Chocolate consumption is inversely associated with calcified atherosclerotic plaque in the coronary arteries
The NHLBI Family Heart Study; Clinical Nutrition, Volume 30, Issue 1 , Pages 38-43, February 2011. Luc Djoussé, Paul N. Hopkins, Donna K. Arnett, James S. Pankow , Ingrid Borecki, Kari E. North , R. Curtis Ellison
Background & aims
While a diet rich in anti-oxidant has been favorably associated with coronary disease and hypertension, limited data have evaluated the influence of such diet on subclinical disease. Thus, we sought to examine whether chocolate consumption is associated with calcified atherosclerotic plaque in the coronary arteries (CAC).
In a cross-sectional design, we studied 2217 participants of the NHLBI Family Heart Study. Chocolate consumption was assessed by a semi-quantitative food frequency questionnaire and CAC was measured by cardiac CT. We defined prevalent CAC using an Agatston score of at least 100 and fitted generalized estimating equations to calculate prevalence odds ratios of CAC.
There was an inverse association between frequency of chocolate consumption and prevalent CAC. Odds ratios (95% CI) for CAC were 1.0 (reference), 0.94 (0.66–1.35), 0.78 (0.53–1.13), and 0.68 (0.48–0.97) for chocolate consumption of 0, 1–3 times per month, once per week, and 2+ times per week, respectively (p for trend 0.022), adjusting for age, sex, energy intake, waist-hip ratio, education, smoking, alcohol consumption, ratio of total-to-HDL-cholesterol, non-chocolate candy, and diabetes mellitus. Controlling for additional confounders did not alter the findings. Exclusion of subjects with coronary heart disease or diabetes mellitus did not materially change the odds ratio estimates but did modestly decrease the overall significance (p = 0.07).
These data suggest that chocolate consumption might be inversely associated with prevalent CAC.
In this study, we demonstrated that chocolate consumption was inversely associated with prevalent CAC in a dose-response manner. Using other cut points to define prevalent CAC yielded similar results except when an Agatston score of 0 was used. In addition, exclusion of subjects with prevalent CHD or diabetes mellitus was suggestive of an inverse association between chocolate consumption and CAC. To our knowledge, this is the first study to examine whether chocolate consumption is associated with lower prevalent CAC as previous observational and interventional studies have focused on the effects of chocolate consumption on blood pressure, endothelial function, and platelet function. Using the same data, we identified an inverse association between chocolate consumption and prevalent CHD (manuscript under review). While health benefits from chocolate consumption might have been suspected as early as in the 17th century (mostly in Europe),23 chocolate is not viewed in the US as a healthy food, but rather as a source of fats and calories.23 However, the literature suggests that moderate consumption of dark chocolate or other flavanoid-rich foods might have cardiovascular benefits.12 In a cross-over design, consumption of 100 g/d of dark chocolate was associated with a 5.1 mm Hg reduction in systolic blood pressure and 1.8 mm Hg reduction in diastolic blood pressure in 13 hypertensive subjects after 14 days of intervention.10 Another study also demonstrated beneficial effects of dark chocolate on blood pressure in healthy subjects after a 2-week intervention.13 In the Iowa Women’s Health Study,24 chocolate contributed 6% of total catechins and when analyzed by catechin source, there was suggestive evidence for an inverse association between chocolate derived-catechin and CHD death [RR (95% CI): 0.88 (0.71–1.08)] in a multivariable adjusted model comparing the 3rd with the 1st tertile of catechin.
Potential physiologic mechanisms by which chocolate might lower the risk of CHD have been suggested. Besides lowering systolic and diastolic blood pressure,10, 13, 25 dark chocolate has been shown to transiently but substantially increase nitric oxide bioactivity in human plasma and reverse endothelial dysfunction.26, 27, 28 In a randomized trial of 41 diabetic subjects, a 30-day intervention with flavanol-rich cocoa resulted in a 30% increase in flow-mediated dilation of the brachial artery.29 Furthermore, a prospective cohort reported a 40% lower risk of cardiovascular risk comparing the fourth to the first quartile of chocolate consumption and about 12% of that reduction was attributable to beneficial effects of chocolate on blood pressure.30 In addition, dark chocolate was shown to improve insulin sensitivity and beta cell function in healthy or hypertensive subjects after 15 days of intervention.13, 25 Furthermore, there is an evidence that chocolate might suppress epinephrine-stimulated platelet activation and platelet microparticle formation.11 Other constituents of chocolate (magnesium and potassium) have been shown to exert beneficial effects on cell membrane and blood pressure. It is thus possible that flavonoids in chocolate, alone or in conjunction with other minerals, might favorably influence the development of atherosclerosis.
Calcium deposition in the arterial walls occurs in the early stages of atherosclerosis just after fatty streak formation31 and has been shown to correlate closely with the total burden of atheroma.1, 2 In sensitivity analyses, we observed an inverse association between chocolate intake and CAC using Agatston scores other than 0 to define prevalent CAC. It is likely that with a lower Agatston score cut point such as 0, cardiac CT might not have been able to accurately discriminate the presence or absence of clinically relevant calcified atherosclerotic plaque in the coronary arteries. This might lead to effect dilution as observed in this paper. Studies have shown that the use of an Agatston score of 0 as a cut point to predict angiographic disease (at least 50% stenosis) had a relatively low specificity (up to 40%).3, 32 In contrast, using a cut point of 100 had a sensitivity of 93% and a specificity of 76%.3
Our study has limitations. First, although CAC measurement was completed about 7 years after dietary assessment, we did not have baseline CAC measurement to differentiate calcification that was present at baseline from calcification that developed after assessment of chocolate consumption. Thus, our ability to determine temporality or infer causality is limited. Second, we were not able to differentiate between dark chocolate and lighter or milk chocolate. Polyphenolic content is higher in dark chocolate than it is in milk chocolate. Therefore, combining lighter and dark chocolate might have biased our results towards the null. Lastly, the possibility of confounding by indication (subjects with prevalent CAC or clinical CHD might have avoided chocolate consumption as bad foods) cannot completely be excluded in this study. However, the fact that exclusion of subjects with prevalent CHD showed a similar inverse association (p for trend 0.07 when CAC of at least 100 was used as cut point and 0.005 when CAC of at least 150 was used as cut point) does not support confounding by indication as these subjects were free of symptomatic CHD. On the other hand, the large sample size, the availability of data on major CHD risk factors, and the standardized techniques used for both CAC and dietary assessment were major strengths of the study.
In conclusion, our findings indicate that consumption of chocolate might be inversely associated with prevalent CAC in a dose-response manner. Future studies are warranted to confirm these findings and elucidate physiologic mechanisms by which moderate consumption of chocolate (up to 2 servings per week) might positively influence the risk of atherosclerosis.