00:00
Hey everyone, it's Mikki here. You're listening to Mini Mikkipedia on a Monday. And today I want to chat about the potential pitfalls or harms, if you like, of a high protein diet and what the research actually shows. So this was a topic of conversation I was at. I've explored before with Brandon de Cruz and I was intending to chat to Eric Helms about this because

00:29
Eric did a really great debrief in the mass research review last year based on the paper that I'll be discussing today. But of course, Eric and I get on the call and he's in the midst of his bodybuilding shows and things like that. So, and he had just won his pro card. So we ended up actually just chatting a lot about that. So I thought this would be an opportune time to actually bring this to your attention, this paper.

00:57
because I get questions all of the time about the harms of a high protein diet, particularly because, as you know, this is a dietary recommendation that I advocate for for almost everyone that I talk to. And if you spend any time on social media or even reading mainstream health journalism, you'll see headlines like, high protein diet could be harmful, even for your healthy kidneys, or claims that excess protein just turns to fat, or that it raises your risk of diabetes.

01:27
And I understand why people get concerned because these claims sound authoritative and they get repeated so often that they start to feel like they're established facts. So this is what I want to unpack today on the basis of an excellent perspective paper, as I mentioned, that Eric did a great debrief about that was published in the American Journal of Clinical Nutrition last year by Stephen French, Mitchell Cantor, Kevin Markey, Brett Rust and

01:55
David B. Ellison, who is a real authority in the nutrition space. The title pretty much tells you where they land. And the title is, The harms of high protein intake conjectured, postulated, claimed and presumed, but shown? And that question mark at the end is doing a lot of heavy lifting. What I really liked about this paper is that it doesn't just wave the harms away. It goes through each of the major claimed risks.

02:21
kidney damage, bone loss, diabetes, reduced longevity. And it actually asks, what does the research show? And importantly, it flags the methodological problems that make so many of these claims look stronger than they really are. And what I will also say is that none of these guys have anything to lose from outlaying the facts. There's no hidden agenda here. They're not protein metabolism experts. They're not protein metabolism researchers.

02:50
not plant-based. They really don't have a reason to portray the evidence as anything other than what it actually is. So first, let's have a look at some context. As you know, the recommended dietary allowance in the States for protein, also in the UK, Australia, and New Zealand, is about 0.8 grams per kilogram per day. Now, the dietary guidelines

03:16
since his paper has been published in the US is now recommending sort of 1.2 to 1.6 grams that is different from the RDA, just so you know. And this is a minimum, that 0.8 grams. It's the amount considered sufficient for basic bodily function for the majority of athletes derived primarily from nitrogen balance studies. And, you know, it was never intended as an optimal intake for health, body composition or performance. Not that you would know that.

03:46
looking at the health authorities and how they sort of discuss protein. Anyway, there is another number as well, the acceptable macronutrient distribution range, the AMDR. And this allows for protein actually to make up between 10 to 35 % of total caloric intake. This is in the States. In New Zealand and in Australia, that's actually set at 15 to 25%.

04:12
That's an aside. This paper is based in the US and they're talking about recommendations there. If you do the maths on the upper end of that recommendation in the paper, for a typical energy intake, you're looking at somewhere between 1.05 to 3.67 grams per kilogram per day. This is based on an energy intake of around 36.5 calories per

04:40
kilogram per day, puts someone in this sort of enough available energy to maintain their weight. So that's where they get their numbers from. So even quite high protein intakes technically fall within this accepted range. And for what it's worth in the Australian and New Zealand acceptable macronutrient distribution range, the range would roughly work out to be 1.37 to 2.28 grams per kg body weight.

05:09
So while it's a noticeably narrower range, the upper end is still well above that RDA and comfortably within what most people I would recommend eat. So I guess that's the first thing to sort of point out is that even within the actual guidelines, there is allowance for quite high intakes. And the International Society of Sports Nutrition positions intakes of around 1.4 to 2 grams per kilogram per day is sufficient for building and maintaining muscle mass.

05:38
for most exercising individuals. And they also note emerging evidence that intakes above three grams per kilogram may actually have positive effects on body composition in resistance-trained people. And if you listen to the two ERIKs on iron culture discuss a recent meta-regression based on protein intakes, they do talk about this sort of, you know, that to get the most out of protein from a

06:06
muscle perspective, you'd want to be at about 1.9 grams per kg body weight, although there is evidence of even higher intakes up into above three grams per kg body weight of lean mass actually. So there's a whole range of different quite high protein intakes that are recommended based on either public health authorities, sports science societies, or just people who know what they're talking about. when it comes to protein,

06:37
a lot of people are eating is what would be classified as high protein. And the question that French et al were wanting to answer is whether or not this is a problem. So first let's start with one of the claims that you often hear is that excess calories turns to fat, therefore excess protein turns to fat. Now there is a kernel of truth here. So excess protein can contribute to de novo lipogenesis. That's the

07:05
creation of fat tissue, but this depends heavily on context, the degree of protein excess matters, and critically, of course, the energy status of the individual matters. The authors point to studies showing that in an energy deficient state, which is where many of my people sit because I'm helping them with fat loss, an increased protein content of the diet can actually decrease de novo lipogenesis or the creation of fat tissue or fat cells. So the blanket statement that protein turns to fat

07:35
strips away any of that nuance. In a deficit, higher protein helps preserve lean mass and may actually reduce fat synthesis. So essentially the opposite of what you hear. And this is a recurring theme throughout this paper that I want you to hold on to. Context matters enormously and sweeping statements about protein being harmful for health almost always ignore context.

08:00
Outside of this paper, I do think it's worth looking at the controlled overfeeding data, just for an instance, because it really does reinforce this point. The most cited work comes from Jose Antonio's lab, and I've had Jose on the show before, and Jose and I chatted about endurance athletes he worked with, whereby their protein intakes were around three grams per kg body weight per day. And across several studies in resistance trained individuals,

08:27
protein intakes of 3.4 to 4.4 grams per kilogram per day over 8 to 12 weeks consistently showed no meaningful gain in fat mass compared to control groups eating around 1.8 to 2 grams per kilogram, despite the high protein groups being in a significant caloric surplus. And in one crossover study running a full year at roughly 2.5 to 3.3 grams per kilogram, there were again no adverse effects on body composition,

08:56
kidney function markers or blood lipids. So the implication is that excess calories from protein don't appear to be stored as fat the way excess calories from carbohydrate or fat are. This likely comes down to the thermic effect of protein. Around 20 to 30 % of energy from protein is expended just in digesting and metabolizing it and in that muscle protein synthetic response, compared to roughly 5 to 10 % for carbs and around 1 % for fat.

09:26
Now, of course there are caveats. The Antonio studies are in resistance-trained men, the sample sizes are small, and dietary reporting relies on self-report at very high intake levels. Which clearly is always something to be mindful of. The pattern is consistent. And the broader overfeeding literature does tell a complementary story. So the Bray-Ed-El study published in the Journal of the American Medical Association in 2012 overfed

09:53
sedentary participants about 40 % of their caloric needs at different protein levels. 5, 15, 25 % of energy from protein for 8 weeks. Note that that 15 and 25 % of energy from protein sits in that acceptable macronutrient distribution range. All groups gained weight, but the low protein group gained significantly less lean mass. Not muscle mass, but lean mass.

10:21
the normal and high protein groups gained more lean mass. Importantly, all three groups gained similar amounts of fat mass, so excess calories still led to fat gain, regardless of the protein level, the higher protein groups partitioned more of the surplus towards lean tissue. And I'd say that the difference between Braves findings and Antonio's likely come down to the population. Resistance training plus high protein appears to create conditions

10:50
where more of the surplus gets directed towards MPS, muscle protein synthesis, and thermogenesis, rather than adipose storage. And in sedentary populations, excess protein still contributes to overall weight gain, but favors lean mass accrual over fat storage compared to lower protein at the same surplus. So the protein turns to fat claim doesn't hold up well under scrutiny, regardless of whether we're looking at French air tales review or the wider overfeeding literature.

11:20
One other thing I will say on this is that oftentimes if you are increasing your protein intake and gaining fat, then proteins' satiating effect isn't allowing you to drop your calories to a point where you could lose body fat, or the foods that you're eating that you think are high in protein aren't actually that high in protein and they might have more of their calories coming from fat. And I'm thinking of things like pork belly, sirloin, scotch, I mean delicious, obviously.

11:49
cheese, nuts, like these things are not protein sources. are much more fat sources. So that's often what um I see clinically anyway. All right, well, let's get to the kidney damage story. And this is a persistent myth that high protein damages your kidneys. This one has been around for decades and traces back to what's known as the Brenner hypothesis. The idea that high protein intake increases urea production, which increases

12:19
filtration rates of the kidneys. And over time, this hyper filtration leads to kidney problems. And yes, the acute increase in estimated glomerular filtration rate, GFR, from higher protein diets is well documented in both human data and in animals. In animal models, high protein diets have induced renal hypertrophy alongside that hyper filtration. And in rats specifically,

12:47
Chronic high protein diets have increased urinary protein excretion and cause sclerotic changes in the kidney cells. But here's the critical point. The authors note that there are no reported data on acute pathological effects of high protein diets in human kidneys. And when we look at the higher quality evidence in humans, meta-analyses, and even a Mendelian randomization analysis based on the UK Biobank data,

13:16
there's been no causal effect of protein intake on kidney dysfunction in the literature or in the data. An umbrella review of systematic studies looking at protein intake and the risk of kidney diseases found no convincing evidence for detrimental effects. The authors of that review concluded that the changes observed in kidney function markers mostly represent physiologically regulatory responses, so the kidneys are adapting to a higher protein load,

13:45
rather than pathological damage. Now, we do want to caveat this with the fact that most of the studies are relatively short in duration, so truly long-term effects over decades can't be completely ruled out, but the evidence we have currently is quite reassuring. In one six-month intervention study comparing a 25 % protein diet with 12 % protein diet, GFR went up in the higher protein group by about 5%, and kidney volume changed proportionately.

14:14
So the filtration rate per unit of kidney tissue was unchanged. Serum creatinine and urinary albumin were also unchanged. Both of these are important markers which combined may flag actual kidney damage. Interestingly, a recent systematic review and meta-analysis of observational studies found inverse associations, meaning higher protein intakes were associated with lower risk of developing chronic kidney disease. Now observational trials

14:44
cannot be used to assess cause and effect. They can only be used to show associations. So the author's position on this one is that for people without pre-existing kidney disease, the evidence currently rules out this alleged harm for practical purposes. Obviously, if someone has diagnosed kidney disease, it may be a different conversation because kidneys need to do more work if you have a higher protein intake.

15:13
but for otherwise healthy individuals, the data just doesn't support the fear. Now what about bone loss? The theory here is that a high protein intake causes hypercalciuria. What this is, is increased calcium excretion in the urine. And I see this a lot actually on plant-based accounts. And this is because the oxidation of sulfur-containing amino acids creates an acid load that drives the bone resorption to buffer it. And yes,

15:40
High protein diets have consistently been associated with hypercalcauria. But, and this is really important, when you look at actual fractured data, the picture is essentially reversed. Few observational studies have linked high protein intake with increased bone fractures. In fact, more studies show an inverse relationship, meaning the higher the protein intake, the lower the fracture risk. The authors reference the expert consensus from Rizzoli and colleagues.

16:09
which is endorsed by the European Society for the Clinical Aspects of Osteoporosis and the International Osteoporosis Foundation. Their conclusions are pretty striking. In older people with osteoporosis, a higher protein intake above that 0.8 grams per kilogram is associated with a higher bone mineral density, a slower rate of bone loss, and reduced risk of hip fracture, provided that calcium intake is adequate.

16:36
So protein supplementation attenuates these age-related decreases in bone mineral density and reduces those bone turnover markers. And there is no evidence that diet-derived acid load is actually deleterious for bone health. In fact, the consensus position is that insufficient diet protein may be more of a serious problem for the elderly than protein excess, as long as calcium intake is adequate.

17:03
So the authors conclusion here is that for bone mineral density, there is no plausible evidence that a higher protein intake has any negative impact. And if anything, the majority of the data point in the opposite direction. So this really is one where the evidence seems sufficient to rule out the alleged harm for practical purposes. Now, type 2 diabetes, this is one where the picture may be more complicated and the authors do acknowledge that.

17:33
Observational studies have shown associations between higher total protein intake, particularly animal protein, and increased insulin resistance, pre-diabetes and type 2 diabetes risk. Meanwhile, plant protein tends to show either reduced or no association. And there are some plausible mechanisms. Amino acids, particularly leucine, which we know is the amino acid that initiates muscle protein synthesis.

18:01
does it because it activates something called mTORC1, the mammalian target of rapamycin complex. And this plays a role in regulating insulin signaling. Protein and amino acids can also stimulate glucagon secretion. And glucagon's role in the body is primarily to raise blood glucose by signaling the liver to break down glycogen and to produce new glucose. So it essentially works as a counter regulatory hormone to insulin.

18:30
insulin lowers blood glucose, glucagon raises it. And there's an interesting intervention study by Weckert and colleagues that the paper discusses. They compared a high protein diet, about 28 % of energy from protein, with a high cereal fiber diet at around 15 % of energy from protein, an overweight adults with features of the metabolic syndrome. After six weeks of intensive dietary intervention, insulin sensitivity

18:57
decreased in the high protein group compared to baseline, or increased in the high serial fiber group. So it does sound pretty concerning, right? But this is where there's a little bit of nuance. The authors of this paper point out that most of the evidence linking amino acids to mTOR complex 1 activation comes from studies using free amino acids or isolated protein sources, which produce a rapid transient spike in

19:26
amino acidemia. That doesn't reflect what happens when you eat actual mixed meals or protein-dense whole foods like chicken, eggs or beef. And when you look at meal combinations and protein-dense foods, activation of mTOR complex 1 in its downstream targets are much weaker. And then if you look at randomized controlled trials of actual dietary patterns, we've got systematic review and meta-analysis of trials comparing diets containing

19:54
red meat versus diets with less or no red meat that have found no significant effects on insulin sensitivity. HOMA-IR, which is a marker of insulin resistance, got fasting glucose or fasting insulin. And in individuals with type 2 diabetes, there were actually small, marginally significant improvement in insulin sensitivity with higher red meat intake. And when you look at observational associations, you've always got to counter what else

20:24
comes along with the high animal protein intake. And honestly, in population-based research, you see that the people who eat the most red meat are also ones, the people that tend to have more overweight or obesity, they eat less fruits and vegetables, they drink more alcohol, they're more likely to smoke, they're less likely to be physically active, and that red meat comes in between a white bun,

20:51
and with fries and a full-sugared soda on the side. So often what we're seeing is that it's a problem with overall diet quality, processed meat consumption, and those other lifestyle factors rather than protein itself. A meta-analysis found that red and processed meat were associated with increased type 2 diabetes incidence, but soy and dairy were inversely associated, and egg and fish showed no significant association.

21:19
All of that to say it's not protein per se that's driving the risk signal. And in a really well-controlled feeding trial, clinical trial, where all the food was provided, replacing added sugars and refined starches with egg protein and unsaturated fats led to a 24 % improvement in insulin sensitivity. And as the authors discuss, most of that improvement was probably from displacing added sugars and increasing in saturated fat.

21:47
But the point is that higher protein intake did not prevent the benefit. And I would also say the same if you look at the Virta Health trials, which weren't included in this particular paper by French. Whilst those trials are looking at a ketogenic diet, they have a substantial component of their diet come from protein as well. And we only ever see improvements here. So the authors position that while there are plausible mechanisms and some observational associations, a direct

22:16
harm on high protein diet on type 2 diabetes has not been demonstrated. this one, the authors say, sits in the category of plausible with mechanism, but not proven when we look at the totality of the data. And what about mortality and longevity? And this one gets a lot of air time, partly because of animal data on caloric and protein restriction that has been shown to extend lifespan. In animal models, some researchers have concluded that

22:45
protein restriction is more important than caloric restriction for longevity, but others, looking at a subset of rodent studies, concluded the opposite. An increased lifespan was primarily related to reduced energy intake rather than reduced protein specifically. And then of course this gets further complicated if we look at that protein leverage hypothesis. This is when animals are given ad libitum access to low protein food, they tend to overeat total calories to meet a protein target.

23:14
muddies that interpretation. We've also seen these actually in clinical trials in humans. And in humans, there are some observational studies that show that higher animal protein intake is positively associated with overall and cardiovascular mortality. Plant protein generally doesn't show this association. When you look more closely, the lack of association is mainly driven by the fact that protein from grains and potatoes isn't linked to mortality or protein from legumes, nuts, and

23:43
vegetables is actually associated with lower risk. And again, I do keep coming back to this. These are observational studies with all of the limitations that come with them. Dietary recall data is notoriously unreliable. And researchers are asking people about their diet across several years, and they may be asking them just once or twice.

24:05
There's that residual confounding as a real problem, as it's very hard to separate the effects of animal protein from those other diet and lifestyle factors that travel alongside it. And I already mentioned what those were earlier. There's also significant challenges in extrapolating from animal studies to humans. One paper the authors referenced showed that higher blood glucose was associated with lower mortality in mice, but higher mortality in non-human primates and humans. And higher body fat was associated with

24:35
better survival in mice, but worse outcomes in primates. So the metabolic responses don't translate neatly across species. So the authors land on this one, that high protein reduces lifespan, as implausible based on current evidence. It means that it can't completely be ruled out, but the evidence suggests it's unlikely that protein intake per se is driving mortality. More work is needed to disentangle

25:02
total protein from animal protein and from other confounders. Now do want to spend just a few minutes on what I think was a valuable part of this paper, and that's these methodological concerns. First of all, you've got cross species generalization. As I just mentioned, we need to be really careful about taking findings from mice and rats and applying them to humans. The metabolic responses can be fundamentally different. And most good scientists would say that

25:31
Mechanistic trials and preclinical trials that use animals are so useful and so helpful, but that's just one part of the research. We can't take it as proof that this is what happens in humans. Second, is the issue of how protein intake is scaled. This is subtle, but it's an important point. So some studies normalize protein intake to an ideal body weight based on a BMI of 22. But this artificially inflates the grams per kilogram figure

26:01
for anyone with a BMI above 22. So a 100 kilogram person with a BMI of 30 who eats 100 grams of protein per day gets assigned an intake of 1.37 grams per kilogram instead of 1.0 because the calculation uses their ideal weight of 73 rather than their actual weight. So this doesn't just distort the numbers.

26:27
It can actually change which intake category a person is placed into in the statistical analysis. And that person's excess adiposity might itself be contributing to the outcome being measured, like reduced glomerular filtration rate when we're looking at the kidney research. Third, and this is an interesting one, I really liked it.

26:50
This is the grandfathering of knowledge. This is where older claims get cited and recited until they take on the weight of established fact, even when the original data doesn't support the claim being made. side note, this is 100 % that eight glasses of water claim that you hear. Absolutely not. It all backed up by research. Anyway, the authors in this paper gave a great example too. A widely cited statement that the tolerable upper limit for

27:18
protein is 3.5 grams per kilogram per day and that prolonged intake above 2 grams per kilogram should be avoided. But when you go back to the actual source being cited, the data didn't show that 3.5 grams is the upper limit. In fact, the cited paper discusses that Antonio trial where resistance-trained individuals could consume 4.4 grams per kilogram per day for eight weeks without side effects beyond GI symptoms. The claim and the citation don't match.

27:47
but the claim keeps getting repeated. And I will say the number of research papers that I have read where I've gone hunting down those citations only to find that what the authors claim is absolutely not backed up by the citation is innumerable at this point. And finally, of course, confirmation bias, because don't we all have one? Just because a mechanism is plausible doesn't mean the harm is real. The authors make the important distinction that

28:16
plausible and conceivable are not the same thing as true or demonstrated. And when a claim seems reasonable, it's very easy to to selectively cite evidence that supports it while ignoring evidence that doesn't. And this is how myths get perpetuated, not through bad science necessarily, but through insufficiently rigorous interpretation. So this is where things stand. For kidney damage and people without pre-existing kidney disease,

28:43
the evidence seems sufficient to rule out the alleged harm for practical purposes. i.e. no big deal, protein is fine. For bone loss and increased fractional risk, the same conclusion. In fact, higher protein appears protective provided calcium intake is adequate. Type 2 diabetes, the harm has not been demonstrated, but existing evidence suggests plausibility. So we may not be able to fully dismiss the concern, but nor can we say it is proven

29:12
given the totality of data. For reduced longevity and increased mortality, the alleged harm can't be ruled out, but existing evidence suggests it's absolutely implausible. And importantly, for none of these claims did the authors find that the evidence strongly supports the existence of the alleged harm, not one. So if you're an otherwise healthy individual eating a high protein diet, and by that I mean somewhere in the range of, you know,

29:41
1.5 to 2.8 grams per kilogram or even higher if you're resistance training or wanting to lose body fat, etc. The current evidence doesn't support the claim that you're damaging your kidneys, weakening your bones, or meaningfully increasing your risk of diabetes. Now look, does this mean there's zero risk at any level of intake forever? Well, I mean, no. Long-term data though at very high intakes is limited and there are genuine research gaps that need to be filled. The burden of

30:10
proof should sit with those making the harm claims right now that burden hasn't been met. So what I take from this paper and what I hope you take from this discussion is the importance of looking past the headlines and the social media some bites. Context always matters. Study design matters. The difference between association and causation matters and the difference between plausible and proven matters. So

30:36
That's what I've got to say on that. Hopefully you found that interesting and maybe somewhat helpful. Let me know. I am over on Instagram threads and X @mikkiwilliden Facebook @mikkiwillidennutrition. Get to my website, @mikkiwilliden.com. Book a one on one call there with me. All right guys, you have the best week. See you later.