digest
New Trial Results
Scientists just compared two of the most popular diets on Earth head-to-head, keto and Mediterranean, and there was a clear winner. I mean, it wasn't even close. In fact, some of the metrics were several times better on one diet compared to the other.
So, they took pretty unhealthy people, obese, pre-diabetic with fatty liver, and they split them randomly onto different diets. One group went on a ketogenic diet, 4% of their calories coming from carbs and 73% from fat. Another group was on a Mediterranean diet. So this is what we call a traditional Mediterranean diet: fruits and vegetables, some fish, some white meats, some olive oil, nuts and seeds, whole grains like oats, that kind of staples. In terms of macros, it was 50% of calories from carbs and 35% from fat.
And then they actually had a third group, which was a low-fat, plant-predominant diet. They call it a plant-forward diet. It wasn't completely vegan but plant-rich. So the foods were pretty similar to the Mediterranean diet but a little heavier on the carbs and a little lighter on the fat. It was 70% of calories from carbs and only 15% from fat.
Another good thing about this trial is that they provided the meals, prepackaged. They sent them to the participants' homes. That helps with adherence.
Now, here's the interesting twist. They controlled the calories so that the participants' calories were initially cut. They started losing weight on all three groups, on all three diets. But then when they got to about 10 to 11% of body weight lost, they tuned the diets again to stay there, to lock them in so that they didn't lose any more weight. So, this is a very interesting design because it separates the effect of weight loss from other effects that the diets might have. By design, they all lost the same amount of weight. So any differences that we observe are due to other characteristics of the diets.
After about 5 months on these diets, all three groups had lost a similar amount of total body fat, but their liver fat looked quite different. All three groups lost liver fat. And remember, they had fatty liver when they started out, but on the Mediterranean and the low-fat groups, they lost about 45% of their liver fat, which is pretty respectable. But on keto, they lost 68%.
A similar pattern held for insulin resistance: all three groups improved, but more so on keto. And this was very tissue-specific. For muscle insulin resistance, all three groups improved about the same. But for liver insulin resistance, again all three groups improved, but on keto they improved about two to three times more than on the other two diets.
Overall, it was similar for fasting glucose. Fasting blood sugar improved in all three groups, but more so on keto. Fasting glucose dropped six points on the Mediterranean and low-fat diets and 12 points on keto, so twice as much improvement by the end of the trial. Also, 7% of the participants doing the low-fat diet were no longer pre-diabetic. They were in remission. That's what we call it; the colloquial term is "reverse." They reversed, in quotes, their pre-diabetes. They were no longer pre-diabetic. So, 7% on low-fat, 29% on Mediterranean, and 50% on keto. Very striking differences. Some inflammatory markers also dropped more on the keto diet.
Another interesting aspect of the trial was ApoB. ApoB is a metric of your risk of heart disease. It dropped nine points on low-fat, 15 points on Mediterranean, and six points on keto. However, the difference was not significant between the groups. So we can't say that one diet was more effective at lowering ApoB than another.
This might sound surprising, because you might expect a ketogenic diet to be less good for ApoB, especially if it's not the avocado-and-walnut version of keto, which this one was not. They did have healthy fats. They had some fatty fish and some nuts and seeds, but they also had plenty of bacon, sausage, red meats, and things like that. So why don't we see a bigger difference in ApoB? One possibility is that because they were all losing weight, that may mask the effect of the macros and the foods themselves. Regardless, eating to your blood work is always a good principle. If your values look great, that goes a long way.
Reconciling Contradictions
Now, at this point, looking at the results of the study, the fights usually break out in the comments. The Mediterranean fans say that this trial is probably trash and probably funded by the meat industry. And the keto fans say, "Of course keto won. We don't even need trials. We already knew that." I find it a lot more productive and, to be honest, more relaxing to just let studies tell us what to believe and form beliefs based on evidence.
And this study makes me question a lot of things we saw previously, because up until now most trials indicated that when you match for calories and for weight loss, these differences in more carbs, fewer carbs, more fat, or less fat don't usually make much of a difference when it comes to liver fat or fasting glucose, fasting blood sugar.
As an example, there's a really interesting trial from 2022 where they had two diets with very different carbohydrate content, 40% versus 10% of calories. So, a four-fold difference, but the calories were matched and the weight loss was matched, and they both lost liver fat identically. The curves overlapped. So that trial was saying that despite a big difference in carbs, liver fat was similar when calories were matched.
This trial suggests that there's something else in addition to calories and weight loss. Those still matter; all three diets got improvements. But this trial is saying there's something else delivering an extra improvement. The implication here is huge, because it says that just looking at your scale can be misleading. Everybody was losing the same amount of body weight, but what was going on inside their bodies was very different.
So how do we reconcile all these different trials? That's at the end of the day how science works. We don't pick one trial and pretend everything else doesn't exist. We have to look at all the evidence and come up with an explanation that reconciles everything. Both trials have weaknesses. Every trial ever conducted has weaknesses, by the way. Wink wink. There's no such thing as a perfect trial.
This recent trial was pretty small: 44 participants completed it, so about 14 participants per group, 14 on each diet. So we've got to be a little careful before we go crazy. That other trial from 2022 was larger, about 20 to 30 participants per diet by the six-month mark, roughly matching the length of this one. But they started out with a lot more participants, about 200 total, and then they lost a lot of their subjects. A lot of people dropped out. That's a concern in and of itself and a weakness of that trial. Either trial has other weaknesses too.
3 explanations
So how do we reconcile this without pretending one trial doesn't exist? There are a couple of possibilities.
One possible explanation is that it's about the carbs, the kind of explanation staring you in the face. Perhaps there's a key range here. As you go down from 40% to 10%, you don't see much of a difference. But in the sub-10% range, the ketogenic range, something extra kicks in. Maybe it's the total amount of insulin over 24 hours, and maybe it's the effect of these hormones on liver metabolism that's explaining this.
Another possibility is that it's specific to a population, to specific individuals with specific characteristics. Both of these trials looked at obese people, but in this trial, they were a little sicker. There was a specification that they had to have pre-diabetes. The trial from 2022 didn't have that inclusion criterion. So maybe there's something about these individuals where they do better on a very low-carb diet that doesn't apply to other populations. That's always a possibility.
And then there's a third fascinating possibility, which a lot of scientists were discussing in the wake of this trial coming out: that it might not have anything to do with the carbs per se. When we talk about low-carb, high-carb, low-fat, high-fat, we have one change in our minds, but we forget that these diets change a lot of things. There are many moving pieces simultaneously. Your body doesn't care what you call the diet. It cares about what physiological changes are taking place.
In this new trial, the folks on the keto diet were getting a lot more protein than the other groups: 23% of their calories from protein versus 15% on the Mediterranean and low-fat diets. But in that trial from 2022, the protein was matched across the higher- and lower-carb diets. This raises an intriguing possibility: it's not the difference in carbs per se, it's the difference in protein that's having these effects. We know from other trials that increasing the protein content of a diet can help you lose more liver fat, even when calories are matched. So this is a very real possibility.
If that's true, you could go on any diet you like, lower carb, higher carb, Mediterranean or not, and just get the right amount of protein and get all these benefits. This is pretty straightforward to test. You would compare the ketogenic and Mediterranean diets but match them for protein. If the protein is what makes all the differences, the differential benefits should go away. But if the difference in carbs is delivering the extra benefits, they should remain.
Takeaways
There are two major takeaways from this trial, which is somewhat preliminary by itself but fascinating. Number one: all three diets delivered improvements, but the very low-carb, high-protein diet may be better for liver health, at least for some individuals.
Last, but not least, I have to emphasize sustainability, probably the number one factor. The best diet is the one you can actually eat. If you can't stick to a diet, it doesn't matter how good it looks in trials. It's not going to do anything for you. Absolutely keep that in mind as you design your ideal diet for health, especially long term.
What about drinks? What you drink can have a substantial impact on your liver fat, sometimes even when calories are matched. We covered the best drinks to lose liver fat and visceral fat in a previous video, so check that out. I'll see you in there.