Nutritional Ketosis, Treating Type 2 Diabetes and More—A Q&A with Dr. Stephen Phinney

Published on 
April 26, 2024
January 11, 2018
Stephen Phinney, MD, PhD
Stephen Phinney, MD, PhD
Stephen Phinney, MD, PhD

Thanks again to everyone who joined our recent Facebook Live session with Dr. Phinney and for submitting great questions. We'll be doing it again soon, and we'll announce our next live event on Virta's Facebook page.

In the mean time, Dr. Phinney took it upon himself to answer many of the questions that didn't get answered. If you don't see your question, it was probably answered live. Enjoy!

Q: Interested in reversing type 2 diabetes - is a ketogenic diet recommended?? Thx!
—Susie T.

Dr. Phinney: Insulin resistance is the hallmark of type 2 diabetes and manifests as carbohydrate intolerance. Like other food intolerances, the most logical and effective approach to managing carbohydrate intolerance is to restrict sugars and starches to within the individual’s metabolic tolerance. A well-formulated ketogenic diet can not only prevent and slow down progression of type 2 diabetes, it can actually resolve all the signs and symptoms in many patients, in effect reversing the disease as long as the carbohydrate restriction is maintained.

Q: Appreciated your article on the concerns about prolonged fasting. Could you comment on the utility and safety of shorter durations of fasting (i.e. 16 hrs of fasting/8 hrs of eating or 20 hrs of fasting/4 hrs of eating over a period of 1 day)?

Dr. Phinney: I do not have major concerns with either time-restricted feeding or intermittent fasting for durations less than 24 hrs as long as:

  • There are adequate protein and vegetable intakes within the ‘feeding window’ to support lean tissue preservation (where the vegetables are a source of potassium and magnesium)
  • The person is not on fixed doses of diabetic medications such as insulin or sulfonylureas that could cause hypoglycemia on the feeding-restricted day

The goal of achieving metabolic health and improved body weight depends upon optimizing fat oxidation and preserving lean tissue, and in our experience the best first step is keeping protein intake moderate and dietary carbohydrates low. In addition to that, if one can intermittently skip one or two meals per day without stimulating hunger and cravings, this may benefit some individuals.

Q: Can you achieve this on a vegetarian diet too? A lot of people I know from India, including my wife, are vegetarian and prediabetic or T2D. I am a T1 and now on keto diet, but I am not able to convince the vegetarians to eat meat.

Dr. Phinney: A well-formulated ketogenic diet should consist of protein (meat, fish, poultry, eggs, dairy) in moderation, providing something in the range of 10-15% of daily energy intake. To remain in nutritional ketosis, most of us need to keep carbs in the 5-10% of energy intake range. Thus the majority of one's dietary energy intake should come from fat with protein consumption in moderate amounts. This does not have to come from meat sources and should be based on one’s personal preference so those who are vegetarian can in fact follow a ketogenic diet.

See our guide to low carb for vegans and vegetarians here.

Q: Does the amount of fat I eat in a ketogenic diet interfere with my ability to burn body fat?

—Marcos C.

Dr. Phinney: It is possible to eat enough fat when consuming a ketogenic diet to prevent weight (i.e. body fat) loss. But most heavy people find that nutritional ketosis tames their appetite and cravings, such that they tend to eat less than they burn. Even if that deficit starts at just 500 Calories per day, the losses in a year can add up to between 25 and 50 lbs. Eventually, as one's body fat reserves drop from 'high' to 'normal', the body senses somewhat greater need and it takes a bit more fat in the diet to achieve the same degree of satiety.

Q: Why are blood ketone values lower in the morning than in the afternoon/evening?—Nikola S.

Dr. Phinney: People vary with some showing this pattern and some starting the day higher and then coming down a bit after meals. But reduced morning ketones are not uncommon and likely due to the ‘dawn phenomenon’ associated with the early morning spike in cortisol. We touch on this in our blog post on the dawn phenomenon.

Q: When I started my cholesterol was fine. Now my glucose has dropped 40%, but my cholesterol is 241. Should I be worried?—Kerry J.

Dr. Phinney: Congratulations for your dramatic drop in glucose. During rapid weight loss, cholesterol that you had stored in your adipose tissue (ie, body fat) can be mobilized, which will artificially raise serum LDL as long as the weight loss continues. The best time to check is a couple of months after weight loss ceases. Total cholesterol includes HDL (the so called ‘good cholesterol’), which usually goes up 10-15% on a ketogenic diet. That said, some people have high calculated LDL cholesterol values even after weight loss stops. If this occurs, you should discuss further diagnostic tests with your doctor. Current research is looking at LDL cholesterol as a mix of different particle sizes, where the small ones are dangerous and the larger ones are not. With a well-formulated ketogenic diet, we see a shift away from the small dangerous LDL even when the total LDL goes up.

Q: Since families often (and should!) eat together, is there any concern about children who are eating high fat diets at breakfast and dinner with their ketogenic parents, but higher carb foods at lunch/snacks at school?—Tera N.

Dr. Phinney: We discussed this live, but adding to it, the most important dietary changes for children are reduced intakes of sugar, high fructose corn sweetener, and refined carbohydrates. They can have their own separate dishes with starchy vegetables and unrefined carbs if they are otherwise healthy and not dealing with a weight problem.

Q: What are your inflammation biomarkers?—Tekla B.

Dr. Phinney: The two most published inflammation biomarkers are total white blood cell count (WBC count) and c-reactive protein (CRP). Both are powerful predictors of future diabetes, coronary heart disease, and the more common forms of cancer.

Q: Can long-term keto diet contribute to hypothyroidism?—Sonia Z.

Dr. Phinney: While there have been no large prospective studies looking at the effects of carbohydrates restriction on thyroid function, there is quite a bit of evidence to show that this is not the case. We have published data from our studies dating back to 1980 showing that this is not a major concern.

Q: Why some people's LDL goes sky high on keto diet?—Sonia Z.

Dr. Phinney: The changes we see in total and LDL cholesterol levels are much less predictable than the changes in triglycerides and HDL cholesterol. For some people calculated LDL cholesterol goes down, and for others it goes up, sometimes quite a bit. This tendency for some people to see an increase in their LDL cholesterol has been a focus of our research over the last decade. We have found that when LDL cholesterol goes up on a carbohydrate-restricted diet, most of this increase is in the ‘good’ or ‘buoyant’ LDL fraction. We have also observed that some of this early rise is due to mobilizing cholesterol from fat (adipose) tissue as weight loss occurs, and it often normalizes when weight loss stops. In either case, we can confidently say that even an increase in one’s calculated LDL cholesterol represents no increase in risk.In order to be sure, it is a good idea to follow up with a physician and perform the most up-to-date assessment of your blood lipids, including the size and amounts of your LDL cholesterol fractions with a NMR or VAP lipoprotein test.

Here are our 1-year clinical trial results as they relate to blood lipids.

Q: There is a lot of concern about protein intake and gluconeogenesis kicking people out of ketosis. As stated in your book, moderate protein is .6-1 gram/pound of lean body mass. As long as you stay within that protein range is GNG something to be concerned about?—James F.

Dr. Phinney: When most of us are counseled to restrict carbs, because we have been brainwashed to avoid fats, we tend to overeat protein. Thus we need to consciously keep our protein intake moderate and increase our fat intake to achieve satiety. We will be posting a more detailed explanation on our blog within the next week.

Q: Does Diazoxide helps in the ketone production? Thanks a lot for sharing your knowledge. Your influence is bigger than you think.—Salomon J.

Dr. Phinney: Interesting question. I am not aware of published human research on this topic. Diazoxide is a drug that blocks insulin secretion by the pancreas. It is routinely used in infants who are born with hypoglycemia due to high insulin secretion. Diazoxide does not reduce insulin resistance, which is the underlying defect in type 2 diabetes and metabolic syndrome. Taken in the context of a low carbohydrate diet, it might increase ketone levels somewhat, but the safety and sustainability of this should be carefully tested in a human clinical study.

Q: Does athlete fat oxidation rate vary continuously with carb intake? Or is keto the only way to get it above 0.5g/min?—Norman T.

Dr. Phinney: In a study done in the Netherlands of 300 adults, some of whom were athletes, the average rate of fat oxidation was about 0.5 g/min and the highest was 0.99. They did not assess each individual’s carbohydrate intake, so we can’t tell if it was a significant cause of the inter-individual variation. But when we test keto-adapted athletes, we routinely get values over 1 g/min. Jeff Volek is currently doing a study in college students undergoing 3 months of keto-adaptation, so we should have a better answer some time this year.

Q: My daughter has high uric acid on KD, we don’t know if she did before KD, have you seen this before?—Justine L.

Dr. Phinney: Yes, blood levels of uric acid usually double in the first week of a ketogenic diet or with fasting. This is because there is a competition between ketones and uric acid for excretion by the kidneys early in adaptation. As the adaptation process proceeds over a few months, the uric acid level comes down to normal levels even as nutritional ketosis continues, Thus this is a process of clearance (i.e., excretion), not over-production from dietary protein, and this temporary elevation in uric acid is usually harmless unless one is prone to gout. That said, the precursors of uric acid are pretty high in organ meats like liver and kidney, so if these are part of one’s diet, they should be eaten in moderation. You can learn more about keto-adaptation here.

Q: What about Camolina Oil?—Jim K.

Dr. Phinney: Camelina comes from the same wild plant family (brassica) as Canola and mustard. It contains about 30% alpha-linolenic acid which is the omega-3 fatty acid also found in flax seed. Thus in small quantities it can be a good vegetarian source of omega-3.

Q: Thoughts on patients suffering from depression and/or hypothyroidism and keto diets? Also, interested in your thoughts on "The Fast Metabolism Diet" by Haylie Pomroy, which encourages a 5x/day eating plan, which includes a couple days of a keto-like diet after some carb and protein-heavy days each week over the course of a month. Thank you for what you do, sir. Appreciate your work.—Tyler S.

Dr. Phinney: Dr. Laura Saslow working with Professor Rick Hecht at UCSF did a randomized trial of a ketogenic diet vs standard mixed diet in people with type 2 diabetes. Compared to the mixed diet, at 12 weeks the ketogenic group reported significant reductions in negative mood, diabetes distress, and sweet cravings. We have seen similar improved mood and energy at 10 weeks in our Indiana University Health study. We have discussed thyroid metabolism in our blog post ‘Does Your Thyroid Need Carbs?’

I have not seen much published, high quality science supporting the 5:2 eating pattern, and certainly not supporting the remarkable claims of popular authors. Our experience is that the human body ‘likes’ (i.e., functions better and more efficiently) being adapted to a consistent supply of fuel. Riding a roller-coaster between carbohydrates and fats every week causes problems for many people, and it is certainly contra-indicated for anyone on glucose lowering meds for diabetes.

Q: Love your and Dr. Volek's low carb performance book - amongst the 'good' fats you talk about high-oleic safflower being acceptable (better PUFA ratio). Are high-oleic SUNflower oils OK and can you cook with them?—Chris B.

Dr. Phinney: Yes. Any oil for which the label on the back lists more than twice as much monounsaturate as polyunsaturate is OK. By the way, I’ve been told that high oleic soybean oil is coming to market soon.

Q: Could blood ketones, along with blood sugars, be used to fine-tune insulin dosing? Too little insulin with someone with type 1 and we get DKA. With too little insulin ketones disappear. Could we use ketone levels are consistently below 0.3mmol/L as a signal that we have too much basal insulin?—Marty K.

Dr. Phinney: This is crossing the border into specific medical advice, so unfortunately I can’t go there in this format. However this is a very insightful research question, and hopefully will be addressed by folks in the type 1 keto community.

Q: Does Virta know the benefit of whey protein for diabetes? Not sure in the context specifically of DM, but it is a very high quality protein source - BCAA, immunoglobulins—Salomon J.

Dr. Phinney: Whey protein is a two edged sword: it has a very high biological value score, but it also causes higher insulin secretion than other quality protein sources like meat and eggs. My other concern with isolated whey protein is that it does not come along with intracellular minerals such as potassium, magnesium, and phosphorus. In that sense, it is not a complete food like eggs, fish, poultry, or meat. Of course, if included as part of a diet containing adequate vegetables, whey protein can have a place in a well-formulated ketogenic diet.

Q: Why NOT eat certain oils, such as safflower and sunflower oil? Do these oils impact ketosis or is it for other reasons?—India K.

Dr. Phinney: High omega-6 oils such as corn, soybean, cottonseed, peanut, and the older versions of safflower and sunflower are good sources of essential omega-6 fatty acids if one is consuming a low fat diet. In that case, most of your daily energy comes from carbohydrates and fat is just a sliver of your daily energy use. On a well-formulated ketogenic diet where 70-80% of your daily energy comes from fat, most of your dietary fat should be from sources high in monounsaturates and saturates that the body likes to burn for energy. We only need about one teaspoon (5 grams) each of omega-6 and omega-3 fats each day to meet our essential fatty acid needs. Too much omega-6 in particular is not well-tolerated by our digestive tract. The older ‘high linoleic’ versions of safflower and sunflower oil are 65-75% omega-6, so if even one third of your daily energy need (say about 600 out of 1800 Calories) came from one of these oils, that 600 kcal would contain 8 teaspoons of omega-6 fat. That’s way too much, and way out of balance with a normal omega-3 fat intake. We discuss this in chapter 16 of our book ‘The Art and Science of Low Carbohydrate Living.’

Q: Can you please explain hair loss on Keto?—Jessy G.

Dr. Phinney: There are many causes of hair loss. One treatable cause is deficiency in the mineral zinc, which is also associated with very dry skin and reduced sense of taste and smell. Zinc repletion is a bit tricky, and is discussed in chapter 9 of our book ‘The Art and Science of Low Carbohydrate Performance’.

Our hair (well, for those of us who still have some) typically grows for about 12-15 months, then stops for 3 months, falls out, and the same hair follicle starts growing a new hair 3 months later. This 24-month growth, shedding, and regrowth cycle occurs at random for each of the 200,000 hairs in our scalp. But in the event of a major stress such as physical or psychological trauma, surgery, childbirth, or major weight loss, a synchronized hair loss called a telogen effluvium can occur. This hair loss typically occurs 2-4 months after the stress and is temporary, as new hairs start to regrow 2-3 months later.

So whatever the cause, a telogen effluvium rarely causes shedding of more than half of one’s hair, and in almost every case it’s resolved within a year. So don’t go out and buy an expensive wig or toupee. More importantly, since a telogen effluvium is not commonly seen with a well-formulated ketogenic diet, one may choose to avoid extremes of calorie restriction like very low calorie diets or prolonged fasting.

Q: Thanks for all you do!! There are groups that advocate the potassium to sodium ratio at 4 grams to 1 gram per day. Does that fit into your view of those two. Also, is 400 milligrams of Magnesium appropriate? Thank you! My ketone level ranges from .3 to 1.2.—Mark K.

Dr. Phinney: Rather than buying in to the dogma of those on the extreme sodium restriction fringe, it helps to look at actual scientific data. In the PURE Study that I mentioned (102,000 people from 17 countries studied for 3.7 years), the lowest mortality was seen with sodium intakes around 4 grams per day and potassium intakes between 2 and 4 grams per day. This suggests a potassium to sodium ratio of 0.5 to 1.0. But I hate ratios, because what we eat is X and Y, not X over Y, and this should come from real food. Given the accelerated excretion of sodium with nutritional ketosis, most of our non-hypertensive patients should consume 5 grams of sodium per day. For potassium, we recommend eating protein from real foods and consuming 5 servings of non-starchy vegetables, nuts, and berry fruits per day for potassium. When they do this, their kidneys manage the ‘ratios’ just fine.

Q: Are you looking at Cardiac Calcium Score?—Lincoln C.

Dr. Phinney: As a dynamic test of coronary risk progression over time, we have chosen to look at carotid intima-media thickness instead.

Q: Hi Dr. Phinney. I have chronic pain and take a lot of medications. Also have controlled type 2 diabetes. Do you think taking 4400mg of gabapentin a day could be why I am not losing weight that easily. I am eating under 30 grams of carbs a day and I dont eat a lot of fat. I eat just twice a day, and only around 950-975 calories. Thank You!—Teresa R.

Dr. Phinney: The effects of anti-seizure and mood stabilizing pharmaceuticals on body weight remains controversial. Unfortunately I cannot offer specific medical advice via this forum.

Q: While someone pursuing a therapeutic ketogenic diet for the management of cancer, epilepsy, Alzheimer’s or dementia may want to reduce protein to achieve elevated ketone levels, do you think someone trying to manage diabetes or lose weight needs to consciously worry about "too much protein” or can they just follow their appetite when it comes to protein intake?—Marty K.

Dr. Phinney: Too much protein can drive down ketone production in the liver and therefore, may reduce the beneficial effects of nutritional ketosis. A well-formulated ketogenic diet should consist of protein (meat, fish, poultry, eggs, dairy) in moderation, providing something in the range of 10-15% of daily energy need to maintain critical structures and functions of the body. Most healthy humans maintain lean body mass and function during a ketogenic diet providing between 1.2 and 1.75 grams of protein per kg of ‘reference body weight’. This is an excellent question that we have heard often, and we have an upcoming blog post on this topic that should be available this next week.

Q: I note that the average blood ketone levels in the first ten weeks of the Virta study was 0.6mmol/L with a standard deviation of 0.6 mmol/L. This is less than the generally understood optimal ketone range of 0.5 to 3.0mmol/L. Will the blood ketone results from the Virta study be used to refine our understanding of target ketone levels for people managing diabetes?—Marty K.

Dr. Phinney: Yes, there was a lot of variance among participants in this study. But on average, the majority were above 0.5 mM BOHB in the first 10 weeks. And when we look at who did well over a longer period of time, it was those who had higher ketones in their first 10 weeks.

Q: If you are trying to reduce saturated fat but maintain ketosis, what fat sources do you recommend?—Gail K.

Dr. Phinney: Olive oil is pretty close to ideal in its fatty acid composition. Other good fat sources are high oleic safflower and sunflower oils, plus lard (which is typically half monounsaturated) and canola oil. But given that we have demonstrated over and over that keto-adapted patients can rapidly burn saturated fats, why would we want to restrict them other than to satisfy those in the mainstream that still believe that Ancel Keys’ flawed hypothesis was right?

Q: Dr. Phinney, thank you for sharing your knowledge and together with others in the scene giving me a new lease on life. With your Virta program developing, is there a standard recipe for electrolytes for people following a Ketogenic Diet other than 2 stock cubes a day? Especially Magnesium seems to be a bear. I'd also love to hear about machine learning and how that applies to the Virta venture and whether there are any patterns emerging?—Martin G.

Dr. Phinney: Your last question first, we are in the very early stages of using machine learning to guide our patients and coaches to individualized recommendations. As for sodium and potassium (aka electrolytes), rather than trying to track them directly (which is frustrating at best), we recommend salting food to taste, adding 2 grams of sodium as broth or bouillon, and eating 5 servings of non-starchy vegetables daily. For magnesium, muscle cramps are our best indicator of depletion, and we reverse it with 3 tabs per day of Slow-Mag or the generic equivalent daily.

Q: Dr. Phinney you changed my life. Thank you.

Dr. Phinney: Jeff Volek and I (and our many students and collaborators) are a bit like explorers who have mapped the low carbohydrate ‘territory’. Kudos to you, who have translated our map into life-changing results.

Have more questions about nutritional ketosis? Check out our FAQ by Dr. Steve Phinney and the Virta team.

The information we provide at and is not medical advice, nor is it intended to replace a consultation with a medical professional. Please inform your physician of any changes you make to your diet or lifestyle and discuss these changes with them. If you have questions or concerns about any medical conditions you may have, please contact your physician.

This blog is intended for informational purposes only and is not meant to be a substitute for professional medical advice, diagnosis, or treatment. Always seek the advice of your physician or other qualified health provider with any questions you may have regarding a medical condition or any advice relating to your health. View full disclaimer

Are you living with type 2 diabetes, prediabetes, or unwanted weight?

Check to see if your health plan or employer covers Virta
Check Eligibility

Frequently Asked Questions

No items found.