The body stores far more energy as fat than as carbohydrate. If you can change your metabolism so that you rely more on fat to fuel exercise, then you should reduce the need to consume calories during exercise, spare the carbohydrate that is stored in muscle and liver, and improve your endurance capacity. Training when carbohydrate availability is low may force metabolism toward a greater reliance on fat to fuel exercise. Two approaches to creating a state of low carbohydrate availability are often described simultaneously, as if they are simply different but equivalent ways of going about it:
a) exercising after an overnight fast - stores of carbohydrates are used while you sleep and not replaced before a workout, or
b) exercising after a prior heavy exercise bout - carbohydrate is used up during an exercise bout and not replaced before a second workout.
Despite being presented as equivalent options, the two approaches are different in their bioenergetics and will lead to different adaptations. This blog post is meant to clarify the physiology of the two approaches.
In 1853, Claude Bernard discovered that the liver secretes our body’s main carbohydrate, glucose, by observing that the blood flowing out of the liver of fasted or meat-fed animals contained glucose while the blood flowing into it did not (or very little).
When a cell takes in glucose from the blood, it is modified (phosphorylated) in the process. This effectively locks the molecule into the cell because the modified form cannot be secreted back out of the cell. Only the cells of the liver make the enzyme glycogen 6-phosphatase, which modifies glucose back to it’s original form so that it can be secreted. Therefore, only the liver can secrete glucose.
The liver is our highest concentration of glucose. But liver doesn’t taste sweet at all. That’s because it is stored as long strands of glucose, called glycogen and glycogen doesn’t activate sweet receptors in the mouth. Liver contains 100-120 g of glucose, as glycogen. Skeletal muscle contains much more glycogen overall (400-500 g) but we have much more skeletal muscle than liver. So, ounce-for-ounce, skeletal muscle contains roughly 1/5th the concentration of glycogen when compared to the liver. The net calorie storage, at 4 kilocalories(a nutritional calorie is a kilocalorie) per gram of glucose, is about 500 kilocalories in the liver and about 1500 kilocalories in skeletal muscle.
At night, while we sleep, the liver secretes glucose to maintain blood glucose levels and supply the body (especially the brain) with much needed carbohydrate. By morning, 60-80% of the liver glycogen has been secreted and used by cells throughout the body. However, muscle glycogen levels remain high throughout the night because it can’t leave the cells and muscle isn’t consuming it.
At the start of exercise, the liver begins breaking down glycogen to maintain blood glucose levels as the working muscle consumes it. However, muscle preferentially consumes it’s own glycogen stores. During exercise, a person ‘hits the wall’ when they run so low in liver glycogen that blood glucose levels drop. Muscle glycogen levels will also be very low because the muscle has been actively consuming it.
In the running community, the concept of exercising when the body is low in available carbohydrate (i.e., glycogen/glucose) has been promoted as a means to encourage adaptations that favor the use of fats for fuel. In theory, the development of fat utilization pathways will ultimately lead to a sparing of the limited glycogen stores for energy and reduce the need to eat as many extra calories. While some recent research has shown that exercising when carbohydrate availability is low, following approach ‘b’, does enhance fat utilization, there is no evidence yet that this leads to longterm improvements in endurance. But, this field of research is young and there remains much hope that periodizing carbohydrate consumption during exercise training can lead to improved endurance performance.
With that goal in mind, there are two methods of training low* with respect to carbohydrate (i.e., glycogen/glucose) stores. An athlete may either train a) first thing in the morning without eating breakfast or b) after depleting stores through a prior workout. In the later case, an athlete would perform a high intensity workout, then not eat carbohydrate until after a second workout; the two workouts would be performed approximately 10-12 hours apart (e.g., early morning/late evening or late evening/early morning).
It’s important to appreciate that these two methods are very different with respect to their bioenergetics. Let’s evaluate each in turn.
a) Running first thing in the morning, after an overnight fast and without breakfast, is a hepatic (liver) challenge because only the liver becomes depleted (or very low) in glycogen stores over night. In this case, active skeletal muscle will rely on it’s own stores for the glucose it needs during the morning run (which it prefers anyway); it will run out of it’s own stored glycogen a little sooner than if the liver was supplying it with additional glucose. But, the muscle is energetically fine. The stress in this scenario is placed on the liver, which does not have the reserves to respond to the energetic needs of the body.
b) Engaging in a high intensity workout (or a long run) to deplete muscle of glycogen, followed by a null-carbohydrate diet for 10-12 hours, will put greater stress on muscle energy metabolism. Working muscle will have to rely much more on blood glucose (supplied by the liver) and on it’s own fat reserves. During the interim, between the depleting bout and the subsequent workout, the liver will secrete glucose to maintain blood glucose levels. When you start the second workout, both muscle and liver will be low in glycogen reserves, you’ll feel fatigued and weak (at least, not strong), and muscle will be forced to rely much more on fat for fueling your run.
From this evaluation, it may seem that depleting muscle glycogen through prior exercise (approach ‘b’) is the way to go if you want to really stress muscle fat utilization and promote subsequent adaptations. But, be careful and considerate of the fact that the first workout places a substantial stress on muscles. It’s a depleting workout after all. Your immune system infiltrates skeletal muscle for repair and adaptation. By depleting carbohydrate and not replacing it, these repair and adaption systems are unable to function properly. The second bout of exercise, where the goal is to stress fat utilization, comes with the added consideration that the muscle is already functionally compromised. So, the second bout should be of low intensity and carbohydrate feeding should begin immediately upon completion of the bout. Moreover, this should not be performed very often or symptoms of overstress/underrecover will ensue…probably no more than once every 10-14 days for most runners.
We must also appreciate that adaptations to exercise training include an increase in the rate of fat utilization at the same absolute and the same relative workload. You’re already improving fat utilization just be training. So, how do exercise training and dietary manipulation compare in efficacy? Exercise training is the stronger stimulus. Even in a fully fat-adapted athlete, who has adapted to a radical dietary manipulation comprising very high fat and very low carbohydrate, the increase in fat use during exercise is not extraordinary…perhaps as much as accomplished through training alone. Because simply training with low carbohydrate availability every 10-14 days is not nearly as extreme as adopting a fat-adaptation diet, you will probably achieve an even smaller effect.
Overall, exercising while carbohydrate stores are low may be something you choose to dabble with but it’s unlikely that doing so will have a large impact. And, you may compromise the efficacy of your training while you figure out what may work for you because exercising while carbohydrate depleted is hard and the quality of a workout (intensity, form, total work) usually suffers. If your goal is to get on the podium and you are routinely coming in 5th place, then this may be something you want to explore. But, for the majority of ultramarathon runners, it’s my opinion that you will be better off focusing on the core principles of training. Ultimately, unless you have access to metabolic assessment instrumentation (like in a performance laboratory), the outcomes of low carbohydrate training will be along the lines of “I feel like I don’t have to eat as much on long runs” or “I feel like my energy levels hold on for longer”. These are subjective at best. The placebo effect alone, based on a desire for the dietary manipulations to work, is far greater than those subjective perceptions.
This is an area of research where I hypothesize that we’ll see large inter-athlete variability as more studies are completed. I suspect we’ll also see, even if the periodization of diet and exercise is optimized for each subject in a study, that there will be high- and low-responders.
* It’s potentially confusing to use the phrase ‘training low’ because ‘train low, live high’ was coined by Dr. Ben Levine and colleagues to explain the best approach to training and adapting to altitude. One should train at lower altitude (train low) so that workouts can be performed to their fullest without the detrimental affects of altitude while the athlete spends the rest of the day living at higher altitude (live high) to gain the beneficial adaptations caused by spending a minimum of 16 hours (preferably >20 hours) per day at high elevation. Therefore, use of ‘train low’ in the realm of carbohydrate availability may lead to confusion.