Fasting has been all the rage lately. Jimmy Moore and Jason Fung released a book on it that quickly shot up the New York Times Best Sellers list. In fact, it has been so popular that they launched a podcast on it at the beginning of this year. Oh, I did I mention the episode featuring Jason Fung and fasting is still the most downloaded of the 2 Keto Dudes podcast?
Yet other high profile low carbers such as Dr. Stephen Phinney are not part of the fan club. My personal favorite article on the subject was Not So… Fast… (A Rant) from the prolific Amy Berger.
While on low carb myself, I’ve only ever intentionally fasted for 14 hours at a time, which was only done to meet the requirements of a blood draw. But while I don’t feel hungry when fasting, I don’t feel… right.
To be sure, I’ve wondered if I want to eat all the time so that I either maintain or gain weight, given I’m underweight right now. And therefore my feeling odd when not eating is perhaps entirely mental manifestation. Regardless, a fast of a few days probably wouldn’t be that bad anyway, right?
The Experiment
The plan was pretty simple:
- I’d take a blood draw in the morning at the beginning of the fast.
- Fast for three days while both supplementing and keeping electrolytes high, but otherwise drinking only water.
- Take a final blood draw for comparison on the morning 72 hours after the first blood draw. So in all, 86 hours will have passed since my last meal, making it a total of 3.5 days.
Execution
Day One: I was surprised to find I wasn’t hungry at all. This seemed to back up my theory that if I had already made the commitment to myself to forgo eating, my brain wouldn’t send me subconscious “shouldn’t we be eating?” signals.
As happens with me when I’ve lowered my total calories for an experiment (but while still being keto), I feel run down and puny. I have an overall feeling of lower energy. I also feel a little dispirited in this state, but its hard to tell how much of that is annoyance of that phase of the experiment vs it being an actual physical response.
However, that not-feeling-right sensation I mentioned above?… I was certainly getting that signal. But I had hoped I’d only feel it on the first day.
Day Two: I still wasn’t feeling hungry. And while I did continue to feel low in energy, it wasn’t notably better or worse.
However, the not-feeling-right sensation was definitely much, much higher. It was like nothing I’ve experienced before this point. In my imagination it was as though my body found a red phone line and called some special center of my brain to say, “ABORT! ABORT!” No physical pain, no odd changes in the senses, nothing other than a feeling… a feeling this was terribly wrong.
By the evening I decided to go ahead and cut the experiment short. I’d take my blood on the morning of Day Three as opposed to Day Four. Heck, at least it was a 2.5 day fast in the Data Can. I just knew I’d feel annoyed if my numbers had hardly changed. (Just writing that last sentence makes me laugh out loud now…)
General Bloodwork
In every blood draw now, I get a slate of general panels like a CMP and CBC. The latter is known as a Complete Blood Count and has 14 markers. These numbers always been in range… until this time.
| Ref Range | 3/21/17 | 3/23/17 | |
| RBC | 4.10-5.70 | 5.02 | 5.85 |
| Hemoglobin | 13.0-17.0 | 15.3 | 18.1 |
| Hematocrit | 37.0-49.0 | 44.6 | 52 |
All of these markers have to do with red blood cells and their functionality, which I won’t cover here. What I really wanted to see is if I had fallen off on my electrolytes, which would explain both the run down feeling and my general sense of malaise.
| Ref Range | 3/21/17 | 3/23/17 | |
| Sodium, Serum | 134-144 | 137 | 136 |
| Potassium, Serum | 3.5-5.2 | 5.1 | 5.8 |
| Calcium, Serum | 8.7-10.2 | 9.6 | 10.5 |
Interesting — instead of being under, I was over on K and Ca.
Cholesterol
Of course the big one is the lipid profile. And if you follow me, you already know what I’d predict after fasting for the very first time given the Inversion Pattern –> a record increase in LDL cholesterol.
So what happened? I hope you’re sitting down for this….
| Ref Range | 3/21/17 | 3/23/17 | Difference | |
| Cholesterol, Total | 100-199 | 371 | 479 | +108 |
| HDL-C | >39 | 72 | 70 | -2 |
| LDL-C | 0-99 | 284 | 368 | +84 |
| LDL-P | <1000 | 2068 | 3348 | +1280 |
| Small LDL-P | <=527 | <90 | 546 | +546 |
| Triglycerides | 0-149 | 76 | 205 | +129 |
Indeed it was a record!
Let’s unpack a few things:
- Like the Extreme Drop Experiment from last year, this one had a heavy shift in dietary energy, as in a sudden drop off. And likewise this huge degree of change broke the Inversion Pattern with LDL-P, while still demonstrating its general direction (lower fat = higher cholesterol).
- Given the pattern up to this point, we’d have expected LDL-P to land around 2200, but it instead landed much further upward at 3348. This is strikingly consistent with with the drop experiment that likewise overshot in the other direction (from 2597 to 1487 in three days!).
- Yet LDL-C would be expected to land around 355 +/- 22 and sure enough it landed at 368.
- While I don’t like doing these huge shift experiments, I am glad they continue to reinforce the general mechanics of the Inversion Pattern and further establish its nature.
- Once again Small LDL-P pops up in a low dietary energy context.
Final Thoughts
- While the experience wasn’t great, the data from this experiment was golden! As predicted, the Inversion Pattern kicked in and demonstrated just how fast cholesterol can rise while fasting, particularly for a hyper-responder like myself. I guess that last bit doesn’t actually sound like good news, but don’t worry, the blood test taken just 4 days later (not shown) had my Total and LDL cholesterol drop down near where they were on 3/21.
- I already didn’t enjoy fasting for even a half day before this experiment… now I’m very sure I don’t want to fast for longer either. This might be something I’d consider if I weighed more, but probably not. The incredibly alarming feeling I experienced was something I’d prefer to leave in the past.
[UPDATE]
Hat tip to James DiNicolantonio who pointed out my Uric Acid likely had likely risen and I sure enough it proved to be true:
| Ref Range | 3/21/17 | 3/23/17 | |
| Uric Acid, Serum | 3.7-8.6 | 5.8 | 8.2 |
I run a script that captures every marker outside its reference range and (unfortunately) mostly noticed those that did. Technically, Uric Acid was still inside, but obviously moving upward fast. Would it have gone above range had I been fasting another day? Alas, I’m now unwilling to find out anyway.
Very similar experience for me.
1. Ketogenic diet for several month
2. fasted for 4 days (96hs) before blood work
3. blood work 14 days later (eating “normally” and not being in ketosis)
Resutls
Fasted 4 days ===>>> eating “normally” + not in ketosis
Tot Col 434 ====>> 338
Trig 109 =======>> 92
LDL-C 337 ====>> 259
HDL 75 ========>>61
Wow — yet one more datapoint to show the Inversion Pattern!
Thanks for the addition!
For some reason, I’m not concerned with my cholesterol going up while fasting, nor does the ‘hunger’ that lasts maybe a minute bother me. I feel it’s merely psychological since anyone burning fat has plenty to ‘eat’ in their fat stores. My energy is good, I feel fit and my HA1C went down from 6.7 to a steady 5.4. The effortless weight loss and almost total lack of carb cravings is a real relief, as well, but to each their own. 🙂
That’s great! Again, I only speak for my own experience and you’re exactly right that it might have been entirely mental. But that said, it wasn’t my favorite experience, that’s for sure.
And once more, I have to emphasize that I personally know several dozen people who have been using fasting to great success for themselves.
In the lead up to your experiment I do not get any sense of what the benefits of fasting are claimed to be, or what is trying to be achieved. I get that in the context of this experiment you are just trying to document the repeated specifics of the Inversion Pattern. Also that you just “don’t feel right”, as part of your subjective documentation. But what is it that boosters of intermittent fasting claim is the benefit of doing so?
David: in this comment #4249 again you allude to several dozen people who have been “using fasting to great success for themselves” — but it is still unclear to me what criterion for “success” is being used here.
I find it interesting that Dr Phinney, who has consistently impressed me with his well-reasoned, factual, well researched and referenced data is not a fan of fasting.
As far as “fasting benefits” it may be worth checking into some of the presentations from Megan Ramos. Typically the benefits are seen in those with health problems (e.g. type 2 diabetes, obesity, PCOS, etc) or those who can’t/struggle to change their diet (e.g. due to budget, time, stress, etc). They don’t really recommend extended fasting (24+ hours) for people who are lean.
The electrolytes likely reflect dehydratation. I hope you rechecked them since. Those being abnormal can cause heart arrhythmias
Hi Nicole!
Yes, I actually get an electrolyte panel with every blood draw. All markers were back within range four days later, save potassium (which was just a hair above range at 5.3, but was back to normal in the test after at 4.3).
Do you think dehydration contributed to the changes? Particularly in light of the CBC changes?
I give it a very strong possibility. My doc thought it was primarily dehydration and recommended really upping my electrolytes after he saw my panels a few days later.
Brilliant article
Thanks, Herbie.
Why is a TC of 371, DL-C of 284 and LDL-P >2000 GOOD??
Not sure I follow — I didn’t call those numbers “good”. I’m appreciative of the data revealed in how these numbers move, exposing more of the pattern. But I don’t believe I’ve ever said I feel my higher cholesterol numbers are good. After all, discovering both how they work and why is the reason I started my research (and this blog).
Do you think that your numbers are good? If nit have you ever considered cghanging your diet to lower them?
Have you ever had a CT Scan and/or a carotid artery scan done?
– I don’t know if my numbers are good. I’m well aware the vast majority of doctors would insist they aren’t, but I’m working out the context of both how and why of how they got there from an engineering perspective.
– Yes — I’ll likely change my diet if I see progressing signs of heightened inflammation, oxidative stress, and/or atherosclerosis. In short, while I feel better than I ever have while likewise having low levels of inflammation and (so far) stable signals for heart health, then I need more reason to change.
– I’ve had 2 CIMT scans done (and am about to go in for a third) where the numbers are nearly a match. I had a CAC score of 0 from last year and will have another one soon as part of a more advanced CT scan that I’m getting in a couple weeks. Per my point above, I’m very, very actively capturing enormous data on my health and will likely catch one or more signs of declining vascular health early on (I say “likely” because it is still very challenging).
I have very similar numbers (a little worse) so what are you doing about it? Have you posted another thread on it? I found this in a google search.
I’m wondering what this would look like on a non hyper responder Dave. Any hypothesis on whether it might have something to do with the results?
I currently believe this same Inversion Pattern applies to the vast majority of people were they on a low carb diet, and quite possibly if they weren’t. Obviously this would likely be altered if there were an underlying metabolic disease, energy/lipid modifying medication, or weight instability, for example. But absent things of this nature… I’m inclined to bet the pattern would hold.
In other words, it may well be that anyone on any diet that stuck to the same composition of fat/carb/protein but dropping total quantity would likewise see a spike in their LDL cholesterol as the necessity to mobilize VLDLs (and the stored energy they carry) become higher.
I wonder if what you saw here was actually a good thing (if done in the short term, not chronically?!
If in fact fasting stimulates autophagy, then perhaps we are seeing blood cell turnover where you made lots more and then the old ones get broken down.
Similarly, since cholesterol is involved in so many repair and production processes such as hormone precursors.
There’s just a few of us that think the same thing as I do. That cholesterol is the red herring. That mostly, this is due to higher demand for fat-based energy coming from storage in the form of triglycerides being carried by VLDLs. The cholesterol being measured resides in those VLDL-originating LDL particles, which is why it’s quantity is inverted from the total amount of dietary fat I eat.
More fat in my low carb diet? Less need for fat-based energy from storage, less VLDLs mobilized, less cholesterol riding along with it. Lower cholesterol score.
Less fat in my low carb diet? More need for fat-based energy from storage, more VLDLs mobilized, more cholesterol ridding along with it. Higher cholesterol score.
So, if in a fasting state, of course there would be more triglycerides and LDL in the blood.
I am not aware that anyone has made a strict protocol for testing cholesterols if you do intermittent fasting or alternate day fasting. Obviously results in a fed and fasted state will show extremes from one another. What do we do for an in between/average reading?
You should take this further. You need to include all inflammation markers, sugar markers and anything else possible to show health benefits. This should also be done over time.
Most intermittent Fasters I know end up in exemplary health. They lower and/or reverse all signs of heart disease, diabetes, and in some cases they has success at reversing cancer (on dietary changes only).
Think you have MUCH more homework to do.
This isn’t really obvious to everyone, hence the experiment. Many would assume that LDL and triglycerides would go down from extended fasting, as you’re not ingesting fat. However, we know this isn’t true in healthy/lean humans.
The point of the experiment wasn’t to show health benefit or detriment of fasting, it was to demonstrate the inversion pattern, and discuss how it fits in to the lipid energy model.
I wouldn’t be surprised if this were the case – there are plenty of people over at IDM/The Fasting Method who appear to be experiencing health benefits as well. It just wasn’t the focus of the experiment.
I think you’ve gotten the impression that Dave is against fasting, maybe based on the title of the article (which is only referring to how he struggled through the experiment which isn’t surprising, as from what I recall he was fairly lean at the time). But, I’ve heard Dave say a couple times that “if you can comfortably fast, it might be a good idea to do so”, and I’m sure I can say confidently we appreciate the work Megan Ramos is doing with her supervised fasting clinic in Canada.
If you were referring to specific comments from Dave somewhere perhaps you could provide them so we could clarify any misunderstanding. Thanks!
Hi Dave,
Another fascinating experiment! The folks at ApoE4.Info (www.apoe4.info) have been closely following your work with admiration. We applaud your citizen scientist approach to understanding the role of cholesterol. We’d like to invite you to speak with us at the LowCarb San Diego event in August. We’re especially eager to hear more about what you’ve learned about being a hyper-responder of dietary fat. I suspect your work may have implications for our population. Please contact me at: [hidden] for details. Thanks for your consideration.
Best-
Julie
Hi Julie —
Thanks so much for the kind words. I’ve emailed you directly. 🙂
Dave,
Fascinating as usual! I’ve asked you think on Twitter but to reiterate, what I really want to know is: based on lipid response to fasting, when is the ideal time to measure lipids to give the absolute *best” insight into future CHD risk? Is it best to do one in the middle of a normal day, one 12-14 hours fasting, and perhaps one just after a meal?
-Chris
That’s difficult to answer with regard to risk, because risk itself is based on *not* knowing about the Inversion Pattern’s influence on cholesterol testing.
So if the question is “how can I get my best median cholesterol number?” — I’d say try to eat at the median of your calories/composition for five days, then fast for 12 to 14 hours before. You want that fasting period so you can be sure you’ve cleared out the chylomicrons which would impact the score.
Interesting that you found fasting uncomfortable. I have done two three day fasts always breaking them for social reasons rather than for feeling unwell. The difference is I am chowing down mostly on my own on body fat as I am overweight. My understanding via the 2ketodudes protein episode is that there is an upper % limit to the body fat you can convert so It makes sense that if fasting when not overweight your body would signal discomfort at eating too much body protein.
Thanks for putting up with your pain mistress called science!
Macro Four
Funny you say that — I often joke with my wife that “science” is indeed my mistress.
I’m not interested in fasting at all, but am always about 14 or 15 hours fasted when I have my cholesterol checked and I have high total (around 300), high HDL, low trigs, high LDL and high particle count. I wonder if I tried getting tested after breakfast if I would have a better reading? I have not heard of a hyper responder to fasting before.
Hi ReneeAnn-
I suspect you’d have higher trigs and LDL cholesterol if you ate before your lipid test given it would have newly introduced Chylomicrons from your meal being picked up in the measurement.
you should get it checked normally like you do after a 14 hour fast but eat high fat for the three days before like Dave Feldman has done. That will work better.
ReneeAnn, I’m 63 and I have the exact cholesterol profile you describe. Since you made this comment 2 years ago, I wondered what if anything you decided to do about your cholesterol? If you’re still reading this blog? I feel very healthy but am being badgered to go on statins.
Is it known if the ratio of cholesterol to other “passengers” in LDL and VLDL stays constant? Or could it be that when the body needs more fat due to e.g. Fasting, the relative amount of cholesterol in LDL and VLDL goes down.
So you see a rise in VLDL and LDL but there might not be a change in the bodies’ total circulating cholesterol?
I’m very doubtful the composition of cholesterol on a per particle basis is being up and down regulated. Mainly this is due to how they traffic in the bloodstream in their readiness of circulation.
… so the body increases cholesterol production, just because it needs to deliver more fat for energy when in a fasted state
Close–
The body needs to pull more fat-based energy (fatty acids) from storage to fuel cells, which necessitates more LDLp (Low Density Lipoprotein particles) to carry them (as triglycerides), which likewise carry cholesterol. Hence, LDL cholesterol blood tests actually becomes a useful proxy to see what the stored energy is doing, since they are only picking up the cholesterol from storage-based LDL particles (VLDLs).
Incredible short term changes, what do you think the long term effects would have been, for people doing 10+ day fasts? Gene changes would kick in?
Not sure what would happen with regard to myself, but my friend Tom Seest did a version of my protocol that was fasting only and it showed that past 3 days he noticed LDL-C and LDL-P began to head back down.
(Minor typo with small LDL-P: +564 -> +546)
Corrected. Nice catch!
I see exactly the same effect in my tests. So the implication is to stop fasting because it drives LDL to unhealthy levels?
Why assume that the LDL levels are unhealthy? That’s just the old paradigm with which we have all been brainwashed.
Some people in this world are short and some people are very tall. Is it dangerous for them to be so different to the average height of the human population.
I think the same applies to cholesterol.
Per my discussions on the Inversion Pattern, in this context I’m doubtful the higher cholesterol test score is that relevant for overall “health”. And I emphasize “test score” because if my higher levels of cholesterol were indeed unhealthy, then I believe they were nearly as unhealthy before the fast as I was circulating more cholesterol in chylomicrons in the fed state.
It helps to know what I mean by this if you’re already familiar with my research and the concept behind the Inversion Pattern.
Dave,
re my twitter comment I have some data to share with you if you are interested.
Leigh
How about repeating the experiment but trying a 16/8 fast for a few weeks by skipping breakfast….you may lose some weight though.
Dr Mark Hyman who says he needs to keep his weight up as he is very slim, eats some starches like sweet potato to ensure he doesn’t lose weight so when you eat after the 16 hour fast you may need to add in some low GI starchy carbs perhaps.if you start losing weight.
Dr Terence Kealey has written a very illuminating and interesting book on skipping breakfast ” Breakfast Is A Dangerous Meal “….
Thanks — I’ll check out their work on 16/8. But I’ll concede fasting is now a bit lower on my list now given the new set of experiments.
Just got lab results today which flabbergasted me. I have been on the extreme LCHF version of my diet needed to actually lose weight (I shoot for 500-1000 cal, 90% from fat and most of that saturated fat).
BTW, doing LCHF for over 5 yrs.
Cholesterol numbers are through the roof for me:
LDL-P 2260
LDL-C 227
HDL-C went down from my usual range and is at 53.
What interests me about my results are the high Triglycerides (168) and high glucose for me (104). Any speculation about what would drive these numbers up due to my near-fasting? If I need to get myself up to speed science-wise here, any resources to understand this mechansim?
Both the LDL-C and LDL-P are comparable to mine (avg 270 and 2400, respectively), suggesting you are a low carb hyperresponder as well. But the HDL-C of 53 and Triglycerides of 168 are unusual. That said, they aren’t necessarily worrisome depending on the context.
If you’ve been reading my blog, you should know your diet and activity in the 3 day window of time before your blood test has a massive impact on your -C scores (with the 3 day window + 2 day gap for your -P scores). Did you eat an unusually lower amount of fat and/or calories in the days leading up to your test?
I went off half-cocked in my post. Frankly, I was panicked by the test results and came by your blog that way. Apologies. I have now read the whole blog, understand the science and your project.
I have been doing something akin to what Atkins called the fat fast, except I restrict calories even more than he suggests (sometimes as low as 250-300 cal/day). I need to do this to get any weight loss. I even hold steady weight-wise at 1000 cal/day.
I can’t imagine how much weight I’d gain at 5000 cal per day, even in 3 days. Did you and your sister see large weight gains? I have to scrupulously watch protein intake as well as carb.
Planning to go back to maintenance once I reach weight goal. I will re-test at that level (90% fat).
I actually don’t have much data on fat fasting with regard to lipids as neither myself nor anyone who has reached out to me was doing this (that I know of). But in theory, I’d certainly expect lipid numbers to look “bad” relative to your being on a maintenance ratio.
Again, if you’ve been following the inversion pattern, you know that your “bad” markers (LDL-C/-P and Trigs) will rise while the “good” (HDL-C/-P) will fall given you are trafficking more from storage. This is why I fully expected my markers would do exactly this for the fast I did from above.
Please don’t feel you need to do the protocol if you are only looking to know what your numbers are for your median diet. Take the test when you are back around to it and you’ll have a better sense of what your average numbers are.
If you don’t notice weight loss going 1000 cal/day, have you tried sticking to it for at least 5 days?
Sometimes the body holds on to water for a few days, then you’ll suddenly lose 1-2 lbs.
This is my experience and others have experienced similar effects.
Another factor to help track your loss can be to rule out any day-to-day variances by always weighing when you wake up, with the same amount of clothes, and preferably after peeing and taking a dump. Intra-day variances can be 1-2lbs, masking actual weight loss.
Additionally, track *every* calorie. It kind of breaks the laws of thermodynamics to go on 1000 cal/day for a week and not lose any weight, given the same amount of water intake.
I wonder if the rise in your numbers with fasting is a type of stress response. I am a hyper responder and recently noticed a huge rise with acute stress – in a nutshell, I was admitted to hospital and very worried, cholesterol on admission 250, then sent home next day thinking all was ok, relaxed, then 3 days later got the results I wasn’t ok and spent the next 4 days in hyper stressed mode worrying myself sick. Cholesterol was 350 at next appointment! 10 days later in reduced state of stress, cholesterol was 227. This was all psychological stress not physical. So, do you think raised cholesterol is a response to cortisol – hard to do an experiment on this though!?
I absolutely believe hormones can play a roll, and especially those activated by stress like cortisol. But that said, I now firmly believe nothing trumps energy. I now have far too many data points to feel otherwise.
Indeed, in a stressed state (particularly with modifiers like adrenaline in play), the body can require energy to be used at a far faster rate to meet the needs of readiness and intensive focus (both big metabolic burners).
Possibly: Higher Stress => More Idle Energy Burned => More Energy Needed from Stores => Higher Cholesterol via VLDLs
Hi dave
Thanks for providing such a useful information. I’m an hype-responder too and never tried longer than 14 hours fasts. Maybe one day I’ll try it with your experiment in mind to see if I can stand it.
Apart from that, I’m wondering why Lipid tests don’t include NEFAS (Non Esterified Fatty Acids). If I’m not mistaken they take some part in the Lipid trafficking in the blood stream the difference being that they don’t come bundled in Lipoproteins but are bound to Albumin. Isn’t this a missing data point in any study about Lipid levels ? To my knowledge, they can be elevated in a series of conditions like Insulin resistance, uncontrolled Type 1 Diabetes and prolonged fasting.
Thanks
Actually you’re right inside my wheelhouse with NEFAs as it is something I’m actively studying and trying to tease out myself. It is also a strong explanation as to why many who are lean and/or athletic (like myself) are disproportionately hyperresponders. But this is a medium correlation, not a strong one, so there’s still more studying to do.
I actually have some papers on this in my reading short list that I need to get back to here soon. But without question, they are certainly a key factor in the overall Fatty Acid Cycle.
Ah ah the Typo “Hype-Responder”. Damned, I’m unmasked !
Anyway thanks for the reply. Good to know that I’m not alone wondering about those molecules.
Hi Dave; These are curious results (speaking as a doctor who looks at lab results all day every day). The high potassium is surprising, and could be from hemolysis during the blood draw (the needle tip not being exactly in the vein slices the passing red cells open, releasing potassium). Or from the electrolytes?
Also, it’s unusual to see a true bump in Hct from one day to the next unless it’s from vascular volume loss (so the increase is relative to the decrease in blood serum). Fasting, we know, causes insulin to drop and Glucagon to rise, thus mobilizing our fat stores for fuel. But because Insulin causes sodium retention, it raises intravascular volume, and by fasting and thus dropping Insulin, there is a diuretic effect as sodium is dumped via the kidneys, and water follows it. (Hence the quick wt loss seen w a ketogenic diet, and the drop in BP). So you are probably not “dehydrated” as such, and probably don’t need to take electrolytes for such a short fast.
A couple more points;
– it’s hard to draw real conclusions from an “n of 1”, so please, please, repeat this and see if you get the same results (I’m guessing not)!
– a lot of us (and likely you as well) are starting to conclude that chol and lipids made by the body naturally are normal healthy components of energy distribution, and their quantity (as opposed to quality) does not directly cause atherosclerosis. Plaque seems more the result of corruption of the particles from things like oxidation (smoking, artificial Trans and Oxidized fat (polyunsaturated seed oils), HTN (vessel wall stress), inflammation (immune response to damage from AGE’s from excess sugar, processed foods, additives, toxins, allergens, invaders, etc), calcium buildup from under-carboxylated Matrix GLA Protein due to Vit K2 (menaquinone-7) deficiency, and direct negative effects of Insulin on the endothelium (DM2 and Insulin-resistant DM1’s).
– I have been reviewing CAC scores for my patients for almost a year now, (n=40), and have found no relation between high LDL and high calcium scores. There is a strong relationship between the Trig/HDL ratio, however (marker of Insulin resistance), as well as the ALT and AST (marker of hepatic inflammation and cell death, over about 20), cardiac CRP over 2 (inflammation of vessels), fasting Insulin over about 10, and A1c over 5.5 (3 mo avg blood sugar).
PS – I love your inquisitiveness, and your inversion theory of cholesterol, and hope you keep pushing the envelope!! Below is a link to a really great article on Intermittent Fasting and it’s benefits. (I fast 16 – 23 hrs a day, and absolutely LOVE the way it makes me feel! My brain is sharper and more focused, and by exercising while fasting I have built much more muscle mass (and skiing agility) from the elevated Growth Hormone and BDNF (Brain-Derived Neurotrophic Factor). Note that it does take most folks about 2-3 weeks to switch to this form of metabolism)
authoritynutrition.com/10-health-benefits-of-intermittent-fasting/
Hi Nils-
Great comment!
– The high potassium may have been due to my using Morton’s Lite Salt mixed in water as was advised by a friend who does frequent fasting. It was his favorite way to supplement potassium, but I might have overdone it. (I think I averaged 1 tsp / day)
– Another possible confounder with the electrolytes was the amount of water I’ve been drinking heading into the blood test. I would usually average between 2 to 2.5 liters in the 2 hours before the test out of advice I had long ago from a doctor to help with the blood draw for my hard-to-reach veins. I had asked him at the time if it would have any impact on the tests to which he said, “almost none” — but I now believe otherwise. Unfortunately, I’m locked into it for now as I don’t want to change a variable that is present with all prior blood tests until I can do a comparison experiment to control for it.
– Not sure I’ll be repeating this test, or at least, that I’ll be doing it soon. I have something very big on my plate now, but I might come back to it again later.
– I agree with just about everything you said regarding cholesterol. While I can’t say I know for sure that I can rule either way just yet, it is always comforting to hear from other doctors who actually have a practice and see patients on a regular basis what the patterns are. Certainly I’m much more concerned about inflammation and oxidative stress for CVD health overall.
– On that note, my one concern that I’ve mentioned here and in other places is my mysteriously high ferritin levels (generally in the 600s). I’ve done a few iron panels that come in normal range and it doesn’t appear genetically likely as hemochromatosis [ given rs1799945(C;G) rs1800562(G;G) ]. Thus I am concerned it is a possible indicator of oxidative stress. (The scientist in me is more annoyed at it being a research distraction from my normal lipid focus.)
– Thanks for your kind words on the PS. It really has been an unexpected journey in so many ways. 🙂
Hi Dave,
Your comment on high ferritin drew my attention, as I have recently received my blood panel results with ferritin at 892.
I have been on a lchf diet for 6 months, my lipids are total ch 306, LDL 205, HDL 83, Trigs 87.
High ferritin apparently not due to iron overload. So in theory it can be due to inflamation, malignancy, hepatic damage… but hepatic markers are normal, CRP <0.1, no signs of infection, normal weight and apparently very healthy…
So what is your theory about this? I’ve read another cause can be dislipidemia and metabolic syndrome and I think maybe in this line the ketogenic diet somehow may have triggered these results?
I am at a loss because none of the explanations I have found in literature seem to apply, and the only “culprit” I can think of, as something that has been a major change in my life, is the ketogenic diet and being fat adapted…
Any thoughts? Thanks!
I genuinely have no idea. My Ferritin has slowly dropped in the last year and I have no idea why. The closest corollary is that it appears to be right around the time I got serious about electrolytes and made sure to include a LOT more salt in my diet (typically 10g a day). A lot of things got better and seemingly that might have been one of them. Here’s a data dump of every ferritin score I’ve had since Nov ’15:
527, 712, 602, 562, 601, 559, 657, 767, 609, 489, 527, 569, 973, 731, 775, 759, 759, 662, 551, 525, 527, 514, 475, 536, 316, 303, 316, 296, 226, 292
— that first 316 was 8/14/17 and that was a month after I really upped my salt.
Hi Dave,
I know this is and old post, but I’d like to share my results.
I’m 32 years old, clearly with IR, I was weighting 104kg being 180cm, although I’ve been low carb for a couple of years. Since I’m a classical pianist, I sit a lot and I do not go to the gym or so, only ride a bike around 25 minutes a day. In 2006-2007 I did carnivore for like 6 months and trained every day and my body weight dropped to 76 in only 3-4 months (from the same weight), but then I moved to Sweden and here I stopped training and ate a lot of junk and wine for several years. Now since 2015 I’ve been always LCHF although I might eat some shit (like pizza or ice cream and some beer but less than 10 times a year). Almost always have more than 0.5 mmol ketones in blood. So long story short, this November I decided to cut the weight so I went all in in carnivore, and I did a 6 days fast between Dec 2 and Dec 7. Since November 7 until now I lost 12 kg even though I had twice some creme brûlée/ice cream. I drink wine in the weekends, around 2-3 glasses.
This Monday I did my blood works and there were some interesting results (one alarming ferritin):
Cholesterol: 6,4 mmol/L
LDL: 4,83 mmol/L
HDL: 1,08 mmol/L
ApoA1: 1,39 g/L
Apo B: 1,43 g/L (rather high, right?)
ApoB/Apo A1: 1,03 (!!!!)
Fasting Glucose: 5,1 mmol/L
HbA1C: 33 mmol/mol (which is around 5,2%)
Ferritin: 468 ug/l
But to the good thing:
C-peptide: 0,27 nmo/L (range 0,3-0,8).
IGF1: 122 ug/L
I got preoccupied about the ApoB/ApoA1 mostly, but even the Ferritin is quite high, so I started googling a lot to see if I could find any studies on fasting and ferritin since I thought that maybe it had something to do with Autophagy of the liver which contains a lot of iron and could free iron in the blood and then of course Ferritin would go up to pick up the iron and protect the body from the iron overload, and I found this:
https://www.researchgate.net/publication/306274004_Increased_Serum_Ferritin_and_Decreased_Serum_Haptoglobin_Levels_Developing_in_a_40y_Old_Healthy_Man_Fasting_for_3-Weeks
and this (in swine):
https://academic.oup.com/jas/article-abstract/37/3/697/4667532?redirectedFrom=PDF
My liver ALAT is: 0,53 ukat/L (range 0,15-1,1)
and ASAT: 0,30 ukat/L (range 0,25-0,75)
so I don’t suspect the liver is exploding (yet).
Maybe some interesting data for you? Hope you have some commentary!
Merry Christmas,
Daniel
Certainly interesting data 🙂
You may or may not be aware, but Dave actually has/had high ferritin as well that only went down recently.
Do you know what your total cholesterol and triglycerides are by any chance? If you have those you can put the information into our report tool and it helps give some more context/information
Dave,
I have a simple explanation to offer. Here goes:
1) Since you are well adapted to LCHF your metabolism is primarily utilizing fat for energy.
2) During the 3-days of low fat consumption (and especially low fat + calorie deficit – the extreme drop) you are effectively ‘starving’ your metabolism and it will compensate by cranking up lipolysis of adipose tissue and greatly increasing your level of circulating free fatty acids (FFA) – akin to fasting.
3) Most people think of adipose tissue as just ‘fat storage’, i.e triglyceride (TG) storage but it is also a large reservoir of stored cholesterol.
4) The size of the cholesterol reservoir is proportional to the TG content, which suggests that homeostasis of cholesterol within adipose tissue might well be tightly coupled to TG metabolism. If so, any cranking up of TG lipolysis will likely be accompanied by an increased efflux of cholesterol out of adipose tissue along with the FFA.
5) During fasting/semi starvation the rate of fatty acid cycling is much greater than actual oxidative demands – probably by a factor of two to three times. Therefore if cholesterol and TG metabolism from adipose tissue are indeed coupled there is a good chance it will get manifested as an increase in serum cholesterol.
6) I will avoid trying to theorise a detailed mechanism for how the different lipoproteins participate in all of this except to say that it probably begins with an increase of HDL in order to reverse transport the increased cholesterol efflux and increased HDL is one of your observations.
7) Now during the 3-days of high fat (and hypercaloric) re-feeding I think the opposite of the above happens. Adipose tissue lipolysis is cranked down and probably stops as chylomicrons and FFA flood in. Again if the ‘coupling’ assumption is correct the cholesterol efflux from adipose will also be cranked down and get manifested as a decrease in serum cholesterol.
1) Yes
2) Yes
3) Yes – mostly
4) Yes – mostly
5) Here’s where it gets a bit trickier. Yes the fatty acid cycle is in play and to the degree that the components found inside LDL particles (TGs and cholesterol) are more readily available is certainly a key component. (See more with NEFAs, of course)
6) That’s actually a very large subject that takes much more unpacking.
7) (See below)
The mechanism as you are describing it is a supply-driven process, rather than a demand-driven one, which is where my headspace started out. And this is likewise where most doctors and researchers believe it to be.
It’s actually the data itself that started me questioning it otherwise. The more tightly I saw the inverse correlation between dietary fat and LDL cholesterol (as well as LDL particle count), the more I realized it had to be a regulatory mechanism. This isn’t to say there is a single operator (such as the liver) that is orchestrating the the entire process, but there *does* appear to be a target configuration while on a specific macro-energy breakdown (keto diet). And many different players have the regulation instructions baked into it in order to meet that target.
I come back to it again and again as the data is telling this to me — it’s all about the energy. The body clearly wants to keep both long term and short term (readily available) energy in place and this is shown to be more of a demand-driven process than a supply-driven one.
In short, we can theorize the different pathways involving insulin, NEFAs, gut vs liver production of LDLp, etc. But for all these disparate elements to *coincidentally* land in the same high correlation distribution over and over again doesn’t fit Occam’s razor. Rather, it is the reverse — that something is influencing their pathways in a staging-and-pacing manner that resembles more of a cloud network that I see all the time in my line of work with load balancing (which would be consistent with many other operations of the human body).
I don’t follow why you say I am describing a supply-driven process.
The process I have suggested is demand driven because it is trying to maintain the body’s fatty acid futile cycle in the face of semi-starvation. This is precisely an energy demand, for which reason the body releases stored TG in the form of NEFA’s.
The futile cycle circulates these NEFAs at a rate, which is about 2 to 3 times higher than the actual metabolic oxidative requirements and the liver acts as a buffer for some of the excess circulating NEFAs by churning out more VLDLs.
I am suggesting that your periods of low fat intake are akin to semi-starvation because your primary (maybe even 100%) source of energy is fat for direct oxidation and for ketone production.
Your observation of increased LDL-P (albeit with a lag) following low fat intake/semi-starvation would be consistent with the liver increasing its VLDL output, which ultimately has to mean more LDL-P as the VLDL degrade.
Now digressing back to cholesterol and considering just total serum cholesterol I observe the following:
From your ‘identical diet experiment’ the consecutive swings in total serum cholesterol concentration from one 3-day period to the next range from about -60 to +70 mg/dL for yourself and about -30 to +55 md/dL for Darla (as best as I can read the data from your charts).
However, the two of you must have very different blood serum volumes, probably around 3 liters for yourself and 2 liters for Darla. This means that although the changes in cholesterol concentration may look similar the consecutive swings in absolute serum cholesterol are rather more different, viz
Dave: -1800 to 2100 mg Darla: -600 to +1100 mg
I am postulating that these transient swings in serum cholesterol arise from the energy demand driven changes in adipose fat storage on the assumption that cholesterol has to be expelled along with TG during lipid droplet lipolysis and taken-in again when the metabolism switches back to fat storing mode.
I think (but it needs checking) that the average cholesterol content of adipocytes is about 5 mg per g of stored TG, in which case the observed cholesterol swings would correspond to the following range of consecutive fat storage changes:
Dave: -360 to +420g Darla: -120 to +220 g
or since these occur over 3-day periods the corresponding average daily change estimates are:
Dave: -120 to +140g/d Darla: -40 to +70 g/d
The two of you ate identical diets but your energy demands are quite different. (78.5 kg male versus 64.4 kg female). Hence I would expect that your fat ‘turnover needs’ under the experimental conditions should be greater than Darla’s – as indeed the data suggest.
Occums Razer. Dave what if you redrew your charts with the following formula: Fat= Calories burned – dietary carb – dietary protein. So fat is 100% while fasting. And while in a calorie deficit, fat is dietary fat plus deficit fat burned.
This would explain the continued increase in ldl p while fasting.
Interesting theory, Bert. When I next make a pass at my spreadsheet, I might play with that equation.
Hi Dave – are all your blood tests done, fasted (e.g. overnight if not ‘fasting’) and at the same time?
I’ve been in and out of keto for a few years and generally just supplement carb for intensive cycle rides. I’ve stopped counting macros but my last (overnight fasted) bloods where not so ‘good’:
HDL 82
LDL 432
TRIG 122
TC: ??!!
Yes — before *every* blood test I aim for around 14 hours of total water fasting to stay consistent. This experiment was the *only* time I went for longer than those 14 hours by design.
Your profile falls right in line with an athletic hyper-responder. Indeed, I need to do a post on this at some point given how frequently I see it now. Mechanistically, the higher LDL makes sense given you need to traffic more total trigs to feed your cells meeting your energy demands (given you have lower adipose energy stores from being lean and lower relative glycogen stores from being low carb).
In fact, I don’t think I’ve yet seen a single lab where trigs were < 150, HDL > 60, and LDL > 300 that wasn’t a lean and/or athletic low carber.
Thanks for responding, Dave. These results have me questioning EVERYTHING! Going to now knock off Saturates for a couple of months and see if that turns the tide – I just don’t remember how I used to eat pre-LCHF!?!
Read about perhaps upping fibre to help but If I start down the oats or grains route I find that’s a slippery slope back into all out carbdom.
I am near my leanest (64kg/5’10″/14%bf),due to cycling occurring at this time of year but I wouldn’t say I’m super lean.
Thanks again for your time. Do you do consultations?!
I’m happy to chat with people via Skype here and there, but I never charge. While I might someday have a patreon account to show the “sausage making” of my experiments, I plan to always have my data and analysis free to the public.
Dave, this is Nadir – the cardiologist. I am interested in this data if you have. I am also a die hard cyclist and have several team members who are in the LCHF camp. They have not checked their numbers but I am interested in any information you have on athletic hyper-responders who are cyclist. I guess, what you are saying is that if they start out a ride in a fasted state, they will have no chylomicrons and thus will make a lot of VLDL to provide energy. Thus cyclist doing this repeatedly several times a week are burning fat and sparing the glycogen. Thus in order to for the muscles to get the Trigs to burn, the liver churns out more VLDL. I wish someone would do a mechanistic study on the cyclist following the LCHF vs the high carb diet.
Hi Nadir–
Just to update anyone following this thread, you and I spoke over skype between the comment you made above and my response here.
But for posterity, I’ll answer here as well — yes, the lipolysis in the fasted state would ultimately lead to production and recirculation of more VLDL-originating lipoproteins. This would likewise appear as though their “cholesterol went up” on a blood test, given a blood test is picking up almost entire VLDL-originating LDL cholesterol.
Interested in why you think your RBC etc increased?
Hi Dave ,
Thank you for sharing your findings, they are fascinating! I have been lchf for 10 years. I started out on Atkins and over the years have slid off here and there but for the most part have kept to it. Since hitting post menopause I have gone to Keto and IF. I am post trauma after being hit by a car last February so somewhat sedentary since then. I am about 20 lbs overweight and working on losing it. It’s slow going!
I recently moved and on gaining a new family doctor, he ordered a lipid panel. I have never had one done before, I am 59 yrs old. So I took the test and the results were not stellar! Here are a few,things I think might skew them…I was fasted water only at 40 hrs at time of test . The fast was in response to some high carb indulgences especially a dinner that was 2 days before the test…..and was on a steroid inhaler to reduce a cold induced bronchospasm.
Cholesterol 339.9 , Trigs 132.8, HDL 64, LDL 250, Non HDL 128.7
My doctor who claims he approves of the Banting diet ( he is from South Africa and didn’t recognize the word Keto) quickly recommended statins and a vegetarian diet! I refused both much to his horror!
I wouldn’t say I was lean but do you think this is still a hyper response?
Hi Lorraine — while I think it is certainly possible you are a hyper-responder, I’m willing to bet you spiked your LDL cholesterol dramatically by having that 40 hour fast beforehand, as I did in this post. I’d retake the test and be sure to fast for 12-14 hours between the meal you had the night before and the blood test the morning after. 🙂
I recently completed a 3 day 19 hour water only fast from Oct 19 430pm to Oct 23 1130am
I took the blood test 45 minutes just before breaking by fast Oct 23 1043am.
I do have a blood test from Oct 17, a few days prior to starting my fast.
Too bad I cant post a pic of my metabolic panel and cholesterol.
I too see very elevated levels of Bilirubin, AST, Total Cholesterol, Triglycerides, LDL and VLDL
Before/After
Bilirubin 0.5 -> 1.6
AST 46 -> 60
Total Cholesterol 183 -> 222
Triglycerides 86 -> 175
VLDL 17 -> 35
LDL 78 -> 112
Im not nervous about it, yet.
I think this may be more of a matter of going keto and body destroying cells to feed itself.
The idea being was to replace bone marrow, white blood cells with healthier new cells.
That said, its more important to check and possible recheck the same test possibly a week or two later.
Maybe even upto a month later.
So the original post of this thread didnt include the blood test to show recovery – that would then be a more complete picture in my opinion.
Thank you so much for adding this very illuminating experiment! You really picked up a lot of biomarkers as well — great work!
I likewise agree that it is hard to say whether a good or bad thing. But I’m inclined toward the Net Good Thing if you felt generally okay. I likewise feel there is a lot of evidence for the “housekeeping” that comes with both whole body and cellular level fasting (like autophagy, of course).
The [fasting=cholesterol spike] evidence is impressively clear, even more so than the Inversion Pattern (about 85% success rate). But then, I knew this already to some extent because (unlike the Inversion Pattern) there’s been a lot more historical studies that showed this with human subjects. What I *didn’t* know until George pointed it out, is that these studies date all the way back to the 50’s.
Hi Dennis,
Just wondering if you could share the white blood cell count and absolute lymphocyte count from your blood tests at baseline and after the fast? I am wondering if the lymphocyte count is the key to the renewal of white blood cells that you refer to.
I had similar results. And the reason for my 5 day fast was to trigger autophagy and ketosis, that worked well and I lost 14 pounds plus my heart felt very good ( no more palpitations etc, heart in throat feeling).
This 5 day fast was my only solutions after being emitted to the ER, and visiting a cardiologist who both did nothing but assign me statins. Which I cannot take.
I am a 55 year old woman, quite thin. I have always had reasonably low cholesterol (total around 160). But the past three months, I’ve been doing intermittent daily fasting (with a few 24 hour fasts here and there) of 19 hours (eating only between 2 and 7 pm). My most recent total cholesterol came back as 297! HDL was 91, and tryglicerides 60. As those last numbers were okay, my doctor did not recommend statins (which I wouldn’t take anyway).
I do have a heart disease in my family (father died at age 55 of heart failure; not overweight nor a smoker, fairly active). So I would like to see lower cholesterol numbers. On the other hand, my father had normal cholesterol levels (though autopsy revealed extensive heart disease).
Anyway, I offer this to show that even intermittent fasting can drive cholesterol way up. So I’m back to three meals a day and will retests within a few months to see if that changes things.
Thanks!
Kathleen
Kathleen,
Based on all the research documented on here, I would expect LDL-C to increase with IF.
I’m curious why you started IF, since you are already thin? (I’m in the same boat, thin and practicing some IF, CVD in family).
IMO, your lipid panel is enviable, TG/HDL of 0.7. If you are disappointed that you need to give up IF because of a TC score, I’d encourage you to get some better metrics before making a decision. At 55, you might consider a CAC test. Ivor’s talk is a great intro on that.
Dave,
The effect you observed is not uncommon. See this 2014 study: https://www.sciencedaily.com/releases/2014/06/140614150142.htm. Fasting needs to be done over a period of time to see changes. What you probably experienced is your body trying to fight the sudden change by dumping more cholesterol into your blood.The inversion you speak of is probably a temporary self-protection effect, and not a long term phenomena.
I’ve definitely seen studies predating my work where fasting brought about higher cholesterol.
However, I’m more of the mind this is a mobilization of energy — and in particular, fatty acids — in the absence of incoming energy by the diet. Since cholesterol “ride shares” with these fatty acids when packaged into VLDLs (the precursor to LDLs), it makes sense as to why LDL cholesterol would climb.
I am female age 61. 173cms ( 5’8’). Currently 63kgs ( 139 lbs) Very active and well (athletic most of my life) I always look slim, but my body fat with calipers has always had me in the obese range! Normal weight and BMI 21. I work full time in my own business and am happy. No medications beside supplements, mg, d3, Omega 3, zinc.
I have followed LCHF since 2013.
3 months ago I stopped drinking completely (was wine about 2 sometimes 3 glasses, every day) I felt it had become a bad habit. I wanted to focus more on my health and well being. Wanted to lose a couple of kgs and get a bit stronger and toned. I was cheating occasionally on lchf and wanted to get back on track. The easiest way for me to do this was to go into ketosis which I did very easily about 3 weeks ago, I increased my exercise to daily 1.30 hour weight training exercise. I walk 4 KMS daily to and from work. I have lost 2.5 kgs in last 2 weeks am feeling great and strong and sleeping like a baby 8 hours. Could not actually feel better!! I go for annual blood tests for my own interest every year.
I was shattered today when I received my labs taken while in ketosis!
Previous/yesterday
TC was 8.8 (340) – now 11.66 (447.87)
HDL was 2.3 (88.9) – now 1.8 (69.6)
LDL was 6 (232) – now 8.1 (313.22)
TRI was 1.7 (151) – now 2.75 (243.57)
HDL Ratio was 3.8 – now 6.4
Non HDL was 6.5 (251) – now 9.8 (378.38)
HbA1C was 5.4 – now 5.4
Fasting Sugar was 5.3 – now 5.1 (91.89)
CRP was 0.5 – now 1.73mg/L
I typically eat 16:8. Time restricted eating.
So my eating has been high in saturated fats this last 3 weeks! I loose my appetite when in ketosis, but not my energy so I thought everything was perfect until I got these results!! I have followed you videos and blog and Twitter for a while so I am aware of your experiences…. If you are a hyper responder I am a super hyper responder! The results that bother me most, all of them, but CRP and triglycerides!! Any comment? Thanks
Hello Di,
Since we sound similar and I’m going to start LCHF soon, your post was of interest to me so I found this that you might be interested in: https://friedmansprout.com/2015/04/01/how-does-a-ketogenic-diet-affect-you-part-3-c-reactive-protein-a-marker-of-inflammation/. Have you thought of doing IV therapy? You can get an inflammation reducing cocktail and that might help these numbers. Just a thought!
Best wishes!
Hi Di–
– The higher triglycerides would concern me. But I’d want to confirm a few things first:
1. Was there anything unusual about your diet in the five days leading up to your most recent test?
2. Did you do a *water* fast (only drank water, no coffee or anything else) of at least 12 hours before your test? (and not too much more, such as multi-day)
– Your CRP of 1.73 is not that unusual. I love the test, but find it is a hypersensitive needle. It is especially likely to be elevated if recovering from intensive exercise or from being sick. Mine is typically below 1.0 (though sometimes 1.0-2.0), but in the tests I took 24 hours after doing a half or full marathon, they were 7-52!
Hi Dave
1. I was almost zero carb and ++ saturated fat ( meat, eggs, butter, cheese, nuts) for 5 days prior
2. Yes I only had water in the 12 hour fasting period, nothing else.
3. Heavy daily weight training with a 2 kg weight loss in the 3 weeks prior
4. Feel so good on this regime !!
Wow! what an increase following your marathon! I am going for a full lipogram in a few weeks and will post here and keep better details of food etc. Thank you:)
Hi Di–
Very interesting… I seem to keep finding many with ZC showing higher trigs. And like you, they all feel fantastic and don’t want to change gears.
I’m going to see if I can do a bit more investigating on this with other ZCers.
And yes, please share back your new lipid numbers as I’d be curious what does and doesn’t change. Thanks!
Makes no sense. I had the same thing, very healthy diet and mostly vegan but salmon once a week. My BMI was 22, and I ran 2 miles a day etc. No alcohol, no caffeine, no sugar painfully healthy.
2 months ago my cholesterol was 190, HDL 55, TG 65, LDL 110 all fairly decent. But for some reason I got painful heart palpitations, went to ER had a panic attack, saw a cardiologist etc.
They did nothing to help my heart except say “you have to live with it” . So I went on a 5 day water fast, lost 14 pounds and took a Cholesterol test 1 hour before breaking my fast.
The results were horrifying.
TC over 290, TG was 450 , HDL 50, and they could not calculate LDL. You realize LDL is calculated on old lookup tables from the 1970’s and is not done in a centrifuge anymore. So LDL can be off…..that’s why for me they couldn’t give me a number.
Truth is after the FAST I feel great, I will continue to eat well, exercise and fast. I tried statins for 3 years and they made me sick, I felt like I was aging quickly, and had all sorts of issues. Eventually my body would reject even the smallest dose.
Hi Traci
Thank you, the article was interesting and happily, I believe I know what was causing the high CRP so will not do an IV therapy. My CRP is usually very good. Best wishes with the LCHF lifestyle:)
It is interesting to note that your small LDL-P rose that much. My small LDL-P hardly budges: it stays between 1040-1089 even when my LDL-P fluctuates between 1617-2295.
Steve Phinney and others have proposed that during fasting, the body mobilized the cholesterol stores in adipose tissues.
https://www.ncbi.nlm.nih.gov/pubmed/2035468
I never quite understood how this happens. Do the adipose tissues directly store LDL particles which come back to circulation after weight (fat mass) loss? Or fresh LDL particles from the liver visit the adipose tissues and there stuff in more cholesterol from the adipose tissues?
If one’s adipose tissues were formed when they were eating badly (e.g. high sugar diet), could major weight loss unleash damaged (e.g. oxidized) cholesterol or LDL particles into the bloodstream?
In the above article, Phinney cites the following paper:
https://academic.oup.com/aje/article-pdf/129/2/312/200594/129-2-312.pdf
Here is a quote from its abstract:
‘The 25-year crude risk of coronary death was 26% in the “gain and loss” group, 15% in the “gain only” group, 14% in the “no change” group, and 17% in the remaining 1,550 men.’
I am stunned that the risk was greater in the “gain and loss” group than the “gain only” group.
I am especially worried about this possibility because my symptoms (exertional chest pain) of 99% LAD blockage first appeared after I lost 8kg in 5 months.
Hi Abhishek,
Of course, I must first caveat I’m not a medical professional, just an engineer obsessed with cholesterol and the lipid system (and experimenting!).
– Small LDL-P being that high as a total ratio would get me curious to run more labs. Have you had these markers checked?…: hsCRP, Ferritin, GGT, Homocysteine, and Lp(a)
– Fatty acids and cholesterol are stored in adipose tissue, but the lipoprotein that brought them is not. When they are released into the bloodstream, they are ultimately ferried back to the liver (where they may be repackaged back into VLDLs for use, or have other fates).
– The gain/loss study described above doesn’t surprise me. It sounds like yoyo dieting which implies either (1) the up/down group is already predisposed to committing, then binging out of it in a cyclical fashion and (2) they were up/down for other reasons more likely deleterious to health, such as habits like smoking (starting, restarting, etc), chronic diseases, or even just stressful lifestyles (work/family).
Thanks, Dave.
Recently, I have started tracking hsCRP, Ferritin, ALT, Homocysteine, and Lp(a).
During the last 7 years, until I switched to LCHF 5 months ago, my ALT was always elevated. Doctors told me to cut fat, which never worked.
I have tabulated most of my lab tests at:
https://docs.google.com/spreadsheets/d/1IpPeuPXV-deXjexaR3uXwGjhBxaY2EuBlrTeR7dGbzE/
Many in the LCHF community say that a high carb diet causes the creation of small LDLs.
5 months into LCHF, my small LDL-P has remained the same.
However, I have continued to lose weight during these 5 months.
I wonder whether the constant efflux of cholesterol stored in my fat cells, which were created during my high carb days, is causing my small LDL-P to remain high. If so, I hope my small LDL-P reduces when I stabilize my weight.
On LCHF, has your small LDL-P ever risen significantly when cholesterol was not coming out of your adipose tissues: e.g. when you, for many days, consumed at least as much you expended?
I’ve seen a subset of people who are in great shape metabolically but have higher smLDLp, including two Lean Mass Hyper-responders.
As mentioned earlier, I’d be *curious* to run more labs, because it can turn out to be something else that might be bad. Siobhan will actually have a forthcoming piece on this at some point in her series.
But curiosity isn’t always going to lead to concern. I’m not as much a believer in “small dense” being bad in all circumstances because it could potentially be a higher depletion of triglycerides on a per particle basis. Again, being powered by fat changes up the whole paradigm as we know it.
I’ve been doing alternate day fasting for 5 years, and I started that after moving to an all organic diet with essentially no processed foods. Some years before the alternate fasting, my triglycerides were at 100, and in those days I had a sweet tooth. I just got them measured after an eating day (so after ~12 or so hours of fasting) and they were 340.
Another guy on reddit who does intermittent (not alternate) fasting, says his triglycerides are at 52.
I’m not sure what to make of this.
(I’ll add that ADF was very natural for me. It was like discovering what I had been looking for, for years.)
Ok, so I’ve just watched one of your talks. Here’s more data on my lipids, and this is from a person who has been doing ADF for 5 years, loves it, and has a steady weight. This was measured the day after an eating day, and so on 12 or 13 hours of fasting:
Total cholesterol: 184 mg/dL
HDL: 53
LDL: 63
Trig: 340
So my Trig’s are high, but my LDL’s are low. Hm. Do I understand correctly that it’s the LDL’s shuttling around Trigs? If one’s high, shouldn’t the other be?
Your numbers are very, very curious to me.
Are you doing ADF and keto, or are you on a more of a general whole foods diet? Can you comment some of your staple meals overall to get a sense of your eating composition?
Thanks for your reply. They seem curious to everyone else involved, too. To answer your question, no, I am not doing a ketogenic diet on my eating days. I did try a higher fat low carb diet at one point, prior to doing ADF, and found it didn’t work for me. It made me tired and foggy. Presently, my diet consists of basically: veggies, fruit, eggs, chicken, a little rice, a little fermented, gluten free bread, a small amount of nuts & seeds along with some coconut.
I have just made a modification due to a UTI: I have removed all fruit. I will replace that with more veggies and … well, we’ll see what feels right. I plan to retest after the UTI is resolved, and see what happens.
I don’t know if it’s useful, but I do have some other lipids results from some years before, when I was doing nothing special for diet, other than whole, organic, foods. I was also a long-time, 28 yr, vegetarian. I would have to say that at that time, I had a bit of a craving for both sugars and fats.
Total cholesterol: 189 mg/dL
HDL: 50
LDL: 119
VLDL: 20
Trig: 100
So while my total cholesterol and HDL are similar to my recent test, my Trigs are more than three times lower and in the (supposedly) normal range, but my LDL is almost twice as high.
Hi D–
It sounds like you are on a mostly whole foods, mostly plant-based diet that’s very high in carbs, low in fat.
Whether low carb or high carb, high triglycerides can usually mean a few possibilities…
1. You are in an overall energy surplus. I can see this in either LCHF or HCLF (I have a handful of vegans who I work with). Your body is generally “topped off” in glycogen stores and thus you’re turning your carbs to fat (TG via DNL) to shuttle into adipose tissue (body fat), but that too is reaching the threshold. Thus there are less and less places for your lipoproteins to park their energy (TG) leaving more of it floating in the bloodstream. You often see this with Type 2 Diabetes and hyperinsulinemia.
Another clue that suggests this is your low LDL. I’m willing to bet you have a lot of VLDL particles (the precursor to LDL particles) that aren’t getting remodeled from providing energy to your tissues. You can find this out by getting a Nuclear Magnetic Resonance (NMR) blood test. (This can be done privately through a service like RequestATest.com)
2. You are fighting an illness. This can be possible without your even knowing, but it is a reason your LDL could drop down while your TG goes up. Your body is intentionally remodeling and dispatching these particles to deal with inflammation and/or infection. While this is a possibility, it is probably less likely — and (hopefully) it is transient, so it is likely to change by the next time you do a blood test.
3. Genetic? Unlikely given your prior test being what it was… but still possible.
LDL is based on an old calculation lookup table unless it’s actually don in a centrifuge. Very high TG creates a very low LDL, or falsely low LDL. And if your TG is very low you will get a very high LDL reading. If you google how LDL is calculated its very disturbing.
Your “inversion” is nothing more than dehydration. Anytime you fast you must be more hydrated than usual. When you are dehydrated everything in your blood in more concentrated so you are going to see increases in CBC, CMP, Lipids, and decreases in electrolytes. This article is utterly useless.
Interesting comments, Dawson. I certainly agree that I was experiencing some dehydration, but I don’t know how much it would’ve impacted lipid numbers.
To be sure, the inversion pattern has been observed in many experiments to this point, not just this one. If this were unique in my being dehydrated, I’d expect a delta of difference with the other inversions.
On your higher concentration suggestion — wouldn’t we see a likewise change in all particles? If LDL increased by 1.3x, shouldn’t HDL do so as well?
My Name is Douglas Hoover- In the spring of 1975, I read 2 books by Dr. Paul Bragg, Water, The Shocking Truth, and The Miracle of Fasting. I fasted for 19 days to get rid of gout and get off the Zyloprim…Gout disappeared after 6 days. Continued fasting every year for 21 days (distilled water only) and every 5th year fasted for 40 (while working the entire time in my construction company. Fast forward 34 years, I performed 26 21-day fasts and 8 40 days fast. I know very little about biology, medicine or science! However, I do know this for sure, for 43 years I eat meat, dairy, bread, butter, eggs, and bacon etc. along with drinking a minimum of 1 gal. distilled water daily. I started my 9th 40 day fast at 70 yrs. of age January 15 and will be drinking 1.5 gallons of distilled daily and working out at LA Fitness 1 hour three days a week. I had complete blood work and physical at Scripps Clinic in LaJolla CA prior to starting. I’m documenting this fast and will be publishing a book, “Diary of a 40- Day Fast”. First Fast I ever started and finished with medical monitoring. My faith in MDs, ND, and PhDs is not so very strong. If you Google ” “drinking distilled water can bring early death” you will find this exact statement has been quoted by 1,300,000 times. I do, and I have not been to a doctor for 34 years for sickness and never took any drugs! My wife is in the same program! We are on the 6th day. The book will be available on FastingAuthority.com when it is finished. Get a free copy of my last best seller sold on Amazon/ Kindle at DistilledWaterAndHealth.com A man with an experience is not at the mercy of a man with an argument. Dave, you more than likely will delete this but I shared it primarily with you for your own benefit.
Incredible story, Douglas.
Sure, please ping here again when you have the data in an article of some kind. 🙂
Just last week i finished a 12 day water( smart water ) fast. Some himalayan salt on several days only was added. Ht 6’2″ and 210 at the start. End day 12 at still 6’2″ : ), and 187. Mild fatigue yet energy for everything. Wish I did Pre labs but results are …Chol 210, Tri 91, Hdl 40, Ldl 152, Glu 63……4 years ago last results and only major change was Tri from 59 to 91 also Small particle number was quite high “” . 2 years ago I did a 22 day water only and felt fantastic from day 4 to 22…..No change in work schedule through out ….i daily intermittent fast as well…..
Hi Derrick — that makes for a highly “modified” blood test, of course. I’d be curious what your numbers would be at baseline eating to compare with (as in, just 12-14 hours via recommended fast for bloodwork). 🙂
My question is, why would an alarming feeling dissuade you from fasting? It’s a stressor, it’s not supposed to feel amazing. Also, the first 2 days are the worst, so if you stop there you’ll never really feel the “good” part people talk about (good in a way that is hard to describe).
Would you not work out because it made you out of breath, gave you a burning in your muscles, or made you dizzy/lightheaded? Of course you do, b/c you know it’s the recovery from that stressor that makes you stronger. It’s the recovery from the fasting that makes you stronger at the cellular level, that increases mitochondrial density/efficiency, insulin sensitivity, etc.
Hi Greg–
Indeed, I’ve often wondered myself if going beyond that second day would prove important, particularly since many I know insist the second day is the hardest.
That said, I am on the cusp of being a Lean Mass Hyper-responder when fully keto. I find many LMHRs have an especially difficult time with multi-day fasting, and I think this is because they are already in a “blended fasted state” as it were — very low fasting insulin and extremely high fatty acid usage (as a proportion, especially).
This was really fascinating. I had an abnormally high cholesterol reading (also after losing 25 lbs) after fasting for 24 hours before because I thought I might get better numbers. I was also taking Red yeast rice supplement which is a TERRIBLE idea. My numbers were through the roof which was crazy because I had gone Vegan etc. New test tomorrow but I will definitely eat tonight to see if there is a difference. Although my HDL is crazy high at 85 the LDL was off the charts bad which was really confusing.
That’s great, Erica — I’m excited to see the new data. 🙂
I went to get blood work done after my 5th day of water-only fasting. The LDL results alarmed my doctor, but not me.
https://www.ncbi.nlm.nih.gov/pubmed/10539776
Glad you went in knowing the “why” of it. 🙂
I have been on 16-8 Fast 7 days a week for around 8 months now. I am a vegetarian. I quit milk and egg as well. Except for LDL and Total Cholesterol, all other markers improved remarkably. I lost around 15 lbs, now I weigh156 lbs and feel better than I have ever been. I am in my late 40’s.Exercise/gym 5 days a week. Cardio includes Running outside 7.5 miles on Sundays and around 3 miles on Treadmill. Should I be worried?
A1C – 6.5 to 5.7
Fasting Sugar – 117 to 95
Testosterone – 290 to 472
Triglyceride – 129 to 107
HDL – 45 to 52
CRP – 0.93 to 0.7
Homocyst(e)ine – 7.6 to 9.7
TC – 205 to 227
LDL – 130 to 153
Hi Vasu,
First off – congrats on your weight loss! I am glad you are feeling better.
One question before commenting – are you following a low carb, high fat diet? Just curious.
While I can’t really comment on whether or not you should worry (we’re not doctors after all 🙂 ) I can comment on a few things.
For one, I would agree that even by conventional standards all other markers do appear to be trending in a good direction. While I personally prefer to see triglycerides below 100 mg/dL if it continues to go down from the studies I’ve read on various risk markers that would be considered a good thing.
As for LDL and total cholesterol, I do tend to see people who get closer to the leaner side have increased levels, especially if they’re more active, although it’s also worth mentioning that LDL and total cholesterol can fluctuate based off of what you’ve eaten in the past 3 days. Dave talks about this here…
If you are following a high fat low carb diet it could also be contributing, because LDL particles are used to carry fat that we use for energy and based on what I’ve read increased reliance on fat for energy can increase LDL as a result, although it’s worth mentioning that based off of #4 on this list, if you combine low triglycerides, with high HDL, LDL doesn’t seem to have much impact on risk.
Do you take electrolytes prior to a blood draw?
No, I don’t. I drink water only in the 12-14 hours before I get one.
Hi, very interesting read… I just want to know if I should be alarmed over heightened levels of LDL due to intermittent fasting?
Last year from Sept to November I tried IF to various levels… sometimes 16 hours, sometimes 24. I lost 6kg from these, pretty happy but quite shocked over LDL results… both me and my doctor. He wants to put me on satins.
It seems to suggest that the body is using fats for energy during fasting thats why there are so much LDL floating around in the blood stream. But would it cause heart problems in the future?
I most certainly don’t think so. In fact, my research exposes why this is just an echo of one lipoprotein (VLDLs from stored fat) compensating for the lack of different lipoprotein (chylomicrons for incoming food). The added benefit of fasting for autophagy, etc would likely result in better cardio health — not worse — so long as the fast is well managed (keeping up on electrolytes, etc).
Exit question, why don’t we have a whole epidemic of monks dying of heart attacks from all the fasting they do?
Thanks for this article! I’ve done a blood test today and my doctor wants me in the morgue immediately. I am on keto and I did 4 days fast last week. (I’ve ended the fast 4 days prior to the test and ate keto during these 4 days).
LDL from 241 to 324 (!!!)
HDL from 59 to 68
Trig from 94 to 169
Total ch from 319.6 to 426
My doctor is going NUTS 🙂 Actually, I am a bit worried, too.
When can I expect the numbers to go down?
Thanks!
I actually don’t think fasting four days prior to the test would have a major impact like that unless you were under-eating before the test.
I’m not sure how long you’ve been low carb, but I don’t usually see triglycerides going up on a normal 12-14 hour fasted blood test.
So, one thing you might want to try is to re-test… also here’s some common things people run into that might confound their trigs in particular:
* Not water-fasting 12-14 hours before the blood draw
* Some people experience higher trigs if they drink coffee on the morning of the test, or frequently in the days leading up
* Some people experience higher trigs if they consume MCT or coconut oil in the days leading up
Do you perhaps do any of those that might have impacted the results besides the 4 day fast a few days before?
But Dave’s trigs also skyrocketed from 76 to 205….
And mine are even lower. I have done a full 4 days zero calories fast.
After that I was eating normal keto food for 4 days, then took a test. Probably, the fasting influence lasts longer…
Now I MUST show much more lower numbers in the next test or my doctor will die from a heart attack!
What do you recommend: to do a three days of Feldman protocol or to exercise as hard as hell. And how long before the test (including in the morning of the test?) ? Thank you!
Yes, but his bloodwork was done *directly after* (technically during) the fast. Definitely possible that the fasting numbers may have impacted several days after, so you can avoid that for the next test just to make sure it doesn’t influence anything in the next one.
You could definitely try the protocol (5 days of high calorie, as much extra calories from fat as possible, e.g. fatty meats/cheeses/etc), no coffee during this time period (in case it impacted before) only water would be ideal if possible, no MCT or coconut oil if possible, 12-14 hours fasted before the blood draw.
Hi, thanks for the article and your continuing research into this field. I have a few questions, but first a little context.
Ive been keto for the past year and a half with the exception of a 2 week vacation and a few cheat days here and there. In that time Ive lost 44lbs while maintaining muscle mass along with gaining all the benefits of a keto lifestyle.
I did my first ever lipid profile on June 14th which was the last day of Ramadan (no food or water from sunup to sunset). Ive been intermittent fasting (16 – 20hr fasts) for the better part of the past year and dry fasting (15hr fasts) during the month of Ramadan. I generally drink lots of coffee, more so during Ramadan and I dont consume much coconut oil maybe 20g thrice a week. I had a high calorie keto meal the night before along with a usual amount of coffee. I had blood drawn on hour 10 of a dry fast. I am 6’1, 195lbs, 22% BF.
My results all in mg/dl:
Total serum cholesterol 261
HDL – C 57
LDL – C 181
Triglycerides 164
Ive been researching the health related aspects of keto since I started the lifestyle so Im not worried about the high cholesterol, however I was surprised with the high trig results. My questions:
1) How does caffeine cause high trig numbers?
2) Could the dry fasting throughout the month or on the day of have caused a spike in my trig numbers?
3) I plan on retesting after 3 months, Ive already cut out caffeine with the exception of 16g of 99% chocolate on most nights, any tips on how I can lower my trig numbers?
Thanks again and in advance
Yes — While anecdotal for now, we are DEFINITELY seeing a very high correlation between many low carb coffee drinkers and unexpectedly high Triglycerides. Consider cutting it out a minimum of 10 days (preferably longer) and retest.
Fasting can likewise spike TG, but it sounds by what you’re saying that it is intermittent (16-20) and that perhaps you aren’t at a caloric deficit. If you are, I’d expect TG to be up temporarily until weight stable.
Hey Dave thanks for the quick reply.
Yes I am in a calorie deficit. I was under the impression that I since Im fasting I would be using up the trigs for energy, so what is it about fasting the spikes trigs? And why does caffeine increase them as well?
Cheers
I believe there’s a “bunching up” of trigs from fasting/caloric deficit in a temporary response of one mechanism shifting to another.
Imagine you’re a restaurant that runs a large buffet. You have it timed well to get shipments of food to put in the freezer (adipose) along with some of it going straight out to the buffet. And for a long time through practice, you’ve gotten to where you supply your buffet (with fatty acids) just enough to meet the demand (of tissues).
But then suddenly you have lower or no incoming shipments (less incoming energy), so you have a period of time where you need to fill the buffet with more of the walk-in freezer food than usual and that will take a moment to re-pace.
So what do you do? Do you hedge on the side of oversupply or undersupply to your buffet line (fat availability in the bloodstream). You oversupply! Better to deal with oversupply of energy while you get your bearings than under. Then once you have, you can calibrate on this “new normal” of undershipments vs stored (incoming lack of food vs adipose FFA release).
A hyper respinder is usually someone who is lean, right?
Can it be that in a hyper responder, dietairy fat goes really quick to storage and then has to come back from storage as well? When you maintain weight and on low carb, fat either has to be used or stored. The more you use the more you eat, the more that has to go to storage.
The more that comes from storage, the higher Cholesterol goes. Multiple day fasts means all your energy comes from storage,… Thus your cholesterol numbers are super high. But what it really means is more boats in the rivers.
A very good explanation of what is likely going on – I would describe it similarly.
The only difference being in Lean Mass Hyper-responders, there’s also the addition of having very little fat mass. Adipose tissue can supply some energy to the muscle directly adjacent to it, thus in someone who has a lot of fat mass there may be less reliance on “global transport” (lipoproteins), but in lean mass hyper-responders they have very little fat mass and thus the lipoproteins are perhaps more heavily relied upon. Dave touches on that in his talk on LMHRs that you might find interesting.
Interesting article!
I actually went through similar experience. I had my before Keto blood work done and for 3 months i did 1.5 month keto/IF and 1.5 fasting. LEADING to the blood work the second time i had a 12 day fast and saw my cholesterol all increased! My A1c and other markers went to “green” though especially a1c from 8.ish to 5.2
I read and listened to others that the reason behind the increase in cholesterol is due to in a fasted state, your body is burning fat. So there’s tons of fat particles floating around and your liver is processing all that. So naturally when taken blood test the cholesterol numbers will be wildly distorted.
Indeed. The LDL particles carrying the cholesterol have the first job of also transporting fat (triglycerides) from fat tissue to cells that need it.
Hence when you are relying more (or, I suppose, entirely) on fat from storage, LDL skyrockets to meet that demand. 🙂
Thank goodness! I just had my yearly health screening and heavily fasted for the week before and my LDL was triple what it usually is. All other metrics were fine. I’m glad I found this page.
OT: My cousin’s name is Siobhan but it quickly morphed into “Chibby” by her older siblings and her cousins
Thank You, Siobhan and have a great day!
ive been fasting for a few months now round a bout 12-14hrs my waist circumference has dropped by 4.5 inches and a drop of 10lbs body weight ,my current weight is 10st 11lbs ,i train every day walk /run about 25.35 miles weekly ,train and teach karate and also do light weights weekly. my diet consists of mackeral,tuna.sardines walnuts,beans and two eggs every day chicken and lean red meat occasionally ,i try to keep to a low card diet with plenty green veg ,i also have good energy levels my age is 64,my last cholesterol test showed the following ,serum cholesterol 7.2 mmol/l serum triglycerides 0.6 mmol/l ,serum hdl cholesterol 2.9mmol/l serum cholesterol ratio 2.5 se non hdl cholesterol 4.3mmol/l my results were classed as abnormal and im being referred to my doctor , any body have any thoughts on these results .
I copied this from our report tool after inputting your numbers:
–CHOLESTEROL REMNANTS–
Remnant Cholesterol: 11 mg/dL 0.28 mmol/L >>> Lowest Risk Quintile
Remnant Chol to HDL: 0.1 >>> Lowest Risk Quintile
Go to https://tinyurl.com/y8hokam2 for more on Cholesterol Remnants
–ATHEROGENIC INDEX OF PLASMA (AIP)–
AIP: -0.684 >>> Lowest Risk Third
Go to https://tinyurl.com/ycccmmnx for more on Atherogenic Index of Plasma
If I had those numbers I’d be pretty pleased 🙂
It looks like you fit into the profile of a Lean Mass Hyper-responder who does endurance/resistance training (hence lower LDL).
You may find this presentation and our facebook group useful.
Of course, when one fasts, he is forced to burn fat released to the blood stream as triglycerides, so naturally triglycerides would rise.
Indeed, and this is reflected in the literature. It’s not so much a consequence of burning fat alone (people on low carb high fat diets often have low triglycerides), but rather it seems the body overcompensates with how much is released, and after about 3 days (in other studies I’ve seen) it figures out how much is generally needed and peaks out there. Better to have too much energy around than too little, when unexpected fasting is involved, perhaps.
I too felt the way you did. I’ve done a couple of 72 hour water fasts trying to get to the ‘no hunget’ part but never got there. I’m very lean and active so I wonder if that’s a disadvantage for being able to do an extended fast. My gut feeling is that having a fast metabolism and efficient fuel delivery system that there is more of a catabolic effect that happens. I used to wrestle and would train hard and basically do a dry fast and not suffer during a match but that wasn’t for an extended period of time.
Yep. I sympathize.