All carbohydrates are not the same. Table sugar and High fructose Corn Syrup (HFCS) are both half glucose and half fructose, whereas starchy foods like grains and potatoes contain entirely glucose.
Glucose is taken up throughout the body, but fructose must be converted into glucose by the liver before it can be used. The process of metabolizing fructose in the human liver is sloppy and generates nasty blood fats as by-products. Habitual high consumption of fructose could therefore lead to chronic problems like metabolic illness and increased cardiovascular risk.
Unlike glucose, fructose does not trigger the body’s hormonal satiety signals, such as insulin secretion, so the body is unable to include the calories from fructose intake in its energy balancing efforts. The sugar calories are basically invisible to the body’s fat thermostat. This leads to runaway weight gain, like with the obese fifth graders crowding Lustig’s clnic.
Sugar can be addictive. It activates the reward centers of the brain.
The sugar industry has worked diligently for years to cover up information that might suggest sugar could be harmful. (This is also exposed in some detail by the 2012 Mother Jones article, “Sweet Little Lies.”)
There it is. A comprehensive explanation for metabolic syndrome, for increasing obesity, and maybe even for the heart disease epidemic.
Some additional support for the hypothesis:
As Lustig points out, lack of sugar is one of the few things that the “Japanese diet” (low fat, based on rice) and the Atkins diet (high fat, very low carbohydrate) have in common, and they both seem to work for weight control.
The “French Paradox” might be explained by the fact that the French use very little sugar, even in desserts. In the Blue Zones, they use “about the same amount of naturally occurring sugars as North Americans do, but only a fifth as much added sugar.”
Sugar is added to everything in the U.S. Breakfast cereal, bread, tomato sauce, yogurt, you name it. Even people who don’t particularly eat a lot of sweets may be getting more than they think.
Unfortunately for such a tidy hypothesis, there are some major observations that don’t match up.
For starters, we know of populations eating a lot of sugar and suffering no apparent ill effects. It’s easy to imagine that indigenous cultures following traditional diets and enjoying robust health would eat practically no fructose, blissfully out of reach of the industrial bane of white sugar. Not exactly. Obesity researcher Stephan Guyenet has a topical writeup, “Sugar Intake and Body Fatness in Non-Industrial Cultures.” Of his two examples, one — the Hadza — ate a diet extremely high in natural sugars like honey and baobab fruit, and the other — the Kuna — ate a diet that matched the usual template of indigenous fare but included a substantial amount of white sugar obtained through trade. Both groups were lean and had low blood pressure.
Furthermore, we have a well documented example within the Western tradition. Walter Kempner’s rice diet of the 1940s — ultra-low in fat, protein, and sodium — allowed unlimited sugar, fruit, and fruit juice. Although hard to stick to for long, the diet was wildly successful in treating obesity, diabetes, coronary heart disease, and more. If fructose were inherently poisonous, this result would make no sense.
What about the statistical correlations? Wasn’t the increase in obesity rates from the 1970s through the 2000s correlated with increasing sugar intakes? Behold the “Australian Paradox”: between 1980 and 2003, Australian obesity increased 3-fold, while total per capita sweetener consumption dropped by 16%.
And then there’s the curious case of Natasha Harris, the New Zealand woman who died in 2013 from drinking too much Coca-Cola. She was consuming the equivalent of two pounds of sugar a day; inferring from one news source which said she’d been doing this for “years” and that she’d had “an estimated 50,000 pints” at a rate of “18 pints a day,” we can deduce she’d endured more than 7 years of this level of sugar intake. The press reports that she did have an “enlarged liver,” but get this, the sugar intake isn’t what killed her. What killed her was apparently heart arrhythmia from the caffeine. It struck me that if sugar is really so bad for you, she would never have survived so long. And if fructose is inherently fattening, she should have been obese, but the face in those tabloid photos was not really a fat face.
So let’s be honest, fructose consumption is not an immediate threat to the health and well-being of anyone (except, perhaps, folks with hereditary fructose intolerance). If there is any issue, it will have to do with effects of excessive fructose consumption over long time periods (decades), or else calorie overconsumption in general.
Worth noting: Lustig has been criticized for relying too much on animal studies. Two problems with these studies: a) the animals are fed comparatively huge amounts of fructose — upwards of 60% of calories; b) mouse metabolism is unlike human metabolism in at least one critical aspect: animals like mice can convert 50 percent or more of the carbohydrates they consume into fat via the process of de novo lipogenesis (DNL), whereas humans can barely do this with 10% of the carbohydrate they consume. (Yes, that’s right: sugar is not readily converted into fat, in humans, no matter what the likes of Gary Taubes and Ron Rosedale say.)
It turns out, yes, we know of at least two reasons.
High blood triglycerides are a risk factor for heart disease. Fructose can raise fasting blood triglycerides, either through de novo lipogenesis (DNL) or, more likely, through a mechanism of impaired clearance. (Note: As mentioned above, DNL does not produce enough fat to meaningfully affect body weight — since humans are bad at DNL — but it can be enough to raise blood levels.) The link between fructose and blood triglycerides is not controversial; both low-carb and low-fat diet proponents acknowledge it. For example, Dr. John McDougall, prominent low-fat diet advocate, explains that starch (glucose-based carbohydrate) does not raise triglyceride levels, even though fructose often does.
Dr. Luc Tappy provides some gory detail in a 2012 interview. He states: “In healthy subjects, short-term overfeeding studies with large doses of fructose (in the 1.5 to 3 g/kg/day, corresponding to 15 to 30% total energy requirement) have repeatedly reported an increase in fasting and postprandial triglycerides, mainly associated with very low density lipoproteins (VLDLs), and an increase in concentrations of apoB100 (a component of both VLDLs and low-density lipoproteins (LDLs)). […] Moreover, there is strong evidence that 24-hour triglyceride concentration is an independent risk factor for atherosclerosis. In addition, a high plasma VLDL-triglyceride concentration leads to the generation of smaller, more dense LDL particles through the cholesteryl-ester mediated transfer of lipids between VLDL and LDL particles. This process is further enhanced in fructose-induced hypertriglyceridemia, probably because of the impaired VLDL-TG clearance, and hence an increased residence time of VLDL in the blood. Both fructose and sucrose therefore lead to an increased proportion of small dense LDL particles within the LDL fraction, a phenotype that is clearly associated with an increased cardiovascular risk.”
In the same article, when asked what amount of fructose has to be consumed to see these adverse effects, Dr. Tappy suggested a range with a midpoint of 75 g/day, but mentioned that “even with moderate amounts of fructose (40 g/day) that do not change fasting plasma triglycerides, one can observe a shift from large to more atherogenic small, dense LDL particles.”
Various sources around the Internet seem to concur on an average daily sugar consumption of 100-150 g/day for Americans. That’s 50-75 g of fructose, and since it’s an average, half of Americans would be above that. How much fructose did the Kuna eat? Guyenet’s guess is a bit under 40 g/day. For reference: A typical apple or candy bar has 10-12 g of fructose, a 12 oz can of soda has around 20 g, and a typical sugar packet (for coffee or tea) around 2 g.
Let’s consider just sugar-sweetened beverages for a minute. A casual look at medical literature turns up repeated findings that humans do not adjust their food intake downward to fully account for extra calories from beverages. Here are two sample studies:
Those findings seem to apply to any beverage, as long as it’s caloric (i.e not a diet soda). They even saw it with milk. Fructose-containing sweeteners happen to be the most popular way to sweeten beverages, because, well, they taste sweet. The Harvard School of Public Health confirms that sugary drinks are associated with obesity. If you drink calories, you’re probably overeating.
What about solid food? Sweeteners increase the palatability and energy density of food. Both of these effects can lead to passive overeating. Indeed, there is some evidence that the reward centers of the brain could be involved. This is a complex subject, so I’ll leave it at this thought: There’s a big calorie difference between overdoing it on jelly beans, vs. overeating pound cake. And there’s a big palatability difference between pound cake and crackers.
But something everyone agrees on, including Lustig: whole fruit is still good for you. The fiber, phytochemicals, and nutrients in intact fruit more than make up for any negative impact of the fructose involved. With juice, you lose the fiber and can drink 5-10 oranges without realizing it, but with the whole fruit you basically can’t go wrong.
Last time I searched the phrase “gave up sugar,” I saw a parade of breathless anecdotes recounting how lives changed after a harrowing trial of going without sugar. For some people, sugar seems to be a disruptive substance, a hidden addiction, and giving it up is nothing less than a revelation. If you have any suspicions this might be the case for you, it’s worth an experiment.
Do you have high triglycerides? Are you trying to lose weight? Probably a no-brainer to cut out added sugar except for special occasions. You really could do worse than follow the French or Blue Zones example.
Note: This post was originally published with the title “Fructose Considered Harmful,” but it was reworked in 2016 to incorporate new sources and reflect an updated view.