One of the most controversial and interesting topics this week in nutrition news is the new intermittent fasting study published out of UCSF by my friends, Ethan Weiss and Dylan Lowe, which was featured in the New York Times. It showed a negative or null result in terms of difference between intermittent fasting and a standard 3-meals-a-day control diet. I'll be covering this in depth.

In other news, I do want to say hello from Montana. I've been spending the past few weeks here and I'll be in Montana until mid-October-ish. I've been very much enjoying the Montana hospitality, and it's a great change of pace from California. Highly recommend if you have the opportunity, which I know I'm very privileged to have, to be able to see the country a little bit and get some fresh air. Again, it's tough to be sheltering in place for six or seven months plus, and for the foreseeable future. So take the possibility to go outdoors, even the local park and take advantage of the freedom to be able to get fresh air and get that sun, get a little bit of exercise, and reset our mind.

It's not great for our psychology, our morale, our spirit, if we're just locked in our box for six months in a row. It is essentially a form of jail. So of course, be safe, do everything that's legal within your jurisdiction or neighborhood. So don't voluntarily sign up to be a caged chicken, be a free range chicken.

Tune In:

Apple Podcast:

Please subscribe, like, and comment to help our content rank better on those platforms. It's really the best way you can show me and my team support.

The Transcript:

So the first topic for Free Fatty Friday before is essentially a coalition of 50 plus questions that I've received over the last few days on this new UCSF study titled “The effects of time restricted eating on weight loss and other metabolic parameters in women and men with overweight and obesity: The treat randomized clinical trial". First author Dylan Lowe, and the corresponding author is Ethan Weiss. Let's talk about it!

I know there's different variations asking why is this a null study? Is it a valid study? Let's break this down. So I'm going to give myself five minutes on my timer here, but I'll probably go over. I mean, it's going to be a disservice to Ethan and Dylan to just try to jam a breakdown within five minutes, but I'll give it my best effort and we'll be speedy here.

So what are my thoughts? A lot of the New York Times headlines and all this stuff basically say, intermittent fasting doesn't work. Is that the conclusion to take away here? And the answer is definitively no. And I'll tell you why that's an over extrapolation of the result of this study and the limitations of what we can extrapolate and conclude from this study. Now, before diving into why it makes such a strong statement here. I do want to say that I've gotten to know Dylan and primarily more Ethan Weiss, in San Francisco, over the years. And, I think they're great smart individuals. And I think they've done a really great service to do human studies on intermittent fasting. I would say that within the whole corpus of literature, there's a lot more animal models, rat models, rodent models, and a lot fewer human studies and being involved in some human studies myself. I know it's very labor intensive, very hard to coordinate and organize a hundred plus people to do anything, especially over 12 weeks. So kudos to the team to pulling this off, it's really a testament and a huge effort. I know that I think reading some of the background, this took over a year plus of work to pull together into a nice little paper that we can all comment and pontificate about.

So do want to credit the team for contributing a very valuable piece of data into the corpus, enough with the pleasantries. Let's talk about the actual study. So what did the study actually compare? It compared a 16/8 which is a 16/4 fasting window with an 8 hour eating window. And this was a relatively late eating window. So this was a noon to 8:00 PM eating window and comparing it with the control group that does that study titled Consistent Meal Timing, CMT, or the control group, which it didn't specify exactly what it was in the paper, but I did correspond with Dylan Lowe the first author on Twitter. And he said that basically they prescribed a protocol of having their first meal between 7:00 to 11:00 AM. The second meal between 11:00 AM to 3:00 PM. The last meal put between 4:00 PM to 10:00 PM. So three meals a day, plus add a little bit of snacking, which means just whatever snacks in between comparing that to a 16/8 between a noon 12:00 PM to 8:00 PM.

And no recommendations on food choice, nutritional coaching, exercise, which I think is going to be an important point that I'll bring up later. And another interesting point again, is that this has done on overweight population. So people that are not necessarily living their best life in terms of  food choices and exercise. Now immediately from there, I think there's a very important question in terms of, what is the consistent meal timing protocol exactly? Now the time when it was a very, very broad. So the intermittent fasting group, it's a very clear 16/8 so noon start 8:00 PM and eating, but with the CMT group, the control group, you can have your breakfast as early as 7:00 AM, but some people might have their breakfast closer to 11:00 AM.

So I would love more clarification on is, what exact time window are we really measuring with the CMT group? The worst, like crappiest version of this is 7:00 AM to 10:00 PM, which is a 15 hour eating window, which implies a 9 hour fasting window. But what if I did a, the most optimal or most compressed any window within that broad range? 11:00 AM to 4:00 PM. If I get all my calories in first meal and 11, 10:59 will give me an hour. So we'll just say 10:00 AM. And then I finished my meal at starting at 4 and in dinner at 5, that's actually a 7 hour eating window. That's a 17 hour fasting window. So that's even a longer, fast than the intermittent fasting.

So let's take the average per say and again, I would love more clarification on this. The average eating, whenever we start at the midpoint of each time slot would be 9:00 AM to 7:00 PM, which implies a 10 hour eating window, therefore, a 14 hour fasting window. So just from the protocol mythology level, what are we actually comparing? Are we comparing a 16 hour fasting window with a 14 hour fasting window? Are we comparing a 16 hour fasting window with a seven hour fasting window? Are we comparing a 16 hour fasting window with a 17 hour fasting window? So there's a lot of variability in that constant consistent meal timing. So I think that's going to be a little bit hard for me to really pull out.

So that's a critique number 1, like we just really need to get the precise with exactly the fasting protocols or eating protocols on both arms. And then one of the things I don't think I saw a lot of coverage on was the activity levels, and this is very, very interesting. So this is not in the abstract, so you actually need to dive into the paper, but in the results section, there's actually an ordering, which is a wearable ring that tracked activity and sleep. Now the 16/8 group had 2,500 steps less. So, before the intervention, both groups had about 8,800 steps a day. The CMT group, the control group, after that window only decremented 250 steps a day. Now the fasting group lowered their activity 2,500 steps lower. So equally valid conclusion one could take here is that a 16/8 fasting window in a population of overweight obese people with therefore likely not the best exercise habits, exercise less when they have a tighter eating window. And therefore, maybe that's an explanation of why there wasn't a huge statistical difference in the weight loss because we lost energy expenditure by taking less steps. So I would love to understand and have a conversation around why exercise is not more controlled. I think it's a nuanced conversation, and I understand that like the researchers want to just focus on one specific variable to control, which I think is a fine choice, but now we see that one of the end points or secondary end points was activity level.

And that might explain actually why you see non-difference, non-significant difference on weight loss because you saw statistically significant reduction activity or steps taken. That would be an area to unpack a little bit more because in practice in real life, when people actually adopt a new diet and new protocol to increase their overall health, we're not just saying, “Hey, just compress your eating window and eat your standard American crap diet.” We're saying, “Hey, clean up your diet and potentially exercise”. And I think the exercise component is pretty critical because I can imagine if you're an overweight, obese person, you don't eat until noon now. And that's like really, really scary for you. Maybe you don't take that morning walk. Maybe you just like, sit on your couch to watch TV for that entire morning period because you didn't have breakfast. So to me, that is an important confounding variable or explanation that we should explore more. Did we just essentially make a discovery that a tighter eating window in an overweight obese population made that cohort less likely to do activity and therefore explaining the null result in terms of the two different cohorts of, in terms of weight loss. It's important to point there.

And then another point that I want to just flag is that typically, and talking to and hearing Dr. Jason Fung, who is probably one of the most experienced clinicians that actually prescribed fasting to resolve a metabolic syndrome, he talks about fasting at 24 hour periods. And this is something that I think is an important scientific point here, which is that when I talk about fasting, I don't like time windows as a measure of how effective or the efficacy of a fast, what do I count as an effective fast? To me that standard is simple, it is being in ketosis as you go from glycogen depletion, as you fast and effective fast to me is elevation of beta-hydroxybutyrate or ketone levels. That signal shows to me, you've fully depleted your glycogen and now converting your stored fat into ketones. And that is the marker of an effective fast. And again, for overweight obese people, 16 hours is not going to be enough likely to enter ketosis. So what I'd love to have measured at the end of that 16 hour fast was their presence of blood ketones, because if there's not well, the fast wasn't long enough. So you're not really getting the true numerical qualitative bio marker for a fast. So again, from the clinical experience with Dr. Jason Fung, he typically prescribes 24 plus hour fasts. And that is a much more likely window in terms of actually seeing ketosis in a person. So to me, it's basically, we're talking about eating windows, which is one distinct point, but from my perspective, I would say more quantitative perspective, the point of a fast is to get into ketosis and for not actually hitting that end point, what your fats is not long enough.

Now you could also get into ketosis by keeping the same fasting window by restricting carbohydrate. So going on a low carb ketogenic diet or stacking exercise, right before you break a fast, so 16/8 can be effective, but now we actually need to understand the systems behind why a fast works. So if you do a low carb ketogenic diet, you do high intensity interval training that the depletes glycogen before you break a fast and you can do a shorter fast like a 16/8 fast, you will likely get into ketosis. And I would bet that that cohort would see a lot of benefits. So those are some key points that really explore the nuance here in terms of what we can really conclude from this study. And my takeaway here is that for an overweight obese population, that's eating a standard American crap diet, a 16/8 eating window, especially one, that's not a matching circadian rhythm, which was shown by Satchin Panda, you want to scoot up your eating window. So a late eating window, relatively small intermittent fasting window without changing macros, without changing exercise is likely not going to be that effective. So I am AOK with that conclusion. That's not really a great protocol, but to make any other claim that intermittent fasting doesn’t work, it's bad, it's BS. There's no data that I can see from this paper that justifies that. Now, we went over time. let's see, I probably went to almost 2 questions worth of time here, but, and I can answer more questions if you have them. But I think this is important topic it's really, really timely, and we should be very scientific and very precise what we extrapolate. And of course, I mean this feedback while maybe direct or blunt, but really in the best of faith.

And I hope that Ethan and Dylan can incorporate my thoughts and I would love to hear their feedback. So to me as a scientist, I want to search for the truth. If there is really compelling evidence that fasting does not work or will decrease my lifespan, I will stop doing that. But this study does not do that, not even close to it. And we need more work here. So within that study, I already emphasized a few different areas that future researchers and hopefully Ethan and Dylan can take this feedback and incorporate into their next work and help us really understand a holistic framework of how to apply fasting into an overall health and wellness program. So great work UCSF, great work, Ethan, great work, Dylan, keep up the awesome work. I want to learn more with you guys, and that's it. That's probably closer to 15 minutes now, but hopefully well worth it.

So question number two today is from Cooper C:

Cooper asks, how long does it take to create cellular autophagy?

Now I'm gonna start the timer and actually stick to it this question. This is a good question. So this is actually an open scientific question. Well, why is autophagy an interesting topic? Well, this topic actually won a Nobel prize in 2016 by Japanese scientist to Yoshi Mario Asumi. And he based it to find the core mechanisms of what autophagy is and what autophagy is auto self-eating phagy. So the cell eating itself, it's essentially a mechanism in which a cell can clean up it's damaged organelles, bring these damaged organelles, break them down, turn them into reusable parts to reconstitute, healthier, fresher, organelles.

So that was discovered, and it won the Nobel prize. And the context in which it's relevant to today into metabolism is that a lot of people talk about fasting, intermittent fasting as a way to trigger autophagy and people say, “Hey, do fasting and get autophagy. You'll be younger, you'll live longer.” The question is, how long does it actually take to create autophagy? How long do you need to fast to take autophagy? And that's a tricky question because not all cells, not all organs are alike. So actually, if you ask the question more precisely, autophagy is very cellular or organ specific or tissue specific. So for example, the liver muscle pancreas, those organs have very different thresholds  for what kind of nutrients sensing activates autophagy versus for example, the brain. So the way to look at this on a specific technical level is actually opening up a cell and looking at lysosomes and seeing you staining the organelles and seeing which organelles are being transmitted where. So that is the specific way to actually detect that. But to measure this from a biomarker or a blood marker, that's not known, and that's an open scientific area of research, what are good biomarkers that actually correlate to autophagy? That's an open science question. So if you actually discover that you could probably win the next Nobel prize  you can actually basically get a blood marker for autophagy.

So that does not exist. That's an open scientific question, although there are people working on different correlates. One of the interesting concepts that I think again, needs to be further articulated is organ specificity or tissue specificity of a lot of these concepts. And we've talked about this in terms of Mtor, Mtor is on they're very popular buzz phrase in which Mtor is activated when the presence of protein or leucine and that activates muscle growth. And one of the popular discussion items is can we mitigate Mtor therefore reducing growth, and that is potentially helpful for longevity and anticancer and impact, but do we just want to inhibit Mtor forever? Well, it's not that simple because you actually want to maintain muscle and grow muscle. In fact, presence of muscles, one of the long is one of the better markers of longevity that functional strength. It's a  sink for glucose. It's it protects your organs and bones from being damaged. So just like you want tissue specificity of the activation of Mtor, the same story holds for autophagy. So what are my best guesses here? Again, I believe that ketosis just like the first question is likely important threshold to trigger more autophagy. Again, it just makes sense from a mechanistic perspective when there is a sensing of a lack of nutrients, it makes sense to activate some of these longevity and preservation pathways. So to me, a potential analog is to make sure you're actually getting to ketosis, nutritional ketosis, or ketogenesis.

That's likely a good blanket way to likely get into top of G, but to measure each specific cell  you gotta just like open it up and actually dissect, you know, where are the organelles, are organelles being broken down in the, in the vesicles, in the lysosomes, are they being transported around? That's a precise way to measure it, but obviously we're not going to pick out the billions trillions of cells in our bodies and measure every single one. Is there a more generic bowel marker that might guide or correlate to an organs levels of autophagy? That's an open area of research, and having to make a best guess in terms of what are some thresholds considered that triggers autophagy I'll look at it. And then nutrient sensing pathways. So Mtor, AMPK, Insulin. These are likely correlates in terms of having that natural nutrient sensing, detecting, lowered AMPK or lowered Mtor, lower insulin likely will drive autophagy.

So those would be the things I would look for. So if you're in ketosis, you're probably upregulating your autophagy pathways, but I cannot get more specific than that because it's an open scientific question and that's five minutes. And now we're onto question three. And the question is from tiredlookingforname, that's a great name.

And he/she has a question, if I am doing OMAD, keto or carnivore style, what is the best strategy of timing, the meal and workout. And could it be goal specific, for example, gains autophagy, etc. or type of workout, specific strength or cardio.

Now I'm going to start the five minute timer. Tiredlookingforname, this is a great question, because it touches upon the complexity, the nuance of personalizing a protocol, and that fits your personal baseline and your personal goals.

And this is again, the systems engineering approach towards things about human performance. And I think it's great that you start understanding the lovers to ask this question, which is that there's time restriction, the one meal a day, there's the diet, a macro restriction or carnivores the food choices that you're thinking about. And of course, the caliber work amount that you're also consuming, and these are the three possible lovers in terms of a diet. And of course you tend to put on activity, load or exercise bouts as a additional lever to manipulate. And these are all the core inputs in towards an anabolism, cannibalism and gains autophagy of these things that are downstream from these inputs, for every day energy expenditure and consumption. So I'll caveat my answer at the start by saying that there has not been much if any human randomized control trials on any of these specific protocols. Again the human research on intermittent fasting is relatively scarce and relatively basic. So OMAD and carnivore and stacking workouts is very, very cutting edge in the sense that a lot of these experimentations and the anecdotes that I'll be talking about are happening at the elite performance levels, the elite athlete levels that are less so focused about general population and less interested in trying to prove out in a randomized controlled fashion, does a set intervention work. This is more a button elite athlete, elite performer, eating the best gains for their specific, very, very niche athletic or performance use case. So what I like to do when I'm doing an OMAD protocol, a one meal a day protocol, which is not my default protocol, but I've done blocks of OMAD as I like to do the workouts right before breaking fast. So that sort of necessarily means and typically how I do that is I either have when I wake up and I'm doing one meal a day, have like a primary meal, which is like a large late breakfast.

And then basically do a heavy workout right before that meal. Of course, if you're doing a OMAD towards dinner, then it's literally like, after work, you know, f4:00 or 5:00 PM, do that heavy workout and then eat right after for your one meal a day. And the thinking there is that as you go on a fast, you start depleting glycogen, you start oxidizing fat, you start creating ketones, and then at the very tail end of that fast, as you're dipping into ketogenesis and ketosis you create a further demand on your fat reserves, further demand on your glycogen reserves and really deplete glycogen, and really ramp up a fat metabolism and ketone metabolism. So it's basically you were done your reserves and then you really, really demand it. And then right after you have full repletion of nutrients. So I liked that protocol in terms of pushing, forcing outpatient and recovering.

But I could see very similar arguments to be made for doing the workout after a meal, and then fasting after that, that might be better for it for autophagy if we kind of think about the mechanisms where you have less money, nutrient availability, especially after an acute exercise bout that might actually trigger a more recovery autophagy pathways. But the concern there is that you might be going to catabolize or break down and not give yourself the nutrients to fully recover after a heavy workout. So you might gain potential better autophagy benefits, because there is no nutrients available post heavy workout, but you'll likely to tap lies your lean muscle tissue, which is not ideal, but he might get potential autophagy longevity benefits. But in terms of what I like to do, which is I liked the autophagy benefit of just doing the fast, but I do care about maintaining and gaining and having an anabolic process and feed myself in the anabolic window after a workout. That's why I tend to like to do. I think one things the concern about, especially for folks focused on performances, do a light, maybe pre-workout meal with a little bit of carbohydrate before that workout. So you can imagine, you do your 24 hour fast, or OMAD right before your workout. You basically break fast with a little bit of a snack, little bit of carbohydrate and fat, right before the workout. Do you have your workout where you have a little bit of that pre workout fuel and which would ideally boost the overall work volume that you could do during the workout and then have your full meal. That would be a potential tweak in terms of focusing more on performance or more on gains in terms of workout specificity. I know a lot of athletes like to do fasted workouts for cardio athletes, really tests or fat metabolism really adapt into ketosis, but usually this is done in preseason, basically leaning up burning fat getting into the optimal body composition before doing fed workouts for strength, again, ideally you probably want to do a little bit of carbohydrate before a heavy strength workout to gain maximal adaptation, but of course, you're cutting short. Well, let me excuse me on that. The essentially having a little bit of fuel will allow you to have a potential bigger work volume work output, but of course, that tests your fat adaptation because you're starting to break your fast a little bit earlier.

So I'm over time now. But I think the main takeaway is that experiment a little bit and understand the concept of why you're choosing a meal at the beginning or the end of an OMAD. Why are you doing keto? Why are you doing carnivore? Is there a role of carbohydrate before workout or after workout to gain optimum performance? Or are you doing this for autophagy where you're just really trying to stress your acute agenesis, your fat metabolism. These are all individualized personal decisions to optimize and micro optimize little aspects of your protocol, but overall, hopefully this gives you a sense that think about how you're piecing this all together. Happy, go into more specifics. If you tell me about your baseline and your specific goals, and you're trying to win an Olympic powerlifting meet that's one question, if you're older and your own mind to make sure you're not going to metabolic syndrome, that's another question versus trying to gain muscle and avoid sarcopenia. A third type of question. And our last question for today's Free fatty Friday is from Dan Mauney.

He asks, Can you use the G/K ratio such as 107/.6 to obtain an insulin level. Can a table be extrapolated?

That the physicist in me is looking at the unit difference and cringing a little bit in terms of 107 is a unit of glucose milligrams per deciliter, where 0.6 is a popular a unit for ketones beta-hydroxybutyrate, which would be millimolar.

So it doesn't matter though, the ratios are all the same, just the units are off, because millimolar, and milligram per deciliter are essentially convertible, but well answer this questions. Let me put on the timer. Let's go!

So this question teases out a concept that I think bears more understanding within the physiology human metabolism community, where snapshots of data, or I take a finger stick and get a glucose and ketone reading or an insulin reading for for that fact is a valuable piece of information, but it's a very static or very primitive version of understanding what is going on in a very dynamic flowing system. And what I mean by flow is that there's constant conversion and utilization and turnover of all of these metabolites, glucose is being released and consumed and stored, constantly ketones being produced, transported and an oxidized and mitochondria insulin, is released and then signals and is broken down.

So a better picture that I think we will as a community, as a group of scientists or enthusiasts in the space of metabolism will better understand is that we should really be thinking and reasoning about these things as dynamic flows or fluxes. So can use the GK ratio to obtain an insulin level? The answer is no, because it's not a complete enough system to predict. And how, I mean, that is that insulin is a response primarily to nutrients, primarily glucose, and there is a time delay or a timeline from insulin release and insulin consumption and utilization to, uh, glucose going up and down, and also ketones going up and down. So one cannot predict from 107 and 0.6, uh, glucose, the ketones, if that is at the start of a fast or at the end of the fast. What I mean by that is what if I'm in ketosis? So I have 0.6 millimolar ketones, and I start drinking a soda. So I boost my blood sugar and it starts rising. Insulin will be a tail end, lagging indicator of that. So my insulin will be low and just as my glucose is going up, or, and that would look equivalent in terms of just three snapshot data points as being at the tail end of a fast, or having some metabolic syndrome where I have relatively high glucose. I have some key ketones in my system and, I'm at the tail end. I'm not controlling my my blood sugar that well, and I have a high ambient insulin.

So that's a subtlety that cannot be captured by just three simple data points. And that makes sense in the sense that we can get a general good understanding from a very simple system, but the simple system cannot extrapolate and paint the overall more complex. Uh, it doesn't fully model or capture information in the human body, which stands to reason, right? Because the human metabolism is obviously more complicated than three specific timestamps on three specific bio metabolites. But that said, can we get directional understanding, a directional extrapolation from a GK ratio? And the answer is yes. So we can not necessarily pav a hundred percent causative predictive ability from a GK ratio to an insulin, but we get a general good direction in the sense that if you have low blood sugar, which I define as less than a hundred milligrams per deciliter, ideally in less than 95, 90 milligrams per deciliter and above 0.5, millimolar ketones a beta-hydroxybutyrate that likely implies a low fasting insulin level.

And if, of course, if you have a very poor GK ratio, meaning very, very high glucose and very, very low Q tins, that likely implies either a static spike in insulin, which is course as normal, cause you need insulin to process. And basically activate glute for bring glucose into the cell or, and then it's normal, or it's someone with a metabolic issues with poor glycaemic control and you might have high facet insulin, but in general, if you have a lot of snapshots where you have high GK ratio, it likely means double check and start looking at your, uh, like the curve of your insulin. If you have more of your time when the low GTK ratio, meaning low blood sugar, high ketone levels likely your nutritional ketosis, and you probably have a better or a healthier or a lower facet insulin level.

So hopefully this gives you a more complete answer, which is that GK ratio, good directionally to extrapolate and predict what your insulin could be like. But of course it really depends on the context. Is it the perfect model to fully capture that dynamicism complexity of human metabolism? No, but it's a good tool. So I do like the GK ratio, but let's be careful in terms of over extrapolating and overextending, what it can really tell us about the system, hopefully this answers the question, happy to clarify. I know I went in a little bit into the physics world, but I think it's helpful to make sure that we get the framework before going into the takeaways.

And that's a wrap for this week's episode of Free Fatty Friday. Man! These were some great questions and I went over time, I think on every single question. So I hope you do excuse me on that, but these were really great technical, deep questions that deserve that extra attention. If not 10 times more time. So thanking for letting me help answer your questions and learning with me, this is an awesome exercise for me. It's been great to tinker and think about all these edge cases and questions that you guys are surfacing up and brings to me, uh, really helps push my research, my study, my thinking, and I think a very wide and disparate and fun way. So please do keep those questions coming. It's a lot of fun and I always appreciate your support.

So if you enjoy this episode, please do give us a like subscribe five stars, all those great things that is really the best way to support us and me. I mean, I'm doing this really as a hobby. So your feedback is really like the core of motivating thing for me here. So there's a lot of fun and I'll see you guys next week as always stay healthy, be resilient, digging into research and think from first principles.

All right, this is Geoff Woo and I'm out!