.1 ± 0.7%; P = 0.003), total:HDL cholesterol (2.9 ± 0.8%; P = 0.001), LDL:HDL cholesterol (2.4 ± 1.0%; P = 0.015), and apolipoprotein B:A-I (1.4 ± 0.8%; P = 0.076). Applying the Framingham cardiovascular disease risk equation to the data confirmed a reduction in risk of 4.2 ± 1.4% (P = 0.003). Small reductions in blood pressure were found after both diets. The subjects reported no significant differences in palatability or gastrointestinal symptoms between the diets.

Conclusions: The reduction in serum lipid risk factors for cardiovascular disease supports the FDA'sapproval of a health claim for a dietary fiber intake of 4 servings/d. Although relatively small in terms of patient treatment, the reduction in cardiovascular disease risk is likely to be significant on a population basis.

Key Words: Soluble fiber • psyllium • oats • ß-glucan • LDL cholesterol • HDL cholesterol • coronary artery disease • National Cholesterol Education Program • Food and Drug Administration • health claim • functional foods

INTRODUCTION  
The US Food and Drug Administration (FDA), which assesses health claims for foods, has approved health claims for the viscous fibers oat ß-glucan and psyllium as cholesterol-lowering agents that in the context of a good diet may reduce the risk of cardiovascular disease (1, 2). Furthermore, national agencies concerned with cardiovascular health have for the first time recommended viscous fiber intake (3, 4). Much interest has been shown internationally by the general public, the scientific community, and federal regulators in the medical application of these food components in so-called functional foods, foods that favorably modify physiologic function. There is therefore a need to substantiate the validity of current regulations that govern health claims in this area (1, 2, 5, 6). To address this need, we tested the effects of psyllium and ß-glucan intake, in the amounts approved by the FDA for a fiber health claim, on serum lipid risk factors for cardiovascular disease (1, 2).

SUBJECTS AND METHODS  
Ninety-one hyperlipidemic subjects were recruited and 82 were available for random assignment. Of these, 68 (83%; 37 men and 31 postmenopausal women) completed both 1-mo dietary phases, separated by a 2-wk washout period in a randomized crossover study. The subjects' mean (±SE) age was 60 ± 1 y (range: 33–82 y) and their mean body mass index (in kg/m2) was 25.6 ± 0.3 (range: 20.0–33.8). At baseline, all subjects had elevated serum LDL-cholesterol concentrations (>4.1 mmol/L) (7). One subject started with triacylglycerol concentrations >4.0 but <6.0 mmol/L. None had clinical or biochemical evidence of diabetes, liver disease, or renal disease, and none were taking hypolipidemic agents. Dosages of medications and level of physical activity were held constant for both study periods. Blood samples were obtained and blood pressure was measured with the subjects seated after they had fasted 12–14 h overnight before the start and at the end of weeks 2 and 4 of each phase. Serum was stored at -70oC until analyzed. Body weight was measured at the start and during biweekly clinic visits in both phases, and the subjects were also asked to monitor their weight on a home scale weekly.

For the last 7 d of each phase, the subjects recorded their diets after weighing food items on self-tarring electronic scales. During week 4 of each phase the subjects noted their number of bowel movements and rated flatulence, bloating, and abdominal pain on a 9-point bipolar scale. At each biweekly visit, subjects also rated their feelings of satiety with the use of a 9-point scale on which -4 represented extremely hungry, 0 neutral, and +4 completely satiated. The study was approved by the Ethics Committee of the University of Toronto. All the subjects gave informed consent.

Diets
At least 1 mo before the start of the study, the subjects were instructed on the principles of a National Cholesterol Education Program Step II diet (total fat <30% of energy, saturated fat <7% of energy, and dietary cholesterol <200 mg/d) (3). The subjects were instructed to follow these guidelines throughout the course of the study. Test (high-fiber) and control study foods consisted of a variety of breakfast cereals, breads, pasta-based frozen dinners, tea cakes, cookies, potato chips, and smoothie beverages. On average these contributed 36.2% of total daily energy intake. The test foods contained 1.8–2.5 g psyllium or 0.75 g ß-glucan/serving (The Kellogg Co, Battle Creek, MI). Each subject selected 4 servings/d from the available foods, with the stipulation that one breakfast cereal and one frozen dinner must be chosen. On average, 7.2 g psyllium and 0.75 g ß-glucan were consumed daily from these products (Table 1). The control diet provided similar commercial foods without the added fiber. The foods provided replaced similar foods in the subjects' diets such that they maintained their weight. Supplements were recorded when eaten, and any uneaten supplements were returned at the end of the dietary phase.

View this table:
TABLE 1 . Calculated macronutrient intakes from week 4 dietary records with the high-fiber and control ad libitum diets¹  
Analyses
Serum was analyzed in a single batch for total cholesterol, triacylglycerols, and HDL cholesterol after dextran sulfate– magnesium chloride precipitation in accordance with the guidelines of the Lipid Research Clinics Program (8). LDL cholesterol was calculated by using the Friedewald equation. Serum apolipoprotein A-I and B were measured by nephelometry.

Macronutrient intakes were assessed from the 7-d dietary records with the use of a database derived from food-composition tables of the US Department of Agriculture (9) and with food labels for the few foods that were not analyzed directly. The supplements used in the study were analyzed with the use of Association of Official Analytical Chemists methods for fat, protein, and fiber, with available carbohydrate determined by difference (10). The fatty acid composition was determined by gas chromatography.

Statistical analysis
The results are expressed as means ± SEs. The mean of weeks 2 and 4 for each phase was used in calculating treatment differences because no significant differences were found between weeks 2 and 4 in any measurement except LDL cholesterol, for which a significant treatment effect was seen only at 2 wk. Furthermore, this difference was not reflected in a corresponding difference in non-HDL cholesterol; the mean of the LDL-cholesterol values from weeks 2 and 4 was therefore used in analysis of the data. The significance of the percentage difference between treatments was assessed by two-tailed Student'st test for paired data. The absolute difference between treatments was assessed by analysis of covariance with the use of the general linear model procedure and SAS software (PROC GLM/SAS) (11), with the response variable as the mean of the measurements from weeks 2 and 4 and the main effects of diet, sex x sequence interaction, and a random term representing the subject nested within the sex x sequence interaction and the baseline value as a covariate. Analysis with an alternative statistical model, with diet, sex, diet x sex, and the random term described above as main effects and dietary cholesterol and starting value as covariates, yielded similar results.

The Framingham predictive equation for cardiovascular disease risk was also applied to the data by using the total:HDL cholesterol and systolic blood pressure results. The equation also includes age, sex, the presence or absence of left ventricular hypertrophy, and whether the individual is a smoker or has diabetes (12). Our subjects did not have diabetes and were nonsmokers. Left ventricular hypertrophy was not assessed. Cardiovascular disease risk is expressed in the text as the percentage of individuals who would be predicted to have angina, have a myocardial infarction, or die during a 10-y period (12).

RESULTS  
The diets were well accepted and compliance was good. With both the control and high-fiber diets, subjects consumed 96 ± 1% of the supplements provided. Both diets had high and similar palatability scores (high-fiber diet: 3.4 ± 0.1; control diet: 3.2 ± 0.1) and satiety scores (high-fiber diet: 0.6 ± 0.1; control diet: 0.5 ± 0.1). There was no significant body weight change with either the high-fiber diet (-0.1 ± 0.1 kg) or the control diet (-0.1 ± 0.1 kg; Table 2).

View this table:
TABLE 2 . Body weight, serum lipid, and blood pressure data for the high-fiber and control ad libitum diet periods¹  
Blood lipids and apolipoproteins
Significant reductions from pretreatment values were seen in blood lipids after both diets (Table 2). There were no significant differences in pretreatment values between the high-fiber and control diets (Table 2). Mean values were significantly lower after the high-fiber diet than after the control diet for total cholesterol [percentage difference between treatments: 2.1 ± 0.7%; P = 0.003 (Students'st test)]; triacylglycerols (5.2 ± 2.4%; P = 0.037); apolipoprotein B (2.9 ± 0.8%; P < 0.001); total:HDL cholesterol (2.9 ± 0.8%; P = 0.001); and LDL:HDL cholesterol (2.4 ± 1.0%; P = 0.015), with a nonsignificant reduction in apolipoprotein B:A-I of 1.4 ± 0.8% (P = 0.076). There were no other significant differences. The percentage treatment differences in the lipid and lipoprotein ratios at week 2 were similar to the respective values at week 4. The percentage changes from baseline for the high-fiber and control treatments are shown in Figure 1. The significance of the effect of diet on serum lipids was confirmed by using the mean of the lipid values from weeks 2 and 4 and the general linear model procedure (for total cholesterol, triacylglycerol, apolipoprotein B, total:HDL cholesterol, LDL:HDL cholesterol, and apolipoprotein B:A-I; Table 2).

View larger version (14K):
FIGURE 1. . Percentage change in blood lipid, lipoprotein, and apolipoprotein ratios from week 0 to weeks 2 and 4 of the high-fiber () and control () diets.

 
There were no significant differences between the sexes in response to diet. In addition, neither age nor body mass index influenced the treatment effect. Controlling for weight change or differences in dietary carbohydrate, saturated or polyunsaturated fat, the ratio of polyunsaturated to saturated fatty acids, or dietary cholesterol in the general linear model procedure did not alter the pattern of significance in treatment differences of blood lipids.

Cardiovascular risk estimation
Application of the Framingham cardiovascular disease risk equation to the data confirmed a 4.2 ± 1.4% treatment difference in cardiovascular disease risk (P = 0.003) (12).

Blood pressure
Systolic and diastolic blood pressures tended to be reduced after both dietary phases but were not significantly different between treatments (Table 2).

Gastrointestinal symptoms
No significant differences were seen between the high-fiber and control diets in bloating, flatulence, or abdominal pain. Bowel movements were more frequent during the high-fiber than during the control diet (9.6 ± 0.5 compared with 8.6 ± 0.5 movements/wk; P = 0.005).

DISCUSSION  
Our data show that incorporation of viscous fibers into a wide range of foods resulted in small but significant reductions in total cholesterol and in the ratio of LDL to HDL cholesterol. These data add support to the FDA-approved health claim that in the context of a low-fat, low-cholesterol diet, 4 servings/d of foods containing the viscous fibers psyllium and oat ß-glucan can be expected to reduce serum lipids and the risk of cardiovascular disease.

The FDA was one of the first national agencies to recognize a role for fiber in cardiovascular disease risk reduction. Products that contain 0.75 g ß-glucan or 1.78 g psyllium/serving are permitted to carry a health claim stating that the product "will reduce the risk of coronary heart disease" (1, 2). The FDA further determined that 4 servings of these foods is likely to provide the effective daily dose (1, 2).

There is considerable international interest in health claims, both to encourage industry to produce foods with specific health benefits (functional foods) and to ensure efficacy of action (1, 2, 5, 6, 13). Validation of health claims is therefore important, especially in the case of soluble fiber, because national agencies concerned with cardiovascular health have only recently acknowledged a role for fiber in cholesterol reduction (4, 14). The ability of viscous soluble fibers to lower serum cholesterol has been recognized for more than a quarter of a century (15, 16). At first the dietary fibers of interest were pectin and guar (15, 17); later attention focused on oat ß-glucan and psyllium, which in most but not all studies were shown to reduce serum cholesterol concentrations (15, 16, 18–20).

The present study assessed the effects of consuming 8 g/d of a combination of the viscous fibers psyllium and ß-glucan, an amount that meets the FDA requirements for a health claim for cardiovascular disease risk reduction. Although the cholesterol reduction was modest, estimated cardiovascular disease risk was reduced. In addition, the ratios of total to HDL cholesterol and of LDL to HDL cholesterol were also reduced in our study, which was not previously reported with the consumption of fiber in studies achieving these modest levels of cholesterol reduction. Although glycemic index data are not available on the foods used in the present study, it is possible that part of the beneficial effect on the lipid and lipoprotein ratios was caused by a general lowering of the glycemic index of the diet. Incorporation of viscous fibers into a range of carbohydrate foods eaten over the day would tend to reduce the dietary glycemic index, as has been shown for individual foods and meals (21). Studies of low–glycemic index diets reported varied effects on serum cholesterol: some showed reductions (22–24) and others did not (25, 26). On the other hand, low-glycemic-index diets appear to lower serum triacylglycerols in subjects with raised triacylglycerol concentrations (27, 28).

In cohort studies, low-glycemic-index diets were associated with higher HDL-cholesterol concentrations (29, 30) and reduced risk of cardiovascular disease (31). The changes in lipoprotein ratios observed in the present study are considered key indicators of cardiovascular disease risk reduction and are central to equations predicting coronary artery disease risk (12). In this respect our data suggest a potential advantage of distributing fiber over the day in a range of carbohydrate foods.

The viscous nature of the types of dietary fiber that are most effective in reducing serum lipids has been a major barrier to their use in foods. Their clinical relevance has been questioned because of the extreme unpalatability of some of the products tested (32). The finding that palatable diets can be produced by incorporating lower amounts of fiber into more foods while retaining effectiveness is therefore important.

Foods containing sufficient psyllium and oat ß-glucan per serving to justify a health claim for cardiovascular disease risk reduction reduce serum lipids and are as palatable as their low-fiber counterparts. Spreading the fiber intake over the day may be responsible for reducing the ratio of total to HDL cholesterol and of LDL to HDL cholesterol, through lowering of the glycemic index of the carbohydrate portion of the diet. The palatability of and lack of side effects from these foods suggest that consumption of more servings of fiber-supplemented foods will also prove acceptable in clinical situations where larger reductions in lipid risk factors for cardiovascular disease are required. The present level of supplementation is likely to be of benefit on a population basis as one of several dietary strategies to reduce lipid risk factors for cardiovascular disease (33).

ACKNOWLEDGMENTS  
We thank Yu-Min Li and George Koumbridis for technical assistance.

REFERENCES  

1. US Food and Drug Administration. FDA final rule for federal labeling: health claims: oats and coronary heart disease. Fed Regist 1997;62:3584–681.
2. US Food and Drug Administration. Food labeling: health claims; soluble fiber from certain foods and coronary heart disease. Fed Regist 1998;63. (Docket no. 96P-0338.)
3. Expert Panel on Detection, Evaluation and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel II). JAMA 1993;269: 3015–3023.
4. Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Executive summary of the third report of The National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). JAMA 2001;285:2486–97.
5. Japan Health Food and Nutrition Food Association. Foods for specific health use. Tokyo: Japan Health Food and Nutrition Food Association, 1995.
6. Health Canada. Policy paper. Therapeutic products programme and the food directorate from the health protection branch. Nutraceuticals/functional foods and health claims on foods (November 2, 1998). Ottawa: Health Canada, 1998.
7. US Department of Health and Human Service. Expert Panel on Detection, Evaluation and Treatment of High Blood Cholesterol in Adults. Report of the National Cholesterol Education Program (NCEP). Washington, DC: US Government Printing Office, 1989. (NIH publication no. 89-2925.)
8. Lipid Research Clinics Program. Manual of laboratory operations. Volume 1: lipid and lipoprotein analysis. Washington, DC: US Government Printing Office, revised 1982. [US Department of Health, Education, and Welfare publication no. (NIH) 75-628.]
9. The Agricultural Research Service. Composition of foods. Agriculture handbook no. 8. Washington, DC: US Department of Agriculture, 1992.
10. Association of Official Analytical Chemists. AOAC official methods of analysis. Washington, DC: Association of Official Analytical Chemists, 1980.
11. SAS Institute: SAS/STAT user'sguide, version 6.12 edition. Cary, NC: SAS Institute, 1997.
12. Anderson KM, Wilson PW, Odell PM, Kannel WB. An updated coronary risk profile. A statement for health professionals. Circulation 1991;83:356–62.
13. Katan MB. Functional foods. Lancet 1999;354:794.
14. Krauss RM, Eckel RH, Howard B, et al. AHA dietary guidelines. Revision 2000: a statement for healthcare professionals from the Nutrition Committee of the American Heart Association. Circulation 2000;102:2284–99.
15. Anderson JW, Deakins DA, Floore TL, Smith BM, Whitis SE. Dietary fiber and coronary heart disease. Crit Rev Food Sci Nutr 1990;29:95–147.
16. Brown L, Rosner B, Willett WW, Sacks FM. Cholesterol-lowering effects of dietary fiber: a meta-analysis. Am J Clin Nutr 1999;69: 30–42.
17. Jenkins DJA, Newton AC, Leeds AR, Cummings JH. Effect of pectin, guar gum and wheat fibre on serum cholesterol. Lancet 1975;1: 116–7.
18. Olsen BH, Anderson SM, Becker MP, et al. Psyllium-enriched cereals lower blood total cholesterol and LDL cholesterol, but not HDL cholesterol, in hypercholesterolemic adults: results of a meta-analysis. J Nutr 1997;127:1973–80.
19. Anderson JW, Allgood LD, Lawrence A, et al. Cholesterol-lowering effects of psyllium intake adjunctive to diet therapy in men and women with hypercholesterolemia: meta-analysis of 8 controlled trials. Am J Clin Nutr 2000;71:472–9.
20. Swain JF, Rouse IL, Curley CB, Sacks FM. Comparison of the effects of oat bran and low-fiber wheat on serum lipoprotein levels and blood pressure. N Engl J Med 1990;322:147–52.
21. Jenkins DJ, Leeds AR, Gassal MA, Cochet B, Alberti GM. Decrease in postprandial insulin and glucose concentrations by guar gum and pectin. Ann Intern Med 1977;86:20–3.
22. Jenkins DJA, Wolever TMS, Collier GR, et al. Metabolic effects of a low-glycemic-index diet. Am J Clin Nutr 1987;46:968–75.
23. Wolever TM, Jenkins DJ, Vuksan V, Jenkins AL, Wong GS, Josse RG. Beneficial effect of low-glycemic index diet in overweight NIDDM subjects. Diabetes Care 1992;15:562–4.
24. Jarvi AE, Karlstrom BE, Granfeldt YE, Bjorck IE, Asp NG, Vessby BO. Improved glycemic control and lipid profile and normalized fibrinolytic activity on a low-glycemic index diet in type 2 diabetic patients. Diabetes Care 1999;22:10–8.
25. Fontvieille AM, Rizkalla SW, Penfornis A, Acosta M, Bornet FR, Slama G. The use of low glycaemic index foods improves metabolic control of diabetic patients over five weeks. Diabet Med 1992;9: 444–50.
26. Brand JC, Colagiuri S, Crossman S, Allen A, Roberts DC, Truswell AS. Low-glycemic index foods improve long-term glycemic control in NIDDM. Diabetes Care 1991;14:95–101.
27. Jenkins DJA, Wolever TMS, Kalmusky J, et al. Low-glycemic index diet in hyperlipidemia: use of traditional starchy foods. Am J Clin Nutr 1987;46:66–71.
28. Fontvieille AM, Acosta M, Rizkalla SW, et al. A moderate switch from high to low glycaemic-index foods for 3 weeks improves the metabolic control of type 1 (IDDM) diabetic subjects. Diabetes Nutr Metab 1988;1:139–43.
29. Frost G, Leeds A, Doré C, Madeiros S, Brading S, Dornhorst A. Glycaemic index as a determinant of serum HDL-cholesterol concentration. Lancet 1999;353:1045–8.
30. Ford ES, Liu S. Glycemic index and serum high-density lipoprotein cholesterol concentration among USA adults. Arch Intern Med 2001; 161:572–6.
31. Liu S, Willett WC, Stampfer MJ, et al. A prospective study of dietary glycemic load, carbohydrate intake and risk of coronary heart disease in US women. Am J Clin Nutr 2000,71:1455–61.
32. Ellis PR, Dawoud FM, Morris ER. Blood glucose, plasma insulin and sensory responses to guar-containing wheat breads: effects of molecular weight and particle size of guar gum. Br J Nutr 1991;66:363–79.
33. Jenkins DJA, Kendall CWC, Vuksan V. Viscous fibers, health claims, and strategies to reduce cardiovascular disease risk. Am J Clin Nutr 2000;71:401–2.