on April 08, 2013
at 09:21 PM
I posted this on the Time article. edit: ** (first paragraph might not be right, see comments section. The 2nd and 3rd I still stand by) **
That's all nice but it's missing an important fact: TMAO levels aren't elevated in humans by eating meat http://www.ncbi.nlm.nih.gov/pubmed/10456680 only by carnitine supplements, choline supplements, and eating some kinds of fish. Fish contains it, but I'm guessing that in meat the carnitine isn't exposed to gut microbes until it's digested and then shortly after it's absorbed into the blood stream with little exposure to the microbes, not enough to significantly impact levels in the blood. But this isn't the case for carnitine supplements that are all out in the open and exposed to microbes from the start. And this new study didn't show that meat increased TMAO levels, there was consumption of meat but only accompanied by carnitine supplements, which actually have been demonstrated to increase TMAO levels.
So am I to expect Time to come out and tell people that eating fish causes heart disease, and to stay away from choline and carnitine supplements because they will get you? I hope not, there isn't really any evidence that fish causes heart disease despite its potent effects on TMAO levels, in fact most correlations are inverse http://www.ncbi.nlm.nih.gov/pubmed/21914258 If TMAO was so important then fish would cause heart disease, wouldn't it? Maybe the heart disease patients with high TMAO levels that they looked at were eating a lot of fish because after getting heart disease they heard that eating fish was good for their hearts and that explains the association? Maybe poor kidney function results in chronically high TMAO levels in response to fish consumption and is also associated with heart disease? Is high serum carnitine a marker for insulin resistance because insulin can't store carnitine and it builds up in the blood? Lots of questions to ask. And those genetically modified mice in the study that took the carnitine supplements may not be a relevant model for humans.
Also the recent meta-analysis for red meat and coronary heart disease only implicated processed red meat like bacon and sausages, not fresh red meat like steak and roasts http://www.ncbi.nlm.nih.gov/pubmed/23001745 Though there was a weak positive correlation with diabetes risk, so maybe your recommendations for red meat are right. Or maybe it's just a problem with overcooking it which can be inflammatory, or maybe it's the fries and soda that people tend to have with their red meat. Who knows? In any case all of this isn't really convincing.
I'm also adding this http://www.ncbi.nlm.nih.gov/pubmed/11529691 L-carnitine prevents atherosclerosis in rabbits. Though maybe they don't have the right gut bacteria.
Also nice post Mscott.
on April 08, 2013
at 02:08 AM
I find this all a bit at odds with the blinded RCT's on the benefits of carnitine on heart disease:
"Angina pectoris (17.6 vs 36.0%), New York Heart Association class III and IV heart failure plus left ventricular enlargement (23.4 vs 36.0%) and total arrhythmias (13.7 vs 28.0%) were significantly less in the carnitine group compared to placebo. Total cardiac events including cardiac deaths and nonfatal infarction were 15.6% in the carnitine group vs 26.0% in the placebo group"
"There were 6 deaths in the placebo group and 1 death in the L-carnitine group. Survival analysis with the Kaplan-Meier method showed that patients' survival was statistically significant (P <.04) in favor of the L-carnitine group. L-carnitine appears to possess considerable potential for the long-term treatment of patients with heart failure attributable to dilated cardiomyopathy"
"L-carnitine was randomly administered to 81 patients at an oral dose of g 4/die for 12 months, in addition to the pharmacological treatment generally used. For the whole period of 12 months, these patients showed, in comparison with the controls, an improvement in heart rate (p < 0.005), systolic arterial pressure (p < 0.005) and diastolic arterial pressure (NS); a decrease of anginal attacks (p < 0.005), of rhythm disorders (NS) and of clinical signs of impaired myocardial contractility (NS), and a clear improvement in the lipid pattern (p < 0.005). The above changes were accompanied by a lower mortality in the treated group (1.2%, p < 0.005), while in the control group there was a mortality of 12.5%"
The TMNO theory is an interesting idea which I hope to look into further, but as of now I find the mechanism not well supported by controlled trials.
(Just some stuff I wanted to add)
Higher TMAO is associated with greater heart disease, OK. What about CRP, plasma glucose, insulin sensitivity, HbA1c, TNF-a, oxidized LDL, IL-6, and adiponectin? Surely these have been shown to affect heart disease risk. Well, carnitine supplementation has been shown to improve all of these in some clinical trials (4,5,6,7,8).
Since an ApoE (-/-) model was used to say carnitine may promote atherosclerosis, how about we look at this paper, which found that the type of CLA found richly in grass fed animal fat slowed atherosclerosis progression in ApoE (-/-) mice. There are a lot of things in meat which can exert different effects, both in isolation and in combination which should probably be accounted for. But really, how reliable are these animal models? I'm sure they are to some degree, but this study, for example, found very different effects of saturated fat, monounsaturated fat, polyunsaturated fat, and carbohydrates on atherosclerosis progression and blood lipids depending on whether an ApoE knockout or LDL receptor knockout mice were used.
on April 07, 2013
at 11:30 PM
Edited my reply as I read the study more and cleaned up my stream of consciousness analysis...
Most definitely red meat is getting scapegoated in popular news stories about this study. But an interesting study, nonetheless! There are flaws and some assumptions the authors have made that I don't think are quite valid.
TMAO makes sense as an oxidative stressor in the body, no surprise really that it would promote atherosclerosis. No issue with that idea. Though they do have to use knockout mice to induce atherosclerosis and much higher TMAO concentrations are necessary than what is shown in humans.
Plasma levels of carnitine and TMAO are doing some interesting things. Plasma carnitine rises first and then only after it returns to baseline does plasma TMAO begin to rise (see Figure 1e in the study). What does this mean? Who knows! I'd wager through that carnitine is absorbed early in the gut, and TMA is generated by bacteria late in the gut.
Why do antibiotic-treated guts not absorb carnitine as well? The middle graph in Figure 1e suggests this. Odd!
Why to vegans produce no TMAO? The authors claim it's the absence of the particular microbes that suggest this, and that may be the case. However, if you delve into the supporting information (freely available, by the way!), you'll notice that vegans spike carnitine more than omnivores (Supplementary Figure 6). The authors suggest this due to competitive metabolism of carnitine by gut microbes, however enhanced absorption by challenged guts (vegans) also could explain this. They lack a control experiment where antibiotic-treated omnivores exhibit the same serum carnitine response curves than do to better support their arguement.
Perhaps the vegans simply are so deficient in carnitine that they essentially absorb it all before the gut microbes have a chance to? Let's see saturation dose of carnitine, something along the lines of 2+ grams which I've seen mentioned as the maximum useful dose.
I don't doubt the possibility that vegans likely have very different gut microbes than do omnivores.
I think it's quite likely that lower carnitine levels in vegans can explain some of the results in this paper. Lower serum carnitine doesn't sound ideal to me really either.
on April 08, 2013
at 06:00 AM
I believe we underestimate the role our gut bacteria plays in our health. I also think that we need more vegetables in our diets, just like we need more tubers.
I remember listening to Daphne Miller. She said, "Use meat as spice." This made sense to me. Hunters would not catch a buffalo every single day. And even when they caught it, one buffalo would be divided among so many people - so they would end up eating ... not so much. They would diversify and mix their protein sources - raw eggs, fish, seafood, fermented/sprouted grains, local/wild greens.
I think growing your vegetable garden is one of many essential things you can do for your health.
It is not about just carnitine. It is about eating species specific diet.
I wonder if they had the same results if they fed those people RAW meat.
on April 08, 2013
at 04:14 PM
In the first paper (Wang 2011, Nature), they do exactly what you'd want them to do. They screen plasma for unknown metabolites that correlate with cardiovascular disease. I suppose there wasn't necessarily a second step if they didn't find anything. However, they found three metabolites, that along with some nifty work in mice, presented this model...
- Dietary phosphatidylcholine eventually converted to TMA by gut microbes
- TMA converted to TMAO by liver (FMO3)
- Increased TMAO > upregulation of scavenger receptors > choline induced atherosclerosis
This focuses on phosphatidylcholine, present in eggs, milk, liver, red meat, poultry, shellfish, and fish. Interestingly referenced (Martin, 2008, MBS) is a paper showing administration of the probiotic L. paracasei to mice with humanized micro biome resulted in decreased TMA (but not TMAO).
The second paper (Bennett, 2013, Cell Metabolism) is a little drier, and further implicates FMO3 as the main culprit in the conversion of TMA > TMAO. Suggests looking for variants that you can plug into 23andme.
The third paper (Koeth, 2013, Nature Medicine) has the big payoff...red meat. It moves from choline to another trimethylamine, L-carnitine, which is abundant in red meat. It is noteworthy that carnitine did not make the initial cut that led to the choline association with increased atherosclerosis and CVD.
I find it hard to believe they got a > 5 year vegan (self-reported) to eat a steak, but I have no doubt that diet can modulate intestinal microbiota.
So there are correlations between plasma carnitine levels and CVD risk, but they are knocked down some after adjusting for traditional risk factors.
Pretty good stuff. Offers part of an explanation for relationship between animal consumption and increased CVD risk that's not predicated on saturated fat and cholesterol. It's part of the puzzle. If I understand correctly, you'd still need the initial event (oxidized LDL penetrating the endothelial wall) and it presumably doesn't change other variables, such as the benefit of high HDL.
I think if you wanted to be a defender, you'd go to the old standby, and call yourselves a different class, so that;
Maybe your gut is different from an SAD omnivore? If dietary carnitine (choline) can alter your intestinal microbiota, I would certainly expect that fast-food and sugar could, as well.
Maybe you're different. If you actually have less oxidized LDL and higher HDL, maybe you are mitigating the effects of TMAO enhanced atherosclerosis?
If you eat massive amounts of red meat, or animals in general, maybe you want to reconsider.
If you supplement aggressively with L-carnitine or choline (or other trimethylamines?) you probably want to reconsider.
Put this on the list with all of the other risk factors, and continue to follow along.
(Pretty funny how the choline paper generated no media buzz, but the carnitine/red meat has everybody freaking out).
on April 07, 2013
at 11:23 PM
I saw this earlier today and I'll be interested in the discussion here.
Until then, I'll just say it's probably the same as everything else--despite the flat pronouncements so loved by writers of articles, your results may vary. The valid point I do accept from this and similar articles is that today's CAFO beef is not your father's beef and certainly not your grandfather's beef.
I've always confounded the medical profession because despite being overweight much of my life I've been healthy other than mild complaints that proved to be sensitivity to wheat.
While on SAD, tests for inflammation were completely negative which my doctor said she never saw. Also on SAD I had perfectly normal cholesterol with high HDL.
After a year on paleo, which included lots of CAFO beef with fat and bone broths, I had cholesterol of 160, normal blood pressure and a great blood panel. So here I am at 66, not taking any meds at all and much stronger than any of my female friends.
Will you have similar results? Maybe, maybe not. Is supermarket beef ideal, or even preferred? No.
I shouldn't claim everyone will thrive on paleo just because I appear to, and problems experienced by some people from red meat don't mean no one should eat it.
There simply isn't one correct diet, no matter how many times people try to define one. And there are few if any foods that will make every human sick.
on April 08, 2013
at 12:26 AM
"In fact, these scientists suspected that saturated fat and cholesterol made only a minor contribution to the increased amount of heart disease seen" -- Slipped in, but is some good news.
"Meat contains protein, for example, and B vitamins, which are both essential for health. But the study???s findings indicated that the often-noticed association between red meat consumption and heart disease risk might be related to more than just the saturated fat and cholesterol in red meats like beef and pork" -- Another confirmation of what many of us believe
This research seeks to find what variables can lead to heart issues, other than the myth of saturated fats and cholesterol. That's refreshing.
It would be foolish to completely deny that eating a ton of red meat, in an unbalanced way, can lead to heart health issues. Make that meat processed, and you compound the problem.
- "That led to the steak-eating study. It turned out that within a couple of hours of a regular meat-eater having a steak, TMAO levels in the blood soared.
"But the outcome was quite different when a vegan ate a steak. Researchers had hypothesized that vegans would not have as many of the gut bacteria needed to make TMAO, and indeed virtually no TMAO appeared in the vegan???s blood after he consumed a steak."
I will take this final point to mean that eating too much, out of balance with your other nutrients, can lead to heart problems. Eating a healthy diet, that works for you, can include red meat. Overdo it, and you're asking for complications.
on April 08, 2013
at 04:31 PM
This sounds like yet another attempt (if not by the researchers, then by the media) to establish causes via correlation.
The gist seems to be this (I only read the abstract because I didn't want to buy the full text):
- We find more TMAO in habitual meat-eaters after they eat foods high in L-carnitine.
- Having both high L-carnitine and high TMAO is associated with adverse cardiac events.
I think the form of this will look familiar to the cholesterol fallacy:
- We find more circulating cholesterol and fatty acids after people eat fatty foods.
- Higher circulating cholesterol is associated with adverse cardiac events (and is especially damning because we find it in plaques!).
Um, maybe something else is causing both the high carnitine/TMAO levels and the cardiac events? Maybe? Shouldn't we then try to get to the root and stop hacking at the leaves? Because maybe hacking at the leaves will cause other problems? We only have to consider what statin drugs and artificially low cholesterol are doing to people. I predict a similar thing will happen when people try to take this study as a suggestion to do something stupid like take antibiotics.
Also, did anyone notice the functions of carnitine in humans (see Wikipedia)? Apparently it builds bone and is a "substantial" antioxidant. Your body can synthesize it.
This statement from the NY Times article should give anyone pause:
It is quickly converted by the liver into yet another little-studied chemical called TMAO that gets into the blood and increases the risk of heart disease.
Huh? TMAO is little-studied, and yet scientist know it increases the risk of heart disease? The "increases risk" phrase was entirely unqualified, so readers will take it to mean it's established fact. I personally am unable to find studies on TMAO itself apart from L-carnitine. I consider the statement a leap and a reading-comprehension problem on the NY Times writer's part.
Some Wikipedia editor also made the same gaffe on the carnitine page:
When intestinal bacteria were exposed to carnitine from food, they produced a waste product, Trimethylamine N-oxide (TMAO), that is known to cause atherosclerosis.
Other reasons to avoid jumping to conclusions have already been mentioned by the other answerers, such as, "was this CAFO red meat?"
The abstract used a strange statement:
than did vegans or vegetarians following ingestion of l-carnitine through a microbiota-dependent mechanism
What does "microbiota-dependent mechanism" mean in terms of ingestion?
How exactly did these vegans/vegetarians ingest the l-carnitine? Did they eat red meat or take some extracted form? Did even the omnivores ingest red meat? I suspect the full article would probably elaborate on this.
Scientists (and more so the media) really need to be a little more cautious about experimental results. Science is always discovering new things that turn the old accepted dogma upside down (it's especially bad if the old dogma itself was based on a leap of logic). Even worse is when there is actually evidence of confounding factors. For example, researchers should ask, are the people we are testing representative of the entire US population, or the average human anywhere in the world? Are there characteristics of the red meat we are testing with that might not be representative of the variety of red meat in the US, or in the world? Experimental research is not easy, and you have to be somewhat of a philosopher to do it well. Philosophy deals with abstract principles of specificity and generality - Are my results specific to a certain set of conditions? How far can I reasonably generalize this?
on April 07, 2013
at 11:33 PM
I'm always open to new studies with the potential of changing the way we paleo, health-conscious Homo sapiens eat???if they actually apply to how paleo people eat.
The study was conducted using a single cut of meat. Are beef hearts the same as tongues as sweetbreads? And the meat used was extremely unlikely to be grass-fed and finished. And then compare that to a single v*gan. N=1 is nice and all when the results are used to help that original 1, but the authors of the study are using that 1 v*gan to demonstrate the optimal levels of health for every.single.person.
Sure, that annoys me and all, but the part I can't read without leaving a furious comment is the part on antibiotics. I'm paraphrasing here, but I'm pretty sure the article said something like this: let's give every person with a heart and therefore potential of heart disease antibiotics so we can 1) destroy every last bacterium in people's guts so they can't properly digest anything and 2) assist in the evolution of new, antibiotic-resistant strains of bacteria that might have greater human-killing superpowers. Delightful, no?
I type this while chowing down on ribs.
on April 10, 2013
at 11:20 PM
"The biological effects of a food cannot possibly be reduced to one of the biological effects of one of the food???s components."
on April 10, 2013
at 06:00 PM
Mark Sisson weighs in: http://www.marksdailyapple.com/does-red-meat-clog-your-arteries-after-all/#axzz2Q4CYLRW8
We should definitely keep an eye on this and any other future research. I???m particularly interested in what this means for research into the effects of gut flora on human health.
Balanced response, as expected.
Reports aren't always what they seem, but just because studies have been "debunked" in the past doesn't mean you rush out and try to "debunk" every study in which you don't like the conclusion (especially if it's not your day job).
on April 08, 2013
at 03:46 PM
This is indeed troubling. Everyone on paleo eats red meat and therefore consumes large amounts of carnitine. Some eating paleo also supplement with carnitine or acetyl l carnitine to increase energy or enhance brain function. Now this study which seems to be very straight forward in the negative effect TMAO exerts on plaque and heart disease. Surely we need more studies but if further studies indicate the same, paleos must be willing to question whether they are primarily paleo because it's "natural" or they are paleo because it is healthy (and enhances life span and risk of heart disease). If they answer that they are paleo because it optimizes life span (and the study is believed and replicated), paleos should abandon the consumption of red meat.
Btw, anyone who posts that this study is about optimized gut flora or that eating fruits and vegetables or cultured foods or probiotics will somehow prevent TMAO's effect on heart plaque has not read the study. Interestingly (but without all the mainstream press) Nature published a paper demonstrating that choline also promotes TMAO increasing heart disease: http://www.ncbi.nlm.nih.gov/pubmed/21475195 This study may also be relevant as us Paleos consume large amount of eggs and fish which are loaded with choline (and some of us supplement with choline, phosphatidylcholine or phosphatidylserine). Again choline (and derivatives) definitely promote brain function but possibly at the cost of future heart disease.
on April 08, 2013
at 05:19 AM
Before spleens are vented and teeth gnashed, perhaps it's best for everyone to read one of the studies referenced in the article:
"Intestinal microbiota metabolism of l-carnitine, a nutrient in red meat, promotes atherosclerosis" Koeth, Wang, et al. Nature Medicine(2013), doi:10.1038/nm.3145
n=2595 in this study. Not definitive, by any means. Certainly a sufficient starting point for investigation of TMAO production from consumption of CAFO vs. grass-fed meats.
Also, need it be said: atherosclerosis is probably polygenic.
on April 11, 2013
at 05:26 PM
The best response i've seen to the new red meat study is this article by chris masterjohn. IMO, he thoroughly debunks the study.
on April 07, 2013
at 11:39 PM
I really want to know if grass fed beef causes a different reaction, and if having enough K-2 in your diet (ahem, thanks grass fed beef fat!) is protective even if the gut flora in grass-fed-meat-eaters has the same response. I would expect the gut flora in people eating paleo, or even just no grains/refined sugar, would be very different from that (those?) in people eating a conventional diet.
on April 10, 2013
at 02:47 PM
Chris Kresser weighs in: http://chriskresser.com/red-meat-and-tmao-its-the-gut-not-the-meat
No surprise he's a skeptic of the study. Makes the claim that healthy user bias means you have better gut flora than the average. Suggests that the omnivores in the paper suffer from dysbiosis and SIBO and that explains why their guts produce trimethylamine when fed red meat.
(Community Wiki'd this, others can edit to add more of his argument if you like.)
on April 09, 2013
at 03:04 PM
Would have thought this would get a little more play? What is this site about?
I really appreciate how people "review" these studies - not summarize or give their impressions, but critically review - I am indebted to all who undertake this endeavor simply for the entertainment it provides. You should e-mail the authors to point out the flaws. Why not, right?
on April 09, 2013
at 06:46 AM
I found this brief write up, linked by Anastasia at PrimalMedEd:
"The study found that in genetically modified mice, a high (but not impossible) dosage of l-carnitine did double plaque buildup. This may or may not be related to TMAO, we cannot say. This may or may not happen in humans, we cannot say. Overall? It's just preliminary research that should only interest other researchers, not the layperson."
Still, I'm going to keep an eye on this one.