Yet another study showing (very) high LDL does not mean atherosclerosis


(Bob M) #1

A study where they looked at people with high to very high LDL…yet many of them had CAC (coronary arterial calcification) scores of zero or <100. A score of zero means low risk, low to no atherosclerosis.

One quote: “In patients with severe hypercholesterolemia, 45% had CAC = 0, which was associated with a significantly lower ASCVD risk.”

As a scientist, I think that if having high LDL causes atherosclerosis, it should cause it every single time in everyone of the same age. There should be no one having a CAC score of zero. Yet there are. Plenty of them.

To me, this means something OTHER THAN LDL causes atherosclerosis.


(Bacon is a many-splendoured thing) #2

From the abstract, it looks like a fascinating study. One line in particular drew my eye:

A higher CAC category was associated with increased statin use after CAC scoring (p < 0.001).

Too bad the American Journal of Cardiology is an Elsevier publication; I’d love to read the full study.


(Bacon is a many-splendoured thing) #3

Well, you’d certainly think so. And the neurobiologist David Diamond would agree with you. And so would the researchers involved with the International Network of Cholesterol Sceptics (THINCS).

But I recently read a paper arguing that because low to normal LDL is associated with over 50% of first heart attacks, that simply means that our idea of “normal” LDL is far too high, and everyone should be taking a statin.


(Bob M) #4

I know, it sucks that you can’t read the whole thing. It’s hard to know about the higher statin use with higher CAC, other than perhaps the CAC was taken because there was evidence of atherosclerosis? I don’t know why CACs are being taken, because no one seems to know of their existence.

Yeah, David Diamond doesn’t think that LDL is a risk factor…period.

I think the idea of all LDLs are too high and should be zero is something that would have to be rigorously tested, because LDL has many functions. It is an energy carrier, for instance (at least for us keto folks). It also serves an immune function, which is probably why older folks with higher LDL tend to live longer.

Unfortunately, the only way to lower LDL to low levels is with drugs, probably a combination of statins and PCSK9 inhibitor. And I wonder if they’ll have a study long enough to show overall mortality? Even if lower LDL results in fewer heart attacks, it won’t help if the overall death rate is higher in the drug-taking group. (And this assumes the drug’s benefit is due to lower LDL, which is probably not true, at least for statins, which have other effects.)


(Bacon is a many-splendoured thing) #5

Add to that that a number of large, well-funded studies (among them Keys’s own Minnesota Coronary Study) have shown an inverse association between LDL and either cardiovascular risk or all-cause mortality, and you have another reason not to worry about LDL.

Naturally, the fact that higher LDL associates with lower cardiovascular risk or lower all-cause mortality doesn’t give us reason to think that these lower risks are because of the higher LDL, but it is certainly thought-provoking. The inverse correlation is, however, ipso facto reason to dismiss LDL as a cause of cardiovascular disease or of mortality.

The logic runs thusly:

  1. This number is important, because we have a drug to lower it; and

  2. It’s better to manipulate a marker we have a drug for, than to try and assess what the actual cause really is and try to address that (especially since we probably don’t have a drug for it).

It’s relevant to note that in the U.S., every health insurance company either owns a pharmaceutical company or is owned by one, so the incentive to give patients a drug is powerful.


(Michael) #6

Thanks for the study link. Of course, you can greatly lower ldl through carbs or super high fat diets, although who would want to do that.


(Bob M) #7

This has always made sense to me. Various pieces of this puzzle are out there. For instance, pretty much every study I’ve seen of the elderly indicates higher LDL = longer life.

Even flu vaccine studies in people with heart disease, indicating that people getting the vaccine do better than those who don’t:

(Let’s ignore this is both RCT and epi data, and it’s reported as relative risk.) When you realize that a vaccine – which ostensibly does nothing to LDL – can reduce the risk of death in people with heart disease, then you know LDL isn’t everything, and the immune system means something.

And it’s a short step from there to realizing LDL’s immune system effects may be more powerful than its “artery clogging” effects. Even if such “artery clogging” effects exist – and I personally doubt they do – if the immune system effects are larger than the artery clogging effects, then higher LDL could be better than lower.

Sadly, some of us just have “low” LDL, no matter what we (I) do.


(Alec) #8

It’s this arrogance that makes me mad. What they are saying is: “we KNOW that LDL causes heart disease, so everyone needs as low LDL as possible”. It just does not stack up with the evidence, but they are too stubborn to even contemplate that they might be wrong. I am pretty sure there may also be some drug (statin) money in there somewhere too…


(Bacon is a many-splendoured thing) #9

I just watched this interview with Robert Lustig this morning. He has some very interesting things to say.


#10

Paul, this is an excellent interview! Thanks for posting it!


(Ohio ) #11

Pardon my ignorance: Aren’t cholesterol levels bad biomarkers? Measurements lack consistency? I assumed cholesterol level checks were something to be shelved with the food pyramid.


(Bacon is a many-splendoured thing) #12

Many people find it extremely difficult to admit that Ancel Keys’s diet-heart hypothesis is not backed by any valid data, and is in fact, contradicted by quite a few studies that were designed to support it. (This includes the Minnesota Coronary Study, which was designed by Keys, but from which he had his name removed when the data failed to support his pet hypothesis.)


(Alex K Chen) #13

Isn’t it the combo of high-LDL and high oxidative stress/inflammation that does it in? LDL alone is not bad (and VERY slightly/non-significantly associated with reduced epigenetic age)


(Bacon is a many-splendoured thing) #14

There have been a number of large, well-funded studies, including the Minnesota Coronary Study, that not only failed to show a correlation between LDL level and cardiovascular risk, but they showed an inverse correlation (i.e., the higher the LDL the lower the coronary risk).

While statisticians will not accept a positive correlation between A and B as sufficient proof that A causes B or B causes A (because of the risk of confounders in the data), they do agree that an inverse correlation is sufficient proof that A does not cause B. At best, all a positive correlation can do is to generate hypotheses for further research.


(BuckRimfire) #15

DongCAChypercoles.pdf (1004.7 KB)

Compliments of the University of Washington.

I hope that file works! And that Elsev. doesn’t send legal flying monkeys after me…


(Todd Allen) #16

Thanks! I was waiting for SciHub but sometimes that is a long wait.


(Bacon is a many-splendoured thing) #17

Thanks for the paper. I’m not sure I follow the logic of all their conclusions, but the data are very interesting.

What I’d like to see is a study comparing outcomes of people stratified by CAC score who do not take a statin, versus people stratified by CAC score who do take a statin. We know that taking a statin will definitely lower cholesterol levels, but we don’t know that lowering cholesterol levels produces better cardiac outcomes. That has been asserted to be the case, but I’d still like to see some good comparative data.

This paper, for example, asserts that “Patients with severe hypercholesterolemia are at increased risk for atherosclerotic cardiovascular disease,” but the reference given is to a set of guidelines that may or may not be based on valid data (I haven’t checked that source out, yet).

The authors of this paper discuss familial hypercholesterolaemia, but they don’t appear to be familiar with the work of Harlan, Graham, and Estes (“Familial Hypercholeterolemia: A Genetic and Metabolic Study,” Medicine, vol 45, no. 2 [1966]) and similar papers from that era, in which it is noted that the factors that lead to cardiovascular disease in patients with familial hypercholesterolaemia do not appear to be related to their cholesterol level, no matter how high.


(Bob M) #18

One of the more interesting studies I’ve seen was where they stratified people into groups based on their actual LDL reduction, and found the “benefit” to a single statin did not improve with lower LDL. In other words, if 100 people take a certain daily dose of statin, there will be a range of values over which LDL will be lowered, from low to high. If you divide those into groups, if lower LDL = better, then lower LDL should have better results. But that did not happen.

To me, this means the “benefit” to statins is not caused by lowering LDL.


(Central Florida Bob ) #19

Skimming the article I see no mention of “healthy user bias.” That people who got get a flu shot may be taking “better care of themselves” in other ways, or that getting a milder case of flu (or no flu at all) makes work easier on their hearts.

I find it hard to take reports like that seriously.

The meta-analysis of 16 randomized and observational studies covered the experiences of more than 237,000 people. It concluded those with heart disease who were vaccinated for the flu were 18% less likely to die from heart problems and 28% less likely to die from any cause. They also were 13% less likely to experience any type of major heart problem than those who didn’t get a flu shot.

When I took statistics, we were taught that a meta-analysis was less reliable than any one of the studies it was based on since it’s all but impossible to determine how well those groups compare on the countless other confounding factors. I would consider those to be insignificant findings, regardless of their P value. I ignore those studies until the relative risk gets into the “twice as likely” range, not 18 or 28%.


(Bob M) #20

The flu shot was given to people with heart disease. Now, it’s a meta-analysis of both RCTs (better) and observational (crap) studies.

But there are postulated mechanisms for viruses causing atherosclerosis:

The beginning of part 9:

Like you, I find the RRs (relative risks) to be small, but I think there is something there between viruses and atherosclerosis. And it has nothing to do with LDL (other than higher LDL probably means better immune system, so lower chance of atherosclerosis via viruses).

And, here’s the great (in my opinion) Malcolm Kendrick discussing how a virus (covid 19 in this case) might cause endothelial damage, leading to blood clots:

I’m of the opinion (as is Dr. Kendrick - and he’s the one that got me to this point) that anything that limits damage to the endothelium is beneficial, and if the flu vaccine causes less viral load, and viruses cause damage to the endothelium, then the flu vaccine is good. Testing this is, of course, ridiculously difficult.

For decades, I did not get a flu shot. Decades. However, as I get older, if the flu shot means less infection, then that could mean less atherosclerosis and therefore less heart disease. Since my LDL is “low”, I need all I can get. :wink: