| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Nutrition, Arizona State University-Polytechnic, Mesa, Arizona
Address reprint requests to: Donna M. Winham, DrPH, Department of Nutrition, HSC 1386, 7001 East Williams Field Road, Arizona State University Polytechnic, Mesa, Arizona 85212. E-mail: donna.winham{at}asu.edu
 |
ABSTRACT |
|---|
 TOPFOOTNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSCONCLUSIONACKNOWLEDGMENTSREFERENCES |
|---|
Methods: Randomized, crossover 3x3 block design. Sixteen participants (7 men, 9 women) received each treatment for eight-weeks with two-week washouts. Fasting blood samples collected at beginning and end of periods were analyzed for total cholesterol (TC), low density lipoprotein cholesterol (LDL-C), high density lipoprotein cholesterol, triacylglycerols, high-sensitivity C-reactive protein, insulin, glucose, and hemoglobin A1c.
Results: A significant treatment-by-time effect impacted serum TC (p = 0.026) and LDL (p = 0.033) after eight weeks. Paired t-tests indicated that pinto beans were responsible for this effect (p = 0.003; p = 0.008). Mean change of serum TC for pinto bean, black-eyed pea and placebo were –19 ± 5, 2.5 ± 6, and 1 ± 5 mg/dL, respectively (p = 0.011). Mean change of serum LDL-C for pinto bean, black-eyed pea and placebo were –14 ± 4, 4 ± 5, and 1 ± 4 mg/dL, in that order (p = 0.013). Pinto beans differed significantly from placebo (p = 0.021). No significant differences were seen with other blood concentrations across the 3 treatment periods.
Conclusions: Pinto bean intake should be encouraged to lower serum TC and LDL-C, thereby reducing risk for CHD.
|
INTRODUCTION |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSCONCLUSIONACKNOWLEDGMENTSREFERENCES |
|---|
The majority of legume studies have examined the relationship between soybeans and heart disease. Less is known about the effects of other commonly consumed varieties, such as pinto beans or black-eyed peas on heart disease and type 2 diabetes biomarkers [1, 3]. Some studies using navy beans and chickpeas have been conducted. In an early report, Anderson et al. demonstrated that consumption of navy beans in tomato sauce (baked beans) for 21 days decreased serum total cholesterol (TC) and low-density lipoprotein cholesterol (LDL-C) concentrations in hypercholesterolemic men [4]. Nestel et al. demonstrated that a test meal of chickpeas improved insulin action 2 hours post-prandially [5]. However, they did not see a long-term beneficial effect of daily chickpea ingestion on insulin sensitivity after 6 weeks in their healthy participants. The absence of a significant long-term effect of chickpea consumption may be attributed to the study population being insulin-sensitive (mean fasting insulin concentrations 
7 µU/L as compared to the standard reference range 6–24 µU/L) rather than insulin-resistant [5]. Because an insulin-resistant population may show a greater risk reduction in insulin, a participant population, albeit healthy and disease-free, may show greater effects if they have a mild to moderate risk profile.
The objective of this research was to demonstrate if daily pinto bean or black-eyed pea ingestion for 8 weeks would result in improved biochemical markers of CHD and diabetes mellitus (DM) risk in apparently healthy, moderately insulin resistant participants. Primary endpoints were fasting TC, LDL-C and insulin. Secondary endpoints included high-density lipoprotein cholesterol (HDL-C), triacylglycerols (TG), high-sensitivity C-reactive protein (hs-CRP), glucose, and hemoglobin A1c (HbA1c).
|
MATERIALS AND METHODS |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSCONCLUSIONACKNOWLEDGMENTSREFERENCES |
|---|
15 µU/ml were eligible to participate. Exclusion criteria included pregnancy; use of cholesterol-lowering medications or supplements; expressed desire to lose weight; weight loss or gain in excess of 5 kg in the past 6 months; cigarette smoking or other chronic tobacco use; diagnosed diabetes, heart disease, stroke, hypertension, cancer, or a fasting insulin concentration in excess of 50 µU/ml. Power calculations indicated that a sample size of 14 would be sufficient to demonstrate a decrease in total cholesterol of 20 mg/dL with a power level of 0.80 and significance level of 0.05 [6]. Potential participants who met eligibility requirements were scheduled for a screening interview and a fasting blood draw at the University. At this appointment, participants provided written informed consent and completed a self-administered medical history questionnaire which inquired about medication use, alcohol consumption, and willingness to keep weight and activity levels consistent during the study. A fasting venous blood sample was analyzed for insulin in the Department of Nutrition laboratory via a standard radioimmunoassay. Participants were notified of their eligibility status based on insulin results within one week of the blood draw.
Sixty-five potential participants had fasting blood samples analyzed. Of these, 29 did not qualify because their insulin concentrations were too low. Three individuals were ineligible due to insulin concentrations in excess of 50 µU/ml indicating severe insulin resistance and possible diabetes mellitus. They were instructed to contact their personal physician. Of the 33 eligible people, 10 declined to start the study. Reasons given were desire to lose weight (n = 5), lack of time (n = 4), and medical reasons (n = 1). Twenty-three participants began the study. Seven people dropped out after completing varying weeks. Reasons for dropping out included lack of time (n = 2), moving out of the area (n = 2), and medical problems not associated with the study (n = 3). Sixteen individuals (70%; 7 men, 9 women) completed all three 8-week dietary interventions and wash-out periods encompassing 7–8 months of study involvement. There were no statistically significant differences by sex, BMI, or age between those eligible individuals who declined to participate, withdrew, or completed the study.
Participants were enrolled in the study on an ongoing basis over 6 months. Treatment randomization was done by having each participant draw the name of one of the three treatments from a bowl containing equal numbers of the three options. Participants received $80 in gift cards to a major retailer after completion of each 8 week study phase ($240 total). At the end of the study, participants were provided copies of all blood analyses, a personal food intake analysis report and a consultation with the staff registered dietitian. The institutional review board of the University approved all aspects of the study protocol (Human Subjects number 07858-04).
Participants completed a 2-day baseline diet record before beginning the study. A registered dietitian assessed these records for completeness and accuracy and provided feedback and further instruction on how to complete diet records at the first visit. During the three 8-week trials, participants completed 24-hour diet records at 2-week intervals resulting in a total of 4 diet records reflecting different days of the week and weekends per test period. Participants were strongly encouraged to bring in labels or nutrient information from items consumed to aid in data entry for food analysis. Each participant received a food scale and a set of measuring cups to weigh and measure foods on the days they kept food logs. Participants received test foods, recipes, and instructions to not eat other legumes, soy products, or carrots during the trial. Food log data were entered into Food Processor nutrient analysis software (v. 8.4, ESHA Research, Salem, Oregon). Nutrient intake was averaged over each of the three study periods and for baseline entries.
Participant compliance was monitored and reinforced by staff through frequent contact. Participants met with the dietitian every other week to pick up additional cans of food product and to review their one day food log. They received a follow-up telephone call by a staff member on the alternate week. During the telephone call or meeting, participants were asked how many times they had eaten the food product, if they had difficulty fitting the food product into their diet, and if they experienced any gastrointestinal discomfort or changes as a result from eating the study foods. Staff encouraged compliance and offered serving suggestions for foods as needed.
The test pinto beans and black-eyed peas and their fiber and nutrient analyses were provided by Archer Daniels Midland (Decatur, Illinois) and canned specifically for use in this study by Bush Brothers & Company (Knoxville, Tennessee). The pinto beans contained 5.4 g total dietary fiber (3.7 g insoluble, 1.7 g soluble) and the black-eyed peas contained 2.9 g total dietary fiber (2.1 g insoluble, 0.8 g soluble) per 1/2 cup serving. The control food, canned North Pride brand sliced carrots, was purchased at a local store, and, based on the nutrition label, the carrots contained 2.0 g total dietary fiber. No values were available for insoluble and soluble amounts per 1/2 cup serving. Participants were asked to consume a single 1/2 cup serving of pinto beans, black-eyed peas, or carrots, as well as the brine or liquid the food was packaged in, daily for 8 weeks as part of their usual diet.
Height (GPM, Carlstadt, NJ), weight (Seca scale, Hanover, MD), waist and hip circumferences (Gulick non-stretch consistent tension measuring tape; Sammons Preston, Chicago, IL) were recorded at the start and end of each 8-week trial. At the first meeting, participants picked which US Census 2000 race and ethnicity categories best identified them. The Physical Activity Recall (PAR) questionnaire as developed by Sallis et al. [7] was administered at the beginning, middle, and end of each treatment phase to obtain information regarding the duration and frequency of physical activity engaged in during the previous seven days by participants. Weight and blood pressure were recorded at each biweekly visit. Participants were reminded to keep their weight stable and to not alter their physical activity and exercise patterns during the course of the study.
At the beginning and end of each test period, a trained phlebotomist performed a fasting blood draw with the participant seated. Blood samples were analyzed using a Roche Modular Analytic System for TC, LDL-C, HDL-C, TG, hs-CRP and HbAlc by Sonora Quest Laboratories (Tempe, AZ). TC and TG were assessed photometrically and HDL cholesterol was measured by an enzyme immunoassay. LDL cholesterol was calculated using a modified Friedewald equation where TG are divided by 6 instead of 5 ([LDL-chol] = [Total chol] – [HDL-chol] – ([TG]/6) (Sonora Quest Laboratories, Tempe, AZ). Blood samples were analyzed for insulin and glucose by a trained laboratory technician at the ASU Polytechnic Department of Nutrition laboratory. Glucose and insulin were analyzed via the standard colorimetric glucose oxidase assay and a standard radioimmunoassay, respectively. Insulin sensitivity was calculated with the homeostasis model assessment {HOMA = [fasting insulin (uU/ml) x fasting glucose (mM)]/22.5} [8].
The Statistical Package for the Social Sciences (SPSS, version 13.0, SPSS Inc., Chicago, IL), was used for all data analysis. Food treatment effects on biomarkers were estimated by the multiple analysis of variance (MANOVA) test for repeated measures. Paired t-tests were used as appropriate to detect differences in the means of variables at baseline and the end of each 8 week intervention period. Data are presented as the mean ± standard error of the mean (SEM). Statistical significance was indicated by a p-value of 
0.05.
|
RESULTS |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSCONCLUSIONACKNOWLEDGMENTSREFERENCES |
|---|
|
|
|
The mean change of serum TC for pinto bean, black-eyed pea and placebo were –19 ± 5, 2.5 ± 6, and 1 ± 5, respectively (p = 0.011). For serum LDL-C, the mean changes over the 8 week intervention were –14 ± 4 for pinto beans, 4 ± 5 black-eyed peas, and 1 ± 4 for placebo (p = 0.013). These changes with pinto bean consumption over time are equivalent to percent changes of –8.4% for TC and –8.6% for LDL-C for all participants. Division of the participants into normocholesterolemic (n = 4; TC <200 mg/dL) and hypercholesterolemic (n = 12; TC 200 mg/dL) showed no differences in the percent reduction of TC (–7.8 vs. -8.6%) and LDL (–8.5 vs. 8.7%). No significant differences were seen with other blood concentrations across the 3 treatment phases.
Daily consumption of black-eyed peas for eight weeks did not significantly alter serum concentrations of TC, LDL-C, HDL-C, TG, hs-CRP, glucose or insulin in this study population of generally healthy, mild to moderately insulin resistant adults. Following eight weeks of daily ingestion of the placebo, 1/2 cup canned carrots, serum parameters changed minimally, as expected.
|
DISCUSSION |
|---|
This trial appears to be the first to examine the effects of pinto beans and black-eyed peas independently. However, the mean percent reductions in serum TC and LDL-C of 8.4% and 8.6% observed are similar to reductions observed in most other trials of mixed pinto beans and other legumes, but not all. Mean percent reductions of 7% and 5% in serum TC and LDL-C were seen in free-living, hypercholesterolemic men fed such a type of mixed legumes [10]. Greater reductions were observed in serum TC and LDL-C of 19% to 23% and 23% to 24% in two similar, but separate, trials that examined the effects of mixed pinto and navy beans [4, 11]. All three trials fed larger quantities of legumes than 1/2 cup and only included hypercholesterolemic men.
In a contrasting study, Mackay and Ball noted almost no change in serum TC and LDL-C when pinto beans were fed with other legumes (80 grams or about 1/3 cup) to free-living, hypercholesterolemic men and women [12]. The lack of change may be due to the small quantity fed, the combination of legume types, and use of a free-living population. Due to the different methodologies and study designs in these other studies, direct comparison to our research is difficult.
Unlike pinto beans, consumption of black-eyed peas over 8 weeks did not lower serum TC or LDL-C substantially. We know of no other published studies that have examined the effects of black-eyed peas on CHD biomarkers. At 3 grams per 1/2 cup, the amount of total fiber in black-eyed peas is less than half that of pinto beans (7 grams). A higher dose may have reduced lipids as well, e.g. 1 cup per day, assuming that fiber content is the driving force behind the reductions in cholesterol.
The observed cholesterol reduction seen with some beans probably occurs through multiple mechanisms that are dependent upon the type of components in each bean variety. These include fat displacement from the diet, direct binding of dietary cholesterol by viscous fiber in the intestine, interruption of bile acid enterohepatic circulation, and inhibition of endogenous cholesterol synthesis by short chain fatty acids (SCFA), among others.
If the daily legume treatment displaced consumption of dietary saturated fats, TC and LDL-C could drop due to the lack of fat intake rather than because of the beans. In studies that fed 200 grams (3/4 cup) or more of legumes, it is likely that fat would be displaced. This may partially explain the larger reductions in serum TC and LDL-C seen over shorter time periods in some other clinical trials. In our study, it is less likely that the drop in cholesterol reduction was due to fat displacement alone because participants only ate a 1/2 cup (140 gm). Diet records indicated no significant differences in dietary fat intakes between the three treatment phases and baseline.
Several other modes of action are more complex and beyond the scope of this research study. Viscous fibers found in some legumes bind cholesterol in the intestine and promote elimination via excretion. Reduced absorption of dietary cholesterol decreases the internal cholesterol pool and forces the body to make more. In addition, the adsorption of bile acids by dietary viscous fibers in the intestine can increase bile acid excretion and ultimately, interruption of the enterohepatic circulation. As the liver utilizes cholesterol precursors and cholesterol from the blood to synthesize new bile acids to replace those lost through excretion, serum cholesterol levels become lower. Cholysteramine, a cholesterol-lowering medication, has a bile acid binding capability near 100% compared to cellulose, a non-viscous fiber that does not bind bile acids. Binding capacity of bile acids for pinto beans, black beans, soy beans, black-eyed peas, garbanzo beans (chickpeas) and lima beans were 5.5%, 8.2%, 1.9%, 3.3%, 10.0%, and 3.7%, respectively relative to cholysteramine [13, 14]. The amount of bile acids bound by the pinto beans would be relatively small.
A fourth proposed mechanism is the fermentation of resistant starches (RS) by bacteria present in the human colon. Legumes contain high amounts of RS, which is any starch that resists digestion by amylase in the small intestine and progresses to the large intestine for fermentation by the gut bacteria. One of the end-products of RS fermentation is propionate. Propionate has been shown to inhibit hepatic cholesterol synthesis in animal models but this potential mechanism has not been extensively tested in humans [15].
Since our study was not designed to examine such potential mechanisms, it is difficult to speculate as to which mechanism or combination of mechanisms were responsible for the substantial reduction in both serum TC and LDL-C after the pinto bean treatment phase. From a practical standpoint, whether legumes, such as pinto beans, lower serum TC or LDL-C via fat displacement or another mechanism may not be as important as the simple fact that individuals who consume them are achieving these reductions and reducing their overall risk for CHD.
Serum HDL-C, TG, and hs-CRP did not change with consumption of pinto beans and black-eyed peas. In other similar studies, HDL-C has either not been effected by legume consumption or has been lowered in some cases by as much as 7% to 20% [4, 10, 14]. All studies in which HDL-C values decreased also observed reductions in TC, which may help explain the decrease. Anderson et al. reported a significant 20% reduction in mean serum HDL-C in addition to a 23% reduction in mean serum TC [11]. Some other trials have found reductions in serum TG using larger amounts of legumes. Often legumes replaced simple carbohydrate sources according to the researchers [10].
Because of their RS content, legumes are frequently recommended as a low-glycemic index food thus benefiting individuals who are insulin-resistant and/or diabetic. However, pinto bean or black-eyed pea consumption over 8 weeks did not produce changes in the HOMA index nor glucose and insulin concentrations in the mild-moderately insulin-resistant study population. It is possible that the 1/2 cup dose was insufficient to produce a change. Nestel et al. noted similar findings with chickpea consumption in a non-insulin resistant population [5]. Other explanations may be that the RS of the beans needs to be ingested more frequently than once per day, or simply that non-diabetic individuals will not show a significant change in insulin resistance. The HOMA index may not be a reliable predictor of insulin-resistance among people who do not have impaired glucose tolerance [17]. We believe further research should be conducted in this area with more frequent consumption of legumes throughout the day, e.g. 1/2 cup twice daily, and with persons who have type 2 diabetes.
Several design elements distinguish this study. At 8 weeks, the trial was substantially longer than many others, which have typically lasted three to four weeks. Both free-living men and women were included in a wider age range. Self-reported racial and ethnic identity was obtained. Baseline dietary intake data were collected. Without pre-intervention dietary data, it is impossible to determine if dietary intakes changed with treatments.
Efforts were made to control for possible confounding factors by monitoring dietary intake, physical activity and weight status. Although conditions were not controlled stringently as in a metabolic ward setting, physiologically significant reductions in serum lipoproteins were seen. These results give credence to the ability of the layperson to reduce his/her cholesterol with the addition of pinto beans to the daily diet.
It is important to note that improvements in serum lipoprotein profiles occurred even within individuals who had ‘normal’ concentrations less than 200 mg/dL and not only among those participants who were hypercholesterolemic. Of the human clinical trials that have investigated the effects of legumes on serum lipoproteins, most included only hypercholesterolemic individuals [4, 10, 11, 16]. As both obesity and the metabolic syndrome have become more widespread and prevalent in the US, more people have become mild to moderately insulin resistant now than in previous decades.
The quantity of legumes fed to participants in this clinical trial, a 1/2 cup daily or 3.5 cups weekly, was consistent with recent US Government recommendations for legume intake. This research provided evidence that pinto beans can serve as a functional food and decrease risk of CHD through a reduction in both total and LDL cholesterol and should be recommended to the general public.
The study design is limited by several factors. Dietary intake information and compliance were based on self-report. Self-reported food intakes may underestimate actual consumption. Completion of weighed food records can be a substantial participant burden. The length of the trial may have caused participant fatigue with less compliance in later phases. Sufficient participant numbers were met according to power calculations for change in total cholesterol, but a larger sample size may have resulted in changes in glucose or insulin concentrations with prolonged ingestion of the legume varieties.
|
CONCLUSION |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSCONCLUSIONACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSCONCLUSIONACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSCONCLUSIONACKNOWLEDGMENTSREFERENCES |
|---|
Preliminary results were presented at the annual meeting of the Federation of American Societies of Experimental Biology (FASEB), San Francisco, CA, April 2, 2006; FASEB Journal Session 269.6.
Research supported by the Beans for Health Alliance (BHA) with partial support from the United States Agency for International Development (USAID) through the Global Developmental Alliance Secretariat under grant no. REE-A-00-03-00094-00.
Received October 15, 2006. Accepted January 29, 2007.
|
REFERENCES |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSCONCLUSIONACKNOWLEDGMENTSREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
