Greetings Ketonians! Dr. Anthony Gustin writing to you today.
Many of you know I created Perfect Keto to help ketosis be accessible to everyone who wants it, and that includes me. One of my favorite quotes from Fr. Peter Drucker is,
That’s why I encourage everyone on a ketogenic diet to test their ketone levels. I’m taking it about 10 steps further than just ketone blood tests by conducting an intensive two-month ketosis experiment on myself to learn precisely what changes occur in my body in ketosis. I began with a four day fast and I’m pleased to share that the results are in! They truly rocked all the old ideas I had about fasting and ketosis and I can’t wait to share them with you.
If you think I’m nuts and wondering why the hell I would fast, start by reading my article about why I am choosing to fast here. The reason is I wanted to maximize my transition into ketosis. I’ve tracked body composition, lab blood work, ketone levels, microbiome tests and other noteworthy biomarkers to answer the questions I had for myself.
Does fasting lead to massive lean tissue losses?
Does it destroy your metabolism?
Is it unbearable?
Does it sap your energy?
All of these questions and more are answered in this article.
The Results of My Four Day Fast to Start A Ketogenic Diet
Since there’s about a 0.003984% chance you read linked article about fasting, let me explain something from the start, this fast was never going to be a water fast. There are a few reasons to not do a water fast, but we won’t get into that here. We’ll get into what type of fasts I’m now doing regularly and why.
My fast was similar to what researcher Valter Longo has shown to actually be preferential to a water fast in many ways, in what he dubbed the “fasting mimicking diet.” Much of his research indicates that “fast mimicking” provides all of the benefits (and sometimes more) of a water fast while minimizing a vast amount of downsides.
His official fasting mimicking diet has fat as a macronutrient at only about 40% of total calories and closer to a Mediterranean Diet macro distribution. I wanted to maximize transition into a ketogenic state, so I used Valter’s caloric load (500-750 kcal) with a classical Johns Hopkins ketogenic distribution of macronutrients (85% fat, 10% protein, 5% carbs).
Ok, there’s some “how” of fasting, back to a quick recap of why. Again, I wrote a longer post about this, but here are the top reasons for me personally to incorporate modified fast as a regular part of my life:
Purging of cancerous or precancerous cells (a.k.a. reversing cancer)
Improved immune system performance
Increased lean tissue mass (possibly from vastly increased HGH secretion)
Decrease in fat tissue
Rapid shift into nutritional ketosis
Increased gene expression for longevity
Autophagy and apoptotic cellular clearing/repair
Decrease in oxidative stress and inflammation
Increase in mental performance and BDNF (brain stem cells that create new neurons)
The Glucose-Ketone Index While Fasting for Ketosis
I was initially going to do a three-day modified fast simply because three days sounded nice, but after reading the research of Dr. Thomas Seyfried, I was motivated to take this until I reached a glucose-ketone index (GKI) of 1.0 or under. His research is evidence that most of the therapeutic effects of fasting to correlate with getting this ratio between glucose and ketones under that specific level.
Measuring the Glucose Ketone Index
If you want to measure the glucose-ketone index yourself, it is pretty simple. Take your glucose readings (usually in mg/dL if you are in the US), divide by 18 to get glucose in mmol, then divide that number by your ketone (mmol) number you got from your blood BHB ketone reading. There you go. Dr. Seyfried thinks the lower this number, the better, shown through data here.
Example: If glucose = 81 mg/dL and ketone = 2.0 mmol then,
81/18 = 4.5
4.5/2.0 = 2.25
GKI = 2.25
My goal was to track glucose/ketone levels multiple times per day and then body composition and lab values (to see changes in lipid panel, inflammation and immune response) both at baseline, at the last day of the fast, and then five days after. The reason is that a lot of benefits can be “rebound” effects from fasting so I wanted to see how much I bounce back and the velocity of said bounces. All of these results are posted below. Here we go!
Day 1 Fasting Results
On the first morning, I clocked in surprisingly a little low on my ketone levels even though I had gone pretty “keto” in the week or so leading up to this fast. My hypothesis here is that I probably have been eating too much protein. I’ve been using the Precision Xtra ketone meter for these readings. The Precision Xtra works great, but the strips are priced at a ridiculous amount. You’ve been warned!
AM Fasted Measurements:
Ketone levels – 0.2 mmol
Glucose levels – 85 mg/dL
Glucose ketone index – 23.6
Note on the GKI here — the average American is at about 50-75, and I’m aiming for 1.0 and under. Long way to go. Hopefully this will be the highest number I see for the next 8 weeks.
After taking my initial measurements I went to LabCorps and got a bunch of blood work done for the baseline to this entire experiment. After the labs, I went to get my body composition scanned with a DEXA machine. Relevant test results are highlighted later in the post.
I consumed a coffee and shot of espresso shortly after the bloodwork. Coffee is something I might try to eliminate during this keto experiment, switching to tea for a while to see how I respond, but for the fast I’m keeping my damn coffee. The little things in life, you know.
Mid-morning I consumed 5 grams of BCAA with 1 scoop exogenous ketones and one tablespoon of raw organic coconut oil. The 5 grams of BCAA and exogenous ketones were to provide some easily accessible energy and to help the transition into ketosis.
After 6 hours of work, I had ¾ can of coconut milk and ¾ container of Kettle and Fire bone broth to get some more fat in. That’s all for food on day one, folks.
20g protein / 80 kcals / 10%
70g fat / 630 kcals / 85%
10g carb / 40 kcals / 5%
Total / 750 kcal
My focus was high. This is why I normally eat a keto/high fat diet, though so nothing new to me there. Not hungry or distracted at any point, but also didn’t do much activity.
Relatively little. Went on a few walks to get where I needed to be, did four rounds or so of a movement routine I have hashed out, and some light band work. Intentionally kept movement down to minimize the transition symptoms of lack of energy as the body goes from massive energy input and burning glycogen to burning stored body fat.
PM Fasted Measurements:
These were taken at 5:00 pm. Baseline glucose held steady, however, ketones rose at a decent rate. Still not in a state of ketosis. Will be curious to see the AM Day Two measurements. Seeking a decrease in glucose and increase in ketones.
Ketone levels – 0.4 mmol
Glucose levels – 84 mg/dL
Glucose ketone index – 11.6
Did a cyrotherapy session followed up by an isolation tank session mid-day. Very low stress day all in all. Also very intentional. All in all a pretty normal day. Saved a ton of time and was actually probably much more productive without having to cook/eat/clean at all.
Day 2 Fasting Results
On night one, I was surprisingly much more tired than usual. I went to bed around 8:30p and woke up around 7:30a. As a comparison, my usual sleep time is about 9:30p – 5:30a. The sleep quality was a little like if I were to take NyQuil. Super deep and very groggy if I had to wake up and go to the bathroom.
Upon awaking, I didn’t feel hungry at all, which was shocking to me. Running on ketones already? Let’s see…
AM Fasted Measurements
Ketone levels – 0.9 mmol
Glucose levels – 74 mg/dL
Glucose ketone index – 4.56
A huge jump in ketones and a huge drop in baseline glucose measurements. Definitely trending the right direction here.
Food intake was pretty much the exact same as the day before, except I had black tea, then went on to my cold brew that I so desperately needed. Seriously, the little things when you can’t have real food for days on end matter. Or perhaps I’m addicted to caffeine and coffee. What do I know?
The only difference on day two was that I had the coconut oil with the coffee, and the exogenous ketones and BCAAs while I was doing a mini working out.
20g protein / 80 kcals / 10%
70g fat / 630 kcals / 85%
10g carb / 40 kcals / 5%
Total / 750 kcal
Didn’t want to push things too much but felt so much energy in the morning (that’s ketosis for you) that I decided to do a quick little 45-minute movement routine/kettlebell loaded session. The workout biased upper body pressing and pulling, with some kettlebell RDL and cossack squats added in.
Definitely felt a little hunger for the hour after the workout. I usually notice a giant increase in hunger on days I workout so this wasn’t anything out of the ordinary. Fasting and working out intensely doesn’t seem like the best of ideas, but maybe this is an idea I need to test instead of holding onto untested beliefs.
PM Fasted Measurements
Ketone levels – 1.7 mmol
Glucose levels – 73 mg/dL
Glucose ketone index – 2.38
There come the ketones! Everything was tracking in the right direction.
There were certainly more times on day two of my fast where I noticed the feeling of hunger, but nothing that was abnormal or something I couldn’t solve by using my brain or drinking some water.
What was profound was the level of mental clarity once I was deeper into ketosis and noticing that at the level of about 1.5 mmol and above (I would assume) providing something like a sustained focus that only comes when I’m at my peak in the AM after tons of stimulants and in a state of flow.
I was in a super positive mood for most of the day. Nothing out of the ordinary there, but definitely not in a negative mood being I haven’t really eaten in two days. No hanger. Yet.
Day 3 Fasting Results
Admittedly, sleep was not so great on night two. No cramping or anything weird, just very tired at night and woke up a bunch of times. This led to quality of sleep being diminished and feeling super tired upon awakening. My biggest annoyance in life is feeling tired.
Ketones – 2.1 mmol
Glucose – 71 mg/dL
Glucose ketone index – 1.87
Cranking along as expected here. Nice to see that glucose-ketone index continued to drop.
Since I was feeling relatively normal and still not hungry at all, I started with my scoop of coconut oil on this day three of my fast but didn’t add anything else in. I wasn’t going to have the bone broth “soup” from Day 1 and 2 unless I felt hungry, and I didn’t. Time to speed this train up.
My fasted workout ended up being the same intensity as the day before, with some light lower body kettlebell work, around 40 minutes or so. There was a mix of kettlebell swings, one leg RDL, goblet squats, cossack squats, and lunges. Sipped on exogenous ketones and BCAA’s throughout.
Same as the Day 2 with post-workout hunger (go figure) but went away once I drank water and started working.
A little slower/foggier today than usual but had some work stuff blow up in my face so that may have been the distraction and lack of wanting to do any work more than anything else. No real changes in hunger for day three and I had a realization that FASTING IS NOT A BIG DEAL.
Ketones – 4.1 mmol
Glucose – 69 mg/dL
Glucose ketone index – 0.93
First time hitting below 1.0 on the almighty GKI! This means I’m in prime therapeutic mode, and only have to ride the fast out through Day 4 to sleep overnight in this state as well as wait to collect some markers (more blood work, DEXA scan, etc.).
Day 4 Fasting Results
AM Fasted Measurements:
Ketones – 2.9 mmol
Glucose – 68 mg/dL
Glucose Ketone Index – 1.3
A little bit lower than the PM here, but it is completely normal to have higher PM ketone readings than AM readings due to counterregulatory hormones, like cortisol. The great thing to see here is how crazy low my blood glucose was. I call this fast a success and ready to get my blood work done, so it was off to LabCorps for some data and to get back to eating!
In the waiting room for final blood tests on Day 4 of the fast I realized how ridiculous some people are. A woman was pacing around furiously. This concerned me until she screamed at the lab tech who had some problem with her paperwork for her (very important) lipid panel saying “ARE YOU KIDDING ME?! Are you saying I can’t get my blood drawn? I haven’t eaten since 6 pm the night before and have been fasting since then! I’m SO hungry!”
If only this woman knew… On a positive note, a lifelong lesson for the entry-level lab tech. Don’t mess with San Francisco yoga moms and their morning quinoa bowls.
AM Post Meal Measurements:
Ketones – 1.5 mmol
Glucose – 88 mg/dL
Glucose Ketone Index – 3.25
When I got back from getting my labs drawn I had: 3 eggs, 3 bacons (I call them bacons, not strips), half an avocado and two cups of sauerkraut. Sounds awesome coming off a four day fast, but WAY too much food. Felt uncomfortably full afterward. Will definitely ease back into eating after next fast.
Glucose had a completely normal raise here, but hoping to keep post-prandial ketones a bit higher. See graph below for more information.
Just did a lot of walking, intentionally didn’t want to push my body as it already had all this new stuff (food) to deal with. Felt fine physically.
In between meals definitely felt a little bit lower energy and was harder to focus for a decent amount of time. Not sure if this is normal for me coming off the fast or if it was for the length of the fast. Not unreasonable and still 100x better than if I were to eat a bagel.
After breakfast I was so damn full that I didn’t want to eat until dinner, and even then wasn’t that hungry, but knew I needed some nourishment. Wanting to stay in ketosis, dinner was a grass-fed ribeye topped with Dungeness crab meat topped with hollandaise sauce and a side of kale cooked in butter. Not bad, but too full.
PM Pre-Dinner Measurements:
Ketones – 2.5 mmol
Glucose – 76 mg/dL
Glucose Ketone Index – 1.68
PM Post-Dinner Measurements:
Ketones – 2.0 mmol
Glucose – 88 mg/dL
Glucose Ketone Index – 2.44
This is a pretty good before/after measurement considering how much protein I ate in the meal and how my body hadn’t dealt with protein for a while. Will hopefully be able to hold these metrics and even increase them in the next 7-8 weeks.
The Cumulative Results of My Four Day Fast to Start a Ketogenic Diet
Main data here is split into three sections:
- Ketone/glucose measurements
- Lab work
- Body composition
Most measurements here are taken from Day 0 through to Day 4 and then again at Day 9. Why track after the fast? A lot of the benefits come in the “rebound” effect of fasting.
Summary: Just like when you are working out and breaking down muscle fibers when the going gets tough, you get sore, then your body thinks “hey, let’s not let that happen again” and makes you stronger by building new tissues to deal with the perceived repeated threat. The same thing happens with fasting. You start clearing out old and defective cells and turning on a lot of genes that normally aren’t active and your body has a supercompensation effect after, building tissues and systems back up thinking you’re going to go without food again.
Since the whole point of this fast was primarily to get into ketosis as fast as possible, let’s start with an overview of the ketone and glucose levels I took throughout my fast.
Ketone and Glucose Measurements
You can note on Day 1 how high my glucose ketone index is. Again, the lower the GKI, the better for overall benefits when looking at ketosis and especially for fasting. Day 3 and on is probably why we get a lot of benefits of ketosis. I was “low carb” and not in keto before and I was up in the 20s. This is why you MUST TEST when you are talking about your health. If you don’t get the numbers, you don’t know what is going on. Simple as that. Low carb does not equal keto my friends.
The Day 1 to Day 4 drop of blood sugar and spike in ketones is exactly what I was looking for. This fast was primarily directed at getting into a state of deep ketosis quickly so I used the glucose ketone index (GKI) and >3.0 mmol as an indicator I was where I wanted to be.
Next fast I will prioritize spending time in the therapeutic range of <1.0 GKI.
Post Meal A – This was the egg, bacon, avocado sauerkraut (way too much) meal. Great to see blood sugar stay super low here while ketones not taking that much a hit. Clearly had enough fat to hold ketosis.
Post Meal B – This was a little overkill and definitely too much protein. Had half of a ribeye and about ¼ pound Dungeness crab with hollandaise sauce and kale. Crab is essentially all protein and created a 0.5 decrease in ketones but most importantly the overnight drop from 2.0 to 1.1mmol.
When you see this amount of drop, you know you ate too much protein (It was worth it).
Post Meal C – Another example of eating too much protein. Dinner was two lamb chops, less protein than the night before, and you can see pre- to post-dinner move ketones up a smidge, but you can still see the drop overnight from 1.9 to 1.3.
Lab Results from Fasting
Below are the most interesting inflammation, immune and lipid changes from Day 0 to Day 9. Each dot corresponds to days one, four and nine. You can use companies like WellnessFX if you elect to get your own blood work done.
Most markers were not affected, however two did stand out to me in regards to inflammation and immune function. These markers are high-sensitivity C-reactive protein (hsCRP) and white blood cell count (WBC).
hsCRP is an overall measure of inflammation in the body. As you can see with the graph below I had a pretty marked decrease in this inflammation value over the course of the fast. The starting point of 0.52 is already fairly low, and for it to decrease to 0.42 in just nine days is drastic. I would love to see data like this with someone who has a reasonably high hsCRP and high levels of inflammation pre and post fasting.
WBC count is a general reflection of immune activity. The higher levels, the more active your immune system is. For example, if you’re incredibly sick, you’ll see these numbers jump really high as your immune system properly ramps up. What was interesting to note here is the decrease then rebound past baseline post fasting. This is very likely due to the hormetic effect of fasting. This means fasting is a stress on the body, but a positive stress you can easily adapt to become stronger.
The lipid panel was extremely interesting as well.
LDL stayed fairly constant, and this is about the level I’ve had for years now. Contrary to what your general physician may say, this isn’t a bad thing. What you need to look at with LDL is the amount of large fluffy LDL versus small dense LDL. The type of LDL that is associated with heart disease and arterial inflammation is the small dense kind. You can see this breakdown with a fractionated lipid panel, or an NMR lipid profile. This gives you “particle numbers,” which is the most important reading. For me, this level of LDL isn’t high, as my small LDL particle amount typically reads around 400 and at day one was at 424.
HDL cholesterol dropped a bit, but this was to be expected as HDL is correlated to what you eat. When you don’t eat, it should go down. I was looking for this to rebound nicely after the fast though, which it didn’t. Will be interested to see this value in a few weeks when I measure again.
Triglyceride levels went from a fantastic 78 to almost double at 143, which is not something you want to see. The hypothesis I have here is that fat was being mobilized for energy, as I was burning body fat for energy (proof in the graph above of ketone levels, as well as below with the body composition graph). Nice to see it trending back down after the fast, but again, would like this response to be a little better. I’ll be tracking this again in the next few weeks to see how it has leveled off.
Body Composition Results from Fasting
I came into this fast already about 10 pounds underweight from where I usually am. I had been traveling for over three weeks in Europe a week before I started and was deathly ill for about 10 days of that. Combine that with eating much less overall and not lifting heavy weights and I lost about 8-10 lbs.
To track body composition, I chose reading through a DEXA scan with the company Body Spec. This method is the gold standard for the ability to very accurately localize where your fat/lean/bone tissue is distributed.
If you’ve never used a DEXA scan before, please note that body fat will be about 5-7% higher than other methods. So if you are 20% body fat on a DEXA readout, that means you are about 13-15% with calipers, for instance. Don’t freak out.
Raw numbers don’t really matter anyway. What matters is that you track consistently and accurately the same method and trend in the way you want. Also, if you are going for body composition, who cares if 5%, 12%, or 20% spits out of the machine, the mirror test should be your biggest measure to track! Without further ado,
Whoa! I didn’t wither away into an emaciated heap of skin and bones after fasting for two days. And wait! I actually gained lean mass while losing fat when compared to baseline measurements. What?!
You’ll note the Day 1 to Day 4 drop in lean mass. However, a good portion of that was located in the trunk, aka intestines, aka food. This is the beauty of the DEXA scan that you can’t get from any other measurements. However, the gain of lean mass above baseline was actually bodywide.
How Did I Gain Mass from My Fast?
Why would I gain lean mass from fasting? My hypothesis is the massive surge of human growth hormone (HGH) you get from fasting coupled with the hormetic response that your body exhibits trying to prepare for another bout without food.
The increase in fat of 0.3 lbs from day one to four is negligible, but the reduction of 1.1 lbs compared to baseline over four days is not. The mechanism here is clear. My body had energy demands, and I was in ketosis, so I was using my stored body fat to keep me humming along.
From four days of fasting I gained 2.4 lbs lean mass and lost 1.1 lbs fat mass.
This is a huge finding for me, as I want to protect lean mass that I’ve worked hard to maintain. This belief that fasting leads to lean mass reduction is completely wrong, at least for me. If I can gain 2.4 lb muscle and lose 1.1 lb fat every time I fast, sign me up on the regular. #FastingForGains
What I Learned and Modifications for My Next Fast
Fasting is NOT THAT BAD
I’ve done 1-2 days of fasting in the past, so this was by far the longest fast of my life. That’s pretty silly when you think about it. I haven’t gone more than four days without food in my entire life. Humans probably did this regularly for much longer periods of time, for most of our species’ existence. The silliest part? It wasn’t even bad. I basically felt like I feel on normal eating days, with more mental clarity and focus than I otherwise would have. I was hungry a handful of times and yoga mom hangry zero times.
A friend pointed out that I’m already pretty healthy and eat a pretty low-carb diet to begin with, so this transition for me was probably not the same as someone coming from a standard American diet. I agree 100%. This isn’t going to be a cakewalk if you are addicted to sugar.
These four days psychologically helped me with fasting moving forward. Considering the results, now I think it is insane NOT to fast.
I haven’t been in ketosis as much as I thought
I’ve used my Precision Xtra machine to test the effectiveness of my ketone supplements a good amount, but this is the first time I’ve obsessively tracked WITHOUT the supplements (4-6 times per day).
What I learned here is that even though I assumed I was in ketosis when I was low carb before, there’s no way I was actually in deep ketosis. I will not stop saying it — test, don’t guess. I’ll be tied to the Precision Xtra through my little ketosis experiment to understand how I respond.
Track IGF-1, citrate synthase, HGH and RMR next time
IGF-1 is linked with aging and pushing a few different disease processes forward. That being said, some research points to IGF-1 dipping considerably during a fast, then rising up after a fast, but lower than baseline. What this means is that you giving your body the signal to clear out old cells as you are not making any new ones. It’s also probably pretty good if you want to live longer.
Citrate synthase is an enzyme that indirectly measures mitochondrial health. This is also an area of interest to me that I’ll get into in another post.
HGH (human growth hormone) will be measured to see if this was the cause of the increased lean tissue. A drastic spike in HGH has been reported as a response to fasting, but I want to see it for myself.
RMR (resting metabolic rate) is how much energy (calories) you are burning without exercise in a day. Did this go up or down? Did it not change? I’m interested.
Reduce calories and optimize for GKI
The day where I only had coconut wasn’t that bad and my ketones rocketed, so I’m going to start the next fast I do with much fewer calories.
The biggest goal here for me on this fast was to get into nutritional ketosis quickly and test the waters to see how I function under fasting conditions. The goals of the next fast will be to maximize the time spent in a glucose ketone index under 1.0. My hypothesis is if I eat less calories, that will happen quicker. Time will tell.
Last but not least? Don’t mess with yoga moms. Enough said.
Alright, so there you go, there’s the data from the first fast. I’m going to be doing this again in a month or so and will be tracking and posting everything again. Before then I’ll post a protocol of how I’ll execute the next fast in a downloadable PDF.
Until then, any thought or questions? I’m not that crazy, am I?
Disclaimer: This type of modified fast is safe for most people, but not all people. For example, this wouldn’t be a great choice for someone with super low lean tissue, or in a compromised immune status. Don’t do this if you are pregnant, nursing, or are a child. But if you’re a child and you got this far in this article, I’m probably not going to talk you out of fasting, am I? I’m a doctor but I’m not your doctor. Check with someone in your area who can help you out before you do something silly.