The personal blog of Peter Attia, M.D.

The straight dope on cholesterol – Part III

The straight dope on cholesterol – Part III



Previously, in Part I and Part II of this series, we addressed 4 concepts:

   #1What is cholesterol?

   #2What is the relationship between the cholesterol we eat and the cholesterol in our body?

   #3Is cholesterol bad?

   #4 How does cholesterol move around our body?

This week we’ll address the following concept:

   #5 How do we measure cholesterol?


Quick refresher on take-away points from previous posts, should you need it


  1. Cholesterol is “just” another fancy organic molecule in our body but with an interesting distinction: we eat it, we make it, we store it, and we excrete it – all in different amounts.
  2. The pool of cholesterol in our body is essential for life.  No cholesterol = no life.
  3. Cholesterol exists in 2 formsunesterified or “free” (UC) and esterified (CE) – and the form determines if we can absorb it or not, or store it or not (among other things).
  4. Much of the cholesterol we eat is in the form of CE. It is not absorbed and is excreted by our gut (i.e., leaves our body in stool). The reason this occurs is that CE not only has to be de-esterified, but it competes for absorption with the vastly larger amounts of UC supplied by the biliary route.
  5. Re-absorption of the cholesterol we synthesize in our body (i.e., endogenous produced cholesterol) is the dominant source of the cholesterol in our body. That is, most of the cholesterol in our body is made by our body.
  6. The process of regulating cholesterol is very complex and multifaceted with multiple layers of control.  I’ve only touched on the absorption side, but the synthesis side is also complex and highly regulated. You will see that synthesis and absorption are very interrelated.
  7. Eating cholesterol has very little impact on the cholesterol levels in your body. This is a fact, not my opinion.  Anyone who tells you different is, at best, ignorant of this topic.  At worst, they are a deliberate charlatan. Years ago the Canadian Guidelines removed the limitation of dietary cholesterol. The rest of the world, especially the United States, needs to catch up.  To see an important reference on this topic, please look here.
  8. Cholesterol and triglycerides are not soluble in plasma (i.e., they can’t dissolve in water) and are therefore said to be hydrophobic.
  9. To be carried anywhere in our body, say from your liver to your coronary artery, they need to be carried by a special protein-wrapped transport vessel called a lipoprotein.
  10. As these “ships” called lipoproteins leave the liver they undergo a process of maturation where they shed much of their triglyceride “cargo” in the form of free fatty acid, and doing so makes them smaller and richer in cholesterol.
  11. Special proteins, apoproteins, play an important role in moving lipoproteins around the body and facilitating their interactions with other cells.  The most important of these are the apoB class, residing on VLDL, IDL, and LDL particles, and the apoA-I class, residing for the most part on the HDL particles.
  12. Cholesterol transport in plasma occurs in both directions, from the liver and small intestine towards the periphery and back to the liver and small intestine (the “gut”).
  13. The major function of the apoB-containing particles is to traffic energy (triglycerides) to muscles and phospholipids to all cells. Their cholesterol is trafficked back to the liver. The apoA-I containing particles traffic cholesterol to steroidogenic tissues, adipocytes (a storage organ for cholesterol ester) and ultimately back to the liver, gut, or steroidogenic tissue.
  14. All lipoproteins are part of the human lipid transportation system and work harmoniously together to efficiently traffic lipids. As you are probably starting to appreciate, the trafficking pattern is highly complex and the lipoproteins constantly exchange their core and surface lipids. This is a big reason why measuring how much cholesterol is within various lipoprotein species will in many circumstances be so misleading, as we’ll discuss subsequently in this series.



Concept #5 – How do we measure cholesterol?

All this talk about cholesterol probably has some of you wondering how one actually measures the stuff.  Much of the raw content I’m going to present here is actually material I’ve had to learn recently.  One of the best resources I’ve found on this topic is the text book Contemporary Cardiology: Therapeutic Lipidology, in particular, chapter 14 by Tom Dayspring and chapter 15 by Bill Cromwell and Jim Otvos.  Anyone aspiring to be a lipid savant like these three pioneers probably ought to get a copy.  The other book that tells this story well is The Cholesterol Wars: The Skeptics versus the Preponderance of Evidence. For most folks, however, I’m hoping this series is sufficient and I’ll do my best to get the important points across.

As far back as the 1940’s scientists understood that cholesterol and lipids could not simply travel freely within the bloodstream without something to carry them and obscure their hydrophobicity, but it certainly wasn’t clear what these carriers looked like.

The initial breakthrough came during the Second World War when two researchers, E.J. Cohn and J.L. Oncley at Harvard developed a complex and elaborate technique to fractionate (i.e., separate) human serum (serum is blood, less the cells and clotting factors) into two “classes” of lipoproteins: those with alpha mobility and those with beta mobility.  [“Alpha” versus “beta” mobility describes a pattern of movement seen by different particles, relative to fluid, under a uniform electric field, which is the essence of electrophoresis.]

You’ll recall that LDL particles are also called “beta” particles and HDL particles are also called “alpha” particles.  Now you see why.

This work set the stage for subsequent work, by a physicist named John Gofman, using the techniques of preparative and analytic ultracentrifugation to fully classify the major classes of human lipoproteins. The table below summarizes what was gleaned by these experiments.


lipoprotein characteristics

Cool, huh?  Well, sort of.  While this was an enormous breakthrough scientifically, it didn’t really have an inexpensive and quick test that could be used clinically the way, say, one could measure glucose levels or hemoglobin levels in patients routinely. What became crucial with Gofman’s discovery is that lipoproteins were now a recognized entity and they got their names according to their buoyancy: very low density, intermediate density, low density and high density.

There is more interesting history to this tale, but let’s fast-forward to where we are today.  When you go to your doctor to have your cholesterol levels checked, what do they actually do?

Let’s start at the finish line. What do they report? The figure below is a representative result.  It reports serum cholesterol (in total), serum triglycerides, HDL cholesterol (i.e., HDL-C), LDL cholesterol (i.e., LDL-C) and sometimes non-HDL-C (i.e., LDL-C + VLDL-C).  But where do these numbers come from?


Blood is drawn into a tube called a serum separator tube (SST) and immediately spun in centrifuge to separate the blood from “whole blood” into serum (normally clear yellow, top) and blood cells (dark red, bottom). A gel film, from the SST, separates the serum and blood cells, as shown below.  The tube is kept cool and sent from the phlebotomy lab to the processing lab.

SS tube

As early as the 1950’s scientists figured out clever chemical tricks to directly measure the content of total cholesterol in the serum.  The chemical details probably are not interesting to non-chemists, but I was able to find a great paper from 1961 that details the methodologyThe point is this: initially it was only possible to measure the total content of cholesterol (TC), or concentration to be technically correct, in plasma. By that I mean it is the total mass (weight of all the cholesterol molecules) of cholesterol trafficked within all of the lipoprotein species that exist in a specified unit of volume: in the United States, we measure this in milligram of cholesterol per deciliter of plasma abbreviated as mg/dL, or in the rest of the world as mmol/Liter or mmol/L.  Why?  Think back to our analogy from last week:

Cholesterol is a passenger on a ship — the “ship,” of course, being a lipoprotein particle. The early methods of measuring cholesterol had to break apart the hull of the ship to quantify the cargo.  The assays to do so, like the one described above, were pretty harsh.  If you had a bunch of LDL ships, HDL ships, VLDL ships, and IDL ships, these assays ripped them all apart and told you the sum total of the cargo.  Obviously this was a great breakthrough in the day, but it was limited. From this assay, one could conclude, for example, that a person had 200 mg/dL of cholesterol hiding out in all their lipoprotein particles.

Good to know, but what next? It turns out there were two other important factors that could be measured directly in blood: triglycerides and the cholesterol content within the HDL particle, HDL-C. Early on laboratories could easily separate apoA-I-containing particles (i.e., HDL) from the apoB-containing particles (i.e., VLDLs, IDLs and LDLs), but they could not easily and economically separate the various apoB-containing particles from one another.  A full description of these methods is not necessary to appreciate this discussion, but for those interested, methodologies can be found here (TG) and here (HDL-C).


Important digression for context

What becomes critical to understand for our subsequent discussions is that the apoB particles have the potential to deliver cholesterol into an artery wall (the problem we’re trying to avoid), and 90-95% of the apoB particles are LDL particles.  Hence, it is LDL particle number (LDL-P or apoB) that drives the particles into the artery wall. Thus, physicians need to be able to quantify the number of LDL particles present in a given individual. For decades there was no way of doing that. Then LDL-C (read on) became available and it served as a way (not entirely accurate, but nonetheless a way) of quantitating LDL particles.

Back to the story

How can one figure out the concentration of cholesterol in the LDL particle? As you may recall from last week, LDL is the “ship” that carries the most cholesterol cargo. More importantly, as I mentioned above, it is also the key ship that traffics cholesterol directly into the artery wall. Thus, there has always been an enormous interest in knowing how much cholesterol is trafficked within LDL particles.

For a long time it was not possible to directly measure LDL-C, the cholesterol content of an LDL particle.  However, we did know the following had to be true:

          TC = LDL-C + HDL-C + VLDL-C + IDL-C + chylomicron-C + remnant-C + Lp(a)-C

where X-C denotes the cholesterol content of a respective cholesterol-carrying particle.  There are 2 particles in the equation above that I didn’t specifically mention last week, the remnant particle and the Lp(a) particle (pronounced “EL – pee – little – a,” which sounds less silly than, “Lip-a”). Lp(a) is an LDL-like particle but with a special apoprotein attached to it, aptly called apoprotein(a), which is actually “attached” to the apoB molecule of a standard LDL particle.  Think of Lp(a) as a “special” kind of LDL particle.  As we’ll learn later in this series, Lp(a) particles are bad dudes when it comes to atherosclerosis.

“Remnants” are nearly-empty-of-triglyceride particles of chylomicrons and VLDL.  In essence they are larger TG-rich particles that have lost a lot of their TG core content as well as surface phospholipids and are thus smaller than, or remnants of, their “parent particles.” Hence,they are cholesterol-rich particles. Under fasting conditions, in a not-too-terribly-insulin-resistant person, IDL-C, chylomicron-C, and remnant-C are negligible.  Furthermore, in most people Lp(a)-C does not exist or is not very high.

So we’re left with this simplification:

          TC ~ LDL-C + HDL-C + VLDL-C

which is clearly an improvement in convenience over the first equation.  But what to do about that pesky VLDL-C?

There are a number of variations, but essentially a breakthrough (mid 1970s) formula, called the Friedewald Formula, estimates VLDL-C as one-fifth the concentration of serum triglycerides (some variants use 0.16 instead of one-fifth, or 0.20).  This assumes all TG are trafficked in one’s VLDL particles and that a normally composed VLDL contains five times more TG than cholesterol.

Rearranging the above simplified formula we have:

           LDL-C ~ TC – HDL-C – TG/5

Let’s plug in the numbers from the above figure, as an example.  TC = 234 mg/dL; HDL-C = 48 mg/dL, and TG = 117 mg/dL.  Hence, LDL-C is approximately 234 – 48 – 117/5 = 163 mg/dL.

Kind of a long run for a short slide, huh? Perhaps, but it is important to understand that when you go to your doctor and get a “cholesterol test,” odds are this is exactly what you’re getting.

Therefore LDL-C can be estimated knowing just TC, HDL-C, and TG, assuming LDL-C matters (hint: it doesn’t matter much in many folks). 

Furthermore, what if the LDL particle is cholesterol-depleted instead of its normal state of being cholesterol-enriched?  Unfortunately, especially in an insulin resistant population (i.e., the United States), both TG content within lipoproteins and the exchange of TG for cholesterol esters between particles is very common, and using this formula can significantly underestimate LDL-C.  Worse yet, LDL-C becomes less meaningful in predicting risk, as I will address next week.


What about direct measurement of LDL-C?

To chronicle the entire history of direct LDL-C measurement is beyond the scope of this post.  Many companies have developed proprietary techniques to measure LDL-C directly, along with apoB, and ultimately LDL-P.  I’ll address two “major players” here: Atherotech and LipoScience.

Atherotech developed an assay, called a VAP panel (VAP stands for Vertical Auto Profile) to do everything described above, but also to directly measure the amount of cholesterol contained within the LDL particle.  Furthermore, they have developed assays to directly measure the cholesterol in IDL particles, VLDL particles, and even Lp(a) particles.  Below is a snapshot of how VAP reporting looks.

VAP results

A couple of things are worth mentioning:

  1. Subparticle cholesterol content information is also generated, including 2 different classes of HDL particles (HDL-2, HDL-3) and 4 different classes of LDL particles (LDL-1, LDL-2, LDL-3, LDL-4).
  2. LDL particles, based on the subparticle information, are classified as “pattern A,” “pattern B,” or “pattern A/B.”  Pattern A implies more large, buoyant LDL particles, while pattern B implies more small, dense LDL particles.

Remember, though, while cholesterol mass concentration numbers may correlate with the number of particles, they often do not.  They only convey the mass concentration of cholesterol molecules within all of the particle subtypes per unit of volume.  VAP tests do not report the number of LDL or HDL particles, but they do attempt to estimate atherogenic particle number (apoB) using a proprietary formula based on subparticle cholesterol concentration and particle sizes.  I should point out that the formula, to my knowledge, has not been validated in any study and not published in a peer reviewed journal.

A high estimate of apoB100 (i.e., what the VAP reports) is said to correlate with the actual measurement of apoB.  Since apoB is found on each LDL particle, this serves as a proxy of LDL-P. The American Diabetic Associate and the American College of Cardiology Consensus Statement on Lipoproteins and the new National Lipid Association biomarker paper stipulates that apoB must be done using a protein immunoassay, not an estimate, such as that of VAP.


But how can one actually count the number of LDL particles and HDL particles?

There are several methods of doing this, but only one company, LipoScience, has the FDA approved technology to do so using nuclear magnetic resonance spectroscopy, or NMR for short. The other available methodologies are ion mobility transfer and ultracentrifugation (by Quest) and separation of LDL particles with particle staining (by Spectracell).  Virtually all guidelines (e.g., ADA, ACC, AACC and NLA) only advise LDL-P via NMR at this time.

NMR, which is the basis for not only how to count lipoprotein particles, but also diagnostic tests such as MRI scans, is really one of my favorite technical topics.  In residency I wrote a surgical handbook and on page145-146, if you’re interested, you can read a brief description of how MRI technology works, which will explain how NMR technology can actually count lipoprotein particles.

As an aside, and just to give you an idea of what a great sport my wife is, I wrote this surgical handbook over the course of a year while in residency.  To do so, I had to read approximately 8,000 pages of surgical textbooks and try to distill them down to just this 160 page summary.  Doing so required reading about 22 pages every day while working about 110 hours per week, typical of a surgical residency “back in the day.”  Besides exercising, I spent every single moment of my “free” time reading for and writing this handbook.  Finally, a few months into it, my wife asked, “Why the hell are you doing this?  You never watch TV, you never go out, you never do anything else!”  I responded that it was the best way I could learn this material, but also, that I wanted to have a legacy when I left residency.  Half joking, I asked her, “What’s your legacy?” Blank stare.  A few months later, for Valentine’s Day, she gave me this t-shirt. I think it’s safe to say not a single person has read this handbook.  So much for my legacy…

What's your legacy

A brief explanation of how NMR works to count (and measure) particles can also be found here.

Below is a snapshot of how NMR reporting looks.  This particular report is from Health Diagnostics Laboratory (HDL), Inc.  LipoScience performs the actual NMR test, but HDL, Inc. runs a number of complimentary biomarkers I will discuss in subsequent posts.  I now use the HDL, Inc. test exclusively for reasons I will explain later.

NMR data

In addition to counting the actual total number of LDL particles (LDL-P) and HDL particles (HDL-P) per liter, HDL, Inc. (not LipoScience) directly measures apoB and apoA-I.  Furthermore, the size of each particle is measured using NMR in nanometers (to give you a sense of how small these things are, and why we need to use nanometers to measure them, about 1.3 million LDL particles stacked side-by-side would measure only one inch).

The final point I’ll make about the value of NMR reported subparticle sizes and diameters is particularly telling when it comes to insulin resistance.  In the panel below, you can see that this person has small VLDL particles, small HDL particles, and LDL particles. Why is this interesting? The presence of increased large VLDL-P, large VLDL size, increased small LDL- P, small LDL size, reduced large HDL-P, small HDL size are early markers for insulin resistance, and such findings may actually precede more conventional signs of insulin resistance (insulin levels, glycemic abnormalities) by several years.  In other words, the number and size of the lipoprotein particles is perhaps the earliest warning sign for insulin resistance.

LP-IR data


In summary

  1. The measurement of cholesterol has undergone a dramatic evolution over the past 70 years with technology at the heart of the advance.
  2. Currently, most people in the United States (and the world for that matter) undergo a “standard” lipid panel which only directly measures TC, TG, and HDL-CLDL-C can be measured directly, but is most often estimated.
  3. More advanced cholesterol measuring tests do exist to directly measure LDL-C (though none are standardized), along with the cholesterol content of other lipoproteins (e.g., VLDL, IDL) or lipoprotein subparticles.
  4. The most frequently used and guideline recommended test that can count the number of particles is the NMR LipoProfile.  In addition to counting the number of particles – the most important predictor of risk – NMR can also measure the size of each lipoprotein particle, which is valuable for predicting insulin resistance in drug naïve patients, before changes are noted in glucose or insulin levels.

I know some of you are getting antsy.  I thank you for your patience, and I hope you appreciate that it was a necessary step to get through this somewhat technical material and nomenclature.  Next week we’ll get to the “fun” stuff – what does all of this cholesterol have to do with heart disease?

In addition, we’ll get further into the importance of using LDL-P as the best predictor of risk.  If anyone wants to read up on another very important topic, especially for understanding why LDL-P is more important to know than LDL-C, get familiar with the concepts of discordant and concordant variables.  You’ll be hearing a lot about these.

(To Part IV »)


About the Author:

Peter Attia, M.D., is a physician in private practice in NYC and CA. His practice focuses on longevity and healthspan. His clinical interests are nutrition, lipidology, endocrinology, and a few other cool things.


  1. David Nelsen  May 10, 2012

    Peter I’m getting blood drawn (for the 4th time) to get my LDL-P from LipoScience. The other 3 were lost, or something… My local lab sends the sample to Quest in California, who theoretically should send it on to LipoScience. How does the Health Diagnostics laboratory fit into the process if LipoScience is doing the scan? Again great stuff. Dave

    • David Nelsen  May 10, 2012

      I think your wife probably figured out a long time ago that you have an obsessive type personality and throw yourself into whatever your current passion is head first. Fortunately for you now, a lot of people are reading your blog – so you’ve got that going for you which is nice (Caddyshack reference).

    • Peter Attia  May 10, 2012

      Doctor, doctor, doctor, doctor, doctor, aaaaaaand doctor! Did I miss anyone? (Spies Like Us reference).

    • Peter Attia  May 10, 2012

      HDL, Inc. sends blood to LipoScience specifically for the NMR. They do a large number of other tests that compliment the NMR.

  2. Michael J Lies  May 10, 2012

    Comment: I am a post-bypass 71 year old. My 15 year old by-pass arteries are doing fine, however, the desending posterior coronary is too pluged to receive a stent !
    I am working hard at the low carb protocol as explained by yourself (for 6 months now) & weight is coming off but with heavy fat ingestion, I am concerned about causing more blockage; should I be concerned ?

    • Peter Attia  May 10, 2012

      Check NMR and inflammatory markers. Ask you doctor to get the full panel through HDL, Inc.

  3. Alexandra M  May 10, 2012

    Damn! When the site was down today, I “sensed” that the next installment was coming. The F5 key on my keygboard is broken now.

    One of the thing I find a little concerning about the tests is that they contain the predictive information of the moment – the same sort of information that makes doctors prescribe statins from the standard tests. They see somewhat elevated LDL-C and they want to write a scrip – it doesn’t matter that HDL is higher and triglycerides are low.

    In short, I wish they would just report the numbers.

    But thank you for another awesome post – I’ll read it again tomorrow (have pity on your East coast readers).

    • Alexandra M  May 10, 2012

      Note to self – always compose in Word which spellchecks and allows editing before posting.

  4. aek  May 10, 2012

    This will take awhile to digest. Incredible work – thank you!

    Have you seen the press release about Frank Sacks’ HDL study looking at apoC-III?

    How does that figure into this analysis?

    • Peter Attia  May 10, 2012

      These analyses are interesting, but difficult to draw firm conclusions from. They are often observational, not experimental. However, it does suggest how primitive our understanding remains. We know too many LDL-P are bad, but we still don’t really understand how it all fits together. Another caveat: no drug that has been designed to raise HDL-C (to date) has shown any efficacy in reducing heart disease. Interesting, huh?

    • Magarietha  May 12, 2012

      Hi, thank you for a most insightful site. I don’t think Americans are stuck with the problem some of us here in South Africa have (caucasian Afrikaners and some Jews too) – we have familial hypercholesterolaemia. So now where does this throw us into the equation? I do know for certain that if children are born with a double gene from both mom and dad they are in very serious trouble. Our levels are always double digit. In metric it’s about 11 to 13 without medication. BUT then my mom who only found out about this in her 60’s has recently turned 80!!! And from her 60’s she was on and off meds. She didn’t like the idea. Yet still? It does frighten me however. My 2 sons also have this. The meds also never bring it down to acceptable levels. I have now gone on low carb diet just to see if it makes a difference. What is your opinion please. No-one thus far has answered me on this. Would so appreciate. Cheers

    • Ed  May 13, 2012

      “no drug that has been designed to raise HDL-C (to date) has shown any efficacy in reducing heart disease. Interesting, huh?”

      This is actually interesting… might this imply that other interventions (lifestyle, high levels of saturated fat, especially high intake of lauric acid… coconut oil) wouldn’t work either?

      Which is interesting… if you are managing the numbers that are the inputs for “risk”… are you really changing your risk or changing your numbers?

      To the extent you are not modulating causal factors, but biomarkers, then you probably arent modulating risk. Dont put out fires by hiding the smoke.

    • Peter Attia  May 13, 2012

      Not necessarily. It’s more likely to suggest that the rise in concentration of HDL-C is symptom of the positive metabolic change. Give a drug that only raises HDL-C –> no change in underlying metabolic problem. Fix HDL-C with “lifestyle” –> you fix a lot of other stuff with it.

  5. KevinF  May 10, 2012

    Now it’s getting interesting! I had a chat with my sister (a physician). Surprisingly she understood right off the truth of how a low carb diet would be expected to affect the lipid profile, and she practically wanted to give me a VAP panel on the spot — but she’d never heard of the NMR.

    • Peter Attia  May 10, 2012

      Never too late to learn! I didn’t know about NMR until a year ago.

  6. Jim L.  May 10, 2012

    What is the NMR price?

    Thanks. 🙂

    • Peter Attia  May 10, 2012

      Cost of NMR is probably under $100.

    • Joe Visma  March 12, 2015

      NMR lipoprofile from LabCorp November 2014 = $100
      After LabCorp acquires LipoScience
      NMR lipoprofile from LabCorp February 2015 = $150
      50% increase, must have been an expensive acquisition.

  7. John M  May 10, 2012

    Thanks again for the great posts. Can you give us any info on the relative costs for these tests? I’m going to guess that my health insurance will not cover the NMR test.

    • Peter Attia  May 10, 2012

      I’m not sure of the exact relative differences. Obviously a simple lipid profile is the cheapest. All of these tests probably cost under $100. The most deluxe of these panels, with advanced genotyping, can be done for $500. If an NMR is warranted, insurance almost always picks up the cost.

    • Ed  May 11, 2012

      $127 at (search NMR).

      BTW- last year, Blue Shield in California denied coverage of my VAP panel as well as my hsCRP tests, because “these treatments weren’t supported by peer reviewed research”. I received a form letter to this effect and got so mad I dug up some research (NEMJ) and tried to get on the phone with them. But you know, I went through 60min of automated response systems before it transferred me to a number that was busy. I just gave up.

  8. lorraine  May 10, 2012

    Thank you for the explanation of VAT. I was unaware that it was an estimate.

    I couldn’t help but thinking as I read this post of the enormous impact of whatever the current state of technology has been on public health. We had the ability to measure total cholesterol and then we had Ancel Keyes and the Food Pyramid. We had the ability to fraction cholesterol and then we had “good cholesterol”/”bad cholesterol”. The whiz-bang of the day drives health policy and food policy, people try and do the right thing, and diabetes and heart disease marches on. Uncharacteristic of me as a lover of whiz-bang, this post has me feeling a tone of cautiousness that we need to be really careful this time, and we need to be very sober about understanding as much of the mechanisms as possible because a lot is riding on it.

    I certainly need to do much more reading. My biggest question at this point is why are increasing particle number considered to be an early predictor of IR? Additionally, I get the metaphor of the increasing traffic in the artery [it’s the cars not the passengers] that’s more likely to cause an injury to the epithelium, but to me that sounds like “plaque clogging cholesterol”. What’s the mechanism that causes increased particle number and the mechanism that increases the risk of injury as a result of increased particle number? Asking these questions not so much for you to answer them, Peter, although if they come up in following posts that’ll be great, but rather just to try and sort out in my own head the things that I need to further hunt down. And yeah, that feeling of caution.

    • lorraine  May 10, 2012

      edit: Above, I meant “artery clogging plaque”.

  9. steve  May 10, 2012

    Very informative: I hope you will cover why some of us generate more LDL particles for a given level of LDL than you would think. Is it just genetics; and if so what is the genetic basis for particle production? Are drugs the only way to deal with it? In my case, diet manipulation after learning I generated tons of particles was unsuccessful- still no matter the diet generated tons of particles with size differing- larger with less carbs. Particle count per last NMR was 569, but achievable only via Crestor 10 and Zetia 10.
    Your wife has a great sense of humor! I love the T-shirt! May have to get some made- w/o the PA at the end of course!. Thanks for the hard,very important work; many will be benefited from it.

    • Peter Attia  May 10, 2012

      Genetics and diet play a large role in the discordance of LDL-P and LDL-C, which I’ll discuss subsequently.

  10. Jim  May 10, 2012

    So, if I go to my primary care physician for a check-up and he orders blood testing, can I request the NMR testing? Also, if he can do this, will insurance cover this?

    • Peter Attia  May 10, 2012

      Yes, and probably. Of course, your doctor needs to explain to them why this test is necessary. I’ve very rarely seen any issue with this.

    • Debbie  May 11, 2012

      I’m looking forward to reading your post on genetics and diet and LDL-P and LDL-C.

  11. Glenda Glass  May 10, 2012

    Thank you thank you for doing this! What you have written is going to be extremely helpful to me and other health educators. There is SO much misinformation out there in the general public (and in the minds of most medical practitioners for that matter) about cholesterol, heart disease, and diet. Your articles are going a long way to refute the “high cholesterol, take a Statin, continue to eat crap and get sicker” mind set that I encounter every day.

  12. Alex Carvalho  May 10, 2012

    I think I’m getting to know enough about cholesterol to become an annoyance to my physician (and perhaps a hazard to myself…). Thanks for that, Dr. Attia. I kept wondering what you consider to be a “too terribly insulin resistant person”. Could I be one of those (I’ve been in ketosis for 3 months, strictly doing the induction phase of Atkins, and yet my HOMA-IR stubbornly remains at 2.4)? What should such a person do differently when monitoring lipids?

  13. Alex Carvalho  May 10, 2012

    If someone starts a low carb diet, a significant fall in TG should be expected, right? If the LDL-C ~ TC – HDL-C – TG/5 formula is then applied to estimate the dieter’s LDL-C before and during the diet, is it right to conclude then that the result during the diet will necessarily be higher than before the diet, only because TG has fallen? Looking at my own standard cholesterol tests before and during my diet, it seems this was exactly the case, which renders the slight increase I have verified in my LDL-C completely meaningless, doesn’t it???…

    • Peter Attia  May 10, 2012

      Not always. HDL-C often rises, too. The point, however, is that is probably doesn’t matter what’s happening to LDL-C. All that matters is what’s happening to LDL-P. Even the TG/HDL-C ratio becomes unimportant once you know LDL-P or possibly (real) apoB.

  14. Mike N.  May 10, 2012

    It is wonderful to read real information on this subject rather than the usual hysterical arm-waving. But knowledge always begets more questions. I hope you address these other questions: 1) Why does the body raise the LDL-C concentration? 2) When that happens, what is it telling us, assuming that it is doing so in response to some pathological or other condition that needs LDL for correction? It would seem that the body is sending a message, and we need to know how to interpret the message. Thank you for making the effort to write this clear series. I am very grateful.

  15. Peter  May 10, 2012

    My ideal TG/HDL ratio from standard LP could be misleading as I could still have excessive LDL-P particles, even if they are mostly pattern A. Previously I was skeptical about NMR. No longer.

    A direct count from NMR makes sense. Inferences, like the formulaic calculation of LDL-C in standard LP, provide a range at best. Accuracy trumps inferences.

    Thanks much for the education.

  16. Jill  May 10, 2012

    Thank you Peter. Almost everyone I know is trying to lower cholesterol with drugs and/or diet but the cardiologists are still very busy. Our friend had a very low cholesterol. His doctor said he would never have a heart attack. He had a heart attack. He is now on Lipitor. His very low cholesterol number didn’t prevent his heart problem. I think Vitamin C deficiency has something to do with this. Thank you again Peter.

    • Peter Attia  May 10, 2012

      Not uncommon to see LDL-C in the “low” range with sky-high LDL-P. Will discuss later.

  17. Eugene  May 10, 2012

    Interesting article, I was wondering if you had any disclosure statements or conflicts of interest with these companies?

    • Peter Attia  May 10, 2012

      Great question, Eugene. Thank you for asking. I have no financial interest in any of these companies, nor do I receive any compensation in any form from any of them. As a provider, I now only use Health Diagnostics Laboratory, Inc. for any client (or friend or family member) I get bloodwork on. This choice is based on my belief in their comprehensive assays and the quality of their service.

  18. Sue  May 11, 2012


    Thanks for all this important (but heavy) information.

    My MD currently uses Berkerley HeartLab(BHL) for advanced lipid panels. How does the BHL “Segmented Gradient Gel Electrophoresis” (S-GGE) process fit with your recommendations? BHL says that it “is the only commercial standard method that correlates with” analytic ultracentrifugation (AnUc) which BHL cites as the “gold standard” for lipid fractionation (page 3 of BHL’s Clinical Implications Reference Manual ). I do note that the cites in support of this statement are quite old!

    • Peter Attia  May 11, 2012

      I touch on this part II. There is only FDA-approved way to counter particles — NMR by Liposcience.

  19. Ron Wichin  May 11, 2012

    Thanks for the comprehensive explanation! I am not sure if this is the best place to ask this but here goes. Based on what I read here I have been using super starch and I love it. My question is should I count the 31 gr of carb toward the 50 gr max to get into ketosis or not. I only use the super starch when running, biking or swimming an hour or more. Thanks and I always look forward to your next post.

    • Peter Attia  May 11, 2012

      It’s not really clear. Probably not the whole amount, and it depends how much you’re exercising when you ingest it.

    • lorraine  May 15, 2012

      Speaking of this (and sorry, I also realize this is off topic of the lipid discussion), do you count fiber into your total CHO? (I’m assuming you’re not using any other “net carb” products).

    • Peter Attia  May 15, 2012

      I count ’em all. Certainly fiber has a lesser effect, but if I only need to keep track of one thing, this is easiest. The “high bar” so to speak.

  20. Barbara Hvilivitzky  May 11, 2012

    I may be way off base for this comment but really, where is all this testing getting us? Is this what passes for preventive medicine these days? I thought preventive medicine was what we did BEFORE we had to go to the doctor. You know, eating right, getting lots of exercise, lots of fresh air and sunshine, reducing stress, staying slim, getting plenty of sleep, not working too hard – that kind of thing. Now we’re off to the doctor getting “check-ups” and “tests.” and Peter, you seem to be fostering all kinds of new more sophisticated tests, and telling us all about the ins and outs of even more intricate ways of finding out how complicated cholesterol really is.

    I have not been to the doctor for a check up for 15 years and I don’t see the need to go. I’m not sick. I’m not fat. I don’t get flu shots. I don’t care if I have high cholesterol. I try to eat right. Now don’t flame me for this, but I see the sky high cost of the medical systems both in the US and in Canada as a result of people going to the doctor and having tests when they are not sick. If high cholesterol is not a worry for 90% of people why would anyone go for a check up and have a blood test for cholesterol?

    • KevinF  May 11, 2012

      Barbara, 4 weeks from now I’ll be exactly the same age as my father was when he dropped dead of a heart attack. I for one am VERY interested both in knowing all I can about such things as cholesterol (one component of which seems to be largely responsible for causing heart attacks); about recognizing and quantifying as well as possible my own risk of dropping dead; and of learning what if anything I can do to somewhat lessen the risk that I’ll drop dead next month.

      Obviously, Peter here is providing some of that education; and he’s letting us know about new technology we can use to help identify and quantify those risks (and God knows we’ve been living in the dark ages using leeches up til this point); and he’s discussing dietary practices that may mitigate these problems. How is that not very relevant? You’r free not to care, but that being the case how do you wind up on this blog in the first place?

      As for as the cost — yes I’m sure that we do a lot of needless screening and tests. But the solution is to develop and identify the RIGHT test to do, and not waste money administering a bunch of useless myth-based testing that leads to the obligatory useless prescription.

    • Peter Attia  May 11, 2012

      The data would suggest that preventative testing is not at the heart of what we spend on healthcare. Yes, the US spends more on healthcare than any other country by just about any metric you want to consider: total amount ($2.5 Trillion), %GDP (18%), per capita (about $8,000 per person per year). But here’s the thing…that money is not spent uniformly across one’s life. Most of the healthcare dollars a person spends are spent in the last 90 days of their life. It’s not the tests and check-ups that are killing our economy — it’s the catastrophic diseases we try to fight off when it’s probably too late.

      We’re being decimated by 2 factors: 1) our behavior is driving an unhealthy lifestyle because poor infrastructure and poor information feed it; 2) and we only play “defense” — that is, we only spend money when it’s too late.

      Is a $500 lab test worth it to let me know if I’m a ticking time bomb? It is to me. But you’ll have to make your own choice. Just make sure you’re starting with the right information, though.

    • Dan  May 11, 2012

      Question to KevinF:

      I found today a bottle of high oleic acid cold pressed safflower oil (organic), from Europe, but when I went to look at the breakdown of PUFA in it, it was all omega-6 (about 2.6g per tablespoon). Does that make any sense to you? Naturally, I didn’t buy it, as I do not wish to mess around with my OM6:OM3 ratio. You mentioned that high oleic acid cold pressed safflower oil should have very little if any PUFA of any kind in it, so the ratio hardly matters. But that does sound to me like alot of omega-6 – now perhaps I misread it (in Europe they use commas instead of decimal points), but I don’t think so! I believe it was 2,6g per tablespoon (15 mL). Does that seem like a huge amount to you?

      Thanks in advance.


    • Dan  May 11, 2012

      I see now I screwed up, Kevin. You said the ratio of MUFA:PUFA is 11:2, and that is exactly what’s in a Spectrum brand. The brand I saw was not spectrum and I do not recall the MUFA:PUFA ratio. I do think it had either 2.4 or 2.6 g of OM6 per tablespoon, which seems really high to me (if OM6 is as biologically active as OM3 – no idea).

    • KevinF  May 12, 2012

      That sounds about right Dan. I think I got the 11:2 ratio by noting that a tablespoon has 11 grams MUFA and 2 grams PUFA. And I suppose just about all that 2 grams PUFA is Omega 6. So yeah you probably get that much O6 in a tablespoon, and maybe they rounded down on the label. Nevertheless, hard to get much less than that if you to cook higher heat and avoid saturated fat (which I think was your original issue).

      I’m thinking that’s not bad in terms of O6:O3 ratios, assuming 1) you are supplementing with a good Omega 3 or eating salmon, and 2) you are not using huge amounts of the oil. Not sure what your plans there are, but I’d be apt to use maybe a tablespoon in the wok doing a stir fry every now and then — can’t see me using a whole lot of it.

      Aside from that kind of higher-heat application, you can stick with olive oil, which would have less Omega 6 (but still isn’t O6 free).

    • Nina  May 12, 2012


      Information is power; why shy away from it?

    • Dan  May 12, 2012

      Kevin, thanks so much for checking in and for writing me back and hope you will tolerate (I know you will) – two more questions

      1) How is olive oil better? Is it just the lower OM6 content, or high resilience to heat? Or its use in the Mediterranean diet trial showing improved vascular outcomes (Lyon Heart Diet Study)?

      2) 2+ grams per tbsp of OM6 still seems really high to me. If one wanted to maintain a 1:1 relationship of OM3:OM6, that would mean having to eat oily fish every day or popping a supplement (something I am really loath to do, for several reasons). Many of my stir-fry recipes call for well more than a tbsp (sometimes up to 3 tbsp), and a single serving of salmon or tuna would not provide sufficient OM3 coverage. Then again, I’m still very uncertain whether OM3:OM6 is simply an epiphenomenonal marker of “healthier” patterns of eating in general (for example, the very nice 1:1 ratios in Greenland and Japan have been correlated with better vascular health, but those are simply correlations not necessarily cause-and-effect relationships – we really need a randomized trial to dissect that).

      Thanks again. It’s great to have such astute people to consult with.

    • KevinF  May 12, 2012

      Dan – Well keep in mind I don’t have a PhD in oil or anything, so I can just pass on what I’ve read. (A good book for example is called “The Queen of Fats”). I realize that a 1:1 ratio of O6:O3 is often said to be ideal, however I’m not sure how you get there in the real modern world. It’s hard to GET Omega 3, while it’s hard to AVOID Omega 6. There is some advice out there that a more achievable ratio of, say 4:1 in favor of Omega 6 is OK or at least an improvement — I think that’s now official advice in some countries. And really the typical America diet is probably somewhere between 15:1 and infinity to 1. (And now Peter is saying, just make sure you get your O3).

      And you can drive yourself crazy looking up fat-sources in something like
      … everything seems to have lots of Omega 6.

      So my approach is to somewhat minimize the O6, while taking about 2.0 – 3.0 grams of liquid O3 supplement from Carlson Labs — citrus flavored, very nice, not fishy. I eat salmon when the opportunity presents itself but probably only once a week at most.

      Olive oil is usually seen as good for two reasons: one, it’s nearly all MUFA and not PUFA or saturated fat. Some people fear saturated fat and some people fear PUFA, but just about everyone agrees mono is benign (at worst) or positively good for you. Even if it’s just benign, then this is a way to get some fat in your diet without megadosing on O6.

      And then there is the Mediterranean diet theory — I won’t attempt to judge if that’s valid or not, but IF it’s valid there might be some mysterious X-factor that makes olive oil healthy, even beyond the whole MUFA/PUFA/SFA thing.

      A drawback to olive is: yes it does have SOME omega 6. It does not have a good O6:O3 ratio. But the key is it doesn’t have much of either. So as far as oils go, it’s not a good source of Omega 3, but also it doesn’t deliver very much Omega 6, despite what the ratio looks like. But if you actually used a LOT of olive oil, yes the O6 will add up. Second drawback is, it has a lower smoke point than your PUFA type oils. It’s not a high-heat oil. So, not good for high-heat applications like true stir-fry.

      So … to your case, you say you want to avoid saturated fat, I believe. But you also want to minimize O6. It’s hard to do both unless you just avoid added fats altogether. (And I don’t see how it’s possible on a true low-carb high-fat diet). It would be nice if there was a high-heat, O6-free, pure mono oil … but I’m not aware of any such thing. And you certainly are not supposed to be cooking with fish oil (it burns and becomes corrupted).

      Finally one thing I’d recommend (at the risk of being obvious) is to use good modern non-stick pans. That way you don’t need much if any oil, unless you want want to add that fat on purpose, like I do with ghee or coconut. I’m sure I don’t use anywhere near more than a tablespoon of oil when I DO stir fry, in a pretty large wok. If you haven’t tried them, new white-surfaced ceramic pans are simply unbelievable. You could fry road tar in those things and wipe it off with a paper towel. Also, you can simply accept working at medium heat with olive instead of stir-frying at high heat.

    • JohnJ  May 12, 2012

      Some of the bigger problems in modern medicine is the over diagnosis and overtreatment. You have it with PSA, mammography, cholesterol etc… Patients are being harmed more by the treatments than from the ailment they went to treat. Modern medicine is a business having patients to treat with pharmaceuticals and interventions is big business. You very be well informed before falling victims, there may be times were there could be more natural lifestyle modification and treatments available that you can try first.

    • DHackam  May 12, 2012

      Kevin, this is great and very helpful. Indeed I am trying to limit consumption of substances which might boost my LDL-P count. You are right that it is very difficult to both limit carbs and limit SFA and dietary cholesterol. It is like walking two tightropes rather than one.

      Now on top of those two tightropes, imagine walking the OM6:OM3 ratio tightrope.

      I am metabolically very sensitive to both carbohydrates and to fats, therefore I am attempting to maximize lean protein intake, coupled with MUFA and OM3 PUFA. Tricky to do, but my main sources of protein are now nuts, tofu, fish (both lean and oily), (skinless) chicken. Pretty much the Mediterranean diet (which unlike most diets has randomized trial evidence showing a reduction in hard outcomes). Fats for me will include olive oil and avocados (as well as, obviously, what’s in nuts and oily fish). I’ve cut out all red meat, butter, coconut oil, hard cheese, cream cheese, sour cream – the main chol/SFA sources in “low carb, high fat”.

      I’m guessing for barbequeing (a truly high heat application – in many cases), you are applying ghee or coconut oil. I was actually applying olive oil, which I now see from your post is a very bad thing to do (as Peter once put it, bad to add heat to a double bond). This puts me in a bit of a quandry as I love to barbeque salmon and red snapper. The recipes I have seen even call for coating the fish with olive oil (before applying seasonings and prior to grilling on the ‘Q’). Is there anything other than SFA (coconut oil, ghee) that you would recommend for frying at high heat? I guess we come back now to high heat oleic acid containing safflower oil (cold-expressed)! Time for me to pick up some of that stuff!

    • Peter Attia  May 12, 2012

      Dan, Kevin, I’m planning to rock your paradigm a bit in the coming months with a series on n-6 and n-3. Hope you can wait until then.

    • Dan  May 12, 2012

      Peter, I really look forward to it.

      I just picked up a can of Spectrum organic high heat canola oil (expressed). Ingredients: Expeller Pressed Refined High Monounsaturate Canola Oil, Soy Lecithin, Propellant (no chlorofluorocarbons). This is the only thing I could find that said it was stable at the temps I barbeque at (400-425 F). The other choice was a cold-expressed organic safflower which did not promise heat stability and had 4.9 g of OM6 per teaspoon (10 mL). So given the devil of heat-produced free radicals and/or trans fats vs having to use canola with an unknown OM6 content but heat-stable, I chose the latter devil.

      I love this site, as you can tell.

    • KevinF  May 12, 2012

      @Peter, I was just pointing here from Facebook, saying you tended to write in cliffhanger fashion. Are you sure you never worked in media?

      @Dan, after all that, just an hour ago I think I found a winner for you. Was in a specialty grocer called Sprouts, and found a product called Napa Valley Naturals High-Oleic SUNflower oil. It has about the highest smoke point yet (460 degrees), and has only 1 gram PUFA per tablespoon, for a MUFA:PUFA ratio of 12:1. That there might be the best you can hope for.

      It’s this product here, but not sure where you buy it if you don’t have Sprouts.

      My high heat cooking, such as I do it, is just wok and pan-frying. Sadly I haven’t lived in a grill-friendly place since I was out in Peter’s neck of the woods, living in Coronado.

    • Dan  May 13, 2012

      Kevin, thanks for the tip. I am going to see if I can order this online.

      I hit PubMed last night to see if pure saturated fat – in the form of coconut oil – and in the absence of dietary cholesterol – has any effect on serum lipid levels. There were about 7 or 8 direct feeding studies which suggested that coconut oil (which is high in myristic but also several other types of FAs) does increase apoB, total:HDL ratio, LDL, triglycerides and VLDL. Unfortunately none of them studied these in carbohydrate-restricted people; rather these were randomized, crossover trials in which all fat was from coconut oil (that would be a rare situation). Not surprisingly, Walter Willett has a piece in the Harvard Health Letter declaring caution regarding the burgeoning use of coconut oil. Lastly, none of these studies looked at LDL-P – I think that marker is just too new for major feeding trials to have pronounced any effect (though the changes in apoB are pretty telling).

      Anyway, thanks for the tip re the sunflower oil!

    • lorraine  May 15, 2012

      DHackman, if you don’t mind me asking, how have you determined that you are sensitive to dietary fat? Thanks.

  21. David Nelsen  May 11, 2012

    Peter, one question I hope you address in the next section is heart health due to cholesterol with Syndrome X types (me) versus people like your wife who can eat Oreo’s all day. Does eating carbs if you are insulin sensitive have the same effect on ones arteries if you don’t have a lot of excess insulin floating around? I know that a lot of non obese people have heart attacks every day, so remaining a little less non fat (as your wife likes to say) would not seem to have inherent advantages with respect to heart attacks and strokes. Now Syndrome X types might have a higher rate, but they are not immune.

  22. Bong  May 11, 2012

    Would you change the font color to black? It is gray and hard to read your article. Thanks.

    • Peter Attia  May 11, 2012

      What browser are you using? Are others having this trouble?

    • Bong  May 12, 2012

      Chrome. Please just change the font color to black. Gray color fonts are not beautiful. Black fonts over white background is easy to read and clean. Thanks.

    • Peter Attia  May 12, 2012

      Yes sir. Sorry for the inconvenience.

    • lorraine  May 15, 2012

      I’m using Chrome. Looks black to me

  23. Adam  May 11, 2012

    font color looks black to me…

    Barbara, occasional doctor check-ups and (when appropriate) running tests IS preventative medicine! the other things you mention (sleep, eat well, fresh air, sunshine) are called lifestyle factors. google “mark’s daily apple” if you want more of those.

    you haven’t been to a doctor in 15 years and you’re fine? that says nothing. that’s like saying “i haven’t taken my car to a mechanic in 10 years, and it runs fine!”. well, cars need oil changes and tire-checks and other preventative measures so that it doesn’t breakdown and leave you stranded on an interstate sometime.

    are you really saying you’ve never even been to a gynecologist for an exam? you don’t believe in mammograms? colonoscopies? if you want to go rogue and roll the dice with your health, that’s your choice, but don’t be surprised if by the time you find a lump, the doctor says “it’s too late to do anything, if only you had come in sooner.”

    as Peter says, these tests are a fraction of our nation’s healthcare costs and Peter is not recommending that everyone in America needs one on an annual basis. this is an in-depth discussion of the role of cholesterol in our bodies, and these new innovative testing methods help shed light on the subject for patients, doctors, and researchers alike.

    • Barbara Hvilivitzky  May 11, 2012

      Well, I guess I AM in the wrong place! First of all when Peter first started this blog it was all about nutrition and he told us he was testing himself very carefully – but he was doing it because he was a very serious athlete, a medical doctor, and a bit of a medical nerd to boot. He told us what he was eating too which was really interesting. I tried to imitate him and have lost ten pounds – and I feel great. Then things got more complicated and testing, testing, testing, and extremely detailed information became the order of the day.

      Can I harken back to the day when heart attacks were rare, cancer was also rare, obesity was rare, people worked very hard, seldom went to the doctor etc. Food was not toxic at source, meat was fatty and people ate plenty of it, and on and on. What about trying to live like that? So when I say I have not been to the doctor why does that sound so weird? My grandmother didn’t go either and she lived to be 93. She just wore out. Of course I know people get sick but we are all going to die of something, no?

      Oh, well, I guess I’ll just shut up and go away. Thanks Peter for all your good work. I really hope you and Mr Taubes have wonderful success with your efforts this summer. God bless you both.

    • Peter Attia  May 11, 2012

      Hmmmm, I’m not sure I understand why a detailed understanding of how our body works, and how to test it, is mutually exclusive from what you’re interested in (and what most of us are interested in). Sorry you feel this way.

    • Bob Johnston  May 12, 2012

      I can definitely see both points of view here and I don’t believe they need to be mutually exclusive.

      I think it’s awesome that we’re developing new tests and markers for determining whether you’re healthy (Peter’s point). This is long overdue and if we can break through the conventional wisdom should lead to great things for humanity. And since the technology and theory is new of course we need to do a lot of testing to verify its efficacy. Very reasonable.

      If I can be so bold as to speak for Barbara I think what she’s saying is that because she’s already doing everything from a lifestyle point of view necessary for good health and vitality then there’s no point in going in for tests because if something was out of kilter what could she possibly do otherwise? I think that’s a valid point too. If you’re doing everything correctly then what’s the point of getting tested for cholesterol levels? I tend to follow this point of view myself. Since I feel great and am at a healthy weight and probably know more than my doctor on the topic why would I go in and subject myself to something that would probably be a complete waste of time?

      And one thing that’s off topic but was mentioned by the OP, I don’t think I’d ever get a colonoscopy performed on me. I ran across this guy’s experience a while back and he makes some very good points about the pro’s and con’s of this test. Talk about an eye-opener.

  24. Peter  May 12, 2012

    Please stay put and don’t go away frustrated. You likely have good genes and will live to 90s or more. You shed ten pounds as a result of following Peter’s advices. And you may be one of those w/ ideal TG/HDL ratio which should correlate w/ low LDL-P count. But there a is 20% chance where you may have high LDL-P particles count.

    Should you choose to, you can derive additional benefits by taking advantage of the latest testing techniques, instead of the standard LP. If not his explanation, many of us would not know about LDL-P and its implications. We are able to gather medical information to diagnose and then seek treatment from our doctor, who probably thought we were not at risk at all. We are empowered.

    Yes, the acronyms and concepts can be overwhelming. I admit my mind has a hard time grasping them. But where else can we acquire this knowledge, certainly not from my medical providers. Reading is tough, but digesting and then writing it for this audience ( probably non-biology or chemical majors) is even tougher. Thanks much, Peter!

  25. Marilyn  May 12, 2012

    Yes, Barbara. Stick around. There are others — well, at least one other — around here who think like you. As I see it, probably the greatest risk in not going to a doctor for 15 years is that if a person then needs a doctor for an accident or sudden illness, s/he will be doctor shopping under less than ideal conditions.

    That said, I’m finding this series fascinating, and I’m eager to see where Peter Attia goes with it. That doesn’t mean I’ll feel compelled to get a test. If as Peter above suggests, I have a 20% chance that I’ll have high LDL-P, that means I have an 80% chance of not having a high reading. If I do get a test and find myself in the 20% with high LDL-P, I would want to know what that means in terms of all cause mortality. If it seems that most of those with high LDL-P are dying much younger than those with low LDL-P, I would want to know what the treatment options are. Before embarking on any treatment, I would want to know the risks and benefits of treatment. . . .

    Adam, I don’t see the similarity between oil changes and routine maintenance for a car, and gynecology exams, mammograms, colonoscopies, and the like. These tests are neither completely foolproof nor completely harmless. Persons without symptoms might responsibly do their research, weigh the odds and decide the best choice is to skip the tests.

    • Barbara  May 12, 2012

      I’m a different Barbara and I’d like to chime in on the topic of testing. I agree that testing, just for the sake of testing, isn’t always a great idea. That doesn’t mean testing is not warranted for people with specific risk factors such as routine HIV testing for those engaging high-risk sexual behaviors or lipid testing for those with a strong history of heart disease. I highly recommend the book “Overdiagnosed” by Gilbert Welch. He thoroughly discusses the potential harms associated with several tests used for early diagnosis of a variety of diseases. After reading the book, I decided that I will respectfully decline any routine testing that my doctor offers.

    • lupo  May 12, 2012

      There is an old medical proverb: “There is no healthy person, only an insufficiently examined one”. You will always find something uncommon in the labs, just because statistics says that the more variables you measure, the higher the chance that at least one will be out of the normal range. I see the problem especially in the US healthcare system, where diagnostics are often not driven by medical usefulness (and an understanding of NNT, NNH, NND, sensitivity, specifity, pretest probability etc.), but by being afraid of lawyers. If you suspect any health problem, seeing a doctor is a good idea. Doing a small-scale checkup is usually quite safe, too. Going through a full-body MRT each year is not.

    • Dan  May 13, 2012

      We have all seen cases where routine screening saved lives of asymptomatic patients. For that reason, major consensus guidelines recommend such things as pap smears, lipid panels, blood pressure measurement, oral glucose tolerance tests, abdominal aortic aneursym screening and so forth. However, there are other instances where routine screening of asymptomatic patients led to anxiety, cost, and further investigations and unnecessary treatments because of false positive results. On balance I think selected evidence-based screening combined with aggressive lifestyle modification is the best “middle road” to take. As an example – carotid artery plaque measurement in someone with a family history of premature vascular disease. I’ve also read Welch’s book and I disagree somewhat with his core message but do agree that people should not be screened with every test imaginable and in every situation. This is why 20-year-old males do not get PSA testing but will get testicular examination by their (right-minded) doctors. Prostate cancer is fairly unheard of in a 20-year old but testicular cancer is actually the most common malignancy in that age group (alongside lymphoma and thyroid).

    • lorraine  May 15, 2012

      There are folks who, no matter how aggressive their lifestyle intervention, will still carry significant risk for heart disease. Factors such as age, genes and other metabolic factors (menopause, thyroid disorders etc.), job stress, and poor sleep may mix a brew of risk that continues to drive insulin resistance despite good dietary habits. Weight cannot always be the identifier for risk because there are many thin people who have significant insulin resistance and carry significant risk. Good tests (and I acknowledge and commented earlier that tests have not always been good) identify these folks. They may need more help than lifestyle, like meds, to keep them safe.

  26. Ed  May 13, 2012

    The role of LDL-receptor activity and how we might modulate it… to lower our LDL-P, our LDL-C (indirectly) and our lipo-induced stress levels… any thoughts or suggestions?

    If you’ve come this deep into the comments your probably wondering how to lower your particles, just like I am. So, in terms of particle numbers circulating in blood at any given time (which a test would be measuring, probably around 8am before going to work), you have to ask yourself how did they get there and how do they get out? From what I know so far, there is only 1 place where they come from… the liver. This is your apo-B factory, and your on ramp for apo-B particles into the blood. If I tried to take the strategy of targeting LDL-P lowering by focusing on the production end, this seems to be complicated by the fact that the liver doesn’t actually make LDL-P. It makes apo-B containing VLDL particles, some of which eventually turn into LDL-P particles.

    So let’s turn our attention to the consider the disposal path. If the liver is the onramp for your apo-B particles… what’s the off ramp? One clue is that people with familial hypercholesteremia (FH) have high cholesterol levels because they have mutations in the LDL receptor gene that diminishes its activity. I googled “half life of ldl” and what do you know… up comes a study of 12 normal and 12 FH subjects. The abstract says that LDL half life in the 12 normal patients was 4.7 days and the 12 FH patients was 3.1 days. The LDL-C of FH subjects was signficantly higher; 252 vs 105 and their apo-concentrations (I interpret this as a surrogate for particle count) 153 vs 63. So, if your LDL-receptors don’t work as well, your particles stay around longer in the blood and you have higher LDL-P and and LDL-C.

    Are LDL receptor activity levels stuck by your genes? I did a little pubmed on LDL-receptor, and came across a drug called amiodarone, (ironically a drug used to deal with irregular heart beats), which happens to increase plasma LDL levels. According to the abstract, this increase is due to the “decreased expression of the LDL receptor gene”. Hmmm… so this means that LDL receptor activity can be modulated… at least if you want more particles for some reason.

    Now a word on statins. I was really surprised to hear from Dayspring himself in his “Cholesterol Synthesis and Absorption Markers” on lecturepad that statins actually work to lower LDL-P by upregulating LDL-R! I always thought it was because it blocked cholesterol production… which it does. My understanding now is that the liver sends out cholesterol depleted particles, compared to when you weren’t on statins. But it also upregulates the LDL receptor activity in the liver. I visualize it as such… here is the liver, short on cholesterol because it can’t make as much as it wants to. What does it do? It ends up sucking up more from the gut (increased absorption rates) and more from the blood (increased LDL-receptor activity). If you happen to be good at really good at recycling cholesterol through your intestines, your blood cholesterol (and presumably your LDL-P) doesn’t go down as much as it doesn’t need to upregulate the LDL-R as much to keep the liver happy. To me, that sounds like a very indirect method to lower your LDL-P.

    Preferably, I would like to modulate my health through natural methods… food, exercise. It’s not that I want hyperactivate my LDL-R to abnormal levels. After all, the body is a fine tuned machine and it might not be a good idea to force feed your cells stuff they don’t want. Perhaps your LDL-R activity is low because cells have plenty of TG, CE, and phospholipids and they don’t want any more. But maybe your cells would like more and the LDL-R is hampered in some way… by a lack of nutrients, or a hormonal balance, or something like that. Maybe your cells do want to more cargo from the containers, but your LDL-R isnt working as well as it should, so your liver senses this (obviously the gut and liver communicate via some messenging system) and ships out more apo-B particles to help out.

    If anybody wants to shed light or explore this issue with me, please comment.

  27. David Nelsen  May 13, 2012

    This thread seems to have gotten hijacked a bit. I think everyone has a different opinion about their health and what they need to do to maintain it. My family history includes cancer and heart disease. Should I assume that these diseases will skip past me and ignore the risks? I fall on the testing where it makes sense (to me anyway) side of the ledger. I have life insurance and home insurance, knowing that the odds of my house burning down or dying young are not high. I have had my blood tested for LDL-P, not because I’m convinced that I have a problem but because if I’m in the 20% with problems I want to know about it and do something about it. Men my age die all the time from heart attacks and strokes. My youngest daughter is only 8, so I need to live a good while to see her grow up and get married and have a family – and those are all things I very much want to be able to do. My wife’s dad died from a stroke when she was only 3 years old. She grew up never knowing him. In those days the kind of testing and care we have now was not available. If it had been, my wife could have grown up knowing her father. As the author Calvin Trillin says – you can see a Gentile oncologist or go to a white run BBQ restaurant – but you’re really not playing the percentages.

    One of the biggest problems that I see is that the food pyramid it upside down. For many of us to be healthy, we have to ignore the conventional wisdom and read and make decisions for ourselves. For every Gary Taubes you have a Dean Ornish. It takes critical thinking a some leaps of faith to decide who you want to believe. Peter presents his data and information here and you can read it and decide what applies to you and how and if you are going to change what you eat and how you live as result – or not. Peter isn’t selling anything or twisting anyone’s arm. How you use his information is up to you.

    • Peter Attia  May 13, 2012

      Thanks for your comments, Dave. I’ve been sitting back with a tincture of bemusement, wondering what folks are going to do when I *actually* get to the controversy! All we have talked about so far is what cholesterol is, how it moves around the body, and how to measure it…

      There is nothing of controversy in this series. Sure, folks can debate the cost utility of using NMR on everyone versus select cases, and I’m happy to write about that, though I haven’t heard that concern, save one person.

      I think what’s most interesting is that compared to all the other stuff I write about — where there is real disagreement with “conventional wisdom” — this is the topic getting the most resistance. Very interesting!

      This is ok. Discussion is good. The part I want to caution you on is this: don’t forget to be a thinker. I know it’s tempting to say, “My uncle’s cousin lived to 104 and all he did was eat bacon,” or “My neighbor’s mother never ate meat and lived to 97,” or “I know someone who had an adverse reaction to drug X,” and therefore…fill-in-the-blank (i.e., bacon is good, meat is bad, drug X is bad).

      Guys, this is not science. Everyone seems to understand this point when we talk about nutrition, but all of a sudden I sense a lot of folks have thrown out that kind of thinking as we touch on this new topic.

      Two issues are being entirely confused: 1) science, which can only be commented on the context of a population (e.g., a statistical commentary on a population through a controlled experiment) and 2) personal choice – deciding if you’re one of the folks for whom the science implies a successful treatment.

      On average, using amiodarone for certain types of heart rhythms DOES save lives. But it has horrible side effects, as Marilyn shared in the case of her friend. Same is true of antibiotics. I’ve watched people die of Stevens Johnson Syndrome from taking an antibiotic for a common infection. Does that mean we shouldn’t use antibiotics? Well, it’s a personal choice. Nothing in life is free. Everything has a risk. The goal is to be able to understand the risks and benefits of every intervention (e.g., a drug, a food choice, driving while texting, heroin use).

      All have something to give. All have something to take. You’ve got to use the DATA to make a decision for yourself, but to say “Texting is dangerous” is out of context. It’s dangerous while driving.

    • Dan  May 13, 2012

      Dave, I think a combination of Dean Ornish and Gary Taubes is best — you can get the best of both worlds by going down the oft-detested “moderate middle”. In my own case, being a cholesterol hyper-absorber with a tendency towards metabolic syndrome and a family history of metsyn, diabetes, heart disease and stroke, that means reduce both dietary cholesterol, saturated fat AND carbohydrates. Any diet that is extreme may have extreme consequences for the genetically unfortunate. Judging from the diabetes statistics, there are many, many genetically unfortunate people. Judging from the dyslipidemia statistics, there are many, many genetically unfortunate people.

      Strict vegetarianism probably saves lives but may actually harm those who are carbohydrate-intolerant (which represents many people). On the other hand, large quantities of pure saturated fat and dietary cholesterol could harm someone who hyperabsorbs or has other derangements in metabolism and/or receptor function (e.g. heterozygous FH). That is why I think the moderate middle is best (this is NOT the food pyramid by the way, but I do advocate personalized food choices to my patients).

    • David Nelsen  May 14, 2012

      Dan, I think we all have to fine tune our diet to what provides us the optimal health. The fundamental difference between Gary and Dean is that Gary questions everything and doesn’t assume he knows the answer. The NuSi initiative I believe is one to get solid answers to the questions he’s been driving. I think Dr. Ornish (and other in his camp) could not or would not consider an alternative hypothesis to the all vegetable plan he’s supported for decades now. This gets to the bottom of what I like about Gary and by extension Peter.
      By degree I am an engineer, but if you really boil it down to what my company pays me for it would be to solve problems. There are a lot of different problem solving techniques out there (Model based problem solving, 7 Step, a Russian one with a Russian acronym called TRIZ). The techniques may have some differences but one thing that is common to all of them is that to solve a complex problem you really (really) have to have a good problem statement that fully explains the issue at hand. If you don’t do this, you end up with a lot of the poorly designed studies done by scientists who didn’t fully understand the problem or issue they wanted to resolve (especially in the field of nutrition). I don’t care how much data crunching you do, you will never get a value added result if your experimental design and problem statements were not well defined. An example of this might be a study that examines Low Carb eating where the Carb content of the diet is lowered from 60% to 40% of calories consumed. You could never achieve ketosis or achieve many other benefits of this approach with this lower bound on Carbs. This type of critical thinking must be used by the reader when they read these studies so that they can discern what the study did or did not prove (if anything).
      Problem solving efforts that generally fail are ones where people assume the problem statement is obvious and bypass the work needed to generate a good problem statement. Instead they jump to the solutions that they already know. An example of this type of bias in the nutritional realm would be that direct observation of obese individuals reveals that they eat too much and are not active. There is no need to dig deeper if the “Solution” is so readily know and universally agreed upon. If model based problem solving is used (and essentially Gary works through something like this in his books). This first pass Model would be written. You would then have to search for data that either supports the model or contradicts it. Gary in his books provides ample data that refutes obesity and heart disease models that refute the Sloth & Gluttony and eating Saturated Fat clogs your arteries models. In MBPS, you then come up with a new model and then either build it up or knock it down. Even if you come up with a new model that has no data against it, you can’t be sure it’s the right one. It just then becomes the prevailing theory until something better comes along.
      This is a long winded way of saying that everyone owns their own health and you need to skeptic for the health claims that anyone makes. You can read about some secret herb in Africa that the natives chew and that makes them lean or taking a Lipozene pill will slim your fat effortlessly. Even if the person isn’t selling a pill and like Ornish are promoting a theoretically healthy way to eat and live you need to skeptical. A fair percentage (3%) of the US population are vegetarians’ and most would assume they are eating that way to achieve optimal health, if not for religious reasons. Which reminds me of the joke, Question: How can you tell if someone is a Vegan? Answer: Don’t worry, they’ll tell you. Now a lot of people don’t have the time, interest or inclination to do the digging for themselves. Most people who read this blog are probably interested in LCHF way of eating. No 2 people who read it will implement in exactly the same way, however, given different genetics and general likes and dislikes. How much testing you chose to get to assess your health is also up to the individual. I’m getting my annual physical for work tomorrow. Work pays for it and encourages it and I feel it’s a good idea to get it done. Can this catch every possible problem I could have? Of course not, but it’s the prudent thing to do. I’m getting into an Attia length post here. So in short – determine what you believe will provide optimal health. Eat and live that way and see if you’re happy with the results.

    • Peter Attia  May 14, 2012

      David, can I hire you to work for NuSI? 🙂 Seriously, I’m glad you’ve responded. I try not to get my feelings hurt when people put me in same sentence with Dean Ornish. Not because I think he’s a bad person or wrong, but because he’s like so many completely non-critical thinkers who have allowed us to get to where we are today. I think Dan was probably just using his name to refer to a “low fat diet,” but your point is very important, and actually quite subtle. Certainly missed by most.

      My opposition is not people who disagree with me. My opposition is to a system that does not foster science – the real discipline of doing what folks like Richard Feynman did. While we were searching for a name for NuSI, we came very close to calling it The Feynman Foundation. It’s what he would have done. He’s rolling over in grave right now at the horrible stat of affairs in nutrition “science.”

      Personalization of nutrition is vital, but to do without any scientific basis is frightening, at best, and foolish, at worst.

    • David Nelsen  May 14, 2012

      Peter, my work experience is nearly all related to manufacturing computer chips (if you throw out the 6 months I spent working for the EPA keeping track of sewer treatment plants). The biggest draw dropping moments for me when reading Gary’s books and articles (Politics of Salt, etc) and critiques of the China Study by Denise Minger for example – is the lack of objectivity of the “Scientists” performing and analyzing the data. I understand if you work for 20 years trying to prove a hypothesis (Salt consumption is bad for blood pressure) that you could delude yourself and your results. I don’t think this should happen, but its human nature. The prospect of realizing you spent that much of your life working on a false hypothesis might be too big of an issue to confront. The peer review process should work these things out, but as I said in a previous post you can’t fix a poorly designed experiment – and I should know because I do a lot of experiments.

      I work at a for profit company and our goal is to maximize our good die (computer chips) out per lot. If I work on a change to save money, improve yield etc – I can get a false outcome if I run a small data set. Statistics clearly says can happen on small data samples this like this. For changes that could be impactful we run pilots. These are designed to clearly prove/disprove that the change we want to make does not have adverse side affect’s. Luckily for me, every batch has a report card when it finishes processing. I can tell if my change caused any shift in device performance or yield. I don’t care how married I am to a change, if the data does not support it we won’t implement it. This is a systematically enforced discipline as we are all aligned to our goals of good die out and maximizing profit. There does not appear to be any type of course correction in the field of nutrition science (at least given the last 60 years of data). Otherwise we would not have been encouraged to eat PUFA’s, HFCS, Trans Fats, etc.

      Nutritional science is not nearly as cut and dried as semiconductor manufacturing. That said, this is still not a justification for the poor quality of science served up to the general public to date. It’s actually pretty sad to me that we have very few voices crying out in the wilderness (Insert Gary & Peter & Co. as a metaphorical John the Baptist’s) about the crisis. The fact that the American Diabetes Association doesn’t advocate Carb restriction is disturbing in and of itself. They do recommend watching what you eat for Carbs and Sugars but don’t say that they should be avoided. I think this is some real cognitive dissonance at work here. In the meantime, self education and awareness are needed to ensure you are doing what you can to optimize your health.

    • lorraine  May 15, 2012

      David Nelsen, nice post about models of problem solving. I think that many folks are succeeding in developing some basics in evaluating study design, stuff like randomness and control. But constructing the testable null hypothesis, and passes of problem solving, that’s the meat.

      The long terms effects of diet manipulation in humans is an almost untameable beast, but it does not excuse poor study design, data trimming, and conclusions that are not supported by the evidence – all of which have been, and still are, rampant in nutrition publication.

      Haha, Peter, don’t get your feelings hurt. No haters here.

  28. Marilyn  May 13, 2012

    @Ed. Amiodarone. One of our friends took that for irregular heartbeat. He no longer has an irregular heartbeat. He no longer has a heartbeat at all. He was in his mid-50s.

    • Barbara Hvilivitzky  May 16, 2012

      Sorry to hijack your comment, Marilyn, but can I just say that I really learned something from David Nelson’s comment! Wow! Great stuff about defining “the problem” carefully WAY before you attempt to come up with solutions….I may actually think before I speak….or I’ll try!!! ;}

  29. Jeff Johnson  May 14, 2012

    Cholesterol exists in 2 forms – unesterified or “free” (UC) and esterified (CE) – and the form determines if we can absorb it or not, or store it or not (among other things).

    Much of the cholesterol we eat is in the form of CE. It is not absorbed and is excreted by our gut (i.e., leaves our body in stool). The reason this occurs is that CE not only has to be de-esterified, but it competes for absorption with the vastly larger amounts of UC supplied by the biliary route.

    Re-absorption of the cholesterol we synthesize in our body (i.e., endogenous produced cholesterol) is the dominant source of the cholesterol in our body. That is, most of the cholesterol in our body is made by our body

    If the above statment’s are correct – then what difference does injested cholesterol make to a person with genetic Familial hypercholesterolemia – if the gut gets rid of ‘it’ – what does liver LDL-R(receptors) have to do with them – meaning – the gut gets rid of it – not the liver and since a higher saturated fat-lower carb diet produces lower fat levels in the blood – what does saturated fat have to do with ‘it’

    Reading Dan’s post(a doctor who actualally treats Familial hypercholesterolemia) above mine – I’m somewhat confused –

    • Peter Attia  May 14, 2012

      Cholesterol ingestion is not the issue in FH. I don’t know where this confusion is coming from. The problem in FH (there are several kinds, each with its own defect in the LDL receptor) is that the liver can’t clear LDL particles. These patients typically have high LDL-C and LDL-P. What’s going on in their ABCG5,G8 is irrelevant. All patients with FH need very aggressive treatment with statins.

    • Bob  May 15, 2012

      According to Uffe Ravnskov, patients with FH do not need statins. See Fat and Cholesterol are Good for You.

    • Peter Attia  May 15, 2012

      Good luck with that, my friend.

    • moreporkplease  May 16, 2012

      Hi Peter:

      Dr. Ravnskov does indeed argue some interesting points about FH. First he notes that some drugs may be effective in those with certain types of FH because they do other things than lower cholesterol. For example, those with too much Factor 8, or prothrombin, or fibrinogen were trialled, 1/2 with simvastatin, the other half with Vytorin (a statin + ezetimb). The Vyortin half did well, but it’s apparently because the ezetimib that drug contains prevents their blood from clotting up on them. Those on statins showed no benefit. He cites the famous ENHANCE trial here. I find Dr. Ravnskov to be a colorful guy with strong language, but his footnotes and cites seem sound. 🙂 Therefore, if you are an FH person with the above 3 characteristics, statin won’t help, but the ezetimib will. Or so the ENHANCE trial found. Do you disagree, Peter?

    • Peter Attia  May 16, 2012

      This is different than the original assertion that there is no role for statins in FH. Almost without exception a patient with FH (remember there are many types of FH) reduces LDL-P and cardiac risk with a statin. Ezetimide can help only if they are hyper-absorbers, also. Otherwise it can make things works.

    • Debbie  May 19, 2012

      Hi Peter,

      Does FH correlate with extremely high HDL as well as LDL-C?


    • Peter Attia  May 19, 2012

      Not necessarily, but it can.

    • Vernon McVety Jr.  August 2, 2015

      Confusion is part and parcel of the disinformative trickery used by the drug companies and their allies. The more medical confusion they can muster up the more they can make legitimate research (anti-bad cholesterol facts by alternative physicians) appear to be “disinformation.” I’m not saying any of the controversy in these replies are intentional disinformative trickery, but the words and ways of expression are very close to what the pharmaceutical adversaries like to use. The drug companies will do any thing possible to keep the truth about heart disease from becoming entirely publically known.

  30. Ed  May 14, 2012


    I brought up the issue of LDL-R because as far as I can gather, it is a pathway for the disposal of the LDL PARTICLE itself. Peter has done a great job elucidating how cholesterol gets into and out of the body.

    But what about the “particle” in your “bloodstream”? That’s the question I would like to explore.

    As far as I can tell, a major way that the particle gets out of the bloodstream is through something called the LDL receptor (and there are probably others, which are outside my current level of understanding). These are on cell membranes throughout the body. I’m not sure if the liver is the most important site for LDL-receptor activity, but I don’t think it matters for purposes of this discussion. The LDL receptor effectively gets rid of the particle from the bloodstream because when a circulating particle binds to the receptor (as it’s designed to do), then the cell takes in the particle via endocytosis (it eats the particle!).

    The FH paper is just an EXAMPLE to show if your LDL-R don’t work well (because of genetic mutation), this will have a big impact on your level of circulating particles. I believe the injection of cholesterol in that particular study is only related to the technique used to estimate the half life of the particle, and that can be ignored for purposes of this discussion.

    The key point is that people with poor LDL receptor function due to genetic mutations have higher circulating particles and cholesterol levels in the bloodstream because these receptors don’t do as good a job at pulling particles out. Its like vacuuming a rug with a Dyson or a Dustbuster. You will eventually get the room clean with a dustbuster but the dyson will do it a lot faster. If dust is falling at a constant rate onto your rug, those vacuuming with dysons will generally have less dust on the rug at any given time. Obviously, living people do get to some level of balance, because their circulating levels of LDL-P dont keep rising forever, but clearly, circulating levels of LDL-P can be influenced by your LDL-R activity.

    The Amiodarone is just another EXAMPLE of LDL receptor activity. You take this drug (and it was a completely random selection of drug), and your LDL receptor activity goes down. And what do you know… you’re starting to look like someone with FH.

    Low thyroid hormone is also a downregulator of LDL-receptor. Go hypothyroid, and your LDL soars.

    So how do we help those LDL-receptors of ours to keep working to the best of their abilities?

    • steve  May 14, 2012

      Am sure Peter will get in to this as the series continues, but LDL-P in the blood stream is not only related to issues of clearance capability, but also to size,core composition(how much cholesterol vs TRG there is) and LDL particle production. Some of this is genetic and some related to cholesterol metabolism- CETP.
      Lecturepad which Peter referenced is a tremendous resource for info.
      Statins upregulate LDL receptor activity. If thyroid ok and to many particles floating around meds may be in order.

  31. George Henderson  May 14, 2012

    there is an interaction between thyroid hormone, PUFA and LDL-R levels.

    PUFA- if diet is high in PUFA it is taken up in cell membranes which become too soft, requiring more cholesterol for stiffening.
    This causes the liver to make more C, and cells to make more LDL-R to pull it out of the blood.
    Thus serum LDL-C is lowered – but there is now more cholesterol in the body overall (so there might be a rebound if you lower PUFA).

    Low thyroid T3, for example if selenium, iodine are deficient, or sometimes in zero-carb states, can also decrease LDL-R.
    This means that T3 and LDL-R levels must correlate, and there is indeed cross talk between PUFA and thyroid status.

    • Ed  May 15, 2012

      @ George-

      Thanks for thinking about LDL-R… helpful.

      Biology is so wonderfly complex! I’ve definitely been racing to figure out the thyroid connection… you see a lot of anecdotal murmurings and T3 discussion on various low carb forums, so I don’t think this is a link that can be dismissed. T3 is definitely part of the gene sequence that regulates LDL-R expression. A simple pubmed search on those 2 terms generates some striking abstracts related to the issue. So low T3 can downregulate LDL-R and dramatically increase LDL-P and LDL-C.

      PUFA concentration in cell membranes is another interesting topic. I didn’t think of PUFA of having much impact on LDL-R, because cells need more C for stiffening. I will do some digging on that if I get some free time.

  32. J Perry,MD  May 14, 2012

    Thanks for your good work.What do you think about increasing LDL and high LDL-P (>95 percentile,half small dense) on a low carb diet,weight stable.APOE 2/3.I have read all of Dayspring”s stuff and know how to treat it.But what do you think is happening to cause this in light of the macronutrient content?very low carb,no restriction on sat fat,minimal vegetable carbs,some wine.blood sugar is normal.
    I have searched Volek and Phinney and other literature. While they state this can happen to some extent,this seems high TCL 299,LDL-C 199,LDL-P 2300,HDL 57,TG 66.
    trying to learn more about that component of low carb dieting as I see it happening often.

    • Ed  May 15, 2012

      @J Perry,

      An MD seeing it happen often! Wow, that’s a very useful datapoint. Google the LC threads out there and it seems to me there is a subset of people that do react this way, probably via T3 levels. Can we take this offline, via a discussion board or email. I’d love to exchange some data with you, and dont worry, I wont bore you with my personal details as I will not ask for medical advice… only discuss the issue on general terms.

  33. J Perry,MD  May 15, 2012

    there are some who are recommending doing these tests non-fasting.I talked to someone at HDL who recommended it.what do you think?what about rising TSH with the VLCKD?

    • Peter Attia  May 15, 2012

      Fasting state obviously impacts TG, glucose, and insulin levels, but plays no role in determining the values of particles or even cholesterol contained within them.

    • Ed  May 15, 2012

      @J Perry

      VLCKD and rising TSH? Is this something you have anecdotal evidence for?

      If you want to explore the POSSIBLE, and as I search various LC-related forums, I am concluding, PROBABLE, connection between carbohydrate intake and thyroid (and possible LDL-C and LDL-P via the LDL-R mechanism), read this.

      I’ve done enough reading to suggest this is a plausible explanation for rising LDL-P via T3 and LDL-R.

  34. Jeff Johnson  May 15, 2012

    Scenario: Assumptive Reasoning: – Assuming there is a thin layer of calcium build-up in the arteries close to the heart(assuming the heart likes calcium to maintain it’s electrical function)

    Assuming there are tooth decay bacteria flowing with-in the blood stream – these bacteria see this calcium – attack or eat it – rot it – the inner lining of the artery sends out a red alert-red alert message saying something bad is happening to my lining – LDL comes to the rescue – plugs hole –

    Tooth Bacteria don’t care – plaque tastes good too – so this destroy/repair process just keeps going on

    There may other bacteria or chemicals or even viruses maybe that attack the artery lining as well –

    The quality of the plaque build-up might be an issue – as well – better plaque – better pizza – as Papa John likes to say –

  35. Kypros  May 15, 2012

    Do you think that any test results showing a specific particle concentration and size would lead you to recommend a diet different than a very low carb one?

    • Peter Attia  May 15, 2012

      Not that I’ve encountered. The issue of diet choice may be more around other factors, though.

  36. Ed  May 15, 2012


    Is it possible to build a function to keep tabs on new comments on a thread via email or other mechanism? I know some blogs have that capability… not sure if wordpress does, though.

    Great work! Keep us educated!

    Look forward to next episode.

    • Peter Attia  May 15, 2012

      I don’t know…I’m a bit too overwhelmed to even contemplate this right now.

  37. Busflier  May 15, 2012

    This is getting too complicated…I’m going back to my “SAD”…pizza…spaghetti…beer…BREAD…live happy! (just kidding).

    • Peter Attia  May 15, 2012

      And cotton candy… 🙂

    • Debbie  May 16, 2012

      It is too complicated for me! But, at least people are talking about the complexity of cholesterol. I’m awaiting the results of a more advanced lipid profile; my cardiologist sent my blood across the country to California. My numbers are weird, and I know I’m not supposed to post them, but…

      HDL – 161
      LDL – 150

      Everything else normal. I don’t want to take a statin!!

    • Peter Attia  May 17, 2012

      Who says you need one!?!? Just make sure LDL-P is ok…

  38. LynneC  May 15, 2012

    Will you touch, at some point, upon apoE genotypes and diet, or is it your view that apoE is irrelevant?

    • Peter Attia  May 16, 2012

      Absolutely. This is a very important and relevant topic.

  39. Gabriel Zirpolo  May 16, 2012

    I know your overwhelmed but if you have any spare time and wanted to appease Ed there is a wordpress plugin to subscribe to comments. Not sure how it works or if it’s any good but you can find it here.

  40. Larry Sumners  May 16, 2012

    I hope part of the future posts will cover some of the possible treatments etc to lower ldl-p. i maintain a low carb diet with statins and niacin with regular exercise and still have high numbers- maybe just the curse of dna.

    • Peter Attia  May 16, 2012

      Yes, but first we need to understand the problem a bit better. This week I’ll write about the “how” of atherosclerosis. Thanks for your patience.

  41. sugaraholics  May 16, 2012

    I don’t understand the science of all this, but I know that pills are not the answer, and a healthy vlc diet and getting some exercise are the best things I can do. My doc suggested I go on statins, and I told him I didn’t believe in them; happily, he didn’t argue. I’ve decided I don’t want to become a elderly person, like so many in my family, who lives on meds and from dr visit to dr visit. The 60s onward is supposed to be the the “freedom” time of life.

    –but, I will take pills for migraines, so I’m no purist.

    • Peter Attia  May 16, 2012

      Pills are sometimes the cure, but with 2 caveats: 1) all else has failed, and 2) you actually know what you are treating (i.e., LDL-P not LDL-C or HDL-C).

  42. johnk  May 16, 2012

    This just hit the NY times – I’ll let the knowledgeable ones explain it all.

  43. Barb Janoski  May 16, 2012

    Here we go…‘Good Cholesterol’ May Be an Overstatement, Study Finds”

    My brain is still hurting from Parts I – III. This article doesn’t help any. 🙂

    • Peter Attia  May 17, 2012

      Yes, I very briefly allude to this in part IV, which I’ll be posting tonight, hopefully…
      There has never been any evidence that pharmacologically increasing the cholesterol content of an HDL particle (i.e., HDL-C) could fix atherosclerosis. You’ll understand better, I hope, in about 24 hours. Stay tuned!

  44. Jeff Johnson  May 17, 2012

    Thinking: Cell Permeability leads to improper use of and depositing effects of Calcium(plaque)

    Race Horses Diet: Race horses are usually fed a fat supplement in their feed to increase energy density

    The fat used(common) is heat processed corn oil – an omega 6 type fat that increases cell membrane permeability

    Race horses are always breaking legs – suffering bone fractures – extremely common

    My thinking is: Cell Membrane Permeability is linked to Calcium Deposits in the lining of the Arteries

    Conclusion: Sugar feeds harmful Calcium Rotting Bacteria – Cell Membrane Permeability
    weakens the inner lining of the Arteries – causing Calcium to stick to it – as well as allowing to easy an access pathway for LDL repair work -: Calcium Sticks to Artery Wall – Inflammation Increases at site deposit zone – LdL repairs it(plaque)

    Closing Comment: – My last ski race of the season – (Super G – Park City) – there is this Doctor in line in front of me – I’ve heard him talk this omega 3 fish oil study his is doing –

    Frankly – he didn’t look all that great – bad left knee and he just looked sort of like sick or something –

    It didn’t seem fish oil(I assume he uses it) was doing him any great favors – purely observational – but something I recorded in my play book(brain)

  45. bobbydean  May 19, 2012

    Can you get HDL test done from Canada? How would you do this. My doc doesn’t know of NMR either.
    Peter your work and energy you put into this site is amazing and so appreciated. I hear lots of great things about you on many cardio / lo carb boards and people are learning and watching you, keep up the good work!

    • Peter Attia  May 19, 2012

      Not at this time, unfortunately. Your doctor should be able to send samples to LipoScience for NMR, though.

  46. LalBeral  June 1, 2012

    Peter, can you tell us if the NMR test by LipoScience requires fasting? I found their web site but didn’t readily see that info. on it. Thanks!

    • Peter Attia  June 2, 2012

      Lipoprotein measurement is not impacted in the short-term by fasting. Other parameters are (e.g., insulin, glucose, TG).

  47. Jennifer  July 18, 2012

    Hi Peter,
    Thank you so much for this wonderful free education about lipidology!
    I have 2 questions–

    1. With HDL, is HDL-C a meaningful measure? That is, is a high HDL-C really reassuring?

    2. When you say that it is the LDL particle number that drives the particles into the artery wall, do you mean that literally? Atherosclerosis is almost completely a function of there being too many of those particles, regardless of something to do with the endothelium or other factors?

    My husband’s doctor was concerned about his standard panel (good TG, good HDL, LDL-C 132), which I wasn’t too worried about but his mom died of MI at age 63 and in the past year he’s getting some signs of metabolic syndrome and those things worry me. We would have to pay out of pocket for anything other than advice to reduce sat fat/cholesterol/calories and get more whole grains & exercise, and I’m wondering if we should try to get the NMR test, or the ultrasound of the carotids, or just to try to improve diet the LCHF way and get more exercise.

    • Peter Attia  July 19, 2012

      1. In an of itself, no. It’s just a measurement of how much cholesterol is being carried in an HDL particle. One of the posts in this series covers this in more detail. I think part 6 or 7. A high HDL-C can, however, be a marker for efficient mobilization of TG and CETP transfer between LDL and HDL particles, in both direct and indirect RCT.

      2. Yup. The artery wall is made up of endothelial cells. The particles “sneak” in between them and the rest is bad.

  48. Jennifer  July 19, 2012

    THank you so much for answering my questions. This series is amazing–I am learning so much!

    I keep having to stop to look up terms I don’t understand and when I looked electrophoretic field I got: “A swarm of space-dwelling lifeforms encountered by the USS Voyager in 2371 generated an electrophoretic field. The field increased electrophoretic activity in Kes’ nervous system, causing her to enter premature elogium.” Turned out to be a Star Trek fan site.

    Anyway, Star Trek notwithstanding, I have learned so much and I’m only on part IV so far. This is awesome!! Thank you!!!

    • Peter Attia  July 20, 2012

      Thank you. Hope you enjoy the rest.

  49. Lex  August 17, 2012

    Frustration here. Husband had a blood draw at LabCorp for the NMR LipoProfile, and we just got the report. Somebody made a mistake somewhere; they reported only Lipoprotein(a). I don’t even know if this is useful info by itself (11.1 nmol/L). At least LabCorp says they will re-draw if it is their mistake. Sample already discarded by LipoScience; I checked. (LabCorp does ship out to LipoScience.)

  50. bhill  August 21, 2012

    Hi Peter, I had a couple question regarding LDL size vs LDL-P. Is someone that is classified with Pattern A LDL at a lower risk than someone who is classified as Pattern B? Sorry if this is a dumb question but I frequently read (on Mark’s Daily Apple) about large buoyant LDL being “better” than small dense LDL. When I finally read your blog I was like “YES!, it’s the actual particle count the matters..”. Now I’m confused. Does the size relate at all to particle count?

    • Peter Attia  August 22, 2012

      Size is irrelevant once you know number. Small size often (but certainly not often enough to justify it) predicts large numbers (bad), but the focus on particle size is misguided.

    • bhill  August 22, 2012

      Thanks for answering my question. Your series on cholesterol is very enlightening. I look forward future posts!

  51. Michael Martz  November 16, 2012

    Thanks for the series.

    I think you *may* have an error in your description of the NMR IR sample. You say that the person has “small VLDL particles, small HDL particles, and LDL particles”. From my understanding, the person has large VLDL particles (undesirable), small HDL particles and barely acceptable LDL particle size (NMR uses 20.5nm as the cutoff between pattern A and B, so probably A/B or I).

    I’ve tried to understand the IR section on my 2 tests, but since I’m not a doctor, NMR won’t tell me anything about these numbers. But as far as I can tell from listening to Dr. Dayspring and researching on the net, the particle numbers, “size” X-P, (i.e. molarity) are for a specific size of the particle, but I’ve never found a definition of exactly which subclasses these map to (e.g. H5-H1 or V6-V1).

    I suspect the size numbers in the IR score are an average for the lipoprotein class as a whole and NOT the size of the corresponding subclass(es) particle count in the IR Section. (My LDL size in the IR section is always the same as my overall LDL size score, at least that’s the way Genova Diagnostics reports it)

    An HDL size of 8.7nm would be either 3a, Medium or alpha 2-3. Also, seemingly supporting an overall average HDL size, as opposed to the average size of just Large HDL particles.

    What’s your opinion of what sublclass(es), if any, these size numbers represent?

    Back to the example, wouldn’t this person then have: mixed signals wrt VLDLs (large size, yet few particles), barely poor HDL (small size and below median count) and a almost average LDLs, more than average small particles yet barely acceptable in size?

    • Peter Attia  November 20, 2012

      Michael, thanks for the heads up. I’ll be back and check. Regardless of what I wrote, you are correct (and I’ll fix if necessary). IR is best predicted by LARGE VLDL and SMALL HDL and LDL.
      Now, to your real question — which is a great one — it’s important to keep in mind that particle size is a marker, but one that needs to be take in the context of other factors. For example small LDL (ore via VAP type 3/4) is a marker for IR, but by itself says nothing more or less about atherogenicity. On the HDL side, it gets much trickier because generally large particles are thought to * correlate* more with insulin sensitivity, but small particles are thought to be more function from a RCT standpoint.
      If this poses some confusion, you’re not alone…

  52. Greg  February 22, 2013

    NMR’s website about what to do if you get poor results from the lipo test:

    “Eat more whole grains, fruits and vegetables. Eating more of these foods, like whole-wheat pasta, beans, nuts, berries, spinach and broccoli, are good for your heart. Limit your red meat — like steak and hamburgers — eat fish at least once a week and lean turkey or chicken without the skin. Use olive oil or canola oil in place of butter. Avoid fried foods or those meals served with heavy sauces.”


    • Greg  February 22, 2013

      (Sorry, I mean LipoScience’s website, obviously…)

    • Peter Attia  February 22, 2013

      Yes, I’ve pointed this out…

  53. John Barlowe  May 9, 2013

    Excellent work on this series. I’m learning so much! In case you don’t know, the link ( in part III doesn’t work. At least today it doesn’t.

  54. Jerry  July 11, 2013

    Just discovered your blog and work, wonderful science. To the point, any research or opinions on hypothyroidism and cholesterol. I’m been on Synthroid for 15 years and have had fluctuating cholesterol numbers over the years. I believe Synthroid has had some consistency issues and thought this could be a factor. My doc put me on 10mg Lipitor. Am I barking up the wrong tree.

    • Peter Attia  July 11, 2013

      Yup. Pretty clear, especially in women, that hypothyroidism needs to be corrected before lipid meds are started. In other words, correcting hypoT can fix the lipid abnormality. Recently saw a dramatic response to this in someone I work with.

  55. Gionei  January 6, 2014

    Please, how many different kinds there are LDL- P . Are 9 or 11?
    I read everything that you brightly typed. I learned a lot!

    • Peter Attia  January 6, 2014

      Doesn’t matter. Each one probably counts the same toward risk.

  56. Mary  April 10, 2014

    I tried to print out this article to take to my doctor for my next visit. I’m afraid that me just trying to regurgitate this info to my doctor will be met with rolled eyes, as I’m not a medical professional. Internet Explorer will not print, print selected, nor allow me to save the page. I assume this is by design. Could you make a printable link or could you send me a pdf of this article to print out? Thanks for this very informative series.

    • Peter Attia  April 10, 2014

      Copy and past content into WORD and print.

  57. Mary  May 7, 2014

    The link in item #7 about the study of dietary cholesterol and coronary disease is broken. Here’s where it sent me:

    There are several similarly-named studies, so could you provide a good link that I could print off for my cardiologist?

  58. Shyam  November 9, 2014

    Hello Dr. Peter,

    Thanks for reflecting this wealth of information about cholesterol and benefits of low carb diet in simple terms. I am from India and unfortunately we do not have NMR based lipid profile tests yet. But fortunately there are a few labs which checks for Apo-B / Apo-AI levels. Along with these markers the report also provided a measure of Homocysteine levels. I read online that elevated levels of Homocysteine might be prone to endothelial injury. Is this true? If yes, is this also a precursor to Atherosclerosis?

    Thanks again for your help.

  59. Andrew Kuhn  June 17, 2015

    Dr. Attia,

    For those of us in the community who have physicians who are either unable or unwilling to sign up with HDL for the NMR LipoProfile testing, could you please list what specific panels you order for patients when doing the full workup?

    (I ask because my physician wanted a standard VAP, and after a lack of call back to his office, he isn’t interested in signing up with a 3rd lab currently). I still want the full workup results, knowing they are superior to the VAP test.

    Thanks for your time.

    Andrew Kuhn

    P.S. – I sent you an email to your NuSI email connecting you two as I believe you could bring great value to each other. He’s a neurosurgeon via profession with a passion for anti-aging and I forwarded him your whole series as introduction. if you’re interested. Thanks


  60. Kirstin Mapstone  July 19, 2015

    Good morning Dr Attia,

    Thank you for sharing this series on cholesterol. I am finding it very insightful in this journey of figuring out what the ‘real deal’ is with regards to cholesterol.

    I am a 22 year old female with genetically disposed high cholesterol. I exercise regularly and eat well. I have been on statins (first Crestor, and now Zuvamor, as they are known in South Africa) since the age of about 17. These medications do indeed bring my cholesterol levels to those deemed as ‘acceptable’ by my GP, however they come with some very undesirable side effects. About a year ago I was put on a high dose of Zuvamor to bring my ‘Total Cholesterol’ down to < 5.5. This lead to very bad bruising and muscle pain, which is not ideal for someone who loves exercising. In December of last year I requested that I be put on a lower dose because of the bruising. Now, on 20 mg of Zuvamor, I do not bruise as easily and muscle pain is not as bad as before.

    I know that this series on cholesterol refers to people in the general population who have high cholesterol due to lifestyle/diet reasons. How does this information relate to one who has unusually high cholesterol due to genetic reasons?

    I have just one more question. In this article you mention NMR testing for cholesterol particles.

    "The most frequently used and guideline recommended test that can count the number of particles is the NMR LipoProfile. In addition to counting the number of particles – the most important predictor of risk – NMR can also measure the size of each lipoprotein particle, which is valuable for predicting insulin resistance in drug naïve patients, before changes are noted in glucose or insulin levels."

    As a fourth-year BSc student I'm glad I know what this is, and understand how it works, because it's quite a complicated thing to understand :). Ideally, I would like to be able to have my cholesterol levels tested using this method, however I have not been able to find any laboratories/pathology labs that perform such a test in South Africa (where I live). I know that this is a big stretch, but do you know of anyone in South Africa who will know where this can be done?

    Kind regards,

  61. Allen Carriker  January 2, 2016

    Hi Peter,
    It was announced that HDL, Inc. (True Health Diagnostics) and LipoScience (LabCorp) no longer have a business relationship. Do you have a recommended alternative to HDL, Inc. that provides the additional data beyond the NMR LipoProfile test from LabCorp ?

    • Peter Attia  January 5, 2016

      THD bought HDL after the BK. The technology of LS is used by THD and others.

  62. Cassie  April 5, 2016

    Hi Peter,

    Are parts I and II still available the links bring me to an error page, I just wanted to start from the beginning. I want to have a solid defense next time one of my friends claims they can’t eat a piece of bacon due to cholesterol, ha.

    • Peter Attia  April 6, 2016

      Should be but you might need to search directly for them rather than use embedded links.

  63. abby harris  April 11, 2016

    Hi Peter,

    I am confused about my results and hope you can explain.

    My LP (a) is 78 however my Lp (a) cholesterol is 21
    What is the difference and is it high either way? Here they are:

    Test Low Normal High Reference Range Units
    Lipoprotein (A) 78 0-30 mg/dL

    Vap Cholesterol Profile

    Test Low Normal High Reference Range Units
    Ldl Density Pattern A A
    Ldl4 Pattern B 5.8 mg/dL
    Ldl3 Pattern B 29.3 mg/dL
    Ldl2 Pattern A 39.8 mg/dL
    Ldl1 Pattern A 13.0 mg/dL
    Vldl-3 (Small Remnant) 9 25 mg/dL
    Hdl-2 (Most Protective) 28 >15 mg/dL
    Probable Metabolic Syndrome No No
    Remnant Lipo. (Idl+Vldl3) 15 <30 mg/dL
    Idl Cholesterol 6 <20 mg/dL
    Lp(A) Cholesterol 21.0 <10 mg/dL
    Ldl-R (Real)-C 88 <100 mg/dL
    Apob100-Calc 79 <109 mg/dL
    Non Hdl Chol. (Ldl+Vldl) 131 <160 mg/dL
    Triglycerides 50 <150 mg/dL
    Cholesterol, Total 222 <200 mg/dL
    Vldl Cholesterol 15 =40 mg/dL
    Ldl Cholesterol 115 <130 mg/dL

  64. boon yew lee  October 2, 2016

    I have been in ketosis for a month but my ldl-c shot up by more than 50%, over here, we do not have the means to perform ldl-p test, should I continue with this diet?

    I know you are busy but hope you could reply me. Thanks.

    Ps: my triglyceride went down and hdl went up.


Add a Comment

Disclaimer: This blog is for general informational purposes only and does not constitute the practice of medicine, nursing or other professional health care services, including the giving of medical advice, and no doctor/patient relationship is formed. The use of information on this blog or materials linked from this blog is at the users own risk. The content of this blog is not intended to be a substitute for professional medical advice, diagnosis, or treatment. Users should not disregard, or delay in obtaining, medical advice for any medical condition they may have, and should seek the assistance of their health care professionals for any such conditions.