Chocolate consumption and cardiometabolic disorders: systematic review and meta-analysis

BMJ 2011; 343 doi: (Published 29 August 2011)
Cite this as: BMJ 2011;343:d4488

Recent rapid responses

Rapid responses are electronic letters to the editor. They enable our users to debate issues raised in articles published on Although a selection of rapid responses will be included as edited readers' letters in the weekly print issue of the BMJ, their first appearance online means that they are published articles. If you need the url (web address) of an individual response, perhaps for citation purposes, simply click on the response headline and copy the url from the browser window.

Displaying 1-10 out of 10 published

To the Editor. Comment on “Chocolate consumption and cardio-metabolic disorders: systematic review and meta-analysis1”: a role for unrecognised vitamin D content?
Marion Rowe1, Senior Laboratory Scientific Officer, Peter M Timms1, Consultant Biochemist, Barbara J Bouche2r, Professor of diabetes
1 Department of Clinical Biochemistry, Homerton University Hospital Foundation Trust, Homerton Row, London E9 6SR, UK .2 Centre for Diabetes, Bart’s & The London School of Medicine & Dentistry ,Blizard Institute, Queen Mary University of London, Newark Street, London E12AT, UK

Enjoyment of chocolate may become a less guilty pleasure in the light of this meta-analysis reporting reductions by at least a third in the risks of ‘any cardiovascular disease’ and of stroke in prospective studies of chocolate consumption over 8-16 years, especially with consumption of dark chocolate1. This risk reduction was not abolished by adjustment for recognised risk factors for cardiovascular disease including unspecified ‘dietary factors’. Chocolate is thought to have anti-inflammatory and anti-atherogenic effects that have been suggested to be due to contained polyphenols and flavinoids2. We wonder, however, whether fat-soluble vitamin D might be present in chocolate and contributing to these beneficial effects, since it has both anti-inflammatory and anti-atherogenic effects and, furthermore, better baseline repletion is associated prospectively with reductions in cardiometabolic risk factors. 3-5. In the 1930s, vitamin D2 was reported to be present in cacao beans on bio-assay but only after they had been undergone fermentation and been dried in the sun, the vitamin D being thought to be synthesised by invading moulds,6,7 since vitamin D is well known to be formed in yeasts and fungi under the influence of UVB from sunshine. However, cacao shells [nibs] and cacao butter are also reported to contain sterols [at 6-8% and 0.3-0.4% respectively] that develop anti-rachitic activity when irradiated with UVB, and the total vitamin D content of prepared cacao beans has been reported to be comparable to that found in natural cod-liver oil. Cacao beans are used in chocolate manufacture and nibs are also used in making some dark chocolate. Furthermore, winter feeding of cows with dried cacao bean shells in the winter increased the anti-rachitic effects of their butter to match that seen in the summer7. We have, therefore, analysed an especially high coca dark chocolate currently on sale in the UK using modern methodology [Liquid chromatography coupled with triple quad mass spectrometry]. Vitamin D2 was present at 1.042 microgram per gram, providing at least 1000 IU per ounce [25 G]; thus, anyone eating dark chocolate at 1-2 oz/day is getting at 1000-2000 IU of vitamin D2/day which is rather more than the recommended daily amount of vitamin D [600 IU/day] recently advised for adults by the Institute of Medicine in the US8. Buitrago-Lopez et al. report adjustment for ‘dietary factors’ [including caffeine] does not reduce the effects of chocolate consumption.1 However, can the authors say whether the beneficial effects of high chocolate intake that they report persist after adjustment for the vitamin D content of the chocolate eaten, since vitamin D could be contributing to the dose-dependent reductions in cardiovascular disease [CVD] risk reported with higher chocolate consumption. If adjustment for vitamin D content does reduce or abolish the association it would support the suggestion that improving vitamin D repletion could contribute to reduction in CVD risks9, supporting the rationale for carrying out adequate trials of vitamin D supplementation for CVD prevention.

1. Buitrago-Lopez A, Sanderson J, Johnson L, Warnakula S, Wood A, Di Angelantonia E, Franco OH. Chocolate consumption and cardiometabolic disorders: systematic review and meta-analysis. BMJ (2011) 343:d4488. doi: 10.1136/bmj.d4488.
2. Katz DL, Doughty K, Ali A. Antiox Redox Signal (2011) 15:2779-811
3. Jeffery LE, Burke F, Mura M, Zheng Y, Qureshi OS, Hewison M, Walker LS, Raza K, Sansom DM. 1,25-Dihdroxyvitamuin D3 and IL-2 combine to inhibit T cell production of inflammatory cytokines and promote development of regulatory T cells expressing CTLA-4 and FoxP3. J Immunol (2009) 183: 5458-67
4. Gouni-Berthold J, Krone W, Berthold HK. Vitamin D and cardiovascular disease. Curr Vasc Pharmacol 7:414-22
5. Forouhi NG, Luan J, Cooper A, Boucher BJ, Wareham NJ. Baseline serum 25-hydroxyvitamin d is predictive of future glycemic status and insulin resistance: the Medical Research Council Ely Prospective Study 1990-2000. Diabetes (2008) 57:2619-25
6. Knapp AW, Coward KH. Vitamin D activity of cacao shell: i. the effect of the fermenting and drying of cacao on the vitamin D potency of cacao shell. (1935). ii. The origin of vitamin D in cacao shell. Biochem J. (1935) 29:2728-35
7. Kon SK, Henry KM. The effect of feeding cacao shells to cows on the vitamin D content of butter (milk). Biochem J (19935) 29:2051-6
8. Ross AC, Manson JE, Abrams SA, Aloia JF, Brannon PM, Clinton SK et al. The 2011 report on dietary reference intakes for calcium and vitamin D from the Institute of Medicine: what clinicians need to know. J Clin Endocrinol Metab. (2011) 96:53-8
9. Judd SE, Tangpricha V. Vitamin D therapy and cardiovascular health. Curr Ther Hyperten. (2011) 13:187-91

Competing interests: None declared

Barbara J Boucher, Professor of Diabetes

Mariaon Rowe, Peter M Timms`

Bart's & The London School of Medicine & Dentistry, Queen Mary University fo London, UK, Centre fir Diabetes, Blizard Institute, Newark St. London E12AT

Click to like:

Migraines are caused (in many if not all) cases by the dilation of blood vessels. Many anti-migraine drugs cause the narrowing of said vessels (and can cause cardiac associated problems).

Therefore Chocolate and wine - which in the case of wine (red) is also known to dilate blood vessels, will cause migraine, but can be beneficial for certain cardiovascular symptoms. Therefore possibly chocolate can act like that as well.

I don't think, in fact, we have enough data and studies to draw final conclusions for either outcomes.

Competing interests: None declared

Marketa Zvelebil, Research Scientist


Click to like:

The epidemiological studies suggesting that chocolate consumption prevents cardiometabolic disorders are difficult to believe.1 It has long been established that eating chocolate precipitates migraine in many people - which can be due to sticky platelets and amine changes.

In the 1970s I spent 10 years at Charing Cross Migraine Clinic, and most patients already knew they could not take either chocolate or red wine without getting a headache. Chocolate is one of the commonest food allergens in the Von Pirquet sense. Avoiding food and chemical precipitants not only prevented patients getting migraines but also lowered labile high blood pressure into the normal range.2

There has also been a similar epidemiological mistake that some alcohol is beneficial - which is wrong - because alcohol is a cell poison and blocks cell membranes. I think this flawed epidemology is due to the fact that sensitive people have learned to avoid alcohol and the same applies to chocolate.

1 Buitrago-Lopez A, Sanderson J, Johnson L, et al. Chocolate consumption and cardiometabolic disorders: systematic review and meta-analysis. BMJ 2011 343:d4488doi:10.1136/bmj.d4488

2 Grant ECG. Food Allergies and Migraine Lancet 1979;1:966-968.

Competing interests: None declared

Ellen C. Grant, Retired physician and medical Gynaecologist


Click to like:


As would be expected, the claim that chocolate reduces cardiovascular events [1] has been warmly received. Any excuse to indulge in one of life's great pleasures comes as welcome news in these bleak times.

But we should beware those who tell us what we want to believe. And this is certainly true in the case of Buitrago-Lopez et al. [1] After scouring the literature, they identified a handful of relevant epidemiological studies. When we look closely, though, we see that the meta-analysis was based on data dredged from very large cohort studies. Unsurprisingly, the results from individual studies were, to say the least, unimpressive. For example, the German study [2] reported that chocolate reduced the risk of myocardial infarction by <1% in absolute terms and the size of the effect was similar in the case of stroke. But it's the old, old story: trivial treatment effects disguised by the use of relative risk reductions. [3] And the same applies, of course, to the meta -analysis. [1]

The authors of the paper duly acknowledged its limitations [1] and so did the writer of the accompanying editorial, [4] although this prevented neither from drawing enthusiastic conclusions. And, as has become the custom in these situations, they also called for randomised trials. But who is going to fund this research? The pharmaceutical companies have nothing to gain by doing so. Chocolate manufacturers would be wise to settle for what they have at present. As for independent research, it is to be hoped that scarce resources would not be squandered on such flimsy grounds. But, in any case, the results of large-scale RCTs would not be available for a decade or more.

In the meantime, the misguided believers in statistics-based research will peddle the nonsense that the meta-analysis from Buitrago-Lopez et al. is something of importance. [3] But the claim of a link between chocolate and the prevention of cardiovascular events is unfounded. Let this be a warning to the Royal Colleges: it would be premature to begin preparing mandatory training courses in rock climbing for medical practitioners. The days when doctors have to scale the side of country mansions to dispense Milk Tray to their patients are a long way off.

James Penston



1. Buitrago-Lopez A, Sanderson J, Johnson L, Warnakula S, Wood A, et al. Chocolate consumption and cardiometabolic disorders: systematic review and meta-analysis. BMJ 2011;343;d4488

2. Buijsse B, Weikert C, Drogan D, Bergmann M, Boeing H. Chocolate consumption in relation to blood pressure and risk of cardiovascular disease in German adults. Eur Heart J 2010;31;1616-23.

3. Penston J. Stats.con - How we've been fooled by statistics-based research in medicine. The London Press. London, November 2010.

4. Mackenbach JP. The temptations of chocolate. BMJ 2011;343;d5883.

Competing interests: None declared

James Penston, Consultant Physician/Gastroenterologist

Scunthorpe General Hospital, Scunthorpe, North Lincolnshire DN15 7BH

Click to like:

I read with interest the recent systematic review and meta-analysis of chocolate consumption and cardio-metabolic disorders.(1) The number of cohorts included in the meta-analysis is small (six in total and all adults) and none were randomised controlled trials. The heterogeneity was higher than 50% for the effect of chocolate on any cardiovascular event, indicating a large range of effects; however, the results for stroke only, were more robust with low levels of variation. No attempt was made to compare different type of chocolate due to the limited evidence available.

Despite the limitations of this review, the results call into question the scientific evidence to encourage the population to reduce chocolate consumption. I refer particularly to the school meal standards in England (and similar standards introduced throughout the UK) which are described by the School Food Trust.(2) The food based standards were introduced in 2006 closely followed by nutrient based standards in primary schools in 2007 and secondary schools in 2008. An important component of the standards was the exclusion of chocolate (with the exception of cocoa powder). The reasons are clear; the association of cocoa solids with sugar and fat in all but the strongest dark chocolate products is well established.

There are three important points to make here. Firstly, by excluding chocolate covered confectionery and desserts, sugar and fat consumption is not necessarily reduced if another sweet, fatty food is substituted such as flapjack or jam sponge. Secondly, if the results from the recent review are confirmed by randomised controlled trials, restricting chocolate products may actually be detrimental to health. Thirdly, any nutrition policy introduced must always be supported by the most rigorous of scientific evidence. If this is not the case, those involved in nutrition policy, are in danger of losing all credibility. I strongly urge the governmental departments involved in legislating school meal standards to regularly review the nutritional epidemiological evidence to ensure that school food provides the best diet for children's future health.

1. Buitrago-Lopez A, Sanderson J, Johnson L, Warnakula S, Wood A, Di Angelantonio E, et al. Chocolate consumption and cardiometabolic disorders: systematic review and meta-analysis. BMJ 2011;343.

2. School Food Trust. Revised guide to standards for school lunches, 2008.

Competing interests: None declared

Charlotte E.L. Evans, Lecturer in Nutrition Policy

University of Leeds

Click to like:

6 September 2011

A powerful implication of this study is that very few -if any- expensive trials using expensive drugs have obtained the same reduction in the risk of cardiometabolic disorders. We should pay more and more attention to life styles and diet habits of our patients, and be less prone to lures of pharmaceutical companies.

Competing interests: None declared

Antonio Migheli, G.P.

ASL TO5 Moncalieri (Italy)

Click to like:

31 August 2011

Could you please clarify more in more detail the following:

1. The properties of the chocolate consumed (cocoa content, sugar content, etc).
2. The amounts consumed by the participants in relation to the local population, as from my understanding there is a wide global variation in national chocolate consumption.
3. The length of time required to gain the benefits of increased chocolate consumption.
4. You mention that the participants where aged 18 and above. Which age group benefited the most?

As you must know, such publications as this make for sensational lay news. This leads to friends asking for more information and puts an onus on the medical profession to have adequate advice.

Thank you.

Competing interests: None declared

Chayim Freed, Medical Student

King's College London

Click to like:

31 August 2011

This systematic review and meta-analysis highlights something important; patients calorie count but do not density count.

I think more emphasis should be paid to the calorie breakdown on the front of food packaging. Although manufacturers have made an attempt to improve information provision about the calorie content of chocolate, it only appears as calories per bar or half a bar on the front packaging.

As a consumer, I believe it may be the saturated fat and sugar content of chocolate that makes most of us want to indulge; the more of it the better it tastes. If the packaging was to show this on the front, instead of e.g. 240 calories (10% of GDA), it may make it easier to convey this information to patients. We've seen this on some foods, but chocolate is lagging behind.

We have already established that too much sugar and saturated fat has detrimental effects on patients' health. So, rather than focusing on trials to estimate how much chocolate is good for someone, the focus should be on deterring patients from eating bad chocolate.

Competing interests: None declared

Saiji Nageshwaran, Student

University College London

Click to like:

31 August 2011

So the lay public wants to know (or maybe they don't) - what kind of chocolate is recommended and/or was used? The photo of the bars I see in the publicly available portions of this article show bars that look like a Kit Kat bar! Not the same as an 88% extreme dark chocolate bar with relatively low levels of additives. Please clarify and make public in the abstract or other portions that are available to us lay folks!

Competing interests: None declared

Erik Olsen, interested layperson

Town of Blacksburg

Click to like:

30 August 2011

Though the data is not new, the huge number of participants included, gives it a powerful quality. But, if one wants to adopt the findings, taking the authors reservations regarding high calories intake and possible influence on weight seriously,it brings us to the core question the authors tried to circumvent:

What is high and what is low chocolate consumption to reduce cardiometabolic risks?
Should we recommend chocolate eating once daily? weekly? or else? what is the ammount of cocolate eating balance to maximize beneficial effect to the detrimenal one?
Last but not least - how long should we consume chocolate? i.e - is a given period of chcolate consumption can last and for how long?

Competing interests: None declared

Reuben Steinherz, Pediatrician

Israel 71799

Click to like: