January 19, 2020


Intermittent fasting

Metabolic switching and different forms of fasting.

Read Time 3 minutes

In my previous email, I laid out a nutritional framework, consisting of 3 levers—dietary restriction (DR), caloric restriction (CR), and time restriction (TR)—and alluded to a recent review in The New England Journal of Medicine (NEJM) highlighting TR (i.e., when you eat and don’t eat).

(We also just released an AMA exclusively on fasting.)

A staple of the NEJM is research relating to treatments for disease, and how they can be used in the clinic. The review article from the NEJM—“Effects of Intermittent Fasting on Health, Aging, and Disease”—is no exception. After describing the potential mechanisms underlying its beneficial effects—metabolic switching and stress resistance chiefly among them—the authors go on to describe the reported effects of so-called intermittent fasting (IF) on health and disease, and finish with recommendations for how health care providers can help their patients implement this eating pattern to treat or prevent a number of metabolic maladies.

(I say “so-called” IF because I think this term is confusing and inaccurate. The authors conflate three distinct forms of fasting into one in their definition of IF: Alternate-day fasting (ADF), 5:2 intermittent fasting (5:2), and daily time-restricted feeding (TRF). This is confusing because these are quite different forms of fasting or restriction. It’s important to distinguish between them. The other problem within these definitions is the flexibility around the term “fasting.” Most studies on ADF and 5:2 allow up to 700 Calories per day on fasting days, while others don’t allow any Calories. I’m very particular about the definitions because I think different forms of fasting and restriction may have different physiologic effects, and I worry that by lumping all forms together, we may dilute such insights.)

One of the aforementioned mechanisms, metabolic switching, refers to the preferential shift from the use of glucose as a fuel source to the use of fatty acids and ketone bodies. Your ability to toggle back and forth between these two metabolic states is an indicator of your metabolic flexibility. For those interested, I’ve discussed this concept in more detail on the website (here and here) and most recently with Iñigo San Millán on the podcast.

The authors question how much of the benefit of IF is due to metabolic switching and how much is due to weight loss, which often, but not always, accompanies fasting. They point out that many studies have indicated the benefits of IF are independent of its effects on weight loss. They also emphasize that ketone bodies produced from fasting are not only used for fuel, but are also signaling molecules that have profound effects on metabolism and are known to influence health and aging.

What is the level of ketosis one must reach in order to elicit these benefits? The minimum threshold appears to be about 1 mM, at least in mice. This raises the question of whether different types of fasting achieves levels above this threshold. According to the authors, plasma ketones generally increase to 0.2-0.5 mM during the first 24 hours of fasting.*

* In the review article, the authors wrote, “In the fed state, blood levels of ketone bodies are low, and in humans, they rise within 8 to 12 hours after the onset of fasting, reaching levels as high as 2 to 5 mM by 24 hours.” We wrote to the corresponding author questioning this and he confirmed this was an error in the paper and he’s submitting the change to the editor.

Given the importance put on ketone bodies by the authors, it’s surprising that very few trials in humans measure plasma ketones during IF, TRF, or ADF. But if it takes 24 hours for ketones to reach up to 0.5 mM, and the beneficial threshold is 1 mM, it suggests there may be a dosing problem with some forms of fasting. This raises the possibility that the metabolic switch isn’t occurring in some types of TR.

In one trial that did look at morning ketones after 4 days of TRF (18-hour fasting window) in 11 overweight adults, levels were 0.15 mM. In a different trial, a one-week 5:2 IF regimen in 19 healthy female participants, in the morning after the full (and non-consecutive) days of fasting (about 36 hours of fasting), self-measured BHB levels were 2.58 and 1.14 mM, respectively.

However, these studies are very small and lasted a week or less. Whether the results are reproducible and what happens when people engage in these regimens long-term hasn’t been determined. Do people improve their metabolic flexibility during long-term TRF, IF, and ADF? Do the levels, and utilization of, ketones increase over time? Does one regimen work better than the others in this regard? Does a ketogenic diet mimic the effects of TR (or does TR mimic the effects of a ketogenic diet)? If we have any chance of sorting this out, we need to start carrying out more (and longer) studies that look at different forms of restriction and measure the thing that presumably matters, in this case, ketones.

Disclaimer: This blog is for general informational purposes only and does not constitute the practice of medicine, nursing or other professional health care services, including the giving of medical advice, and no doctor/patient relationship is formed. The use of information on this blog or materials linked from this blog is at the user's own risk. The content of this blog is not intended to be a substitute for professional medical advice, diagnosis, or treatment. Users should not disregard, or delay in obtaining, medical advice for any medical condition they may have, and should seek the assistance of their health care professionals for any such conditions.


  1. I have to believe that the time necessary to begin seeing the benefits of IF varies DRAMATICALLY depending on your lean body mass/weight ratio. Example:as a 6’1 200 lb male with 6% body fat I likely exhaust glycogen so much faster than a female half my size with twice the body fat that my body would perceive starvation much much sooner.

    • William Byrd, glycogen is stored in the muscles, not the fat mass. It looks to me like you are bragging about your body composition. It doesn’t matter how much body fat a “female half your size” carries, with respect to glycogen storage — the only reason for you to point out body fat here is to brag about how lean you are.

      • Cant explain why. But he is actually right.
        My guess:
        – With low BF the body is used to have less food
        – With low BF the person are used to eat less and have glycogen store that may already be deplected a little.
        So I guess with low BF the metabolism is already used to burn fat and produces ketones much faster, if not all the time already, even if the glycogen store are not empty.

    • This is a really good point and something I’ve been thinking about. If we had a perfect test for measuring autophagy, would we see different levels of autophagy for different people given the same fasting time? I have to believe that the leaner you are the more autophagy you would experience.

  2. Another great contribution. I suppose like many things this recent study perhaps raises even more questions. My biggest question is a TRF regimen of like 16:8 or 18:6 which is what I have been doing in recent months still hold value besides the inevitable calorie restriction that inevitably follows. Or is TRF essentially IF lite ? There are benefits to physical and mental health/aging/weight loss but how much and is the true value really only when you fast for over 24 /48 hours ?

  3. Maybe Peter should consider sharing his historical data of changes in ketone levels under the different fasting regimes. He has done them all and must have interesting historical data. Granted it is only a one person study, but it would be interesting to see a summary of this.

  4. Since my comment above I have read the full article from the NEJM. It’s a very interesting read and refreshing to see such proposals and recommendations being published in NEJM focusing on prevention as much as treatment. Hopefully many more studies will result and even greater focus on dosing, timing, etc. The article contributor certainly seems to have strong beliefs in the power and positive influence on overall health and lifespan of different forms of fasting including those falling into the categories of TRF and ADF (Attia’s 3rd lever of timing restrictions seemingly the point to explore with greater intensity in upcoming studies). What is clear is that there are clear benefits to DR, CR, and TR….the extent of those benefits to be more closely scrutinised. As is probably evident in my comments, I am neither a health profession nor a medical student, but just a fascinated individual by the knowledge we are potentially on the brink of discovering, and how it could benefit my life, the lives of those I love, and the world at large. Any commentary/replies are very much appreciated.

  5. I have been using the IF approach since 2006. Back then nobody would even consider it. It takes time to adapt. I think it took 5 years for me to get stable energy throughout the day. I was overweight until 17 and it took a while to reverse the damage. Now I know that it works because the energy spikes have disappeared. I usually eat when I am hungry and as much as I need. This translates to about 18 hours of no eating (other than black coffee and water), maybe some nuts and a fruit in the evening and a BIG meal at night (usually is protein and fats plus greens). I do eat carbs now ( some fruit, rice and tubers, no pasta or bread) but in moderate amounts. Even if I eat carbs at night, 12 hours later the ketostix check shows purple. I have to note that I exercise four times per week and there are days that I feel I do not need to eat as much (it can be less than half of my usual days).

  6. If you exercise in a fasted state (i.e. 14 hours into an 16 hour fast) I assume that would increase blood levels of ketone bodies more than just TRF on its own?

    • It doesn’t appear that way. Glucose levels, in the blood (of course), rise with exercise. I assume this is due to signals that tell the body to make glycogen to replace what the body is consuming in exercise.

  7. I do the Valta Longo Fasting Mimicking diet for 5 days each month.
    Today on day 4, Ketones are 3.9 & blood sugar 2.8. I don’t do it for weight loss but for autophagy & regeneration & help with autoimmune disease. I have managed to reduce medications for Rheumatoid arthritis from 4 to 1. Just N=1 but works for me. In-between FMD I do LCHF mainly beef/fish & greens with tasty recipes & eating within 10 hours.

  8. I have been consistently puzzled by the time it takes for ketone levels to increase. I had been doing TRF with fasting windows of 18-20 hours daily for a couple of months before I purchased a ketone monitor. What I noticed is that not until about 36 hours fasting did my ketones consistently increased above 1. More disturbingly, I noticed that my fasting blood glucose have been consistently high. Whereas before starting TRF my morning glucose was occasionally abnormal (in the 95-108 range, not quite enough for diagnosing diabetes), 2 months into TRF my fasting glucose is above 100, and on one occasion up to 130!. I’m tempted to order an insulin level and HgA1c for myself, as I have been reading a lot about “functional insulin resistance” (sounds like an inappropriate addition of an inappropriate word onto an inappropriate name for a problem). At any rate, I now got into the habit of doing a 3-day fast on the first week of every month, and well into the 40-48th hour I start getting ketones of 1.2-1.3, with glucose in the 80-85 range. I enjoy my prolonged fasts, but I don’t think it is practical or safe to do those more than once a month. Mostly puzzled; would like to understand better.

Facebook icon Twitter icon Instagram icon Pinterest icon Google+ icon YouTube icon LinkedIn icon Contact icon