TMAO and Red Meat - What is it and is it Harmful | Chris Kresser
Interested in becoming an ADAPT Certified Functional Health Coach? Early Bird Enrollment starts on August 17th Learn More

Red Meat and TMAO: Cause for Concern, or Another Red Herring?


Last updated on

tmao, how to lower tmao
What is the impact of red meat consumption on TMAO in the bloodstream?

This article is part of a special report on Red Meat. To see the other articles in this series, click here.

I’m sure many of you have seen reports on a recent study published in the journal Nature suggesting a possible mechanism linking red meat consumption to heart disease. The day after one such report was published in the New York Times, I received numerous emails and numerous Facebook and Twitter messages from concerned red meat enthusiasts. This is understandable, but rest assured it’s not yet time to switch over to soy burgers.

The researchers in this study published a paper a while back proposing that a chemical called TMAO (trimethylamine N-oxide) increases the risk of heart disease. In this study, they hypothesized that eating red meat may increase levels of TMAO in the bloodstream, which would in turn ramp up your chances of having a heart attack. Sounds plausible, right?

There’s another hypothesis that also seemed plausible for why red meat increases the risk of heart disease (if we even accept that, which I do not; more on this in a moment). It’s called the “diet-heart hypothesis”, and you’re all very well aware of it whether you know it by name or not. It holds that eating cholesterol and saturated fat increase cholesterol levels in the blood, and high cholesterol levels in the blood cause heart disease. This theory became so widely accepted that few people even question it anymore. The problem is it’s simply not true. Recent research has shown that dietary saturated fat and cholesterol are not associated with heart disease after all, and even if they were, high cholesterol levels in the blood are not the culprit. I’ve written about this extensively in the past, and I will be starting a brand new series with updated information this month.

Is TMAO the new cholesterol? Find out why red meat is still innocent.Tweet This

The mistaken blame of saturated fat and cholesterol as drivers of heart disease led to a decades-long campaign to encourage low-fat, high-carbohydrate diets. Unfortunately, the effects of this campaign were not harmless. Not only did it unnecessarily deprive people of nutrient-dense, nourishing (and delicious!) foods like meat, butter and eggs, it may have indirectly contributed to the epidemics of obesity, heart disease and diabetes. Studies have shown that when people replace saturated fat with carbohydrates, the risk of heart disease doesn’t go down—it goes up. (1) This is not because of the carbohydrates, per se, but because 85% of the grain consumed in the U.S. is in the highly refined form. (2)

The diet-heart hypothesis should be a cautionary tale that prevents us from jumping to rash conclusions based on limited evidence. Alas, the almost complete lack of criticism or scrutiny in the popular media reports on this study indicate that caution has been thrown to the wind. Let’s now examine three reasons why I’m not yet ready to take the conclusions of this study (i.e. red meat causes heart disease via TMAO) at face value.

Epidemiological Evidence Is Inconsistent

If red meat consumption elevates TMAO, and elevated TMAO increases the risk of heart disease, we’d expect to see higher rates of heart disease in people that eat more red meat. The epidemiological evidence examining this question is mixed. A large meta-analysis published in Circulation by Micha et al. covering over 1.2 million participants found that consumption of fresh, unprocessed red meat is not associated with increased risk of coronary heart disease (CHD), stroke or diabetes. (3) On the other hand, a smaller prospective study including about 121,000 participants from the Nurses Health Study and Health Professionals Follow-up Study did find an association between red meat consumption (both fresh and processed) and total mortality, cardiovascular disease (CVD) and cancer. (4)

If eating meat increases heart disease risk we might expect lower rates in vegans and vegetarians. Early studies suggested this was true, but later, better-controlled studies suggest it’s not. The early studies were poorly designed and subject to confounding factors (i.e. vegetarians tend to be more health conscious on average than general population, so there could be other factors explaining their longevity, such as more exercise, less smoking, etc.). Newer, higher quality studies that have attempted to control for these confounding factors haven’t found any survival advantage in vegetarians. For example, one study compared the mortality of people who shopped in health food stores (both vegetarians and omnivores) to people in the general population. They found that both vegetarians and omnivores in the health food store group lived longer than people in the general population. (5) This suggests, of course, that eating meat in the context of a healthy diet does not have the same effect as eating meat in the context of an unhealthy diet. (Hold this thought: we’ll be coming back to it shortly.) A very large study performed in the U.K. in 2003 including over 65,000 subjects corroborated these results: no difference in mortality was observed between vegetarians and omnivores. (6)

Taken together, these data do not suggest a strong relationship between red meat and heart disease. It’s also crucial to remember that epidemiological evidence does not prove causality. Even if red meat intake is associated with a higher risk of CVD (or any other health problem), such studies don’t tell us that red meat is causing the problem. If you’re new to this concept, I suggest reading these excellent articles by Denise Minger and Chris Masterjohn.

The “Healthy User Bias” Strikes Again

The healthy user bias is the scientific way of explaining the phenomenon I described above, where people that engage in one behavior that is perceived as healthy (whether it is or not) are more likely to engage in other behaviors that are healthy. (7, 8) Of course the flip-side is also true: those that engage in behaviors perceived to be unhealthy are more likely to engage in other unhealthy behaviors. The healthy user bias is one of the main reasons it’s so difficult to infer causality from epidemiological relationships. For example, say a study shows that eating processed meats like bacon and hot dogs increases your risk of heart disease. (9) Let’s also say, as the healthy user bias predicts, that those who eat more bacon and hot dogs also eat a lot more refined flour (hot dog and hamburger buns), sugar and industrial seed oils, and a lot less fresh fruits, vegetables and soluble fiber. They also drink and smoke more, exercise less and generally do not take care of themselves very well. How do we know, then, that it’s the processed meat that is increasing the risk of heart disease rather than these other things—or perhaps some combination of these other things and the processed meat? The answer is, we don’t. Good studies attempt to control for some of these confounding factors, but inevitably some will not be controlled for. And one of the most important potential confounding factors that is never controlled for is the gut microbiome.

Numerous studies, which I’ve written about on this blog and spoken about on my podcast, suggest that the balance of bacteria in our gut may be one of the most important factors—if not the most important—that determines our overall health. Gut dysbiosis (an imbalance between healthy and unhealthy bacteria in the gut) and small intestine bacterial overgrowth (SIBO, a condition involving an inappropriate overgrowth of bacteria in the gut) have been linked to health problems as diverse as skin disease, depression, anxiety, autoimmunity, and hair loss.

The study we’re discussing here found that those who eat red meat produce TMAO, whereas vegans and vegetarians who hadn’t eaten meat for at least a year do not. The researchers claimed that this means eating red meat must alter the gut flora in a way that predisposes toward TMAO production. However, there’s another explanation that I believe is much more plausible: the red meat eaters are engaging in unhealthy behaviors that have led to dysbiosis and/or SIBO. This could include eating fewer fruits and vegetables and less soluble fiber, and more processed and refined flour, sugar and seed oils. All of these behaviors have been shown to be more common in the “average” red meat eater, and all of them have been associated with undesirable changes in the gut microbiota. (10, 11, 12) In other words, the problem isn’t the red meat, it’s the gut bacteria. This is supported by the finding in the study that the red meat eaters did not produce TMAO after a course of antibiotics. It is also supported by data indicating that a breakdown in the intestinal barrier, which occurs in dysbiosis and SIBO, may increase heart disease risk by elevating the number of circulating LDL particles in the bloodstream. (13) I will be covering this (i.e. the connection between LDL particles and heart disease) in my updated series on heart disease.

In the last section I presented evidence suggesting that eating meat in the context of a healthy diet does not have the same effect as eating it in the context of an unhealthy diet. This study is likely yet another example. In order to know whether red meat is really to blame for changes in the gut flora that increase TMAO production, we’d have to do another study with two groups: one that follows a Paleo diet rich with fruits, vegetables and soluble fiber, as well as red meat; and another vegan/vegetarian diet with equivalent amounts of plant matter and no meat. If the Paleo diet followers still had higher levels of TMAO, this hypothesis would be a lot stronger.

The Jury Is Still out on TMAO

The evidence linking TMAO production to eating red meat, and serum TMAO levels to heart disease, is not as cut-and-dry as the study authors suggest. For example:

  • The Nature paper on TMAO contained data from two studies: an epidemiological study on humans, and a clinical study on mice. The human study compares a single vegan that they managed to convince to eat a steak to five “representative” meat-eaters. A sample size of six people, with only one in the vegan group, is hardly adequate to draw firm conclusions from.
  • The mouse study used a carnitine supplement. While it is well established that free carnitine increases TMAO production, previous studies have not shown that carnitine-rich foods like red meat increase TMAO. In fact, in one 1999 study, out of 46 different foods tested, including red meat, only one food elevated TMAO levels in the participants: seafood (see graph to right, from Chris Masterjohn’s article referenced below). This makes perfect sense since trimethylamine occurs naturally in seafood. Does this mean we should cut back on fish and shellfish because they’re going to give us a heart attack?(15)


Another obvious question we should ask is whether there are alternative explanations for why we see elevated TMAO levels in meat or seafood eaters (if indeed we do see them in a wide sample of meat eaters, which at least one earlier study didn’t support)? According to a 2011 article by Chris Masterjohn touching on TMAO in a different context: (16)

Elevated TMAO could reflect dietary trimethylamine or TMAO from seafood, but it could also reflect impaired excretion into the urine, or enhanced conversion of trimethylamine to TMAO in the liver.

The enzyme Fmo3 carries out this conversion, mainly in the liver, as reviewed here. There are a number of genetic variants affecting the activity of this enzyme, some of which appear only in certain ethnicities, and the enzyme also processes a number of drugs used to treat psychoses, infections, arthritis, gastro-esophageal reflux disease (GERD), ulcers, and breast cancer. Iron or salt overload may also increase the activity of the enzyme. TMAO could, then, be a marker for ethnicity, drug exposure, genetically determined drug efficacy, or other conditions.

As you can see, it’s overly simplistic to suggest that eating red meat causes elevated TMAO; there are many other factors at work.

But even if Paleo meat eaters have higher TMAO levels than vegans and vegetarians, we still don’t have evidence proving a causal relationship between TMAO and CVD. Once again, the supposed link between cholesterol and saturated fat and heart disease should serve as a reminder not to jump to hasty conclusions that unnecessarily deprive people of nutrient-dense, healthy foods. It is virtually impossible to control for all of the possible confounding factors, and the study we’re discussing in this article only further highlights this problem.


I’d like to end with an observation from the discussion section of the TMAO paper. The authors state:

Numerous studies have suggested a decrease in atherosclerotic disease risk in vegan and vegetarian individuals compared to omnivores; reduced levels of dietary cholesterol and saturated fat have been suggested as the mechanism explaining this decreased risk. Notably, a recent 4.8-year randomized dietary study showed a 30% reduction in cardiovascular events in subjects consuming a Mediterranean diet (with specific avoidance of red meat) compared to subjects consuming a control diet.

This might sound like damning evidence against red meat. However, when you look at Table One in Mediterranean Diet trial, you’ll find that the Mediterranean diet allowed more red meat than the control diet (a low-fat diet)! The Mediterranean Diet allowed for “one serving or less of red or processed meat per day“, whereas the low-fat diet only permitted “one serving or less of red or processed meat per week“. (You can see this for yourself. Click here to access the PDF version of the study, then scroll down to Table One.) Clearly this paper does not support the authors’ conclusion that red meat increases the risk of heart disease. [UPDATE: Stephan Guyenet brought my attention to the Article Supplement, which I had missed before. On  Page 9, it does show that the investigators asked those on the Mediterranean diets to choose white meat instead of red. However, on Page 26 the data indicate that the amount of red meat consumed during the study by the Mediterranean diet groups was virtually the same as the amount consumed by the low-fat/control group. This is a common problem in diet studies: the participants don’t always do what they’re told! What this means, of course, is that the 30% decrease in CVD observed in the Mediterranean group was caused by something other than reducing red meat consumption—which the Mediterranean group did not do.]

They also claim that vegan and vegetarian diets reduce the risk of atherosclerotic disease compared to omnivorous diets; but the studies they reference fail to adequately control for the “healthy user bias”. The study I mentioned in the beginning of this article compared heart disease risk amongst omnivores and vegetarians that shop at health food stores (which is a big step toward reducing healthy user bias), and did not find a difference in deaths from heart disease, stroke or all causes.

If you read the media reports and full-text of this study, you might have noticed something interesting. The study itself, and even most of the media article about it, quite simply and without much fanfare stated that saturated fat and cholesterol have little to do with the supposed increase in heart disease observed with red meat consumption. Hold the press! Shouldn’t THAT be front-page news?!? Apparently not. Of course, they’re only willing to admit this publicly in the context of an article where they’re proposing yet another mechanism for how red meat will kill you.

Finally, the most remarkable and sad part of this for me is seeing just how deep most people’s fear and distrust of red meat is, even if they’ve been following a Paleo diet for a long time. The day after the TMAO study was published, I woke up to no fewer than 20 emails and the same number of Facebook messages and Tweets from people expressing concern that their choice to eat red meat might be killing them. It really is a testament to the power of brainwashing. Most of us grew up with the idea that red meat is harmful, and it’s perhaps not so easy to leave that behind—even when you think you have.

Chris Masterjohn has also published a superb, detailed analysis of the underlying data in the study, and I highly recommend reading it. Frankly, the conclusions of the authors (that eating red meat increases the risk of heart disease via TMAO production) are so incongruous with the data in the study that it’s difficult to imagine how it could have passed peer review.

I believe we may be seeing more “red meat is bad because of TMAO” studies in the near future, so as always, when you see a media report on such a study, take it with a heavy grain of salt (which, by the way, doesn’t cause high blood pressure in most people!).


Join the conversation

  1. I think one reason that many ancestral eaters became concerned was not due to a lack of faith in their prior knowledge, but because TMAO is a term to them. At least it was for me. I had never heard of it before. And the article seemed to be convincing that elevated TMAO caused bad things at least in mice. So it would be concerning if (big if) red meat actually did elevate TMAO in a large sample of humans and *if* that elevated level really has any negative consequences when it’s part of an otherwise healthy well rounded diet… even a diet that is read meat heavy, or seafood heavy. These are important things to know as well since many people, esp bodybuilders, consume lots of meat protein.

    Good write up, thanks. Looking forward to Chris M.’s feedback as well. I will be keeping an eye out for learning more about TMAO.

  2. This was an awesome deconstruction on what is a very convoluted, elaborate hypothesis that grabs at least a few straws to keep holding the link between SFA and disease. Thank you for this, Chris.

  3. Weird, where did the study mention it was red meat only that was providing L-Carnitine?
    L-Carnitine is in all complete amino acid foods, namely whole proteins. Yes?
    And yet in the article it seemed (magic, mirrors and sleight of hand) to be all about red meat.
    Weird but oh so very human, if you want a good headline.

    • No, l-carnitine is not in all whole proteins. One thing that few people seem to mention – though it was brought up in the interview with the study’s author – is that l-carnitine is also abundant in salmon.

  4. Hello, what about supplements of l-carnitine for Fibro, following the mentioned above, should one avoid using them?
    thanks much.

    • I will have to investigate this further. Free carnitine is readily converted into TMAO, but the clinical significance of that is not clear.

    • There’s no gold-standard test right now, so it’s still in part a clinical diagnosis. There’s a breath test used in conventional settings, but it’s not very accurate. I use the Organix Dysbiosis profile from Metametrix as a starting point.

  5. Thanks Chris,
    I love knowing that I can count on you to tell it like it is, picking apart these studies and telling us the facts one by one. I think most people read the first paragraph or two of a study and take it as gospell.
    My husband and I eat red meat at least once a day, often 2 or 3x/day…have never been in better health at 60 and 67.
    It always seems to come back to grains and gut flora!!
    Thanks again for all the great work you do. I’m in Richmond/El Cerrito. I hope we get to meet some day. 🙂

  6. And then there are people like me who have Trimethylaminuria (TMAU), which is a genetic disorter that limits the activity of the FMO3 enzyme. I don’t have enough (or properly working) enzymes to convert (oxidize) the TMA to TMAO. Wish I did though, because TMA smells like rotten fish which them makes me smell like rotten fish. So my question is, is there a way to increase my TMA oxidation to make up for my faulty FMO3 Enzyme activity? Currently, I just have to limit my Choline intake and I take probiotics.

  7. Thanks for taking the time out of your busy schedule to write this, Chris! It’s frustrating to see how easily people can get disheartened by these things, even when they’ve trusted in Paleo a long time.
    I’m rather pleased that when I originally read the original TMAO article, my first conclusion was that it was about gut flora, not about the red meat, and that their conclusions were not soundly thought out. It’s because of people like you who’ve better taught me how to critically evaluate these sort of things that I was able to come to that conclusion. So thank you. 😀

  8. Will read the article, but wanted to point out a typo: in second paragraph, you’ll want to change “intern” to “in turn”.

    Love your blog!

  9. I get a brief moment of panic because now I am feeding red meat to my 4yo a number of times a week whereas previously we rarely ate it. But I’ve been reading you (and others) for a year now so I look at these articles and can see thru much of it (though not as easy without access to the full text study). These “journalists” do a disservice in writing these articles on such studies with no critical analysis, but then again I think there is an agenda on the part of the NYT writer, so I am not surprised. So, when I saw the article the other day, my initial dismay was quickly replaced by skeptiscism.

    Glad to see your aticle and I appreciate the timely analysis.

    • Colleen, The research the NYT reported on was very extensive and well done. They did not try to draw conclusions ie red meat will lead to heart attacks, but to show statistically that you are much more likely to die from stroke or heart disease if you consume red meat. In the study, people were 2.5 times more likely to die then non-meat eaters.

      Statistically, that is a HUGE increase in risk. For instance, people who smoke their whole lives are not even 2.5 times as likely to die of lung cancer then those who don’t smoke.

      So yes, from what we KNOW now, eating meat will kill you.

      • Oh please, show me a double blind, controlled study that accounts for confounders that proves what you supposedly KNOW

    • Dead soldiers during the Korean and Vietnam wars(?) proved that men in the 20s already had CVD.

      Do you really think eating the same diet they did will provide different results? Maybe you think most of them smoked or lived in a smoke filled environment? I do not know for sure what causes the CVD but I see plenty of it in the world so I will choose to follow the diets of the cultures where CVD is low. Check those out and you will not find loads of meat being consumed like in the Standard American Diet (SAD).
      Make your own conclusions or wait for the meat industry to tell you what to do, it is up to you!

      • “Make your own conclusions or wait for the meat industry to tell you what to do, it is up to you!”

        Funny you should say that, Richard. Cuz what I’ve read from you indicates that once we made up our own conclusions, if we happen to disagree with yours, you’re just gonna call us morons. Here, that’s what you said: “Haha, do some research….that is a typical response from a moron!”

        In my mind, you’re doing it wrong, Richard.

        • Ahhhhh you are right, I should not have called him a moron but when someone says “Do some research” it touches a sensitive point because I think that is as bad as calling someone a moron. It certainly was not because I disagree with the author since the little he did say makes me think that we are in agreement about grains.
          Please notice that I included an url in my response so that the original commenter has something to go to besides “Do some research”.
          However, your point is well taken and I will try to control my choice of words…

  10. Chris, look at the Supplementary notes PDF. Do you see the huge fall in glucose and insulin when mice were fed antibiotics?

    Could the gut flora be producing glucose as well as TMAO? If so, and the patient is insulin-resistant, perhaps this is what’s responsible for increased CHD risk?

  11. This is fine. Less people eating red meat ought to mean more for me!

    On a more serious note, this is an excellent discussion of the facts. Much appreciated!

        • Animal production in the US creates more CO2 than all the modes of transportation. Creates even worse gases than CO2 and uses loads of water and land space.
          In my mind, there is little doubt that if people consumed less meat, cheese and milk they would be healthier and thinner as a result of more vegetables and fruits and the country would save bundles on health costs.

  12. What have you got against refined grains? Ever heard about antinutrients and insoluble fiber in whole grains? Also billions of asians eat refined rice as a staple like 60% of caloric intake, and they aren’t very fat, are they?

  13. This was my explanation too at the Reddit thread about this research paper the other day: that the research was conducted on SAD eaters with an altered gut flora, not on Paleo people who eat fermented foods. I went unnoticed, I hope your article gets read a lot.

  14. Thanks for this very thoughtful piece. I will continue to eat red meat in the context of a healthy diet.

    QUESTION: I take an acetyl-l-carnitine supplement per my doctor. Could that contribute to an increase in TMAO?

  15. Thank you for this, Chris! I didn’t get a chance to tell you @ PaleoFX last month, but you have helped me a TON on getting back on the road to health! Also, much like Carrie, you are one of the few people, and literally, the only Doctor, I trust in this world!

  16. Thanks Chris. It’s increasingly difficult to find thoughtful critique and discussion of “science” and nutrition amidst all the hype and posturing going on in the media today. You are on the short-list of people I trust to cut through the crap. Thanks for being a voice of reason in the wilderness.

[if lte IE 8]
[if lte IE 8]