New Meta-analysis Shows Clinically Significant Benefits of the Effects of Cocoa on Blood Pressure

  • Cocoa (Theobroma cacao)
  • Blood Pressure
Date: 02-15-2013 HC# 101246-466

Re:  New Meta-analysis Shows Clinically Significant Benefits of the Effects of Cocoa on Blood Pressure

Ried K, Sullivan TR, Fakler P, Frank OR, Stocks NP. Effect of cocoa on blood pressure. Cochrane Database Syst Rev. August 15, 2012;8:CD008893. doi: 10.1002/14651858.CD008893.pub2.

High blood pressure (HBP), a risk factor for cardiovascular disease (CVD), contributes to about 50% of cardiovascular events and 37% of related deaths in Western countries. Even small reductions in blood pressure (BP) reduce CVD risk. Current guidelines suggest using lifestyle changes and complementary treatment with HBP medicines. Epidemiological studies indicate that cocoa (Theobroma cacao)-rich products may reduce CVD risk. Flavanols in cocoa increase formation of endothelial nitric oxide (NO), promote vasodilation, and reduce BP. NO production may be increased through the insulin-mediated signaling pathway. Cocoa intake improved insulin sensitivity in some, but not all, studies. Cocoa flavanols inhibit angiotensin-converting enzyme (ACE) hence reducing BP. Also, they may have indirect antioxidant effects on the cardiovascular system, again reducing BP. Earlier meta-analyses found that cocoa-rich foods may lower BP, but recent trials report conflicting data. This paper is an update of a Cochrane Database Systematic Review covering the effects of cocoa products on the BP of individuals with or without hypertension that was first published in 2010 and now includes several new trials.

Cocoa’s flavanols include epicatechin, catechin, and procyanidins, thought to be responsible for its BP-lowering effect. Flavanols occur in other foods, such as beans (Phaseolus vulgaris), apricots (Prunus armeniaca), blackberries (Rubus corchorifolius syn. R. villosus), apples (Malus pumila syn. M. domestica), and tea (Camellia sinensis), but in lower concentrations than in cocoa products, which had 460-610 mg/kg of flavanol monomers and 4-5 g/kg of flavanol polymers. However, flavanol content of cocoa products depends not only on the variety and maturity of the cacao beans used, but on their processing, and the processing of the resultant raw cocoa powder.

Traditionally a cold, unsweetened drink of raw, dried, powdered cacao beans, cocoa was transformed by Europeans who added sugar, milk, vanilla (Vanilla planifolia), and lecithin; roasted the beans before grinding; mixed (“conched”) and alkalized (“dutched”) the powder; heated cocoa beverages; and made other modifications. Modern processing reduces monomeric flavanol content and alters the epicatechin/catechin ratio. While fresh and fermented cacao beans have from 2.5-16.5 mg/g of epicatechin, processed cocoa has only 2-18% of this amount. Dark chocolate has more cocoa than milk chocolate, but a 70% cacao bar from one producer, for example, may have very different amounts and proportions of flavanols than a 70% bar from another company.

Interest in the effect of cocoa on BP began with a report that the Kuna people of the San Blas Islands in Panama had a noticeably low rate of HBP and consistent healthy BP, unaffected by age. Traditional island-dwelling Kuna drink 3-4 cocoa drinks daily on average, but mainland-dwelling Kuna consume up to 10 times less. Among this latter group, age-related BP increases and prevalence of HBP are comparable to Western populations. The cocoa powder used by the Kuna on the San Blas Islands has about 3.6% flavanols, while cocoa-rich dark chocolate products may have about 0.5% flavanols. Alkalization may reduce this to less than 10 mg/100 g (0.001%).

To perform this review, the authors searched the electronic databases Cochrane Hypertension Group Specialised Register, CENTRAL, MEDLINE, and EMBASE through November 2011, as well as international trial registries and the reference lists of review articles and randomized, controlled trials (RCTs). Inclusion criteria included parallel or crossover, single-blind, double-blind, or open-label trials of greater than 2 weeks’ duration that reported the clinical mean or median ± standard deviation (SD) (or ± standard error [SE]) systolic BP (SBP) or diastolic BP (DBP) at baseline and before and after the intervention. The primary outcome was a difference in SBP and DBP at the final follow-up between the cocoa and control groups, adjusted for baseline.

The authors identified 136 potentially relevant articles and 3 unpublished studies of which 20 met the inclusion criteria. Together they examined the effects of cocoa on BP in 856 mostly healthy adults. While most trials lasted 2-8 weeks, 1 was 18 weeks. Two RCTs included patients with HPB. Active groups received 30-1080 mg of flavanols daily (mean=545.5 mg) in 3.6-105 g of cocoa products. In 10 trials, they consumed 500-750 of mg flavanols daily. Nine studies used commercially available chocolate and 11 studies used flavanol-rich cocoa powder (tablet, bar, or powder mixed with water or milk) and compared the effect to a control group, which either took flavanol-free placebo (white chocolate, milk, or a placebo pill) or low-flavanol powder, respectively. In 12 RCTs, control groups received flavanol-free products; in 8, they received cocoa products with low-flavanol content (6.4 and 41 mg/d).

No RCTs were found on the effects of long-term daily use of cocoa products on BP, nor any on clinical outcomes related to HBP.

There was evidence of some publication bias, though it was not statistically significant by Egger’s test (P=0.081 for SBP; P=0.105 for DBP).

The meta-analyses found that there was a significant lowering of BP (2-3 mmHg) with flavanol-rich products compared with control in studies with a duration of 2-18 weeks (mean difference in SBP [95% confidence interval (CI)]: -2.77 [-4.72, -0.82] mmHg, P=0.005, n=20; mean difference in DBP [95% CI]: -2.20 [-3.46, -0.93] mmHg, P=0.006, n=19).

Sub-group analysis revealed that this BP-lowering effect was found in trials with a duration of 2 weeks (n=9), but not in trials lasting longer than 2 weeks. It is not clear whether this was due to trial length or another factor, such as the type of control group used in the shorter trials or the level of blinding of participants to the treatment. The BP-lowering effect was more striking in trials using flavanol-free control products than those using low-flavanol cocoa products (mean difference in SBP [95% CI]: -3.70 [-6.02, -1.36] mmHg, P=0.002, n=12; mean difference in DBP [95% CI]: -2.71 [4.26, -1.5] mmHg, P<0.001, n=11). The low-flavanol control products in 9 trials had potentially sufficient flavanols to produce a vascular response. There was no significant difference between the treatment and control groups in trials using low-flavanol controls, nor any association between monomeric flavanol dosage and BP change using meta-regression analysis. There was also no association between BP and the theobromine content of the cocoa.

The test for subgroup differences (flavanol-free trials compared with low-flavanol trials) provided moderate confidence that there is a genuine difference between the groups: SBP/DBP: I2=69.2%/67.5%.

At least 4 studies were required for subgroup analysis. Subgroups analyzed included participants with a baseline SBP≥140 mmHg vs. SBP<140 mmHg, with a baseline DBP≥80 mmHg vs. DBP<80 mmHg, in double-blind vs. single-blind RCTs, in industry-sponsored vs. non-industry-sponsored RCTs, and other variables indicating significant contribution to heterogeneity, such as those in parallel vs. crossover trials.

Patients with HBP showed a significant lowering of SBP (mean difference in SBP [95% CI]: 3.99 [-7.02, -0.97] mmHg, P=0.01), while those who were normotensive did not. However, this division was not supported by the test for subgroup differences (I2=0%).

One study in this review compared the effects of a cocoa product with an 80% sugar content with sugar-free cocoa powder on BP and endothelial function, finding a larger beneficial effect with sugar-free cocoa. Six studies compared overweight and obese participants with participants of normal weight and reported that beneficial effects of low-sugar cocoa products were more pronounced among overweight and obese subjects. Meta-regression analysis, however, found this effect to be of only borderline significance. Of 9 trials that investigated heart rate, 8 found no significant difference, and 1 reported significantly higher heart rate in the cocoa group compared with control (P=0.007).

Beneficial effects of cocoa on BP were found, by meta-analysis, to decrease with age; with significant reductions in trials with younger participants (mean age: <50 years; mean age range: 18-45.4 years), but not with older ones (mean age: >50 years; mean age range: 51-69.7 years) (mean difference in SBP for participants <50 years of age [95% CI]: -4.57 [7.41, -1.73] mmHg, P=0.002, n=10; mean difference in DBP for participants <50 years of age [95% CI]: -3.85 [-5.45, -2.26] mmHg, P<0.001, n=9).

Adverse effects were reported by 5% (n=22 out of 429) of subjects in active cocoa groups and 1% (n=5 out of 427) of those in control groups; these included gastrointestinal issues, dislike of the trial product, headache, and nervousness. A product used in 1 trial, with a theobromine content about 30 times that of commercial cocoa, had a laxative effect on 2 of 12 participants. Theobromine is toxic in small quantities to some animals, such as dogs, but it is estimated that a 60 kg human would need to consume about 4.5 kg of dark chocolate with naturally occurring theobromine to be harmed. Compliance was reported as 100% for 9 trials, ≥90% for 4, >80% for 5, 79% for 1, and 69% for 1.

Given the high heterogeneity, insufficient evidence of adequate allocation concealment in 55% of the included RCTs, single-blinding in 45% of the studies, and some publication bias, the authors rate the quality of the evidence they considered as low. RCTs with flavanol-free control groups, trials designed to directly compare effects of cocoa in different age groups, and those considering the effect of sugar content in subjects with a body mass index (BMI) >25 are needed. Tracking BP for longer than 2 weeks would help assess cumulative effects as well as product tolerability and acceptability. Finally, long-term trials are needed to assess the effect of cocoa products on cardiovascular clinical outcomes. The authors conclude by noting that, “Further research is very likely to

have an important impact on our confidence in the estimate of the effect and is likely to change the estimate.”

thanks to

American Botanical Council,

—Mariann Garner-Wizard