Question: I’ve seen the idea kicked around that a cyclical ketogenic diet might be beneficial for improving endurance performance. Do you think this is possible and, if so, what’s the best way of going about it.
Answer: This is one of those long-standing questions that’s been on my mind for years and the short answer is ‘maybe’. Some of it depends on what you’re talking about which will make more sense when you get to the end of my answer.
The idea of fat adapting endurance athletes has been around for years, I presented most of the early data in my first book The Ketogenic Diet. Summing that research up, the general consensus was this:
- In the short-term (a few days to about a week), low-carb diets tend to destroy performance.
- With sufficient adaptation (usually 3+ weeks), there may be performance benefits.
But even #2 is a bit questionable. In the most often cited study (by Phinney), the results were skewed by one of the five cyclists who got massive improvements in endurance, the other four stayed about the same. So although the average performance improved, most of the subjects showed no improvement.
The results also depend on how performance is actually tested. If endurance was tested at lower intensities, performance sometimes improved. When researchers tested high intensity activity (where glycogen is required for optimal performance), performance was invariably worse. The conclusion was simple, no amount of adaptation to low-carbohydrate/ketogenic diets would benefit high-performance activities.
This led to the idea that perhaps fat adapting was most beneficial for ultra-endurance types (who typically go at a lower intensity for longer periods) and less valuable for endurance sports where high intensity efforts were required (think of a cyclist who may have to cover a breakaway or make a final sprint).
This also led to a second, more recent data set which was an examination of short-term fat adaptation (5 days) followed by a 1 day carb-load (looks just like a cyclical ketogenic diet). The idea was to get some adaptations in fat utilization AND refill muscle glycogen to support high intensity activity.
Research had shown that many of the adaptations to low-carbohydrate diets hang around for a while (at least a week) after carbohydrates are re-introduced so it seemed possible that the benefits of low-carbs could be generated while refilling muscle glycogen to sustain high-intensity performance.
And the data was mixed as hell.
On average, performance didn’t generally improve which led to the conclusion that the approach was still invalid. However, in the studies that showed individual data, a pattern emerged that I found interesting. The subjects who got the best performance on carb-based diets showed the biggest decrease on the fat adaptation diet. But subjects who did relatively poorer on high-carbs usually got a performance boost on the with fat adaptation.
This is very consistent with my experience with dieters (as discussed in the article series Comparing the Diets, on this site). Individuals who do best on carb-based diets often do very poorly on low-carb diets and folks who do poorly on high-carb diets often thrive on low-carb diets.
Related to this, there is data suggesting that people differ in how well they adapt to increases in dietary fat, some seem to do a better job of increasing fat oxidation compared to others and I suspect this explains some of the difference here. The subjects whose bodies ‘run better’ on fat may be the ones getting performance improvements from fat adaptations.
However there was another problem with those studies, they invariably tested exercise in a way that wasn’t exactly analogous to typical racing situations so applying the data to real-world performance was difficult.
In a very recent study, cyclists followed 5 days of fat adaptation with 1 day of carb-loading and then performed a time trial. The difference was that the time trial included several short sprints (this was meant to more closely mimic a true competition). Power output during the sprints was lower after the fat adaptation even after the 1 day carb-load and despite the refilling of muscular glycogen.
It appeared that the body had temporarily lost the ability to generate energy from carbohydrates quickly enough to sustain optimal power outputs during the sprints. So for the most part, it doesn’t look like a cyclical keto diet can really enhance performance, at least not those requiring high intensity bursts in the short-term.
However, there’s another way that cyclical ketogenic diets might be beneficial for endurance athletes but it has more to do with training than competition. Increasing amounts of data are finding that training and diet interact in terms of the adaptations seen. What you eat on a day to day basis as well as around training affects what sorts of overall adaptations are seen.
For example, early data had shown that several days on low carbs with endurance training increases gene expression for pyruvate dehydrogenase (PDH) which is a key enzyme for energy production in skeletal muscle.
More recent data has examined a molecular energy sensor called AMPk (adenosine monophosphate kinase) which turns out to play a crucial role in endurance adaptations (you can read more in AMPk: Master Metabolic Regulator). AMPk is activated more easily when glycogen is depleted which might mean higher levels of gene expression for the adaptations that endurance athletes are looking for.
Based on a variety of data, it looks like training under conditions of glycogen depletion may enhance endurance adaptations and a theory of training with low glycogen and competing with high has been advanced by some researchers in the field.
The problem is that gene expression and protein synthesis aren’t the same thing and protein synthesis tends not to proceed very well when cellular energy state is low (as would occur when muscle glycogen is depleted and AMPk activity is high).
Which brings us back to cyclical ketogenic diets. What if someone combined glycogen depletion with a large volume of low- to moderate-intensity endurance training; in theory this should generate a great deal of gene expression for stuff important to endurance athletes. However, this wouldn’t be the optimal situation for actual protein synthesis.
What if now that person swung into high-calorie/carb-loading phase with a reduction in training volume (and perhaps an increase in intensity) so that there would now be sufficient energy to synthesize mitochondrial and energetic proteins?
Would this help drive adaptations further than training under normal carbohydrate intake conditions? I don’t know but that’s sort of the idea I’ve been kicking around anyhow.
Basically, although it doesn’t look like a cyclical ketogenic diet is optimal for overall competition results it might be used during a base training phase in an attempt to drive adaptations higher than they’d otherwise go.
During the competition phase, higher intakes of carbs make the most sense based on the most current data, at least for sports which require bouts of high intensity performance during the event (if you’re an ultra runner or something, fat adaptation may work just fine).
Finally, it is at least worth noting that the current kings of endurance running (the Kenyans) are known for exceedingly high daily carbohydrate intakes (70% of total calories) although they are also known for doing consistent morning runs fasted as well as multiple training sessions per day. So they may have actually stumbled on a training pattern where they get the benefits of everything I’ve talked about, some training done fasted/under glycogen depleted conditions with sufficient carbs/calories to sustain optimal rates of protein synthesis.
Of course, since most can’t train three times per day, it’s debatable how relevant that is to the average trainee. A cyclical low-carb approach might be a more practical way of achieve some of what I’ve talked about and I’d love to hear from anybody who’s tried it.