“Mark already ignores me. I think it is because I am in bed and I am a ‘stationary’ object, because once or twice at home when he saw me up and on the way to the toilet he evinced great interest at once. I am absolutely determined to check myself in any tendency to be sad when I see him forgetting and ignoring me, because what matters is that he shall be happy where he is. It is a mercy that he is at an age when ‘now’ is everything. It takes only a day or so to adjust himself completely to new people. I know Bill is right when he says that when I do come back to him he will, within a week, be as though I had never been away.”
… I feel hopeful about Mark remembering me again when I come back, because though last time he got quite out of the way of thinking about my being anything but an inanimate object, and used to distress me terribly by crying for someone else if I got him out of his pram or bathed him, yet within a week or so of our being back he was always trying to come to me.
These are the thoughts of a British woman of 32 who was suddenly separated from her husband, Bill, and fifteen-month-old baby, Mark in 1944, an era when those infected with tuberculosis were isolated from society and placed in sanatoriums. Extracts from her diary were published in the Journal of the Royal Society of Medicine in 2004 (1).
Tuberculosis – The Robber of Youth
Tuberculosis is an ancient plague that has flogged humankind throughout history. It has surged in great epidemics and then receded, thus behaving like other infectious diseases, but with a time scale that differs from many other known epidemic cycles.
The disease reached epidemic proportions in Europe and North America during the 18th and 19th centuries, earning the sobriquet, “Captain Among these Men of Death (2).” The bacterial species, Mycobacterium tuberculosis, may have killed more persons than any other microorganism.
Tuberculosis had many nicknames that produced despair and horror; the “great white plague”, “the robber of youth”, and “the graveyard cough”. Therapeutic options were limited, the mainstay of treatment was rest and fresh air.
In the late 19th and early 20th centuries, sanatoria were developed for the treatment of patients with tuberculosis. In 1859 Herman Brehmer opened his Heilenstat in the Silesian Mountain village of Gobersdorf in Germany, emphasizing a regimen of rest, a rich diet, and carefully supervised exercise (3). This facility is considered the first sanatorium devoted to the treatment of tuberculosis.
The first sanatorium in North America was opened in Asheville, North Carolina, by Joseph Gleitsman. Gleitsmann, a German-born and trained doctor, is believed to have selected Asheville because of its optimal combination of barometric pressure, temperature, humidity, and sunlight which he believed to be essential to healing tuberculosis (3).
Although the sanatoriums were supposed to provide treatment and care, they also isolated tuberculosis sufferers not only from those whom they might infect but also from matters of the world they lived in. In his famous novel, The Magic Mountain, Thomas Mann writes about Hans Castorp’s symbolical transport away from the everyday life and mundane obligations he has known, to the rarefied mountain air and introspective little world of the sanatorium
It is hard to comprehend the agony associated with being taken away from your loved ones and isolated from the rest of the world at the same time as realizing that death would be the most likely outcome. Thankfully, tuberculosis is not an epidemic anymore. Although not completely eradicated, it can be prevented and cured. Communicable diseases are no longer the most common cause of death and disability around the world.
Today’s Epidemic – The Metabolic Syndrome
Today’s epidemic is a horse of a different color. The sanatoriums have been closed. Instead, we are facing a collection of noncommunicable diseases that to a large extent are caused by the way we live our lives, what we eat, whether we smoke, and whether we are couchbound or not.
There is abundant evidence that type-2 diabetes, cardiovascular disease (CVD), non-alcoholic fatty liver disease (NAFLD), hypertension (high blood pressure), and some types of cancer may all be the result of biochemical alterations associated with the modern Western diet. These metabolic changes, collectively known as metabolic syndrome, are characterized by insulin resistance, derangements of glucose metabolism, hypertension, high blood triglycerides, and low blood levels of HDL cholesterol.
Of course, we might easily conclude that the metabolic syndrome is a consequence of the obesity epidemic. In fact, recent evidence shows that the worldwide prevalence of overweight and obesity combined rose by 27.5% for adults and 47.1% for children during the past three decades (4). Interestingly, these increases were seen in both developed and developing countries.
However, although metabolic syndrome often seems to travel with obesity, it occurs in normal-weight people as well. Furthermore, many obese individuals do not have the metabolic abnormalities that characterize the metabolic syndrome. Hence, it has to be assumed that it is not only about body weight and calories. But, if weight gain itself is not the culprit, what is?
For many experts, the answer is obvious (5). It’s added sugar (6), fructose, the silent ‘sweet killer’.
The Silent ‘Sweet Killer’
Fructose is massively used by the food industry due to its very sweet taste and lack of inhibition of satiety compared with other types of sugar (7).
Fructose is a monosaccharide that is found naturally in fruits. It is typically consumed as sucrose (table sugar), a disaccharide composed of equal parts of fructose and glucose, or as a component of high-fructose corn syrup (HFCS). According to US Department of Agriculture (USDA) data, 60% of HFCS used is 55% fructose, and the remainder is typically 42% fructose (8).
For the last few decades, there has been an enormous rise in fructose consumption in the U.S. It is highest among adolescents (12–18 years). The largest source of fructose is sugar-sweetened beverages (30%) followed by grains (22%) and fruit or fruit juice (19%) (9)
There is strong evidence from both experimental and animal studies suggesting that high fructose consumption can lead to insulin resistance, high blood pressure, lipid abnormalities, and NAFLD.
The metabolic effects of fructose are very different from those of glucose. Fructose is metabolized almost exclusively in the liver whereas glucose can be metabolized by most cells of the body.
The entry of fructose into cells is not dependent on insulin and does not promote insulin secretion, unlike glucose. Fructose promotes an increase in blood levels of triglycerides (10). Elevated triglycerides caused by excessive fructose intake may be a precursor of insulin resistance (11).
Fructose-sweetened beverage consumption habits are associated with a central fat distribution and visceral obesity (12). Visceral fat tissue appears much more damaging to health than other types of fat tissue. Obese individuals with excess visceral obesity have a higher risk of diabetes, lipid disorders, and CVD than those with less visceral fat accumulation (13).
However, this may all be circumstantial evidence. The question remaining is whether fructose plays a causative role in the biochemical abnormalities associated with obesity and metabolic syndrome? Does restricting fructose intake improve these metabolic abnormalities? Will reducing fructose reduce the risk of noncommunicable diseases such as diabetes, CVD, NAFLD, and cancer?
Interestingly, these important questions were addressed recently in a very important paper published by Robert Lustig and colleagues from the University of California (14).
Lustig and coworkers studied 43 obese children (ages 8-19) with metabolic abnormalities typical of the metabolic syndrome. All were high consumers of added sugar in their diets (e.g. soft drinks, juices, pastries, breakfast cereals, salad dressings, etc.).
The children were fed the same calories and percent of each macronutrient as their home diet; but within the carbohydrate fraction, the added sugar was removed, and replaced with starch. For example, pastries were taken out, and bagels put in; yogurt was taken out, baked potato chips were put in; chicken teriyaki was taken out, turkey hot dogs were put in. Whole fruit was allowed. Dietary sugar consumption went from 28% to 10% of calories.
To dissociate the metabolic effects of dietary sugar from its calories and its effects on weight gain, it was important to keep weight constant. If the children were losing weight, they were told to eat more. The goal was to remain weight-stable over the ten days of study. On the final day, the children came back to the hospital for testing on their experimental low-added-sugar diet.
The Bottom Line
According to the World Health Organization (WHO), noncommunicable diseases kill 38 million people each year (15). Sixteen million of those deaths may be considered premature as they occur before the age of 70. CVD account for most of these deaths, followed by cancers, respiratory diseases, and diabetes. According to WHO, tobacco use, physical inactivity, the harmful use of alcohol, and unhealthy diets all increase the risk of premature death from noncommunicable diseases.
A worldwide study published last year suggests that sugary soft drinks kill 184,000 adults every year (16). The study shows that 133,000 deaths from diabetes, 45,000 deaths from CVD, and 6,450 deaths from cancer were caused by fizzy drinks, fruit drinks, energy drinks, and sweetened iced teas in the year 2010.
By now, it should be obvious that sugar-sweetened beverages are a single, modifiable component of our diet that can impact preventable death and disability in adults in high-, middle-, and low-income countries, indicating an urgent need for strong global prevention programs.
The success in preventing and treating communicable diseases like tuberculosis may be considered the biggest achievement in the history of medicine. However, it has left us with another enormous problem of a completely different nature. This time, we are not dealing with a ruthless microorganism but with an environment of our own creation, a culture that manages to encourage unhealthy lifestyle and junk food.
Defining our targets is the first step. It’s about time we realize that added sugar, fructose, in particular, is the silent ‘sweet killer,’ the 21st century’s white plague, today’s robber of youth.
If you think that the statistically significant changes in the table reflect biologically meaningful changes then you have to accept that kind of data when it shows that fat is bad and fruits ‘n vegetables are good. If WHP is an authority on sugar, it has te be accepted as an authority on red meat, etc. Statements like “kills 184,000 adults every year” are as totally idiotic — you could not possibly know such a thing. I hate to fight with my friends but this is dangerous because if you replace sugar with starch in a real way, you are more likely to create problems. The only question is whether you think that I am as much of an authority as Lustig and if I explained to you how all this was wrong, you would have an open mind and consider that you’d made a mistake. If you’re sure you’re sure that you’re right and that I’m wrong then nothing else to say.
Richard. I appreciate your input an certainly respect your knowledge and authority. I certainly have an open mind because I don’t know where the truth lies and would be happy to admit that I’ve made a mistake. I’m definitely not sure that I’m right and you’re wrong.
My article is supposed to highlight the great contrasts between the public health issues of today as compared to 70 years ago, and how our targets/culprits have changed. It’s meant to provoke thoughts. For the layperson (and all of us i guess) it may be helpful to see things from I historical perspective.
I’m aware that I use strong words when it comes to fructose. It was not supposed to be fear-mongering. However, I think there is abundant evidence that added sugar is a causative factor when it comes to insulin resistance, visceral obesity and NAFLD and thereby type-2 diabetes.
Anyway, wouldn’t you agree that added sugar is of concern from a public health perspective?
Everything’s of concern butt it comes down to this game: you have to give your patient one single piece of advise and the insurance company will only pay for one sentence. Choose between:
1. Replace carbohydrate, any carbohydrate, in your diet with fat, any fat.
2. Replace fructose (sugar) in your diet with glucose (starch).
Which one? I can defend my choice. (Exclude trans-fat I you like)
Richard. That’s pretty tricky. However, I don’t think I ever advise 2. But, rephrasing it and saying “replace fructose (sugar) with diets high in minimally processed carbohydrate and dietary fiber” would be the traditional approach which I might consider. But, for those with insulin resistance, visceral obosity, metabolic syndrome or DM-2 I generally prefer 1 (although it might go against most clinical guidelines). What do you say?
If you were to suggest any of your patients to replace ANY carbohydrate with ANY fat, you’d be giving them bad advice.
I agree. The word ANY is a problem here, both with regards to carbs and fats.
Even on the ANY, that is not what has been shown. The strong data is on any carbohydrate and any fat. (We agree to leave out trans-fat although I think that even that is exaggerated). I, of course, share your opinion that type makes a difference but I don’t think we have the data. (I could be wrong. I would love to hear good data on this issue).
“The strong data is on any carbohydrate and any fat.”
That is simply not true. If you’re referring to any of the meta-analyses of RCTs, you do realize that these findings are severely compromised due to several factors (e.g. participant not adhering to the intervention, statistical underpowering etc. etc.). In other words, they provide evidence – at best – that virtually ANY given intervention is of limited benefit because people simply don’t comply. They DON’T provide evidence that the intervention per se is useless.
As I stated above, metabolic ward studies clearly point out the problems created by e.g. trans fat intake.
Dr Feinman’s having problems with modelling studies? How about you offer detailed criticism instead of just saying something is “idiotic”?
Perhaps you’d also like to explain how one would be more likely to create problems when replacing sugar (ADDED sugar, that is) with starch? Please. Show us the way.
Mie seems to be suggesting that my comment was a knee jerk reaction for Axel’s blog rather than for a scientific paper and that because we are among friends and roughly in agreement that I spoke very loosely. Hmmm. Have to think about that. My point is really that this is an example where statistical significance is mis-used. The numbers are too similar, the error is too large and the experimental design is too loose to make such a strong conclusion. The statistical significance does not take account of the biology which comes first. As for deducing absolute figures as to how many people ‘s lives will be saved from rough epidemiologic data, the burden of proof is on anybody who thinks it is not idiotic. I presented a very detailed analysis as to why the exaggerated condemnation of fructose does not go with the science — not that it is wrong but that we are doing the same hysterical reaction that was done with saturated fat — Our paper, peer-reviewed (and not the most congenial peers), was published in Nutrition & Metabolism. The most obvious point is that, depending on conditions, half of ingested fructose can be converted to glucose — that is why the GI of fructose is 20 and not zero — so you just don’t know what is doing what The real reason for believing fructophobia is exaggerated is that in the choice above, replace carbohydrate with fat or replace fructose with glucose, that experiment could have been done by the people who got funded to do the more ambiguous studies (discussed in our N&M paper). That they didn’t do so — as they say in court — goes to credibility.
“The numbers are too similar, the error is too large and the experimental design is too loose to make such a strong conclusion.”
Elaborate, please. What numbers? What do you mean by “loose”?
“The statistical significance does not take account of the biology which comes first.”
Staring at statistical significance can, of course, be misleading. But, as above, what is this “biology” that should “come first”?
“As for deducing absolute figures as to how many people ‘s lives will be saved from rough epidemiologic data, the burden of proof is on anybody who thinks it is not idiotic.”
Nice try but … No, actually, it wasn’t a nice try but kinda lame and half-assed attempt. If you consider the paper seriously flawed, its YOUR duty to explain why. In detail.
“I presented a very detailed analysis as to why the exaggerated condemnation of fructose does not go with the science — not that it is wrong but that we are doing the same hysterical reaction that was done with saturated fat –”
Yes yes, dose and context. However, as there’s clear evidence from e.g. metabolic ward studies showing what’s the issue with saturated fat, I find it … shall we say, comical, that you – with your history of reliance on varied short-term risk marker studies (your low carb fundamentalism) – would dismiss them in preference of e.g. epidemiological data and/or poorly designed RTCs. 🙂
“The most obvious point is that …”
… your finish didn’t really offer any specific criticism.
I agree with you on the issue of fructophobia (dose and context – and the Lustig study has several problems, the most obvious being that the design didn’t include a control group). But you’ve got to try harder if you want to be considered an expert in this field. 🙂
Mie, I never said that I was an expert, I am giving you the opinion of a professional biochemist. I am fundamentally in agreement with Doc’s Opinion. I am giving my perspective — I think he exaggerated fructophobia although he has the additional burden of actually evaluating the patient. He probably takes my view seriously but you can evaluate it or not. If you think that sugar is a poison, or uniquely bad, don’t eat any.
I thought the “saving lives” was obviously foolish but, outside of inborn errors of metabolism and isolated cases like pure fructose administered in surgery, nobody has ever been admitted to a hospital for fructose poisoning. Nobody has died from fructose poisoning. But, again, if you think fructose is life-threatening, don’t eat any. .
Carbohydrate is a problem in diabetes but almost everybody in the world consumes more glucose than fructose, Almost all fructose in the diet is in combination with at least equimolar glucose.
If you have some personal objection to what I am writing, don’t come to my birthday party. Hell, I wasn’t going to invite you anyway.
“If you think that sugar is a poison, or uniquely bad, don’t eat any.”
I don’t think it’s “poison” but it certainly brings about many health problems (added sugar, that is).
“Nobody has died from fructose poisoning.”
Ingenious. Is that the best you can do when asked to provide DETAILED criticism of a study you label as idiotic?
“If you have some personal objection to what I am writing, don’t come to my birthday party.”
I object nonsense, be it you or someone else.
Sorry. I meant that there is no evidence that fructose, or sugar, per se (that is, outside their contribution as carbohydrate) has any effect on mortality. The collection of epidemiologic studies with odds ratios of 1.5 and 15 confounders are, in my view, meaningless. But it could be a matter of taste and if you think they are real, go with that.
On treatment, I told you the experiment that needs to be done. I told Rob Lustig and I told Richard Johnson. If they do it, we will know. It’s all the same to me. (Scientifically, that is; personally, anecdotally I find starch to be a bigger threat to my weight than sugar).
“Sorry. I meant that there is no evidence that fructose, or sugar, per se (that is, outside their contribution as carbohydrate) has any effect on mortality. The collection of epidemiologic studies with odds ratios of 1.5 and 15 confounders are, in my view, meaningless.”
So … an odds ratio of 1.5 in an individual study.Ok. So what?
In addition, what is wrong with the modelling in Singh et al. and – in general – adjusting for confounders? Could it be that you still have no better understanding of epidemiology than what you’ve demonstrated in the past in your blog? 🙂
“(Scientifically, that is; personally, anecdotally I find starch to be a bigger threat to my weight than sugar).”
If you consume it in excess (and don’t consume a lot of sugar), yes. Scientifically (that is, scientifically) that is irrelevant in this discussion.
I agree that the biggest weakness with Lustig’s study is the absence of a control group. In the discussion part of the paper they explain why using a control group would present difficult challenges. However, I have a hard them agreeing with that.
I don’t think there’s a problem with the statistics and the statistical differences are indeed quite impressing. However, it all comes back to the design of the study and the method; using each patient as his own control. In my opinion, every interventional study needs a control group. So, therefore, I would regard this as a pilot study with very promising results, clearly suggesting that fructose may be more problematic than other types of carbs in this patient population (which is very important to comprehend). Results of randomized controlled studies are eagerly awaited.
Now, this is a very important issue because of the rapid rise in metabolic syndrome among children. NAFLD and NAS are becoming the most common causes of liver failure and liver transplantation.
Yes, there simply is no reason for not having a proper design with a control group.
Or well, there IS a reason, but I’m hoping that wasn’t the case here.
The problem with the statistics is that it is an average and you don’t know how many kids got better and how many got worse. The trajectory of an average doesn’t tell you about the individual changes. If you’re okay with that, you are easy to please. Also, some of the changes are small and you don’t have control over the input to a degree that justifies taking small changes seriously. (“If the children were losing weight, they were told to eat more”).
That’s a bit better. You’re, on course, kind of beating a dead horse as Lustig’s paper is – simply put – inadequate in soooo many ways. 🙂
Now, how about you tackle the problems in the modelling study by Singh et al. next? Or explain where you’ve found this “strong data” on “any” carb or fat?
In the text Iread that you conclude that everything was better.. HDL-c also.. I don;t see it… From 1.2 to 1.0
Hdl shoud then have been higher.. to be seen as better..