Monday, April 27, 2015

New Study Strengthens the Case that LDL Causes Heart Disease

There is little remaining doubt in the scientific/medical community that high levels of LDL, so-called "bad cholesterol", cause heart disease.  Yet in some alternative health circles, the debate continues.  A new study adds substantially to the evidence that LDL plays a causal role in heart disease.


Fat and cholesterol are transported through our blood via carriers called lipoprotein particles.  The primary cholesterol-carrying lipoprotein particle in humans is low-density lipoprotein, or LDL.  LDL delivers cholesterol to the body's tissues, where it's used to regulate cell membrane fluidity, synthesize hormones, and play many other important roles in the body [4/27 note: see comment section for information on LDL's role in reverse cholesterol transport].  Yet, we don't need very much LDL to fulfill this function, and most traditionally-living cultures have lower LDL levels than the modern average.

There is a huge mountain of evidence from multiple lines of investigation that high levels of LDL contribute to the thickening and degeneration of arteries called atherosclerosis.  LDL particles get trapped in the blood vessel lining, depositing cholesterol and activating immune cells called macrophages.  Somewhere along the line, the LDL becomes oxidized.  Macrophages accumulate the oxidized LDL, grow into giant "foam cells", and eventually die, releasing their cargo of cholesterol and fat into the artery wall.  Arterial lesions swell, and the tissue inside them becomes fibrous, calcified, and necrotic.  Blood flow through the artery is restricted.  Sometimes, the lesion bursts, causing a clot that stops blood flow to the heart, brain, or other tissues.  This is a heart attack, stroke, or other embolism.

There are certainly contributors to cardiovascular risk besides LDL particles.  HDL particles, for example, are probably protective*.  Also, cigarette smoking greatly increases the risk of having a heart attack, and this increase in risk is largely independent of effects on blood lipids.  Similarly, the protective effects of exercise are partially independent of blood lipids.  Yet nothing changes the fact that, all things being equal, higher lifetime LDL = higher risk.

In some corners of the alternative health community, the idea persists that LDL is unrelated to cardiovascular risk.  I think this is wrong, and frankly, dangerous.

Statin drugs, as we know, reduce LDL cholesterol and also reduce cardiovascular risk in high-risk groups.  Some people have argued that statins have effects besides LDL lowering that could explain their protective action.  It's not an unreasonable idea.  Yet a new class of cholesterol-lowering drug promises to erase any remaining doubt that lowering LDL protects against heart attack risk.

Enter PCSK9

PCSK9 (proprotein convertase subtilisin/kexin type 9) is an enzyme that regulates levels of the LDL receptor.  The less PCSK9 there is in a liver cell, the more LDL receptor that cell will express, and the more LDL cholesterol it will remove from the blood.  In short, lower PCSK9 = lower LDL.  Researchers have known since 2006 that people with naturally occurring mutations that inactivate PCSK9 have a much lower risk of cardiovascular events (1).  Therefore, PCSK9 was considered an extremely valuable drug target.

Well, a decade has passed, and we now have drugs that inhibit PCSK9.  They are antibodies that, once injected, bind and inactivate the enzyme.  Turns out, these drugs can lower LDL by about half, which beats statins.  Yet PCSK9 inhibitors have not gone completely through the drug development pipeline, so all we have are preliminary studies on whether or not they actually protect against heart attacks and stroke.  A new meta-analysis cobbles together all of the preliminary reports, allowing us to get a better idea of whether or not these drugs actually work.

The study

Eliano Navarese and colleagues collected data from 24 phase II and phase III randomized, placebo-controlled trials of PCSK9 inhibitors (2).  These trials included data from over 10,000 patients.  Navarese and colleagues compiled the data on efficacy and safety, and analyzed it statistically.

The results

PCSK9 inhibitors reduced LDL cholesterol by nearly half (49.5%).  They also reduced heart attacks by more than half (51%), and all-cause mortality by more than half (55%).

PCSK9 inhibitors didn't increase the rate of serious adverse events.


We now have two independent drug therapies, statins and PCSK9 inhibitors, that substantially reduce LDL levels by completely different mechanisms.  Both appear to reduce cardiovascular risk, although we'll need additional randomized, controlled trials to be 100% certain about the efficacy and safety of PCSK9 inhibitors (the trials in this meta-analysis were generally short and weren't designed to measure effects on cardiovascular events or mortality).

The mountain of evidence implicating LDL in cardiovascular disease just got larger.  I hope this will be a wake-up call for people who don't think LDL is important.

* Some people have questioned this, based on the results of the CETP inhibitor trials, which increased HDL cholesterol but didn't reduce cardiovascular risk.  This is not a very convincing argument against the importance of HDL, and here's why: there's no reason to expect that pumping HDL particles full of extra cholesterol would reduce cardiovascular risk.  One of HDL's jobs is to take cholesterol out of the artery wall and bring it to the liver for disposal.  Artificially increasing the cholesterol cargo of the particle, as CETP inhibitors do, isn't going to help HDL take more cholesterol out of the artery-- if anything, it should reduce its capacity to do so.  It's essentially "treating a number", with no regard for the biological mechanism of action of the HDL particle.  We don't yet understand how to increase the ability of the HDL particle to transport cholesterol out of the artery, but if we can enhance that process, we will probably have another effective means of protecting against heart disease.


Tucker Goodrich said...

Causing LDL "causal" is a bit of a stretch. Sort of like saying the barrel of a gun was the cause of death. It's the bullet travelling through the barrel that does the damage.

Of course reducing the number of barrels would reduce the number of bullets...

You've documented well in the past how different environments and diets seem to affect CVD risk. And there's mounting evidence for just such a connection:

"Oxidized LDL Regulates Macrophage Gene Expression through Ligand Activation of PPAR╬│"

"Lowering dietary linoleic acid reduces bioactive oxidized linoleic acid metabolites in humans"

Tiny Almond said...

Do you recommend any foods that lower LDLs in the place of taking synthetic drugs? I've heard that you can significantly impact LDLs by your diet as well.

tremendo said...

This is interesting regarding PCSK9. I notice you mention "LDL Cholesterol" without making any distinctions at any point between number of LDL particles and possibly a high volume of Cholesterol carried in fewer LDL particles. Others have IMO convincingly made this distinction (Peter Attia, Chris Masterjohn, Chris Kresser, even I believe Paul Jaminet). Of course, in general one would expect that a higher volume of Cholesterol "implies" a higher number of LDL particles, but it's frequently not the case among many of those you summarily dismiss as "the alternative health community" (Paleo?) I guess in the end your post presents another reason for anxiety, or another rabbit hole to go down under for many. I know I'll be checking my own sometime soon

David L said...

While it may be true that LDL levels cause heart damage, the effect may be small overall, since heart attacks are not closely related to cholesterol (including LDL) levels:

Newbie said...

You might want to reconsider the statement "LDL delivers cholesterol to the body's tissues."
Here is the view of a lipidologist's a quick 2 paragraph read.
"What is the purpose of LDL cholesterol?"
The author, Seth Baum, did get an answer from the prominent lipidologist, Thomas Dayspring, in the comments section - "The vast majority of LDLs are the lipolyitc by product of a VLDL and/or IDL that has niot been internalized by LDL receptors and thus achieved an LDL size or buoyancy by losing core TG and surface phospholipids. Normally about 60-80% of IDLs (product of VLDL catabolism) are cleared by the liver - maybe 20-40% are not and become LDLs. Their purpose with a 2-3 day half life is to garner additonal cholesterol from HDLs (via CETP and then return that to the liver or small intestine in a process called indirect reverse cholesterol transport. LDLs do the majority of RCT. Their purpose for the most part is not to deliver cholesterol to tissues."

I am glad that you added your voice to the benefits of lowering LDL-P.

ivor cummins said...

Hi Stephan

Do you not think that LDL only partakes in the atherosclerotic milieu - if you have an inflammatory issue, then indeed driving LDL down will mitigate it somewhat; especially if the driving agent has anti-inflammatory effects? Thank you for listening, and here is a bauble for your amusement:

patentsumu said...

Interesting post.
I'm at risk for heart disease based on family history.
Any recommendations on blogs/people to follow?

spfldo said...

There is no proof to show statin drugs are beneficial to anyone other than those who have had both, an early onset heart attack and familial hypercholesterolemia and any benefit is very marginal. I have both and when I was on statins I had every side effect on the bottle, severe cramps, muscle and joint pain, amnesia, severe fatigue and peripheral neuropathy. Statins are dangerous and essentially a useless waste of money. If LDL i.e., cholesterol is so dangerous, why were there practically no heart attacks a hundred years ago? I think there is something else at play here e.g., diet. Hopefully this new drug treatment will be more effective and safe.

brec said...

Typo alert: "Stain [sic] drugs"

Peaceful Pony said...

Finally, some cholesterol clarity. Thanks Stephan.

Stephan Guyenet said...

Hi Tremendo,

In general, in my article I referred to the LDL particle rather than LDL cholesterol. There is some indication that LDL particle number is a better indicator of risk than LDL cholesterol, but both are meaningful.

Hi Newbie,

Good thoughts. I didn't know LDL is involved in reverse cholesterol transport. I have read elsewhere that cells in many tissues supplement their own internal cholesterol synthesis by expressing the LDL receptor and endocytosing LDL. Dr. Daysrping didn't provide references to support the idea that LDL is not a significant source of cholesterol for tissues. He may be right but I'd like to see the evidence.

Hi Ivor,

LDL and its downstream effects (e.g., cholesterol crystals, foam cells) appear to be a primary driver of inflammation in the artery. I don't think LDL and inflammation are two separate processes, although inflammation can be driven by things other than LDL of course, such as the irritants in cigarette smoke.

Hi spfldo,

A lot of people including myself would disagree with your statement that statins only reduce risk in people who have both FH and a previous MI. There is a wealth of evidence supporting its efficacy for secondary prevention in people without FH, and some for primary prevention.

I won't dispute the effects you've experienced, but I will note that the rate of adverse effects for statins in RCTs is usually very similar to placebo.

Hi brec,

Thanks-- fixed.

Stephan Guyenet said...

Hi Ivor,

The Reaven article discussed in the link you passed along is very interesting. Thanks for sharing.

Metanoia said...

This is a good starting point for dietary strategies:

julianne said...

In my experience many on statins are unaware that issues they are experiencing are side effects of statins. As a nutritionist I do a full health questionnaire covering as many issues as possible - I have clients who say things like "lately I've gotten really forgetful" " I had to stop going to the gym because my muscles were sore and tired" "My libido has bombed, I'm really worried" "I used to be able to do 10 pullups - now I'm lucky to get out 3"
I ask questions like - when did this start? I have usually already covered what meds they are on and for how long.
The client has put their deterioration down to age. When I look at the facts - it starts soon after statin drugs. I point this out - and yes they agree. And when they go off them - or reduce the dose (I ask them to talk to their doctor first) they find the issues go away.
I think this is a bigger problem than most realise

Unknown said...

Your ref #2 link does not appear to work for me. Perhaps this is the right one?

JanKnitz said...

You footnote link did not work. Are the benefits reported for PCSK9 for primary or secondary prevention? What was the ratio of males to females in the study population? Are the benefits reported in terms of relative or actual risk? And how were ldl numbers measured/calculated?

What is the role of triglycerides in predicting cardiovascular risk?

RLL said...

Here is why many of us ignore recommendations on lowering cholesterol.

Patients do not have easy access to good measuring of LDL. The standard tests may be useless. Particle size of LDL does seem to matter.

Statins benefits are somewhat controversial. Doctors for the most part refuse to consider that a patient may be getting bad side effects.

If you control blood sugars, blood pressure, exercise, have higher HDL, and triglycerides are OK/so-so you likely are at low risk for serious heart disease already.

Older people (me) do not tolerate larger doses of statins. And I am not about to trust any 'newest and latest' drug until it has been out for 5-10 years as a preventative measure.

Statins seem to have a number needed to treat to benefit one person somewhere between 40 - 150. There are reputable people who think any NNT under 5 may be dubious.

I lost 2 years of superbly good health, likely from statin overtreatment. Doc put me on the highest dosage of statins because my triglycerides were 100. BAD call. Cholesterol was 180.

Stephan Guyenet said...

Hi Unknown and Jan,

Fixed the link. I had used the link provided to me by Ann Int Med that was supposed to activate after the embargo lifted. Annoying.

Stephan Guyenet said...

Hi Tremendo,

I'm not picking on Paleo specifically. Please note that many (all?) of the Paleo founders, such as Eaton and Cordain, acknowledge the role of LDL in cardiovascular disease. It was only when Paleo was partially co-opted by ideological low-carbers that this began to change. Most low-carbers are perfectly rational, but the intersection of low-carb and Paleo does seem to attract a disproportionate number of people who have an imaginative way of viewing scientific evidence.

Newbie said...

Stephen, your response to Tremendo is like a breath of fresh air - you hit the nail on the head! The morphing of the Paleo community has not been positive in my mind, and I am an original Paleo enthusiast.
Thank you.

@NaturalWorkwear said...

Amazing. People are truly impervious to facts.

Thanks, Stephen, for highlighting the new data on PCSK9. These data are, as you said, more boulders in the mountain of evidence relating LDL to cardiovascular risk.

ivor cummins said...

I'm afraid LDL is primarily dysfunctional via IR, and even LDLp is only an associational factor - again intimately linked to IR. Reavan, Despres and others were correct. Insulin drives the bus, but there are many passengers along for the ride (Despres slide at bottom of this post: also, Reaven's brilliant and properly-measured IR prospective - the maths speaks:

JanKnitz said...

Since I can't get past the paywall, my questions still stand:

Are the benefits reported for PCSK9 for primary or secondary prevention? What was the ratio of males to females in the study population? Are the benefits reported in terms of relative or actual risk? And how were LDL numbers measured/calculated?

Also, what is your opinion of triglycerides as a predictor of CV risk?

George Henderson said...

The idea that LDL is unlikely to be a primary driver of CHD isn't specific to the alternative health or diet community.
Bill Stehbens was an experienced pathologist employed at a high level to teach and investigate the causes of CHD in New Zealand and he expressed his doubts as someone who knew the pathology well (which few of us do).

In the statin trials, people who are in the upper quartile for HDL get no benefit from the drug, indeed the placebo treated, high-HDL subgroup do better than any treated group, and this is true for another lipid-lowering drug gemfibrozil when low TG is also factored in.

Some people, I think in children it is near half, do not respond to drugs or dietary change with a significantly lower LDL; statin trials like to separate responders because they have better results, but of course what is the use of the LDL theory to the perhaps large part of the population who will not benefit from chasing it, but can still benefit from fixing HDL, TG, HbA1c, BP, BMI, particle size and so on?

Why do so many people have an excess of PCSK9? Is there an evolutionary reason? PCSK9 inhibits the hepatitis C virus and other viruses that infect cells through the LDL receptor. Ebola is said to be an LDL receptor virus. LDL endocytosis is a method by which large particles are allowed into cells, and may well have been exploited by other pathogens in the past.
In an ancestral milieu, CHD is perhaps not common. When there is little risk of CHD, the benefits of high LDL might outweigh any risks. One of these benefits may be resistance to sepsis; LDL particles can bind endotoxin and older people with high LDL are less likely to die from infectious disease.
So there are perhaps good reasons why so many people have inherited genes that keep LDL at levels that are now supposed to be dangerous.

But LDL is perhaps only dangerous in the sense that atmospheric oxygen is dangerous if your house is on fire. Firemen restrict oxygen when fighting a fire, but no-one says to prevent fires by restricting it. Not even a true analogy, because oxygen is absolutely essential for fire, and high LDL is by no means essential for CHD. Nor is saturated fat, which famously elevates LDL, associated with CHD, despite health conscious people long trying to limit it (which really ought to have created a solid statistical association).
In any case, I think the Boss, Bruce Springsteen, had his facts straight when he wrote, in his 1984 hit song Dancing in the Dark, "you can't start a fire/ you can't start a fire without a spark."

valerie said...

What happens to total mortality in people who have the weird LDL mutation?

What about total mortality for those who take the new drug?

After all, we can come up with scenarios where the reduction in the rate of CVD is only a statistical artifact left from a very deadly side-effect, or something like that

William Gillean said...

It is my understanding that LDL is an aggregation of similar particles of differing sizes.

What is the evidence that all LDL is equally atherogenic?

Is not the evidence stronger for the small dense particles?

Thank you very much for your efforts.

Nawp Dolittle said...

Stephan, I believe it is well established that dietary saturated fat increases LDL cholesterol, but recent large studies/meta studies have not found a correlation between dietary saturated fat and CVD. If "LDL causes heart disease", how can this be?

Zahc said...

You might as well call this review the ODYSSEY analysis, since the results depend on the ODYSSEY trials (mostly the ODYSSEY long-term) done by same group. This is illustrated by the so called reduction in heart attacks, which depends on that one trial (ODYSSEY long-term), and if removed, the results would actually tend to make the inhibitors look worse than placebo. For some reason the recent study on Evolocumab was not included, but it found no difference between the groups in the 14 heart attacks that occurred. So in my view, your claim that "They also reduced heart attacks by more than half" is very misleading. Nevertheless, we are talking about very limited post-hoc data here, which is unconvincing either way, and certainly does not "strengthen" anything.

Erik Arnesen said...

Hi. You mention CETP inhibitors, but another argument against increasing HDL-C for lowering risk comes from Mendelian randomization studies, which show that genetically, lifelong increased HDL is not associated with CVD risk. Here's one frequently cited paper on 14 allels related to HDL:

In contrast Mendelian randomization studies strongly suggests that LDL is indeks causal, including the above study. A recent paper by Brian Ference concludes that "the clinical benefit of lower LDL-C levels may be more closely related to the magnitude of reduction than to the mechanism of LDL-C lowering."

Richard said...

Nawp and Zahc: I agree. "Causation" here is a very tricky concept. Nawp's example is a good one, as it proposes to falsify the findings. Similarly, with Zahc, these "results" do not prove any specific thing, really, except that people who really want to "prove" something will find a way.

For myself, my belief is that CHD and increased blood pressure is caused by excess calcium in the soft tissues, in this case in the lining of the arteries, which results in (frantic and somewhat futile) efforts by the body to repair or cover over the damage. Cholesterol and various mechanisms for moving it about increase with that damage and such efforts by the body. So the cholesterol is there for a reason. Treating the numbers cannot work.

Is this LDL information and knowledge perhaps, in some way, useful? Possibly, but not as "proof" as it is used here.

psychic24 said...

Hi Stephan,

Im slightly confused by your clarification for Tremendo. In your article you discuss inhibition of pcsk9 to reduce LDL particles, but the meta analysis (or at least your excerpt of it) notes a reduction in LDL cholesterol, not necessarily particle number. Reducing particle number would be in line with the logic you prefaced the study with, but reduction in cholesterol on board the particle not so much.

Newbie said...

Stephen - typo I think ..."Studies in animals suggest that HDL efflux from arterial macrophages to HDL plays a strong causal role in atherosclerosis. "
I think you meant to type "cholesterol" efflux....just thought you might want to tidy it up.... unless you meant to say something else altogether.

Stephan Guyenet said...

Hi Newbie,

Yup-- thanks.

Stephan Guyenet said...

Hi Erik,

Yes I'm familiar with the Mendelian randomization evidence for HDL. It suffers from the same problem as the CETP inhibitor evidence: it's based on HDL cholesterol, not HDL function. HDL cholesterol is poorly correlated with HDL function.

Human observational studies have shown that HDL efflux capacity, or the ability of HDL to remove cholesterol from cholesterol-laden macrophages, is a better predictor of CVD risk than HDL cholesterol.

Studies in animals suggest that cholesterol efflux from arterial macrophages to HDL plays a strong causal role in atherosclerosis.

Tangier disease is caused by a loss-of-funcion mutation in ABCA1, a key protein that mediates cholesterol export from macrophages to HDL. These people accumulate cholesterol in tissues and develop premature atherosclerosis. There is still a strong case to be made for a causal role of HDL in CVD.

I acknowledge that some uncertainty remains, however the evidence overall still points to a causal role IMO.

Stephan Guyenet said...

Hi psychic,

Good thought. In this case, LDL cholesterol reduction should be well correlated with particle number reduction, and here's why: PCSK9 inhibition acts by increasing expression of the LDL receptor. The LDL receptor participates in gobbling up whole LDL particles from the bloodstream into cells. So if LDL cholesterol is going down due to an increase in LDL receptor expression, it is doing so because the number of LDL particles in circulation is decreasing.

psychic24 said...

Thanks for the response, Stephan, but I'm still slightly confused. Why would the residual particles still in the blood house less cholesterol just because there are less of them? I'm not very familiar with how LDL cholesterol is measured so I may just be missing something (e.g. if the measurement is based from a set volume of cholesterol in the blood, and not specifically on individual particles)

Stephan Guyenet said...

Hi psychic,

Typical LDL-cholesterol values are estimates of the total quantity of cholesterol carried by LDL particles in a given volume of blood. If we reduce the number of particles by 50% without affecting how much cholesterol each particle carries, that leads to a 50% lower LDL cholesterol value. Make sense?

George Henderson said...

Looking at some of the individual PCSK9 trials, the abstracts say that other lipids, not just LDL, decreased.
How this happens I'm not sure, but it does raise the possibility that the LDL subfraction effects of the drugs are not evenly spread, or that there are effects on post-prandial VLDL etc.

@Erik Arnesen, there are subclasses of HDL, and some of these are useless.
Excess alcohol increases the useless types, as does hyperinsulinaemia.
And no doubt drugs too.
If there are genetic high-HDL cases with no benefit it is likely these other HDLs are involved. Also, there is a likely to be a plateau beyond which even more "good" HDL can't produce much benefit, at the upper end of the normal range.
There are dietary components that enhance the functionality of HDL - CLA from dairy fat is one, another is olive oil polyphenols.

Jerry said...

Berberine does the same thing and is available now over the counter. Over a 2 year span, I took it and dropped total cholesterol from 230 to 178, LDL dropped from 150 and is now 102.
Per "Berberine appears to suppress PCSK9 at the promoter level secondary to suppressing the activity of its cofactor, HNF1╬▒.[236]"

Jerry said...

Also wanted to add, I tried statins and felt like crap. I feel great taking Berberine. It is literally a wonder, compares favorably with Metformin as well. No, I don't sell it, either.

psychic24 said...

Thanks for the clarification, Stephan; Crystal clear now =)

Pauline said...

Stephen, you said in your write up "the trials in this meta-analysis were generally short and weren't designed to measure effects on cardiovascular events or mortality." And yet that's exactly what they are being used to show here. I looked up one study at random and found that the results were being written up before the planned end of the study period, which is one of the tricks trials can employ - notice that the data look particularly good at a certain point in time and write them up, when in fact it's just part of a statistic fluctuation that would be evened out over time. I am skeptical that these results are truly representative of long-term use of these drugs - I have read far too much about the "cholesterol con". I guess we will find out eventually - though I hope we get the right answer sooner rather than later.

Nawp Dolittle said...

I think Richard misunderstood me. My question really was a question. Of the following claims, both of which can be supported with substantial research, only one can be correct:

1. Dietary saturated fat elevates LDL which causes CVD.

2. There is no correlation between dietary saturated fat and CVD.

One of those is wrong. Which one?

Paul Darrigo said...

I have Plaque as well.

I have spent years on copious diets and supplements, exercise, blood tests etc. Once, I reversed to a lower number on my Ultra Fast CT score which I attribute to eating mostly salmon, vegetables etc.

I am now on 10 mg of a statin, eat only fish, removed food allergens, wheat, peanuts and soy.

I feel better just from that. No constipation and sleepiness. Allergens effect IGG and others to a level that might not be outlandish but do cause inflammation.

If you are on a statin you have to take CoQ10 - That is why the muscle pain. Also you need good fats for reverse transport and brain health as well as a good grade B vitamin.

I have also added a lot of resveratrol - heck I take ton of stuff but all geared for this purpose.

My scan is time soon, so it will be interesting to see how I am doing.

Paul Darrigo said...


How much Berberine do you take?

Do you take blood levels of the usual markers?

Colin Macdonald said...

I'm curious about the original statement that most traditionally living cultures have low cholesterol.

I'm interested where this data is. This post here questions it's validity ( thoughts?

If LDL cholesterol isn't caused by saturated fat. What's the difference between traditional living cultures and urban dwellers that causes this difference that leads directly to increasing LDL and heart disease?

I'm having a hard time fitting this post into the context of your other writing.

marie curieous said...

The effect of PCSK9 seems well substantiated, thank you for explaining the evidence. I have a more general question though, regarding the start of the atherosclerosis process, and if you have any thoughts on it that you could share, that woykd be appreciated. You mention that at the start, LDL particles get trapped in the vessel walls - how/why do they do so? Are there injuries to the cells of the vessel wall and if so what causes such injury? Or is it just a stochastic process where the more partciles there are, the higher the chance some will get trapped? I have trouble visualizing the latter because those are some very smooth vessel walls normally, or so I recall from my, admittedly dated now, graduate studies :-)

Dan said...

In science, truth is gained through replication. All that we have in this area are studies funded by pharmaceutical companies that have never been independently replicated.
In addition, the best study we have, that is rather independent, is the Framingham Heart Study. Well, what did that study tell us: 1. Over 50 percent of people having a cardiovascular event had LDL in the optimal ranges, 2. the best predictor of heart disease was triglyceride levels and HDL levels (LDL was not a good predictor at all).
Then we have the French paradox. A group of people with high cholesterol and LDL levels with low levels of heart disease. I am not going to debate why but only to demonstrate object information about a group of people and their rates of heart disease.
Then we have the fact that oxidized LDL occurs when glycemic load increase. That is, to create large fluffy pattern A LDL you lower your glycemic intake. To lower triglycerides you lower your glycemic intake. To increase HDL you lower your glycemic intake. Do you see a pattern? Oh yeah, to lower HS-CRP, you might want to try lowering your glycemic intake.
In addition, even the best run pharmaceutical funded studies on statins show that 100 people have to be treated to maybe reduce 1 non-fatal heart attack. None of these studies are a slam dunk proof of anything.
In addition, and no one can deny this, is the fact that statins are being given out for “preventive care.” As such, all studies done on them are short, no longer than three to four years. Yet, statins are given to people to take for decades. There are no long term studies demonstrating the efficacy and safety of these drugs for such a long period of time.
In fact, what studies do show is that statins, when the outcome is reduction of all cases of mortality, do nothing to extend a person’s life.
I would also add, when doing a study, the longer it goes, the harder it is to keep a hold of the benefit you are studying. That is, these pharmaceutical companies know to end these studies because the effect they might have demonstrated, all but low, would disappear over time and might actually demonstrate harm when statins are taken for long periods of time.
Ask yourself, if statins are such wonderful drugs, why would a pharmaceutical company end the study instead of keeping it going, if the outcomes were so promising? A study cost nothing in the grand scheme of things as far as the billions they make off these drugs. There is a clear reason why they end these studies early.

Suzee said...

There may be a role for menaquinones in prevention of arteriosclerosis. Menaquinones are needed for the carboxylation of Matrix-Gla-proteins which are responsible for inhibiting calcification in arteries. Our normal source of menaquinones is fermented cheeses, milk and some meat. Alteration of meat and milk production via grain feeding rather than grass may be the cause of many cases of arteriosclerosis and bone loss. It may be more important to prevent the calcification than to change the particles that are clustering. There are many new studies looking at the role of K2 in CVD and other illnesses.

Suzee said...

I tend to be leery of statins because it seems that bile has a role in transporting excess dietary cholesterol out of the body. Since bile is synthesized from cholesterol, reducing it would seem to reduce our ability to remove dietary cholesterol from our body. It often seems as if the elements necessary for digestion and processing of nutrients are included in the foods we eat (if they are not processed).

Jerry said...

Responding to Paul Darrigo, about my other markers & dosage of Berberine:

I've been taking 500mg 3 times daily with meals. For the past two years, My HDL has been steady at 60, triglycerides stay right at 149. The only change has been the LDL going down. That and I lost 15 pounds, about 7% total bodyweight. I know losing weight can lower cholesterol, but I don't think it would be to the extent it has. (LDL dropping 50 points)

Richard Nikoley said...

Well well.

I love the pure exercise of iconoclasm, of challenging beliefs that reach bromide and slogan proportions. Stephan, you've always been good at that, and the fact that you are now saying "wait a minute" with respect to those things you were posting about way back is testament to your underlying honesty. This is independent of being right or wrong. Honesty finds a way.

Personally, I suspect it's rather a U or J curve thing, where most people are in reasonable ranges. Probably, the standard industrial health BUSINESS wants the range of "probably OK," to be as narrow as possible. It's unrealistic to hold them to higher standards than auto-manufacturers and marketers. It's their business to sell drugs. OTOH, just as in cars, you really have to benefit the consumer long term (economics is all the same, whether cars or pills).

I also suspect that modern lifestyles and food engineering serve to either push people out of normal C ranges they would otherwise have, or, exacerbate somewhat out-of-normal that in an HG setting would not be any problem. Chronic widespread inflammation is--to my opinion--probably the most underlying factor and C exploits this. So what if treating C is treating a downstream symptom. Would people complain about a drug with low side effect that reduced MI 50% independent of other factors?

My HDL remained near 100 and my LDL went way down by focussing on feeding my gut, primarily beans.

Finally, it's really difficult for me to see what Stephan politely suggested in terms of the marriage of LC and Paleo in terms of super high C in so many as anything more that irrational denial and justification.

Keep on clashing with icons, Stephan.

Richard said...

I may have used your post as the beginning of a point I wanted to make, but your question does not strike me as a real question. There is no doubt that dietary saturated fat consumption is a minimal factor in increased "cholesterol " levels. But there is little evidence that increased LDL actually "causes" anything like heart or arterial problems. Now it may be that LDL increases correlate with such issues, which raises some issues regarding the cause for the increase, but it does not necessarily show anything like "cause."

As Suzee points out, the reality is that LDL increases may be the body's reaction to arterial damage, not a cause in itself. Statins may "work" because they might function as an anti-inflammatory, but they also seem to have some undesirable effects. But why take a drug (or medicine?) that does not actually do what it is advertised to do? Calcium in the wrong place (in soft tissue) is a possible cause of arterial damage, and is not addressed by statins.

Better to eat some kimchee or natto or cheese on a regular basis and avoid statins. As if they do any good.

Tim said...

50 comments and counting. No surprise, really. Somehow on the blogosphere everyone whose cousin's friend is on statins and who has understood the difference between LDL and HDL seems to consider himself an expert on cholesterol, knowing about the "real causes" of atherosclerosis all medical experts have missed for decades (because they are only fostering their dogmas without commiting any empirical research). Well, sure!

Is the oxidative state of the LDL particle the key factor to atherosclerosis? Very unlikely, otherwise the vitamin E trials wouldn't have failed so badly. Does a pro-inflammatory milieu promote the progression of atherosklerosis? You bet! But LDL tearing though the artery wall remains the initiating factor, and the less LDL the less atherosclerotic lesions. Thank you, Stephan, for once again confronting people with the actual scientific evidence.

Colin Macdonald said...

Tim, yes everyone has an opinion.

Though a lot of us formed these opinions with help from previous posts on this very blog so I'm sure you can appreciate the confusion.

E. Hosewater said...

Just a few random comments.

I recently saw a video put out by Ray Kurzweil (narrated by Edward James Olmos) that made the claim that it really is oxidized LDL that causes atherosclerosis. LDL particles normally get into the artery wall and macrophages clean them out. But the oxidized particles "kill" the macrophages which leads to the formation of foam cells and starts the process. Too many LDL particles circulating for too long tends to make more of the oxidized. The video didn't cite any references as far as I could find.

Also, some have speculated that one of the ways statins work is by increasing the number of LDL receptors in the live, much like PCSK9. The body notices that there is less cholesterol being produced so it increases production of some protein that promotes LDL receptor creation. I think researchers should try to figure out if something is nuking our LDL receptors, something that was not around until the early 20th century. Seed oils, vaccines, electricity, sugar, whatever.

And, as Dan said a few posts up, blood sugar seems to be a better predictor of risk than LDL (C or P).

Sorry no links or references for any of this, just recalling what I have seen recently.

jewiuqas said...

Sorry for interrupting the sophisticated scientific exchange of views. I am just a down-to-earth guy somewhat at a loss as to which school of nutrition to adhere to with the view to escape the lamentable fate of many of my ancestors of both sexes who died in their early sixties, or even in their fifties of various diseases termed “of civilization”. So, does anyone have a recommendation as to what I should exclude from my diet in the light of the subject discussed in this post? I never cared a fig for cholesterol because someone knowledgeable said once that it was one of the biggest mistakes of the 20th century to accuse dietary cholesterol with causing cardiovascular issues. LDL and HDL, discussed in this post, are not the same as cholesterol, as much as my understanding goes. So, should I avoid eating cholesterol or LDL, or both? If so what foods contain them?

Tim said...

@jewiuqas: For authorative, science-based and largely ideology free information on healthy eating - including CVD prevention - you should consult the Harvard Nutrition Source:

Regarding dietary cholesterol: unless you carry some rare genetic polymorphism, dietary cholesterol won't have much effect on your cholesterol levels. What is more likely to raise your LDL cholesterol is saturated fat and refined carbohydrates.

jewiuqas said...

As to refined carbs, I got no worries because I have almost totally excluded them from my diet apart from some white rice flour that I use as thickener in shepherds pie and suchlike dishes. But I use very little, so that a 250 g pack lasts for 6 months.
So, the Richard Price devotees are proved to be wrong with their saturated fat rich regime ? But saturated fats are more stable, it is the PUFA-s that get oxidized and turned into trans fats, is it not the case?

Richard Sanchez said...


I would recommend a balanced style of eating with plenty of leafy greens, berries, citrus fruit and supplementation with chia seeds and fatty fish or fish oil. This is because your cells need these fats in small amounts. A top researcher who studies the endothelium told me this. Stephan has a quality article here exposing hucksters.Native newly formed LDL does not appear atherogenic, but the endothelial cells themselves have the ability to modify LDL. Oxidized and altered LDL lipoprotein is a problem and the greater the levels of LDL the more risk and potential there is for modification. Of course, many other things matter-blood sugar, elevated blood pressure.

Richard Nikoley said...

Richard S:

Where does all tis "leafy greens" stuff come from? It's basically rabbit food and they're coprophagic.

For instance, consider the Ugandan diet. Lots of various starches, and meat. No leafy greens in sight.

Will Kriski (Potato Strong) said...

Yep it's a no brainer. I lowered my cholesterol to 112 mg/dL on a starch based diet. Below 150 is considered heart attack proof woohoo! You can get omega 3 from flax and fats from nuts, seeds, avocados without the toxins that reside in the fats of fish, etc.

Amazing people think if you eat cholesterol it doesn't affect your cholesterol but they make exceptions for iron, protein and calcium from meat and dairy that they eat. You are what you eat, it's common sense. Our biology points to us being herbivores from teeth, saliva, stomach acidity, intestines, jaw, etc.

Richard Sanchez said...

Richard Nikoley:

They are anti-inflammatory, and extremely nutrient dense. They are a great addition to anybody's diet likely. Far too little is known about nutrition for the Blogosphere's certitude and dietary tribes.

We want as much VARIETY in our diets as possible until we know more- much more. Nutrition is not rocket science, it is FAR worse. Far more complicated.....

You might know who I am. We have met before I believe. ;)

Taylor said...

It doesn't really seem that there is a scientific consensus about what the precipitating factor of an atherosclerotic lesion is :

The risk associated with high LDL-p would exist if LDL contributed only to the progression of the disease or only to the initiation of the disease or to both.

I must admit that I find the idea that LDL particles start to tear through the arterial wall when the concentration of them increases by 50% or so to be somewhat dubious.

ivor cummins said...


You are bang on - the 'LDLp as main cause of atherosclerosis' is a lucrative farce...


Richard Sanchez said...

I forgot to mention dark leafy greens specifically lower the risk of pancreatic cancer, even more so than yellow vegetables. This is something we do not want to get. They are not the only thing I eat. I only recommend adequate to generous helpings in the context of an overall balanced diet.

Dark leafy greens are an excellent addition to any diet- even one that does pretty well without them. Variety is essential. I am neither low carb, nor high carb. I eat as much variety as possible, until we know more. And we have MUCH to learn about nutrition. This may well be the most complicated area in biology or one of them.

Joe Doro said...

From the study you quote:

Limitations: Results were derived from study-level data rather than patient-level data, and clinical outcome data are rare.

And from one of the studies included in this meta analysis (Efficacy and Safety of Alirocumab in Reducing Lipids and Cardiovascular Events) NEJM.

"In a post hoc analysis, the rate of major adverse cardiovascular events (death from coronary heart disease, nonfatal myocardial infarction, fatal or nonfatal ischemic stroke, or unstable angina requiring hospitalization) was lower with alirocumab than with placebo (1.7% vs. 3.3%; hazard ratio, 0.52; 95% confidence interval, 0.31 to 0.90; nominal P=0.02)" Absolute risk reduction 1.6%

And from one of the others while showing a decreased overall mortality, did not show a statisitically significant decrease in cardiovascular deaths.

Once again,we are being bombarded with relative risks, odds ratios, etc. all of which always sound pretty dramatic. I mean 50% sounds so much better than 1.6 % But when the TV ads come out how many will quote the absolute risk or for that matter, the NNT (number needed to treat), a measure that most clinicians find much more useful.

Remember when Jarvik was pitching Lipitor while rowing in a single scull? BTW he wasn't a rower, they had a stand in for him. Anyway, the TV ads shouted a 33% reduction of having a vascular event if you were high risk for a vascular event (primary prevention). Yet if you looked at the package insert and saw the data you found that over a 3 year period the control group had 3 events and the treated group had 2. But through the magic of the relative risk reduction calculation an absolute 1 % risk reduction becomes 33%.

So let's look at the NNT

Assuuming that the annual absolute risk reduction is near 1.5 % for these new drugs then the NNT will be about 67 so that means at say $10K a patient we as a society will spend about $670,000 a year to prevent 1 vascular event. To put that in perspective, that's my family's health care premium for nearly 56 years.

And finally, I wish I could have back the money I have invested over the years in companies that had very promising phase 2 trials only to the have the phase 3 trials fail.

No yonder the FDA will probably not give these drugs approval until the clinical data becomes available or at least will have very stricit indications. Remeber the debacle with the statins? In order to get a new one approved all you needed to show was that it lowered cholesterol and was safe. You never had to show they actually reduced clinical events. The LDL itself became the endpoint.

jewiuqas said...

In my comment above I meant Weston Price not Richard. Thank you for your well meaning replies, but my confusion still persists as to cholesterol. It is but a poor consolation that, as much as I can see from the posts, I am not alone. My question remains the same: What is the practical message that this article boils down to? Someone recommends that saturated fat intake should be kept low. An explanation, please. Someone else says ingested cholesterol matters after all. I am still somewhat skeptical about this later opinion, because shrimps (containing lots of cholesterol, don’t ask me if of the good or the bad type) for example have been somewhat of a staple for certain South Sea tribes practically free from western diseases.

Colin Macdonald said...

Thanks for digging that up Joe Doro. When I see a "50% reduction" in risk, alarm bells tend to go off in my head. I wonder, where would those numbers would sit if there wasn't a strong bias to only publish positive results?

I suppose policy makers perhaps excuse this kind of math by looking at entire populations of a country, thinking that if heart attack rates go down 1.5% they're saving countless thousands of people! Without considering the hundreds of thousands that would be spending fortunes without any statistical chance of benefit while risking significant side effects.

It's irresponsible to completely demonize pharmaceutical companies for this kind of shotgun approach to treatment, it can certainly cause one to loose faith in what their real interests are.

Tim said...

Of all "arguments" one can contrive against the significance of some study's results, complaining about the well-established scientific standard of relative risk ratios is certainly the most inept and misguided one.

In fact, anyone coming up with such a "last resort" argument against evidence that doesn't fit his preconceptions only reveals his own scientific illiteracy.

In the case of Joe Doro's comment above it is not only misguided but outright fraudulent, as he deceptively ommited the context in which the study was conducted by selectively citing parts of the abstract.

This is was the methods paragraph states:

"We conducted a randomized trial involving 2341 patients at high risk for cardiovascular events who had LDL cholesterol levels of 70 mg per deciliter (1.8 mmol per liter) or more and were receiving treatment with statins at the maximum tolerated dose (the highest dose associated with an acceptable side-effect profile), with or without other lipid-lowering therapy."

The 48% risk reduction was seen in patiens already on high-dose statin drugs which were ADDITIONALLY given alirucumab in order to get their LDL below 70 mg/dl, compared to those ONLY taking statins and a placebo (any ony trial design would have been unethical anyway). The absolute incidence in CVD events is of course expected to be low in such a collective - the more remarkable it is that alirucumab further cut the risk in half.

Of course developing a new drug such as alirucumab is immensely expensive and this cost (plus a generous profit margin for the patent holder of course) is reflected in the actual cost of the medication. However, this is the case with any new drug on the market. Prices go down massively - basically to the mere production cost - as soon as the patent on a drug expires and generics become available.

This is why economical objections against a new drug based on its initial market price don't make any sense with regard to the drug's value in the long term.

Colin Macdonald said...

Tim, I don't think the point is that relative risk isn't scientific. But very fair criticism can be made against relative risk as a way of presenting results. Absolute risk and NNT are also scientifically robust in my understanding, and I feel they are more useful in clearly communicating risk reduction. I think saying that criticizing relative risk is 'inept, misguided and fraudulent' is a deeply inflammatory remark that serves no purpose but to elevate the level of tension in a very controversial subject.

On a different note, I'm curious how ezetimibe and cholestyramine factor into the discussion on the direct effect of LDL levels and heart disease. From what I can gather, they both are effective at lowering cholesterol levels, yet both are ineffective are reducing heart disease. If LDL was a direct cause, wouldn't these drugs be effective?