Welcome to part two of Fructose Phobia!
Ready for more fun with fructose? I know I am! In today’s post, we look at fructose research, starting with the latest fructose study making headlines to see what all the fuss is about. Researchers at the University of Southern California found that fructose may make it harder to control appetite and food cravings than glucose, paving the way to overeating and obesity. Should we believe these scientists? Spoiler alert: nope.
[This is article two of a five-part series about fructose. If you’d like to start at the beginning and read about what fructose is and how the body processes it differently from glucose, please see article one: "Has Fructose Been Framed?"]
The study is titled: "Differential effects of fructose versus glucose on brain and appetitive responses to food cues and decisions for food rewards." And away we go!
Twenty-four males and females aged 16 to 25 yrs old, with BMI's ranging from 19.6 to 45.5 kg/m2, without history of eating disorders, fructose intolerance, diabetes, or other medical conditions, and not on weight-loss diets or taking medications (except birth control pills).
Volunteers fasted overnight for 12 hours, rated their appetites on a scale of 1 to 10, then drank 300 ml (10 oz) of cherry-flavored water containing either 75 grams of glucose or 75 grams of fructose.
Thirty minutes and 60 minutes after the drink, they rated their appetites again, and were shown pictures of foods they personally found appetizing. They then had to decide whether they would prefer to eat those foods immediately after the study or receive a small cash reward a month later. Blood samples were collected at 0, 30 and 60 minutes to measure sugar and appetite hormone levels. fMRI (functional Magnetic Resonance Imaging, aka "brain scan") studies were also conducted at 0, 30 and 60 minutes to observe which brain areas were stimulated by the food images.
- After drinking fructose, volunteers were hungrier and less likely to choose a cash reward over food reward than they were after drinking glucose.
- As expected, fructose had almost no impact on insulin levels, whereas glucose caused a large insulin response [Figure 1: Luo S et al.]:
- As expected, fructose caused lactic acid levels in the blood to rise more than glucose did (lactic acid is one of the by-products of fructose metabolism—see article one).
- Levels of the appetite-regulating hormones leptin, ghrelin, and GLP-1 (glucagon-like peptide) were similar after the fructose and glucose drinks.
- Although not highlighted by the study's authors as a key finding, levels of the appetite reduction hormone peptide YY (PYY) rose higher in response to fructose than glucose, and were still rising at 60 mins, whereas PYY levels after glucose were already falling by 60 mins. (This information was buried in a table in the supplementary data section and was not addressed by the authors, so I had a graph created [data from Table S1: Luo S et al.]:
- fMRI brain scans after fructose consumption showed a greater increase in activity in the visual cortex, as well as in the left orbital frontal cortex and left ventral striatum, which are brain regions thought to be involved in reward processing.
The study is sophisticated in design and seems thoughtfully conducted, but upon closer scrutiny, it’s easy to find serious flaws that render the study utterly meaningless.
- Dosage too high. The amount of fructose (and glucose) used in the study was very high: 75 grams. To consume that much fructose in one sitting, you'd have to eat a dozen apples, or drink almost five cups of apple juice. Even if you drank Mountain Dew (perish the fluorescent green thought), you'd have to down more than a whole liter to reach 75 grams. Yes, some people do this (you know who you are . . . and we do too, from your eerie green glow), but most of us don’t. Seventy-five grams is not a dose of fructose, it’s an OVERdose of fructose.
- Fructose vs glucose absorption not considered. Researchers voted the two volunteers who tested positive for full-blown symptoms of fructose malabsorption off the island and didn’t include them in the study, but they didn’t take into account that as many as 50% of Americans can’t completely absorb 25 grams of pure fructose, and that at least 80% of us are unable to completely absorb 50 grams of pure fructose, whereas glucose is completely absorbed. Since they used a whopping 75 grams of pure fructose in this study, the chance that any of the volunteers absorbed all of it is very small. No wonder the poor fructose-gulpers were hungrier! This was not a fair fight. [More about fructose malabsorption coming up in article three.]
- Time course too short. Researchers only tested hunger and hormone levels at 30 and 60 minutes after the sugary drinks were consumed. Sugars have effects on appetite-regulating hormone levels for many hours (perhaps even days) after we eat them, so limiting the study to 60 minutes is a very important shortcoming of this study. In the graphs above in the results section, notice that both insulin and appetite reducing hormone Peptide YY (PYY) levels are still rising 60 mins after fructose consumption; we don't know if either of these hormones would have continued rising after 60 mins, and if so, by how much. I wrote about panic attacks and how sugar affects stress hormones on Psychology Today: "Stabilize Your Mood with Food." In that post I showed how sugar causes a spike in adrenaline levels (and therefore hunger) more than 4 hours later. For all we know, as the blood sugar levels of the people in the glucose group began to plummet, they would have become so ravenous that they started consuming members of the fructose group. Who knows? If you don’t wait long enough, you may miss out on some really fun results.
- Oh, and one more thing: who cares? Studying the differences between fructose and glucose on appetite is ridiculous in the first place. We don’t encounter pure fructose in the real world! As you can see in the next post about fructose malabsorption, every naturally occurring food on the planet that contains fructose also contains glucose. In fact, the vast majority of artificial fructose-containing foods and beverages also contain glucose. We virtually always eat fructose in combination with glucose. Therefore, even if fructose were a less satisfying sugar than glucose, to quote Joey Tribbiani from Friends, it would be “a moo point . . . it’s like a cow’s opinion . . . it doesn’t matter!” This is, I’m sure, why there are no episodes of Friends devoted to fructose metabolism.
Hello trees? This is the forest calling…
So it turns out this study doesn't prove that glucose is more likely to curb appetite than fructose. What this study does prove is that intelligent, well-meaning scientists can waste lots of time, money, and intellectual capital conducting studies that seem important on the surface and succeed in getting media attention, but that in the end are unhelpful and irrelevant at best. The New York Times titled their review of this study: "Fructose May Increase Cravings for High-Calorie Foods." What do you think of this headline now that you've looked at the study more closely? While I don't expect even the New York Times to be able to decipher complicated scientific studies such as this one, I do wish that they and all media organizations would be careful not to turn every new study they cover into what could be easily interpreted as a public health recommendation.
Unfortunately, this study is not alone; most fructose research suffers from one or more of the following important shortcomings:
Common fructose research flaws
- Animal-based studies: Many fructose studies use rats or dogs instead of people. Therefore, since each species is a little different, unless you have furry ears and a tail, the results of such studies may not fully apply to you.
- In vitro conditions (test tubes, Petri dishes, etc): Our cells behave QUITE differently when they are taken out of our bodies. Cells inside a living human being are listening to all kinds of sophisticated, carefully-coordinated signals from our hormones, our brains, our muscles, our digestive systems, etc. Take cells out of the body and you have created highly unnatural conditions that rarely apply to whole humans.
- Unrealistically high doses of fructose: The majority of human studies use doses of fructose that most of us do not typically consume in one sitting, exaggerating its effects and making results appear more dramatic than if more reasonable doses had been used.
- Intravenous fructose: Some studies inject fructose directly into people’s veins—don’t try this at home! If you read part one, you learned that the majority of the fructose we consume is removed by the liver before it can circulate in our bloodstream, so our bodies are not accustomed to seeing high levels of fructose in the blood. We eat or drink fructose, we don’t shoot it into our veins.
- Excessive calorie intake: Many fructose scientists ask people to eat their usual number of calories per day and then add a big dose of fructose on top of that. The problem with this is that, if you see a difference in the fructose group, you won’t know if it was because of the presence of fructose, or simply because of the extra calories.
- High starch baseline diets: Every study I read fed people a high-carbohydrate diet, usually containing about 55% percent complex carbohydrate. Since most complex carbohydrates (such as grains and beans) contain starches that break down into glucose, these diets are high in glucose, and therefore high in insulin. On the one hand, this is how most people eat, so it makes sense on that level. However, it makes it very difficult to determine the effect of fructose alone. When fructose is added to this type of diet (or substituted for some percentage of the starch), what we observe is the effect of fructose when added to a large amount of glucose. Since glucose and fructose metabolism affect one another, the results will be confusing.
- Epidemiological methods: Epidemiological studies are “observational” studies in which researchers give ridiculous food and lifestyle questionnaires to large groups of people and then look for patterns in the answers that fit their beliefs about food and health. They are not experiments, and therefore do not have the power to establish cause and effect. The most they can do is notice trends that then need to be tested in clinical trials. Sadly, the vast majority of nutritional information disseminated to the public is based upon epidemiological observations which have yet to be tested, and which is therefore frequently proved wrong later (more than 80% of the time). This is why the public is so often frustrated by the ever-changing headlines about which foods are healthy. It is unfortunate and even dangerous that the media continues to publish headlines based on this kind of guesswork. For more information please see my epidemiology page.
Better fructose science
How could the quality and relevance of fructose studies be improved? I am not a researcher, but I have a few ideas:
- Use humans, use ordinary amounts of fructose/glucose (25 grams or less?), and do not inject sugar into people’s veins.
- Instead of the typical high-starch background diet, use a high-fat, adequate-protein diet in which the only source of carbohydrate is glucose, fructose, or some combination of the two.
- Dose-response studies could be very helpful in determining whether we respond to sugars differently at different doses—this could improve our understanding of whether there are safe levels of consumption of fructose and glucose and whether there is a threshold amount of sugar above which health problems may be more likely to occur. Now THAT would be useful.
- Compare mixtures of glucose and fructose in various ratios to help tease out differences between pure fructose (which doesn’t exist in nature) and naturally occurring sugars, which contain a combination of fructose and glucose in various proportions. Better yet, use real foods! How about comparing a diet in which the only source of carbohydrate is fruit (which contains fructose and glucose) to one in which the only source of carbohydrate is a starchy whole food such as a potato (glucose only)?
Drs. Richard Feinman and Eugene Fine wrote a superb review article titled "Fructose in Perspective" in which they argue convincingly, using my favorite combination of weapons—nutrition science PLUS common sense—that there is no reason to fear fructose more than other sugars, no evidence to support the banning or taxing of fructose, and compare the politics of fructose to those that plagued saturated fat in recent decades. It’s an open access paper, and quite readable, so if you’d like to learn more: http://www.nutritionandmetabolism.com/content/10/1/45
Where does this leave us? In article one: "Has Fructose Been Framed," we shot down the scary metabolic claims that fructose is a poison that turns instantly into fat, robs cells of energy, and raises uric acid levels under ordinary circumstances. Today in article two, we’ve discovered that there is no proof of the fructose-appetite connection. But what about the thing we really care about—our risk for diseases?
Does fructose cause Diabetes? Obesity? Gout? Fatty liver disease? Heart disease? Cancer? The next post in the series, "Is Sugar Bad for You: A Summary of the Research" answers all of these questions!
In the meantime, what do you think about the appetite study? Did it convince you to make any changes to your diet? Will fructose turn out to be the bad guy when it comes to diseases like diabetes and cancer? If not, who do you think it will be?