
Peter Attia Dives Deep on STATINS (side effects & the best alternatives)
What this covers
This is a clip from a longer conversation in which Peter Attia, with questions from Rhonda Patrick, works through the statin class one concern at a time. He starts with mechanism: statins block the first committed step of cholesterol synthesis, mostly in the liver, and the liver responds by upregulating LDL receptors, which does most of the LDL lowering. He stresses that the seven available statins are effective but not a benign class, and he names the four he prescribes.
The middle stretch takes the side effects in turn. Muscle aches affect about 7% of users and resolve within one to two weeks of stopping. About 4% develop type 2 diabetes through insulin resistance, which he says calls for monitoring beyond hemoglobin A1c. Raised transaminases are described as a sign of inflammation, not true liver dysfunction. Rhonda raises the mitochondrial question, since the statin pathway also makes ubiquinol). Peter says supplementation trials show no benefit for muscle soreness, though it is low-risk. He adds that any mitochondrial effect looks small when measured by lactate at Zone 2 effort. A tangent on metformin follows: he is skeptical of it as a longevity drug for healthy people because of mitochondrial toxicity.
The later sections concern the brain. No randomized trial has made dementia the primary outcome, but roughly 15 to 16 cardiovascular trials with dementia as a secondary outcome found neutral-to-beneficial effects. Even so, Peter argues for avoiding statins in APOE4 carriers and people with a family history of neurodegeneration, and he proposes tracking desmosterol, a marker of brain cholesterol synthesis. He notes that desmosterol is hard to get measured outside a few labs, and he mentions that Tom DePriest has been analyzing it. The alternatives come last. PCSK9 inhibitors are supported by natural-experiment evidence but cost around $500 a month in the US, with insurers requiring strict criteria. Ezetimibe is best matched by phytosterol testing, and bempedoic acid is the third option.
Speaker argues that statins are effective but have non-trivial side effects (muscle soreness, insulin resistance, liver inflammation) that require careful individual monitoring, and that newer alternatives like PCSK9 inhibitors, ezetimibe, and bempedoic acid offer better risk-benefit profiles for certain patient populations, particularly those with APOE4 genetic risk for neurodegeneration.
- Statins work by inhibiting cholesterol synthesis but affect ~7% with muscle soreness and ~4% with diabetes risk, requiring close monitoring of dexlosterol levels in high-risk individuals
- Desaturol (dexlosterol) levels below 0.8 mg/dL correlate with increased dementia risk, necessitating alternative drugs for APOE4 carriers or those with family history of neurodegeneration
- PCSK9 inhibitors, ezetimibe, and bempedoic acid provide effective lipid-lowering without the side effect profile of statins, though access is limited by cost barriers
This asset isn't compiled yet
You're seeing its claims, ranked. Compile it to build the argument threads, weight them, and check each claim against your library — the full view.
Dexlosterol is a cholesterol synthesis intermediate that is a better proxy for brain cholesterol synthesis than lathosterol, which is a better proxy for peripheral cholesterol synthesis.
“remember how I said there were two cholesterol synthesis Pathways well in the CNS really only you have one pathway and it's the pathway that goes through desol to cholesterol so desol levels are actually a decent proxy for brain cholesterol synthesis lathosterol which is the penultimate molecule in the other pathway is more of a proxy for peripheral cholesterol synthesis”
Ezetimibe is most effective in patients with defective ATP-binding cassette cholesterol transporters in the gut, which can be identified by measuring phytosterol levels; elevated phytosterol (particularly cytostanol and campesterol above normal) indicates impaired excretion and preserved absorption of plant sterols.
“the only times we see really Head Over Heels responses are in patients who have defective ATP binding cassettes in their gut and we measure that by looking at phytosterol levels so we measure two things one is called cytol one is called called compol those are phytosterol so these are cholesterol we don't make it's zerol uh pardon me it's it's it's phytosterol not zerol and so when we measure those levels we know that it speaks to how much plant sterile is being absorbed and not being excreted”
Statins increase transaminase levels (liver enzymes), with elevation indicating inflammation rather than true hepatic dysfunction, despite the term 'liver function tests' suggesting functional impairment.
“the third thing we see with statins is a change in or an increase in the transaminases or the liver function tests liver function test is a bit of a misnomer because the transaminases really tell us more about inflammation than function”
No randomized controlled trial has examined statin use with the primary outcome of Alzheimer's disease or dementia incidence, limiting causal inference; approximately 15-16 trials have used statins with cardiovascular outcomes as primary outcome and dementia/Alzheimer's as secondary outcomes, and all found neutral-to-beneficial statin effects on dementia incidence.
“there has never been a study done that has looked at the use of statins and the incidence of Alzheimer's disease or dementia as a primary outcome why is that important it's important because in clinical research the primary outcome is the only thing you can really take to the bank cuz that's what the study is powered to detect there are more than a dozen probably less than 25 so a big number of studies call it 15 16 that have used Statin have had a primary outcome of ascvd but a secondary outcome of Dementia or Alzheimer's disease and I looked at every single one of those and I can tell you that every single one of those found neutral to benefit of Statin use on the incidence of dementia”
Approximately 7% of statin users develop muscle aches, representing a large absolute number of people given statin prescription prevalence, though this side effect is completely reversible and resolves within 1-2 weeks of discontinuation.
“about 7% of people develop muscle aches on Statin so that's if you think about how many people are on those drugs or how many people are prescribed those drugs that's a huge number of people the good news is that's a completely reversible side effect so you put a person on a Statin they experience muscle soreness you take them off it's gone within a week or two”
A small subset of statin users (approximately 4%) may develop type 2 diabetes as a result of statin-induced insulin resistance, requiring careful monitoring of insulin sensitivity markers beyond just hemoglobin A1c.
“the other big side effect the one that I probably think about the most is insulin resistance so um a very small set of subset of people about 4% of people put on a Statin might go on to develop type 2 diabetes as a result of it”
Healthier individuals can rely more on mitochondria for ATP generation under increasing cellular demand, and the hallmark of aging and disease is the inability to maintain oxidative phosphorylation, resulting in earlier shift to glycolysis.
“we understand that the healthier an individual is the more they can rely on their mitochondria for ATP generation under increasing demands of the cell this is one of the Hallmarks of health and by extension one of the Hallmarks of aging and one of the Hallmarks of disease is an inability to do that meaning as the ATP Demand on a cell goes up there is an earlier and earlier shift to gly is as opposed to oxidative phosphorilation”
Seven statins currently exist in clinical use, each with different strengths and advantages, and they are not a benign class of drugs despite being very effective at lowering LDL cholesterol.
“there are currently seven Statin in in existence uh and each of them you know offers some strengths and advantages over others and they're not a benign class of drug so to be clear they're an effective class of drug they're very effective at lowering uh LDL cholesterol”
Statins lower LDL cholesterol by inhibiting the first committed step of cholesterol synthesis, which occurs throughout the body but primarily in the liver. When hepatic cholesterol synthesis is suppressed, the liver responds by upregulating LDL receptors, and this receptor upregulation is what primarily drives LDL lowering.
“they work by inhibiting the first committed step of cholesterol synthesis they do that everywhere but primarily in the liver and the response of the liver when cholesterol synthesis is being shut down the liver says I got to get more cholesterol in here and what does it do it puts a whole bunch more LDL receptors all over the liver and that's what's primarily driving down LDL in the presence of a Statin”
PCSK9 inhibitors are approved (since 2015) and have been tested in rigorous trials, including natural experiments with individuals carrying PCSK9-loss-of-function mutations, demonstrating safety and superior efficacy compared to other lipid-lowering drugs, with zero side effects.
“uh and highly effective in insanely safe zero side effects um cheaper than when they came out uh so they were approved in 2015 um and we have long-term data with the natur people walking around with a natural mutation right just amazing yeah exactly we have the natural experiment we have all of the data from these drugs um and these these drugs have been tested in really good trials and they've gone head-to-head with every drug and they always win and there's no side effects”
PCSK9 inhibitors cost approximately $500 per month in the United States, but are cheaper outside the US; drug pricing is more favorable internationally than in the United States.
“they're expensive right they're it's a $500 a month drug in the in the United States it's cheaper outside of this country so everything's better out of the US when it comes to drug pricing but in the US you're talking about 500 bucks a month for that drug if it's not a if it's not covered by your insurance company right”
Insurance companies typically will not cover PCSK9 inhibitors unless patients meet specific criteria such as familial hypercholesterolemia or having already had a cardiac event and being statin-intolerant.
“unfortunately most insurance companies will not cover it unless you meet certain criteria such as having familial hyper cholesterolemia or having already had a cardiac event like a heart attack and not being able to tolerate a Statin or what about myopathy like if you have muscle yes significant myopathy on multiple Statin but you'd also have to be at high enough risk to justify it”
Phytosterols (plant sterols like sitosterol and campesterol) are more atherogenic than cholesterol itself because they are more prone to oxidation, more inflammatory, and more oxidizable, making them a cardiovascular risk factor independent of elevated cholesterol.
“phytosterols are actually more arenic than cholesterol and that's also like Boston Hart would measure all these ferol okay um they're more prone to oxidation more inflammatory are they being carried in lipoproteins they're uh so so oid are they oxidized they're more oxidized they're more oxidizable”
It has not yet been studied whether APOE4 carriers have lower dexlosterol levels compared to APOE3 or APOE2 carriers, but this is a testable hypothesis that would make mechanistic sense given APOE's role in brain cholesterol metabolism.
“there is a correlation with apoe status on that no one has done that study yet in your in your clinical but in our clinical practice we just decided like why would we take the risk but but yes no one has done the study to show our desol levels lower in apoe4 individuals that's actually a very testable hypothesis and it makes a lot of sense cuz we know apoe is heavily involved in cholesterol activity in the brain”
Clinical trials examining ubiquinol supplementation to reduce statin-induced muscle soreness have shown no benefit; however, some patients report subjective improvement, and supplementation carries minimal harm risk.
“there have been a number of clinical trials looked looking at using or supplementing ubiquinol with patients taking statins they have mostly done this to assess uh the muscle soreness issue so they've mostly done this as a way to ask the question can you reduce the insulin the the incidence rather of muscle soreness with Statin I haven't looked at those literature in a couple of years the last I looked at them there was still no difference”
Statins inhibit the HMG-CoA pathway, which is also responsible for synthesis of ubiquinol (coenzyme Q10/CoQ10), which is essential for mitochondrial function and electron transport chain activity that couples oxygen consumption with energy (ATP) production.
“Statin are affecting the HMG co co pathway that you mentioned the cholesterol synthesis with which also is important for the synthesis of uh ubiquinol y right this is is an important or cutan as I should probably call it um this is important for mitochondrial function I mean it's necessary for mitochondrial function for transferring electron electrons across the electron transport chain which is essentially coupling the oxygen we breathe with the food that we eat to make energy”
There is no justification for taking unnecessary risks with statins in high-risk individuals (those with APOE4 or family history of neurodegeneration) when alternative lipid-lowering drugs (PCSK9 inhibitors, ezetimibe, bempedoic acid) are available that do not affect brain cholesterol synthesis.
“it makes no sense to get all to take unnecessary risks with Statin in a in a higher risk individual when we have these other tools we have as I we talked about or we will talk about aetam pcsk9 Inhibitors bidic acid these are unbelievable tools that have no bearing on brain cholesterol synthesis”
This is an example of the desmosterol problem: you can lower cholesterol but if you're raising desmosterol too much, it can become more atherogenic than cholesterol itself, illustrating the danger of targeting single biomarkers without understanding upstream and downstream metabolic consequences.
“so this is an it's sort of an example of that dmol point earlier where you can lower cholesterol but if you're really raising dmol too much it can be more atherogenic than cholesterol in the first place so dmol has shown up twice today it showed up in a good sense and in a bad sense so too much of it”
The top-10 most important metric for patient health is the workload (watts on a bike or METs on a treadmill) at which lactate stays below ~2 mmol/L, as this reflects the maximum capacity of mitochondria to handle cellular ATP demand.
“we use this test with patients this is one of the most important metrics we care about literally it would be in the top 10 things we care about for our patients which is how many watts can you produce on a bike or how many Mets can you exercise at on a treadmill or whatever vehicle you're using while keeping lactate below about 2 m 2 Millo is about the threshold Beyond which you are now shifting away from the maximum capacity of the mitochondria”
Metformin may not necessarily be harmful for all patients, but speaker does not believe in metformin as a geroprotective agent because of its mitochondrial toxicity; metformin is appropriate for diabetics who cannot exercise enough and cannot achieve energy balance, but not for healthy individuals like the speakers.
“so all things equal their lactate is just getting higher to me by the way I don't know if that's necessarily harmful I don't think it's a good idea which is why I don't believe in metformin as a geroprotective agent I think metformin is a good drug for someone who's diabetic if they can't exercise enough and they can't get into energy balance um but I don't think Matt foran is a great drug for someone like you or someone like me”
Tom DePriest recently gave an internal presentation (months in preparation) examining the relationship between statin use and desmosterol levels in relation to dementia risk, suggesting this is an important emerging area of personalized medicine analysis
“this is something that we spend a an awful lot of time looking at in our practice and and actually just last week Tom dpring um did gave us an internal presentation that was so incredible um it was months in the making uh looking at the relationship between Statin use and desol levels and dementia risk”
If statins have an impact on mitochondrial function through CoQ10 depletion, that impact is very small based on the best measurement available for mitochondrial function, which is Zone 2 testing with lactate production measurement.
“if there is an impact on mitochondrial function with stattin use it's very small based on what I consider to be the single best measurement we have to measure mitochondrial function which is Zone 2 testing with lactate production”
Atorvastatin has the highest incidence of diabetes among statins, but this may be attributable to atorvastatin being the most widely prescribed statin, confounding the assessment of whether it has intrinsically higher diabetogenicity.
“the highest incidence of diabetes is probably with atorvastatin um but that might also be because atorvastatin is the most widely used like I don't we we basically first of all there's only four statins that I think are even worth prescribing these days maybe only three”
Dexlosterol and lathosterol are difficult to measure in standard clinical laboratories; Boston Heart is a lab that measures these metabolites, making comprehensive statin safety surveillance inaccessible in standard-of-care settings.
“are these measure you can measure these on a like they're very difficult to measure in most Labs we use a lab that measures them so we measure desol and and lathosterol in every patient with every blood drop unfortunately this is not standard of care most Labs can't measure this Boston Hart does that's why we use Boston”
Rhonda notes that many patients with APOE4 status are prescribed statins by their physicians without awareness of their genetic risk, because most physicians do not personalize lipid management based on genetic status
“Peter aren't people that have an apoe4 Lil more likely to be prescribed statins based on their their LDL particle number by their physician because a physician doesn't look at their sure none of this this isn't personalized it's not medicine 3.0 right”
The speaker's clinical practice assumes equal risk of side effects across all statins at the individual patient level, and routinely monitors all statin-treated patients for all known side effects regardless of which specific statin is used.
“I treat them all equally in terms of risk in other words I would assume anytime you put somebody on a stat and you should be looking for any of the side effects and I don't particularly because again at the you might say at the population level it's different but at the individual level who cares you it's it's either one or zero you're going to get it or you're not”
Metformin is a mitochondrial toxin that impairs mitochondrial complex I, resulting in measurable decrements in Zone 2 lactate performance, reduced Zone 2 output, and elevated baseline resting lactate levels.
“metformin which is a mitochondrial toxin right metformin impairs complex one of the mitochondria we immediately see a change in the lactate performance curve of an individual on Metformin we see a complete reduction in their Zone 2 output they hit that lact of 2 much sooner we also see an increase not big but significant meaning clinically significant in their fasting resting lactate level”
The speaker describes this personalized, desmosterol-guided approach to statin management as 'personalized medicine 3.0,' requiring individual patient monitoring rather than population-level treatment algorithms
“this is you know kind of what I would describe as personalized medicine medicine 3.0 at its finest which is you have to treat every patient IND indvidually”
Importantly, it is not established that CoQ10 supplementation's hypothetical benefits translate into actual improvements in mitochondrial function, which is the mechanistic gap between theory and evidence.
“it's not clear to me that that effect translates to a functional deficit in the mitochondria”
The distinction between a side effect and a direct effect of statins on mitochondria is that a direct effect would be an unavoidable consequence of the drug's mechanism, whereas a side effect might be incidental or dose-dependent.
“I remember saying oh so statens have a side effect of targeting mitochondria and he said to me no it's a direct effect”
Data suggest that very low dexlosterol levels are associated with increased risk of Alzheimer's disease and non-Alzheimer's dementia, necessitating careful statin dosing in patients with APOE4 genotype or family history of neurodegeneration.
“enough data suggesting that if desol levels are very low the risk of AD does indeed go up and the risk of dementia Beyond ad goes up so this is you know kind of what I would describe as personalized medicine medicine 3.0 at its finest which is you have to treat every patient IND indvidually and we are doubly careful in patients with an apoe4 Gene Andor a family history”
A physician who is not willing to prescribe PCSK9 inhibitors to appropriate patients is 'a fool' given their superior safety and efficacy, though cost and insurance barriers often prevent prescription.
“at this point a doctor who doesn't who's not willing to prescribe a pcsk9 inhibitor just is a fool um so it's just a question of the the cost because unfortunately most insurance companies will not cover it”
If cost were not a constraint, the speaker would likely prescribe PCSK9 inhibitors, ezetimibe, and bempedoic acid in combination or preference to statins, and expects this to become standard eventually once prices decline
“so again we'd have every look honestly at this point like if money were were no issue you'd probably just be on PCS can9 Inhibitors aetam and um and bempedoic acid I mean eventually we'll get there right yeah yeah they just have to come down in price”
Statins were first developed in the mid-to-late 1980s, around 1987, and represented the real turning point in pharmacologic treatment of atherosclerotic cardiovascular disease.
“it really wasn't until the mid to late 80s probably I think 1987 if my memory serves me correctly that the first Statin came to be developed and that was the real turning point in um basically the uh you know pharmacologic tool that became valuable against ascvd”
Early statin generations (first, second, and third generation from the 1980s era) are no longer in clinical use today because their side effect profiles were harsh compared to modern statins.
“the first second and third generation Statin of that era are no longer in use today because their side effect profile was very harsh relative to what we can do today”