Keto-Adaptation to the Low-Carb Diet: A Clinical Conversation with Stephen Phinney,MD,PhD,and Robert Rountree,MD

Dr. Stephen Phinney received his MD from Stanford University. He then went on to complete a PhD in Nutritional Biochemistry at Massachusetts Institute of Technology (MIT) and postdoctoral research at Harvard University. Dr. Phinney has held academic appointments at the Universities of Vermont, Minnesota, and California at Davis. Most of his research has involved diet, exercise, and inflammation. He is now the Chief Medical Officer of Virta Health in San Francisco, California.

Robert Rountree: Could you start by letting us know how you ended up where you are in your career today? Have you always wanted to be a nutritional biochemist/physician?

Stephen Phinney: My interest in nutrition began with endurance sports, as a noncompetitive endurance cyclist. I used to think that riding 1,000 miles on a loaded touring bike by myself was a fun thing to do. Or going out and riding 200 miles in a day was fun. At the time, I was living in California and going to school in the Bay Area. To deal with the frustrations of medical school, I would get on an old, heavy road bike and ride over the hills to the Pacific Ocean and back on a Sunday morning to clear my head.

At a reasonably young age, I learned that if I did not eat carbohydrates or carbs, I would run out of fuel pretty quickly, and I would feel lousy. I learned quickly that if I did not eat carbs, I would find myself walking up the hills, pushing the bike, on the way home. There was no Uber back then! I just could not sustain the effort to ride.

Now, this was even before the late 1960s when Swedish scientists were developing the concept of carbohydrate depletion, or hitting the wall, and preventing that with carbohydrate loading. I empirically discovered that if I did not eat carbs, my performance would tank and I would feel lousy. So, I was a very strong advocate of carbs for sport just on personal experience.

When I got to my internal medicine training at the University of Vermont, I met some of the endocrine metabolism faculty there. They were very interested in diet, exercise performance, and obesity. One of the faculty members was Dr. Ethan Allen Hitchcock Sims, MD. Around 1975, Ethan and I got into a conversation about carbs and performance. This was shortly after the first edition of the book on the Atkins' diet ¹ was popular. I was completely adverse to the idea of eating a carb-restricted diet. I felt that although someone could survive or lose weight on this diet, it would make people feel bad, and they could not do anything vigorous.

Ethan mentioned that he had been out cross-country skiing with a friend of his who had been on the Atkins diet for a couple of months and had lost a lot of weight. He told me that his friend clearly was not impaired performance-wise in cross-country skiing and suggested that I rethink my opinion.

Based on this, we conjured up a protocol and undertook a study. We did the study in 1977. It involved a group of overweight people in the metabolic ward. They were put on a very low-calorie ketogenic diet, a low-fat/low-carb diet, with roughly 600–800 calories per day of lean meat or fish and vegetables. We got people to agree to the study because we promised them they were going to lose a lot of weight. Over the course of six weeks, the average weight loss was about 25 pounds, with about 80% of the weight loss from body fat. Also, over this time, without any change in daily activity, our group did not lose peak aerobic power when tested walking uphill on a treadmill. In fact, although there was a decrease at the end of the first week, they were able to walk 50% longer on the treadmill at six weeks. By now, the average person had lost 25 pounds, so to make it fair, we made them wear a backpack and we put heavy objects in to bring up their weight. Of course, being mainly sedentary people, they complained bitterly. However, they still went 50% longer.

This finding forced us to coin the term “keto-adaptation” to explain what we were seeing. Most of the studies done by the Swedes to study high- and low-carb diets on muscle glycogen were very short—often one week or less. And so, I proved myself wrong and this sparked my interest in this field.

Dr. Rountree: When you first did the study, I imagine there was huge pushback from people? I would imagine a lot of people felt that you were going “against the grain”?

Dr. Phinney: Yes, actually against the grain. I love the pun. Yes, this was 1978/1979. At this time, there was not quite the pushback against dietary fat and carbs. The first official Department of Agriculture dietary guidelines did not come out until 1980.

Our paper was published in a high-quality journal and still the pushback came. ² However, this study taught me a few things. The first was a certain amount of humility. The second was that I knew nothing practical about nutrition. I had been to a good medical school. I had done a reasonably acceptable internal medicine residency. I had basically collaborated with and been mentored by physicians who were very interested in nutrient metabolism and performance. However, I knew nothing about practical nutrition.

So, I decided to go back to school. My original plan after finishing my internal medicine training was to go back to school for a year and take some courses on nutrition. The best academic nutrition program in the country at that time was at MIT. So I went to Boston, and I worked with Dr. Bruce Bistrian, MD, at Deaconess Hospital in Boston. At the time, the goal of the department was to bring in MDs and give them training in nutrition. The goal was to train clinical academicians to populate faculty positions at medical schools and teach nutrition.

About 50 of us went through this graduate program over a period of more than a decade, and very few of us found sustained academic jobs. In the end, I spent about 20 years in academia attempting to make that my career, to be able to teach nutrition to medical students. I was reasonably well-received as a colleague, a teacher, and a clinician. However, I could never get grants funded because nobody wanted to fund studies of low-carbohydrate nutrition. Overall, it was believed that it was not valid and why waste the time researching a diet that nobody wants to follow. Despite this, I was able to publish about 60–65 papers while at UC Davis—a good, solid performance. Most of my papers were human research, which is much harder to do than lab bench research and more expensive. That is always where my heart has been—developing interventions that have utility in humans.

My goal all along has been to figure out how to exercise the power of nutrition for human health. I have managed to discover some interesting things along the way, things to do with polyunsaturated fats and membranes. I learned that you are not what you eat—you are what you save from what you eat. I am basing this on the area of dietary fat. The assumption is that the mix of fats that you eat is the mix of fats that are going to build up in the body. However, this assumption completely ignores the fact that through evolution of multicellular organisms over the course of a billion years, our bodies have figured out how to partition things between things they want to save and things they do not want to save. In other words, we have been able to demonstrate that in spite of a diet high in saturated fat for an extended period of time, these fats do not necessarily build up in the circulation or in cell membranes. The body is very selective.

The other thing we discovered is that following the keto-adaptation process while following a low-carb diet, the body's ability to burn fat is basically doubled. We don't know exactly how this works. ³ It looks like there is a marked increase in mitochondrial function and potentially mitochondrial density. So, when you take most of the dietary carbohydrates away from someone and they go through this adaptation process, which probably takes about 12 weeks for a full change in metabolic and physical machinery, there is a dramatic change in the capacity to burn fat, both at rest and during exercise.

This has two obvious benefits. One is that if someone is eating a lot of fat, the body is capable of burning whatever components that might not be needed for structural or long-term energy storage. The second thing relates to the body's ability to burn fat as fuel. All the exercise physiology textbooks say that better-trained athletes have the ability to burn more fat. However, even in highly trained athletes, when you increase the intensity of exercise to an “intermediate” range of about 50% of a person's maximum oxidative capacity or VO₂ max, the ability to burn fat drops off rapidly, and the body becomes essentially almost wholly carbohydrate dependent.

With a ketogenic or low-carb diet, a diet in which carbs are low enough that people make moderate levels of ketones and they become keto-adaptive, not only is the rate of fat oxidation doubled at that 50% stage, but the capacity to burn fat at a much higher rate extends to 70–80% of VO₂ max. So, endurance athletes running at their steady race pace during an endurance event such as a marathon or triathlon are exercising between 70–80% of VO₂ max. This means that the capacity to burn fat at a very high rate is sustained out to their race-pace level.

We saw the harbingers of this in my first two studies. The first one, in Vermont with untrained people, involved a very low-calorie ketogenic diet. ² The second, when I was doing my graduate work at MIT, involved experienced cyclists eating a diet patterned on the traditional diet of the Inuit, 85% fat/15% protein. ⁴ In fact, the only carbs they got during this four-week study was the glycogen that was in the meat at the time the animals were killed. During this study, which was a bit shorter than the others, we did not see an extension of performance above baseline, but they were able to bring their performance back up to baseline. Their baseline was >900 calories of energy expenditure per hour, on a stationary bike. They were able to do the same average duration of exercise, even though they had only had four weeks of adaptation. During the ride, we saw that they were burning 85–90% of their energy as fat.

We published this study in the journal Metabolism in 1983. ⁴ It just sat there. People treated it as a piece of radioactive debris, piled sand on it, walked around it, and tried to ignore that it was there.

Dr. Rountree: Was this about the time that ketogenic diets were just starting to be used for epilepsy?

Dr. Phinney: Well, more specifically, they were just coming back for treating epilepsy. Remember, they were the only treatment for epilepsy in the mid-1920s. The first modern ketogenic diet for pediatric seizures was developed at the Mayo Clinic. So, through World War II, and into the 1950s, this was the standard treatment. However, when pharmaceuticals came along, we cavalierly decided that the ketogenic diets were worthless. In the late 1990s, Dr. John Freeman, MD, working with Dr. Eric Kossoff, MD, at Johns Hopkins, brought the ketogenic diet back, based on findings that at least half of seizure drug-resistant children achieved complete remission or marked benefit from a ketogenic diet. This is what helped the ketogenic diet for pediatric seizures come back into mainstream acceptance in the late 1990s. ⁵

In 2003, I met an exercise physiologist/registered dietitian named Dr. Jeff Volek, PhD, RD, then at the University of Connecticut. By then, I had pretty much given up on my low-carb research. However, Dr. Volek had read my papers from long ago, and he convinced me that we had to come back and prove the safety and efficacy of a well-formulated ketogenic diet. Dr. Volek was doing low-carb studies in people with metabolic syndrome. He was seeing dramatic improvements in blood lipids, blood pressure, and other measures of metabolic syndrome. He wanted to look at the long-term safety implications of the low-carb diet. I have been collaborating with him since then. Interestingly, about five or six years ago, the low-carb diet became a novel tool for use by ultra-endurance athletes, particularly ultra-runners, people who run distances of 50–100 miles. The reason that this was attractive to this group of people was that if you run more than a marathon distance, you cannot keep running if you have been eating carbs before the race, without eating a lot of carbs in the race. Remember, once you run out of glycogen, you are going to hit the wall, and when you hit the wall, your performance is toast. Ultra-runners had learned that they had to consume about 300–400 calories of carbohydrates per hour after the first hour of their event if they wanted to be competitive. The body can only store about 2,000 calories of glycogen maximum. So, at 500 calories an hour, you are going to run out at four hours. With a 10- to 20-hour event, the athlete will have to eat their way to the finish line. Also, this amount of carbohydrates would cause tremendous gastrointestinal problems. Early on, some athletes read some of our research. They started trying ketogenic diets without our guidance, and eventually people started reporting it. They started telling people that they did not “hit the wall” or “bonk.” They said it made them “bonk-proof.” Even a male runner weighing 160 pounds, with 7% or 8% body fat, still has 30,000–40,000 calories of fat in their body. By keto-adapting, it gives them access to that much larger fuel tank. ³

For a very efficient runner, the energy cost of running 100 miles is about 10,000 calories. If the runner depends on carbohydrates as fuel, 2,000 calories stored is not enough. The runner can calculate how much carbs are needed to get to the finish line. However, the high-fat runner, who can run on 90% of their energy as fat, has a 300-mile distance in their fuel tank. Although we don't know if carbohydrates are even necessary for these runners, they can consume about 100 calories of carbs per hour, which is less than a quarter of what the high-carb runners take. ³

Dr. Rountree: We have a lot of big running and triathlon events in my town, and I still see advertisements about getting together for pasta dinner the night before the race.

Dr. Phinney: Yes, and yet there is a woman from Colorado named Abby McQueeney Penamonte. She is an ultra-runner. One year, over a five-month period, she ran four 100-mile runs on harsh trails, in Vermont, Utah, California, and Colorado. Her cumulative time for all four races set a record for any woman who had done all four. She did all four of these races on a low-carb diet.

There are two reasons that we think this is successful. One is that there is no hitting the wall, meaning a lot of carbs are not needed, and there are fewer gastrointestinal problems. The second observation came from the runners themselves. They said that their recovery times were dramatically shortened, and this was a predictable and robust effect of being keto-adapted. The general rule of thumb among ultra-runners is that they need to take six weeks off from training to recover after running a competitive 100-mile race. Although not specifically studied, we know that Abby was able to do four of these in five months, with better times than anybody who had previously done them. We know of a U.S. Marine fighter pilot who ran the Western States 100-mile race in under 20 hours. The next weekend, he felt so good that he thought he would go out and just do a few easy miles, but instead put in the fastest eight-mile training run he had ever done in his life. Remember, this happened seven days after completing a 100-mile run. He said he could never have done that while using carbs.

A third example is a U.S. Army Special Operations soldier. After feeling an overall lack of energy, he tried a low-carb diet. About four months into his keto-adaptation, he not only won a race in Texas, he set a course record and beat his next-nearest competitor by an hour. By midweek, he was feeling limber enough to do some easy miles. He then thought, what the heck, and did another 100-mile race and won again—beating his next-nearest competitor by 45 minutes.

These examples fit into what Dr. Volek and I have discovered in our research and published in multiple papers. When somebody is keto-adapted, their defenses against oxidative stress and inflammation are dramatically enhanced.

Dr. Rountree: You say that keto-adaptation takes about six weeks or so to kick in fully. Can you discuss some of the commercial products available suggesting shorter time lengths are needed when these products are used?

Dr. Phinney: We know that we can't drink exogenous ketones for a week and get all those benefits. We are in the early stage of understanding exogenous ketones, and we may not want to get into that quite yet because the marketing has gotten ahead of the science. There was a product that was developed by Drs. Richard Veech and Kieran Clarke, from Oxford, for example. But time will tell if you can actually eat those ketones consistently day-in and day-out, and experience similar adaptation.

Dr. Volek and I have found that biomarkers of inflammation in the body, such as C-reactive protein (CRP) or inflammatory cytokines such as interleukin (IL-6), are uniformly and strikingly reduced during the keto-adaptation process. ⁶ The decrease in CRP appears to take 12 weeks, although other changes such as a decrease in white blood cell count probably happens more quickly. This is not like taking ibuprofen or anti-inflammatory steroids such as prednisone. It is not a single chemical step. It appears to be a broad-spectrum effect. And it appears to take quite a period of time for all the benefits to accrue. The other piece that is really fascinating is that much of the change seems to stem from reduced damage related to oxidative stress or reactive oxygen species.

Dr. Rountree: Do you think mitochondria are at the core of this mechanism?

Dr. Phinney: Yes. It took us a long time to understand the mechanism. About five years ago, a group of gene regulation scientists, without an interest in nutrition, were looking at the regulation of defenses against oxidative stress.

There is a group of enzymes called histone deacetylases, which are extremely important in physiology. They are the enzymes that determine which genes are allowed to be active and which are silent. When specific histone deacetylase genes, the class I histone deacetylases that control many of our inborn antioxidant defenses, are active, our defenses are shut off. ⁷ However, when those genes are suppressed, the gene-silencing enzymes are no longer active, and our defenses against oxidative stress come up and come online.

It turns out that beta-hydroxybutyrate, the primary ketone that we have circulating in our bloodstream when we restrict carbs enough to be in nutritional ketosis, or following intake of a product containing the butane ester of ketones, increases in the blood to levels in the 1–2 mmol/L range. ³ In comparison, ketoacidosis is life-threatening, with levels in the 15–30 mmol/L range. However, that small level of 1 mmol/L beta-hydroxybutyrate in the blood seems to have a pharmaceutical-like effect on the class I histone deacetylases. ⁸ This was determined by scientists at University of California San Francisco. This level resulted in about a 50% downregulation of those gene activities, and dramatically enhanced the quenching of reactive oxygen species.

We now know that beta-hydroxybutyrate is not just a fuel that the liver makes for use as an “alternative to glucose.” In fact, if anything, it is probably preferred over glucose. We now know that beta-hydroxybutyrate is also a potent epigenetic signaling molecule, changing gene expression, and reducing inflammation and oxidative stress throughout the body. So, this may be the reason why the volume of damage, and the rate of recovery, is dramatically shortened in ultra-runners in nutritional ketosis.

Dr. Rountree: Are you aware of any other information as to how beta-hydroxybutyrate affects pathways in the body?

Dr. Phinney: Beta-hydroxybutyrate also works through the mTOR pathway. Based on that, two groups designed a study to examine the effect of a ketogenic diet in a mouse model of longevity. Both were published recently in the journal Cell Metabolism. ^9,10 One group found that the ketogenic diet increased the longevity of the mice by 13%. They also found that agility and mental acuity were maintained much longer, suggesting improved life-span, and health span.

Dr. Rountree: Do you think that the ketogenic diet is for everybody or do you think it is more for ultra-runners, or people with diseases such as metabolic syndrome and epilepsy? Is there a broader application? Do you do genetic testing to see if some people might be more likely to benefit than others?

Dr. Phinney: The idea of genetic testing is being done by others, so that is in progress. Of course, it is really complex because we have so many genes. So, not only do we need to test for their presence, expression and activity are also important. In other words, it isn't just if someone is born with it or not.

So, we are going to be a long way away to answering this question. In the meantime though, given the pushback from friends and family and physicians, there really has to be a significant benefit to the individual to make it worth their while to follow this diet.

Initially, Dr. Volek's and my focus of our research was with people with metabolic syndrome, or prediabetes. Once we demonstrated that we can essentially reverse metabolic syndrome in 12 weeks with a well-formulated ketogenic diet, things moved forward. We published a couple of studies on this in the journal Lipids between 2008 and 2010. ^6,11

We were then approached by an entrepreneur, Sami Inkinen, who is a world-class triathlete and a very successful Internet entrepreneur. Diabetes runs in his family. So, as it happened, he was diagnosed with prediabetes, while fully trained and competing at the world-class level, and of course, following the normal dogma of eating a very high-carbohydrate diet.

Sami read our papers, where we described type 2 diabetes as a disease of carbohydrate intolerance, and suggested that treatment should start with removing enough carbohydrates from a person's diet that they were no longer intolerant to those carbohydrates. This goes against most of the information provided today, which says that consuming carbohydrates has nothing to do with diabetes.

So, we started a company. Sami suggested that we develop a safe and effective way to manage people with type 2 diabetes without having to put them in a hospital. We know that if someone with type 2 diabetes is on multiple medications, and carbohydrates are removed or reduced, the need for those medications is dramatically reduced. The medications need to be reduced in a timely fashion in order to prevent severe hypoglycemia, which can be life-threatening. This means that the medication use needs to be choreographed in a real-time basis with the diet, and biometric monitoring is needed.

This was a concept that we worked on. Our first goal was not to develop a product and sell it, but to develop a product and prove that it worked. We are now running the largest nonsurgical type 2 diabetes reversal study ever conducted, with 400 people. Surgical methods, such as obesity surgery, are also being developed for type 2 diabetes reversal. However, this treatment is not without side effects, such as malabsorption of nutrients.

Our diabetes reversal study is a collaboration between Virta Health and Dr. Sarah Hallberg, DO, at Indiana University Health in Lafayette. The patients in our study are following a well-formulated ketogenic diet and being compared to a control group following mainstream advice. We have an online specialty clinic with Dr. Hallberg and our coaches. It is all done virtually through a computer or smartphone app. In the first three months, the average person interacts with the coach three times per day. This interaction can involve questions about food eaten, adverse effects, and so on. They receive education through short video segments that they can go through at their own pace.

To Contact Dr. Stephen Phinney

Stephen Phinney, MD, PhD

Website: www.VirtaHealth.com

E-mail: Steve@virtahealth.com

We are now collecting data from year 2 in this study. We published our preliminary data at 10 weeks last year. ¹² We've had a bit of a struggle getting our most recent paper with our one-year outcome data published, but it is now in press. We report that more than half of the people who entered our program on medications for type 2 diabetes now have a sub-diabetes level of hemoglobin A1C and are off all of their medicines except for metformin. In addition, the mean weight loss is about 12%, serum triglycerides and HDL cholesterol are improved, and biomarkers of inflammation (total WBC count and CRP) are dramatically reduced.

Dr. Rountree: Is this a subscription-based model with fees?

Dr. Phinney: Yes. We have signed contracts with a number of employers who realize that their employees with type 2 diabetes are costing them a huge amount of money in terms of medications, regular doctors' visits, and emergency room visits, if they have had hypoglycemia or something. We basically guarantee them that they will save more money in the first year than our program costs.

So, we have a group of those patients, and we also accept people who have approached us online and signed up and pay individually. It is not an inexpensive program. We have physicians and coaches working constantly with these people. Also, there is no end to the program. This is not a six-month program or a three-month program. We have year 2 data for some people. In order to support our clients fully, and keep them up to date on new discoveries with our data, we want to keep them in our family for years or decades. We have now extended our Indiana study to a five-year study from the initial two years planned. We expect more than half the people will stay with us for five years.

When people have to change their lifestyle, we have to keep them confident. We need to track them to make sure that it is safe, so that they can convince their physician and their family and their friends it is safe. This cannot be done in three months before sending them packing. This is something that it is a constant, ongoing commitment and intervention.

Dr. Rountree: What I am hearing is that the trick to getting people going on this diet is lots of support.

Dr. Phinney: And the correct information. We have them measure blood glucose and blood ketones. The ketones are measured every day for at least the first three months. They get real feedback on what works and what does not work. And if they have a piece of supposedly low-carb pizza the night before, and their ketones are in the basement in the morning, they learn not to do that.

There is so much misinformation on the Internet now. There are suggestions that people who stay in nutritional ketosis for more than a few weeks develop adrenal fatigue. There is no such thing as ketone-induced adrenal fatigue. That is inadequate sodium in the face of the kidneys accelerating their salt excretion when somebody is keto-adapted. There is another concept that the thyroid needs carbs. Yes, thyroid function tests change, but the low-carb diet does not cause hypothyroidism. We publish this information in the form of published papers and books, but we are now also posting this type of information on our company blog at https://blog.virtahealth.com/category/nutritional-ketosis-science/

Dr. Rountree: Where do you recommend physicians go if they want to get good, solid evidence-based information on this type of diet? Do you have a website that you recommend to people or a particular book?

Dr. Phinney: Dr. Volek and I are realizing that most people do not go and read the primary research. Between us, we have a few hundred papers on this topic. However, the story is so fragmented, that we decided that we would write a book for healthcare practitioners in 2011. But then, we could not get anybody to publish it for us because they said we couldn't give this information. Getting past the editorial censors was a problem until we discovered that Amazon has a self-publishing arm called CreateSpace. We have since published a book, available through Amazon, entitled The Art and Science of Low Carbohydrate Living. ¹³ Chapter 15, entitled Treating Type 2 Diabetes as Carbohydrate Intolerance, is the chapter that attracted the attention of a world-class athlete and entrepreneur, Sami Inkinen, and led to the formation of our company. The book is written for nurses, pharmacists, dietitians, physicians, and healthcare practitioners. It is to help them understand why what we do is safe and, in general terms, how to do it. It is not a prescription of what to do week by week. This book is the only place where we think the story is told in a coherent and understandable way. For our high-level athlete friends, we have a companion book called The Art and Science of Low-Carb Performance. ¹⁴

Dr. Rountree: Is there anyone who should not follow this diet?

Dr. Phinney: Right now, the only people who should do this are people who will really benefit from it or want to learn about it. People should not do it on their own if they are on medications for hypertension or type 2 diabetes. Seek medical advice to do it safely. We may find increasing benefit for other diseases that are associated with inflammation, such as Alzheimer's or polycystic ovary syndrome. These are active areas of research with some early information.

Dr. Rountree: I think your message is loud and clear and very powerful, and I want to thank you for all the work you have done in the last few decades. If people want to get in touch with you, do you have a public e-mail address or a website they should reach you through?

Dr. Phinney: We are just beginning to flesh out the blog on our Virta Health website, VirtaHealth.com. Also, my email address is Steve@virtahealth.com. I really enjoy interacting with physicians if people want to reach out and ask me questions. I have been known to give away books if people do not have easy access to them. My goal is to get the message out because this is a potent metabolically effective tool.   ■