This is another example of an observational study, where the “observations” are what they are…but the interpretation esp among the chattering class on social media is what leads to the facepalms.
U-shaped associations are nothing new. Same exists for BMI (https://pubmed.ncbi.nlm.nih.gov/27146380/) because obviously, if you have an underlying condition that results in you wasting away….that’s not going to bode well for your survival. Doesn’t impugn the observation itself. But it does behoove you to understand what you’re observing, and also what you’re NOT observing.
Just as a person who, with intervention, goes from high BMI to low BMI is NOT comparable to someone who already starts with low BMI, so too that a person who goes from high cholesterol to low cholesterol with therapy is not the same as someone who starts with low cholesterol (for whatever reason).
Sadly, social media rewards the rage (and idiotic) takes. Just as there are flat-earthers and QAnon enthusiasts, there will be folks who eat up this and assorted other insanities.
"The second Hill criteria violated by this study is coherence. The causal relationship (if present) should not contradict existing knowledge or facts about the nature of disease."
Preemptively dismissing contrary findings out of hand sounds like bad science to me. Certainly, additional scrutiny is appropriate for anomalous results, but you seem to be going rather further than that to advocating actively excluding anything that doesn't fit the current consensus from receiving serious consideration.
I'll give you another observational study. In the 70s and 80s, the coronary care units were filled with 50 to 60-year-old men coming in with massive myocardial infarctions. We would see rupture of the papillary muscle, rupture of the interventricular septum, severe pericarditis, and rupture of the anterior free role ---routinely. I add routinely because I'd be very interested in hearing from any cardiology fellows of how commonly they've seen this in their training. By the time I retired 40 some years later, the coronary care unit mainly had 85-year-olds with troponin bumps. Massive difference. In 1980 we all went to see an 85-year-old man who was admitted because none of us knew anyone that old .. think about that for a second . Now you could argue the population all went out lost weight became vegetarians and joggers. We sort of know that the opposite is true. Look at a beach picture from the 70s just google it and look at your neighbors. Do you see the difference? I'm going to a football game this weekend at my alma mater. The last time I went and we sat on the seat, which is a number on a bleacher, 10 people both showed up for seat seats 1 to 10 . Two people were left standing because there was no room! And that does not includes my grandson who is eight who did didn't even take a seat. Perhaps the reduction in death and there has been a huge reduction in mortality that no one seems to acknowledge, is due to better blood pressure control or smoking cessation. At the same time, though this has been offset clearly by the increase in weight and diabetes. The bottom line is the vast majority of cardiologists have seen death every day none of it has been from the statins. I'm glad the coronary care unit is quieter. Medicine is personal. A non-smoking non-diabetic 40-year-old whose parents live to be 95 and who decided to have a "CTA to be sure" and it was normal-. Probably not going to get a lot of benefit from the statins. The same 45-year-old man with elevated CRP who already has coronary calcification different story. The sad part is most of the discussion is not with the healthy 45-year-olds who understand risk and benefit and who are already are doing everything to make themselves healthy. The people that I used to see are the obviously high risk person with multiple risk factors plaque on Imaging, who don't want to take statins because they see some Internet article. Well cardiologists have mortgages and student loans I suppose. I'm just glad that for the most part the coronary care unit is not full of dead bodies like when our treatment was a Swan-Ganz catheter Lasix and fluids to get an ideal wedge pressure, morphine, oxygen and aspirin....
Excellent as always. So I have a question. I started off as a 60s plus year-old man with high blood pressure and a cholesterol of 300; ldl 200. This was after forced retirement depression and a 60 pound weight gain. Add a GLP, 50 pound weight loss, Repatha, zetia and a low dose of Lipitor and my last LDL is 20. Feel great cardiologist says way to go a family doctor said that's "too low." I hate to use anecdotes but I've never seen anyone die of a low cholesterol , never seen anyone since Baycol was taken off the market die indirectly from a statin (or directly) . I've lost track of the number of people dying or not dying and being disabled from atherosclerosis. At present, I'm taking the win. Anyone disagree?
You probably could have stopped at "study making the rounds on social media" and ignored the rest. Or you could have noted that the study was an observational study reported in Nature and, again, spared some effort. Large observational studies reported in Nature are not known for being much any more except clickbait. There was an era when Science and Nature were notable for thoughtful and comprehensive science publication, and when the studies reported there were impactful and useful. That era is long past.
This is like the long-ago height-weight studies that showed it was bad to be skinny and good to be a little fatter. True, in general, but only because smokers and the severely chronically ill tended to be underweight. Arguing that correlation isn't causation and that there were no controls for anything was arguing accurately but to no point. Just like the current study.
If you want to live a long time, pick long-lived ancestors. If you have short-lived ancestors, maybe there's a modifiable reason that will give you an edge, but maybe not. For those with high lipids and family that all dies of stroke and MI, a statin may help you live longer (or better, in that surviving a stroke or a major MI is not a great thing either). If your lipids are a little above normal but everyone in your family dies of cancer under the age of 60, statins are probably irrelevant to you. Mortality, and prevention of mortality, is individual, not communal.
High cholesterol is a risk factor for cardiac death. Very low cholesterol is often seen in the very malnourished, including patients with metastatic cancer, lymphoma/leukemia, malabsorption syndromes, or eating disorders. In both instances, it’s not purely the cholesterol itself: It’s the relevant associated factors. Those with cholesterol levels of over 200 are often obese, and/or demonstrate poor dietary habits.
The entire walking population does not require statins. Some do. They have utility, partially reflected in a >30% decline in cardiac mortality in the US in 35 years or so. But zealots gotta zealot, and insist upon their own unitary beliefs, reflecting incapacity for nuance, rather than possession of The One Answer.
All other things equal, you’re better off with a cholesterol of 175 than 215. 240 bad. 130 bad. Better to get there by weight control and knocking off the bacon & Krispy Kremes.
(Yes, I’m a physician. No, not a cardiologist nor bariatric surgeon).
Can we talk about fully informed consent? Whatever the truth of the purported statin benefits, doctors must be providing patients with full side-effect disclosure. A relative of mine, suffered years of muscle pain/wasting as well as sudden onset deafness, and dementia, which are all known side effects, yet STILL is pressured by his doctor to take statins.
The problem is the prescribing doctors have their heads in the sand (or worse) and dismiss the side effects as happenstance or anything besides statin-caused. That way they don't feel any need to tell people. Ta-da.
Good grief, stop the cholesterol nonsense. There is no one-size-fits-all in medicine even though that is one of its linchpins and unquestioned tenets. And why the system is beyond broke. We are all individuals and each of us requires different amounts of everything at different times.
These cholesterol standards are gimmicks to get more people on dangerous heart drugs. The body rules, not silly drugs and useless measurements.
"They made minimal attempts at adjusting for variables: age, sex, smoking status, alcohol, physical activity, systolic BP, fasting glucose and BMI."
?? The paper and its supplement have numerous notes that the hazard ratios are " after adjustment for age at baseline (continuous variable), smoking status, alcohol use, physical activity, known history of heart disease, stroke, or cancer, body-mass index, systolic blood pressure, and fasting glucose levels."
But the paper's author neglects to describe those adjustments which may greatly affect the interpretation of the results, which greatly reduces the usefulness of this paper.
I suspect you’re quite correct that this study is demonstrating confounding - younger men having total cholesterol lower than around 200 probably have something else rather serious going on.
Actually, the paper shows mean and median TC levels of 170 for Korean men of 30, and even lower for younger than 30! Hopefully the majority of young men are fine.
Well look what starts happening to the hazard ratio for men around a cholesterol level of 160-170 in 18-34 and then like sub-200 in 35-44 year old men.
EDIT: What would have also been helpful that I didn’t see on my rapid run through that paper is something about how much cholesterol treatment was going on in these groups.
How do you explain the multiple observational studies and meta-analysis showing an associated lower risk of dementia and Alzheimer’s with individuals who take statins?
The second article you linked https://www.nature.com/articles/s41586-023-06120-6. This study was done in mouse models. Mouse models do not equal human models. Data about dementia and Alzheimer’s is likely harder to interchange from mouse to humans because the mice must be given injections of amyloid or genetically engineered for dementia or Alzheimer’s disease. They do not acquire a neurodegenerative disease like humans do. Further making the transfer of information more difficult from mice to humans.
You are suggesting that the ONLY way to lower cholesterol (a number that has been significantly lowered year over year) with taking a statin and IGNORING the inherent risks of taking the said statin!
This also suggest that certian ones may have more issues.
I am not suggesting taking a statin is the “only” way to lower cholesterol. It is an effective way of lowering cholesterol.
What I am suggesting is that you do not support statements for humans with animals models and animal studies.
The Harvard opinion piece you linked is actually supporting statins and refutes the risk of dementia and statin use. “This also suggest that certian ones may have more issues.”The suggestion of using different statins for cognition is theoretical and has no randomized control evidence or even observational data supporting it.
"While you would expect that statin use would reduce the risk of cognitive decline and dementia because statins lower cardiovascular risks and the risk of stroke, it hasn’t been clearly shown to be the case," says Dr. Manson. "It’s surprising that there’s not a clearer reduction seen. If anything, some of these studies have raised concerns about cognitive risks."
The author is wrong in his statement that there are many studies showing taking statins leads to lower all cause mortality. There is at least one meta study that looked at 29 RCT done by the drug companies and in 27 of the 29 studies there was zero mortality benefit. IE the folks on Statins did not die less. So given that the author would state something so obviously wrong destroys the entire credibility of this article. Maybe the NIH should do a study that studies the effect of low cholesterol on overall mortality in a large population. As to the authors statement that the study did adjust for a whole host of other factors. If you have a population of over 13 million you have enough data not to need those factors because you’re observing the whole. If your study is on only a few hundred or even a few thousand. Then you would need to take those factors into consideration. Because one observation out of 100 can skew your results it’s called the law of big numbers. One more example of the author gas lighting in this article. One a grading scale. I am giving this one an F for attempting to gaslighting and ignoring reality.
This is another example of an observational study, where the “observations” are what they are…but the interpretation esp among the chattering class on social media is what leads to the facepalms.
U-shaped associations are nothing new. Same exists for BMI (https://pubmed.ncbi.nlm.nih.gov/27146380/) because obviously, if you have an underlying condition that results in you wasting away….that’s not going to bode well for your survival. Doesn’t impugn the observation itself. But it does behoove you to understand what you’re observing, and also what you’re NOT observing.
Just as a person who, with intervention, goes from high BMI to low BMI is NOT comparable to someone who already starts with low BMI, so too that a person who goes from high cholesterol to low cholesterol with therapy is not the same as someone who starts with low cholesterol (for whatever reason).
Sadly, social media rewards the rage (and idiotic) takes. Just as there are flat-earthers and QAnon enthusiasts, there will be folks who eat up this and assorted other insanities.
"The second Hill criteria violated by this study is coherence. The causal relationship (if present) should not contradict existing knowledge or facts about the nature of disease."
Preemptively dismissing contrary findings out of hand sounds like bad science to me. Certainly, additional scrutiny is appropriate for anomalous results, but you seem to be going rather further than that to advocating actively excluding anything that doesn't fit the current consensus from receiving serious consideration.
Wait,What??
"all show a consistent reduction in cardiac events and reduction of all-cause mortality."???
I'll give you another observational study. In the 70s and 80s, the coronary care units were filled with 50 to 60-year-old men coming in with massive myocardial infarctions. We would see rupture of the papillary muscle, rupture of the interventricular septum, severe pericarditis, and rupture of the anterior free role ---routinely. I add routinely because I'd be very interested in hearing from any cardiology fellows of how commonly they've seen this in their training. By the time I retired 40 some years later, the coronary care unit mainly had 85-year-olds with troponin bumps. Massive difference. In 1980 we all went to see an 85-year-old man who was admitted because none of us knew anyone that old .. think about that for a second . Now you could argue the population all went out lost weight became vegetarians and joggers. We sort of know that the opposite is true. Look at a beach picture from the 70s just google it and look at your neighbors. Do you see the difference? I'm going to a football game this weekend at my alma mater. The last time I went and we sat on the seat, which is a number on a bleacher, 10 people both showed up for seat seats 1 to 10 . Two people were left standing because there was no room! And that does not includes my grandson who is eight who did didn't even take a seat. Perhaps the reduction in death and there has been a huge reduction in mortality that no one seems to acknowledge, is due to better blood pressure control or smoking cessation. At the same time, though this has been offset clearly by the increase in weight and diabetes. The bottom line is the vast majority of cardiologists have seen death every day none of it has been from the statins. I'm glad the coronary care unit is quieter. Medicine is personal. A non-smoking non-diabetic 40-year-old whose parents live to be 95 and who decided to have a "CTA to be sure" and it was normal-. Probably not going to get a lot of benefit from the statins. The same 45-year-old man with elevated CRP who already has coronary calcification different story. The sad part is most of the discussion is not with the healthy 45-year-olds who understand risk and benefit and who are already are doing everything to make themselves healthy. The people that I used to see are the obviously high risk person with multiple risk factors plaque on Imaging, who don't want to take statins because they see some Internet article. Well cardiologists have mortgages and student loans I suppose. I'm just glad that for the most part the coronary care unit is not full of dead bodies like when our treatment was a Swan-Ganz catheter Lasix and fluids to get an ideal wedge pressure, morphine, oxygen and aspirin....
Dr. Mandrola, if you are missing consistency you may want to look at the Danish study basically supporting the associations found in Korean paper.
https://www.bmj.com/content/371/bmj.m4266?fbclid=IwY2xjawNZ96BleHRuA2FlbQIxMQABHgANkw-srXnOnwp-hII6VusLrxzPL_l89E53kdUAx5qskyzZ7ECOKU6RxD1H_aem_0lcNBtnBK1iYeebPwtjXjQ
If you prefer data from the USA, here you are
https://bmjopen.bmj.com/content/14/3/e077949?fbclid=IwY2xjawNZ9-RleHRuA2FlbQIxMQABHjGwSHuDQNKi0QSF6s9zV13bvr8uIcZAyjtUg4pn3yP3uc-gdzox60unayiH_aem_F_kUCCgOKWT35GuROHj6pA
Still inconsistent?
Excellent as always. So I have a question. I started off as a 60s plus year-old man with high blood pressure and a cholesterol of 300; ldl 200. This was after forced retirement depression and a 60 pound weight gain. Add a GLP, 50 pound weight loss, Repatha, zetia and a low dose of Lipitor and my last LDL is 20. Feel great cardiologist says way to go a family doctor said that's "too low." I hate to use anecdotes but I've never seen anyone die of a low cholesterol , never seen anyone since Baycol was taken off the market die indirectly from a statin (or directly) . I've lost track of the number of people dying or not dying and being disabled from atherosclerosis. At present, I'm taking the win. Anyone disagree?
To say "cholesterol is a known risk factor for heart disease" is misleading. Respectfully disagree. See this year's (https://www.nejm.org/doi/full/10.1056/NEJMoa2415879) NEJM, largest and best study of the topic essentially disproving the concept or, to be generous, seriously undermining it. I wrote on it here (https://researchtranslation.substack.com/p/the-cholesterol-myth-unraveled-for), and elsewhere. Your point on causality is excellent and well taken, but we differ on cholesterol's role.
You probably could have stopped at "study making the rounds on social media" and ignored the rest. Or you could have noted that the study was an observational study reported in Nature and, again, spared some effort. Large observational studies reported in Nature are not known for being much any more except clickbait. There was an era when Science and Nature were notable for thoughtful and comprehensive science publication, and when the studies reported there were impactful and useful. That era is long past.
This is like the long-ago height-weight studies that showed it was bad to be skinny and good to be a little fatter. True, in general, but only because smokers and the severely chronically ill tended to be underweight. Arguing that correlation isn't causation and that there were no controls for anything was arguing accurately but to no point. Just like the current study.
If you want to live a long time, pick long-lived ancestors. If you have short-lived ancestors, maybe there's a modifiable reason that will give you an edge, but maybe not. For those with high lipids and family that all dies of stroke and MI, a statin may help you live longer (or better, in that surviving a stroke or a major MI is not a great thing either). If your lipids are a little above normal but everyone in your family dies of cancer under the age of 60, statins are probably irrelevant to you. Mortality, and prevention of mortality, is individual, not communal.
As a physician with over 40 years of clinical experience, I agree with your conclusions.
High cholesterol is a risk factor for cardiac death. Very low cholesterol is often seen in the very malnourished, including patients with metastatic cancer, lymphoma/leukemia, malabsorption syndromes, or eating disorders. In both instances, it’s not purely the cholesterol itself: It’s the relevant associated factors. Those with cholesterol levels of over 200 are often obese, and/or demonstrate poor dietary habits.
The entire walking population does not require statins. Some do. They have utility, partially reflected in a >30% decline in cardiac mortality in the US in 35 years or so. But zealots gotta zealot, and insist upon their own unitary beliefs, reflecting incapacity for nuance, rather than possession of The One Answer.
All other things equal, you’re better off with a cholesterol of 175 than 215. 240 bad. 130 bad. Better to get there by weight control and knocking off the bacon & Krispy Kremes.
(Yes, I’m a physician. No, not a cardiologist nor bariatric surgeon).
Can we talk about fully informed consent? Whatever the truth of the purported statin benefits, doctors must be providing patients with full side-effect disclosure. A relative of mine, suffered years of muscle pain/wasting as well as sudden onset deafness, and dementia, which are all known side effects, yet STILL is pressured by his doctor to take statins.
The problem is the prescribing doctors have their heads in the sand (or worse) and dismiss the side effects as happenstance or anything besides statin-caused. That way they don't feel any need to tell people. Ta-da.
Good grief, stop the cholesterol nonsense. There is no one-size-fits-all in medicine even though that is one of its linchpins and unquestioned tenets. And why the system is beyond broke. We are all individuals and each of us requires different amounts of everything at different times.
These cholesterol standards are gimmicks to get more people on dangerous heart drugs. The body rules, not silly drugs and useless measurements.
"They made minimal attempts at adjusting for variables: age, sex, smoking status, alcohol, physical activity, systolic BP, fasting glucose and BMI."
?? The paper and its supplement have numerous notes that the hazard ratios are " after adjustment for age at baseline (continuous variable), smoking status, alcohol use, physical activity, known history of heart disease, stroke, or cancer, body-mass index, systolic blood pressure, and fasting glucose levels."
But the paper's author neglects to describe those adjustments which may greatly affect the interpretation of the results, which greatly reduces the usefulness of this paper.
I suspect you’re quite correct that this study is demonstrating confounding - younger men having total cholesterol lower than around 200 probably have something else rather serious going on.
Actually, the paper shows mean and median TC levels of 170 for Korean men of 30, and even lower for younger than 30! Hopefully the majority of young men are fine.
Well look what starts happening to the hazard ratio for men around a cholesterol level of 160-170 in 18-34 and then like sub-200 in 35-44 year old men.
https://drive.google.com/file/d/1dW4VKx_r3XmKt0dbnVer4WPb0yOWEmYt/view?usp=drivesdk
EDIT: What would have also been helpful that I didn’t see on my rapid run through that paper is something about how much cholesterol treatment was going on in these groups.
Do you make any correlation between low cholesterol and the myelin sheath being denigrated?
1) We know that Cholesterol is needed to support the myelin sheath.
https://pubmed.ncbi.nlm.nih.gov/15793579/
And damage to the Myelin sheath is higher in those with dementia.
https://www.nature.com/articles/s41586-023-06120-6
How do you explain the multiple observational studies and meta-analysis showing an associated lower risk of dementia and Alzheimer’s with individuals who take statins?
The second article you linked https://www.nature.com/articles/s41586-023-06120-6. This study was done in mouse models. Mouse models do not equal human models. Data about dementia and Alzheimer’s is likely harder to interchange from mouse to humans because the mice must be given injections of amyloid or genetically engineered for dementia or Alzheimer’s disease. They do not acquire a neurodegenerative disease like humans do. Further making the transfer of information more difficult from mice to humans.
You are suggesting that the ONLY way to lower cholesterol (a number that has been significantly lowered year over year) with taking a statin and IGNORING the inherent risks of taking the said statin!
This also suggest that certian ones may have more issues.
https://www.health.harvard.edu/staying-healthy/do-statins-increase-the-risk-of-dementia
I am not suggesting taking a statin is the “only” way to lower cholesterol. It is an effective way of lowering cholesterol.
What I am suggesting is that you do not support statements for humans with animals models and animal studies.
The Harvard opinion piece you linked is actually supporting statins and refutes the risk of dementia and statin use. “This also suggest that certian ones may have more issues.”The suggestion of using different statins for cognition is theoretical and has no randomized control evidence or even observational data supporting it.
"While you would expect that statin use would reduce the risk of cognitive decline and dementia because statins lower cardiovascular risks and the risk of stroke, it hasn’t been clearly shown to be the case," says Dr. Manson. "It’s surprising that there’s not a clearer reduction seen. If anything, some of these studies have raised concerns about cognitive risks."
AND
https://jnm.snmjournals.org/content/62/supplement_1/102
The author is wrong in his statement that there are many studies showing taking statins leads to lower all cause mortality. There is at least one meta study that looked at 29 RCT done by the drug companies and in 27 of the 29 studies there was zero mortality benefit. IE the folks on Statins did not die less. So given that the author would state something so obviously wrong destroys the entire credibility of this article. Maybe the NIH should do a study that studies the effect of low cholesterol on overall mortality in a large population. As to the authors statement that the study did adjust for a whole host of other factors. If you have a population of over 13 million you have enough data not to need those factors because you’re observing the whole. If your study is on only a few hundred or even a few thousand. Then you would need to take those factors into consideration. Because one observation out of 100 can skew your results it’s called the law of big numbers. One more example of the author gas lighting in this article. One a grading scale. I am giving this one an F for attempting to gaslighting and ignoring reality.
Right on the money. Statins are dangerous. Studies need to be independent and lasting 20-30 years, at least. No big pharma manipulating the data.