Effects of Almonds and Almonds with Dark Chocolate and Cocoa on Cardiovascular Risk Factors

Heart disease claimed approximately 610,000 lives in 2015 and remains the leading cause of death in the United States. In 2010, suboptimal diet quality resulted in 678,000 deaths, some of which was attributed to low nut consumption. Previous research has found an inverse relationship between nut consumption and risk of coronary heart disease and all-cause mortality. Furthermore, incorporating mixed nuts with a Mediterranean diet decreased the risk of major cardiovascular events by 28% compared to consuming a diet low in fats and nuts.

Consuming nuts or cocoa (Theobroma cacao, Malvaceae) products can improve the markers of coronary heart disease, according to the authors. Almonds (Prunus dulcis, Rosaceae) are high in unsaturated fatty acids; provide an excellent source of ɑ-tocopherol; and contain minerals, plant protein, fiber, and phytosterols. Dark chocolate is rich in flavonoids, which are associated with a lower risk for coronary heart disease. The authors conducted a randomized, 4-period, crossover, controlled feeding trial to investigate the individual and combined effects of dark chocolate, cocoa, and almonds on lipid, lipoprotein, and apolipoprotein concentrations; vascular health; and oxidative stress in overweight and obese subjects.

Overweight and obese participants with a body mass index (BMI) of 25-40 kg/m2, aged 30 to 70 years, and with a low-density lipoprotein cholesterol (LDL-C) level of 105-194 mg/dL for men and 98-190 mg/dL for women were eligible for inclusion. Participants were excluded if they smoked, had high blood pressure or had a history of diabetes, heart attack, stroke, liver, kidney, thyroid, or inflammatory gastrointestinal disease. Recruitment was accomplished from March 2013 to July 2015 through flyers posted on university bulletin boards, local newspaper advertisements, and email lists from Pennsylvania State University in University Park, Pennsylvania.

Screening visits at Penn State Clinical Research Center were attended by 149 people. Of those, 48 met the study inclusion criteria and were randomized to receive one of four diets for four weeks. After a 2-week washout period, the subjects began a subsequent test diet.

At baseline and at the end of each diet period, the subjects fasted overnight and visited the research center on two consecutive days for measurements of weight, waist circumference, blood pressure, vascular endothelial function, and for blood draws.

The Penn State Metabolic Diet Study Center provided all meals and snacks for the study. Although the outcome assessors and study coordinators were blinded to the diet, the subjects were not.

The diets differed in the presence or absence of the treatment foods, which included 42.5 g raw almonds, 18 g natural cocoa powder, and 43 g dark chocolate. [Note: While the text does not provide manufacturer information for the treatment foods, the Hershey Company (Hershey, Pennsylvania) and the Almond Board of California (Modesto, California) funded the study. It is likely that the treatment foods came from these groups.] The test diets, which used the same 6-day cycle menu, were as follows:

  • Average American diet (AAD) included the same foods as the other diets, but with no almonds, dark chocolate, or cocoa, which were isocalorically substituted with butter, cheese, and refined grains.
  • Almond diet (ALD) included almonds and was lower in saturated fatty acids (8% compared with 13%) and higher in monounsaturated fatty acids (16% compared with 13%) and polyunsaturated fatty acids (9% compared with 7%) than the AAD.
  • Chocolate diet (CHOC) included the same foods as the ALD but with natural cocoa powder and dark chocolate isocalorically substituted for the almonds.
  • CHOC+ALD contained natural cocoa powder, dark chocolate, and almonds, and less butter, cheese, and refined grains. This diet had the highest fiber content of the diets.

The following values were assessed: total cholesterol (TC), triglycerides, high-density lipoprotein cholesterol (HDL-C), LDL-C, non-HDL-C, lipoprotein(a), intermediate-density lipoprotein cholesterol (IDL-C), very-low-density lipoprotein cholesterol (VLDL-C), glucose, plasma fasting insulin, serum high-sensitivity C-reactive protein, plasma nitric oxide, plasma flavonoids, phenolic acids, tocopherols, LDL oxidation, urinary F-isoprostanes, flow-mediated dilation (FMD) of the brachial artery, average blood flow velocity, maximum flow velocity, velocity time integral, and blood pressure.

Study dropouts totaled seven after the baseline visit, eight after the first diet period, one after the second diet period, and one after the third diet period. Reasons for dropping out included noncompliance (n=2), dislike of the foods (n=5), time restraints (n=3), relocation (n=1), and personal reasons (n=6). The 31 subjects who completed all four diet periods were included in the final analyses.

At baseline, the subjects were overweight and had elevated TC and LDL-C but were otherwise healthy.

After the intervention, subjects following the ALD had lower levels of TC (P=0.004), non-HDL-C (P=0.006), and LDL-C (P=0.003) compared with those following the AAD; however, no differences were observed among the diets for HDL-CIDL-C, lipoprotein(a), VLDL-C, and triglycerides. For lipoprotein subclasses, no treatment effects were observed for large HDL, small HDL, or VLDL-C; however, concentrations of large buoyant LDL particles were lower after the ALD compared with the AAD (P=0.04). After the CHOC-ALD compared with the AAD, the concentration of small dense LDL particles was lower (P=0.04).

After the CHOC+ALD, the concentrations of apolipoprotein B (ApoB) were lower compared with the AAD (P=0.02); however, no treatment effect was observed for apolipoprotein A1. The CHOC+ALD resulted in lower ratios of ApoB/apolipoprotein A1 (P=0.02) and TC/HDL-C (P=0.02) compared with the AAD.

Higher levels of fasting glucose were seen after the CHOC and CHOC+ALD compared with the AAD (P<0.05 for both); however, no treatment effects were seen for other markers of blood sugar control.

When comparing the percentage changes from baseline in each group, the authors reported greater reductions in TC (P=0.004) and in non-HDL-C (P=0.009) after the ALD compared with the AAD (P=0.004). For LDL-C, the percentage reductions were greater after the ALD (P=0.004) and after the CHOC-ALD (P=0.04) compared with the AAD. The CHOC resulted in a higher percentage increase in glucose compared with the ALD (P=0.003); however, no significant differences were observed for the other markers of blood sugar control or inflammation.

Systolic blood pressure significantly decreased from baseline after the AAD and ALD (P<0.05 for both) but not after the CHOC diet and CHOC+ALD. Diastolic blood pressure significantly decreased from baseline after the ALD (P=0.04); however, the percentage changes were not significantly different among the treatments. The brachial artery diameter, measured before cuff occlusion and at peak dilation after cuff release, was larger after the CHOC than after the ALD (P<0.01 for both). No differences in FMDs were observed between the CHOC and ALD. No treatment effects were observed for LDL oxidation or urinary 8-isoprostanes.

“The mechanisms by which almonds may affect lipids and lipoproteins have not been fully understood, despite some plausible hypotheses,” write the authors. The high fatty acid content of almonds likely results in reduced LDL-C levels, as seen in this study. Consistent with this current study, a previous meta-analysis of 19 randomized controlled trials reported a neutral effect of cocoa and dark chocolate on LDL-C.

The glycemic control markers did not improve after any of the diets compared with the AAD. In previous studies, consuming dark chocolate or cocoa improved glycemic control parameters by decreasing HOMA-IR and increasing insulin sensitivity. “This discrepancy may reflect different study designs and test methods used,” state the authors.

According to the authors, the lack of effects on blood pressure and vascular function, which is inconsistent with earlier studies, may be due to the dose of flavanols used in the study, which may be too low to change vascular parameters.

This study is limited by its 35% dropout rate, the lack of control for multiplicity, the lack of objective evidence to support a high participant compliance assessed by daily food logs, and the inability to maintain baseline weights of the subjects during the study period. Because butter, cheese, and refined grains were used as comparison foods, generalizing the results to populations in which these foods are not widely consumed may be limited.

Consumption of the almonds and almonds with dark chocolate and cocoa resulted in beneficial effects on lipid, lipoprotein, and apolipoprotein profiles but had no effect on vascular health and oxidative stress. The improvements in lipid and lipoprotein profiles could be expected to decrease coronary heart disease risk.

This study was funded by The Hershey Company and the Almond Board of California and supported by the Penn State Clinical and Translational Research Institute, Pennsylvania State University Clinical and Translational Science Award, and National Institutes of Health/National Center for Advancing Translational Sciences.


Lee Y, Berryman CE, West SG, et al. Effects of dark chocolate and almonds on cardiovascular risk factors in overweight and obese individuals: A randomized controlled-feeding trial. J Am Heart Assoc. November 2017;6(12): e005162. doi: 10.1161/JAHA.116.005162.