The great medical disconnect
Posted by Peter Attia on Dec 26, 2011 in Nutrition & Health, Weight loss | 14 comments
There is probably no greater disconnect in medicine than the root cause of obesity. Even if you think you already know the answer to this “obvious” question, it’s still worth reading on. The reason this question matters, of course, is clear to everyone. Obesity (and more broadly the syndrome we define as metabolic syndrome) predisposes us to virtually every disease afflicting us in the modern age. Below is a simple graphic from the journal Nature showing the linkage between obesity and all of its sequela:
When you are obese, your risk of disease goes up. This is not disputed. Here is where the controversy starts…what actually makes us obese?
Obesity is a disorder of fat accumulation – fat cells accumulate too much fat, relative to how much fat the body breaks down. Conventional wisdom, however, says obesity is a disorder of eating too much and/or exercising too little. These are not the same thing.
Let’s turn to a well-respected source of medical information, Lehninger’s Principles of Biochemistry (the so-called “bible” of biochemistry).
Go to the index and query, what makes fat cells fat? (the technical way of asking this question is, what causes adipose cells to accumulate triglycerides) and you’ll get the following response:
“High blood glucose elicits the release of insulin, which speeds the uptake of glucose by tissues and favors the storage of fuels as glycogen and triacyglycerols, while inhibiting fatty acid mobilization in adipose tissue.”
Let me translate this: Eating glucose (carbohydrates) increases insulin levels in our body. Insulin drives glucose into liver and muscle cells as glycogen (in small, finite amounts) and into fat cells as triglycerides (in unlimited amounts). Insulin also inhibits the breakdown and utilization of fat. Parenthetically, this means no one disputes this graph.
Next, query the exact same text book, what makes people fat (the technical way of asking this question is, what causes human obesity) and you’ll get the following response:
“To a first approximation, obesity is the result of taking in more calories in the diet than are expended by the body’s energy-consuming activities.”
Pretty straight forward, right? People get fat because they eat too much relative to how much they exercise.
But how can both of these statements be correct? Fat cells get fatter because of insulin’s influence on them, but people get fat because of overeating? One of these statements is correct and the other is not. I hope to help you understand why the first one is correct and the second one is not.
Let’s look at another example of this tragic medical disconnect. I say “tragic” because the failure of most physicians (including me up until recently) to understand this is a large part of why the problem persists. When people grow vertically (i.e., when people get taller) no one disputes that a hormone drives this action. The hormone primarily responsible for this, of course, is growth hormone. Sometimes, in rare cases, people develop a tumor in their pituitary gland (from where growth hormone is secreted) and they end up secreting too much of this hormone. The result is that they grow vertically in an unregulated manner. That is, they get too tall. Again, no one in medicine disputes this.
Why is horizontal growth (i.e., obesity) different? I, and many others, argue that unregulated horizontal growth is also regulated by a hormone – insulin. In both cases – growing vertically or growing horizontally – the responsible hormone drives (literally induces) the person to eat more than they burn. The fellow on the left in the figure below (nearly 9 feet tall) ate a lot to get that tall, but he did not get that tall because of eating too much. He got that tall because he had too much of a certain hormone that drove him to over-eat relative to what he was able to burn. The medical establishment tells us the girl on the right got to be that size because she ate too much. Wrong. She ate too much (relative to her energy expenditure) because she had too much of a different hormone – insulin – driving her to eat too much.
This subtle difference in actual causality, and the resulting misunderstanding that has led to current health and nutrition policies, is the root cause of the health problems afflicting us today.
Hi Peter,
A short question: is it important to switch from regular diet to low sugar / high lipid diet in a gradual manner? Will it be possible for the body to change its metabolism to ketosis if the switch is a sudden one, not a gradual one?
Happy Holidays, Amir
If you took the most radical shift possible – complete removal of carbohydrates (or even complete starvation) – your body would generate sufficient ketones to satisfy your brain within about 48 hours (by which point you will have little glycogen left in your body). Without this ability to produce ketones in the absence of any glucose you would actually die of hypoglycemia within a few days of starvation. I think the real question you’re asking, however, is what is the best way to transition from a state of high-carb eating to one of low-carb (or even ketotic) eating? There is no right answer. Metabolically, it only takes a short while to switch over, but the full period of adaptation can take anywhere from a week to several weeks. Furthermore, for some people (like me) it was easier – at least psychologically – to do it in stages. I’ll be writing about this in much more detail later on.
“If people let the government decide what foods they eat and what medicines they take, their bodies will soon be in as a sorry state as the souls who live under tyranny.”
-Thomas Jefferson
Peter,
I think you should read a lot more studies on the actual distribution of nutrients rather than fixating on all of the known physiological effects of insulin. In particular, you should read the metabolic tracer studies of Dr. Hellerstein at Berkeley. While it is true that insulin is an anabolic hormone, the most negative effect of a high carbohydrate diet is that it strongly promotes the oxidation of carbohydrates. You see, the body actually cares about and monitors carbohydrates, while being oblivious to fat intake and storage. On a diet high in carbohydrates, ingested fat will be sent directly to adipose tissue while almost all the carbohydrate will be oxidized. It may not be a large issue in practical terms (you will still be adding fat) it will not be due to insulin directing glucose into the fat cells to be converted to fat. In fact, that would be the best possible outcome, since the process is inefficient enough that you would only gain around 65% of the weight you would gain by directly storing fat. You see, very little conversion from carbohydrates to fat happens in humans. It has been calculated (see Hellerstein) that less than 10% of the fat in your body was synthesized by lipogenesis rather than ingested as fat. For folks on a standard diet, a full 50% of their ingested fat will go directly into a fat cell rather than be oxidized. For folks on a high carbohydrate diet, they will have to consume more calories than they can burn or store as glycogen to even begin to show appreciable conversion of carbohydrate into fat. So, can insulin stimulate this process? Yes, but it is a minor metabolic pathway in humans, and most glucose will be disposed into muscle and liver.
Jack, thanks for your question. Please email the study you’re referring to and I’ll be happy to take a look when I get a moment. I am familiar with some of the work of Hellerstein, and contrary to your comment, I do actually read a lot – as much as I can (but I do wish I could read more)! In fact, I do have a paper by Hellerstein and Schwartz that looks at what you’re describing. The thing you need to keep in mind – and I’m sure you already know this, but some folks do get confused – is that many of these studies actually under-feed glucose, relative to what “real” people eat. In other words, if I took a group of subjects and removed all (and I mean *all*) fructose from their diet (e.g., no fruit, no sucrose, no HFCS), and fed them small amounts of glucose, you’re right, I could prevent a great deal of de novo lipogensis. But, and this is a very big BUT, I would need to ensure that I was only feeding them a small amount of glucose, probably less than 60 gm/hour. This way, I could force a metabolic prioritization from direct glycolysis (first) to glycogen formation (second). But keep in mind what happens when these two sources are “full” – you will make fat out of glucose. Even the most highly trained and muscular athlete can store roughly 400 kcal of glycogen in their liver and 1200 kcal in their skeletal muscle. And that’s assuming both “tanks” are completely empty – a nearly impossible physiologic condition. Unfortunately, most people consume so much glucose (and fructose), that the tight experimental conditions of Hellerstein do not apply.
As we’ve learned from Rob Lustig (and I’m actually going to post on this very topic this week), gram for gram, fructose is converted to fat more readily than glucose, but this does not mean glucose is not converted to fat (i.e., when glycogen stores are full and you are not immediately oxidizing it). Finally, and perhaps most importantly, keep in mind that there is another “sinister” effect of glucose (that is not true of fructose, by the way): glucose raises insulin levels. When insulin levels are high, we stop beta-oxidation of fat. So even in the “best” case (i.e., a case when we are not making new fat as we eat glucose) we’re not burning fat when we eat glucose. This is profoundly important and should not be overlooked.
Peter,
At the end of your reply to Jack you say, “Finally, and perhaps most importantly, keep in mind that there is another “sinister” effect of glucose (that is not true of fructose, by the way): glucose raises insulin levels.”
From my research and understanding (I believe from Dr. Richard Johnson and/or Dr. Robert Lustig), Fructose does not spike blood glucose, but *does* raise insulin levels. Glucose raises blood glucose levels and thus causes insulin to be secreted. I believe Dr. Robert Lustig, says that Fructose keeps insulin levels elevated thus causing more fat storage.
Dr. Richard Johnson finds that Fructose, by raising uric acid and inhibiting nitric oxide, completely blocks the effect of glucose to stimulate the secretion of insulin to increase the uptake of glucose in cells, thus causing insulin resistance.
I should add that there is a lot more in all of the biochemistry involved that I truly grasp
and so I may have misunderstood some of this stuff.
That’s correct. Fructose does not actually raise insulin levels. The beta-cell of the pancreas does not “recognize” it, and it is transported into the liver via the GLUT 2 transporter (rather than the insulin-activated GLUT 4 transporter that puts glucose into cells). I think what you may be thinking about is that fructose DOES indirectly contribute to insulin resistance, but at the moment you are ingesting it, it’s not raising insulin levels, per se.
Petter thanks for your great blog. I have been very low carbing for a few years and had achieved normal weight and good health but since reading your blog I reduced the proteins and increased the fat. Wow I now feel fabulous it have given me so much more energy, have more endurance, less hunger, clearer mind. I could not believe I could be improving so much more doing a ketonic diet.
I wanted to ask you about lipids since I have no way of doing a NMR or VAP. There is some discussion now (Jimmy Moore)that the best indicator for CAD is TC/HDL but I had read that the most accurate predictors is TG/HDL even equal to LDL-P like it said here what do you think? What is your opinion on what best if no NMR or VAP is available?
http://www.theheart.org/article/767865.do
Dr Karim El Harchaoui (Academic Medical Center, Amsterdam, the Netherlands) and colleagues report their findings online January 22, 2007 in the Journal of the American College of Cardiology.
Coauthor Dr Matthijs Boekholdt (Academic Medical Center) told heartwire that, in patients with only moderately elevated LDL-C, clinicians would do better to pay attention to HDL levels and triglycerides as markers for future CAD risk rather than worrying about testing LDL-P.
“Although it’s good to be aware of LDL-P, it’s expensive to test. We show that HDL-C and triglycerides provide similar predictive information. Most doctors do not look at HDL-C or triglycerides because, unlike LDL-C, treatment for these parameters is not mainstream clinical practice. It would be more useful to be aware of these results, particularly in those with the metabolic syndrome, abdominal obesity, or diabetes,” he said.
Sam, it’s certainly a good place to start by looking at TG to HDL-C ratio, and that it probably more helpful than LDL-C, but at least 30-40% of people with low or normal LDL-C have elevated LDL-P. So the questions is how predictive is TG/HDL-C of LDL-P? To be clear, the damage is CAUSED by LDL-P, so by *definition* nothing is more accurate for predicting artherosclerosis than LDL-P. It’s like saying hair color is a good predictor of eye color. Sure it can help…but it’s no where near as good as eye color is for predicting eye color! There is great resistance in the medical community to move away from LDL-C and HDL-C, just like there’s great resistance to move away from the idea that fat is harmful, despite much evidence. If you’re at even modest risk, and/or your TG/HDL-C is above 2.0, I’d strongly recommend you pay for the NMR, even out of your own pocket.
Thanks Peter. My TG/HDL thankfully is below 1. My total TC is below 200 so I don’t have a doctor bothering me about fat in my diet. But I have always advice friend and family that TG/HDL was a better predictor than TC/HDL because that what I had read. So just wanted to be sure that I was giving the correct advice since they all only had the standard lipid panel that the doctor orders.
Sam, make sure your units for both TG and HDL are mg/dl. If in mmol/L you must convert.
Hi Peter,
Thank you for sharing your perspective. You started this blog with 2 explanations for “what makes fat cells fat” – and then followed up by stating, “One of these statements is true, the other is not”. I respectfully disagree that the second statement was false, and I will explain why.
I think the biggest mistake that many make when studying the etiology of obesity, which is a multi-faceted and multi-dynamic phenomenon, is to simplify it. While I agree with your explanation on insulin’s anabolic effects, the story doesn’t start and end with Insulin. There are other huge “players” involved as well, such as leptin, ghrelin, melanocortins, neuropeptide Y, dopamine, opioids, & amylin – and I’m sure many more that I’m forgetting. And what about the role that the lateral hypothalamus in our brain plays with regards to motivation, drive and reward?
We can’t pluck one hormone as the culprit because really, insulin is just part of the end game. When we look at the whole picture, or the whole “system” if you will, it is a brilliant concert of cascading hormones, neurotransmitters and psychoactive chemicals which respond in unison to a primal and evolutionary drive – and that is to store energy [survival].
When we look at the changes in our current food environment, and compare it to the natural food environment for which this brilliant system was designed to survive in, it’s obvious we don’t stand a chance. Our hedonic drive to over consume is purposeful and unchanged, and far more powerful than our homeostatic cues. For our children, obesity is the default, and without early intervention, education and big changes in policies, future generations will follow along the same collision course. The bottom line is, we have to change the food environment, our attitudes and our beliefs about food before we can save them.
I see your point, and I agree, but with one distinction. The second statement, “To a first approximation, obesity is the *result* of taking in more calories in the diet than are expended by the body’s energy-consuming activities,” implies that over-eating is the cause of obesity. I agree – while obesity always implies over-eating took place – the cause and effect are backwards. The hormonal drive to obesity is what creates an energy imbalance through a combination (it varies in everyone) of over-eating and under-expending. Insulin is the major hormone driving this process, though it does not act alone.
Deirdre – for me, it’s a bigger mistake to over-complicate obesity than simplify it. If I simplify it, I know what to do – quit eating carbs (and get my kids to as well). If I complicate it, and start hand-wringing over the food environment and the hedonic drive, I lose my way and start to think “I don’t stand a chance”. Nonsense. I read this brilliant blog from the perspective of an individual and a father. The bottom line is – save yourself first, save your kids next, save the world next. This is how we make progress . . .