Jan 03

The Capstone Experiment – it’s time to #rethinkLDL

As of today, I’m declaring Phase I and II of my research complete.

  • Phase I spans December 2015 to May 2017 where I was focused primarily on the Inversion Pattern.
  • Phase II spans May 2017 to January 2018 where I was focused on Energy Status.

The Capstone Experiment combines both legs of this theory and demonstrates how fluid and agile cholesterol markers are when modifying dietary sources of energy in real time.

Naturally, there are many questions still to answer given how much there is to unpack. But for now, I want to put a white-hot spotlight on the most central and unmistakable observation of this experiment:

Cholesterol is a passenger, not a driver. Its markers are extremely agile and highly influenced by the larger energy metabolism. This runs counter to medical mainstream opinion.

Next Steps

  • Certainly, I’d like to put this into a formal study. But the process of organizing these and raising money tends to be slow, which is why I haven’t taken a lot of time pursuing it to this point. I’m an engineer, not an academic, so I’d prefer to find someone who can better navigate those waters to make this happen.
  • Falsify this! If you’re reading this and you have alternate theories or a way I can disprove this working theory, please let me know. I can’t emphasize enough that I appreciate contrarian views, so long as they are materially relevant and productive.
  • Lastly, I’ll be taking a bit more personal time in the next week to rest and catch up some time with family. Everyone has been very patient and understanding of my work through this point, chief among them being my wife. These experiments take a lot out of me, with this one being the most intensive one yet.


Skip to comment form

  1. Maria

    Hi Dave,

    First of all, thank you for your work. It is absolutely amazing and it cant be appreciated enough.

    But one think appears to me to not make sense:
    Why is higher LDL associated with a longer life? (David Diamond)
    First some thoughts on why it shouldnt happen:
    -high VLDL due to high DNL. Basically the metabolic syndrom typ. This clearly wont make one live longer, rather the opposite.
    -the closer someone is to death, the less appetite this person has, the less he eats, the higher his LDL will be (classical inversion pattern). But again the lower appetite thus higher LDL near the end of life should predict the opposite to be true (shorter life and early death).
    -other things that rise LDL: stress, LPS and other toxins that cause inflammation etc. Again dealing with those things wont make one live longer..

    So far, every condition that elevates LDL is clearly not a sign of health that would explain why LDL is associated with longer live. For me only 3 possibilities remain:
    -those people with high LDL that live longer are on a LCHF diet and may restrict kcal due to satiety reasons. This would elevate LDL and could explain why they live longer. (because they live (eat) healthier)
    -maybe they (their bodies) have not lost the ability to upregulate LDL when repair/healing is required. maybe the really sick people or people with damaged livers have lost this ability. So their livers cant supply the required LDL anymore thus those die earlier. (which would allow those that can upregulate LDL to survive (live) longer… which fits Diamonds research)
    -Ch is somehow healthy and protective on its own.

    But overall, i cant explain why higher LDL is associated with longer live. As i have written, they opposite should be true. The last 3 points, that theoretically could explain it are not very convincing in my opinion (what do you think?). For example i have strong doubts that anyone at a higher age eats a LCHF diet.

    So my question for you: whats your thoughts on that? do you have any explainations?

    Thanks again for all your work.

    1. Dave

      Hi Maria-

      – Actually, high VLDLs that *don’t* make their way to becoming LDLs is a big sign in met syn. An upcoming article I’ll be writing is on Cholesterol Remnants which break out that nonHDL and nonLDL from TC, leaving only VLDLs, IDLs, and Chylomicron Remnants. These last three sitting in the blood at larger quantities is HIGHLY associated with CVD and overall insulin resistance.

      – Not sure I totally follow your second point…

      – “other things that rise LDL: stress, LPS and other toxins that cause inflammation etc. Again dealing with those things wont make one live longer..” — agreed. This is why in literature here and on my Basics video I emphasize there can be both good and bad reasons LDL would be higher (and these two aren’t mutually exclusive).

      – “So far, every condition that elevates LDL is clearly not a sign of health that would explain why LDL is associated with longer live.” — How about being powered by fat and NOT having the other things you’ve mentioned? If so, you’ll generally have a lower inflammatory state that probably has a higher degree of autophagy.

      1. Annie Q

        Maybe the body creates more LDL to cope with stress and other unhealthy conditions. Thus LDL would be associated with these unhealthy conditions not as a cause but as the body’s remedy.

  2. Maria

    Oh WOW, i did not think of that. Thank you !!

    In this context, how trustworthy is a standard lipid panel in terms or LDL in your opinion? Does the Friedewald-Formula provide a clue if TG are not to high (<150) or would you always test them directly?

    1. Dave

      The jury is still out for me with the Friedewald equation. I want to do a run of lab tests soon that do a direct measurement for comparison (NMRs are still FE calculated, annoyingly).

  3. Dominik

    really interesting Dave,

    Do i get it right: stress, LPS, toxicity, injuries etc lower LDL and TC when it occurs acute and short term (used up by body); whereas increase LDL and TC when they are chronic (upregulation by body), right?

    concerning the VLDL: so ramped up DNL, producing lots of VLDL doesnt (necessarily?) increase LDL?

    1. Dave

      To be sure, I can’t speak on this too much yet. I know that typically if a *managable* illness (such as infection) results in a change of lipids, it is usually higher. An end-of-life or deadly level illness can often result in surprisingly low LDL (such as sepsis) since you can have that and/or organ shutdown, etc. Good stuff!

  4. Jeremy

    How can you make your conclusions based on only 2 or 3 data points per phase?

    1. Dave

      This experiment was actually directed by prior experiments leading up to this one. Both the Inversion Pattern and the Energy Status phases of my research have a lot of data to them (see prior blog posts detailing each).

  5. Heather Varaleau

    Dear Dave Feldman,

    I’m trying to gather information on the risks (if any) of very low LDL.

    I am part of a vegan-keto facebook group on which a few participants have reported very low LDL numbers (in the 40-50 range). Their doctors are encouraging and supportive of these very Low LDL numbers, and I am concerned that there are health consequences to such low LDL.

    I have tried to find reliable information on the impacts of very low LDL, but have found little. I have read this article (http://onlinelibrary.wiley.com/doi/10.1111/joim.12614/abstract) which generally concluded little to no risk, but is sponsored by statin producers.

    Do you have an opinion, or can you point me to resources/research detailing/studying the risks (if any) of having very low LDL?

    Thank you

    1. Dave

      What you’ll find (and I agree with) is that very low LDL is associated with reduced (or practically non-existent) CVD. But this is like saying you’ll reduce the chance you’ll die in a hospital by outlawing hospitals. As always, I come back around to All Cause Mortality (ACM). Any/all studies that don’t include ACM but claim a mortality benefit is effectively wasted time to me.

      This is why I have the Cyanide Diet example — where something that kills you sooner, necessarily reduces the risk of everything else killing you instead. The ONLY way to isolate these shenanigans is to always have ACM. Thus, if I’m more likely to die of CVD at 70 because I’m less likely to die of cancer at 60, I’ll take it!

      So with that said, the real question for me is whether low cholesterol is associated with low ACM — and there are quite a few studies that definitely indicate it is the opposite. Here are two I like to often that are fairly recent and very large in population:

      Japan Health Study (91,219)
      Go straight to Figure 1-1

      Norwegian HUNT study (52,087)
      Note the risk of All Cause in this figure: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3303886/figure/fig02/

      In both studies, the lowest cholesterol group faired the worst in mortality from all cause.

      Okay, now here is where I try to be a good scientist and say what’s good for the goose is good for the gander. For the same reason cholesterol can be high for good reasons, bad reasons, or both — I likewise believe cholesterol can be low for good reasons, bad reasons, or both. Much of my experiments are about isolating the energy metabolizing impact with diet. In the case of vegans, they are typically high carb, low fat, and thus are simply not mobilizing as much fat-based energy in favor of glucose instead. Thus, it could be just fine for their cholesterol to be lower overall and their not suffering the fates of those from the studies linked above.

      However, I do speculate there’s a bottom threshold to this because those same lipoproteins that transport energy from fat also transport fat-soluble vitamins (A, D, E, and K) and cholesterol for cellular repair as well as recycling back for hormones and bile salts. One common scenario with vegan women is losing their menstrual cycle, which I do suspect is a red flag regarding sex hormone production/status.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>