Journal of the American College of Nutrition, Vol. 22, No. 3, 195-200 (2003)
Published by the American College of Nutrition
Effects of Plant-Based Diets High in Raw or Roasted Almonds, or Roasted Almond Butter on Serum Lipoproteins in Humans
Gene A. Spiller, PhD, CNS, FACN,
Anna Miller, MS, RD,
Karla Olivera, MS, RD,
Julie Reynolds, RD,
Beverly Miller, RD,
Stanley J. Morse, PhD,
Antonella Dewell, MS, RD and
John W. Farquhar, MD
Sphera Foundation (G.A.S., A.M., K.O., J.R., B.M., S.J.M., A.D.), Los Altos
Stanford Center for Research in Disease Prevention, Stanford University (J.W.F.), Stanford, California
Address reprint requests to: Gene A. Spiller, PhD, Sphera Foundation, P.O. Box 338, Los Altos, CA 94023. E-mail: spiller{at}sphera.org
 |
ABSTRACT
|
|---|
Objective: To compare the lipid-altering effect of roasted salted almonds and roasted almond butter with that of raw almonds, as part of a plant-based diet.
Methods: Thirty-eight free-living, hypercholesterolemic men (n = 12) and women (n = 26) with a mean total serum cholesterol (TC) of 245 + 29 mg/dL (mean + SD) followed a heart-healthy diet including 100g of one of three forms of almonds: roasted salted almonds, roasted almond butter or raw almonds for four weeks. Measurements of serum TC, triglycerides (TG), selected lipoproteins and blood pressure were taken at baseline and after four weeks.
Results: All three forms of almonds in the context of a heart-healthy diet significantly lowered low-density lipoprotein-cholesterol (LDL) from baseline to the completion of the study. Both raw and roasted almonds significantly lowered TC, whereas the decrease by almond butter (in a smaller cohort) did not reach statistical significance. High-density lipoprotein-cholesterol (HDL) did not significantly change with raw or roasted almonds but slightly increased with almond butter. At the end of the study, blood pressure did not change significantly from baseline values for any of the groups.
Conclusion: These results suggest that unblanched almonds—whether raw, dry roasted, or in roasted butter form—can play an effective role in cholesterol-lowering, plant-based diets.
Key words: serum cholesterol, raw almonds, roasted almonds, almond butter, nuts
|
INTRODUCTION
|
|---|
Almonds are a whole food rich in numerous beneficial nutritive and bioactive compounds like fatty acids, dietary fibers, micronutrients and phytochemicals. A growing number of human nutritional studies have revealed that almonds have a cholesterol-lowering effect [1–9]. The forms of almonds used in the studies published so far were either raw or roasted almonds or almond oil, but no published study has yet compared the effect of roasted to raw almonds on serum lipids and lipoproteins.
Several beneficial compounds found in almonds may be responsible for mediating the reduction in serum lipids observed in these studies. Almonds are rich in 
-9 fatty acids, which have demonstrated beneficial effects on lipoprotein profiles [1–9], and their proteins have an arginine-rich amino acid profile that is thought to be cardioprotective [10]. In addition, they are good sources of dietary fiber and phytochemicals, such as plant sterols, which have been shown to contribute to reduced risk of coronary heart disease [11,12]. Finally, almonds are rich in tocopherols, especially 
-tocopherol, and the latter has also demonstrated potent anti-atherogenic effects [12,13]. Recently, Hyson et al. [9] postulated that the beneficial effects of almonds might be mediated by the lipid fraction.
We had previously shown that an almond-based diet had favorable lipid-altering effects using whole and ground unblanched raw almonds [3–5]. Because salted roasted almonds are consumed in larger quantities than raw almonds, and some people prefer almond butter, it is important to know whether these forms of almonds have similar beneficial effects to raw almonds. In the present study, we compared the effect of consuming salted roasted almonds and roasted almond butter with raw almonds, in the context of a plant-based diet, on serum lipids and lipoproteins.
|
MATERIALS AND METHODS
|
|---|
Subjects
Forty-five hypercholesterolemic males and females, 32 to 74 years of age, were recruited in the San Francisco Bay Area through newspaper advertising, local advertisements and through mailings and phone calls to individuals known to have high serum cholesterol from our database. Applicants were screened to exclude individuals with major chronic diseases (diabetes, heart disease or cancer other than minor forms of skin cancer), allergy to nuts or diseases of the gastro-intestinal tract. In addition, individuals with body mass index (BMI) >25, or taking cholesterol-lowering medication, blood pressure medication or any drug known to affect serum cholesterol were also excluded. Inclusion criteria were total cholesterol (TC) >230 mg/dL but <310 mg/dL. Thirty-eight subjects completed the study, 26 females and 12 males. Their baseline values (mean + SD) were as follows: age, 61 + 11 years; body weight, 68 + 10 kg; TC, 245 + 29 mg/dL; LDL, 156 + 28 mg/dL; HDL, 58 + 15 mg/dL, and triglycerides (TG), 155 + 28 mg/dL. Drop outs after randomization were unrelated to the physiological effects of the almonds, and were mostly due to inability of the subjects to follow the study procedures.
The study protocol had been approved by an independent review committee and was explained to each subject who then signed an informed consent. Participants also signed the subject’s Bill of Rights and were told that they were free to withdraw from the study at any time.
Study Design
A randomized, controlled, parallel design was used to compare the three study diets over a four-week period after a two-week baseline period. Following recruitment, the subjects were randomized by serum cholesterol levels, gender and age, into three groups of 15 subjects each. At the beginning of the baseline period a serum sample was obtained to measure routine serum chemistries to confirm that the participants were healthy and had a TC >230 mg/dL and to randomize them. At the end of the baseline period, duplicate fasting serum samples were drawn, three to four days apart, to establish baseline values. After randomization, the participants were instructed not to change their diet during the baseline period and to keep detailed four-day diet records, which were reviewed in individual meetings with a study nutritionist during the second week of the baseline period and analyzed. Of the 38 subjects who completed the study, 14 subjects were in the raw almond group, 14 in the roasted almond group, and 10 in the almond butter group. Fewer subjects were in the almond butter group because some participants could not comply with incorporating almond butter in their busy life schedules.
Study Diets
After the two baseline weeks, the participants received two 50 g-packets of either raw almonds or roasted-salted almonds for the daily amount of 100g or a jar of unsalted roasted almond butter (Almond Board of California, Modesto, California), with careful instructions on the daily dose so that subjects would consume 100g per day. All almonds—raw, roasted and butter—were prepared from the same batch of almonds by one company. Their composition is shown in Table 1. The study nutritionists advised the subjects on how to replace some of the saturated fat or trans fatty acid products and animal proteins in the diet with the almonds or almond butter. In addition, the participants received rice cakes and brown rice (Lundberg Family Farms, Richvale, California) and lentils (Arrowhead Mills, Austin, Texas) to replace animal foods with plant foods. Consumption of eggs, meats, poultry and other high-cholesterol, high-saturated fat foods was limited to ensure that dietary cholesterol, saturated fat and trans fatty acid intakes would be kept to a minimum. The nutritionists also advised subjects on how to keep total caloric intake similar to the baseline diet, but allowing, as in previous studies in our Center, a possible increment in total fat intake from almonds. Participants were instructed to maintain their usual pattern of coffee, tea, alcohol and soft drink intake, their typical exercise routine and not to make any special efforts toward changing their weight or their food supplement routines.
Subjects met with the nutritionists after two weeks on the almond diets to discuss questions and ensure compliance, which was monitored with weekly food frequency records. Four-day food records were then repeated during the last week of the study.
Measurements
Duplicate fasting serum samples were drawn, three to four days apart, at baseline and at the end of the study period, measured and averaged. Two 10-mL blood samples were drawn into vacutainer tubes containing 15 mg sodium ethylenediaminetetraacetic acid (EDTA) from the antecubital vein after a 12-hour fast. Each vacutainer tube was then centrifuged for 10 to 15 minutes, and the specimens shipped by overnight air carrier under refrigeration. All samples were analyzed the following day. High-density lipoprotein cholesterol was separated from the serum by a precipitation procedure using dextran sulfate (50,000 MW) and magnesium chloride [14]. Total cholesterol in the remaining serum and in the separated HDL fraction was measured by an enzymatic procedure on the Spectrum Analyzer (Abbott Laboratories, North Chicago, IL). Triglycerides corrected for the glycerol blank were analyzed by an enzymatic UV procedure (Agent—a TG reagent, Abbott Laboratories, North Chicago, IL) on the same Spectrum Analyzer [15]. These analytical procedures were standardized and met the performance requirements of the Lipoprotein Standardization Program of the Centers for Disease Control (Atlanta, GA) and are traceable to the National Reference System for Cholesterol. Low-density lipoprotein cholesterol was estimated according to the Friedewald algorithm [16] using a value for very low density lipoprotein (VLDL) obtained by dividing TG in mmol/L by 2.22. The long-term interassay correlation coefficient (CV) during the study was 0.7% for TC, 1.5% for HDL and 1.1% for TG at all concentrations measured. The analyses were performed by Pacific Biometrics, Seattle, WA, a specialized lipid laboratory.
Triplicate research blood pressures were measured at baseline and at the end of the study. Subjects had a chance to rest and have a repeat systolic and diastolic pressure taken three times, approximately five minutes apart.
Statistical Analysis
The composition of the study diets was monitored with two sets of four-day diet records (three week days and one weekend day), during the baseline period and week 4 of the almond diets, and were computer-analyzed using the Nutritionist V software. Paired comparison t-tests (two-tailed) were used to evaluate diet and weight changes from baseline to week 4.
The effects of the study diets on serum lipoproteins and blood pressure were computed by analysis of variance for repeated measures (ANOVA). For each cholesterol measure (TC, LDL, VLDL and HDL) and TG, we conducted the analysis with form of almonds (raw/roasted/butter) as a between factor and time (before/after) as a within subjects measure. When a main effect was found, t-tests for each pair of before and after scores were computed to compare the amount of change in each group.
|
RESULTS
|
|---|
Changes in Diet and Weight
Compliance with the protocol and acceptance of the study diets, assessed by weekly food frequency questionnaires and verbal reports by subjects, was very good.
Table 2 shows the composition of the three almond diets—raw, roasted, butter—at baseline and at week 4. Caloric intake did not change significantly during the study period for any group. Total fat intake increased in all three groups but reached statistical significance only in the roasted almond group (p < 0.001). Both monounsaturated fatty acid (MUFA) and polyunsaturated fatty acid (PUFA) intakes increased significantly in the raw, roasted and almond butter groups (MUFA: p < 0.02, p < 0.001 and p < 0.005; PUFA: p < 0.02, p < 0.001 and p < 0.05, respectively). There was a significant decrease in saturated fatty acid (SFA) intake in the raw almond group (p < 0.05) and a non significant decrease in the roasted almond group (p < 0.2). Dietary fiber increased significantly only in the raw and roasted almond groups (p < 0.005, p < 0.02, respectively). Dietary cholesterol intake decreased in all three groups but reached statistical significance only in the raw almond group (p < 0.02). There were no significant changes in body weight during the study period in any of the groups.
Changes in Lipids, Lipoproteins and Blood Pressure
The changes in lipids and lipoproteins for each group between baseline and week 4 are reported in Table 3. The results of repeated measures ANOVA show a main effect for time, with no main effect for form and no time by form interaction effect for TC and LDL. This means that there was a significant decrease in mean TC and LDL (p < 0.0001) over the course of the study for each form of almond. No statistically significant time effect was observed for HDL, VLDL or TG.
The results of the paired t-tests show highly significant differences between baseline and week 4 in TC and LDL levels for the raw and roasted almond groups, and in LDL, but not TC, for the almond butter group (Table 2). In addition, they reveal that the before vs. after differences are larger for LDL (-12% for raw almonds and -7% for both roasted almonds and almond butter) than for TC (-7% for raw almonds and -5% for both roasted almonds and almond butter). Furthermore, the largest decreases in both TC and LDL occurred in the raw almond group (-7% and -12%, respectively). Although no main effect for time was detected for HDL, VLDL or TG, there was a small increase in HDL (+8%) and a small reduction in VLDL (-15%) and TG (-18%) in the almond butter group.
The baseline systolic blood pressures (mean + SD) for the raw almond, roasted almond and almond butter groups were, respectively, 118 + 14, 115 + 18 and 120 + 10 mmHg. Baseline diastolic pressures for the three groups were 68 + 7, 71 + 9, and 74 + 8 mmHg, respectively. No significant changes were found in any of the three groups in either systolic (117 + 11, 116 + 18 and 122 + 16 mmHg, respectively) or diastolic (67 + 5, 72 + 10, 74 + 8 mmHg, respectively) blood pressure at the completion of the study period.
|
DISCUSSION
|
|---|
The results of this study show that plant-based diets rich in raw almonds, roasted almonds and almond butter, produce similar reductions in TC and LDL while leaving HDL and TG unchanged. They also show no changes in blood pressure for the duration of the study, suggesting that consumption of lightly salted roasted almonds did not adversely affect blood pressure. This study confirms the results of our previous studies using raw whole and ground almonds [3–5]. In those studies we observed slightly higher reductions in TC (10% to 12%) and LDL (12% to 14%), possibly because half of the almonds were provided in ground form and may have had a higher digestibility. A recent meta-analysis [1] reviewed seven studies evaluating the effect of almonds on serum lipids and lipoproteins and showed reductions similar to those observed in our study in TC (7%) and LDL (10%) in individuals with baseline cholesterol >250 mg/dL. Kendall and co-workers also found that 50–100g of almonds/day lowered TC by 
6% and LDL by 9% in hyperlipidemic men and women [2]. Hyson and co-workers showed that whole almonds and almond oil induce similar reductions in TC (4%) and LDL (6%) and concluded that the cholesterol-lowering effect of almonds is due mainly to constituents of the lipid fraction [9]. However, the amount of almonds fed in this study (66 g/day) was lower than in the present study and may not have been sufficient for the non-fat associated constituents (dietary fiber, arginine) to exert a greater cholesterol-lowering effect in comparison to the almond oil. In addition, it needs to be noted that the present, as well as our previous investigations [3–5], studied a plant-based diet rich in almonds, whereas Hyson et al. [9] replaced 50% of the participants’ usual dietary intake with almonds or almond oil.
Whereas the ANOVA results show a similar significant lowering of TC and LDL for all forms of almonds, the results of the paired comparisons reveal a significant decrease in TC for all three types of almonds and a significant decrease in LDL for raw and roasted almonds but not for almond butter. This result may be explained by the lower sensitivity of the paired t-test compared with repeated measures ANOVA and by the lower number of subjects in the almond butter group (10 vs. 14 in the other two groups). In addition, TC may not have decreased significantly because of the slight increase in HDL observed in the almond butter group (+8%). This increase in HDL, although small, was observed only in the almond butter group and was not significant. Unfortunately, the smaller number of subjects in this group limits the inferences that can be drawn on any special attribute of almond butter. Any differences in almond butter versus other forms of almonds would require further investigation in larger-scale studies.
Although the largest reductions in cholesterol were observed with raw almonds, the magnitude of the differences between raw almonds and roasted almonds or almond butter is small (-7% vs. -5% for TC and -12%, vs. -7% for LDL). The higher palatability of roasted almonds and almond butter versus raw almonds may lead to more consistent consumption. Interestingly, most participants reported that almonds would become a more regular part of their diet, but they would prefer to mix the three types of almonds.
The analysis of the diet compositions shows that the participants incurred similar changes in the percentage of macronutrient intakes (Table 2). They increased their fat intake from 30%–36% to 44%–45%, a higher level than the one recommended by the National Cholesterol Education Program Step 1 and 2 diets (up to 35%) [17]. This increase, although significant only for the roasted almond group, was mainly due to increases in MUFA and PUFA (significant in all groups) brought about by the consumption of almonds and was similar across the three almond groups. MUFA increased from 10%–12% (baseline) to 21%–22% of energy intake at the end of the study, and PUFA went from 6% to 9%–11% of caloric intake. In addition, SFA intake decreased similarly in the raw and roasted almond groups, from 11%–12% to 7%–8% of energy intake, but did not change in the almond butter group (9%). Dietary cholesterol was reduced significantly in the raw almond group (-54%) and to a lesser extent in the roasted almond (-32%) and almond butter (-41%) groups. Dietary fiber intake was significantly increased in the raw and roasted almond groups, whereas it remained unchanged in the almond butter group. It needs to be noted that the almond butter group incurred dietary changes that did not always follow the trend of the other two groups, mainly in SFA and fiber, which did not change from the baseline diet. This group also comprised fewer subjects than the other two groups. As mentioned above, the results observed in the almond butter group are just preliminary, but worth mentioning, and would need to be replicated in larger-scale investigations.
In conclusion, this study suggests that almonds—either raw, or moderately roasted or in butter form—can be a desirable part of cholesterol-lowering plant-based diets and that replacing some animal foods with almonds appears to be a desirable step in heart disease prevention.
|
ACKNOWLEDGMENTS
|
|---|
This study was made possible by an unrestricted grant from the Almond Board of California (Modesto, CA). We also wish to acknowledge the assistance of Dr. Karen Lapsley of the Almond Board and Dr. Sam Cunningham of Blue Diamond (Sacramento, CA) for their advice and suggestions.
|
FOOTNOTES
|
|---|
The study was supported by an unrestricted grant by the Almond Board of California.
Received July 1, 2002.
Accepted September 27, 2002.
|
REFERENCES
|
|---|
- Fulgoni III, VL, Abbey M, Davis P, Hyson D, Jenkins D, Lovejoy J, Most M, Sabate J, Spiller GA: Almonds significantly lower elevated blood cholesterol and LDL-cholesterol but not HDL-cholesterol or triglycerides in human subjects: results of a meta-analysis [Abstract]. Proceedings of The Fourth International Congress on Vegetarian Nutrition, April 8–14,2002
.
- Kendall CWC, Jenkins DJA, Marchie A, Parker T, Connelly PW, Spiller GA: Dose response to almonds in hyperlipidemia: a randomized controlled cross-over trial. Am J Clin Nutr75
:384S
,2002
.
- Spiller GA, Gates JE, Jenkins DA, Bosello O, Nichols S, Cragen LN: Effect of two foods high in monounsaturated fat on plasma cholesterol and lipoproteins in adult humans [Abstract]. Am J Clin Nutr51
:524
,1990
.
- Spiller GA, Jenkins DA, Cragen LN, Gates JE, Bosello O, Berra K, Rudd C, Stevenson M, Superko R: Effect of a diet high in monounsaturated fat from almonds on plasma cholesterol and lipoproteins. J Am Coll Nutr11
:126
–130,1992
.[Abstract]
- Spiller GA, Jenkins DA, Bosello O, Gates JE, Cragen LN, Bruce B: Nuts and plasma lipids: an almond-based diet lowers LDL-C while preserving HDL-C. J Am Coll Nutr17
:285
–290,1998
.[Abstract/Free Full Text]
- Berry EM, Eisenberg S, Haratz D, Friedlander Y, Norman Y, Kaufmann N, Stein Y: Effects of diets rich in monounsaturated fatty acids on plasma lipoproteins-the Jerusalem Nutrition Study. II Monounsaturated fatty acids vs carbohydrates. Am J Clin Nutr56
:394
–403,1992
.[Abstract/Free Full Text]
- Abbey M, Noakes M, Belling GB, Nestel PJ: Partial replacement of saturated fatty acids with almonds or walnuts lowers total plasma cholesterol and low-density-lipoprotein cholesterol. Am J Clin Nutr59
:995
–999,1994
.[Abstract/Free Full Text]
- Sabate J, Fraser G: The probable role of nuts in preventing coronary heart disease. Pri Card19
:65
–72,1993
.
- Hyson DA, Schneeman BO, Davis PA: Almonds and almond oil have similar effects on plasma lipids and LDL oxidation in healthy men and women. J Nutr132
:703
–707,2002
.[Abstract/Free Full Text]
- Kritchevsky D, Tepper SA, Czarnecki SK, Klurfeld DM: Atherogenicity of animal and vegetable protein—influence of the lysine to arginine ratio. Atherosclerosis41
:429
–431,1982
.[Medline]
- Farquhar JW: Plant sterols: their biological effects in humans. In Spiller G (ed): "Handbook of Lipids in Human Nutrition." Boca Raton: CRC Press, pp101
–105,1996
.
- US Department of Agriculture: "Composition of Foods: Nuts and Seed Products," (Agricultural Handbook No. 8–12). Washington, DC: United States Department of Agriculture,1984
.
- National Research Council: "Recommended Dietary Allowances," 10th ed. Washington, DC: National Academy Press,1989
.
- Warnick GR, Benderson J, Albers JJ: Dextran sulfate-Mg2+ precipitation procedure for quantitation of high-density-lipoprotein cholesterol. Clin Chem28
:1379
–1388,1982
.[Free Full Text]
- Warnick G: Enzymatic methods for quantification of lipoprotein lipids. In Albers JJ, Segrest JP (eds): "Methods in Enzymology." Orlando, FL: Academic Press, pp101
–123,1986
.
- Friedewald WT, Levy RI, Fredrickson DS: Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem18
:499
–502,1972
.[Abstract]
- Van Horn L, Ernst N: A summary of the science supporting the new National Cholesterol Education Program dietary recommendations: what dietitians should know. J Am Diet Ass101
:1148
–1154,2001
.[Medline]
TOP
ABSTRACT
INTRODUCTION
