Wednesday, January 30, 2013

Why Do We Eat? A Neurobiological Perspective. Part II

In the last post, I explained that eating behavior is determined by a variety of factors, including hunger and a number of others that I'll gradually explore as we make our way through the series.  These factors are recognized by specialized brain 'modules' and forwarded to a central action selection system in the mesolimbic area (the reward system), which determines if they are collectively sufficient cause for action.  If so, they're forwarded to brain systems that directly drive the physical movements involved in seeking and consuming food (motor systems).

The term 'homeostasis' is important in biology.  Homeostasis is a process that attempts to keep a particular factor within a certain stable range.  The thermostat in your house is an example of a homeostatic system.  It reacts to upward or downward changes in a manner that keeps temperature in a comfortable range.  The human body also contains a thermostat that keeps internal temperature close to 98.6 F.  Many things are homeostatically regulated by the body, and one of them is energy status (how much energy the body has available for use).  Homeostasis of large-scale processes in the body is typically regulated by the brain.

We can divide the factors that determine feeding behavior into two categories, homeostatic and non-homeostatic.  Homeostatic eating is when food intake is driven by a true energy need, as perceived by the brain.  For the most part, this is eating in response to hunger.  Non-homeostatic eating is when food intake is driven by factors other than energy need, such as palatability, habitual meal time, and food cues (e.g. you just walked by a vending machine full of Flamin' Hot Cheetos).

We can divide energy homeostasis into two sub-categories: 1) the system that regulates short-term, meal-to-meal calorie intake, and 2) the system that regulates fat mass, the long-term energy reserve of the human body.  In this post, I'll give an overview of the process that regulates energy homeostasis on a short-term, meal-to-meal basis.

The Satiety System (Short-Term Energy Homeostasis)


The stomach of an adult human has a capacity of 2-4 liters.  In practice, people rarely eat that volume of food.  In fact, most of us feel completely stuffed long before we've reached full stomach capacity.  Why?



The digestive tract uses a complex suite of signals to communicate with the brain, indicating the type and amount of food a person has eaten.  These signals are part of a system that attempts to keep meal size in the optimal range, not too large and not too small, either of which could be harmful to the organism.  We feel full before the stomach has reached capacity because or brain decides, based on input from the digestive tract, that we've had enough to eat.

One of the most important signals is stomach distension.  The stomach contains stretch receptors that inform the brain of the volume of food consumed.  These are indirectly connected to the hindbrain via spinal sensory nerves, as well as the vagus nerve (12).  The vagus nerve is intimately involved in two-way communication between the brain and the digestive system.

In addition, there are a variety of 'gastrointestinal peptides' that are secreted as food passes through the stomach and small intestine.  These respond both to the amount and the type of food consumed (i.e. protein, carbohydrate and fat).  These peptides play many roles in the body, but one important role is to communicate information to the brain about the food you just ate.  These peptide signals are transmitted to the brain both via the bloodstream, and via local actions on the vagus nerve in the intestine.  It's beyond the scope of this series to discuss these peptides individually, but if you want more information there are several good review papers available (345).

The hindbrain, and particularly the nucleus tractus solitarius (NTS) and area postrema (AP), is the primary brain site responsible for receiving these signals and promoting satiety and meal termination.  In a series of gruesome but fascinating experiments, researchers disconnected the forebrain and the hindbrain of rats, and demonstrated that the hindbrain contains all of the necessary hardware to terminate meals appropriately (6).  However, the hindbrain is not able to alter meal size in response to changes in long-term energy status, so that function (discussed in the next post) resides elsewhere in the brain (7).

What Food Properties Determine Satiety?

The satiety system is designed to make sure the body gets enough energy from an individual meal, without overdoing it.  The primary variable it responds to is calories.  The more calories you eat, the more full you feel.  Eating a huge bowl of lettuce with no dressing may have a lot of volume, but it won't really make you feel full, in the sense that you no longer have a desire to seek food.  Conversely, drinking a pint of heavy cream may not fill up your stomach as much as the aforementioned lettuce, but it will make you feel full!  The cream contains 900 calories, while the salad contains virtually none.

That being said, there are other factors besides calories that influence the satiety system.  In a very interesting paper titled "A Satiety Index of Common Foods", Susanne Holt and colleagues identified some of the factors besides calories that determine satiety (8).  It turns out, the degree of satiety per calorie consumed varies widely between foods.  The 'satiety index' refers to the amount of satiety a food produces, per calorie.  The two largest determinants of satiety were energy density (calories per gram of food) and palatability.  The lower the energy density, the higher the satiety index; conversely, the higher the palatability, the lower the satiety index.  Foods that were the highest in palatability were less than half as filling per calorie as those that were the lowest in palatability.  Palatability and energy density are related since energy density is one of the main determinants of palatability.

Many controlled studies have confirmed that palatability and energy density are major influences on meal-to-meal food intake (9, 10, 111213, 1415, 16).  In a study of people living their normal lives, John de Castro's group demonstrated that people eat 44 percent more calories at meals rated as highly palatable, compared to meals that are rated as average or low palatability (17)*:


Interestingly, according to his data, there was no difference between moderate and low palatability.  It may be that most of the benefit comes from avoiding highly palatable foods.

Energy density determines the volume (technically, weight) of a food you have to eat to obtain a certain number of calories.  Foods with a high energy density include oils/fats, cookies, candy, bacon, sausage, bread, and nuts.  Foods with a low energy density include fruit, vegetables, potatoes, eggs, oatmeal porridge, seafood, and lean meats.  Energy density has a major impact on the amount of calories consumed in a meal, and over many meals in a row, because the more volume you ingest per calorie (up to a point), the more full you will feel per calorie (18).  Researcher Barbara Rolls has a diet book based on energy density called The Ultimate Volumetrics Diet, which is one of the best diet books I've seen.  Controlled trials have shown that reducing energy density is an effective way to lose body fat and maintain fat loss in free-living people, at least compared to other single-factor weight loss interventions (19, 20, 21).

Returning to the satiety index paper, there are other factors that also determine satiety per calorie.  The more protein and fiber a food contains, the higher its satiety index.  The more fat a food contains, the lower its satiety index.  Other studies have shown that if you control for energy density and palatability, fat and carbohydrate are equally satiating however (22), so fat is only less satiating because it's palatable and occupies less volume than other nutrients (this can be avoided by eating fat in the context of a lower energy density meal, for example olive oil in a salad dressing).

Now let's bring this discussion from nutrients back to foods.  Foods with a low satiety index include pastries, cookies, white bread, added fats, candy, and ice cream.  Foods with a high satiety index include fruit, seafood, lean meats, potatoes, grain porridges, eggs and beans.  Sound familiar?  These are all ancestral foods eaten by healthy, lean cultures throughout the world.  Here's a nice quote from the discussion section of the satiety index paper (23):
These results therefore suggest that 'modern' Western diets which are based on highly palatable, low-fibre convenience foods are likely to be much less satiating than the diets of the past or those of less developed countries.
In addition, food variety increases food intake.  The more types of foods a person has available at a meal, the more total food he tends to eat.  I typically lump this phenomenon together with palatability/reward for simplicity, but technically it's a related property called sensory-specific satiety (24).

The Satiety System is Influenced by Other Systems in the Brain

As one might imagine from the discussion above, the satiety system is dampened (and/or overridden) by high palatability acting in other regions of the brain to be discussed later.  Both humans and animals will continue eating past the point of fullness if the food is palatable enough.  This is the 'dessert effect'.  No one eats 250 calories of plain boiled potato after a large meal, but bring on the chocolate cake!

The satiety system is also under the control of the systems that regulate long-term energy homeostasis (body fatness).  When body energy stores are low, the energy homeostasis system tells the satiety system to increase food intake, in an attempt to recover the missing energy (25).  I'll cover this again in a post on long-term energy homeostasis.

The Model

Here is the updated model of food intake regulation, including everything we've covered so far.  The ovals represent functional and anatomical brain modules, and the terms on the outside represent the factors that influence those modules:

In the next post, I'll describe the system that regulates fat mass, which is the long-term energy reserve of the body: the energy homeostasis system.


* His findings suggest that differences in palatability only account for a small percentage (4%) of the variability in meal sizes between meals.  This is due to two reasons.  First, most people in the study consistently selected high-palatability foods.  The palatability of meals was typically high, so the differences in palatability between meals were usually small and could not account for much of the total variability in size between meals.  Despite this, palatability remains one of the most influential factors he has identified in his research (26).  The second reason is that there are other important factors that influence meal size, like hunger, body size, and social environment.  The more factors impact food intake, the less any individual factor will contribute to total variability.

43 comments:

Bethany said...

I had great success losing weight (about 75 lbs) and keeping it off (4 years) by tuning into my physical hunger cues--making sure that all my eating was for homeostatic reasons (as described in part I). I used a hunger scale and started eating when I had certain sensations of hunger and stopped when I reached a certain sensation of satisfaction.

I had been doing a lot of non-homeostatic eating for emotional reasons, and this method helped me stop doing that and find more fitting solutions for those emotional hungers.

I've read a number of books promoting this method (books by Geneen Roth; a book called ,Intuitive Eating etc,). It was a lot of emotional work, but otherwise very simple (no restricted foods, or counted calories, etc).

Is this method an oversimplification that doesn't work reliably? Or is it just such so difficult in our culture of abundance that people dig for a formula on which to rely?

Stephan Guyenet said...

Hi Bethany,

Thanks for sharing your experience, that is remarkable. I do think the main reason people eat more energy than they need to stay lean, at least initially, is non-homeostatic factors. Reducing non-homeostatic eating will reduce calorie intake and cause a certain amount of fat loss for most people, and that has been supported by controlled trials where they ask overweight people only to eat when hungry and to stop eating when full.

If everyone ate simple food when hungry, to satiation, we could probably eat just about anything and remain lean. The problem is that for most people, once they gain fat it becomes 'fixed' by the energy homeostasis system, as I'll explain in the next post. At that point, fat loss becomes more difficult.

I think the method you described can be useful, but some will respond better than others because non-homeostatic eating doesn't fully explain why overweight people have a hard time losing fat.

Unknown said...

Does palatability cause weight gain because it makes you eat more, or because it causes your brain to signal the body to store more fat? Or both?

I am the classic skinny person with an enormous appetite. I commonly eat at buffets, in particular BBQ buffets where I would eat 3-4 lbs of meat, never trimming the fat. I am 5'11" 160 lbs. I am also 29, so it's not just a growing boy metabolism.

But, I never had a taste for modern 'foods' like sweets, refined breads, packaged foods. They just don't taste as good to me and make you feel gross. Give me a steak over that any day. Or even a salad. My only weakness is ice cream.

I imagine then food preference has some genetic factors as well.

Anyway, I look forward to your posts on how to reset body mass set points. I'm hoping it can help people who don't have the same eating ability as me.

Oh yes, I also don't feel full until my stomach is just about physically full. If there is food in front of me I can just keep eating. Of course though this is a choice, because I stop feeling hungry after eating in the patterns you describe.

And, sorry I am rambling, but I also wonder if other brain functions affect palatability and fat storage. Stress is known to cause weight gain, for good evolutionary reasons. But what about personality, daily interactions, fun levels, depression, etc?

Maybe I just don't get the same psychological kick as others do when they have things like sugar or chocolate.

Gretchen said...

Stephan, Nice summary of a lot of research. Two questions:

1. You say "most of the benefit comes from X." Benefit of what for whom? Benefit to a starving lion would be different from benefit to an overweight human wanting to lose weight.

2. A paper I read recently (Woods et al, Phil Trans R Soc B (2006) 361, 1219) differentiates "satiation" (meal termination) from "satiety" (time to next meal).

Is this differentiation not generally accepted?

Steven Hamley said...

The more satiating foods are higher in nutrient density than the less satiating foods.

Do you think nutrient density is another major factor in satiety?

Steven Hamley said...
This comment has been removed by the author.
Steven Hamley said...
This comment has been removed by the author.
Steven Hamley said...
This comment has been removed by the author.
ProfessorEd said...

Does the presence of a desert effect say we should start meals with our desert? It would seem that if we started with the cake and then ate the potatoes, we would stop eating when our body said we had met our caloric need?

spacediver said...

nice post Stephan. Again, thanks for all your passionate sharing of quality knowledge.

We often hear that one of the reason that animal fats are great to eat is because they "fill you up". Yet, according to the research you present (and your emphasis on lean meats), it may be the opposite. Now I'm not suggesting that this is a reason to abandon animal fats (although I suspect you have evolved your position on animal fats over the last couple years to recommend a more moderate cautious approach to them), but where does this notion that animal fats (or fats in general) fill you up effectively. Does this question interact with Gretchen's distinction between satiation and satiety?

Marwan Daar said...

sorry, meant to post the above comment using this name, not the spacediver account.

Anonymous said...

Thanks Stephan - another very interesting post. I'm looking forward to the next installment.

elbatrofmoc said...

I've just dicovered your blog and I'm amazed at how great it is. I'm sorry for putting this comment here, as it is not related to this particular post, but I couldn't find your email address anywhere. I have a question/request: if you ever find time, can you write what is you take on canola oil (CO)? I am really confused after sarching the web trying to find out for myself if I should avoid it or is it ok. For example, Nora Gedgaudas (the author of "Primal body, Primal mind") is a strong opponent of using CO, claiming that it contains unhealthy trans fats. I found similat claims made by Sally Fallon and Mary G. Enig in an article found on the Weston A Price Foundation Site (http://www.westonaprice.org/know-your-fats/the-great-con-ola). Dean Ornish, on the other hand, advocates the use of CO as the healthiest oil out there (http://asktaxmoms.com/wellness/heart/The-Great-Olive-Oil-Misconception-Dr-Ornish-answers-questions-about-the-hea.html).

Do you use canola? Why?

Thanks!

Gretchen said...

I've never been impressed with Holt's satiety index study, as they measured for only two hours, and one way fat seems to make you feel full longer is that it delays gastric emptying so it delays later hunger.

It's like the old cliche that Chinese food makes you feel stuffed but 30 minutes later you're hungry. The baked potato might have filled the people up in the short term, but 3 hours later they might have been hungrier than the people who ate more fat.

Also, were I offered a free buffet, I'd choose different things than I'd choose because of simple hunger. First, I'd take a lot of protein, because that's expensive and this was free. Second, I'd go for novelty, foods I'm too lazy to prepare myself. Not really a true-life situation.

Al Towers said...

Hi Stephan, great article. You do a great job at breaking down all this research.

My question is about your statement in the "What food properties determine satiety" section. You say "The primary variable it responds to is calories". However, you see all the time around the paleoblogosphere that, as humans, we don't have a "calorie receptor". Can this really be the variable of most importance if we can't actually detect it on physiologic basis? Or, is that saying wrong and there is actually a way we can directly measure energy intake through something like calories?

It would seem to me like much of this must depend on the types of energy we are consuming at each meal, as we can directly sense things like glucose and certain fatty acids that could be the variables we sense in order to regulate energy intake. Or else could micronutrient intake at each meal help to regulate our intake per meal?

Just looking for thoughts on this subject, as I have been thinking about the validity of such statements myself over the past couple of weeks. Thanks again!

Al

Sanjeev said...

> animal fats are great to eat is because they "fill you up"

maybe they only fill you up in the context of generally low activation of the reward systems

> we don't have a "calorie receptor" ... ... we can directly sense things like glucose and certain fatty acids that could be the variables we sense in order to regulate energy intake

long term regulation canNOT rely on short term sensing; IMHO the vast majority of the sensing in response to a meal is to direct dispensation of what just came in.

A small fraction of what's sensed will contribute to long-term regulation.

A calorie sensor makes zero sense for long term regulation - a long-term integrator's needed (in analog computer days this was done by capacitors).

> Or else could micronutrient intake at each meal help to regulate our intake per meal?

on the pages this page points to search for Kurt Harris's comments and one from Stephan

I have a couple there on this issue too- in your browser just search for Sanjeev

Sanjeev said...

> a long-term integrator's needed (in analog computer days this was done by capacitors).
____
just an analogy that'll make sense for 0.001% of the population.

Sanjeev said...

apologies - the link above has too many pages this was the page I meant to link to

the discussion starts
here but search because it continues a while later

Catt said...

I recently started buying (gluten free) white bread for post-race sandwiches (I run) since its easy and portable calories. And so easy they are!: those couple hundred calories of bread go in totally unperceived. You can eat one whole loaf and it will feel like nothing. I know because I recently did so on a plane trip after running a marathon: I was hungry and the flight food sucked -- but I can't say that loaf did anything for me -- I was still starved.

Anyway, after going a couple years eating primal-ish, noticing the waifishness of bread has been an eye-opeing experience.

Jane said...

@Sanjeev
Thanks for posting that link. Very interesting discussion between you, Stephan and Kurt.

Here's what Stephan says.

'I agree with you and Kurt that there is not much compelling evidence that people eat to fulfill nutritional needs, outside of a few basic things like energy and protein. Essentially the only human study I know of that supports that idea is the micronutrient supplementation study in Chinese participants where they lost some weight. ...The more I think about that study, the less sense it makes. There are probably hundreds of studies where participants were given multi-micronutrient supplements, and I'm pretty sure fat loss is not generally observed. We would have figured this out a long time ago. The Chinese study is a puzzling isolated finding.

'As you and Kurt both mentioned, you can come up with many examples of nutrient deficiency where obesity was not observed. The fact is, most Americans have a good micronutrient status compared to many other leaner populations around the world...'

The point here is that micronutrient 'status' may not be the relevant variable. What matters is whether you have unrepaired damage in for instance, your hypothalamus. Your micronutrient status might be quite good (as far as it can be measured) but you are still not repairing the damage that is causing your obesity. This could be because certain multi-purpose nutrients like Mg or Mn are too busy keeping your heart beating to undertake a major repair job on a lesion that will not kill you.

Mild chronic deficiencies are very different from the acute gross deficiencies studied in the lab. A micronutrient pill would not be expected to have much effect on obesity in the short term.

Ivan Nikolov said...

I spent years trying to figure out how to make palatable (and rewarding) food that also has the macro nutrient composition that's 'favorable' to food health and better performance. Examples: cookies, muffins, etc. that are high in proteins, low in NET carbs, no refined sugar, very low omega-6 and high Omega-9 and saturates instead, more fiber, etc.

But, only now, reading Stephan's blog I am starting to realize why I could never feel full enough after eating my macros from healthy but palatable (and rewarding) foods that I prepared.. and in the same time I could never finish a whole boiled potato without feeling extremely stuffed.

I guess, making - and eating - healthy but rewarding food is not the answer after all. It still makes you overeat. And, at the end of the day it all comes down to the total calories overconsumption that leads to energy storage (fat gain).

Stephan Guyenet said...

Hi Unknown,

Probably both, but the first explanation is supported by more evidence than the second. Genetics influences every factor related to food intake that has been measured so far.

Hi Gretchen,

1) Benefits the average person who has an elevated energy intake.

2) One can distinguish between satiation and satiety, but it's an artificial distinction not a biological one. Satiation is just satiety that occurs before the end of the meal. The physiological processes that underlie both are similar, and the distinction isn't necessary for the purposes of this series.

You make an interesting point about the satiating effects of fat possibly being more prolonged than that of carbohydrate. I looked into this, focusing specifically on longer studies that measured food intake over the course of a day. By and large, they agree with Holt et al. that fat is less satiating and people eat more calories over the course of a day when they eat a higher-fat breakfast.

However, these studies aren't the final word because they don't take into account how energy intake would (perhaps) change over time as people adjust to the new foods. There is still uncertainty about the long-term effects (compounded by the evidence from low-carb studies), but the short-term effects seem to be pretty consistent.

Hi Steven,

I haven't seen any evidence that it is.

Hi ProfessorEd,

Ha, good question. I don't know.

Stephan Guyenet said...

Hi Marwan,

Animal fats fill you up because 1) they're rich in calories, and 2) they're often associated with protein. If you take isolated animal fat with no protein, calorie-for-calorie it is not very filling as far as we currently know. But if you eat a lot of calories' worth, then yes it will fill you up.

Hi Al,

I don't know who said we don't have a calorie receptor, or who repeated it. It's sort of true in the sense that there's no such thing as a calorie-- it's an abstraction we use to assign potential energy values to food. But the body does have ways of detecting ingested energy as well as body energy stores in the form of fat. So I think that statement is only correct on a technicality, and in a broader sense it's misleading because the body can sense ingested calories in a rough manner.

Hi Catt and Ivan,

Thanks for chiming in, that's interesting.

Gretchen said...

@Jane This has nothing to do with obesity, but I find that every spring I crave bitter greens.

I was curious about this and Googled it and found it's common among many very different cultures to serve bitter greens in the spring. There must be a reason.

My favorite explanation was a Native American comment that it's to refresh your system after a winter spent cooped up with your relatives.

Gretchen said...

". . . they agree with Holt et al. that fat is less satiating and people eat more calories over the course of a day when they eat a higher-fat breakfast."

I think we need to distinguish higher fat in the context of a LC diet and simply adding fat to a "normal" diet.

If you normally have cereal and toast and milk for breakfast and you decide to put a lot of butter on your toast, and if you're not diabetic or prediabetic, then the butter simply adds calories and I'd agree with this statement.

But if you're diabetic or prediabetic, the roller-coaster BG levels on a high-carb diet contribute to hunger and a LC high-fat diet will be more satiating than a low-fat high-carb diet. Many people, including me, have found this to be true. I was ravenously hungry on a low-fat diet and no longer so when I ate more fat and less carbohydrate.

So we have to qualify generalizations by saying which groups they apply to.


Jane said...

Hi Gretchen
That's SO cool.

Grinch said...

"The more calories you eat, the more full you feel"

Then you say calorically dense foods have a lower satiety index.

Why the contradiction on calories? Either calories do or they don't reduce hunger / make you feel full.

Puddleg said...

This just in:
A calorie found not to be a calorie:
Atwater falsified by almonds in nut trial.

http://ajcn.nutrition.org/content/96/2/296.full

"Results: The energy content of almonds in the human diet was found to be 4.6 ± 0.8 kcal/g, which is equivalent to 129 kcal/28-g serving. This is significantly less than the energy density of 6.0–6.1 kcal/g as determined by the Atwater factors, which is equivalent to an energy content of 168–170 kcal/serving. The Atwater factors, when applied to almonds, resulted in a 32% overestimation of their measured energy content.

Conclusion: This study provides evidence for the inaccuracies of the Atwater factors for certain applications and provides a rigorous method for determining empirically the energy value of individual foods within the context of a mixed diet."

Now, this fact was determined by counting calories in faeces.

"Both fecal wet weight and dry weight increased with almond consumption, as did fecal fat, carbohydrate, fiber, protein, and energy (P < 0.0001). The amount of total dietary fiber and protein did not increase between the 42- and 84-g/d doses of almonds.
he fat digestibility of the total diet decreased by nearly 5% when 42 g almonds were incorporated into the daily diet and by nearly 10% when 84 g almonds were incorporated into the diet daily (P < 0.0001). Carbohydrate, fiber, and protein digestibility decreased between the control diet and the diet containing 84 g/d (P < 0.0001). Total carbohydrate digestibility of the 42-g/d diet decreased compared with the control diet and was intermediate to the 84-g/d diet."

Next time you count calories, try counting ALL the calories!

Stephan Guyenet said...

Hi Gretchen,

You said "But if you're diabetic or prediabetic, the roller-coaster BG levels on a high-carb diet contribute to hunger and a LC high-fat diet will be more satiating than a low-fat high-carb diet. Many people, including me, have found this to be true"

I have no doubt that what you say is true in your own situation, but I'd be interested to know how widespread that phenomenon is. The Internet is very good at creating selection bias because it concentrates people with shared experiences. Many obese people are pre-diabetic (in fact, the CDC estimates that 1/3 of the general population is pre-diabetic or more), and the satiety studies that included obese people didn't find anything different from those conducted in lean people as far as I can gather.

I know there are people who feel they react specifically to carbohydrate in any form, developing hunger and gaining weight, yet these effects aren't typically observed in the general lean or obese population in controlled studies. I would love to see that phenomenon investigated in more detail so we can see its prevalence and how well it holds up under controlled conditions. I'm genuinely curious.

Hi Dave,

If you have another look at the definition of "satiety index", you'll find that it means fullness per unit calorie. If you then think about what that means, you'll have the answer to your question.

Stephan Guyenet said...

Gretchen, here are a couple of references relevant to this discussion.

http://care.diabetesjournals.org/content/24/4/619.short
http://ajcn.nutrition.org/content/60/4/476.short

Jane said...

Great stuff Stephan.

Gretchen said...

Stephan, re your provided references. Not sure of the relevance of the first one, which showed that prediabetics can lose weight on a low-fat diet. We already know that. Calories do count.

I lost weight on a low-fat ADA diet. But I was constantly ravenously hungry, thinking of food all day and counting the minutes to the next meal. Most people can tolerate this short term but they finally give in to temptation and abandon the diet. You note that after 5 years the beneficial effects disappeared.

The LC diet allowed me to maintain the weight loss for about 10 years, and eventually lose a little more, without hunger or counting calories.

Re the second: Note that they say that results from other studies "varied widely." (I haven't read all the articles they reference.)

They also note that perceptions of what you've eaten may affect later food intake rather than physiological effects. But the dieter doesn't care if later hunger is physiological or psychological, just that it's lowered.

For research, of course, this is important.

The second study measured satiety at 30 minutes post preload. This says very little about real-life eating.

http://www.nutritionandmetabolism.com/content/2/1/31 shows that people with metabolic syndrome respond differently than others.

Gretchen said...

Stephan, I totally agree with you that Internet comments can be misleading. People who do well on standard medical care don't scour the Internet seeking alternatives.

Also, they make vague statements. I've often had people say "Fat makes my BG go up." Go up when? After 30 minutes? Six hours? 24 hours? On average over a month? Or they'll say some vitamin helped them, not mentioning that they took the vitamin along with 6 others and changed their diet at the same time.

And I have no idea what percentage of people with roller coaster BG levels would or would not feel hunger.

But I think we also can't totally reject Internet comments when they come from many people.

I've seen satiety studies that *did* show that obese people reacted differently. Unfortunately, my filing system is not good and I can't easily find them.

Clint said...

Here's an interesting study I found:

http://www.ncbi.nlm.nih.gov/pubmed/18198223

Basically, in healthy subjects, protein is the strongest macro for suppressing ghrelin overall, while carbohydrate is initially the most potent macro in suppressing ghrelin in the first three hours postprandial, but there is a strong rebound effect. Perhaps the rebound effect of carbohydrate intake on ghrelin suppression is the reason low carb diets work well -- no rebound. As you can see from the graph, although lipids weakly suppress ghrelin initially, and overall, there is no rebound effect. Combine the macro that most strongly suppressed ghrelin with the one that doesn't elicit a rebound, and voila, hunger control.

Anecdotal, but this is exactly what I've experienced as an obese-prone person. I can barely get through a couple of plain, boiled potatoes -- my stomach feels like it will explode if I eat anything more, despite the fact that there's obviously a lot more volume for more food (I can easily put away a large pizza, this is where food reward has a lot of merit), but I will be starving a few hours later and I will be back to eating whatever looks good (i.e. rewarding). However, on a low carb diet, I'm not hungry. My stomach isn't growling within a few hours, thus no impulse to eat. I want to eat less frequently, thus it's much easier not to succumb to the desire to eat highly rewarding foods on low carb.

Although the study I linked was done on healthy people, I wouldn't be surprised if the ghrelin rebound was more pronounced in people with insulin resistance. Or, at the very least, Ghrelin levels are already elevated for some reason, i.e. lots of overweight/obese people have sleep apnea which increases ghrelin, thus ghrelin levels end up even higher after the carb rebound.

If you're eating simple, non-calorically dense foods, the rebound isn't that big of a deal, particularly if you're lean and have lowish ghrelin levels in the first place. If you're wired to seek out rewarding foods when ghrelin levels rebound, it might be a better strategy just to avoid the rebound.

Ed said...

Something I think is interesting is that energy density is largely related to water content. Higher water content = lower energy density. Cooking oil & fat has the highest energy density, due to the complete lack of water. Avocado, while it has a "lot" of fat, is 73% water and thus has a lower energy density.

In fact, most natural foods seem to tend to cluster around 75-90% water. Vegetables for sure. Carrots are 94% water. Even meat -- according to nutrition data, a raw rib eye steak is 58% water.

Processed foods have much lower water content. I figure this has to do with shelf life, palatability, and transportation & storage costs.

Most baked goods have relatively low water content and thus end up with higher energy densities. Baked potato flesh is 75% water. Potato chips are about 1% water.

If we evolved eating foods with high water content, I wonder how that affected the evolution of our digestion, and what happens to that digestion with low water content foods.

I also wonder if your body can distinguish between high water foods (eg potato) and low-water foods consumed with copious water along side it (eg potato chips with a large class of water).

Just like "low carb" diet is really effectively high fat, "low energy density" seems to me to be effectively high water content.

Sue said...

The Volumetrics diet by Barbara Rolls, Ph.D., is about choosing foods that help control hunger and enhance satiety.

Jack S. said...

Hi Stephan,

I can fully well understand that perhaps eating for optimal health can be at odds with eating for weight loss, though only marginally if done properly. Specifically i'm referring to fat soluble vitamins like k2 found in butter and marrow. Is there anyway to eat these foods while pursuing weight loss without sabotaging my efforts or should it just be put on hold while losing weight and then reintroduced? Thanks so much for any advice, Stephan!

Stephan Guyenet said...

Good discussion everyone. I think it's clear that there are knowledge gaps in the literature.

Hi Ed,

Yes, energy density is mostly about water content. Barbara Rolls deals with this topic in the book I mentioned. When water is part of a food, like a fruit, meat or even a stew, it increases the satiety index. According to what Rolls relates in her book, drinking water along with a meal doesn't have the same effect because plain water gets rapidly shunted past the food in the stomach.

Hi Jack,

In general, if a person is overweight, and particularly if he is overweight with metabolic issues, fat loss will improve metabolic health almost regardless of what is eaten (within reason). I wouldn't worry too much about the fat-soluble vitamins during active fat loss. That being said, I think you could design a fat loss strategy that wouldn't sacrifice fat-soluble vitamins.

Gretchen said...

@Clint. I found that article you cited interesting because it provides a mechanism for the postprandial hunger some people report annecdotally after eating carbs.

For example this one: "I could eat a pound of pasta (with the 'regulation' fat free sauce) and an hour later be standing in the kitchen in front of the fridge, starving. Needing to eat something, anything. Unable to stop thinking about food. Feeling very very very hungry, even though my stomach was full."

Of course, we all know that the regulation of hunger is complex, and one hormone can't explain everything. It's more like an orchestra of interrelated signals, some hormonal and some neural.

Still, I think we need to pay attention to outliers whose results may be lost in formal studies in which the researchers report only averages.

Stephan Guyenet said...

Hi Clint,

You said "However, on a low carb diet, I'm not hungry. My stomach isn't growling within a few hours, thus no impulse to eat. I want to eat less frequently, thus it's much easier not to succumb to the desire to eat highly rewarding foods on low carb."

I ate a low-carb diet for about 6 months, and I did notice that I experienced hunger differently. I didn't feel like I had the same urgency to eat at prescribed times, and it was easier to skip meals and do 24 hr fasts. Now that I eat a high-carb diet, I doubt my total calorie intake is any different but I do feel hungry when I get home in the evening. And I don't just want any food, I want carbohydrate.

Carbohydrate is stored in limited amounts in the body, unlike fat, and I think when the body's adapted to burning carb and it starts to run out, it lets you know. This is my own speculation, not based on research that I know of, but it does seem to be consistent with the experience of many people including my own. If that's true, it does make sense that it would be an advantage to rely less on carbohydrate for fuel because the body wouldn't feel the same urgency to eat when carb stores are running low.

As far as I know, there's no known mechanism in the body to account for this effect (e.g. a glycogen sensor that tells the brain when levels are low), but I think it would be really cool to look for one... we might just find something...

Gretchen said...

If there were a glycogen sensor, then wouldn't one expect that people on LC diets, who are somewhat glycogen depleted, would feel hungrier? But the opposite happens.

Apparently the beta subunit of AMPK can act as a glycogen sensor. (BTW I'm not an incredible fount of obscure knowledge. I just Googled "glycogen sensor.")

Jane said...

Very interesting discussion. I'm not sure about this question of the body becoming adapted to burning carb so when it runs out it lets you know. Why shouldn't it just switch smoothly to fat burning and not bother you? That seems to be what mine does.

I do remember what carb addiction is like, I used to be addicted to sweet coffee. I ate just one meal a day, in the evening, because I was terrified of getting fat, and I'd drink sweet coffee all day to 'keep my blood sugar up'. Idiot. WTF did I think my liver was for.

mem said...

The NON-satiating effects of carbohydrate in lean young women.

http://www.ncbi.nlm.nih.gov/pubmed/21219950

One might project that if they luck out and have skinny genes, all will be well. If not, obesity in the future to come.