Jun 26

Cholesterol Endgame?

Something exciting is happening.

There’s a special group of people that may be answering one of the most powerful medical questions of our time.

The Hypothesis

A hypothesis was made several decades ago, well before I was born. This “Lipid Hypothesis” suggested cholesterol in the blood independently caused heart disease.

From Wikipedia:

The lipid hypothesis is a medical theory postulating a link between blood cholesterol levels and occurrence of heart disease. A summary from 1976 described it as: “measures used to lower the plasma lipids in patients with hyperlipidemia will lead to reductions in new events of coronary heart disease”.[1] Or, more concisely, “decreasing blood cholesterol… significantly reduces coronary heart disease”.[2]

More cholesterol, more heart disease. Simple.

It has since morphed and changed a bit. Now some point to Low Density Lipoproteins as the problem (LDL-P, the “boats”) while others still insist it is the cholesterol found inside them (LDL-C, the “cargo”). Regardless, if you have a lot of LDL-C, you likely have a lot of LDL-P, so it is pretty much the same for most people anyway.

This is “deadly” and “needs immediate treatment” to lower the LDL-C/-P with prescription drugs. Even with the newer guidelines putting less emphasis on LDL, there’s no lack of insistence by doctors around the world on taking drugs to lower it when at high or even moderate levels. In fact, what was considered high has been continually dropped (from 130 mg/dL to 100, and now to 70). Naturally, this has resulted in more and more drugs prescribed.

Case Closed… or is it?

Last year we had a massive Consensus Statement from the European Atherosclerosis Society. In very declarative language, they state the evidence is quite overwhelming and clear: LDL is the bad guy.

Interestingly, this appears to be in the wake of many papers that are pointing to several new channels with higher associations to CVD such as Remnant Cholesterol, something I’ve written about here before.

In fact, even this EAS statement didn’t get very far before it got a paper that sought to show its many problems from the Karger group, A Critical Review of the Consensus Statement from the European Atherosclerosis Society Consensus Panel 2017.

There’s even been a quiet subtraction of LDL as a target for treatment in the new guidelines from 2013 AHA/ACC.

So what’s going on here? Is LDL a problem or not?

HDL High and Triglycerides Low? Then No, LDL Does Not Appear to be a Problem

When looking at only LDL, then one can find a weak correlation with CVD/CHD. But when looking at LDL alongside HDL and triglycerides (TG), it becomes nearly meaningless. HDL & TG together are just a far stronger measurement compared to LDL alone.

Over the last 4 months, I’ve been pinging LDL-lowering experts, organizations, and social media at large for studies showing high LDL is giving high CVD to normal, non-treated people who have high HDL and low triglycerides. Can’t find any studies. Not one. I even turned it into a hashtag campaign on Twitter: the #LCCholesterolChallenge. Or you’ve seen the graphic I’ve posted to help attract attention to the challenge.

Instead, I’ve found compelling studies that show the opposite — that high HDL and low triglycerides are associated with low CVD and low all-cause mortality.

Take this Framingham Offspring study, where I even marked it up to showcase where I (and many other hyper-responders) stood in the risk category. (Colored markings and overlay mine)

In other words, if you grab everyone who had a moderately low TG (less than 100), a moderately high HDL (more than 40 for men, 50 for women) — then even if their LDL was above 130, their risk was nearly identical to someone with LDL of below 100 (0.7 and 0.6, respectively).

But it gets better…

In this observational cohort study of 2906 men aged 53 to 74 years free of IHD at baseline, we see this relationship in perfect display. (Colored markings mine)

You see how on the left we have people with an LDL of 170 or less vs those with an LDL of 170 or more? That’s literally just 20 mg/dL away from one of the two criterions for a diagnosis of familial hypercholesterolemia! In other words, and LDL of 170 will likely get your doctor talking to you about multiple drugs to lower your cholesterol.

Yet there it is in black and white — the group on the left is nearly the same as the group on the right. If you’re in the dark grey bar with HDL of 57 or more and a TG of 97 or less, you’re at nearly identical risk for ischemic heart disease whether above or below LDL of 170. But worse, if you’re hanging out in the white bar and take your lipid-lowering drugs to drop below 170 with no change in your HDL of 46 or less and TG of 142 or more, you’ve made effectively zero impact.

So Why Doesn’t the Medical Community Focus on HDL and Triglycerides?

Good question. I wish I knew the answer.

Sure, there’s plenty of fodder for the cynical answer here: there isn’t a pharmacological solution to it. And yes, there appears to be some support for this given all the trials that chemically alter the lipid system to bring up HDL and reduce LDL, such as CETP inhibitors, have failed spectacularly. But weirdly, this has led to a new train of thought insisting maybe HDL isn’t so beneficial after all. (Note: I couldn’t possibly disagree more with applying the assumptions of a population who had their lipid systems chemically altered through drugs to that of the general population without supporting evidence.)

Enter the Lean Mass Hyper-responder

About a year ago I wrote about this emerging profile that was becoming very central to my research. It represents a kind of endpoint to very high fat-adaption as shown in higher circulation of VLDLs, resulting in high LDL. As a class, these individuals have the highest LDL-C and LDL-P of everyone in the low carb community.

Yet they defy expectations of what we’d imagine with high LDL — the insulin-resistant, junk food couch potato. Rather, they are typically lean, fit, and have very low fasting insulin. Many of whom insist this is the best they’ve ever felt and performed in their life. I see their bloodwork on a near daily basis and it is stunning how many of them have great metabolic markers, extremely low inflammation, CIMT, and CAC scores.

I called this profile “Lean Mass Hyper-responder” (LMHR) and I had no idea how just how many of them there were when I wrote the article. Now I know they are a sizable population in the low carb community.

Many LMHRs are concerned about their cholesterol and are taking steps to change it. However, many others are completely fine with it and have no interest in living any other way. (For the record, we try to help everyone meet their goals and urge community respect for their decision, whatever it may be).

Collision Course

So clearly, the ideal group of individuals to test the Lipid Hypothesis are those who are:

  1. Free of all major CVD risks, such as hypertension, hyperinsulinemia, high waist circumference, high triglycerides, and low HDL (to name a few).
  2. Have extremely high LDL-C and LDL-P

And here we are. If the Lipid Hypothesis is indeed an independent risk factor, then LMHRs should be showing signs of progressive atherosclerosis and higher mortality from heart disease — and rapidly, for that matter. There’s nowhere left to move the goalposts.

I think once we get a decent sample size of LMHRs and a solid follow up period of time, we’ll have a strong answer to that question.

You can probably guess what I’m predicting.

Leave a Reply

Photo and Image Files
Audio and Video Files
Other File Types
13 Comment threads
15 Thread replies
Most reacted comment
Hottest comment thread
15 Comment authors
DENI ROCHERogerRoger JohnsonMARK WOFFORDSiobhan Huggins Recent comment authors

newest oldest most voted
Notify of
John Pavao
John Pavao

I found this most encouraging: “In other words, if you grab everyone who had a moderately low TG (less than 100), a moderately high HDL (more than 40 for men, 50 for women) — then even if their LDL was above 130, their risk was nearly identical to someone with LDL of below 100 (0.7 and 0.6, respectively).”

After being keto for a year, my HDL is an all time high of 51 (never before been higher than 44), and my TG was 99 (even though I wasn’t quite 12 hours fasted). Without your work I would have no idea how to back up my hunch that my “high cholesterol” isn’t the problem the doctors make it out to be. Thanks Dave!


Hi Dave: I am still getting caught up on the discussion here after listening to you talks on the Keto Dudes podcast. I am a 52yo 121 lb. 5’2″ female. I sit in a chair all day but am active when not at work. I started eating keto at the end of March and am still in the fat loss stage of my ketogenic journey. Prior to going keto, I noticed a poor trend in my cholesterol panel although they remained within normal limits when last checked in 2017. Since I did not get a baseline lipid panel prior to starting Keto, I thought it was best to wait some time before having my numbers checked. I now wish I had requested the labs before as she just ordered labs (glucose and cholesterol panel). While I do want to see where things are now, I am a bit worried. I don’t necessarily want to biohack, I would like to time my blood draw to show the best possible outcome limiting as many confounding factors as possible. I normally IF 6 days/wk and do a 24h fast 1 day/wk. Would it be best to have the draw done after a day of (~16h) IF or following a longer (~24h) fast? Other than coffee and MCT oil, are there any other food/drink that I should avoid?


Hi Juls! I am someone who did the Intensive Dietary Management program with Jason Fung and Megan Ramos. I went through some of what you are going through and Megan’s advice was to not fast for a week before I got my blood work done. She said most people she says three days. I got 11.88 I think on my total cholesterol which freaked out everyone but I truly believe that fasting messed with the numbers. Doing a hack on my next test I brought down to about 6.45. However, they still wanted to give me statins! I have no other risk factors for cardiovascular disease. So I forced them to give me a CT Scan and the test came back with NO evidence of coronary artery disease. I am in Canada so your experiences may be different but I am here to say don’t let a high LDL or total cholesterol fill you with fear!!


Hi Hayley! I just got back from my blood draw and am now reading your response. Thanks for your advice. It reassures me that I did the right thing in holding off on my weekly 24h fast (although I have fasted within the 7 day prior, it has been 5 days). I had to look up the conversion for Canada. I have family history of cardiovascular disease but have been really healthy without risk factors. If the value turns up high, I believe my doctor will be open to monitoring it before jumping to suggest medication. However, if she does suggest it, I plan to push back. Thanks again. I’ll post my results when available.

Siobhan Huggins

Hi Juls,
along with avoiding coffee, MCT and coconut oil, just try to eat as you normally would, and track if you can (just taking pictures of what you eat would be fine). If possible try to time your IF window the day before the blood test so you hit about 12-14 hours fasted when you get the blood draw, and make sure that 12-14 hour fasted is water only (no tea, coffee, etc). Definitely avoid fasting longer than that as it can mess with the results as demonstrated in Dave’s Fasting Disaster post.

Considering your height and weight I wouldn’t be surprised if you fall into the camp of a Lean Mass Hyper-responder, so just be aware beforehand.


Hello Siobhan. Thank you for your response. I did my test earlier today and am still awaiting results. My fast prior to the blood draw was just shy of 14 hours. I have not had coffee and/or MCT oil for the 2 days prior. I don’t know if it matters but other than yoga, I have not done much exercise in the past 2 days due to a sore ankle. I did practice prior to my test today.

Thanks for the warning on the probably lean mass hyper-responder. I’m going to have to look that up. I am guessing that something is up with my tests as my doctor has generally released results by this time and still hasn’t.


P.S. I just watched the presentation on LMHR. I love the simplified explanation. It definitely helps. BTW, results still not made available to me.


Here are my results:
Total Chol: 302 mg/dL
TG: 102
HDL: 88
LDL: 194

Siobhan Huggins

Nice! That’s pretty much what I would have expected as far as results go, as I mentioned before 🙂


Siobhan: In looking at your posts on LMHRs, I was expecting my TGs to be much lower. Prior to keto, they were generally below 50. I only stopped coffee and MCT oil for the 2 days prior. I wonder if that could be a confounding factor still. Also, my morning glucose (fingerstick) that morning was 91. Yet, 3 hours later after a moderate ashtanga yoga practice the serum glucose was higher at 99. Would this be the result of the body ramping up in anticipation of more intense exercise, as I generally do on the weekends and after work?

Siobhan Huggins

Yes, I suspect the coffee was a confounder. There may be a bit of a withdrawal period for coffee, so it’s suggested you quit for at least 2 *weeks* prior to the blood test. Same for coconut/MCT oil (cut these for at least a week).
It is pretty typical for blood glucose to up after exercise, or before if you have a routine, but even without that blood sugar fluctuates throughout the day, or it could have just been variation in the reading from the monitor.


I find the finger sticks surprizingly can vary 10 to 15 mg between 2 sticks close together. Not expected.


Is there such a thing as a lean mass familial hyper cholesterol anyway responder ? Being apoe3/4 I looked in terror at my doctor who told me I might as well eat lard as my risk of heart disease was highly based on my ldl, yet my triglycerides are the lowest they’ve been and how is highest , I’m fit with no inflammation markers and skinny. But my cholesterol goes down (slightly) on keto, not up
Ldl 5
Hdl 1
Trig 1

Jennifer Ozuna

Two days ago. Fasted 12 hours, black coffee only.
Tennis 4-8 hours a week. 39, 5’4, 136 lbs.
keto for 1 year. Lazy Keto, but I am high fat and I don’t cheat bc I get nausea if I taste sugar.

LDL: 202
HDL: 95
TG: 59


Hey Dave and Siobhan,

I am an Apoe4 carrier and a lean mass hyper responder.

As an Apoe4 carrier, with high LDL (Low trigs, high HDL), I have been trying to research what kind of diet would be most protective for my cardiovascular and neurological health.

Dr. Gundry and Dr. Bredesen, who specialize in Alzheimer research and Apoe4 suggest that a ketogenic diet is preferable, but the fat source should be monosaturated as opposed to saturated (animal) fat.

Do you have any insight into the benefits of mono vs saturated fats? Does the science support their recommendations? In your opinion, are Apoe4 carriers better off avoiding butter in favor of olive oil?

Siobhan Huggins

I’ve yet to see any real justification for it, honestly. From what I’ve seen mentioned in passing it partially comes from slightly higher LDL with satfat and apoe4, but as for what that does for CVD risk… I have my doubts that it worsens it, although I can’t say for sure. A lot of it also seems to come from research involving high saturated fat, *and high sugar/carb* – which is obviously not a wise combo, regardless of what genes you have. Yet more from epidemiology which is far too confounded by other variables (processed food? trans fats? high fat high carb?) to be worth much (as far as conclusions go).

From what I *have* seen apoe4 carriers are more predisposed/vulnerable to modern disease but – as Ivor Cummins points out – this would put emphasis on avoiding insulin resistance in general. FWIW Dave is an apoe4 carrier, from what I remember, and he eats plenty of saturated fat.

So, in my opinion, I haven’t found much of anything convincing as far as reasons to avoid animal/saturated fats for apoe4 carriers. That could always change in the future, but my stance is “unconvinced” as of now.


[…] a look this Framingham Offspring study.  As Dave Feldman points out in a recent post, the study found if your triglycerides are less than 100 and your HDL is more than 40 (more than 50 […]

Sheelagh Littlewood
Sheelagh Littlewood

I am uk based but think I fall into LMHR.
Been low carb 3 years, work out 5-6 days a week and am 5ft 10 and 62kg. Waist to height ratio 4.4. Had an insurance medical(wish I hadn’t) and shocked to find cholesterol 360 but T/G 71 and HDLP 104. Worrying level of LDL 239.
Reassured to find your site but now awaiting an appointment at a specialist lipid clinic and think I know what they will say. Stop diet and taken a statin- even though I’m female, 65 and this would be primary prevention!. Would like to be brave enough not to go but have a medical background and years of the old way of thinking difficult to shake off. Will take the Farmington study you highlight with me but would you suggest anything else that might sway them

Siobhan Huggins

Hi Sheelagh,
Apologies for the very late reply on this, we ran into some issues with accidentally comments and were luckily able to recover them albeit a bit late…
It certainly does look like you fall into the camp of a lean mass hyper-responder 🙂
We have a few highlighted studies you might feel will be useful to bring (or just read for yourself), right here:

Top Ten Lipid Related Studies

Especially #10, #3, and #1.

As two of them point out that, in women, the higher the cholesterol the lower the death from all causes. With no real cut off point… which is interesting to note, and raises some interesting questions.
Additionally, another one mentions that in those with genetic causes of high LDL it isn’t the LDL level that determines their risk of heart disease, but rather their level of insulin resistance and fasting insulin.

But the other thing is whether or not you take the medication they prescribe is entirely your personal decision. So, I would really just recommend in general that people do their own studying and decide for yourself if you want to take it or not.


maybe the game starts now ….

Linda Cooper
Linda Cooper

Hi Dave, Can you explain to me why in the last decade my TG’s hover between 0.7 & 0.9 and my HDL between 1.4 & 1.65. And, I have a CAC score of 230…LDL by the way in the same time has been between 3.0 & 3.9

Dolores Yanez Noonan
Dolores Yanez Noonan

Listened to your podcast recently on Ketogeek. My quick story – I am a 52-year-old female runner and started keto in Feb 2018. I was religiously keto for 4 months and as I slowly dropped the baby weight from 4 pregnancies, I allowed a few extra carbs into my diet but have kept it low carb. It is now October and I am down to 137 lbs at 5′ 6.5″ which is almost 20 pounds less than where I started! My goal is to get a little closer to my pre-kid weight but I am very close. I feel great and hope it will help my ultra running goals. But enter the dreaded blood test! I have not had a blood test in a few years so I can only compare to 5 years ago and my numbers are crazy high.

2/12/13 9/28/18
Cholesterol/HDL Ratio <5.0
2.9 3.2
50 mg/dL
54 81
LDL Cholesterol
<130 mg/dL
89 168 H
<150 mg/dL
61 68

The second set is the new numbers. My HDL & triglycerides still look good. Also, my fasting glucose was 84 down from about 90 a few years ago.

The QUESTION – how do I deal with my dr if she wants to put me on meds? I would like to avoid this if possible. I am taking a heaping tablespoon of MCT oil in my coffee each morning so I can stop this. Any other discussion points with my doctor? I see her tomorrow. THANKS!

Siobhan Huggins

Something to keep in mind is that your doctor is working for you as a health consultant. She can suggest what she thinks is best, but it is ultimately up to you whether to follow that recommendation.
What I did when I was in your position was I did my own research – not just from one source, but from many different sources – and came to my own conclusion about whether medication was right for me. Just recently I also brought some of the studies mentioned in this article, as well as this one, to my doctor as a conversation starter.


These are the 2 studies you mentioned in the podcast with PA, but you didn’t get to go into detail on them. They present the data and conclusion exactly as you described it. Frustrating that you couldn’t get there.

Has Peter helped you (since the podcast) to get access to other data that can be analyzed.

Keep up the good work,


Roger Johnson
Roger Johnson

If LFH were to serially test there CAS say yearly and it is not increasong @ oh maybe non, it would be the fastest way to get support fir your theory. Invidence og CV events will not prove it till i die of old age. (74 years of age)


Hi, my name is Deni. I’m a 57 year old woman. I’ve been on the Keto diet since August 2018. I had my blood drawn on 10/12/18. I had been doing intermittant fasting (16:8) for at least a week before my test, though I would have a tbls of butter & MCT oil in my coffee every morning, so not a true fast. I fully expected great results especially because I was feeling amazing, had dropped 13 pounds, and was more active and energetic than I’d been in a long while. My results: LDL 185, HDL 79, Tri 58.
Also, my TSH is 5.19 (My voice definitely has changed)
I meet with my doctor this week and I know she is going to want to put me on drugs but after researching online and reading and listening to your studies as well as so many others, I have a real peace of mind and will not take drugs. I want to continue this course I’m on and give my body more time to adjust to all these changes. I feel too good to go back. I feel better than I did at 30!