I’ve been having a repeated online discussion about fat and carbohydrate burning, and therefore decided to write a post that goes into it in detail.

There is a very common – perhaps even famous – graph in exercise physiology that looks like this:

The explanation that goes along with says that at low intensities, our bodies get most of their energy from burning fat, but that as we get to higher intensities, the percentage gradually changes, until at the top intensities, we get pretty much all of our energy from carbohydrate (CHO in the diagram). This explanation led to a lot of advice; there was a lot of advice that people should exercise at low intensities because that is where they would burn the most fat, and then contrary advice that said that while the fat burn was a smaller percentage at higher intensities, it was larger in absolute values and therefore to burn more fat you should work out at higher intensities.

What this discussions missed was something very simple…

This graph cannot be true for the vast majority of the population.

And it is wrong for a very simple reason; our bodies adapt their energy sources based on the diets we eat.

To illustrate why, let me discuss two different people:

Chris eats a very low fat diet of about 2000 calories per day; of these, she gets 10% from fat, 25% from protein, and 65% from carbohydrates.

Felicia eats a low carb diet of about 2000 calories per day, of these, she gets 65% from fat, 25% from protein, and 10% from carbohydrates.

Both have stable weight and body composition; they are neither gaining nor losing weight.

Let’s explore how the graph might relate to Chris, starting at the left side. If Chris is getting 80% of her calories from fat, then she needs 2000 * 0.8 = 1600 calories a day from fat. But she is only eating about 200 calories per day in fat and her body composition is stable, so there is no place that she could be getting an extra 1400 calories a day in fat. If she had that big of a fat deficit, she would be losing about 2.7 pounds a week (1400 * 7 = 9800 calories / 3600 is about 2.7 pounds). Further, if she is only getting 20% of her calories from carbs, that would be 400 calories per day, but she is eating 1300 calories per day, so she has an extra 900 calories per day of carbs. Those carbs need to go someplace, but the only big carb sink in our bodies is to store those calories as fat.

This graph simply cannot be true for Chris. Given that she has a stable weight and body composition, the energy she gets has to come in the same proportions as the food that she eats.

On to Felicia. The left side of the graph can work okay for her; she eats a lot of fat calories and those could provide the bulk of her energy at rest. It doesn’t work well for her at the right side; she is only eating about 200 calories per day of carbs, and let’s say that she goes on a one-hour run every day at moderate intensity. On that one-hour run, she burns around 500 calories, and half is 250 calories from carbs. But she is only eating 200 calories per day of carbs, and there are other tissues (the brain and red blood cells) that need some glucose to survive, so she doesn’t eat enough carbs to make this graph a reality.

In neither of these cases is the graph a realistic depiction of what is going on. So what really happens?

Well, here’s some research from back in 1997 where they played around with the amount of fat in the diet, and this is what they found:

“The results of the present study show that, in situations in which energy balance is reached, substrate oxidation can be adjusted to substrate intake. After 7 d(ays) on a high-fat diet, fat oxidation was, on average, equal to fat intake.”  (Discussion section, first paragraph)

In other words, the amount of fat that the subjects burned adjusted to be the same as the amount of fat the subjects ate.

Or this study. From the abstract:

Diet composition did not affect total daily energy expenditure but did affect daily nutrient oxidation by rapidly shifting substrate oxidation to more closely reflect the composition of the diet.

The same result as the other study. Our bodies adapt to burn the mixture of food that we provide it.

So, what does the graph really look like?

Here is some data gathered from testing a couple of athletes; the full article is here. Basically, you put them through a VO2max testing protocol, and you measure their fat and carbohydrate metabolism along the way.

Here is the first athlete:

This is athlete burns a small amount of fat even at very low intensities, and it only gets worse from there. Based on what I wrote about adaptation, what kind of diet is this athlete on?

Yes, it’s a high carb one; in fact, the athlete said that he had a sweet tooth and ate lots of sugar. Even at rest, he is burning a lot of carbs, and it only gets higher from there. As part of the study, this athlete modified his diet to reduce the amount of carbs and increase the amount of fat, by minimizing sugar and eating rice/break/pasta/potatoes only once or twice a week. After 10 weeks of training on the new diet, he looked like this:

He now gets a lot more energy from fat across the board, though he still gets a lot of calories from carbs. His body adapted to use the kind of diet that he now eats.

Here’s a second athlete:

What kind of diet was he on? Well, there’s no description of that in the linked article, but my guess is it’s a pretty standard athlete diet, and he doesn’t eat a lot of sugar. He is decent at burning fat, burning 40-50% over most intensities, but most of his energy still comes from carbs.

He switched his diet to take out grains and fruit (a break from the study goal, which was a low-carb ketogenic diet), and trained for 10 weeks. Here’s his second graph:

He now burns a ton of fat *everywhere*, regardless of intensity. Even at 4.5 watts per kilogram – a very high energy output – the fat and carbohydrate burn are about equal. The contrast of this graph to the first athlete’s “before” graph is huge.

Note that none of these graphs look like the one at the beginning of this post; what we see is that the poor fat burners stay poor fat burners as intensity rises, and the same for the good fat burners. We do see that at the high end fat burning goes down and carb burning goes up, but the overall graph shape doesn’t look like what we are told it should be.

So, where does the graph come from? Not being in the field and having not researched this thoroughly, I do not have a definitive answer, but following a reference led to this article which has this graph in it:

Digging a bit into the article, it is about how training effects the relative use of fats and carbs during exercise. I followed a few of the cited articles, but did not find the ones I wanted for free. I *did* find that the article cited Phinney’s early research on exercise and ketosis, which to me implies that the author knew about the adaptation due to different diets.

Anyway, I didn’t find any support for the idea that this graph was intended to be a representation of different sources of fuel *in general*.

So how do I burn more fat?

The whole concept of a fat burning zone and different intensities is not supported by studies. What *is* supported is that humans adapt to be good at burning the kind of fuel that they have available over time after a short adjustment period.

If you want to burn more fat *in general*, the answer is pretty simple; if you eat fewer carbohydrates – especially simple carbohydrates with high glycemic load – you will get better at burning fat during the day. Whether you lose weight will depend on your overall energy balance – you will still need to eat fewer calories than you burn to lose weight – but that puts you in a better position to be burning fat.

If you want to burn more fat during exercise – which can be a great way to burn fat – there are a few approaches:

1. You can change your base diet to eat fewer carbohydrates – especially the simple ones.
2. You can reduce the amount of carbohydrate that you take in during exercise. If your exercise is of a long enough duration, you will burn off enough carbohydrate to encourage your body to shift to better fat metabolism.
3. You can exercise fasted. You start in your best fat-burning state, so the exercise will have more of an effect on changing how your body generates energy.

You can mix and match these at will to see which one works the best for you.

There are a few caveats when it comes to exercise:

• To become a better fat burner, we need to train as if we were already a better fat burner – without as much available carbohydrate – and if we train for a significant duration, we can exhaust our carbohydrate stores and bonk. So, my advice is to have a source of carbs with you in case you start getting hungry or feeling really tired, and if this happens, to have a few of those carbs.

• When you start, your body is poor at fueling your exercise from fat. That means that you are going to have less power overall than you are used to. If you continue to push hard, you push your metabolism over to burning more carbs, losing the adaptation that you are trying to get. So, you will need to slow down to get the best adaptation, and remember that it’s going to take a little while (ie weeks) to become decently adapted.

Back in May of 2017, I started a journey to learn more about how the biochemistry of weight works; how our bodies metabolism carbohydrates, protein, and fats, and how that influences our weight.

When I started, I had a pretty simplistic view of how our bodies work.  I’m still only starting to have a functional understanding of the underlying biochemistry, but what I’ve found is that our bodies are very complex and fascinating systems shaped by the environment in which we evolved, and that if you understand the underlying biochemistry, things make more sense.

So… This is my attempt to present a simplified view of the underlying biochemistry. If you want to learn more, I recommend starting with a copy of “Mark’s Basic Medical Biochemistry”, but I’ll caution you that the section on carbohydrate metabolism alone is around 84 pages, so it’s pretty dense. If you want a slightly lighter approach, you can try “Mark’s Essentials of Medical Biochemistry”, which is a bit shorter. In either case, read them and then find some YouTube videos that talk about the subject, and iterate on this process a few times.

A few basic principles:

Our bodies are adaptive systems

If you’ve ever trained for an athletic event, you know that when you start training it’s really hard, but it gets easier over time as your body adapts, by making physical changes to your body. Our bodies also have an adaptive response to the kinds of foods that we eat, and that response can take time.

Our bodies try to be energy efficient

We evolved in an environment where food was not always plentiful, and our bodies generally attempt to be efficient and not waste any food calories.

Reactions to diets vary significantly between individuals

Genetics, age, sex, medical history, and activity level are all significant when considering how a specific person reacts to a specific diet. Or, to put it another way, one person may be able to remain healthy on a diet that would make somebody else very sick.

Carbs

I’m starting with carbs because their role is central to how your body reacts to the food that you eat. I’m going to talk about the different kinds of carbohydrates, how they are digested/absorbed, and how they are metabolized (used) by our bodies.

The nomenclature around carbs is a bit confusing, but they break into four broad classes:

Simple Sugars (aka “monosaccharides”)

The simple sugars are the ones that aren’t broken down by our digestive systems to something simpler. Most people are familiar with fructose and glucose, and there is also galactose, which I’ll discuss more in a bit. There are also some rarer sugars and sugar alcohols that I’m ignoring for now.

Compound Sugars (aka “Disaccharides”)

“Di” meaning “two”, these are sugars that are compounds of two simple sugars.

Sucrose (aka “table sugar”) is a compound of one molecule of glucose and one molecule of fructose

Lactose (aka “milk sugar”) is a compound of one molecule of glucose and one molecule of galactose

Maltose (aka “malt sugar”) is a compound of two molecules of glucose.

The much maligned High Fructose Corn Syrup is not a compound sugar, but a mixture of simple sugars. It comes in different fructose/glucose ratios, with the most common one being 55% fructose, so it’s similar to sucrose in its underlying composition and effect on the body.

Complex carbohydrates (aka “Oligosaccharides” and “Polysaccharides”)

All of the complex carbohydrates are chains of simple sugars hooked together (“oligo” means “a few”, and “poly” means “a lot”).

In this class we have ingredients like maltodextrin (a small chain of glucose molecules), or starches (big chains of glucose molecules).

Cooked starches are easily broken down into their simple sugars by our digestion. Some raw starches – also known as “resistant starches” – are poorly digested in the small intestine, and digested by bacteria in the large intestine and converted to short-chain fatty acids, which are absorbed. I would call them “starches that don’t act like more familiar starches”.

Undigestible carbohydrates (aka “fiber”)

Humans don’t have the digestive equipment to extract energy from carbohydrates like cellulose, so they just pass through our systems.

Carbohydrate absorption

With the exception of resistant starches and fiber, all of the carbohydrates are broken down to glucose, fructose, or galactose by the digestive system before they enter the bloodstream. The rate at which they enter the bloodstream depends upon a number of factors. A refined sugar is more accessible than one that is bound up within fiber in a food, so it is absorbed faster. A sugar that is by itself is more accessible than one that is mixed in with fat and protein. Different sugars have different transport mechanisms to get them from the digestive system into the blood stream and therefore have different rates of absorption.

But at the bottom, it’s glucose, fructose, or galactose. The sugars that you get from eating an apple aren’t chemically any different from those you get from a can of Coke. The two foods – if you can call a can of coke “food” – differ in the total amount of carbs, the ratios of different types of carbs, the rate at which the carbs are absorbed, and the non-sugar ingredients, but they both end up as simple sugars in your system.

Complex carbohydrates also aren’t inherently different than simple sugars; they may not taste sweet, but they end up as simple sugars (typically glucose) when they are absorbed.

Carbohydrate Metabolism

Metabolism is all about what happens to the nutrient after it gets into our bloodstream. It turns out that the different simple sugars are processed very differently.

I’m going to start with glucose metabolism.

Glucose metabolism

The amount of glucose in our blood – our blood glucose (or blood sugar) level – is one of the most important values for us as living organisms. Too little (aka “hypoglycemia”), and we get hungry, headachy, sleepy, confused, or worse. Too much (aka “hyperglycemia”), and we have other issues; a lot of our body’s systems do not work well with too little or too much blood glucose.

Our bodies therefore have a system to keep blood sugar constant, and this is a high priority system; it is fair to say that “keep blood glucose within range” is Job One for our regulatory systems.

Let’s say we ate a small plain bagel or drank a can of Coke. That’s going to send somewhere around 30 grams of glucose into our system. That is over 7 times the amount of glucose we normally have in our bloodstream, so we need a place to put the excess glucose, and biochemical system to pull the glucose out of the blood and put it in that storage place.

Storing Glucose

There are two different methods of storing glucose in our bodies.

We can store it in our liver or our muscles as glycogen, a compound that is very close to glucose chemically. Glycogen is a little like a starch; it’s just a bunch of glucose molecules surrounding a protein known as glycogenin, and it’s quick and easy to get the glucose back out. The storage for glycogen is fairly limited; the muscles can store about 400 grams, and the liver can store about 100 grams, or 500 grams / 2000 calories total. If you have ever “hit the wall” or “bonked” during extended exercise, you’ve experienced what happens when you run out of liver glycogen.

If your liver and muscle glycogen stores are already full, the excess glucose out of the blood needs to go somewhere else.

There is only one other place for storing the excess energy that the glucose represents, and that is our fat stores. We tend to think of sugars and fats to be very different things – one is white and crystalline, and the other is oily or greasy – but they are both molecules made from carbon, oxygen, and hydrogen. The liver and fat cells can take in the excess glucose and convert it to fat, which is stored in our fat cells.

The biochemistry to do this is built on a hormone that we have all heard of, insulin. When blood sugar is high, the pancreas releases insulin, which has 3 main effects our system:

1. The muscles and liver increase their absorption of glucose to store as glycogen (assuming there is room to store it).
2. The body turns off fat burning, so that current energy use will help use up glucose.
3. The fat cells increase their absorption of glucose to store as fat.

The speed at which glucose can be stored depends on where it is being stored; storing it as glycogen is quick, while storing it as fat is slow. Here’s an interesting chart from a study:

The subjects in this study had slept overnight, which had burned some of the glycogen in their stores. They then gave them one of three breakfasts:

• A can of Coke
• A service of instant oatmeal
• 2 poached eggs.

Both the Coke and the oatmeal have a lot of carbs, and the eggs have almost none.

Then, they fed them a *second* breakfast of oatmeal – a lot of carbs – and measured their blood sugar over time.

What they found was that if the first breakfast was eggs, the second breakfast had little effect on their blood glucose, because all of those carbs went straight back into filling up the glycogen stores. If the first breakfast had carbs, they had mostly refilled the glycogen stores already, and it therefore took quite a while to get the blood sugar back to normal. Which means that much of the second breakfast was stored as fat.

These charts showed what happened with a healthy person – one that we would call “insulin sensitive”. The blood glucose gets back to normal.

For some people, that doesn’t happen. There are different theories as to why it doesn’t happen; one is that the glycogen stores just get full, one is that the fat and liver cells can’t do the conversion to fat well, and there are others. Regardless, their glucose-absorbing cells become resistant to the effects of insulin, which we call “insulin resistance”. The first reaction of our bodies is to try harder by using more insulin, but this is an arms race that can eventually lead to problems in insulin production. Many people progress from insulin resistance to type II diabetes and metabolic syndrome.

I want to stress here that insulin resistance and type II diabetes are about the abilities of our bodies to regulate blood glucose levels, they are not about weight. It is true that people who carry a lot of extra weight are more likely to have blood glucose issues, but there are obese people who have good blood glucose control, and – perhaps more surprisingly – there are thin or normal weight people who have insulin or type II diabetes. They have bodies that are not good at converting excess glucose to fat.

That is why testing for blood glucose over time is important. Unfortunately, testing for blood glucose every day is intrusive and not something you can use with the general population, so for a long time all that was used was the blood glucose measured at one time, which is not a good predictor.

Then a weird bit of biochemistry came to the rescue. It turns out that in our red blood cells, the hemoglobin that transports oxygen can become glycated – it can have a glucose molecule attached to it. The chance of that happening depends on the average amount of glucose in the blood; if you have a low average blood glucose, few of your hemoglobin molecules will be glycated, while if you have high average glucose, more will be glycated.

And, it turns out that when red blood cells die, we can look at the hemoglobin and see how much was glycated, and therefore have a good idea the overall glucose levels, as a weighted average for the past few months. 

The test to do this generates a value known as HbA1c, or simply A1c, and it’s the prime diagnostic measure for insulin resistance and type II diabetes.

Low blood glucose

Thankfully, this is a lot simpler.

The reaction to low blood glucose is somewhat the opposite to high blood glucose. It is mediated by a hormone released by the pancreas known as glucagon, which has roughly the opposite effects as insulin:

The glycogen stored in the liver is converted back into glucose and released into the bloodstream (the glycogen in muscles can only be used locally; it cannot be released back into the bloodstream).

The body encourages the release and burning of fatty acids rather than carbohydrates.

If that is enough to raise the blood glucose, that everybody is happy. If the low blood glucose continues for longer – say for a few days – the body switches over to an alternate fueling approach known as “ketosis”, which involves the following changes:

The liver starts producing what are known as “ketone bodies”, which you can think of a glucose substitute for some of the body tissues; the brain can largely use ketone bodies for fuel, as can some muscles.

The muscle switch to burning more fat and less carbohydrate to produce energy. Like any exercise adaptation, this occurs over time.

The liver starts producing glucose from whatever it has lying around; it might be glycerol from fat metabolism or excess protein, or both. This is known as “gluconeogenesis”.

Together the switch to ketone bodies and the creation of new glucose is enough for the body to function normally without eating carbs. Adapting to this takes a couple of weeks for most individuals, though adapting to burn more fat during exercise takes much longer.

Ketosis is a response whenever carbs are severely limited, whether it be a fast or a ketogenic diet. Ketosis is not an all-or-nothing response; a person on a relatively low-carb diet might be in ketosis overnight when their carb reserves get low and then switch out of it during the day when carbs are more available.

And that is the story on glucose.

Galactose metabolism

Galactose metabolism is pretty much like glucose metabolism, except a trip to the liver is required to break the galactose molecule apart into two glucose molecules, which are either used by the liver or released into the bloodstream.

Fructose metabolism

Our body cannot metabolize fructose directly, so first it has to take a trip to the liver to be converted to something that is more useful.

It ends up as one of three things:

Glucose, which is stored in the liver, converted to fat by the liver, or released into the blood stream.

Lactate, which can be used by other tissues

Triglycerides (ie fat).

There is considerable discussion around what the proportions are between those three products and what controls those proportions.

Some researchers theorize that in some cases, the triglyceride pathway is especially active and that leads to the accumulation of fat in the liver and a condition known as Non Alcoholic Fatty Liver Disease (NAFLD). My limited understanding says that we don’t have definitive evidence on this, but we do know that alcohol is metabolized into fat in the liver and the accumulation of fat causes alcoholic fatty liver disease, and since fructose can also be metabolized into fat in the liver, it’s an interesting hypothesis.

Futures

That’s all for carbohydrates.

Upcoming, we have fat and protein to talk about, and probably a discussion around energy partitioning, which is how our bodies decide what fuel to burn.