https://www.youtube.com/embed/CjL41H_Dz4k
a brand new study just came out comparing
a ketogenic diet and a mediterranean diet, it's called keto-med (appropriately) and it's
led by christopher gardner's lab at stanford. the main question they were trying to answer
was, which diet is best for people with diabetes or pre-diabetes? both diets were designed to
be clean, meaning that the participants were asked to avoid junk food, refined grains and added
sugars, and both included non-starchy vegetables like leafy greens.
The main differences were, the
mediterranean diet included legumes, whole fruit and whole grains, while the keto diet avoided
most of those foods, except for maybe some berries. the mediterranean diet also had
plenty of olive oil, characteristically, and the main animal product was fish, whereas the
keto diet had more animal foods like meat, eggs, cream etc. now, the macronutrient breakdowns
were approximately 18% protein, 39% carbs and 45% fat on the mediterranean diet,
so fairly typical, and on keto 24% protein, 12% carbs and 62% fat.
Now, i already know there's
some viewers saying it's not low carb enough, because every time we make a video about
a trial that's either low carb or low fat, regardless how low they go, there's 50 comments
saying wasn't low enough. joking aside, it was actually pretty solid, the people on keto
were asked to stay under 50 grams of carbs a day and they actually averaged 43, and they actually
had an objective measure of ketosis, they measured the level of ketone bodies in the blood of the
participants. they asked them to try to stay above 0.5 millimoles per liter and most of the time they
succeeded, they were between 0.5 and 1.5 which the authors consider light ketosis, above 1.5 being
optimal ketosis. we can compare this for example to the virta study, which is a study run by a
company that specializes in low-carb diets and it's actually overseen by some of the intellectual
leaders of the low-carb/keto movement.
In Virta they were also shooting for ketones above 0.5,
same range as the keto-med trial we're looking at, so that seems reasonable. in Virta the
participants stayed in average right above 0.5 for the first six months, by one year most were
under 0.5, so no longer in ketosis technically, and by two years only 14% were in ketosis.
now, Virta is much longer than keto-med but even early on they were right above 0.5, just
barely making it into this light ketosis range, so it's just difficult to get people to
adhere to these strong elimination diets, this is not specific to low carb. we had
a recent video where we looked at a trial comparing mediterranean to a diet they called low
fat and we saw that they had a really hard time getting the participants to significantly cut
back on the fat so this is just something we see consistently. so the issue is always that you're
not dealing with mice in cages so there's a limit to how much experimental control you have.
On the
other hand, when we're just choosing a diet for ourselves, this aspect of populational adherence
matters much less, the only factor is whether you can stick with it it, doesn't matter how many
other people can do it. okay, back to the trial, so they looked at 33 people that were either
diabetic or pre-diabetic, so it's not a large number of subjects, so that's a caveat to bear in
mind, although they used what's called a crossover design, where each participant goes on both
diets sequentially, so in a way you get double the amount of data. so each participant went on
one diet for 12 weeks, then the next diet for another 12 weeks and finally 12 weeks of follow-up
where they didn't get any dietary instructions so they could eat what they wanted, and the order
was chosen at random so approximately half of the participants did the keto diet first and
then the mediterranean, and the other half did the mediterranean first and then the keto.
the main parameter they were interested in, what's called in nerd speak the primary outcome,
was glycated hemoglobin or hemoglobin a1c, as it's also known, and that's an average of the
glucose levels over the previous three months. both diets lowered hemoglobin a1c compared
to the baseline, right before the experiment, and there was no significant difference between
the two.
Now, honestly when i read that at first it was surprising to me, i expected both diets to
generate an improvement but i thought keto would have an edge. in type 2 diabetes the carbohydrate
processing machinery is sluggish so there's two things we can do, in the short term we can avoid
excessive or supraphysiological glucose peaks while the machinery is still defective. that
doesn't solve the underlying issue but it avoids a problematic manifestation. the other
thing that can be done is to fix the underlying issue so the machinery works normally, and the
main lever to do that shown in randomized trials is to moderate overall caloric intake and to lose
a little bit of the excess fat mass that people tend to carry, we covered all of this in a recent
video where we talked to professor roy Taylor, a diabetologist from the UK.
So any diet that helps
with weight loss can help with the function of the machinery, with the underlying issue, this is
why a lot of times you'll see some people saying "a low-fat diet reverse my diabetes" and somebody
else says "a low-carb diet reversed my diabetes" or "a vegan diet" or "a paleo diet"... they're not
lying, the common theme is they all eliminated our moderated ultra processed foods, calorically
concentrated foods, so that helps moderate overall caloric intake. so i expected both diets
to deliver a benefit because they're both clean diets, they're both low in ultra processed foods,
and i thought keto would also help moderate the postprandial glucose peaks in that initial phase
while the underlying machinery is still sluggish, just because there's less glucose being
eaten, and in fact diabetes is one context where there's quite a bit of data showing that
low carb diets can be a valid tool to help with glycemic control.
Yet overall, the results
showed no statistically significant difference between the two diets, but i think when we look a
little deeper it does suggest a difference. see, in order to avoid hypoglycemia, in order to
avoid their glucose levels from crashing too low, the participants who were on glucose lowering
medication were asked to reduce it before getting on the diets. now, that's pretty standard in this
type of dietary intervention trial, the problem is the participants who were about to go on the
mediterranean diet cut their medication in half while the ones that were about to go on the keto
diet completely eliminated it, so it's possible that the drugs that were still left helped the
mediterranean diet perform a little bit better, push the glucose a little bit lower, whereas
the people on keto didn't have this extra help. to the investigator's credit, they ran a
separate analysis where they matched for a dose of glucose-lowering medication, and when they
did that, the results do suggest an edge for keto being able to lower hemoglobin a1c a little
bit more than the mediterranean diet. now, the numbers of participants in that analysis are very
low so we have to be cautious, but the results do suggest somewhat of an edge.
They also asked the
participants to wear a continuous glucose monitor, which looks at the dynamics of glucose levels
over the course of the day, and here again they saw that both diets improved glucose control,
with keto performing a little better. so overall, both diets clearly beneficial compared to baseline
and the results indicate that the ketogenic diet may have an edge with regards to glucose control,
at least in this time frame, and remember, the participants were in light ketosis, so it's
possible if they were in deeper ketosis the results would have been even stronger. possible.
the researchers also looked at blood lipids, and this was interesting, the ketogenic diet was
better at lowering triglycerides but it was worse for ldl-cholesterol.
Now, that's a pretty common
pattern in low carb trials but it's by no means a necessity, and we'll come back to this in a
minute. they also looked at body weight. both diets delivered weight loss, it seemed like maybe
a bit more on keto, it wasn't completely clear, if you remember, it was a crossover study, people
who did keto first did lose significantly more weight than people on the mediterranean
diet but the people who did keto second, there was no significance there, so it wasn't
completely clear, there's a bit of uncertainty, but for me looking at it, it does look like
there's a trend for keto to lose a little bit more weight. now, ideally we'd want to know body
composition, how much of the weight is actual fat mass versus lean mass, we know for example that
with low carb diets it's common to see early on a loss of glycogen and water, so we want to know
how much of the lost weight is actual fat mass, but those details were not reported.
And finally,
they looked at what the participants chose to eat in that follow-up phase when the pressure is
gone and they get to choose, and they found that they generally gravitated more towards
a mediterranean-ish diet and not so much a low-carb diet. and the researchers suggest this
indicates more sustainability long-term or maybe a higher personal preference in default conditions.
so bottom line, i really like this trial, it shows us clearly that each diet had something to offer
and neither one was perfect, mediterranean lowered hemoglobin a1c but keto probably lowered it a
bit more, and it was better for cgm parameters, but it was worse for cholesterol, mediterranean
was better for cholesterol, but keto was better for triglycerides and mediterranean might be
a bit more sustainable long term. now, at face value we might think, okay so i just got to pick
one diet and compromise on the shortcomings, but i think this trial and all the
evidence we've been looking at lately are giving us a much deeper and much more
empowering message, that we can tailor our diet to our needs using different features
from all these different dietary patterns, that we don't have to compromise on health, and we
shouldn't.
Let's say i prefer a low carb diet and i'm actually one of the people who can sustain
it long term and i do well on it, do i have to expose myself to high ldl-cholesterol, am
i stuck with that? no! in this trial, people on keto were eating twice the saturated fat of the
mediterranean diet and half the fiber, they ate as much fiber as the average american on the standard
american diet so it's entirely predictable that their cholesterol would go up, but it's completely
doable to design a low-carb diet, even a ketogenic diet, with more fiber, less saturated fat and more
unsaturated fats, and there's lots of precedent for this, studies have shown a low-carb diet
rich in unsaturated fats helps lower cholesterol, not increase it, so this general idea that
low-carb diets jack up your cholesterol and that's something you just have to live with is not a
necessity at all, it's basically a misconception. i've been saying this for like two or three years
now, i have old videos where i go over this.
Real quick aside, ldl-cholesterol is a marker of apob
lipoproteins that carry cholesterol, and high apob raises cardiovascular risk, the evidence on that
is abundant and unequivocal, we've covered that in a lot of previous videos. as far as the different
types of fat, saturated fat generally refers to fatty meats, butter, coconut oil etc and
unsaturated fat, it's things like nuts and seeds, fatty fish and most vegetable oil. okay, but
maybe there's a disadvantage to unsaturated fats, maybe they're better for cholesterol and
for cardiovascular risk but they're worse for triglycerides or glucose, and maybe it's
a trade-off? we actually have trials comparing different keto diets, same proportion of protein,
fat and carbohydrate, just different types of fat. high saturated fat keto versus polyunsaturated
fat keto, and what this trial found was that people on the poly unsaturated fat ketogenic diet
had lower cholesterol, that's not surprising, but also lower triglycerides, lower glucose and
better insulin sensitivity.
The authors concluded "a polyunsaturated fat ketogenic diet may
be superior to the classical saturated fat ketogenic diet for chronic administration".
so yes, you can get the best of both worlds. and speaking of chronic administration, this is
another caveat that always has to be brought up, we don't know what happens long
term on very low carb diets, five years, 10 years, 20 years on 10 or 5 percent
of calories from carbs, what happens to disease incidence rates. low carb diets have been around
for 100 years or more but we don't have cohort studies with reasonably matched populations over
the long run. i'm sure those studies will come eventually, until then it's a bit of a question
mark. it doesn't mean people can't do the diet, it's just something that we have to bear in mind.
okay, so the keto diet is completely optimizable, the exact same thing goes for the mediterranean
diet, if i'm on a mediterranean diet and i like it but my triglycerides or my glucose aren't quite
where i want them, nothing stops me from tweaking, from for example taking a percentage of the
whole grains on my plate and replacing that with some unsaturated fat or some protein, that
could be nuts and seeds, some legumes, some fatty fish...
So understanding these simple principles
allows us to fine-tune our diet to fit our needs, whether it's low-carb, high-carb, intermediate
carb, without having to go through these whiplash changes from one extreme all the way
to the other. we marry the science with the individual circumstances to personalize and give
the person the best odds, right? we don't ignore individual circumstances and we don't ignore
the science either. and as we said, avoiding the glucose spikes doesn't necessarily mean the
underlying issue is resolved, I could still be insulin resistant and diabetic technically
and just not see the manifestation if there's no carbohydrate challenge to the system, so for
type 2 diabetes, bearing in mind for most people, losing a little bit of that extra fat mass is
going to be beneficial, so whichever diet keeps you satiated and is sustainable long term is going
to help you achieve and maintain a healthy body weight and improve your glucose metabolism.
some people do better on a low carb diet, some on a low fat diet and someone neither.
We see
this individual variation in studies and i hear it every day from you guys in the comments, so
it's crucial to give people some options, some healthy options, so they don't have to
fall prey to crazy fads, the all-bacon diet or the rice and sugar diet, just because those
help lose weight. and yes, both of those are real things i've seen on the internet. the authors
say it themselves: "there should be less focus on promoting one particular diet as best and rather,
allow patients to make an informed choice to help them establish which approach is most suitable for
them".
A diet should not tie your hands. the diet should serve you, not the other way around. i see
people all the time struggling to stick to these diets and compromising, "oh this diet is good for
this but it's bad for that, oh well"... you don't have to play that game, you don't owe allegiance
to a diet, why shouldn't you take something good from both sides if it helps you? in fact, people
have already coined the mediterranean keto diet, a blend of both. the internet can convey this
idea that diets are rigid structures, there's the low carb diet and the low fat diet and if you
deviate, you're a traitor and a failure.
These are internet constructs, if your diet is giving you
the results you want short-term and long-term, optimizing your health overall and you like it,
stick with it. if not, tweak it, change it. yeah, you can have your cake and eat it too. even if it
is a low carb cake. so what are your thoughts on this trial? did it give you some new ideas? let
me know below, take care, see you next time, bye
