Ketofest Cholesterol Experiment – The Hypothesis

As detailed previously, participants of this experiment will consume food in two stages:

  1. Low Days – Low calorie consumption or fasting for three days, while still at a low-carb, high-fat ratio (LCFH diet) : July 11th, 12th, and 13th
  2. High Days – High calorie consumption for three days, while still at a low-carb, high-fat ratio (LCFH diet) : July 14th, 15th, and 16th

Cholesterol tests will be taken on the morning of the day following each stage.

Following the second phase of higher dietary fat, I predict the resulting LDL cholesterol for the majority of participants will show a decrease when compared to the first.

0 0 vote
Article Rating
Notify of
Newest Most Voted
Inline Feedbacks
View all comments
Lee Sandwith
3 years ago

Hey Dave,

Just stumbled across your work and as a hyper-responder myself I have to say I am very happy that you have put some much thought, effort and dedication into this. Amazing work!

I’ve been doing some experiments myself but not to the level of detail as you. I have missed the one you are posting about here but I’ll definitely look to get involved in one shortly.

I’ve documented my experiences to date on my own website; if you could take a look at the below article when you get a chance I would be very grateful although I do understand that time is precious.

To summarise, at my peak, my Total-C was 455 mg/dl. My GP wanted me to get onto statins immediately but I promised her that I could make a dramatic improvement in a matter of weeks. I stayed keto but cut out one thing: a keto coffee. After two weeks my Total-C was only slightly over 300 mg/dl. My GP was speechless.

Anyway, would love to here back from you. Here’s a link to the post in question:

Best, Lee

3 years ago
Reply to  Dave

This is interesting on a personal take – I had been eating a bunch of coconut oil in my food along with some MCT on workout days – also ended up with a really high LDL-C.

I’m not worried about it – because the fatty-acids are saturated, they are unlikely to become oxLDL. – The receptors in the macrophages/foamcells are specific for oxLDL – apparently the particles appear as a dead or dying bacteria.

There are several drugs that lower LDL, yet have no effect on CAD – thus I’m quite doubtful that LDL is causative. Much more interesting is Kendrick’s ideas about insulin and stress.

Lee Sandwith
3 years ago
Reply to  Dave

Hi Dave,

Thank you for coming back to me so quickly and for taking the time to review my post. I had no idea about the workings of MCT, I just learned that it was beneficial in terms of promoting ketosis; from Tim Ferriss’ podcast with Dom D’Agostino I think. I just suspected that it was at the route of my 455 mg/dl result as I was actually consuming quite a few of those keto coffees in the days running up to the test.

On tracking: I’m a it of a freak as I have been tracking EVERYTHING in MyFitnessal for over a year now, essentially so all of the macro and micronutrient info is available when needed. What I haven’t been doing, though, is tracking supplements, water etc as per your tracking recommendations. I do weigh my food though so the info I have should be pretty accurate, although there is exposure when eating out of course.

I am considering doing a more controlled experiment to isolate a couple of things so could look at the MCT powder again, and maybe play with the volume. I’m also keen to isolate coffee as I suspect that this may also be having an impact (as per BJJ Caveman’s experiments).

If I design an experiment, would you be interested in taking a look at it before I start to ensure it’s on point? I live in Dubai so unfortunately the blood work is very expensive, however, I now have a way to get ApoB tested so the data could be very good.

Also, do you know of any online physicians/doctors who could analyse my lipid profile? I’m having little luck in Dubai as all of the GPs and even Cardiologists are very old school, looking at Total-C and LDL-C only, and talking about statins!!

Cheers, Lee

3 years ago

It may be the butter in the coffee, not the MCT oil, that is causing the LDL spike.

Lee Sandwith
3 years ago
Reply to  Doug

Hi Doug, thanks for taking the time to post here. I created my own version of the keto coffee which only used coconut oil and MCT Oil Powder so it can’t be butter. Planning on isolating it in an experiment shortly so will post results 🙂 Cheers, Lee

3 years ago
Reply to  Lee Sandwith

Fascinated reading the exchange between you and Dave–thanks for sharing it here. May I ask what you use for your fats, i.e. in cooking, etc. Did you cut out coconut oil/mct oil across the board or just the extra ‘bombs’ in your coffee?

3 years ago

I just ran across a similar experiment in pubmed where a high energy meal before the next morning blood draw lowered LDL-C. They blamed it on the Friedewald equation failing in the case of a high calorie prev meal.

They used carbs to raise calories, rather than fat.n Of course the difference is nothing like what you’re observing, but it’s only one meal. I thought it was interesting.

Karen H
Karen H
3 years ago


My total cholesterol hangs basically around 240-260. A couple of years ago I stopped at my office before going for my lab work (for which I had dutifully fasted), and had a cup of coffee with cream. I didn’t realize what I had done until I got to the lab and the phlebotomist asked if I had fasted. I said yes, but then made a mental note that, oops, I had the cream. When I got my lab results back, I was blown away. The total cholesterol was 194. First time in decades it had been that low. I am guessing it was the cream.

I will have to try the cream experiment again.

3 years ago
Reply to  Karen H


I did that once, too. My doc told me “we’ll adjust for that” (not that they know how).

Depending on the timing and the amount of cream, there might be some extra trigs and cholesterol in Chylomicrons. So if anything, I’d expect your TC to be slightly higher. A few other things can lower cholesterol though, like exercise.

Those extra trigs would cause your calculated LDL to be lower, since the formula assumes you shouldn’t have any of those. I’d be interesting to look at the full lipid number if you still have them.


Karen H
Karen H
3 years ago


I will dig out my numbers. But yes the trigs are a problem.

3 years ago

Chol 231/HDL 52/LDL Calc 155/Trig 122
Next year – 245/56/157/160
Cream Year – 198/51/119/138 (remembered “194,” but was 198)
Next year – 246/48/165/165
Next year – 252/52/154/232

I’ll share more off line. Email me.

Would love your thoughts, please comment.x