Paul from Facebook asks:
What are your thoughts on fructose? Is it really as bad as Paleo is making it out to be?
Dr. Robert Lustig has worked hard in recent years to demonize fructose, and his efforts have paid off. His YouTube video “Sugar: The Bitter Truth” has over 2.5 million views as of this writing. Lustig et al. claim that fructose is a uniquely fattening poison (when compared to glucose) that is as toxic to the liver as alcohol.
But is this true? Does the current evidence support this position? I’ve changed my views on this over time as I’ve become better acquainted with the literature, so I’d like to share my current understanding with you.
When it comes to fructose, calories matter
There’s no doubt that refined sugar – including fructose – can be problematic. But studies suggest that this is only true when calories are in excess.
This may be the most dangerous aspect of refined sugar: it leads to unintentional overeating. In a recent post on fructose, obesity researcher Stephan Guyenet points out that most people in these studies aren’t deliberately overfeeding. They are inadvertently overfeeding because they aren’t spontaneously compensating for the calories added to the diet via a large fructose- or glucose-sweetened beverage.
This doesn’t happen with fruit or other whole foods that contain glucose or fructose.
When people add a soda or two a day to their diet, they tend not to reduce consumption of other foods, and thus their calorie intake increases.
This is where fructose does appear to be more harmful than glucose. Although people don’t compensate for calories added via glucose or fructose, the fructose-sweetened beverages have more serious metabolic effects.
Is fructose uniquely fattening?
Dr. Lustig argues that, when compared to glucose, fructose is uniquely fattening. He claims that fructose is the most efficient substrate for de novo lipogenesis (DNL), which is the process by which the liver converts carbohydrates to fat.
However, Dr. Lustig relies on animal evidence that doesn’t apply to humans. There’s a big difference between mouse carbohydrate metabolism and human carbohydrate metabolism. When mice are on a high-carbohydrate diet that doesn’t provide excess calories, it’s common to see DNL rates of 50 percent and up. In other words, they are efficient at converting carbohydrates into fat, even when they’re not overeating. (1)
But in humans on an isocaloric diet (without excess calories), de novo lipogenesis falls into the range of 10 to 20 percent. The conversion of carbohydrate is less efficient in humans than it is in mice.
The research in this area is robust and uncontroversial. Nearly 50 controlled feeding studies have been performed on various aspects of cardiometabolic control. Most investigators working in this field believe that DNL in humans is negligible in response to fructose, and doesn’t comprise a significant source of dietary calories.
There’s another problem with extrapolating the animal evidence to humans in this case. The mice in the studies Lustig cites are eating huge amounts of fructose: up to 60 percent of total calories. You’d have to drink more than four 44 ounce Super Big Gulps a day to get that much fructose. Ain’t gonna happen.
According to researcher Dr. Sievenpiper in an interview with science writer David Despain at Evolving Health, the 50th percentile intake for people in the U.S. is 49 grams per day, which works out to 10 percent of total calories. Even the 95th percentile intake of 87 grams per day doesn’t exceed 20 percent of calories. That’s a lot of fructose, but it’s nowhere near the 60 percent of calories fed to mice.
Like what you’re reading? Get my free newsletter, recipes, eBooks, product recommendations, and more!
Is fructose an evil toxin?
Dr. Lustig refers to fructose is a “poison” that is nearly as toxic to the liver as alcohol. But again, human evidence doesn’t support this claim.
In a recent paper, Dr. Luc Tappy and colleagues labeled acetate, fructose and different metabolites with stable isotope tracers so they could see how fructose is metabolized in the human body. (2) They found that 50 percent ends up as glucose, 25 percent goes to lactate and greater than 15 percent goes to glycogen. The remainder is oxidized directly (to CO2 through the TCA cycle) and a small portion – as low as 2-3% – is converted to fat via de novo lipogenesis.
Glucose and glycogen are easily processed by the body, and 2-3% conversion to fat is not significant. And while some have claimed that lactate may be problematic, a paper published more than a decade ago contradicts this. (Hat tip to Evelyn from CarbSane.) According to the authors:
The bulk of the evidence suggests that lactate is an important intermediary in numerous metabolic processes, a particularly mobile fuel for aerobic metabolism, and perhaps a mediator of redox state among various compartments both within and between cells… Lactate can no longer be considered the usual suspect for metabolic ‘crimes’, but is instead a central player in cellular, regional and whole body metabolism.
Translation: lactate from fructose isn’t a problem.
What does this mean for you and fructose?
Fructose-sweetened beverages like soft drinks and juice cause metabolic problems when calories are in excess, and studies have shown that people are not likely to compensate for the additional calories they get from such beverages.
I don’t think there’s any basis for avoiding whole fruit simply because it contains fructose. As I’ve shown in this article, there’s nothing uniquely fattening or toxic about fructose when it isn’t consumed in excess. And since whole fruit contains fiber and other nutrients, it’s difficult to eat a lot of fruit without simultaneously reducing intake of other foods.
Fruit has been part of the human diet for longer than we’ve been, er, human. We’re well-adapted to eating it, and capable of processing the fructose it contains. (Unless you are FODMAP intolerant – but that’s a different issue entirely.)
Better supplementation. Fewer supplements.
Close the nutrient gap to feel and perform your best.
A daily stack of supplements designed to meet your most critical needs.
The fructose-glucose combination is not always bad and has been shown in tests done on humans (albeit small and brief) to actually provide benefits. Wild fruits tend to contain both fructose and glucose and as I mentioned above, wild fruits tend to have more fructose and less sucrose than domesticated fruits (though there are even some wild fruits high in sucrose).
Research suggests that fructose helps human subjects absorb fructose and modulate the glycemic effect of glucose without stimulating insulin secretion (in contrast to the findings of the Wei Wong mouse study–and as Chris pointed out, “The conversion of carbohydrate is less efficient in humans than it is in mice” and “The mice in the studies Lustig cites are eating huge amounts of fructose”):
Moore, Cherrington, Mann, et al. Acute & fructose administration decreases the glycemic response to an oral glucose tolerance test in normal adults. J Clm Endocrinol Metab 2000;85(12):4515-19. http://jcem.endojournals.org/content/85/12/4515.full
“low dose fructose improves the glycemic response to an oral glucose load in normal adults without significantly enhancing the insulin or triglyceride response. Fructose appears most effective in those normal individuals who have the poorest glucose tolerance.”
Acute Fructose Administration Improves Oral Glucose Tolerance in Adults With Type 2 Diabetes
Moore, Davis, et al. Diabetes Care November 2001 vol. 24 no. 11 1882-1887 http://care.diabetesjournals.org/content/24/11/1882.full
“Low-dose fructose improves the glycemic response to an oral glucose load in adults with type 2 diabetes, and this effect is not a result of stimulation of insulin secretion.”
Is Fructose Absorption in Humans Improved by the Addition of Glucose?
Logan L. Davis-Wallace
http://www.usc.edu/CSSF/History/2009/Projects/S1810.pdf
“The addition of glucose could be used to improve the absorption of fructose in humans in high fructose containing foods and reduce symptoms of malabsorption (abdominal pain, gas, diarrhea, and bloating) from these foods.”
Fructose Might Contribute to the Hypoglycemic Effect of Honey
University of Science, Malaysia
http://www.mdpi.com/1420-3049/17/2/1900/pdf
“In view of the similarities of these effects of fructose with those of honey, the evidence may support the role of fructose in honey in mediating the hypoglycemic effect of honey.”
Look at this; http://stke.sciencemag.org/cgi/content/abstract/sigtrans;5/213/ec64
G. A. Kyriazis, M. M. Soundarapandian, B. Tyrberg, Sweet taste receptor signaling in beta cells mediates fructose-induced potentiation of glucose-stimulated insulin secretion. Proc. Natl. Acad. Sci. U.S.A. 109, E524–E532 (2012).
Researching the effect of fructose, without looking at the fructose-glucose combination is missing half the point; the sweetness of fructose increases the insulin response to glucose.
Whether fructose is fattening might depend on one’s taste receptor phenotype.
This nicely integrates the food reward and carbohydrate-insulin hypothesis aspects of fructose.
@ adrian, exactly, which is why claims that fructose “has no effect” must be carefully weighed. This is clearly a personal interpretation of data that others might interpret differently.
@Paleophil, I agree, but Chris is here providing a link, through Stephan Guyenet, to some authorities who do seem to be minimizing sugar harms. I am not criticizing Chris’s position, but the conveyed interpretation these authorities have put on the research. Perhaps the usage “food toxin” or “dietary toxin” is better than “toxin”. Alcohol is a food which is also a toxin at increasing intakes.
It is worth noting that the increase in degenerative diseases that follows adoption of western diet – cancers, tooth decay, heart disease, arthritis – do not seem to depend on hypercaloric intake or sedentary lifestyle. Nor does fatty liver (which can be produced by toxins, food toxins, or pathogens in normal weight individuals). Obesity may be the result of the addition of excess calories to the conditions that cause the other diseases. If we see it as part of this continuum, it perhaps helps to illuminate the role of food toxins.
It is entirely possible that HCFS is more problematic than sugar, even if glucose and fructose levels are equalized. It is a grain product and is never entirely clear of immunogenic grain elements.
George Henderson wrote: “One can defend the intake of sugars from whole fruits and honey, in a hunter-gatherer “intermittent fasting” milieu, as healthy without extending that definition to include sugar, HCFS, apple juice concentrate and agave nectar.”
I haven’t seen Stephan or Chris extend the definition of “healthy” to “sugar, HCFS, apple juice concentrate and agave nectar.” Quite the contrary, my recollection is that they have warned against eating those sorts of foods and extolled the virtues of natural whole foods. Chris even referred to “sugar (especially high-fructose corn syrup)” as a “dietary toxin” here: http://chriskresser.com/9-steps-to-perfect-health-1-dont-eat-toxins. Perhaps he might word that a bit differently today, but it shows that his history on this topic is not one of promoting added sugars.
Going back to the Guyenet discussion, I don’t think that the researchers he quotes actually went through the Lustig references; it looks like they did their own search. And I question this statement: “in each case there was no effect of fructose when it was isocalorically exchanged.”
Many studies have been done where sugar is isocalorically exchanged with various foods, for example the comparisons with honey (a very similar food) linked to Mark Sisson’s blog recently: http://www.marksdailyapple.com/is-honey-a-safer-sweetener/
for example, http://www.ncbi.nlm.nih.gov/pubmed/15117561
(which also distinguishes between dextrose and sucrose, which do NOT seem to have identical effects).
If there are differences between honey and sucrose, isocalorically exchanged, then there are going to be differences between say, sucrose and starch. Whether these amount to much is another story, but I think the claim that there is no effect needs to treated with scepticism.
Also I question whether the subjects in (say) the honey experiment above were carefully fed beforehand to optimise calories, so that the extra sugars were hypercaloric. This doesn’t appear in the abstract, and strikes me as an unusual proceeding.
I am not saying that the honey research proves anything re: Lustig, except that the “isocaloric” claim may be misleading. Guyenet does refer to Sisson’s honey link in his summing up. One can defend the intake of sugars from whole fruits and honey, in a hunter-gatherer “intermittent fasting” milieu, as healthy without extending that definition to include sugar, HCFS, apple juice concentrate and agave nectar.
“(which also distinguishes between dextrose and sucrose, which do NOT seem to have identical effects).”
They do not have the same effect because they are not the same thing. Dextrose (a simple sugar also known as glucose) is a monosaccheride while sucrose (more commonly known as table sugar) is a disaccharide of both dextrose and fructose. Honey however is a very complex substance that also contains both glucose and fructose that would have the same effect wherever they were found. The benefits of honey come from everything else it contains that is neither glucose or fructose. Probably repeatiing something already on this very long.string.
Excellent share. I did not know that fructose can be so damaging and being one of the main culprit that causes fat gain.
You can look at healthy volunteers, and isocaloric states, for relatively short time periods, and conclude that fructose (or the fructose-glucose combo) is not problematic. And you can feed unhealthy volunteers the same thing, and see problems. The experiment that takes healthy people and makes them diabetic in a prolonged, controlled experiment, over decades to match the natural progression, is probably impossibly unethical (and the controls might well be on the same path).
There are perhaps two ways in which fructose might drive this 1) through the hedonic pathways Lustig describes (sugar, like alcohol, is a subversively moreish influence to many of us) leading to the “fructose & isocaloric” state.
2) through sugar lending itself to “snacking” – constant caloric intake disrupting traditional feeding schedules. It does this through the its role as a preservative, flavouring, and appetite stimulant. To understand how this might contribute to diabesity, try the thought-experiment of reverse engineering the various experiments in intermittent fasting, or time-restricted feeding: http://www.salk.edu/news/pressrelease_details.php?press_id=560
If these two behavioural effects of sugar tend to create the sustained hypercaloric environment in which sugar, and perhaps other elements of the diet, contribute to metabolic problems, might that qualify as toxicity?
It is a feature of the Lustig hypothesis, in its published form, that it involves multiple discrete effects of fructose, and the links between them are not in every case drawn with the sort of detail that will be available in future, or that can perhaps be supplied from informed speculation. There are a lot of implications that haven’t been tested. There are probably implications in there that haven’t been realised yet.
fructose only causes problems in caloric excess? But if you’re eating sugar/ HCFS in caloric deficit these might be replacing foods that supply more elements of nutrition (it’s hard to think of foods that supply less). I believe sugar causes tooth decay whether one eats too few or too many calories. Metabolic syndrome and type 2 diabetes hit some people of normal weight who consume fructose/sugar.
I suggest analysing the Lustig hypothesis in its published form, as it has aspects (such as Fox01 activation) that are left out of the video presentation.
http://podcast.uctv.tv/webdocuments/Fructose-and-Ethanol.pdf
I think it is worth asking whether Lustig has dumbed down the evidence for public consumption, and whether it behoves us to make a better job of interpreting it before we dismiss it altogether.
If there are holes in it (and Peter at hyperlipid has pointed out that Glut5 is far from exclusively hepatic in location), I think it’s worth trying to patch them and see if the thing still floats.
If you think PUFA is also to blame, for example, what happens when you add PUFA to the Lustig hypothesis?
It definitely seems to belong in this one: http://www.nature.com/nrgastro/journal/v7/n5/authors/nrgastro.2010.41.html
Before dismissing this idea out-of-hand, why not try to, in the immortal words of Tim Gunn, “make it work”?
Sugar is certainly not optimal as a major source of calories, whether overall calorie intake is in excess or not. I don’t think anyone would make that argument. The point was that the evidence doesn’t support the notion of sugar being “toxic” or contributing to significant metabolic problems in isocaloric diets.
Here’s an interesting question; if someone has elevated blood glucose due to overactive gluconeogenesis or insulin resistance, does this mean that any fructose they eat is processed as if they are in a hypercaloric state?
My problem with fruit is that I frequently binge on it, despite the claim that this should be unlikely. If I buy blueberries in season the pint frequently disappears… during the drive home. Good apples or navel oranges…I can easily eat half a dozen during a single day. I don’t do this with sweet potatoes. Getting my carbs right is very important, but fruit for me is generally a bad choice because of the binge-factor.
If fructose is not the cause, why has there been such an increase in the incidence of non-alcoholic fatty liver disease? Is it just excess calories in general?
Fructose is the cause if you live a hyper-caloric diet. An unintended result of Lustig’s, “Sugar The Bitter Truth” is the Fructose-Phobia in the Paleo community. That’s the point. This is a hard article to navigate the first time through. Just because most of us here are recovering metabolic-aholics. Chris actually did a nice job explaining that if your metabolically healthy — enjoy your fruit. If you have metabolic issues — stay away. And he never, ever advocated the consumption of sweetened beverages (besides a small glass of juice). The best article summing up sugar (especially fructose) is Gary Taubes’ New York Times, “Is Sugar Toxic” which is heavy on the Dr. Robert Lustig references. That is fantastic, but you may not need to go there if you re-read Chris’ article and get the context. Hyper-caloric diets rich in sugar laden foods indeed causes Non-alcoholic fatty liver disease. If your Paleo template is in order, your should be okay. I feel if you eat alot of fructose you are actually twice as doomed. First as Chris points out your metabolism does account for fructose calories, and second fructose sweetness is addictive in high quantities. So you’ll want more food, and the more food you’ll want will be sweet carbs.
Thank you for your thoughtful analysis, as always, Chris!
I have a question tangentially related to this, which has been bugging me for a while, and I hope you might take it up sometime (or refer me if it’s covered somewhere): How exactly are hypo/iso/hyper-caloric diets defined? At first, it seems like a trifle: just compare with basal metabolic rate + activity level.
But on the other hand, the goal of wellness care is to optimize well-being, and this includes a feeling of “having lots of energy”. We all know that some diets cause us to be lethargic and ill-feeling, whereas other diets do the opposite, even with the same amount of calories. (Of course the environment generally play a role as well: stress/emotions/sunshine, etc.)
It seems intuitive that a “hyper-caloric diet” that makes you feel really good and energetic is much healthier than a lower-calorie diet that makes you feel lethargic and ill, and possibly makes you fatter as well. How do diet researchers deal with this issue?
Most cancer is fed by glucose and uses copious amounts of it. Has there been any research as to the metabolism of fructose by neoplasms?
Great article Chris,
I think people take a lot of information out of context when it comes to nutrition.
My experiences say moderate, thin skinned dark fruit consumption has never harmed anyone with healthy body composition. However, overweight and obese clients have almost universally done better abstaining from all fruit for the first 2-4 weeks.
That said, I’ve found it to be the higher mass and carb fruits (bananas, pinapples, etc.) that most people need to watch. Especially if adding them to an already carbohydrate dominant diet.
Excellent information as always, many thanks,
Jeremy
There is something uniquely bad about fructose. Look up the fructose fed rat model, a standard tool for studying insulin resistance and hypertension. It only takes two weeks on a high fructose diet to get rats to develop metabolic syndrome. As the months go by, evidence linking the metabolism of fructose in the liver
with the development of fat accumulation, insulin resistance, hypertension, and altered immune function accumulate. A quick search of PubMed is all that is required to see how the pieces are starting to fit together. Of course it doesn’t matter how you get the fructose, whether it is from excess sucrose or from high fructose corn syrup.
Did you read this article? The whole point is that you can’t extrapolate rat study results to humans, because rats metabolize carbohydrates (including fructose) differently than humans. I also pointed out that the amounts of fructose given to the rats is much, much higher than what humans eat.
I find it interesting that fructose is labelled ‘as bad as alcohol’ yet we are often told that drinking a glass of red wine each day can be beneficial.
I think after everything is considered, balance and moderation is the only approach to take. Too much of anything is never a good idea.
The polyphenols in both fruits and wine are beneficial in ways that we don’t fully realize yet.
Adrian I’ve read this before…
Yet I used to eat 1.5 lb or whole wheat sprouted bread per DAY!
I never gained much weight although I did not and still don’t exercise.. (not good I know).
I am now 61, I eat about 1/3 lb of same bread, I am about 5 lbs heavier than I was at 25:
I am now 5’8″, 142# and consider myself thin although the bread gave me a bit of a beer belly.
I am quite versed in Nutrition and can’t explain how I ever ate 1.5Lb of that bread daily and never gained a pound.
Also, I am skipping midday meal these days (not good but not hungry) 3-toasts loaded w butter keep me going all day! Only sprouted whole grain brad does that — the white stuff and I am hungry within 4 hours at most.
WE ARE DIFFERENT!
I never drink sodas and cut my water with 1/4 to 1/3 juice + stevia sometimes
A similar discussion on this 25 year old Swedish research is taking place on the shape.com website. All the comments are similarly critical of the research. The most succinct comment being this one from Brigitte :-
“The body’s preferred fuel is fat, whether it’s from the diet or used from our stored fat.. the body uses up sugars first only because it must, as sugar is dangerous. You cannot use your stored fat for fuel when insulin is present……if you eat carbs all day (all carbs=sugar even whole grains,) you keep insulin present all day and your body cannot access your fat for fuel so it has no choice but to tell your body to eat to get the energy it needs. That is why you are ALWAYS hungry when you eat carbs. When you switch to low carb, this process corrects itself and you go many hours being satisfied, all the while burning those love handles for fuel instead of eating…seems like magic but it’s simply biology. Low carb paleo changed my life, it can change yours too.”
And your point is …?
If the body’s preferred fuel is fat as you state, why then if we eat only protein and fat will our body convert preferentially fat into glucose for fuel ?.. and if fat is used up then the body will tear down the body structure for fuel ,i.e body tissue/ proteins.This does not sound very desirable. Indeed it sounds quite a stress.
The simplification that ‘all carbs = sugar’ forgets to make mention of the magnesium/potassium and other minerals & vitamins ,etc which come with ,and help us to use, the sugar
Chris explained the effect of soda pop in his post thusly: “When people add a soda or two a day to their diet, they tend not to reduce consumption of other foods, and thus their calorie intake increases.
This is where fructose does appear to be more harmful than glucose. Although people don’t compensate for calories added via glucose or fructose, the fructose-sweetened beverages have more serious metabolic effects.”
Chris didn’t advocate for drinking soda pop, but imagery tends to be more powerful than words, and thus Waldo has a point about the soda image. It’s attention-grabbing, but may be misleading some folks. Perhaps a picture of fruit would have been better?
Yes, I agree. I’m going to change the image now.
The new image is spot on.
Thanks for that Chris, I am getting so confused these days, I totally get the sugar/HFCS part, but was starting to wonder about fruit because of the fructose, it did not seem right to take that out of the diet too. sugar out, fruit stays.
Plants have been producing fruits for millions of years because it is mutually beneficial. No need to be cautious of fruits, it is other 90% of the stuff in grocery store.
Hi Kris
I read your view on Fructose.
How do you explain Dr Lustig’s statement that the only teens that fail in the clinic program (UCSF?) are those who won’t stop drinking sodas?
Thanks. Chris
Because frequent consumption of sodas will tend to cause overconsumption of calories, and in excess energy states fructose has harmful metabolic effects. I completely agree with Lustig on that score. The purpose of this article was to point out that fructose (and sugar in general) does not appear to have those harmful metabolic effects unless calories are in excess.
I don’t recommend people drink fructose-sweetened sodas, period. But fructose in whole fruit – barring fructose-intolerance, a digestive issue or Type 2 Diabetes – does not cause metabolic harm for most people.
And because sipping sodas is like snacking, it breaks the circadian rhythm which is maintained by a proper fed-state fasting-state distinction. Always eating is a form of over eating, even on isocaloric diets.
http://www.salk.edu/news/pressrelease_details.php?press_id=560
Excellent point, High Fat Hep C Diet. Nonperishable sealed soda pops, soda vending machines and such make frequently snacking on high-calorie sugary beverages between meals easy.
Yes, just becuz an infant can being sucking breast milk whenever it needs to suck…if one does give manmade bottles of water/juice and pacifiers…does not mean older children, or adults, for heaven’s sake, can be constantly ingesting calories! Our bodies don’t do well thus treated. A baby’s thrives. We can’t always cross over conclusions without using logic along the way.