No, your chocolate also won't kill you
In my last article I explained how supplements are unlikely to kill you, and here we will show the same for chocolate and cocoa-based supplements. The next article will deal with hardcore geroscience again, promised (maybe with trametinib?). One can be excused for getting carried away with evidence-based, gero-adjacent biohacking given that we finally have enough evidence to make reasonable, strong, science-based guesses about improving our own health! So here we go.
Many articles have raised the issue of elevated heavy metal burden in chocolates and cocoa products. Many articles have been written appearing to debunk the former, too. The reality, however, is a bit more complicated than that.
Before we start, I want to clarify that this is not another post that will tell you how “proposition 65” and consumer labs are wrong. There is, in fact, and as far as I am concerned, no safe intake level of cadmium. The major theme that this post will follow is also a driver for my own research direction and supplementation. All things involve risks and benefits that need to be traded off.
Ironically, and sadly, more articles have been written warning about chocolate than articles praising the results of the COSMOS trial which found that supplementation with cocoa flavonoids reduces cardiovascular mortality (although the statistical evidence was a bit shaky). Hard outcome data from real trials trumps mechanistic speculation based on observational studies that is used to argue for lowering Cd intakes.
There is now no way around supplementing cocoa polyphenols if you are serious about your health. The literature overall suggested that flavonoids are beneficial for a long time but there was no evidence from controlled trials to support this idea until now. With the results of COSMOS this has changed and the polyphenol/flavonoid hypothesis is validated. What remains is the practical implementation, which is were worries about Cd are reasonable.
Why worry about cadmium (and lead) to begin with?
Long long ago, when I was working on metal-binding and -detoxifying metallothioneins in longevity, I also looked into the risks of Cd intake. Briefly put, given enough time, Cd accumulates in the kidney promoting renal failure (Satarug et al. 2017). Cd and lead have both been linked with adverse outcomes like cancer (e.g. Guo et al. 2022) and cardiovascular mortality, although some of the observational data is unreliable as per usual (e.g. for lead-IQ [insert less wrong post I cannot find here]). The mechanisms are not entirely clear.
An old-school study on renal Cd levels with age. (Interestingly in many studies the levels peak around middle age and then decline, perhaps due to kidney dysfunction.)
Why flavonoids?
Higher intakes of flavonoids are associated with reduced all-cause mortality (e.g. Mazidi et al. 2020) and cocoa flavonoids specifically are associated with improvements in intermediate outcomes like blood pressure in controlled trials (Amoah et al. 2022). Surely, funding bias — publication bias — has been discussed for cocoa studies and remains a concern, but overall the data looked more than solid even before COSMOS. Taken together with the almost-kind-of-sort-of significant reduction in the primary CVD outcome of COSMOS this means we should “update” our beliefs about flavonoids to “yes, they work to reduce hard outcomes” (since all the direct and indirect data points in the same direction) and to “the effect sizes must be smaller than expected” (since the confidence intervals overlap 1. For those who care those were the CIs on the primary composite endpoint (HR: 0.90; 95% CI: 0.78, 1.02; P = 0.11) with a significant reduction in CVD mortality as 2ndary endpoint, and a numerical 11% decrease in all-cause mortality.
Consistent with human data, several mouse studies also suggest that taking flavonoids can extend mouse lifespan (e.g. the recent Zumerle et al. 2024) although gold-standard confirmation by the ITP is still outstanding. In fact, when the ITP tested green tea extract it failed to extend lifespan of their HET3 mice.
The evidence points to epicatechin-rich flavonoids as particularly beneficial. The failure of green tea in the ITP and real concerns over EGCG hepatoxicity push me strongly towards cocoa as a source of such flavonoids.
The only lifespan study we have with cocoa (Bisson et al. 2008):
The main caveat with these animal studies is that most polyphenols undergo extensive biotransformation leading to very different metabolites present in human vs mouse blood. Nonetheless, this is a good starting point.
Risk minimization strategies
Generally, <0.5 ppm of Cd is considered acceptable for cocoa and <0.3 ppm for chocolate. To further reduce the risk while still being able to consume cocoa flavonoids, we can aim for a lower Cd content in the products we buy. After a brief survey of available products I believe that <0.2 ppm should be achievable. If you were to consume 20g of cocoa powder with 0.2 ppm of Cd this would result in an additional daily intake of 4 ug of Cd. This compares somewhat favourably to observed daily intakes of 10ug/d in the general population. Although dietary intake assessment is not easy and thus not a gold standard for observational studies (urinary excretion is preferred), it is worth emphasizing that no health risks were found with Cd intakes below 16 ug/d according to this review (Satarug et al. 2017).
Ora cacao, "Thriving Tanzania" provides a CoA showing 0.109 ppm Cd and non-detectable lead levels. Several other products they offer also show impressive results. Test-retest and batch-to-batch variability is not known.
https://ceremonial-cacao.com/pages/heavy-metals-results
Indeed, test results may change over time and even per batch. For example, Navitas is quoted at 0.13 ppm in this reddit thread, while current results are closer to 0.4 ppm according to their webpage.
As for chocolate, freely available, US-centric, test results from 2022 can be also found on this webpage: https://www.asyousow.org/environmental-health/toxic-enforcement/toxic-chocolate#chocolate-tables
The best strategy might be to find a low Cd batch, buy in bulk and keep the powder in the fridge or freezer.
Finally, switching to the cocoa supplement used in the COSMOS trial, and developed by CocoaVia, is another option if you prefer a pill. Theoretically this is the best option because it is the very product that was tested in the trial. But if you prefer to eat pure cocoa powder for the taste just like me, another option is to buy the ultra-low cadmium (1), polyphenol enriched cocoa powder that CocoaVia offers.
One thing that I would like to mention in regards to “honesty” is the role of the sponsor in the COSMOS trial. Although Mars co-funded COSMOS together with Pfizer and NIH, I do believe that the investigators are independent. It is clear that Mars simply wanted the good publicity from the trial. Any time you run a trial with big or small pharma, or together with supplement companies, they will try to spin it to their advantage. The same is happening here, when the CocoaVia webpage misrepresents the observational findings from COSMOS claiming that cocoa can improve cognition when COSMOS specifically refuted this observational finding in their trial. Cocoa is good, but maybe not that good. Thus we have to always take the product information with a small grain of salt.
https://www.cocoavia.com/pages/the-cosmos-trial
Where do we go from here?
If 500mg cocoa flavanols are beneficial in COSMOS, what about 1000 or 1500mg? It is very tempting to dose much higher, especially since many human trials used much larger doses and those higher doses seem to work better, at least for blood pressure. I am still in the process of figuring out the optimal intake myself — maybe a topic for another blog update in the future.
Amoah 2022, https://www.mdpi.com/2304-8158/11/13/1962
References and notes
Satarug, Soisungwan, David A. Vesey, and Glenda C. Gobe. "Health risk assessment of dietary cadmium intake: do current guidelines indicate how much is safe?." Environmental health perspectives 125.3 (2017): 284-288.
Zumerle, Sara, et al. "Targeting senescence induced by age or chemotherapy with a polyphenol-rich natural extract improves longevity and healthspan in mice." Nature Aging (2024): 1-18.
Guo, Xianwei, et al. "Association of urinary or blood heavy metals and mortality from all causes, cardiovascular disease, and cancer in the general population: a systematic review and meta-analysis of cohort studies." Environmental Science and Pollution Research 29.45 (2022): 67483-67503.
Mazidi, Mohsen, Niki Katsiki, and Maciej Banach. "A greater flavonoid intake is associated with lower total and cause-specific mortality: a meta-analysis of cohort studies." Nutrients 12.8 (2020): 2350.
Amoah, Isaac, et al. "Effect of cocoa beverage and dark chocolate consumption on blood pressure in those with normal and elevated blood pressure: A systematic review and meta-analysis." Foods 11.13 (2022): 1962.
Effects of long-term administration of a cocoa polyphenolic extract (Acticoa powder) on cognitive performances in aged rats. Bisson JF, Nejdi A, Rozan P, Hidalgo S, Lalonde R, Messaoudi M. Br J Nutr. 2008 Jul;100(1):94-101. Epub 2008 Jan 8.
Sesso, Howard D., et al. "Effect of cocoa flavanol supplementation for the prevention of cardiovascular disease events: the COcoa Supplement and Multivitamin Outcomes Study (COSMOS) randomized clinical trial." The American journal of clinical nutrition 115.6 (2022): 1490-1500.
For Cd levels:
https://pachamana.com/articles/
https://www.consumerlab.com/reviews/cocoa-powders-and-chocolates-sources-of-flavanols/cocoa-flavanols/
(1) https://www.ahealthylife.nl/wp-content/uploads/2023/01/DarkChocolate_Cocoa-ConsumerLab.pdf