Chocolate—Guilty Pleasure or Healthy Supplement?
Dark chocolate and other cocoa products are popular in the population as a whole, but their overall health benefit remains controversial. Observations from the Kuna Indian population have shown an impressive cardiovascular health benefit from cocoa. For various reasons, this benefit has not been as robust as in other populations. Additionally, several mechanisms have been proposed that might confer cocoa's possible health benefit, but no consensus has been reached on cocoa's physiologic role in promoting cardiovascular health. Flavanols, as well as theobromine, may contribute to enhancements in endothelial function and subsequent improvements in various contributors to cardiovascular disease (CVD) including hypertension, platelet aggregation and adhesion, insulin resistance, and hypercholesterolemia. While the benefits of cocoa may be altered at the various stages of growth, development, and production, it appears that for many people “healthy” dark chocolate may, indeed, provide a pleasurable role in CVD risk reduction. The objectives of this review are to discuss the associations of cocoa with decreased blood pressure and improved CVD risk, to describe the possible mechanisms for these potential benefits, and to highlight considerations for the use of cocoa as a dietary supplement.
Patients are frequently on the lookout for a pleasurable way to approach their health. With this in mind, dark chocolate and cocoa products have garnered attention as a dietary supplement and approach to decrease blood pressure (BP) and modify other cardiovascular disease (CVD) risk factors. An extract from the Theobroma cacao tree, cocoa is a rich source of plant polyphenols, a heterogeneous group of molecules found primarily in fruits and vegetables. Cocoa is especially rich in flavanols, a polyphenol subtype proposed as the mediator for cardiovascular benefits. Flavanols are also found in other plant-based foods (Table 1).[1, 3] While the contributions of the various components of these foods continue to evolve, the flavanol content appears to augment the cardiovascular benefits. Regular dietary intake of plant-derived foods and beverages is inversely associated with the risk of CVD among the general population.
Table 1. Flavanol Concentrations Found in Food
Interest in the effect of cocoa on CVD began with observations among the Kuna Indian population in the San Blas Islands of Panama. This group had distinctively low rates of hypertension and CVD, coupled with an absence of the age-related increases in BP observed in other populations.[1, 5] Environmental, rather than genetic, factors appeared to confer this protective role.[1, 5] Unique to this population was their cocoa intake. On average, traditional island-dwelling Kuna Indians consume approximately four 8-ounce cups of unprocessed cocoa beverages per day.[5, 6] Even in patients older than 65 years, the mean BP was 110/70 mm Hg.[5, 6] Conversely, migrant Kuna Indians consume up to 10 times less cocoa and experience a BP increase with age and hypertension prevalence comparable to Western populations.[1, 5, 6] Salt consumption among the island-dwelling population was equivalent or greater compared with migrant Indians and there were no significant differences in body mass index (BMI).[1, 5] Independent of cocoa, sodium, and weight, other factors may influence the observed changes in BP. Physical activity, smoking status, stress level, and diet may also be related to the CVD risk increase experienced with the migrant Kuna population.
For many, dietary changes and other lifestyle modifications are effective for the prevention and treatment of hypertension and are critical for reduction of CVD. In this review, we discuss the clinically relevant effects of cocoa, focusing on possible mechanisms involved in the cardiovascular response to cocoa and the potential implications associated with consumption.
Structurally, flavanols exist as low-molecular-weight monomeric compounds, such as epicatechin, which have effects on the vascular endothelium, or as complex higher-molecular-weight oligomeric and polymeric compounds (eg, procyanidins), which are less vasoactive.[2, 8] The profile and ratio of the various flavanols in cocoa products vary considerably and can be altered at many stages of growth, development, and production.[1, 7, 8]
Flavanol content in cocoa products is dependent on the crop cultivar type, post-harvest handling practices, and manufacturer processing techniques. Fresh and fermented cocoa beans contain approximately 10% flavanols (100 mg/g) prior to processing, while the cocoa powder consumed by the Kuna Indians contains about 3.6% flavanols. In contrast, cocoa-rich dark chocolate contains approximately 0.5% flavanols. Milk and white chocolate have lower flavanol content or even flavanol-free composition, respectively. Because flavanols are bitter and often considered unpalatable, processes are used for flavor enrichment. The cocoa powder consumed by the Kuna Indians would be unpleasurable to most individuals in the Western population. Fermentation, roasting up to 120°C, and “dutching” (ie, alkalizing), as well as the addition of sugar, milk, vanilla, and emulsifiers, are techniques used to improve cocoa's palatability.[7, 9] These modifications, however, can practically eliminate the presence of flavanols in finished products, and thus may abrogate the potential CVD benefits of cocoa. The percentage of cocoa is not a reliable indicator of the flavanol content present in a given product. As a result, a 70% cocoa bar from one company may contain an entirely different flavanol composition than a 70% cocoa product from another supplier.
Mechanisms and Effects on CVD Risk Factors
Observational studies support the association between high cocoa intake and reduced CVD.[10, 11] Based on the Kuna Indian observations, investigational trials were designed to further define the cardiovascular and antihypertensive effects of cocoa consumption. Although a variety of mechanisms have been proposed, many of the findings involve improvements in endothelial function. Prospective studies have demonstrated an association between endothelial dysfunction and increased risk of CVD. Endothelial function may improve following the consumption of flavanol-rich cocoa, with subsequent improvement of various contributors to CVD including hypertension, platelet aggregation and adhesion, insulin resistance, and hypercholesterolemia.
For the purpose of this review, we searched for published studies from 2003 to 2013 addressing the use of cocoa in the management of CVD risk factors. Using PubMed, we searched by the following MeSH terms individually and in combination: “cocoa,” “dark chocolate,” “flavanols,” “hypertension,” “human,” “platelets,” “lipids,” “cardiovascular disease,” and “insulin.” We also reviewed the Natural Medicine Comprehensive Database. Table 2 highlights examples of studies designed to assess the effects of cocoa on cardiovascular risk factors.
Table 2. Study Examples: Cocoa and Cardiovascular Risk Factors
|Taubert||Randomized, single-blind, parallel groupa/18 wk||44||DC (6.3 g, 30 mg flavanol) vs WC (5.6 g, flavanol-free)||↑ NO poolband ↓ in SBPb and DBPb|
|Heiss||Randomized, double-blind, crossover/ 2 h, 2 d total||11||Cocoa drink 100 mL with high (176–185 mg) vs low (<11 mg) flavanol content||↑ FMDb and circulating NO poolb|
|Hermann||Randomized, parallel group/2 h, 24 h total||20||DC (40 g, 74% cocoa, Nestlé Noir Intense) vs WC (40 g, 4% cocoa, Nestlé Galak)||↑ FMDb and improvement in antioxidant statusb and platelet functionb|
|Persson||Open-label/3 h||16||DC (75 g, 72% cocoa)||↑ NO pool and inhibition of ACE activityb|
|Taubert||Randomized, single-blind, crossovera,c/2 wk||13||DC (100 g, 500 mg flavanol) vs WC (90 g, flavanol-free)||↓ SBPb and DBPb|
|Grassi||Randomized, single-blind, crossovera,c/15 d||20||DC (100 g, 88 mg flavanol) vs WC (90 g, flavanol-free)||↑ FMDb and insulin sensitivityb and ↓ SBP,b DBP,b LDL,b and total cholesterolb|
|Muniyappa||Randomized, double-blind, placebo-controlled, crossovera,c/2 wk||20||Cocoa drink 150 mL twice daily with high (~900 mg/d) vs low (~28 mg/d) flavanol content||No reduction in BP or improvement in insulin resistance|
|Engler||Randomized, double-blind, placebo-controlled/2 wk||21||High-flavonoid DC (46 g, DOVE) vs low-flavonoid DC (46 g)||↑ FMDb and no reduction in BP|
|Flammer||Randomized, double-blind, placebo-controlled/2 h and 4 wk||20||DC (40 g, 70% cocoa, Nestlé Noir Intense) vs placebo chocolate (28.4 g, flavanol-free)||↑ FMD (acute and chronic effect)b and ↓ in platelet adhesion (acute effect)b|
|Innes||Randomized, single-blind/4 h||30||DC (100 g, 75% cocoa) vs milk chocolate (100 g, 20% cocoa) vs WC (100 g, flavanol-free)||Inhibition of collagen-induced platelet aggregationb|
|Ostertag||Randomized, single-blind/2 and 6 h||42||Flavanol-rich DC (60 g) vs standard DC (60 g) vs WC (60 g)|| |
Men: ↓ platelet activity and aggregationb
Women: ↓ aggregationb
|Grassi||Randomized, single-blind, crossovera,c/15 d||19||DC (100 g, flavanol-rich) vs WC (100 g, flavanol-free)||↑ FMD,b insulin sensitivity,b and β-cell functionb and ↓ SBP and DBPb|
|Desideri||Randomized, double-blind, parallel groupa/8 wk||90||Cocoa drink with high (~990 mg) vs mid (~520 mg) vs low (~45 mg) flavanol content|| |
↑ Cognitive function and ↓ insulin resistance
bSeen with mid and high flavanol groups
|Mursa||Open-labela/3 wk||45||Polyphenol-enriched DC (75 g) vs standard DC (75 g) vs WC (75 g)|| |
bSeen by both DC groups
|Baba||Randomized, controlled/12 wk||25||Cocoa group (26 g cocoa powder +12 g sugar) vs control group (12 g sugar)||↓ in LDL-C, ↑ in HDL-Cb, and suppression of oxidized LDL-Cb|
|Neufingerl||Randomized, double-blind, placebo-controlled, parallel groupd/4 wk||152||Drink 200 mL with cocoa (150 mg TB and 325 mg flavanols), pure TB (850 mg TB), cocoa with added TB (1000 mg TB and 325 mg flavanols), or placebo|| |
↑ HDL-C,b ↑ apolipoprotein AI,b ↓ apolipoprotein B,b and ↓ LDL-Cb
bSeen with the pure TB group only
While intriguing, significant limitations apply to all of the cocoa studies reviewed. The paucity of available data may also indicate a publication bias in favor of positive findings. White chocolate was often used as a control, thus preventing participant blinding. Flavanol content of test foods, when reported, varies considerably among studies, complicating interpretations and comparisons. Additionally, most studies had a relatively small sample size, were of short duration, and measured only surrogate markers of CVD risk. This limits the generalizability of the results as well as translation to any long-term clinical implications or benefits. While the following observations are provoking, confirmation is needed in well-controlled, well-designed studies.
The relationship between BP and CVD risk is continuous and independent of other risk factors, and small reductions in BP can lead to substantial decreases in CVD. One suggested pathway for the benefits of cocoa on BP and CVD involves endothelial nitric oxide (NO). Cocoa intake increases NO generation. This process may be triggered by upregulation of endothelial NO synthase (eNOS), which synthesizes NO from l-arginine. Enhancement of NO by cocoa leads to vasodilation and BP reduction, as well as prevention of leukocyte adhesion and migration, smooth muscle cell proliferation, and platelet adhesion and aggregation.[3, 13] In patients with cardiovascular risk factors, a cocoa drink high in flavanol content (176–185 mg) rapidly increased circulating levels of bioactive NO by more than a third. In turn, flow-mediated vasodilation, a commonly used marker to quantify endothelial function, was increased. Furthermore, this flavanol-related increase in vasodilation and improvement in endothelial function can be reversed by L-NG-monomethyl arginine, a competitive eNOS inhibitor, resulting in a significant reduction in blood flow.[3, 14] It can be argued, therefore, that the reduced BP and cardiovascular risk in individuals who consume flavanol-rich foods is the result of the favorable effect on eNOS activity.
Cocoa flavanols also exert antioxidant effects. Consumption of 40 g of commercially available dark chocolate (74% cocoa) significantly improves plasma antioxidant status 2 hours after ingestion. It is likely that, in addition to eNOS induction and NO concentration elevation, a reduction in oxidative stress occurs from cocoa ingestion. NO breakdown by reactive oxidant species is reduced, which contributes to enhanced endothelial function, especially under conditions with a high oxidative stress burden, such as smoking. In addition, antioxidants may prevent NO transformation into peroxynitrite, a powerful oxidant, and thus protect against vascular damage.
Another mechanism by which flavanols may lower BP is by inhibition of angiotensin-converting enzyme (ACE). ACE inhibition is a highly utilized and evidence-based therapeutic approach to the treatment of hypertension. In vitro and, more recently, in vivo studies support the occurrence of ACE inhibition by flavanols.[17, 18] In healthy volunteers, the average inhibition of ACE activity was 18% 3 hours after intake of 75 g of dark chocolate, a level consistent with the magnitude achieved with the ACE inhibitor class of medications.
Several randomized controlled trials have investigated the effects of dark chocolate on BP in patients with hypertension. Trials published in 2003 and 2005 used similar designs with patients randomized to receive either 100 g of dark chocolate or 90 g of flavanol-free white chocolate daily for 2 weeks. After a 7-day washout period, patients were crossed over to the alternative treatment. Dark chocolate consumption decreased systolic BP (SBP) and diastolic BP (DBP; mean±standard deviation) by 5.1±2.4 mm Hg and 1.8±2.0 mm Hg, respectively, in the 2003 trial and 11.9±7.7 mm Hg and 8.5±5.0 mm Hg in the 2005 trial.[19, 20] In both studies, the reduction in BP following dark chocolate consumption reached statistical significance. White chocolate consumption did not reduce BP.[19, 20] In 2007, the effects of lower doses of dark chocolate (6.3 g of chocolate containing 30 mg of flavanols) and white chocolate (5.6 g of flavanol-free chocolate) were assessed over 18 weeks. Compared with baseline, SBP was significantly reduced by 2.9±1.6 mm Hg and DBP by 1.9±1.0 mm Hg in the dark chocolate group. Again, white chocolate had no significant effect. These findings suggest that habitual intake of low doses of dark chocolate may have sustained antihypertensive effects. Other studies, however, have shown no effect of dark chocolate on BP. One study found that consumption of a flavanol-rich cocoa drink for 2 weeks did not significantly reduce BP in patients with primary hypertension. Another randomized controlled trial found no significant changes in BP following flavanol-rich cocoa intake in healthy adults.
A 2012 meta-analysis of 20 studies revealed a statistically significant BP-reducing effect of flavanol-rich cocoa products compared with controls, with mean reductions of 2.77 mm Hg and 2.20 mm Hg in SBP and DBP, respectively. This meta-analysis provides additional support that dark chocolate may have a small but clinically significant BP-lowering effect. Generally, a 3 mm Hg reduction in SBP is estimated to reduce the relative risk of stroke mortality by 8%, coronary artery disease mortality by 5%, and all-cause mortality by 4%.
BP is clearly influenced by diet as documented by the Dietary Approaches to Stop Hypertension (DASH) eating plan. Adoption of this type of diet rich in fruits, vegetables, and low-fat dairy products is associated with an 8 mm Hg to 14 mm Hg reduction in SBP and ≥5 mm Hg decrease in DBP. Additionally, certain dietary supplements (coenzyme Q10, fish oil, garlic, and vitamin C) have evidence of antihypertensive effects. The BP reduction following cocoa intake is not as robust as that exhibited by the DASH diet or these dietary supplements. Although cocoa may be consumed as a part of a healthy DASH diet–type approach to BP management, insufficient evidence exists to recommend it as a supplement for additional antihypertensive effects.
Platelet dysfunction is another hallmark of CVD, and the ability of flavanols to reduce platelet activity might partly explain the beneficial effects of these compounds. Cocoa reduces adenosine diphosphate/collagen-activated, platelet-related primary hemostasis within hours of ingestion. As a result, platelet aggregation and adhesion are diminished. Flavanols exert additional antiplatelet effects by reducing glycoprotein IIb/IIIa expression.[3, 26] A significant reduction in platelet adhesion was observed in young smokers shortly after consuming flavanol-rich dark chocolate; however, this effect was not sustained after 4 weeks of daily consumption. Platelet aggregation was also reduced in participants who consumed 100 g of dark chocolate, with no effect after ingestion of white or milk chocolate. Sex-specific differences in platelet response have been reported. Men experienced a significant reduction in both platelet activity and aggregation while women had only a reduction in platelet aggregation. This study concluded that high-flavanol dark chocolate has a positive effect on platelet function in both sexes, but the benefits may be greater in men.
Impairment of insulin-stimulated production of NO from the vascular endothelium contributes to metabolic insulin resistance. The resulting decreases in insulin-stimulated blood flow reduces the delivery of insulin and metabolic substrates to skeletal muscle. Thus, by ameliorating NO bioavailability, flavanols may decrease insulin resistance. Insulin resistance has been reported to be reduced in patients with hypertension after a 15-day diet that included 100 g of flavanol-rich cocoa daily.
Alteration in the insulin-signaling pathway has been suggested as a contributor to cognitive dysfunction because of insulin's pivotal role in modulating brain function. Chronic dysglycemia is thought to further contribute to cognitive dysfunction by promoting the development of cerebral microvascular disease and inflammation. Elderly individuals who consumed at least 520 mg of flavanols daily demonstrated significant improvement in cognitive performance that was associated with a reduction in insulin resistance. Compounds such as flavanol-rich cocoa that improve endothelial function concurrently decrease insulin resistance, enhance insulin sensitivity, and increase β-cell function.[3, 20, 21] Dietary inclusion of flavanols could be one approach to maintain and improve cardiovascular health and cognitive function.
Increased concentrations of low-density lipoprotein cholesterol (LDL-C) are associated with the development of atherosclerosis, while a negative correlation exists between increased high-density lipoprotein cholesterol (HDL-C) and CVD. Evidence also indicates that oxidized LDL-C progresses to fatty streak formation and has a pathogenic role in atherosclerosis. Improved LDL-C and HDL-C concentrations and LDL-C resistance to oxidation may contribute to a reduction of atherosclerosis. Flavanol substances lower LDL-C by inhibiting cholesterol absorption in the digestive tract, inhibiting LDL-C biosynthesis, suppressing hepatic secretion of apolipoprotein B, and increasing hepatic expression of LDL-C receptors. The pathways through which flavanols elevate plasma HDL-C concentrations remain unclear, but a potential mechanism involves inhibition of vascular endothelial activation via apolipoprotein A1. Lastly, flavanols increase the resistance of LDL-C to oxidation by either scavenging chain-initiating oxygen radicals or chelating transitional metal ions.
Several prospective studies have demonstrated a benefit of flavanol-rich cocoa products on cholesterol. In hypertensive patients, daily consumption of 100 g of flavanol-rich chocolate over 2 weeks led to a significant 12% reduction of LDL-C and total cholesterol. Another study found that daily consumption of 75 g of dark chocolate over 3 weeks increased HDL-C concentration up to 14% and inhibited lipid peroxidation in healthy adults. A 2007 study of hypercholesterolemic patients also demonstrated lipid improvements over 12 weeks with consumption of 26 g of cocoa powder daily. A 12.6% reduction in LDL-C, 23.4% elevation in HDL-C, and suppression of oxidized LDL-C (expressed as 9.4% prolongation in lag time) was observed with all endpoints except LDL-C reduction reaching significance.
Conflicting results exist about the effect of cocoa on HDL-C concentrations. Studies that demonstrated an HDL-C increase with cocoa used theobromine-free products as the comparator or control. In studies where the control contained theobromine, no HDL-C benefit was observed. A recent study showed that daily consumption of 850 mg of pure theobromine independently and significantly increased HDL-C concentrations by 6.12 mg/dL in healthy patients, an effect not seen with the cocoa treatment arm (150 mg theobromine). This finding suggests that theobromine in cocoa, rather than flavanols, may contribute to or even be responsible for the HDL-C benefit.
Other Considerations for Cocoa Intake
Although the positive effects of chocolate and cocoa products seem apparent, precautions exist. An obvious concern with the high intake of most commercially available products is the significant caloric (about 500 kcal/100 g), saturated fat, and sugar content. Excess caloric intake can lead to adverse metabolic effects, including weight gain and blood glucose elevations, negating the positive effects of cocoa. This is a particular concern with our current obesity epidemic and national recommendations for decreased fat and sugar intake among the general population. Although less palatable, cocoa-based products with little or no sugar or added fat are certainly preferred. In most trials previously cited, body weight and glucose concentrations over the study period were found to be unaffected with the doses provided. Again, studies were typically conducted over a short time, and long-term consequences are unknown.
While chocolate has been reported to cause gastrointestinal complaints, migraine headaches, and jitteriness, these effects have not been demonstrated in clinical trials. Chocolate contains caffeine, possibly increasing the risk of tachyarrhythmias and sleep disturbances. Caffeine intake may also lead to medication interactions. Increased central nervous system stimulant effects, decreased sedative effects, decreased theophylline clearance, and increased risk of clozapine toxicity are among the potential interactions. Additionally, cocoa itself may potentiate the effects of anticoagulants. Typical serving sizes of chocolate do not contain enough caffeine to warrant concern; however, overall caffeine intake from all sources must be considered.
In clinical trials, the amount of dark chocolate ingested ranged from 6.3 g to 105 g daily, containing 30 mg to 1080 mg (mean 545.5 mg) of flavanols. In general, 6 g are equal to one small square of a 100 g (3.5 ounce) dark chocolate bar, although exact concentrations are often difficult to identify. In addition to other study limitations identified, cocoa preparations used as dietary supplements have the same concerns as other products for which there is no quality oversight. Content inconsistencies, lack of standardization, and variable formulations make product selection challenging and specific dosing recommendations difficult to determine. From another perspective, dark chocolate may offer a relatively safe and inexpensive approach to support CVD prevention in patients at risk. The estimated incremental cost-effectiveness ratio was $50,000 per years of life saved when only $42 per person per year was spent on a prevention strategy utilizing dark chocolate.
For centuries, cocoa has been recognized for its delectable taste and proposed health benefits, perhaps as just wishful thinking for many. Research suggests that cocoa does indeed exert beneficial cardiovascular effects, mediated largely through the flavanol components. The effects of cocoa are likely the result of modifications of NO activity. Increased NO bioavailability may explain the improvement in endothelial function and the potentially beneficial effects on BP, platelet function, insulin resistance, and lipids. Unfortunately, like other dietary supplements, unresolved issues regarding dosing and the lack of clarity surrounding product components make it difficult to recommend chocolate for health benefits. Further investigation is needed before cocoa products should be recommended as a treatment option for BP reduction or other CVD parameters. Insufficient evidence exists to recommend cocoa as a supplement specifically for CVD risk reduction. A prudent approach is consideration of the inclusion of “healthy” dark chocolate in moderation as a part of overall lifestyle modifications for the prevention or treatment of cardiovascular risk factors. While chocolate can be a healthy supplement, potential benefits may be diminished with the excess caloric intake caused by consuming most commercially available chocolates, making this guilty pleasure a particular concern in our obese population. However, addition of “healthy” dark chocolate to a well-balanced calorically appropriate diet offers a pleasurable and palatable option with little burden for many people.
The authors have no conflicts of interest or financial disclosures to declare.