Hacker News Reader

Comment by simonsarris | Hacker News Reader

Some scientists/doctors love using cholesterol because its easy to measure. But it's still mostly useless. I wonder if its mostly useless here.

Oh look it is. From the article:

> Comparing the cholesterol and survival rates of the burger-eating crows with those of nearby crows who were weren’t supplemented with fast food, the team found that cholesterol levels did not have a detectable effect on the birds’ survival.

Meanwhile, in a 2015 survey by Credit Suisse Foundation found 54% of docs falsely believed eating cholesterol-rich food raises blood levels of cholesterol and damages the heart. In the words of the survey, "This is a clear example of the level of misinformation that exists among doctors." https://www.credit-suisse.com/about-us-news/en/articles/news...

Why won't this dumb meme die?

View on HN

Heart disease is America's biggest killer. There is a lot of data that eating cholesterol effects disease rates including heart disease, cancer, etc.

If you study disease rates for those eating cholesterol and those replacing it with a plant based diet, eating cholesterol is higher.

While you have a particular point about eating cholesterol, measuring cholesterol, and related disease rates - are you suggesting people's health is not negatively affected by eating cholesterol?

Here's a simple study eating an egg versus not, it compares stopping eating cholesterol to stopping smoking. [0]

[0] https://www.ncbi.nlm.nih.gov/pubmed/21076725

Did you actually read that paper? It's an article, not a study. The article is here: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2989358/

The USA is obsessed with cholesterol because of years of bad science. Here's a 2015 Japanese Supplementary Review on Cholesterol and Mortality Rates: Higher Cholesterol = Longer Life

https://www.karger.com/Article/Pdf/381654

It concludes:

> Our fervent wish is that, through this supplementary issue, people can see that the cholesterol hypothesis relies on very weak data—and sometimes considerably distorted data. Indeed, many studies in Japan actually show that cholesterol plays a very positive role in health. We hope that JAS, and the government authorities that defer to JAS’s recommendations, will move toward recognizing cholesterol as a friend not an enemy. In the meantime, we will continue pushing for acceptance of the anti-cholesterol hypothesis, to reverse what we see as the biggest mistake made by medical science in the previous century.

I read it a few years ago. Thanks for the link, I'll take a look.

I agree with many of the critical points they make in this Japanese review. There should be higher standards for nutritional research.

The main idea seems to be that in Japan because there are some positive patterns observed regarding all causes of death and high cholesterol levels and because other studies showing high cholesterol is bad were all flawed, therefore it is warranted to recommend eating more cholesterol.

Well, the notion that eating cholesterol effects your cholesterol level goes against your first comment.

Then the sentiment that you can throw the baby out with the bath water seems extreme, if you can find a flaw in studies that disagree with you then your studies that agree with you prove the point. I mean, I agree that when there is no un-flawed data it is hard to draw conclusions, but are their studies all really as so unblemished. A main criticism of theirs is studies citing rates of death for certain diseases without mentioning the rate for all causes of death, but a lot of their cited studies do the exact same, only some of them reported rates for all causes of death.

Finally, and I know you're surprised, I am not persuaded by them that eating cholesterol or having high/"normal" cholesterol levels is neutral or positive in affecting your health. I agree with them that we need much better data. I understand that in Japan they observed lower rates of date from all causes. But the only question I really have is what should I eat to avoid dying as an American, and this article just doesn't really cover that one way or another. I'm very grateful for sharing it, there are a lot of good insights in there.

The notion that direct intake of cholesterol is bad is itself problematic. Off the top of the head I can recall at least 3 studies getting null results. Can't link you then on mobile, maybe late if you're still interested. (There are very rare genetic exceptions where body does not automatically balance intake with less internal production.)

The problem is likely that cholesterol itself is a coincidence, and that intake of too much red meat, sugar, heavily processed foods and also just to many calories in are the major problem.

And likely trans fats are much more damaging. Plus not enough vegetables or vitamin D or B12 in diet.

It just so happens that there's cholesterol in three of these four.

Right now there is more emphasis on lipid ratios (including IDL and VLDL) rather than total number anyway. Those seem much better correlated to cardiac endpoints.

And I think, I've seen a lot of studies that compare eating "normal" amounts of cholesterol to eating more than normal. But "normal" cholesterol levels in your blood still leave you with a "normal" American rate of heart disease, read, still very high. Those studies might not show much of a difference in disease rates. But studies on eating cholesterol versus no cholesterol are much clearer in showing a significant difference. On a plant based diet people can get their cholesterol below 60 and the disease rates for people with that low of cholesterol are much lower.

I firmly believe that its the sugar/insulin response that creates most of these health problems due to the inflammation. There's a few studies that scratch the surface of just how bad sugar is, but with all the focus on cholesterol, there's been no focus on how much sugar is in people's diets who also happen to consume high cholesterol.

I've been off and on a keto diet for the better part of 12 years. I don't eat like crap when I'm "off", but my entire body gets inflamed and I blow up like a balloon, compared to being on it. My resting heart rate rises. This should be impossible, according to the dietary cholesterol theory. My health should get worse, and blood cholesterol should raise. This does not happen. We've lived for 50 years under low fat high carb dietary recommendations, primarily because its so much easier to scale grain production. Now there's grams and grams of sugar in everything we eat.

I don't think there is one smoking gun for a chemical in food.

To me, I see a clear pattern. Reducing nutrition to consuming more of a specific chemical/molecule or consuming less of a specific chemical/molecule is always way more complicated than that. The best way to get the complicated combination/form of chemicals is to just eat as low on the food chain as possible in a minimally altered form.

Naturally occurring sugar in plants, say an orange, or rice, is good for you. Extracting sugar doesn't ever seem necessary for the American diet. The oil in a nut is good for you, extracting oil and using it to cook doesn't ever seem necessary for the American diet.

So I don't think it is controversial to say eating plans is good for you, especially say broccoli for your inflammation. The more broccoli you eat, the healthier you are; its really simple.

The second part of what I said, to try to eat low on the food chain is more controversial. But as soon as you combine that rice with meat, that sugar starts interacting in new more complicated ways and the it is no longer as simple as, the more rice you eat the healthier you are. It becomes, if you combine it with meat you have to limit the amount of rice you eat because together they spike your blood sugar level [0] study, image of that spike [1]. And I rarely hear, "the more meat you eat, the healthier you are".

So to me it seems like you can try to walk a tight-rope of eating the right amounts of the right processed and meat foods, or you can just eat things that generally make you healthier when you eat more of it.

That is a simple baseline that you adjust based on atypical differences you have, celiac disease, etc.

Regarding cholesterol specifically, and how you mentioned it conflicting with current theories that are being questioned (fairly enough) in this thread - here's some summary thoughts from the editor-in-chief of the American Journal of Cardiology on it. [2] They might be helpful providing some color to the conversation.

[0] https://www.ncbi.nlm.nih.gov/pubmed/2679037 [1] https://imgur.com/ZqHpEzv [2] http://www.webedcafe.com/extern/program_media/ajconline.org/...

I don't think this article really says anything for or against that. The article did note that eating the burgers did raise the crows' cholesterol levels and also that "Crows, “which can live more than 15 years in the wild, might develop disease later in life,” but this needs further research, Townsend says."

I've got high cholesterol from both parents -- when I was first tested (around 25), my total cholesterol was ~250, my HDLs were < 20. I had ran multiple half marathons that year, and I was eating rather well.

My doctor asked me to try cutting cholesterol out of my diet, and it helped remarkably. I went vegan for quite awhile, and added some foods that help with HDL. The next time I got checked, they were amazing.

I've since added some cheese/eggs/fish back into my diet, but I try to keep it low. I hover around 200, with pretty decent HDLs.

Genetic testing showed I (may) have some genetic issues around cholesterol recycling, so it's not a one-size-fits-all thing.

Yea, you can have issues with NPC1L1 or ABCG5/G8 which will impact cholesterol/plant sterol absorption or excretion in the gut. If you are a hyperabsorber, you may want to check your plant sterol levels (xenosterols), as they can lead to health issues.

What is your HDL now?

I don't get what you're upset about. The title doesn't infer that cholesterol is causing death and they state int he article that they aren't related.

Then why mention high cholesterol at all?

Seems more like a tangential complaint about general societal ignorance of that fact... and the title seems to be phrased as clickbait for people that believe in the cholesterol boogieman.

The title of the article says “this happens”, but then body continues with “but it looks like it doesn’t matter”. It’s clickbait.

I don't understand - what is the dumb meme? The link of high cholesterol to heart disease seems to be very well supported by numerous studies, and so are the benefits of reducing high LDL with diet or medication (statins).

https://en.wikipedia.org/wiki/Cholesterol#Hypercholesterolem...

According to the lipid hypothesis, elevated levels of cholesterol in the blood lead to atherosclerosis which may increase the risk of heart attack, stroke, and peripheral vascular disease. Since higher blood LDL – especially higher LDL concentrations and smaller LDL particle size – contributes to this process more than the cholesterol content of the HDL particles,[68] LDL particles are often termed "bad cholesterol". High concentrations of functional HDL, which can remove cholesterol from cells and atheromas, offer protection and are commonly referred to as "good cholesterol". These balances are mostly genetically determined, but can be changed by body composition, medications, diet,[69] and other factors.[70] A 2007 study demonstrated that blood total cholesterol levels have an exponential effect on cardiovascular and total mortality, with the association more pronounced in younger subjects. Because cardiovascular disease is relatively rare in the younger population, the impact of high cholesterol on health is larger in older people.[71]

Elevated levels of the lipoprotein fractions, LDL, IDL and VLDL, rather than the total cholesterol level, correlate with the extent and progress of atherosclerosis.[72] Conversely, the total cholesterol can be within normal limits, yet be made up primarily of small LDL and small HDL particles, under which conditions atheroma growth rates are high. A post hoc analysis of the IDEAL and the EPIC prospective studies found an association between high levels of HDL cholesterol (adjusted for apolipoprotein A-I and apolipoprotein B) and increased risk of cardiovascular disease, casting doubt on the cardioprotective role of "good cholesterol".[73][74]

https://en.wikipedia.org/wiki/Hypercholesterolemia#Diagnosis

Higher levels of total cholesterol increase the risk of cardiovascular disease, particularly coronary heart disease.[33] Levels of LDL or non-HDL cholesterol both predict future coronary heart disease; which is the better predictor is disputed.[34] High levels of small dense LDL may be particularly adverse, although measurement of small dense LDL is not advocated for risk prediction.[34]

https://en.wikipedia.org/wiki/Lipid_hypothesis

The lipid hypothesis (also known as the cholesterol 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]

An accumulation of evidence has led to the acceptance of the lipid hypothesis by most of the medical community.[3]

you're redirecting to a similar but unrelated claim. the claim under scrutiny is:

> eating cholesterol-rich food raises blood levels of cholesterol and damages the heart

not whether or not high cholesterol to heart disease generally is impactful. that's an important distinction, and what you've posted here isn't evidence for or against that.

I asked the OP to clarify, because I think they are trying to make both points: that cholesterol in food does not affect cholesterol in blood and that high cholesterol (of X kind) does not affect the risk of disease.

The way I understand it, cholesterol in food is correlated with cholesterol in blood, it's just that thee are other factors that may be more important, like genetics. Well, genetics.

This is all true but missing one important fact. Dietary cholesterol doesn't turn directly into blood cholesterol. It makes sense why this became a common misconception, but the actual physiological mechanics of this are more complex.

https://www.hsph.harvard.edu/nutritionsource/what-should-you...

I don't actually hear people talk about cholesterol in food, though. They almost always mean high in animal fat, which my understanding is that this then results in higher LDL in the bloodstream.

It's more complicated than that, apparently.

https://peterattiamd.com/the-straight-dope-on-cholesterol-pa...

I've trudged through all parts of his explanations; worth it if you're interested in lipidology, but this contains adequate punchlines to be helpful. TLDR we're probably measuring the wrong thing. And stop eating sugar. Oh yeah, and eating cholesterol is irrelevant.

My favorite bit in this, is when he gets to the really important part about how, precisely, lipoproteins cause disease, the explanation is that it "illegally parks"

"So what drives an LDL particle to do something as sinister as to leave the waterway (i.e., the bloodstream) and “illegally” try to park at a dock (i.e., behind an endothelial cell)? Well, it is a gradient driven process which is why particle number is the key driving parameter."

Except it isn't gradient driven, because it's an active process of transcytosis...

There's just so many holes in the lipid hypothesis.

It's an uncomfortable realization when you first figure out how much of modern "medical wisdom" is based on incomplete understanding, bad stats, marketing crapola from food and drug manufacturers, and weird NGO groups trying to justify their existence.

Modern medicine is really good at trauma care, and certainly the discovery of vitamins, antibiotics and hand washing are good ideas; otherwise, I'm not so sure!

Not to be rude to you or fellow HN users, but between a comment on HN and "modern 'medical wisdom'" I think I'll choose the latter every time, just on the basis that a comment on HN is not as likely to be correct as "modern 'medical wisdom'".

Also, just as a bit of pattern matching that may be completely off, but I haven't heard the most accurate things from people who challenge medical wisdom with scare quotes. Doctors study those things for a living. I don't know why people think they know best because they can read ncbi.gov.

But- I'll listen more carefully if the person commenting is a doctor who works with patients day-in, day-out. That kind of experience still counts for something, flawed as subjective judgements may be.

> Doctors study those things for a living. I don't know why people think they know best because they can read ncbi.gov.

See, that's just the thing: Doctors absolutely do not study this sort of thing for a living. Researchers do. That's why I linked you to a researcher, rather than something written by the guy I see when I need antibiotics or to get a mole checked out.

I ain't asking you to take my word for it: I provided you a link to a guy who really does study this sort of thing for a living. And he says ... it's really complicated, and the standard advice/treatments are probably baloney. Maybe he's wrong! Doesn't look that way though.

Doctors have to study to become doctors and then they practice medicine every day. I expect, if someone has a reliable intuition about how things like cholesterol work out in practice, that's a doctor who has to treat patients every day.

In fact, in this particular kind of knowledge (of how test results correlate with various outcomes) I would trust a doctor's intuition, formed by long experience, more than I'd trust data and research results.

Kind of nonplussed that you'd diss doctors as "the guy I see when I need antibiotics or to get a mole checked out". Do you really think what doctors do is so simple?

I taught medical students in a top-5 medical school and know how the sausage got made.

They're great at the things I listed: trauma care, treatment of acute conditions, the concept of vitamins and use of antibiotics. For stuff involving chronic conditions, or even nutrition advice, they appear to be worse than useless.

This is not my experience. My experience is that- it's a mixed bag.

I ... don't like to discuss personal medical details online of course. But, I recently had a dry cough that didn't go away after many months and I was obviously shitting bricks (though I don't smoke) so I went to my GP. She prescribed ...antacids. I left frustrated, and incomprehending, but of course she knows I have an ulcer, and it seems that the acid coming up my esophagus is irritating my vocal cords.

Note that she didn't sit down at her computer and google my symptoms or anything like that. Our meeting took all of 5 minutes (stupid NHS guidelines...) which is btw why she had no time to explain and I had to look up things myself. But she got it right. In 5 minutes time and under the pressure of a couple dozen other patients in the waiting room. It reminded me of stories about machine learning systems modelling some complex dataset and coming up with surprising results that initially leave researchers bewildered, until digging a bit reveals the accuracy of the model.

Perhaps teaching medical students rather than practicing yourself is skewing your view a bit (you didn't say you are actually a doctor though)? I have respect for doctors' intuition, but maybe, on my part, I'm just impressed because of my own lack of knowledge.

On the other hand, another doctor I see for a chronic condition I don't want to discuss online is completely useless and has all the intuition and skill of a boiled piece of cabbage. But that's not a tale for HN.

I'm not a doctor of medicine; my course (experimental physics methods; basically something where they learn something approximating general diagnosis skills) was a weeder for medical students.

I am in abundant agreement with you that some doctors are amazing and some are cabbage brained imbeciles. The point I'm trying to make here is that for squishy complex chronic conditions like hyperlipidemia, I don't think even the very good ones are well equipped to know what they're talking about unless they spend a lot of time reviewing the literature (and even then, if they act on it, they may be voiding their malpractice insurance terms among other things). T-tests on the differences in outcomes are fairly difficult to interpret for researchers; it's impossible to eyeball and make sense of it with the samples a GP is exposed to. There's all kinds of confounding bias as well; they don't see people who don't go to the doctor, have the condition (such as it is) and do fine without any intervention. As such, they basically stick with the prevailing ideology. And they get this ideology wrong all the time. It's not their fault; it's just how the world works. For myself, I build in a discount term for doctor advice based on what the issue is. Basically if doctors are provably good at treating an illness; I will shut my cake hole and obey. If they're not, I'm at least going to dig into the literature.

FWIIW an anecdote for you: Andy Sessler, the former director of LBNL developed a brain tumor while running the place. He had access to the very best treatment in the world, including the Bevatron proton bombardment technique, but he chose to adopt some kind of macrobiotic diet and take up running (founding the LBNL runaround in the process). I think the tumors eventually killed him; 35 years after he was diagnosed. I didn't know him beyond seeing him around at seminars and runarounds and such, but I assume his gamble was based on a deep understanding of how the stats on the disease treatments played out.

Thanks for a good conversation :)

Not a doctor, sorry :)

I have, however, spent many many hours over the last year reading all of the research I can find on the lipid hypothesis, metabolism, lipoproteins, nutrition, etc. I started a podcast, and have started chatting to researchers, doctors, etc.

>> Not a doctor, sorry :)

No matter. Sorry if my comment comes across as a bit stand-offish.

Your drastically oversimplifying LDL's role in disease.

High LDL is a very weak predictor. This has been known since the Framingham study. Looking at a combination of LDL, HDL and trigs is a much better predictor.

High LDL in combination with low trigs and high HDL, is actually protective (Not a risk factor at all)

The Jeppesen Study (https://www.ncbi.nlm.nih.gov/pubmed/11176761)

the Framingham Offspring study(https://www.ahajournals.org/doi/full/10.1161/CIRCOUTCOMES.11...)

How oxidised the LDL particles are is really the major metric that matters. From looking at cell cultures we know that high levels of non-oxidised LDL have no atherosclerotic properties inherently. https://link.springer.com/chapter/10.1007/978-981-13-0620-4_...

Personally its my opinion that the lipid hypothesis is essentially junk. The level of LDL is a major distraction, from root issues such as inflammation, oxidation stress and diabetes, which are caused mostly by sugar and refined carbohydrates consumption.

>> Personally its my opinion that the lipid hypothesis is essentially junk.

OK, but this seems to be just your opinion and contrary to what experts (doctors, public health bodies, etc) advise.

I'll thrown in my 2c that is really is junk.

Decent overview of some of the flaws: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5969815/ https://www.tandfonline.com/doi/full/10.1080/17512433.2018.1...

Doctors are not experts in this, nor are public bodies. Endocrinology researchers are...

It's really still an open field. New findings take almost a generation (20 years) to propagate to run off the mill doctors, slightly faster to specialists. And public bodies are slow as molasses in comparison, some still pushing advice from 1950s.

Sorry about self reply, edit was not available.

Here's a decent paper on the matter of getting research into practice:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3241518/

Reply on news.ycombinator.com