Ketogenic Low-Carbohydrate Diets have no Metabolic Advantage over Nonketogenic Low-Carbohydrate Diets – Research Review

Title and Abstract

Johnston CS et. al. Ketogenic low-carbohydrate diets have no metabolic advantage over nonketogenic low-carbohydrate diets. American Journal of Clinical Nutrition. (2006) 83: 1055-1061

Background:Low-carbohydrate diets may promote greater weight loss than does the conventional low-fat, high-carbohydrate diet.  Objective:We compared weight loss and biomarker change in adults adhering to a ketogenic low-carbohydrate (KLC) diet or a nonketogenic low-carbohydrate (NLC) diet.  Design:Twenty adults [body mass index (in kg/m2): 34.4 ± 1.0] were randomly assigned to the KLC (60% of energy as fat, beginning with 5% of energy as carbohydrate) or NLC (30% of energy as fat; 40% of energy as carbohydrate) diet. During the 6-wk trial, participants were sedentary, and 24-h intakes were strictly controlled.  Results:Mean (±SE) weight losses (6.3 ± 0.6 and 7.2 ± 0.8 kg in KLC and NLC dieters, respectively; P = 0.324) and fat losses (3.4 and 5.5 kg in KLC and NLC dieters, respectively; P = 0.111) did not differ significantly by group after 6 wk. Blood ß-hydroxybutyrate in the KLC dieters was 3.6 times that in the NLC dieters at week 2 (P = 0.018), and LDL cholesterol was directly correlated with blood ß-hydroxybutyrate (r = 0.297, P = 0.025). Overall, insulin sensitivity and resting energy expenditure increased and serum -glutamyltransferase concentrations decreased in both diet groups during the 6-wk trial (P < 0.05). However, inflammatory risk (arachidonic acid:eicosapentaenoic acid ratios in plasma phospholipids) and perceptions of vigor were more adversely affected by the KLC than by the NLC diet.  Conclusions:KLC and NLC diets were equally effective in reducing body weight and insulin resistance, but the KLC diet was associated with several adverse metabolic and emotional effects. The use of ketogenic diets for weight loss is not warranted.


Pretty much without fail, every decade seems to see the resurgence of the idea that extremely low-carbohydrate diets have a ‘metabolic advantage’ over carb-based diets; that is the claim is made that the very low-carbohydrate diet will generate more weight or fat loss at the same (or sometimes even a higher calorie level).

Now, before I continue let me say that I have nothing against low-carbohydrate diets.  My first book The Ketogenic Diet was about nothing but and many of my dietary approaches often have low- or at least lowered carbohydrate phases to them as they tend to generate certain biological effects that I’m seeking.

As well, research clearly shows that, for some people, lowering carbohydrates can have profound health benefits and in some cases a near removal of dietary carbohydrates (except for things like fruits and vegetables) may be profoundly beneficial.

However, the weight and fat loss claims are a bit trickier.  There is certainly an element of truth to the idea that low-carb diets generate more total weight loss but this issue is confounded by the issue of water loss.  Between a drop in insulin (insulin causes the kidney to resorb water) and a dehydrating effect of ketones themselves, very-low carbohydrate diets can cause significant water loss.

As I discussed at length in The Ketogenic Diet, water loss can range from 1-15 pounds depending on size (even small individual may lose a rapid 3-4 pounds, of primarily water, in the first days of carbohydrate restriction).  This tends to make comparisons of weight loss pretty meaningless.  This is even more true when you consider that the difference in total weight loss between low- and high-carb diets is usually only a few pounds anyhow.

However, other recent studies have used more accurate methods of measuring body composition and several do seem to find a greater fat loss for the low-carbohydrate diet compared to the higher carbohydrate diet.   Often with the low-carbohydrate diets reporting that they are eating the same (or occasionally) more calories.  Aha, a metabolic advantage.


I have two primary issues with most of the studies that have been done, one of which I referred to above.  That is the issue of caloric self-reporting.  The grand majority of studies done to date have allowed people to self-report their food intake and this introduces a staggering number of issues.  Because, simply, people really suck at it.  Relying on them to tell you how much they are actually eating means not having any real idea as to what’s going on.

The other, potentially bigger issue, revolves around protein intake.  In general, and this is especially true if you are comparing a typical very-low carbohydrate diet to a high-carbohydrate diet, the lower carbohydrate diet will contain significantly more protein.

This is simply a function of what foods can be eaten on the diet.  Logically, when most of your diet revolves around meat, you will tend to eat more protein than when it doesn’t.  But this tends to make comparing the diets problematic for reasons I discussed in the article Is a Calorie a Calorie.

Now, at this point there is little to no debate that higher protein diets have a number of inherent benefits to lower protein diets.  Benefits to dieters include increased satiety, better maintenance of metabolic rate while dieting, better blood glucose maintenance, less lean body mass loss and others I’m sure I’m forgetting.

But now we’re not talking about ‘low-carbohydrate diets’ per se, we’re talking about ‘high-protein diets’; that is the comparison is no longer about the carbohydrate content of the diet but about the protein content.  In fact, some researchers argue that the ‘benefit’ of supposed low-carbohydrate diets comes from the increased protein intake, rather than the carbohydrate content per se.


In any case, that’s a brief look at the two main issues I have with a lot of the low-carbohydrate vs. other diet research that is out there.  Between self-reported food intake (which tends to be all over the map) along with variable protein intake (with the low-carbohydrate diet group usually eating more protein), it’s hard to draw any solid conclusion about what’s actually going on.


The Study

Enter the study I want to look at today; while it’s a couple of years old it does a good job of addressing both of the issues I brought up above (in fact, their introduction and mine cover basically the same exact issues).

Researchers recruited 20 overweight subjects (which they admitted was a small sample size), both men and women and placed them on either a ketogenic low-carbohydrate diet (essentially an Atkins type of diet) or a moderate-carbohydrate non-ketogenic diet (like the Zone and it’s worth noting that Barry Sears is one of the authors on the paper) for 6 weeks.  Exercise was not performed.

The researchers examined a number of different variables including fat loss, insulin sensitivity, blood lipid levels, inflammation and energy levels.  Resting metabolic rate was also measured.

One nice thing about this study is that the researchers went out of their way to make sure that both diets were equal in both calories and protein (almost anyhow, the very low carb diet was a touch higher in protein).  The composition of both diets appears in the table below:

  Ketogenic Diet Non-Ketogenic Diet
Calories 1500 1500
Protein (Grams) 125 117
Carbohydrate (Grams) 33 157
Fat (Grams) 100 50

The ketogenic diet was also slightly higher in saturated fat than the non-ketogenic diet (21% vs. 9% of the total calories).

Now, you might be wondering how the researchers ensured that the diets were actually met; another strength of the study was that all food was provided to the study subjects.  Lunches were actually prepared and served to the subjects at the testing facility Monday through Friday; all other meals were prepared for the subjects to take home with them for consumption.  This is likely the reason that the sample size was so small, it gets expensive to provide all food to subjects in studies like this.


The Results

In terms of weight and fat loss, at the end of 6 weeks both groups had lost roughly the same amount of weight (6.3kg for the ketogenic diet, and 7.2 kg for the non-ketogenic diet; this was not statistically significant).  As well, the loss of body fat was the same (3.4 kg in the ketogenic diet and 5.5 kg in the non-ketogenic diet; again this was not statistically different even if the non-ketogenic diet seems to have lost ~4 pounds more fat).  There was no significant change in fat free mass for either diet.

It’s worth noting  that body composition measurements were made with a Tanita BIA device; as I discussed in Measuring Body Composition Part 2, BIA has its share of problems and can be thrown off by hydration status, which is affected by the carbohydrate content of the diet.  It would have been nice had the study used a more accurate method; one not so impacted by hydration state.

Hunger ratings improved for both diets with no difference between diets.  An oft-heard claim is that ketogenic diets cause hunger blunting due to the presence of ketones or what have you; but this study does not support that.  Given that protein is the most filling nutrient, the effect seems to be mediated by the increased protein content, not decreasing carbohydrates per se.

Somewhat oddly, the researchers found that weight adjusted resting energy expenditure went up but as it did so equally in both groups,  this was more likely related to the protein content of the diets (remember, identical) than the carbohydrate intake per se.

Looking at blood lipids, the results were actually quite mixed.  Low-density lipoprotein (LDL) went up in 5 of the ketogenic dieters and down in the other 4; LDL went up in 2 of the non-ketogenic dieters and down in the remaining 8.  HDL levels fell in both groups.

The researchers also found that one marker of inflammation was significantly higher in the ketogenic diet compared to the non-ketogenic diet; no explanation for this was given.  Some studies link saturated fat to inflammation and the slightly higher saturated fat intake in the ketogenic diet might have been part of this.

Insulin sensitivity was improved in both groups with no differences between the diets.

In terms of energy levels, there were no differences except that the non-ketogenic dieters scored higher on a rating of vigor-activity than the ketogenic dieters.  I’d note that, in my experience, the response to full blown ketogenic diets is highly variable: some people feel awful and others feel completely energized.

This tends to show an interaction with insulin sensitivity as I discuss in Insulin Sensitivity and Fat Loss with individuals with poorer insulin sensitivity often feeling fantastic on low-carb diets and vice versa. As well, ensuring adequate mineral intake (sodium, potassium, magnesium, calcium) tends to be key to limiting fatigue on very-low carbohydrate diets.

Summing up, the researchers conclude:

In the current study, the KLC [ketogenic low-carb] diet did not offer any significant metabolic advantage over the NLC [non-ketogenic low-carb] diet.  Both diets were effective at reducing total body mass and insulin resistance, but, because blood ketones were directly related to LDL-cholesterol concentrations and because inflammatory risk was elevated with adherence to the KLC diet, severe restrictions in dietary carbohydrate are not warranted.



My Comments

This study certainly has a couple of strengths, controlled calorie levels (with food provision) and equal protein intake between groups being the two main ones.  Weaknesses are the small sample size and the use of Tanita body fat measurement to track body composition.  Changes in water balance can affect that method significantly and diets that manipulate carbohydrate content tend to impact significantly on water levels in the body.

With that said, this study adds to the data set suggesting that, if calories and protein intake are identical, there is little to no metabolic advantage (in terms of fat or weight loss) to full blown ketogenic diets.  They work at least as well, mind you, but not better.

Now, as I discuss in the series of articles on Comparing the Diets, there still may be circumstances where very-low carbohydrate/ketogenic diets may still be preferred, even given the researchers comments above.

For many individuals, the biggest ‘metabolic advantage’ of very-low carbohdyrate diets may be one of food/calorie control.  Many people seem to show what might be popularly called ‘carbohydrate addiction’ (a term that is massively debated among obesity researchers) where eating even small amounts of carbs makes them want to eat more.  In that case, a full blown removal of carbohydrates from the diet may be the only realistic way to limit caloric intake.

Related to this, for many, spending time on a very-low carbohydrate diet seems to change taste preferences: carbohydrates can often be reintroduced after some period without the loss of food control that occurred prior to the diet.

I mentioned in the introduction that severely insulin resistant individual seem to get health benefits from extreme carbohydrate restriction.  For leaner dieters, as I discuss in both The Ultimate Diet 2.0 and The Stubborn Fat Solution, extreme carbohydrate restriction tends to help with stubborn lower body fat mobilization.

But clearly in other situations, moderate carbohydrates diets (still containing sufficient protein mind you) may be superior.  For people who can control their food intake, many simply feel better with more moderate carbohydrate intakes, their training (especially high intensity training such as weights) doesn’t suffer as much, and they don’t feel quite as lethargic.  Again, these issues are discussed in detail in the Comparing the Diets series.

I guess my point is that there is too much variance between individuals and their needs to claim that any single diet is inherently superior for all people and all situations.  As this study suggests, given identical calories and protein intake, there doesn’t seem to be any inherent metabolic advantage in terms of total fat loss to a full blown ketogenic diet, at least not when compared to a moderate carbohydrate diet with an identical amount of protein.

As I’ve noted above, there may be other ‘advantages’ unrelated to fat loss that are still important.  Basically, the choice of diet may come down to other issues than fat loss per se.  A diet that makes someone feel terrible isn’t one that they are likely to stick to for very long; a diet that does is one that may be the ‘superior’ one for that person.

Finally, finally, if you want to hear more about my thoughts on this whole issue, Jimmy Moore of the Livin La Vida Lowcarb blog has just put up an audio interview we did a couple of months ago.  We talk about everything related to low-carb diets you could imagine.



25 thoughts on “Ketogenic Low-Carbohydrate Diets have no Metabolic Advantage over Nonketogenic Low-Carbohydrate Diets – Research Review

  1. You certainly hit the nail on the head when you say the same diet doesn’t work for everyone. People who preach their diet is THE ONLY WAY and condemn all other points of view should be immediately ignored.

    I have noticed that people with naturally high metabolisms have a much more difficult time sticking to low carb diets, while people like myself, whose metabolism runs more along the lines of sloth-like, do pretty well with it.

    I would also like to point out that the term “diet” has been taken out of context in the modern era. Diet should not be the couple of weeks you eat only cabbage soup to lose a few pounds. Diet should be a permanent part of your lifestyle, how you eat every day, for the rest of your life. You just have to find the way that works best for you!

  2. lets dig a little deeper shall we? what type of protein were they consuming. . .if regular meat from the store, then we can safely say that whatever group they were in would increase inflammation markers. . .grain fed animals pass on the omega 6’s to us and voila, increased inflamation. . .I like the study, gives food for thought. .. 🙂
    as of this summer I will not be buying any meat from the store. . .just a grass fed farm in central ontario. . .
    go low carb. . . .whether moderate or keto, you will not go wrong. . .30% of the population does not do well on really low carbs. . but that is what a good nutritionist will find out for you.. .:)

  3. Lyle,

    Do ketogenic low-carb diets have potential advantages over nonketogenic low-carb diets in terms of fat mobilization?


  4. Lyle,

    This question may come off as silly, so I apologize in advance if it turns out that way. Would there be any potential advantages or benefits to locking in protein and total calorie intake and flip flopping fat and carb intake from time to time for periods of 7-14 days?

    e.g. If you typically consume a moderate-to-high protein diet of roughly X kcals with a low or very low carb intake and a high(er) fat intake, then switching to a diet where the carb intake is higher for 1-2 weeks and the fat intake is lower, mostly from EFA’s and the fat in the protein foods you eat (total kcals that had been from carbs become kcal target for fat intake and vice versa)

    Or, alternatively switching to a higher fat intake and lower carb intake if carbs had been on the higher side and fat intake much lower?

  5. Following up on Hal’s question,

    This is one of my interesting interpretations of carbohydrate manipulation. You could take 2 twins, have them both recomp for a year. Over the year they both eat the same macros and have the same activity (2 resistance training workouts a week).

    Twin A eats a reduced calorie diet: say high-protein, moderate carbs, and low-fat. On a weekly basis, he eats 100-200 g carbs per day, but averages 157.
    Twin B eats a high-protein low-carb diet with 3 weekly refeeds. On a weekly basis, say he eats 20g carbs per 5 days, and 500g carbs on his 2-workout days.

    Imagine their daily consumption of protein and fat is equal.

    Based on my interpretation of what I’ve read we would expect twin B to have better body recomposition. Would the difference be significant? Would it be noticable in the mirror? Or would the majority of Twin B’s benefits from hormone up-regulation be in diet adherence?


  6. Matt,

    There’s a matter of personal variance. Starting bodyfat % of the twins, training volume (500g post workout would require a fair amount of volume), insulin sensitivity (note that it can vary between siblings).

    My take on your hypothetical situation is that Twin A gains more LBM than Twin B. Twin B loses more fat than Twin A. Whether these variances are significant, I’m not sure.

    My rational is that Twin A is consuming sufficient carbs to keep insulin levels stable/elevated throughout the week (good for LBM, less good for fat loss). And, as long as training volume is not super high, there will likely be enough carbs post workout to complement the training effect.

    Twin B is in a much bigger calorie deficit on non-training days, and is probably not reaping as much benefit from round-the-clock higher insulin levels (for LBM gain – but the situation is better for fat loss). The twice weekly refeeds surely complement the training effect acutely, but protein synthesis continues to the next day (peaking at around 24 hours), at which point the twin is back in a suboptimal feeding regime for LBM gains.

    Maybe increasing training frequency (split routines, more full body routines, whatever) for Twin B would result in more LBM gain because he’ll be refeeding multiple times per week. Even if carbs during the refeed are lower (because they are spread across more training sessions), as long as the training volume is not really high, he’ll still be eating enough to complement the training effect. Would this slow down fat loss? Possibly.

    The above is all hypothetical and maybe nonsensical, but, meh.

  7. It is a known fact the increase in LDL on a high-fat diet are of the “safe” variety, e.g. the non-dense LDLs. I see no measurements of the Apo-values for a correct indication whether the increased LDL levels were indeed hazardous to the subjects’ health.

    Moreover, “emotional effects”? Pardon?

  8. That was interesting stuff to read, but I’ll stick to Low Carb Diets. There are my reasons:

    1) With low carb diets you maintain more than 20% of your calories from fats (it’s harder to hit this mark on higher carb diets as when carbs go up, fats must go down) and not suppress my testosterone levels
    2) Low carbs prevent insulin rises (it’s easier to control hunger without them)
    3) It’s more satiating to eat fats over carbs
    4) I just like LCD’s more 😉

  9. Everything having been said, the one thing the I personally experience is that I actually tend to eat LESS when on a low carb diet. The concentration of fat / protein in my diet seems to kill my hunger. I feel better when i restrict my carb intake – especially all the junk carbs. For me, elevated carb intake makes me feel dull and energyless (and no, I am not diabetic or pre-diabetic). High carb intakes also leave me feeling bloated. I believe it comes down to choosing the diet that best “fits” your lifestyle and genetics

  10. This is weird. You seem to be saying the Ketogenic diet is over-rated, yet you wrote a book about it.
    Should I read the book? (I was intendignto buy it. along with UD2)
    Or am I wasting my money?
    Don’t you use fat for fuel instead of carbs when on a ketogenic diet?
    Isn’t that the superiority of it to other diets?
    I am now really confused.

  11. Zane, I’m really not getting that read off this take on this was that while the study has merits (strengths) it also has some weaknesses and in the long run there is no way to say that one diet is better than another FOR EVERYONE.

    Are you confusing this with him saying that low carb is overrated? Because that’s not it all – it’s that the extreme low carb (ketogenic levels) just might not be as advantageous as was once believed.

    My take on it is that while low carb is a good plan/way of life, you don’t NECESSARILY need to push it to the ketogenic level (which is an extremely low carbohydrate/high protein plan) because payback isn’t automatically greater in terms of metabolic functioning than your more traditional low carb plan (although I do believe for some ketogenic is more advantageous).

    I know that low carb (non-ketogenic – I don’t have the discipline for that lifestyle) works the best for me on many levels – my hunger signals, my energy leves, etc. But based on this research and well-thought out analysis by Lyle, any thoughts I had of pushing my diet to the ketogenic level have been put to rest. I’ll stick to low carbohydrate/higher protein.

  12. Always always always go to the actual study rather than the secondary source. And read the paper, not just the abstract. The study is . It does not disclose what the participants actually ate, but it does disclose that the ketogenic diet they used failed to meet the RDA of fiber, vitamin E, folate, iron, magnesium, or potassium, while the “nonketogenic low-carbohydrate” diet did meet all the micronutrient RDAs. They even stick on a little footnote, “Values <=67% of daily recommendations", so you know they knew about the problem, and failed to mention it in the abstract.

    [WORDPRESS HASHCASH] The poster sent us '1762401312 which is not a hashcash value.

  13. Tom – gluten free foods are not necessarily low calorie. Some breads and snacks have more calories and more sugar than wheat products. But some people do report that after going gluten free for two weeks or more, they feel less bloated and have more energy. Gluten free bread is usually brown rice based, so the calories and sugar are still there so be careful.

    Gluten causes inflammation, that is true, and belly fat, but eating gluten free does not guarantee weight loss unless you eat more veggies and protein in wheat (carbs)’s place. If you simply substitute

    If you read Wheat Belly, you will see that wheat can cause all types of problems including belly fat and heart disease.

    My two daughters are Celiac and don’t eat gluten (wheat, barley, malt, rye). But you can find plenty of processed foods that are marked gluten free, including muffins that have WAY more sugar and calories. But I do think the gf foods are healthier, especially pizza.

    The theory I always follow is that gluten gives you a fat belly period. No question. Look at beer drinkers. And then look at runners; you still see marathon runners with bellies and belly fat. How can that be if they train for marathons? How can they possibly not burn that off….. It is obviously not a question of calories burned > calories consumed. It really matters WHAT they eat…. remove gluten beer and bread from their diet, and those runners will have flat bellies.

  14. The missing factor is cardio and weight training. The keto group will have a dramatic difference in fat loss once that is factored in. The study is antiquated. Read The Art and Science of Low Carb Performance for a fresh scientific view.

  15. Oh bullshit. Because adding cardio and weight training to the carb-based diet would have the same effect. And how about reading my book The Ketogenic Diet? Cuz I did it first. I’ve read every study on the topic and simply, you’re wrong.

  16. The mere mention of the Book Wheat Belly dismisses every comment you made. People report feeling better when they drop all kinds of stuff that they have been told are bad for them and which have no effect in blinded studies. That your two daughters are celiac does not change the fact that it’s prevalent in less than like 1% of the population. Their individual results do not apply to the world at large.

  17. BTW did these faggots even study the people after they were keto-adapted or was this another “Subject did not eat carbs for 3 days, subject has headache, KETO R BAD” article?

  18. Can you read

    “Researchers recruited 20 overweight subjects (which they admitted was a small sample size), both men and women and placed them on either a ketogenic low-carbohydrate diet (essentially an Atkins type of diet) or a moderate-carbohydrate non-ketogenic diet (like the Zone and it’s worth noting that Barry Sears is one of the authors on the paper) for 6 weeks. Exercise was not performed.”

    6 weeks long.

  19. “60% of energy as fat” seems low to get one into ketosis and the length of the trial only being 6 weeks long… come now!!!. So it begs the question, “were the subjects actually tested to see if they were, indeed, IN ketosis? I can’t see the majority of the keto subjects having been actually IN ketosis for WEEKS into the trial if they started at 60% and were sedentary (not burning off their glucose/glycogen reserves.

  20. Hey Lyle,

    I know you’re a busy guy but I thought I might try here…

    I’m in a hot debate about how carbs can cause more hunger or not. I know it’s not definitive, but the people I’m talking to seem to think I’m talking out of my a$$. Would you have any scientific articles or sources that I may bring back to them?

    love your work BTW

  21. 6 weeks is plenty long to adapt to a ketogenic state and ketosis will develop in about 3 days so what ARE you on about other than attempting to hand-wave a result you don’t like away?

Comments are closed.