PDA

View Full Version : Heavy metals in your protein powders: follow up



WillBrink
06-12-10, 15:50
Regarding report from CR that has people all concerned about protein powders, here's a write up on the topic by someone who actually knows what he's talking about, senior environmental scientist with CH2M-HILL, John Lowe, who I know well, who knows heavy metal toxicology inside and out:


How Not to Report on Health Risks – Consumer Reports on Heavy Metals in Protein Powders

Consumer Reports seems to have stirred up some controversy over protein powders and drinks. These are staples in the fitness world, and while they’re not intended to be a substitute for real food, they are a convenient way to get some protein into you before a workout. CR’s article is a bit of a scatter-shot complaint about the nutritional benefits and health risks, much of which I’m not particularly disposed to address. However, CR drew my attention by informing its readers how. . .

“[s]ome protein drinks can even pose health risks, including exposure to potentially harmful heavy metals, if consumed frequently. All drinks in our tests had at least one sample containing one or more of the following contaminants: arsenic, cadmium, lead, and mercury. These metals can have toxic effects on several organs in the body.”

"Harmful." "Contaminants." "Heavy metals." "Toxic effects." These are terms that I do not sling around with abandon. And, from my perspective, people who read CR’s report about protein powders - at least the portion that discusses health risks from heavy metal contamination - will come away alarmed, confused, no better educated about this topic than when they picked up the article; and with no roadmap about what kinds of decisions they should make about using protein powders.

What CR did was to interview a few subject-matter experts; buy a handful of protein powder products; sent them to an analytical laboratory for metals analyses; read some government reports; then drew their conclusions about the health risks from the metals detected in the protein powders and communicated those to their readers. What could be so wrong about that?

I’ve found plenty of things to complain about, enough to fill several posts (welcome back to blogging. . .). Let’s unpack one of CR’s claims, and see what they’re really saying.

"We found that three daily services of the ready-to-drink liquid EAS Myoplex Original Rich Dark Chocolate Shake provides an average of 16.9 micrograms (ug) of arsenic, exceeding the proposed USP limit of 15 ug per day."

Sounds dire. But what does that really mean? Is that a lot of arsenic? Are you putting your health in danger if you take in more arsenic than what’s in the USP limit? CR benchmarks the risks from heavy metal contamination using limits developed by the U.S. Pharmacopeia (USP). These are thrown out to the reader without providing any explanation as to what these limits signify, and particularly without noting that these limits are not arbitrary lines that define “safe” from “not safe”. The USP proposed “permitted daily exposure” (PDE) for arsenic is based on a well-regarded epidemiological study of Taiwanese villagers exposed to high levels of arsenic in drinking water. That study a “no-observed-adverse-effect-level:” or NOAEL for significant adverse effect from long-term exposure which in this case were skin lesions and effects to blood vessels (hyperpigmentation, keratosis and possible vascular complications, according to the U.S. Environmental Protection Agency).

Toxicologists express the NOAEL as the amount of the toxic compound ingested per unit of body weight per day. In the case of the Taiwanese epidemiological study, the NOAEL for arsenic exposure was 0.8 micrograms per kilogram of body weight per day or 0.8 ug/kg-day. That no-effect level is then reduced by a factor that accounts for uncertainties in the health effects information and to produce a level of exposure that protects sensitive individuals. The value used to calculate the PDE is 0.3 ug/kg-day, is also used by other regulatory agencies such as EPA and the Agency for Toxic Substances and Disease Registry (ATSDR) to assess health risks from arsenic (EPA’s term for this value is the “Reference Dose” or RfD – we’ll come back to that later). To calculate the PDE, USP assumes that a person exposed to arsenic weighs 50 kg or 110 pounds, to obtain the level of 15 ug/day cited in CR’s article. The weight assumption is selected presumably because USP is trying to set a level that’s protective for everyone from scrawny teen-agers to 300-pound defensive tackles,

So, is that a “safe” level? Are you risking adverse effects if you consume more arsenic than 15 ug/day? CR’s presentation of the information implies that exposure to more than 15 ug/day is unsafe, but the real answer isn’t that clear-cut. The RfD (which is used to develop USP’s limits) is a level described by regulatory agencies such as EPA as “likely to be without an appreciable risk of deleterious effects”. Lower levels of exposure are not likely to be associated with health risks. Correspondingly, as the level and frequency of exposures exceeding the RfD increase, the probability of adverse effects increases. However, you can’t say categorically that any arsenic exposure above the RfD (or USP’s limits) will produce adverse health effects.

That’s probably a bit confusing and understandably so especially when readers haven’t been clued in about the imprecise nature of these health risk-based levels (you certainly won’t find a word about that in CR’s article). One of the clues in deciphering this puzzle is in EPA’s definition of the RfD: “an estimate (with uncertainty spanning perhaps an order of magnitude). . .”. The purpose for that wording is to alert people to the fact that risk-based levels such as the RfD are imprecise – that’s the nature of the science used to develop them.

Now, how can you be expected to judge what should be your level of concern about ingesting 16.9 ug/day of arsenic in your protein shakes when USP’s limit is 15 ug/day, when the authorities say there might be an order of magnitude range (from 4.5 to 45 ug/day – the math behind this range is a topic for another post) around USP’s limit for arsenic? The answer is that the numbers comparison is only an approximate guide for judging the magnitude of risk associated with arsenic exposure. It needs to be accompanied with other information to fully present and communicate the health risks. And, this is information that CR didn’t include in its article.

The kinds of information needed to fully communicate risks would answer these questions: 1) what is known about the capacity of a toxic chemical to cause adverse health effects in humans or laboratory animals; 2) what is known about the quantitative relationship between exposure and adverse health effects (also known as the “dose-response relationship” – the RfD is an example of a quantitative dose-response relationship); 3) what is known about the patterns and magnitude of human exposure? In addition, there should be a discussion of uncertainty that addresses topics such as the quantity and quality of information used in assessing health risks, gaps in our knowledge about the chemical and assumptions made in estimating health risks.

It sounds like a lot of work, but assessing health risks, which is what CR’s article was doing, means presenting this information in a meaningful way that is transparent to the users. Only then can the users, in this case people who are making decisions about buying and using protein powders, properly weigh the benefits versus the risks.

What’s one example of health risk information that should have accompanied the numbers comparison? It might have been helpful if CR had mentioned that some forms of arsenic, and in particular the forms found in foods, are not known to pose a health risk to humans. Arsenic in the environment occurs in ionic form (inorganic arsenic) and bound to organic molecules (organic arsenic). Inorganic arsenic, and in particular trivalent arsenic, is of greatest concern for causing adverse health effects in humans. Most organic arsenic compounds are not known to pose a health risks in humans; for example, the human body detoxifies inorganic arsenic by metabolizing it to an organic form. In addition, arsenic is naturally occurring in soil, water and foods. Typical intake of arsenic from ingestion of food and water is estimated to be 50 ug/day; however only 10 ug/day or less of that is considered to be inorganic arsenic – therefore, some of the arsenic you take in with foods is present as organic arsenic. This distinction becomes important when you look at the laboratory analyses of the protein powder and shake products supplied by CR.

CR also doesn’t provide critical details about the laboratory analyses of the protein powders, such as whether or not the analytical method selected would distinguish from inorganic arsenic species from organic arsenic species. USP’s proposed standard for heavy metals (they refer to them more correctly as ‘elemental impurities”) specifically states that “arsenic may be measured using a nonspeciation procedure under the assumption that all arsenic contained in the supplement is in the inorganic form. When the limit is exceeded using a nonspeciation procedure, compliance with the limit for inorganic arsenic shall be demonstrated on the basis of a speciation procedure,” which is something CR doesn’t mention in its article. In the article, CR makes the assumption that all of the arsenic found in the protein powders is inorganic arsenic which could be a human health risk, then doesn’t call readers’ attention to that assumption. This potentially misleads readers about how much of a health risk is posed by arsenic in protein powders.

CR fails to provide any context regarding the levels of exposure and circumstances under which these heavy metals have been shown to produce adverse effects in humans. I’ve focused on arsenic, but the same kind of story can be told for all of the metals mentioned in CR’s article – arsenic, cadmium, lead and mercury (I’ll try to provide more details in follow-up posts). CR’s description of the risks from metals in protein powders fails because it is not transparent in presenting its data, methods and assumptions, and by starting out with a predetermined frame that protein powders are “bad” in general; featuring the heavy metal contamination is an eye-catching way of making that point. I’m not an advocate for protein powders, but I am interested in using risk assessment in a manner that informs people and helps them make decisions about how to deal with health risks.

Cruncher Block
06-12-10, 23:43
Thank you for posting this. I recently read the CR article and agree with you on the overall alarmist tone.

I've become something of a skeptic concerning CR over time. While they generally provide useful information, I think they sometimes get in over their heads on technical subject matter. I've noticed too that they've never seen a new government regulation that they didn't like.

So, while I take note of the information they present on protein powders, I'm also looking for additional information.