Wednesday, July 6, 2011

Hitting the Wall, Part 3

In part 1 I showed what hitting the wall looks like for me despite carbo-loading.

In part 2 I was curious if a low carb-high fat is optimal for my endurance running.

Dr. Stephen Phinney's low carb exercise research could shed some light. He summarized his low carb exercise studies in 2004 article in Nutrition and Metabolism. It was based on two of his previous studies in 1980 article of weight loss and exercise and 1983 article of professional cyclists. Below I give a summary of Dr. Phinney's perspective on low carb endurance performance and my interpretation of his studies.

Clinical research evaluating the effects of physical performance by high carbohydrate versus low carbohydrate diets has historically supported a high carbohydrate diet. Danish researchers Christensen and Hansen in the 1930s measured endurance time to exhaustion on a stationary bicycle after one week on each diet. The endurance time on a low carb diet was 81 minutes compared to 206 minutes on a high carb diet. Swedish researcher Jonas Bergstrom in the 1960s assessed fuel stores in muscles. This research led to the strategy of carbohydrate loading to replenish the limited fuel of muscle glycogen for high intensity exercise.

It's an easy assumption that grain-based carbs are nutritionally superior to meat-fish-based fat and protein since agriculture has competitively dominated over the hunter-gatherer cultures. The development of modern society was supported largely by the shift to agriculture with the ability to grow and store grains, build permanent dwellings and communities and develop infrastructure for transport and trade. And yet, those hunter-gatherer cultures had physical stamina without carbo-loading. What gives?

Dr. Phinney conducted his clinical trials on low carb exercise based on the lessons from explorers living among one of the last of the hunter cultures, the Inuit, more than a century ago.

First, Frederick Schwatka's expedition in 1879-80 covered over 3000 miles in the Artic on foot. After a month's supply of food, the expedition's only source of food was hunting and fishing. Schwatka said that initially a diet of only reindeer meet seemed inadequate as he felt weak to perform long journeys, but those feelings passed away in about 2-3 weeks. At the end of the journey, he walked the last 65 miles in under 48 hours to make a scheduled rendezvous with a whaling ship. Dr. Phinney refers to this change as keto-adaption.

Second, Vilhjalmur Stefansson, a Harvard trained anthropologist, lived and traveled among the Inuit for a decade in the early 1900s while speaking their language and living on their diet without impairment. Stefansson's writings on his adaptation to the Inuit culture and functioning well on only the products of hunting and fishing caused enough controversy that Stefansson agreed to be scientifically observed along with a collegue at Bellevue hospital for the calendar year of 1929 while reproducing his Inuit diet of 80-85% fat and 15-20% protein. Dr. Eugene Dubois supervised the experiment and concluded that both subjects survived the 12 months in apparent good health with no signs of deficiency diseases.

Dr. Phinney recognized that previous clinical trials on low carb diets were only 1 week long, not enough time to allow for adaptation. His first study on the exercise effects of a low calorie ketogenic diet was 6 weeks long on overweight subjects. He evaluated treadmill performance by measuring peak aerobic power (VO2 max) and endurance time to exhaustion at 75% VO2 max. Peak aerobic power did not decline in 6 weeks. Endurance time to exhaustion declined in the first week of the ketogenic diet from 168 to 130 minutes, trending in the same manner as previous studies, but increased to 249 minutes after 6 weeks. It would seem that keto-adaption greatly improved endurance, but the results were confounded by the fact that the average test subject lost about 10 Kg of body weight. Despite wearing loaded backpacks to equal their weight loss, the subjects had greatly improved exercise efficiency as measured by oxygen consumption decrease.

A second clinical trial was conducted by Dr. Phinney on trained professional cyclists to remove the confounding effect of body weight loss and improved exercise efficiency. The cyclists were put on a ketogenic diet for 5 weeks similar to the Bellevue study with the intention that the subjects would be in ketosis without weight loss. On average the subjects lost less than 1 Kg of body weight in the first week on the diet and then stabilized for the remainder of the study. Endurance time was measured at 65% of VO2 max. There was essentially no change in peak aerobic power, exercise efficiency no endurance time.

Both studies show that a person's physical performance is fully restored after a few weeks to adapt to a low carbohydrate diet. Performance levels were measured around 70% of VO2 max. According to David Swain (1994), 70% of VO2 max correlates approximately to 80% maximum heart rate.

There were two other important variables that Dr. Phinney took note of based on Inuit culture. First, he made sure that his subjects had adequate amounts of sodium and potassium supplements. The Inuit people consumed much of their meat in a soup or broth made from coastal ice, a brackish source of water. It was also common for Inuit to add caribou blood to their soup, a rich source of sodium. Second, the protein ratio of the diets was about 1.5 g per Kg of body weight. Too far below this target will result in muscle loss and reduction in VO2 max. Too far above this target will result in suppression of ketogenesis. In Stefansson's writings he noted that the Inuit were careful to reserve higher fat portions of meat for themselves and feed lean meat to the dogs.

Despite the apparent success of Dr. Phinney's studies, I am still nagged by the effect of exercise intensity on the bodies ability to burn fuel from fat. In the middle of the 2004 paper he states:

"...high carbohydrate diets might be more effective in short-term tests of high-intensity exercise..."
And Dr. Phinney's concluding statement includes a caveat:

"...anaerobic (ie, weight lifting or sprint) performance is limited by the low muscle glycogen levels induced by a ketogenic diet, and this would strongly discourage its use under most conditions of competitive athletics."
This raises a question for me. What is the optimal nutrition for an amateur runner (thinking of myself here) that wants to competitively run a long distance race, such as a full marathon with a goal to qualify for Boston? Is this amateur runner performing at an intensity far and above the subjects in these studies or hunter-gatherer cultures? If so, is low-carb-high-fat an inferior fuel source for such an event?

In my 2004 Chicago marathon charted in part 1, that was the only time I qualified for Boston. I have no idea what my exercise VO2 levels were, or even my heart rate. I was running at a pace that I perceived throughout at least the first half that I could maintain until the end. I wasn't breathing hard. It wasn't until around mile 17 or 18 that I could feel the energy draining from my body, which became progressively worse until I could barely stand to run by mile 23. All else being equal, could I have avoided hitting the wall if I was keto-adapted, or was the pace too fast to keep muscles from glycogen depletion despite an ability to better utilize fat as fuel?

What have other runner's experienced? Kent of ran the Green Bay Marathon in 2011 without hitting the wall. His pace is shown in the chart below. He set a PR of 4:09. Not knowing Kent, I can only assume with a moniker like "Atkins Diet Geek", he is marathon training and racing on low-carb-high-fat nutrition. His pace appears to be at an intensity level similar to Dr. Phinney's trials. It would be interesting to know if Kent has run previous marathons in a keto-adaptive state, but initially at faster paces until hitting the wall near the end of the marathon.

Another low carb endurance runner's account of bonking is on Rambling Outside the Box. In the post by Cynthia and David, they describe running without impairment on several occasions distances up to 50K in a timeframe of about 7 hours without refueling (only water). This pace is in the range of 13 minutes/mile. Again, these experiences seem comparable to Dr. Phinney's research.

Jonas Colting is an example of a very competitive, professional triathlete who embraces low-carb-high-fat nutrition. He has been interviewed by Jimmy Moore and posted on Mark Sisson's site. It's clear though that he doesn't strictly follow low-carb-high-fat nutrition. As he said, all rules are thrown out on race day. He calls it "train low, race high", in reference I think to muscle glycogen. He even is sponsored by the sugary drink manufacturer "Red Bull". However, he does say that his carbohydrate consumption is "a far cry" from the typical amount recommended by Swedish nutritionists for athletes, about 10 grams per Kg of bodyweight, or about 800 grams per day.

Mark Sisson, the former marathon/triathlon "mileage king" doesn't recommend endurance exercise exceeding 80% maximum heart rate to maximize fat burning capacity on training runs. For low-carb-high-fat athletes that choose to compromise some level of fat burning for competitive efforts, he recommends consuming 10-20 grams sugar every 15 minutes after the first 60-90 minutes.

So it seems that some low carb runners like Kent, Cynthia and David correlate well with Dr. Phinney's research, as long as intensity levels are low to moderate. And Dr. Phinney isn't the only researcher to clinically show fat-adaptation doesn't impair endurance. Scientists at the University of Cape Town have published similar results. But it also seems unavoidable that competitive endurance relies at least partially on carbohydrates as per Jonas and Mark. I expect there will be more to say as I do personal experimentation. Til then.


  1. How does one go about factoring in all the unknowables and uncontrollables? Is a disappointing race the result of "improper" training, improper diet, or just having a bad day?

    What makes understanding diet tough is how so many people suffer and succeed with greatly diverging food intake. Also, you bring up adaptability, which could seriously skew the data as well.

    Still, it's all very interesting. Good luck trying to figure it out!

  2. You make good points.

    There are way too many variables to make sense of a specific endurance performance in real life. Even in lab situations (there are a lot of research papers from Cape Town University - same research center as Tim Noakes, author of "Lore of Running"), the studies seem to be designed to control a lot of variables, and yet the averages of results have high standard deviations (lot of variability in results). In fact a study was done at Cape Town where athletes adapted to a high fat diet followed by a day of carbo loading before a high intensity time trial. The idea was to optimize fat oxidation by the high fat diet and load of muscle glycogen (best of both worlds). To your point, some athletes improved performance, some did not change, some were worse. The researchers concluded that some athletes have a "fat burner" metabolic phenotype and some have a "carbohydrate burner" metabolic phenotype.

    So it seems, self-experimentation is in order. What works for me may not work for you. Does this sound familiar Barefoot Josh?

  3. I'm going with Barefeet Josh for now.

    And I say yup, that does sound familiar. I think there are many variables within each individual's body, perhaps with differences in how we process the food we eat. Or a good example if off topic, when I get stung by a bee it hurts, swells a little, then goes away in a few days. If Iris gets stung, she experiences more misery. Other people get stung and, like, die.

    But I think there's another factor that I think might be important but goes unnoticed in the general diet debate/discussion (or at least I think it is; I don't really follow that scene at all), but I'm saving that for my blog.