We’ve thought for decades that red meat is unhealthy. Earlier this month we found out why, a neat story involving a molecule called carnitine found in meat, metabolized by the bacteria that live in our gut. The resulting chemical, according to studies in mice, is linked to heart disease. Here is the paper. Open and shut case? That would be nice, but no.
There is a larger and more uncomfortable story here, and I don’t mean the fact that they got a vegan to eat a steak for science. (Although I would love to have heard that conversation.)
Here’s what I mean. The carnitine study was shouted from the rooftops, or the New York Times which is basically the same thing. But two more studies that came out in the following weeks resulted in less-than-triumphant coverage: one agreed with the first study, but opened up the complexity of the metabolic pathway involved. No longer just carnitine and TMAO, this one mentioned lecithin, choline, and betaine.
The other study claimed that carnitine, as a supplement, protects against heart disease. This meta-study looked at outcomes in patients who had already had one heart attack, and were assigned to take carnitine or placebo. The supplement was associated with lower risk of death, angina, and arrhythmia – but not heart attacks themselves, which is already a little weird.
Who is right? Well, they looked at different things. MedPageToday published what amounted to a mostly polite argument between the authors of the two studies. The pro-carnitine researcher pointed out that his study concluded that carnitine was good for humans, by studying carnitine and its outcomes in humans, while the other study involved mice, biochemistry, and small sample sizes (one single human being stood in for all of vegan-hood). His anti-carnitine counterpart reminded readers that meta-analyses often produce “apparent statistical results that don’t hold up on further direct experimental testing.”
They’re each right about the flaws in the other’s methodology, but that doesn’t help us much if we’re trying to decide whether to avoid meat or start taking a carnitine supplement. Which brings us to the question of what these chemicals are, anyway.Carnitine is an amino acid found in meat, hence the name, and apart from its possible cardioprotective effects is used as a supplement in energy drinks and for athletes, despite a lack of convincing studies saying that it does anything. (If a chemical can be sold in pill or powder form, you can guarantee that there is an article on bodybuilding.com aiming to convince you, the muscle-seeking reader, that you must buy some. Here is the article for carnitine.)
The risk-raising molecule in the first study was TMAO (Trimethylamine N-oxide), which the researchers concluded was made out of carnitine by gut bacteria. Carnitine-rich meals resulted in high levels of blood TMAO, but if the subjects took an antibiotic first, TMAO levels stayed flat. That was in the subjects that were used to eating meat. The vegetarians and vegans’ TMAO levels stayed low when they ate carnitine, whether it was from a pill or a steak. The mouse component of the study established the link between TMAO and heart disease, at least in one heart-disease-prone strain of mice.
Predictably, fans of meat-rich diets did not like this study. This article from Beef Magazine earnestly appeals to beef’s “good taste” as a counter-argument. More seriously, paleo diet fan Chris Masterjohn took the study to its logical conclusion: if meat is bad for you because it produces TMAO, then other foods that produce TMAO in the body should be bad for you too. These foods include lecithin, found in eggs (and added to foods like chocolate, mmm chocolate) and – the biggest culprits of all – fish and shellfish. They have massive amounts of TMAO; in fact, its precursor trimethylamine is known for its fishy smell. People who cannot convert trimethylamine to TMAO have what’s called fish odor syndrome.
Within a week, the other shoe dropped. The same researchers that published the first carnitine-is-bad study (Stanley Hazen’s group at the Cleveland Clinic) came out with another paper about lecithin. Expanding on a similar 2011 paper, this study associated TMAO with heart disease risk in humans. Once again, food (in this case, egg yolks) resulted in higher TMAO levels. The effect was gone, however, after a course of antibiotics.
What’s really amazing here–whether the results hold up or not–is that we are finally hearing about what gut bacteria do, not just who they are. Many studies on our internal ecosystems are just cataloguing the players, creating a modern-day version of the medieval bestiary without really understanding how the species interact and what they do for us. The meat-eaters, for example, had guts dominated by Prevotella bacteria, but who’s to say Prevotella was the microbe responsible for making TMAO, or even that meat-eating was what caused the TMAO makers to grow in meat-eaters’ stomachs? There is evidence linking Prevotella to consumption of whole grains, so the dietary culprit, if there is one, could be a meat/grain combo.There are hundreds of species in our intestines, many of which we can’t grow in the lab, and the tough questions of what exactly they do, how they interact, and how they make use of our food are not ones we are currently able to answer with any sense of completeness. We have crude tools that have only advanced a little since the must-read parables of the geneticist and biochemist. (I’ll wait while you read the whole thing, but here’s a taste: Trying to understand how cars are made, the biochemist grinds one up and announces that cars are 10% glass, 25% plastic, 60% steel, and 5% other. Meanwhile, the geneticist ties the hands of workers going into the factory and examines how and whether the resulting cars crash. He concludes that seatbelts are vestigial since the cars still function without them.)
The other reason these studies are hard to wrap our heads around, aside from the sheer complexity of the question (what else was in that steak, anyway? Does it matter if it was grass-fed or factory farmed? How similar is one carnivore’s gut microbiome to another’s?) is that we like answers we can act on now. I can read a study about steak today and switch to veggie burgers tomorrow. Messing with diet is popular because it’s totally within our control and we can do it three times a day. Popping a pill is not much more of a commitment, but understanding the metabolic pathways of a collection of symbiotic organisms? That’s going to take a little longer.