Cholesterol might be the most argued-about molecule on the internet. One camp says it built the modern epidemic of heart disease. Another says it has been framed. Both camps quote studies. So instead of quoting a study, this article is going to walk the entire staircase of evidence, from single molecules at the bottom to global scientific consensus at the top, and let you watch every step point in the same direction.

By the end, you will know what a micelle is, what ApoB means, why your arteries cannot feel themselves being damaged, why nearly every major scientific body on Earth, across multiple countries and specialties, landed in the same place, and why the most popular arguments against all of this, built from true facts, still reach the wrong conclusion.

The climb covers every level of the evidence pyramid, and it is worth knowing what each one can and cannot tell you. At the base is cell biology: how cholesterol is absorbed, how the liver regulates it, and how cholesterol-carrying particles get trapped in the artery wall. Above that come animal experiments, then genetic natural experiments called Mendelian randomization, in which people born with genes that set their cholesterol higher or lower serve as a lifelong trial. Next are population studies that follow hundreds of thousands of people for decades, then controlled feeding trials in which researchers supply every bite, then randomized trials that count actual heart attacks, strokes, and deaths, then imaging trials that watch plaque grow or shrink, and finally the systematic reviews, consensus statements, and guidelines built from all of it.

The article also looks past cholesterol as a molecule to the foods that carry it and the foods that lower it: eggs, dairy, coconut oil, oats and other whole grains, and unprocessed animal foods, including whether fresh, lean red meat is any safer than processed meat once the fat is trimmed and the additives are gone. It takes up the questions people most often raise, from statin side effects and very low LDL to whether people with high cholesterol live longer and what short-term keto results do and do not show, and it is honest about what remains unsettled, from lipoprotein(a), a largely inherited particle diet can barely touch, to the live debates scientists are still having.

Your Body Already Makes Every Bit of Cholesterol It Needs

What does cholesterol actually do in the body?

Cholesterol itself is not a villain. It is one of the most important molecules you own. It stiffens and stabilizes the membrane of every cell, it is the raw material for testosterone, estrogen, cortisol, and vitamin D, it forms the bile acids that digest your food, and it wraps your nerves in insulation called myelin.

Does your body make its own cholesterol?

It is so important, in fact, that your body refuses to outsource it, even though making it is expensive: the pathway runs through more than 30 biochemical steps. Essentially every cell with a nucleus can manufacture its own cholesterol on demand. The telling exception is the mature red blood cell, which has no nucleus and cannot make cholesterol at all, and yet, as the HDL section below explains, it turns out to be an active player in moving cholesterol around the bloodstream. Your brain is the most dramatic example: it holds roughly a quarter of all the cholesterol in your body, and it makes every bit of it locally, because the blood-brain barrier blocks cholesterol arriving from the bloodstream. Not one molecule of the cholesterol in your brain came from an egg. Your liver, meanwhile, acts as the central bank of the whole system, manufacturing a large share, packaging it for shipment, recalling it from circulation, and converting the excess into bile.

How much cholesterol do you need to eat?

Add it up and your body synthesizes a little under a gram of cholesterol a day, about 840 to 910 milligrams for a 70-kilogram adult by the National Academies’ estimate, several times more than most people eat. Which is why the National Academies (the Institute of Medicine, in its official Dietary Reference Intakes) concluded there is no dietary requirement for cholesterol at all, and advised keeping intake as low as possible within a nutritionally adequate diet. That single sentence is the anchor for everything that follows: the question was never whether you need dietary cholesterol. You do not. The question is what happens when you eat it anyway.

What is the optimal LDL cholesterol level?

There is a companion fact worth holding onto here. The LDL level that healthy human newborns, contemporary hunter-gatherers, and free-living primates all share, groups the review describes as free of atherosclerosis, sits around 50 to 70 milligrams per deciliter, roughly half the American average when the review was written. A landmark review in the Journal of the American College of Cardiology made the case directly in its title: optimal LDL is 50 to 70, and that range is not some aggressive medical target but simply the physiologically normal level our species evolved with. Today’s “normal,” in other words, is not the same thing as natural.

Two findings show what that gap looks like in practice. The PESA study scanned the arteries of 1,779 middle-aged adults, averaging 45 years old, who had none of the conventional risk factors: no smoking, and blood pressure, blood sugar, and cholesterol all in the accepted range, including LDL under 160. Half of them, 49.7 percent, already had plaque or coronary calcium somewhere in their arteries, and LDL was independently tied to how much, even at levels currently considered normal: each 10 points of LDL raised the odds by 14 to 18 percent. It was a one-time snapshot, not a trial, but it found disease quietly underway in the people conventionally labeled lowest risk. At the other end of the disease, a registry of 136,905 hospitalizations for coronary artery disease at 541 hospitals in the American Heart Association’s Get With The Guidelines program found that almost half of patients arrived with an LDL under 100, measured within the first 24 hours of admission, long treated as a reassuring number, and only 17.6 percent under 70. Only about a fifth of them had been taking cholesterol-lowering medication beforehand.

A 2026 study in the New England Journal of Medicine, called REACT, took the question to the general population. It enrolled 16,808 adults aged 18 to 70 in Denmark and Spain, none with known cardiovascular disease, and imaged the arteries of the neck and thigh with three-dimensional ultrasound and the coronary arteries with CT angiography. Plaque was present in 57.1 percent of them. Among the youngest, aged 18 to 29, it was already present in 8.7 percent of the men and 6.7 percent of the women, and plaque volume climbed steeply with age. The study, funded by the Novo Nordisk Foundation, reports where disease is found, not what caused it or what it will do next, but its message is hard to miss: atherosclerosis is common, and it is silent for decades before anyone looks or feels like a heart patient.

What Happens When You Eat It Anyway

How is dietary cholesterol absorbed?

Dietary cholesterol is chewed and swallowed, passing into the small intestine, where bile acids released from the gallbladder dissolve it into tiny water-soluble packages called micelles. Think of a micelle as a soap bubble built to smuggle grease through water. These micelles ferry the cholesterol to the intestinal wall, where a transport protein called NPC1L1 actively pulls it inside the intestinal cells. NPC1L1 is essentially a customs gate for cholesterol, and it is worth remembering that name, because a prescription drug, ezetimibe, exists just to block it.

Inside the cell, the cholesterol is chemically modified into a storage form called a cholesterol ester and loaded, along with dietary fat, into a massive transport vehicle called a chylomicron. A chylomicron is too large to squeeze into ordinary blood capillaries, so it exits through the porous lymphatic vessels called lacteals, rides the lymphatic system upward, and finally drains into your bloodstream through the thoracic duct near your neck. Within hours of a meal, cargo ships full of dietary fat and cholesterol are sailing through your arteries.

Why do some people absorb more cholesterol than others?

Two honest details make this picture complete. First, the pool of cholesterol sitting in your gut is not only what you ate. Your own liver dumps roughly a gram of cholesterol per day into the intestine through bile, typically more than food contributes. Second, how much of that pooled cholesterol gets absorbed varies from person to person. Most people absorb somewhere between 40 and 60 percent of the cholesterol they eat, and the National Academies attribute part of that variation to genes. Some people simply absorb more than others. Hold that thought, because absorption turns up again in one of the stranger findings in recent nutrition research.

The Feedback Loop: How Eating Cholesterol Changes Your Blood

What is a lipoprotein, and how many kinds are there?

A chylomicron is the largest member of a family of particles called lipoproteins, built because fat and cholesterol do not dissolve in blood and need a vehicle to travel in. After an enzyme called lipoprotein lipase strips most of the triglyceride out of a chylomicron, the shrunken remnant is small enough to slip into places a full chylomicron cannot, including the artery wall. The liver runs a parallel assembly line: it exports its own triglyceride-rich particle called VLDL, which sheds triglyceride in circulation and shrinks through an intermediate stage called IDL on its way to becoming LDL, the smallest and most cholesterol-dense member of that lineage. HDL is a different system entirely, built around a different structural protein and generally working in reverse, carrying cholesterol back to the liver rather than delivering it outward.

One mechanical fact ties the rest of this family together, and it is why a single blood measurement called ApoB has become such a useful number. Chylomicron remnants, VLDL, VLDL remnants, IDL, LDL, and Lp(a), every one of them, carry exactly one copy of a structural protein called ApoB on their surface, the shorter ApoB-48 on the chylomicron-derived particles, the full-length ApoB-100 on the rest. HDL does not carry ApoB at all. Counting ApoB in a blood sample therefore counts every particle capable of getting stuck in an artery wall in one number, regardless of its size or how much cholesterol it happens to be carrying at that moment, while a standard LDL cholesterol test measures cholesterol rather than particles and can miss risk sitting in VLDL and remnants. Neither number separates Lp(a) from ordinary LDL, since the cholesterol in Lp(a) is counted inside the LDL measurement, which is why Lp(a) needs its own test, as the Lp(a) section below explains. Our post on how a trapped particle becomes a plaque walks through what each of these particles does once it reaches the artery wall, and introduces the rest of the apolipoprotein family beyond ApoB.

How does eating cholesterol raise blood cholesterol?

When dietary cholesterol reaches the bloodstream inside chylomicrons, and those chylomicrons are stripped down to remnants, the remnants are taken up by the liver, and the liver’s internal cholesterol pool fills up. The liver responds the way any warehouse responds to a full inventory: it cuts back. It throttles its own cholesterol production, which is why the net rise in blood cholesterol is usually modest, and it also displays fewer LDL receptors on its surface, which is why the rise is not zero and why it shows up in your bloodwork.

LDL receptors are the vacuum cleaners of the bloodstream. Their discovery won Michael Brown and Joseph Goldstein the 1985 Nobel Prize, and it remains the single most important concept in this entire subject. When the liver displays fewer receptors, LDL particles, the so-called bad cholesterol, are cleared more slowly and circulate longer. The longer a particle drifts, the more time it has to be chemically altered and oxidized into forms your immune system treats as a threat. On top of that, a cholesterol-loaded liver packs its outgoing VLDL particles with extra cholesterol, and those particles degrade in circulation into cholesterol-heavy LDL and remnant lipoproteins, exactly the species best equipped to damage an artery wall.

How much does 100 mg of dietary cholesterol raise LDL?

Now for calibration, because this is where honest science lives. In controlled feeding studies, each additional 100 milligrams of dietary cholesterol per day raises LDL by only a few points on average, according to a meta-regression that was, notably, funded by the American Egg Board. Averages hide the spread: some people barely budge, while the hyper-responders and hyper-absorbers climb sharply. And in the real world, dietary cholesterol almost never travels alone. The foods that carry it, meat, eggs, butter, cheese, are usually the same foods rich in saturated fat, which also suppresses LDL receptors and raises LDL in its own right. The package deal is what shows up in your bloodwork. How the two food kingdoms package their fat so differently, and why the animal-dominant fatty acids are the ones that suppress those receptors, is the whole subject of a companion piece on animal fat, plant fat, and the journey from your plate to your arteries.

Does blocking cholesterol absorption lower LDL on a keto diet?

If you want a hint of how large the intestinal side of this system can loom, a 2026 case series in Lipids in Health and Disease provides one, and in the interest of full disclosure it comes from the same low-carb research circle behind the retracted KETO-CTA study discussed later in this article. Patients whose LDL had skyrocketed on ketogenic diets were given ezetimibe, the drug that blocks the NPC1L1 customs gate. Their LDL fell by a median of 53 percent, and by about 50 percent among the twelve taking ezetimibe alone, against the roughly 18 percent the drug typically achieves. Because these patients were chosen precisely for their large responses, that gap cannot be read as a measured effect. Nor did they uniformly look like hyper-absorbers on blood markers of sterol absorption, which is why the researchers proposed instead that the diet itself had reprogrammed cholesterol handling. With only 14 patients and no comparison group, it is a hypothesis to test rather than a settled effect. But the finding points at the gate itself: in these patients, blocking it did far more than usual, whatever the reason turns out to be, and most of what passes through that gate is the cholesterol your own bile recycles, not what was on the plate.

How Plaque Actually Forms

What is ApoB and why does it matter?

Most explanations of heart disease skip the crime scene. Here is what actually happens, and it is worth reading twice.

Atherosclerosis begins when ApoB-containing lipoproteins, which include LDL, remnants, and their relatives, slip across the endothelium, the one-cell-thick lining of your artery, into the inner arterial wall. ApoB is the large protein wrapped around each of these particles, one copy per particle, like a shipping label. On that protein sit positively charged amino acids, lysine and arginine, and the arterial wall’s structural mesh is built of proteoglycans carrying intensely negatively charged sulfate chains. Positive meets negative, and the two lock together in an electrostatic bond. The particle is trapped.

Why do cholesterol particles get stuck in artery walls?

This is called the response-to-retention model, and it is not a metaphor. In 2002, researchers published one of the most elegant experiments in the history of this field: they genetically engineered mice whose LDL could not bind proteoglycans. Cholesterol levels were similar across most of the groups, and where particle levels differed, the researchers compared atherosclerosis at matched levels. Either way, the mice carrying binding-defective LDL developed significantly less atherosclerosis. Less trapping, less disease. The trap is the disease.

What is a foam cell, and are there really cholesterol crystals in plaque?

Once trapped, the particle oxidizes, and the immune system flags it as toxic debris. Monocytes crawl into the wall and mature into macrophages, garbage-collector cells that gorge on the damaged particles until they become bloated foam cells. Foam cells eventually die and rupture, dumping their fatty cargo into a growing pool of dead material called the necrotic core. Inside overloaded macrophages, appearing even in early lesions, and later in that core, cholesterol concentrates until it precipitates into actual sharp-edged crystals, which trip an internal inflammatory alarm called the NLRP3 inflammasome and physically damage the tissue around them. Pathologists looking at plaque under the microscope literally see the needle-shaped voids those crystals leave behind. That is what a heart attack is made of.

This is the short version. Our post on how a trapped particle becomes a plaque follows the process step by step, from the first snag in the artery wall to the thinning cap that finally tears, and goes deeper on what this post only touches: why plaque forms in arteries and not veins, how smoking and a single fatty meal affect the artery lining, how sugar glycates the particles, and how iron, much of it from bleeding inside the plaque, fuels oxidation inside the wall.

A Possible Accelerator: TMAO

Does TMAO from meat and eggs cause heart disease?

There is a second player worth mentioning honestly, because it comes up constantly and deserves neither hype nor dismissal. When gut bacteria digest carnitine, concentrated in red meat, and choline, concentrated in eggs, they produce a compound the liver converts into trimethylamine-N-oxide, or TMAO. In large prospective studies, people with higher blood TMAO have had more cardiovascular events and deaths, and the pooled associations are not trivial. The mechanism proposed from animal work is that TMAO impairs the reverse-cholesterol-transport that clears the wall and nudges the whole trapping process along.

Why does fish raise TMAO more than meat?

The honest framing is that TMAO looks like an accelerator, not the engine. Two facts keep it in its lane. First, when researchers use genetics to test whether TMAO actually causes disease rather than merely marking it, the evidence is still thin, and at least one such study found no evidence of a causal effect on ischemic stroke, which is what you would expect if TMAO is partly a passenger rather than a driver. Second, and awkwardly for the simple story, fish raises TMAO far more than meat or eggs do, because fish carries TMAO ready-made, so it is absorbed without any bacterial step. In a crossover feeding trial, a fish meal produced 46 to 62 times more TMAO in blood and urine than eggs, beef, or a fruit control, and those authors describe fish as well known for its cardioprotective attributes. That does not put TMAO in the clear. What it shows is that a food’s TMAO load has to be weighed against everything else that food carries, and fish carries the long-chain omega-3 fats EPA and DHA, which the Academy of Nutrition and Dietetics notes come from fish and other seafood. Those same researchers argue that caution is warranted before restricting animal foods on the strength of their TMAO-raising alone. So TMAO belongs in the picture as a plausible contributor and a useful marker, while ApoB remains the thing actually building the plaque. It is a footnote to the mechanism, not a rewrite of it.

Three Ways a Cholesterol-Heavy Diet Loads the Dice

Everything above compresses into three mechanisms.

Does dietary cholesterol raise your ApoB particle count?

More dietary cholesterol and saturated fat means fewer liver LDL receptors, which means more ApoB particles circulating at any moment. Plaque formation is a numbers game played every minute of your life: the more particles drifting past the wall, the more get in, and the more get trapped. This is why ApoB, which counts the particles themselves, is the better measure, as the HDL section below shows.

Does dietary cholesterol make LDL particles more likely to oxidize?

Particles that linger get modified and oxidized, becoming stickier to the immune system. One clarification the internet usually gets wrong: the notorious small, dense LDL pattern is driven mostly by high triglycerides and insulin resistance, the territory of refined carbohydrates and excess calories, rather than by dietary cholesterol itself. What a cholesterol-rich diet reliably does is raise the particle number and enrich the remnants, which is damaging enough.

What is remnant cholesterol?

Absorbed dietary cholesterol leaves the intestine packed inside chylomicrons, and as lipoprotein lipase strips out their triglyceride in the bloodstream, that cholesterol stays behind in the shrunken particles left over, called chylomicron remnants. Each remnant carries an ApoB label, is small enough to enter the arterial wall, and hauls a far larger cholesterol payload per particle than a standard LDL. When a remnant gets trapped, it delivers a truckload where LDL delivers a suitcase. Remnant cholesterol, as the field uses the term, means the cholesterol riding in all of these triglyceride-rich particles, the liver’s VLDL remnants as well as the intestine’s chylomicron remnants. This is not a footnote: large genetic studies from Copenhagen, covering more than 73,000 people, found that gene variants raising remnant cholesterol while barely moving LDL still raised heart disease risk, with a causal effect of roughly 2.8-fold per 1 mmol/L that did not run through low HDL, which is one reason a fasting lipid panel that ignores them understates risk. The genetics establish that remnants are causal; the dietary link runs through the cholesterol-enriched VLDL described in the feedback section above.

A Word About HDL, Because Someone Will Bring It Up

Is high HDL actually protective?

High-cholesterol, high-fat diets often raise HDL, the so-called good cholesterol, and people wave that number like a shield. The best modern evidence says: put the shield down. Large genetic studies (a method called Mendelian randomization, which the genetics section below explains) show that several gene variants that raise HDL from birth do not lower heart attack risk, and drug after drug designed to raise HDL failed to deliver the expected protection; the one large trial that did show modest benefit produced it in proportion to the drug’s LDL lowering, not its HDL raising. HDL that rises under a saturated-fat-heavy, inflammatory dietary pattern can become protein-depleted, oxidized, and functionally lazy, losing the very anti-inflammatory, cholesterol-clearing behavior that made HDL famous.

ApoB or LDL-C: which number should you trust?

HDL is a system, not a score. ApoB is the score. And when your standard LDL-C number and your ApoB number disagree, which happens most in people with high triglycerides or diabetes, the outcome studies say to trust ApoB, because it counts the actual particles knocking on the artery wall. Recent studies put numbers on that disagreement. In a nationally representative sample of 12,688 American adults, people with an LDL cholesterol of 100 mg/dL had ApoB levels ranging from 66 to 99 mg/dL across the middle 95 percent of the population, and even metabolically healthy people could carry more particles than their LDL suggested, which led the authors to call the practice of testing ApoB only in people with high triglycerides too narrow. A 2025 systematic review of 15 discordance studies covering 593,354 people found, according to its abstract, that ApoB predicted cardiovascular risk better than LDL cholesterol in all 9 studies that compared the two directly, and better than non-HDL cholesterol in 7 of 9. And among 389,529 people in the UK Biobank who were not taking cholesterol-lowering drugs, ApoB, non-HDL cholesterol, and triglycerides each predicted heart attacks on their own, but when all were tested together, only ApoB still did. Several of that study’s authors report research grants from drug companies.

Do red blood cells carry cholesterol?

Lipoproteins are not the only cholesterol carriers in your blood. Mature red blood cells have no nucleus and no internal membranes, so they lack the means to make cholesterol of their own, yet their outer membrane holds a large amount of free cholesterol. As recently as 2013, researchers still described red cells as simply exchanging that cholesterol with the surrounding plasma through nonspecific diffusion, a passive drift in whichever direction the gradient happened to point.

A 2020 study in the Journal of Lipid Research found something more interesting. When researchers incubated human red blood cells with the donors’ own plasma or with isolated lipoproteins, cholesterol moved in a net direction: out of the red cells into HDL, and out of LDL into the red cells. In effect, the red cells took cholesterol in from LDL and passed it on to HDL, the direction of reverse cholesterol transport. And the exchange did not behave like passive drift. It could be saturated, and it depended on temperature, energy, and time, the signature of an active process. The obvious suspect, a cholesterol pump called ABCA1 that red cells turned out to carry, was not responsible: silencing or chemically blocking it made no difference, and red cells from mice lacking it moved cholesterol just as well as normal ones. The authors conclude that red blood cells actively take part in cholesterol transport in the blood, but which proteins actually run that traffic is still unknown.

A separate line of evidence ties red cells and cholesterol together at the scale of whole populations. In a 2013 analysis of 4,469 American adults from the National Health and Nutrition Examination Survey, higher non-HDL cholesterol, the cholesterol riding in all the ApoB-carrying particles, was associated with higher red blood cell counts, hematocrit, and hemoglobin, and most strongly with platelet counts: about 258,000 per microliter in the lowest quarter of non-HDL cholesterol against 281,000 in the highest, after accounting for age, sex, race and ethnicity, smoking, body mass index, folate, and C-reactive protein. The pattern held in a second, independent sample of 5,318 adults. That is a correlation, not a cause, and the authors frame it as a reason to study how these three parts of the blood interact, since all three bear on cardiovascular risk. Red cells matter inside plaques too: when blood leaks into an advanced plaque, the cholesterol in red cell membranes is deposited into its core, a process our post on how a trapped particle becomes a plaque covers in its iron section.

One Particle Diet Can Barely Touch: Lipoprotein(a)

What is Lp(a)?

Honesty cuts both ways, so here is the cholesterol risk factor that no salad will fix. Lipoprotein(a), written Lp(a) and said “L-P-little-a,” is an LDL-like particle carrying its own ApoB plus an extra tangled protein that makes it especially prone to lodging in artery walls and encouraging clots. It is an independent, causal driver of both heart attacks and calcific narrowing of the aortic valve.

Can diet lower Lp(a)?

Two facts make it important for a piece like this. First, it is common: roughly one in five people worldwide carries an elevated level. Second, and unlike almost everything else in this article, your level is set overwhelmingly by the genes you were born with. Diet and exercise barely move it, and the one eating pattern reported to lower it modestly, by 10 to 15 percent, is low-carbohydrate, high-fat eating. That is not the argument for eating animals it might look like. A low-carbohydrate, high-fat diet can be built entirely from plants, on nuts, seeds, avocados, olives, and their oils, so nothing about that finding requires animal food. More to the point, as the trial results below show, lowering Lp(a) and preventing heart attacks turned out to be two different things: the first drug to lower it in a large outcome trial did not reduce cardiovascular events. Even the statins that lower ordinary LDL do essentially nothing to lower it, and a meta-analysis (that link explains how to read one) found they modestly raise it.

Do the new Lp(a) drugs work?

A class of drugs built specifically to lower Lp(a) has been in late-stage testing for years, and the first cardiovascular-outcome results arrived on September 4, 2026, with a sobering answer. The Lp(a)HORIZON trial tested pelacarsen in 8,323 patients with established cardiovascular disease and high Lp(a). The drug lowered Lp(a), and it did not reduce cardiovascular events. The trial missed its primary endpoint.

It is worth being careful about what that does and does not mean, in both directions. It does not overturn the genetic case that Lp(a) is causal, which rests on people who carried high levels from birth across entire lifetimes, not on a few years of treatment in patients already receiving optimal cardiovascular care. Lifelong exposure and late intervention are different questions, and as the genetics section below shows, the same gap exists for ordinary LDL, where lifelong genetic lowering buys roughly three times the protection of a pill started late. But it does mean the therapeutic question is genuinely unresolved. The full Lp(a)HORIZON data have yet to be presented. For now, nobody can tell you that lowering Lp(a) with a drug will lower your risk.

Should you get your Lp(a) tested?

So Lp(a) is the honest asterisk on the whole plant-forward argument. Eating well remains the highest-leverage thing most people can do for the particles diet can move, which is the overwhelming majority of the risk for the overwhelming majority of people. But a person can do everything right and still carry elevated Lp(a), which is exactly why it is worth measuring once in your life, since the number almost never changes. That is now formal guidance rather than a suggestion: the 2026 ACC/AHA multisociety dyslipidemia guideline recommends measuring Lp(a) at least once in every adult, a new recommendation for the American guideline and one European guidelines have made since 2019. And knowing the number still matters even though the first drug aimed at it failed, because an elevated Lp(a) tells you your baseline risk is higher than your other numbers suggest, which is a reason to be more aggressive about every risk factor you can actually move, starting with ApoB and the food on your plate.

Climbing the Whole Pyramid of Evidence

For a closer look at how to read each kind of study, see our series of educational posts explaining how to interpret science by analyzing actual studies across the hierarchy of evidence.

Science ranks its evidence. Mechanisms and lab work sit at the base, then animal experiments, then human observational studies, then randomized controlled trials, then systematic reviews and consensus statements built from all of it. A claim you can trust is one that survives at every level. Cholesterol’s role in heart disease is arguably the most thoroughly tested claim in all of medicine, so let’s climb.

Level 1: What does the cell biology show?

The base of the pyramid is everything you just read. The LDL receptor system (Brown and Goldstein, Nobel Prize, 1985). The discovery of the NPC1L1 absorption gate in 2004. The oxidation hypothesis, showing how modified LDL becomes immunologically toxic. The response-to-retention model of particles binding proteoglycans. Cholesterol crystals activating the NLRP3 inflammatory alarm. None of this is speculative. It is textbook cell biology, worked out in exhaustive detail over fifty years.

Level 2: What do the animal experiments show?

The proteoglycan-binding mouse experiment described above is the crown jewel: block the trapping, and far less disease develops at comparable particle levels. The broader animal literature goes back much further. In 1913, Nikolai Anitschkow fed cholesterol to rabbits and produced arterial lesions closely resembling human atherosclerosis, and in the century since, the pattern has held with blunt consistency. Raise blood LDL high enough, by diet or by genetics, and virtually every susceptible species tested develops arterial plaque, which regresses when those levels fall. Species that resist atherosclerosis are the species whose LDL stays low.

Level 3: What do genetics and Mendelian randomization show?

Here is where it gets beautiful. You cannot ethically randomize babies to a lifetime of high or low cholesterol. Nature already did.

People born with familial hypercholesterolemia, somewhere between 1 in 200 and 1 in 300, carry broken LDL receptor machinery. Their LDL runs high from birth, and untreated, many suffer heart attacks decades early. Children who inherit two broken copies are born with extreme LDL levels and almost universally develop atherosclerotic disease in childhood or early adolescence.

Now flip the coin. The gene PCSK9 first came to light in 2003, when researchers found that mutations in it cause an inherited form of high LDL. In 2006, researchers studying the same gene found people born with mutations that keep LDL modestly lower for life. A 28 percent lower LDL from birth came with an 88 percent reduction in coronary heart disease. Even a 15 percent lifelong reduction cut risk nearly in half. Then in 2014, geneticists found people born with one inactivated copy of NPC1L1, the cholesterol absorption gate itself. Their LDL ran only about 12 points lower, yet their heart disease risk was roughly cut in half. These people are, in effect, born on a low-absorption diet, and their arteries show it.

A 2015 study sharpened that point for anyone wondering whether the route matters. Among 108,376 people in 14 studies, common gene variants that slow cholesterol absorption through NPC1L1, the target of the drug ezetimibe, lowered LDL by 2.4 mg/dL and coronary risk by 4.8 percent. Variants that slow cholesterol production through HMGCR, the target of statins, lowered LDL by 2.9 mg/dL and risk by 5.3 percent. People who carried both had 5.8 mg/dL lower LDL and 10.8 percent lower risk. Per unit of LDL, the protection was the same whichever route produced it, which is what you would expect if the particles themselves, and not the pathway that set their number, do the damage.

This method, Mendelian randomization, uses randomly inherited genes as a lifelong experiment, largely immune to the confounding that plagues diet studies. Applied across hundreds of thousands of people, it shows that every unit of lifelong lower LDL delivers about three times the protection of the same reduction started late in life with a pill: in one pooled analysis of 312,321 people, each 39 mg/dL of lifelong lower LDL came with a 54.5 percent lower risk of coronary heart disease. The relationship is graded and log-linear, and the largest statin meta-analysis found no evidence of any threshold, within the range studied, below which lower LDL stops helping. Exposure over time is the disease. The lesson compresses into three words: lower, earlier, longer.

Time matters as much as level. A useful way to think about LDL is the way smoking is counted in pack-years: the dose multiplied by the years. The CARDIA study, which enrolled 4,958 adults aged 18 to 30 and followed them into middle age, found that the risk of cardiovascular events after age 40, from heart attacks and strokes to heart failure, depended on cumulative LDL exposure, and that the same total exposure carried more risk when it built up earlier in life rather than later. The study was funded by the National Heart, Lung, and Blood Institute.

Level 4: What do population studies show?

Zoom out to whole populations and the same picture appears.

Framingham, launched in 1948, put cholesterol on the map as a core risk factor. The MRFIT screening study then followed over 350,000 men and found the relationship between blood cholesterol and fatal heart disease is continuous and graded, with no safe plateau where risk levels off.

The Ni-Hon-San study answered the genetics objection: Japanese men who had moved to Hawaii and California ate far more animal protein, fat, and cholesterol than men who stayed in Japan, and had markedly higher blood cholesterol, with California slightly higher still. Heart attacks followed the same path: in a later report from the same study, the rate in Hawaii was double that in Japan, and California’s was nearly half again higher than Hawaii’s. Same genes, different environment, food included, different arteries.

Does heart disease really start in childhood?

Autopsy studies like Bogalusa delivered the timeline: fatty streaks and early plaques are already present in children and young adults who died of unrelated causes, and the extent of those lesions tracks their LDL and other risk factors. This disease starts decades before the first symptom, which is why pediatric guidelines from the National Heart, Lung, and Blood Institute strongly recommend checking every child’s cholesterol once between ages 9 and 11. If you are thinking about how this applies to a family, our guide to raising children on a plant-based diet covers the nutrition side in detail.

What do the longest-lived and lowest-heart-disease populations eat?

And then there are the Tsimane of the Bolivian Amazon, described in The Lancet in 2017 as having the lowest levels of coronary artery disease ever recorded in a studied population. Forager-farmers eating a diet around 14 percent fat, built overwhelmingly on plants (plantains, rice, corn, cassava, fruit) plus modest lean wild game and fish, their average LDL sits near 90 and their inflammation from infections is actually high, yet 85 percent of them showed zero coronary artery calcium. Compared with a large American cohort, the Tsimane reached a meaningful coronary calcium score about 28 years later in life. The authors credit a whole way of living, very low LDL alongside low blood pressure, low blood sugar, lean bodies, high physical activity, and almost no smoking, rather than diet alone. The broader question of what the longest-living people actually eat is worth its own treatment, as is the recurring claim that the Blue Zones research has been debunked, and the specific Saul Newman critique that gets cited for it.

Do plant-based eaters have less heart disease?

Meanwhile the Adventist Health Study-2, analyzing more than 73,000 North Americans, found vegetarians overall had about 12 percent lower total mortality than non-vegetarians. That finding is sometimes waved away on the grounds that Adventists differ from everyone else in more ways than diet, so it is worth knowing it holds outside that community. In the general middle-aged population of the ARIC study, higher adherence to plant-centered eating tracked with less incident cardiovascular disease and lower mortality, and across more than 200,000 people in the Harvard cohorts, a crucial distinction emerged: a plant-based diet built on whole foods predicted less coronary disease, while one built on refined grains, sweets, and sugary drinks predicted more. Plant-based is not a synonym for healthy, and the data say so. The Academy of Nutrition and Dietetics makes the same point in its 2025 position paper: adults with cardiometabolic risk factors on vegetarian or vegan diets “should still monitor their intake of saturated fats and added sugars,” because “restricting animal products alone may not mitigate risk.” Two of those Harvard cohorts, 84,628 women and 42,908 men followed for 24 to 30 years, also asked the question that matters most: replaced with what? Swapping 5 percent of calories from saturated fat for polyunsaturated fat, monounsaturated fat, or whole-grain carbohydrates was associated with 25, 15, and 9 percent lower coronary risk, while swapping it for refined starches and added sugars was not associated with any change in risk. The same Harvard cohorts, followed for up to 33 years across 221,054 adults, also isolated one specific swap: the highest butter eaters had about 15 percent higher total mortality, the highest plant-oil eaters about 16 percent lower, and replacing about 10 grams of butter a day, a little under a tablespoon, with plant oils was associated with about 17 percent lower total and cancer mortality. A 2023 analysis of roughly 120,000 adults found that every major healthy-eating index, scored independently, predicted lower total and cardiovascular mortality, and a Women’s Health Initiative analysis found the Portfolio pattern, the same cholesterol-lowering food combination tested in the feeding trials below, associated with fewer actual cardiovascular events over years of follow-up.

Observational studies can be confounded, which is exactly why we keep climbing.

Level 5: What do controlled feeding trials show?

Lock people in metabolic wards, control every bite, and blood lipids become predictable enough to write equations, which is literally what Ancel Keys and Mark Hegsted did in the 1960s. The modern summary is the Mensink and Katan meta-analysis of 60 controlled trials, later updated in an analysis commissioned by the World Health Organization: compared with carbohydrate, saturated fat raises LDL while polyunsaturated fat lowers it. Those trials deliberately held dietary cholesterol constant, so its separate and smaller effect comes from a different body of work, the meta-regression described earlier. This is not epidemiology. This is cause and effect measured in locked kitchens.

How much can food alone lower cholesterol?

Feeding trials also revealed how far food alone can go. The DASH trial dropped blood pressure with fruits, vegetables, and low-fat foods in eight weeks. The Portfolio Diet trial, in which participants were supplied their food but lived at home, went after LDL directly, stacking four plant tools (viscous fiber from oats, barley and psyllium, soy protein, plant sterols, and almonds) and cut LDL about 29 percent in four weeks, statistically indistinguishable from the 31 percent achieved by a starting dose of lovastatin running head-to-head in the same trial. The trial was partly funded by Unilever and the Almond Board of California, whose products were part of the diet. A meta-analysis of plant sterols and stanols shows intakes of 0.6 to 3.3 grams a day lower LDL by roughly 6 to 12 percent, and soluble fiber adds its own reliable reduction. Individual foods carry their own randomized evidence on top of these: oat beta-glucan at 3 grams a day or more lowers LDL by about 10 points across 28 trials, in an analysis supported by DSM Nutritional Products, and this effect is the basis of the oat health claims approved by the FDA, the European authorities, and Health Canada; soy protein and ground flaxseed each lower it a few points more in their own meta-analyses. Even blood vessels themselves respond: dietary nitrate from beets and greens produced sustained blood pressure drops of roughly 8 points systolic in randomized hypertensive patients.

Does unfiltered coffee raise cholesterol?

A quick aside that surprises people, since it shows the logic cuts both ways: not everything plant-based is automatically cholesterol-neutral. In a systematic review of food trials, unfiltered coffee, such as boiled coffee, was one of the few foods that raised LDL, by what the reviewers class as a moderate to large amount, roughly 8 points or more. The point is not to fear coffee but to notice that the science follows the molecules wherever they lead, not a plant-versus-animal slogan.

Do omega-3 supplements raise LDL cholesterol?

The same honesty applies to a supplement plenty of plant-based readers take. DHA, the long-chain omega-3 sold as algal oil, modestly raises LDL. In a randomized, placebo-controlled trial, 0.7 grams a day of DHA for three months raised LDL by about 7 percent and ApoB, the particle count, by about 3 percent, and a meta-analysis of algal DHA trials at a median of 1.68 grams a day found LDL up about 9 points alongside lower triglycerides and slightly higher HDL. It is partly offset because DHA also shifts LDL toward larger particles, but if your whole cardiovascular strategy is built on keeping LDL and ApoB low, which is the strategy this article defends, it is worth knowing. Two things keep this in proportion. The doses that produce the effect are several times higher than the 250 milligrams a day that plant-based clinicians typically suggest. And the underlying question of whether a healthy vegan needs the supplement at all is genuinely unsettled, with the large trials having failed to move hard outcomes in either direction, which is the subject of a companion piece on whether vegans need DHA and EPA.

Level 6: What do randomized trials with hard endpoints show?

The top interventional tier asks the only question that ultimately matters: fewer heart attacks and deaths, yes or no?

Drug trials proved the pathway end to end. The 4S trial in 1994 cut total mortality 30 percent by lowering LDL. The Cholesterol Treatment Trialists’ meta-analysis of 26 randomized trials and 170,000 people nailed the exchange rate: a little over a fifth fewer major vascular events for every 39-point drop in LDL, and that benefit repeats each year the lowering is maintained. IMPROVE-IT showed that ezetimibe, the drug blocking the same NPC1L1 gate your dietary cholesterol uses, reduced cardiovascular events, demonstrating in a randomized trial that the absorption route itself is causally connected to outcomes. FOURIER drove LDL down to a median of 30 with a PCSK9 blocker, and events still fell at those levels. ODYSSEY OUTCOMES, a second PCSK9-antibody trial in 18,924 patients with a recent acute coronary syndrome, such as a heart attack, cut the combined rate of coronary death, heart attack, ischemic stroke, and hospitalization for unstable angina from 11.1 to 9.5 percent over a median of 2.8 years. Deaths from any cause were also lower, 3.5 versus 4.1 percent, though under the trial’s prespecified testing order that result does not count as a formal finding. FOURIER and ODYSSEY OUTCOMES were both funded by the companies that make their drugs. And in 2023, CLEAR Outcomes added a fourth independent mechanism: bempedoic acid, an ATP citrate lyase inhibitor, lowered LDL and cut cardiovascular events in nearly 14,000 patients who could not tolerate statins. Four different drug mechanisms now, each poking the system in a different place, one common denominator: fewer ApoB particles, fewer heart attacks.

Two recent trials pushed into questions the older ones left open. VESALIUS-CV gave evolocumab or placebo to 12,257 people with atherosclerosis or diabetes who had never had a heart attack or stroke. Median LDL at 48 weeks was 45 on the drug and 109 on placebo, and over a median of 4.6 years the combined rate of coronary death, heart attack, or ischemic stroke fell by 25 percent (hazard ratio 0.75). Ez-PAVE, run in South Korea, tested a question on which guidelines had set targets despite limited evidence from randomized trials: for people who already have atherosclerotic disease, is an LDL target under 55 better than one under 70? Among 3,048 patients, the lower target cut the combined rate of cardiovascular death, heart attack, stroke, revascularization, or hospitalization for unstable angina by about a third over three years (hazard ratio 0.67), with similar safety in both groups. Both deserve their caveats: VESALIUS-CV was funded by Amgen, which makes evolocumab, and Ez-PAVE was open-label, with the drug company Yuhan among its funders.

LDL this low raised a fair question: earlier trials of PCSK9 inhibitors had prompted concern that the drugs, or the very low LDL they produce, might affect thinking and memory. The EBBINGHAUS study tested that directly inside the FOURIER trial, following 1,204 patients for a median of 19 months, and found no significant difference between evolocumab and placebo in executive function, working memory, episodic memory, or processing speed. It was funded by Amgen. On dementia, an observational Danish study of 132,585 people newly diagnosed with type 2 diabetes, designed to imitate a randomized trial, found that starting a statin within a year of diagnosis was associated with a 15 percent lower 10-year risk of dementia than not starting one within five years (95 percent confidence interval 9 to 21 percent). As an observational finding, that suggests a benefit rather than proving one, but it runs opposite to the fear.

Fear has costs of its own. Following 674,900 Danes aged 40 or older who started a statin between 1995 and 2010, researchers found that exposure to negative statin news stories was associated with 9 percent higher odds of stopping early (odds ratio 1.09), and people who stopped early went on to have a 26 percent higher risk of heart attack and an 18 percent higher risk of cardiovascular death than those who stayed on treatment. That, too, is observational, since people who quit may differ in other ways, but the direction is hard to ignore.

The PCSK9 story may eventually reach a one-time treatment. In a phase 1 study of 35 people, a single infusion of VERVE-102, a gene-editing therapy designed to durably switch off PCSK9 in the liver, lowered LDL by 9 percent at the lowest dose and 62 percent at the highest, with reductions that appeared durable through follow-up of at least a year in 15 participants. It is an early dosing and safety study, funded by the therapy’s developer, with no data yet on heart attacks or strokes. What it would do, if it works, is recreate on purpose, from the day of treatment onward, the permanently low LDL that people born with broken PCSK9 genes have had since birth.

Do diet trials show fewer heart attacks?

Diet trials with hard endpoints are harder to run, but they exist. The Los Angeles Veterans trial, 846 men randomly assigned and followed under double-blind conditions, saw significantly fewer combined heart attacks, sudden deaths, and strokes in the group given a diet high in vegetable oil and low in saturated fat and cholesterol, with dietary cholesterol cut from about 650 to about 365 milligrams a day. The trial was honest about its limits: the benefit held for men aged 54 to 65 but did not reach significance in the older stratum, heart attacks and sudden deaths alone did not reach significance, total deaths were similar in both groups, and the authors flagged a late excess of non-atherosclerotic deaths in the experimental group, from trauma and from carcinoma, as an open question. The Oslo Diet-Heart study and the Finnish Mental Hospital study, both replacing saturated fat with unsaturated fat, reduced coronary events. The Lyon Diet Heart Study put heart attack survivors on a Mediterranean pattern and saw recurrent cardiac events fall by roughly 70 to 76 percent. The early results were strong enough that the trial was stopped after its interim analysis. One detail cuts against a purely cholesterol-based reading, and it deserves mention: blood lipids stayed similar in both groups, so the protection in this trial apparently ran through something other than lower LDL. PREDIMED, one of the largest diet trials of recent decades, enrolled about 7,400 high-risk adults in Spain and, in its corrected 2018 analysis after irregularities in randomization at some sites, found a Mediterranean pattern with extra olive oil or nuts cut major cardiovascular events by about 30 percent. CORDIOPREV, a single-center Spanish trial, followed 1,002 people who already had coronary heart disease for seven years on either a Mediterranean diet or a low-fat diet. Major cardiovascular events were less common on the Mediterranean diet (adjusted hazard ratios 0.72 to 0.75), but the benefit was concentrated in men; among the 175 women, no difference was found. Its funders included two olive oil foundations alongside Spanish and European public money. And when the Cochrane Collaboration, the most methodologically conservative review body in medicine, reviewed 15 randomized trials covering 56,675 participants in 2020, including the unfavorable ones, reducing saturated fat cut combined cardiovascular events by about 17 percent in the 12 trials that reported them, with bigger cholesterol reductions producing bigger event reductions, exactly what the mechanism at Level 1 predicts. A 2026 review in the Annals of Internal Medicine, pooling 17 trials and 66,337 participants and grading its evidence as low to moderate certainty, reached a more cautious but compatible answer, according to its abstract: for people at low cardiovascular risk, reducing saturated fat made little or no difference over five years, while for people at high risk the benefits were large enough to matter, particularly for heart attacks when saturated fat was replaced with polyunsaturated fat (relative risk 0.75). The authors note the trials varied widely in how much they actually reduced saturated fat and in what replaced it.

What about the trials that failed?

And the honest tier includes the failures, because failures teach. The Women’s Health Initiative randomized nearly 49,000 women, about 19,500 of them assigned to a generically low-fat pattern, and saw no cardiovascular benefit: cutting total fat without targeting what replaced it lowered LDL only modestly and prevented nothing, so the replacement is everything. The recovered data from the Sydney and Minnesota experiments of the 1960s and 70s cut the other way: in Sydney, the group given safflower oil and safflower margarine in place of saturated fat actually died at higher rates, and in Minnesota the swap produced no mortality benefit. Neither result is clean. Sydney never recorded the trans fat content of its diets, and in Minnesota the comparison diet itself was loaded with trans fat from the common margarines of the day. And the great antioxidant-pill trials of the 1990s, ATBC and CARET, found that isolated beta-carotene capsules not only failed but increased lung cancer in smokers. The lesson repeats three times: the operative unit of nutrition is the whole food pattern, not an extracted nutrient in a capsule or an industrial substitution. The fuller version of that argument, including why the “seed oils are poison” claim fails at every rung of this same ladder, is laid out in our post on saturated fat, seed oils, and heart disease.

Level 7: Can arteries actually heal?

Dean Ornish’s Lifestyle Heart Trial randomized patients with established coronary disease to an intensive program built on a very-low-fat plant-based diet plus exercise and stress management. At one year, arteries in the lifestyle group had measurably widened while controls narrowed, and over five years the gap grew, with more cardiac events piling up in the control group. Caldwell Esselstyn’s plant-based case series points the same direction, and it is only fair to label it honestly: case series sit low on the pyramid, but they are consistent with everything above them. Meanwhile intravascular ultrasound trials of intensive drug therapy, ASTEROID and GLAGOV, showed plaque volume physically shrinking once LDL was pushed to around 60, and in GLAGOV’s case below 40. None of these isolates diet alone: Ornish’s program combined diet with exercise and stress management, Esselstyn’s work is a case series, and ASTEROID and GLAGOV were drug trials. We have not found a randomized trial that tested a vegan diet by itself and imaged plaque before and after. Multiple roads, one destination: get ApoB low enough, for long enough, and arteries stop getting worse and start getting better.

Level 8: What did the expert consensus panels conclude?

In 2017, the European Atherosclerosis Society convened its consensus panel to answer the question once and for all. They integrated over 200 studies covering more than 2 million participants and over 20 million person-years of follow-up, spanning genetics, epidemiology, and randomized trials. Their title was not hedged: “Low-density lipoproteins cause atherosclerotic cardiovascular disease.” Not associated with. Cause. The 2020 update deepened the mechanism and said it again. When separate methods with completely different weaknesses all land on the same answer, scientists call it consilience, and this is the cleanest example modern medicine has.

What the World’s Major Scientific Bodies Concluded

Walk the directory of major health and science organizations and notice how rare the disagreement is.

The National Academies (Institute of Medicine) set no requirement for dietary cholesterol and advised intake as low as possible within a nutritionally adequate diet, precisely because your body makes its own and the relationship between intake and heart disease risk is positive with no identified safe threshold.

Why did the 300 mg cholesterol limit disappear in 2015?

The Dietary Guidelines for Americans dropped its old 300-milligram daily cap in the 2015 cycle, a change endlessly misquoted as an exoneration. It helps to know that two different documents are involved. The advisory committee, a panel of scientists who spend years assembling the evidence, writes a scientific report. The government then writes the actual policy document. They are not the same thing, and the difference matters here.

The 2015 committee said the cap should not be carried forward, on the grounds that available evidence showed no appreciable relationship between dietary cholesterol and blood cholesterol, and that cholesterol was not a nutrient of concern for overconsumption. Two pieces of context rarely survive the screenshot. That phrasing overstated its own cited source, the 2013 review by the American Heart Association and the American College of Cardiology, which had actually concluded there was insufficient evidence, a weaker and different claim. And average American intake was already below 300 milligrams, so the number had largely stopped doing work.

Did the Dietary Guidelines say dietary cholesterol doesn’t matter?

The final Guidelines did not repeat the no-relationship line. They said explicitly that dropping the cap does not suggest dietary cholesterol is no longer important, and kept the National Academies’ advice to eat as little dietary cholesterol as possible while eating a healthy pattern, noting that the foods highest in it are generally the same foods highest in saturated fat. The 2020 to 2025 edition kept that language. And in 2019 the American Heart Association effectively corrected the committee’s wording, concluding that dietary cholesterol does raise LDL in a dose-dependent way, on the order of a couple of points per hundred milligrams under a linear model and more under nonlinear ones.

What did the 2025 scientific advisory committee actually conclude?

The 2025 Dietary Guidelines Advisory Committee, the scientific panel again, kept the saturated fat limit and reported that 82 percent of Americans aged one and older already exceed it. It also proposed reordering the protein foods to list beans, peas, and lentils first, followed by nuts, seeds, and soy, then seafood, and finally meats, poultry, and eggs. Two of its underlying systematic reviews are worth knowing about, because they are exactly the sort of thing that never makes it into a screenshot. The committee was the first to formally evaluate saturated fat at the level of foods rather than nutrients, and it graded as strong the finding that replacing butter with plant oils and spreads lowers LDL. It also found, at moderate strength, that swapping red and processed meat for plant proteins like beans, lentils, nuts, and soy is associated with less cardiovascular disease. Its dietary patterns review, graded strong, concluded that patterns higher in vegetables, fruits, legumes, nuts, whole grains, and unsaturated fats, and lower in red and processed meat, refined grains, and sugary foods, are associated with lower cardiovascular risk, consistently across racial and socioeconomic groups. One point of provenance worth stating plainly: the headline conclusion that swapping saturated fat for polyunsaturated fat reduces coronary events was graded strong by the 2020 committee, and the 2025 committee carried it forward rather than rerunning it.

What changed in the 2025 to 2030 Dietary Guidelines?

Then came the part worth watching closely. The 2025 to 2030 Guidelines released in January 2026 dropped the dietary cholesterol language entirely, listed animal protein foods first, starting with eggs, in place of the committee’s plant-first order, and floated butter and beef tallow as fat options, none of which came from the committee’s scientific report. What the political stage could not remove was the number that covers those same foods: saturated fat stays limited to under 10 percent of calories. The American College of Cardiology said the obvious thing out loud, that if you actually eat the newly recommended servings of animal protein and the proposed healthy fats, you will blow past that 10 percent limit, and outside nutrition researchers warned of contradictions between the guidelines and the scientific report they were supposed to rest on. At least one member of the advisory committee itself publicly objected that the accompanying food pyramid, with red meat and cheese at the top, runs against decades of evidence.

The administration published its own account of those choices in a document called The Scientific Foundation for the Dietary Guidelines for Americans, 2025 to 2030. It says the advisory committee’s report was framed through a health equity lens and describes its own replacement as an independent review, peer-reviewed through a process coordinated by the National Institutes of Health. It also includes a table marking which of the committee’s recommendations were adopted. By that table, the recommendations to emphasize beans, peas, and lentils while reducing red and processed meat, to list plant proteins first, to emphasize low-fat dairy and unsaturated fats, and to focus fat replacement on plant sources were not implemented, and two were marked as only partly implemented: limiting red and processed meat, and keeping saturated fat under 10 percent by replacing it with unsaturated fat, particularly polyunsaturated fat. The 10 percent cap itself survived; the committee’s emphasis on what should replace saturated fat did not survive in full. The document also argues that reliance on LDL cholesterol as a surrogate marker sustained confidence in the diet-heart hypothesis despite neutral or unfavorable trial results, a view at odds with the genetic and drug-trial evidence laid out in the pyramid above.

That is the real lesson about the guidelines, and it cuts against the way they are usually attacked. The weak link in 2026 was not the scientists but what happened to their report afterward, in a room where industry had a seat and the panel did not. When a document written under pressure to rehabilitate animal fat still cannot delete the saturated fat cap, that tells you where the evidence floor is.

Are the 2025 to 2030 Dietary Guidelines being challenged in court?

They are. On August 19, 2026, the Physicians Committee for Responsible Medicine sued the Department of Health and Human Services and the Department of Agriculture in federal court in Washington, D.C. The suit alleges that the agencies set aside the advisory committee’s report in favor of recommendations from a quickly assembled second panel, that eight of its nine members had ties to beef, dairy, egg, pork, or low-carbohydrate diet interests, and that this violated the Federal Advisory Committee Act. It asks the court to withdraw the guidelines. Those are allegations from a party to the case, not findings, and the outcome is not yet known. The same group has been down this road before, though: in 2000, a federal judge ruled that the Department of Agriculture had violated federal advisory committee and open-records laws by withholding information about the committee behind that year’s guidelines.

What do the AHA, WHO, and other health bodies say?

The American Heart Association’s 2017 Presidential Advisory concluded that replacing saturated fat with unsaturated fat lowers cardiovascular events on the order of what statins achieve, and its 2019 science advisory on dietary cholesterol recommended keeping intake low while acknowledging that healthy people can fit up to about an egg a day within an otherwise healthy pattern, with more caution for those with high LDL or diabetes. Its newest dietary guidance, a 2026 scientific statement that replaces its 2021 version, lays out nine features of a heart-healthy pattern. On protein it advises shifting from meat to plant sources such as legumes and nuts, eating fish and seafood regularly, choosing low-fat or fat-free dairy over full-fat, and, if red meat is desired, choosing lean cuts, avoiding processed forms, and limiting portion size. It also notes that eating patterns following its guidance are unlikely to exceed 10 percent of calories from saturated fat.

The American College of Cardiology and AHA cholesterol guidelines, and the European Society of Cardiology and European Atherosclerosis Society guidelines, treat LDL and ApoB as causal targets, under the principle that lower is better, and the 2026 ACC/AHA update frames risk around cumulative lifetime exposure.

The World Health Organization (2023) recommends holding saturated fat under 10 percent of energy and replacing it with polyunsaturated fats and plant foods, alongside its long-running push to eliminate trans fats.

The International Agency for Research on Cancer classified processed meat as carcinogenic to humans, its highest-certainty Group 1, and red meat as probably carcinogenic, and the World Cancer Research Fund advises eating little, if any, processed meat. The American Cancer Society’s 2020 guideline centers vegetables, fruits, whole grains, and legumes while limiting red and processed meat. That classification gets challenged regularly, and we took apart one of the most-shared versions of that challenge in our look at why a single subgroup number did not overturn the meat and bowel cancer link.

The American Diabetes Association’s 2026 Standards of Care list vegetarian and vegan eating patterns among the ones it most frequently recommends, alongside Mediterranean, DASH, low-fat, and carbohydrate-restricted patterns. It advises people with diabetes to eat more plant protein from nuts, seeds, and legumes to reduce cardiovascular risk, and to cook with avocado, canola, or olive oil in place of butter, coconut oil, and lard. The National Heart, Lung, and Blood Institute’s pediatric expert panel, in guidelines published in the American Academy of Pediatrics’ journal, strongly recommends universal childhood lipid screening because the disease begins in childhood.

The Academy of Nutrition and Dietetics approved a fresh position paper in January 2025, in effect through 2032: appropriately planned vegetarian and vegan dietary patterns can be nutritionally adequate and can offer long-term health benefits, including improvements in several outcomes tied to cardiometabolic disease.

The American College of Lifestyle Medicine goes furthest, recommending an eating plan based predominantly on a variety of minimally processed vegetables, fruits, whole grains, legumes, nuts, and seeds, as first-line treatment.

Health Canada’s 2019 food guide advises choosing plant-based protein foods more often. And the EAT-Lancet Commission, first in 2019 and again in its 2025 update, concluded that a plant-rich planetary health diet could prevent millions of premature deaths a year worldwide, roughly 11 million in the 2019 estimate and roughly 15 million in the 2025 update. The Global Burden of Disease project puts dietary risks at around 11 million deaths annually, led by too much sodium and too little whole grain and fruit.

That is not a conspiracy of agreement. That is what it looks like when the pyramid holds from bottom to top. If you want the broader picture of how a plant-predominant diet performs across every nutrient and outcome, not just lipids, start with our overview of plant-based nutrition.

When True Facts Lead to the Wrong Conclusion

Some of the most persuasive arguments against the evidence in this article are not made of falsehoods. They are made of true facts, stacked in a particular order, with the facts that would change the conclusion left out. A lecture by physician Paul Mason, “The truth about high cholesterol,” is a useful example because it gets several things right. LDL is not the same thing as cholesterol. The number of particles matters more than the cholesterol they carry, which is exactly why this article calls ApoB the score. Small, dense LDL is associated with heart disease. Where the lecture goes wrong is in what it concludes from these facts, that “the story we’ve been fed about LDL is the major cause of heart disease is clearly nonsense,” and in how several of its sources are described. Each claim below was checked against the paper it relies on, and the same pattern turns up in other popular arguments, which are covered here too.

Do people with high LDL live longer?

The lecture opens with a 2016 systematic review of 19 cohort studies and more than 68,000 people, and says that 16 of the 19 found the highest LDL levels lived the longest, adding that dismissing the finding is “scientific fraud.” The review is real, but three details change what it means. First, it studied only people aged 60 and older. Second, the inverse association was with death from all causes, found in 16 of 28 cohorts, not 16 of 19 studies. For cardiovascular death, which is the question at issue, only nine cohorts had data: two showed the inverse pattern and seven showed no association at all. Third, the authors list their own most important limitation: people with high LDL “may have started statin treatment during the observation period” and so may have added “a longer life to the group with high LDL-C.” Four of the authors also disclose that they have written books criticizing the cholesterol hypothesis.

A far larger Danish study, following 108,243 people aged 20 to 100, did find a U-shaped curve: both low and high LDL went with higher death from all causes, with the lowest risk at an LDL of about 140 mg/dL. But the same study found that the U-shape appeared for deaths from cancer and other causes, not for cardiovascular deaths, and that any increase in LDL was associated with a higher risk of heart attack. The people who died did so at a median age of 81.

In the United States, the study often cited for this argument is a 2021 analysis of the National Health and Nutrition Examination Survey, or NHANES, which followed 19,034 adults for a median of 7.8 years. People with an LDL under 70 mg/dL had higher death rates from all causes than people at 100 to 129, a hazard ratio of 1.71 adjusted only for age and 1.37 after also adjusting for smoking, weight, high blood pressure, diabetes, heart disease, and cancer, and unlike in the Danish data, a U-shape also appeared for cardiovascular deaths. The researchers excluded anyone who died in the first three years of follow-up, a step meant to reduce reverse causation, and still found the pattern. But they did not claim to have shown cause and effect; they concluded that further studies are needed to determine whether the relationship is causal. A single LDL measurement taken at one point in time cannot tell you whether a person’s LDL was low because of good genes, a medication started after a diagnosis, or an illness that was already underway.

Dr. Carvalho, answering this argument in a video, points out that the same curve shows up for measures no one would advise raising. A meta-analysis of 97 studies and more than 2.88 million people found that people in the “overweight” range had slightly lower death rates than people of normal weight (hazard ratio 0.94). In 144,403 adults aged 80 and older in the United Kingdom, death rates were highest in those with a systolic blood pressure below 110, and blood pressure showed an accelerated decline in the last two years of life, which the authors say means the link between low blood pressure and death “may be accounted for by reverse causation”: people are sick and near the end of life first, and their numbers fall as a result. Among 13,288 adults without diabetes, an A1c below 5.0 percent went with 32 percent higher death from all causes and 47 percent higher cancer death, a pattern the authors call “a generalized marker of mortality risk,” and a pooled analysis of six cohorts of older adults concluded that the small excess at the low end of A1c did not support a true J-shaped curve once residual confounding was considered. Nobody reads these curves as a reason to keep blood pressure or blood sugar high. Illness, frailty, and malnutrition pull many of these numbers down near the end of life. Blood pressure shows why that matters: falling blood pressure near the end of life is a sign of illness, yet when the SPRINT trial lowered it on purpose in 9,361 people at high cardiovascular risk, a target under 120 rather than 140 cut major cardiovascular events by 25 percent and deaths from any cause by 27 percent, though it also brought more side effects such as low blood pressure, fainting, and kidney injury. The trial was funded by the National Institutes of Health and stopped early because the benefit was clear.

Honesty requires one qualification to that argument. In the Danish study, the U-shape persisted even when the analysis was limited to healthy people aged 40 to 70 with no chronic disease, and its authors argue that the LDL curve has a different cause than the weight curve. The NHANES analysis likewise found the pattern after excluding early deaths and adjusting for existing disease. So reverse causation is not the whole story of every low-LDL death, and the claim, often repeated online, that the curve simply disappears once researchers account for illness goes further than these studies support. What settles the question is what happens when LDL is lowered on purpose. Across 26 randomized trials, every 1.0 mmol/L reduction in LDL cut death from all causes by 10 percent, with no significant effect on deaths from cancer, respiratory disease, trauma, or other non-vascular causes. People born with genetically lower LDL, described in the genetics section above, have less heart disease. Low LDL caused by illness is a warning sign; low LDL caused by a drug or a gene is not.

Studies built to test the claim in older people point the same way. Among 91,131 Danes without heart disease or diabetes who were not taking statins, people aged 80 to 100 with an LDL of 193 mg/dL or higher had about three times the risk of heart attack of those with an LDL under 116 (hazard ratio 2.99), and the authors found that people aged 70 to 100 with elevated LDL had the highest absolute risk of any age group and the lowest estimated number needed to treat to prevent an event. And in a 2020 meta-analysis of 29 randomized trials, the 21,492 participants aged 75 and older saw their risk of major vascular events fall 26 percent per 1 mmol/L of LDL lowering, statistically no different from younger patients. Neither study had outside funding.

Did people really eat 35 eggs a day without their cholesterol rising?

The lecture describes “this quite fabulous study” in which patients ate 35 eggs a day for a month and their cholesterol stayed normal. The patients were eight young adults with second and third degree burns over 30 to 60 percent of their bodies, fed about 7,000 calories a day to help them heal. Their cholesterol was unusually low before the diet began, averaging 108 mg/dL, and it stayed within the normal range. The researchers’ own explanation was that “the high demand for energy during the recovery of burns” prevented abnormal rises. It is a study of burn recovery, not of everyday eating. Controlled trials in healthy people answer the everyday question: a 2020 meta-analysis of 17 randomized trials found that eating more eggs raised LDL and the LDL-to-HDL ratio, and an earlier meta-analysis found that dietary cholesterol from eggs raises the ratio of total to HDL cholesterol. As the dietary cholesterol section explains, the average effect is modest and varies widely from person to person, which is exactly why a handful of unusual cases can be made to look like a rule.

Does coconut oil prove saturated fat doesn’t raise LDL?

The lecture cites a randomized trial in which people ate 50 grams a day of coconut oil, olive oil, or butter for four weeks, and says the coconut oil group’s LDL dropped while butter raised it. The trial’s actual result: butter raised LDL by 0.42 mmol/L more than coconut oil and 0.38 more than olive oil, while coconut oil and olive oil did not differ from each other. The authors wrote that their findings “do not alter current dietary recommendations to reduce saturated fat intake in general.” The trial lasted four weeks, participants were not blinded, and it was funded in part by the BBC. What it shows is that saturated fats are not all alike, not that saturated fat is harmless. A 2018 network meta-analysis of 54 randomized trials found that every one of the plant oils studied, coconut oil included, lowered LDL more than butter, and concluded that the evidence supports replacing high saturated-fat foods with unsaturated oils. Our post on coconut oil versus animal fat covers the rest of that story.

Is unprocessed red meat bad for your heart, or only processed meat?

A common reply to all of this is that meat itself is not the problem, only the way most people eat it: the additives in processed meat and the fat on red meat. Trim the fat and skip the processed products, the argument goes, and the risk disappears. Part of that is right. Processed meat does look worse per gram than fresh meat in most studies, and saturated fat does drive LDL, as the next question explains. The real question is whether removing the additives and trimming the fat brings the risk to zero.

Most large studies that separate the two find a larger risk for processed meat and a smaller but real one for unprocessed red meat. In EPIC, a study of 409,885 adults in nine European countries followed for an average of 12.6 years, each extra 100 grams a day of red and processed meat combined was associated with a 19 percent higher risk of heart attack or death from heart disease, and 25 percent higher after the first four years of follow-up were excluded to guard against people changing their diets because they were already ill. When the two were split apart, people in the top fifth of intake had the same estimated excess risk whether the meat was unprocessed red meat or processed meat, a hazard ratio of 1.10 for each, though neither reached statistical significance on its own. That study was publicly funded, and its authors report no conflicts. In six pooled American cohorts of 29,682 adults followed for a median of 19 years, two extra servings a week of processed meat went with a 7 percent higher risk of cardiovascular disease and of unprocessed red meat with a 3 percent higher risk; poultry, at 4 percent, also showed a small association, while fish showed none. Among 43,272 men in the Health Professionals Follow-up Study, each daily serving of unprocessed red meat went with an 11 percent higher risk of coronary heart disease, against 15 percent for processed red meat. Pooled analyses point the same way: one meta-analysis covering about 1.4 million people found a 9 percent higher risk of ischemic heart disease per 50 grams a day of unprocessed red meat and 18 percent for processed meat, with no association for poultry, and another covering about 4.5 million found an 11 percent higher cardiovascular risk per 100 grams a day of unprocessed red meat and 26 percent per 50 grams a day of processed meat, concluding that both are associated with higher risk. The association holds even where meat intake is low. A separate analysis of the Adventist Health Study-2 described earlier, covering 72,149 Seventh-day Adventists in the United States and Canada followed for an average of 11.8 years, found that people in the 90th percentile of unprocessed red meat intake had an 18 percent higher risk of death from any cause and a 26 percent higher risk of cardiovascular death than those who ate none, while processed meat alone was not significantly associated with mortality in that group. The study was funded by the National Cancer Institute, and its authors report no conflicts.

Not every study agrees, and a fair account has to say so. A 2010 meta-analysis found that processed meat, but not unprocessed red meat, was associated with coronary heart disease. A later analysis of the PURE study, covering adults in 21 countries followed for 9.5 years, found no association between unprocessed red meat and death or major cardiovascular disease, while processed meat raised the risk of both by roughly half (hazard ratios of 1.51 and 1.46). In the Million Veteran Program, unprocessed red meat pointed the same direction as in the larger cohorts but did not reach statistical significance (relative risk 1.14, 95 percent confidence interval 0.96 to 1.36). A 2022 Burden of Proof analysis, a method that deliberately widens the uncertainty around pooled results, found only “weak evidence of association” between unprocessed red meat and ischemic heart disease, though critics pointed out that the same method awards smoking and ischemic heart disease just three stars out of five. And the 2019 reviews behind the NutriRECS recommendations judged the associations between red and processed meat and heart disease to be very small and the evidence to be of low certainty, and suggested, as a weak recommendation, that adults continue their current intake; neither the review nor the recommendations had outside funding. The fair reading is that unprocessed red meat is less harmful than processed meat and the evidence on it is weaker and noisier, but the weight of the large cohorts leans toward a modest added risk, not zero, and what replaces the meat turns out to matter more than anything else. Country-level comparisons, like the popular claim that Hong Kong eats more meat than anywhere on Earth and still lives the longest, sit below all of these studies on the evidence ladder; our post on Hong Kong, meat, and longevity explains why that comparison measures a trading hub’s food supply rather than what its residents actually eat.

Does lean, unprocessed meat still raise LDL if you trim the fat?

Trimming fat helps. A 2025 randomized crossover trial in 61 adults found that cutting saturated fat from 12 to 6 percent of calories lowered LDL, while halving dietary cholesterol on its own did not, and its authors concluded that saturated fat, not dietary cholesterol, raises LDL. That trial was funded by the Egg Nutrition Center, the low-saturated-fat diet in it also raised the number of small LDL particles, and its finding on dietary cholesterol differs from the egg meta-analyses described above, so it is one piece of evidence rather than the last word.

But the most direct test of unprocessed meat with saturated fat held constant found that trimming the fat is not the whole story. In the APPROACH trial, healthy adults were randomly assigned to a high- or low-saturated-fat diet and, within it, ate red meat, white meat, and non-meat protein for four weeks each, with processed meats deliberately excluded to avoid the effects of added chemicals. Among the 113 who completed all three diets, LDL and ApoB were higher on both red and white meat than on non-meat protein, whether saturated fat was high or low, and red meat was no worse than poultry. The rise came mainly from larger LDL particles, and the authors concluded that their results fit recommendations for diets with a high proportion of plant foods while giving no reason to choose white meat over red. The trial was funded by the National Institutes of Health.

That finding explains why the trial literature on red meat can look contradictory. A 2019 meta-analysis of 36 randomized trials with 1,803 participants found that red meat made no significant difference to blood lipids when compared with all other diets combined, but compared with high-quality plant protein from legumes, soy, and nuts, it left LDL about 8 mg/dL higher. Against fish or refined carbohydrates, swapping out red meat brought no such benefit. The comparison food decides the answer, which matters because much of the research presenting lean beef as heart-healthy compares it with other animal foods or a typical diet. It also explains how two official documents can read the same literature in opposite ways. The Scientific Foundation document behind the 2025 to 2030 Dietary Guidelines, described in the guidelines section above, describes randomized trials of minimally processed animal-protein foods as showing neutral to beneficial effects on blood lipids, and calls the observational links between meat and chronic disease inconsistent, possibly driven by processed meat, and of low certainty. The American Heart Association’s 2026 dietary guidance says the effect of red meat on cardiovascular risk factors depends on the replacement food, and that replacing red meat, including lean, unprocessed meat, with healthier sources of protein, especially plant sources, improves those risk factors, while noting that in cohort substitution analyses the benefit was strongest for replacing processed meat. Against other animal foods, lean meat looks neutral; against plant protein, it does not.

The Beef in an Optimal Lean Diet trial, funded by the Beef Checkoff Program, found that lean beef within a low-saturated-fat diet lowered LDL compared with a typical American diet, but it included no plant-protein comparison. A 2018 controlled feeding trial, funded in part by the Beef Checkoff and the Pork Checkoff, gave 41 overweight or obese adults a Mediterranean-style diet for five weeks with either about 500 or about 200 grams a week of lean, unprocessed beef or pork, replacing the difference with poultry and other protein-rich foods. LDL fell on the higher-red-meat version and stayed unchanged on the lower one, but again, neither version compared red meat with plant protein. A 2025 systematic review of 44 trials of unprocessed red meat found that 66 percent had links to the red meat industry. Every study without those links reported unfavorable or neutral effects of red meat on cardiovascular risk factors, while every industry-linked study reported favorable or neutral ones, and red meat looked neutral against other animal proteins in 16 of 23 comparisons but unfavorable against plant proteins in 7 of 10. A letter responding to that review accepts the finding but argues that industry money partly fills a gap left by slower and scarcer public funding.

Cohorts that model the same swap within the same people agree with the trials. In the Health Professionals Follow-up Study, replacing a daily serving of unprocessed red meat with nuts, legumes, or soy was associated with a 13 percent lower risk of coronary heart disease. In the Million Veteran Program, replacing half a daily serving of red meat with nuts was associated with a 14 percent lower risk of cardiovascular disease. And a 2023 meta-analysis of substitution studies found moderate-certainty evidence that replacing processed meat with nuts, legumes, or whole grains is associated with lower cardiovascular risk. In EPIC, replacing 100 calories a day of red and processed meat with fatty fish, yogurt, cheese, or eggs was associated with about 20 percent lower risk, and the federal evidence review described earlier concluded, at moderate strength, that swapping red and processed meat for plant proteins is associated with less cardiovascular disease. The swaps with the strongest and most consistent support replace meat with plant protein, rather than processed meat with fresh.

What does red meat carry besides fat and additives?

Saturated fat is only one of the pathways researchers have proposed, and several of the others travel with the meat itself. In a controlled feeding study within the APPROACH trial, four weeks of unprocessed red meat, but not white meat or non-meat protein, more than doubled TMAO in blood and urine. An earlier study found that carnitine, which is abundant in red meat, is turned into TMAO by gut bacteria, that people who eat meat produce significantly more of it from carnitine than vegans and vegetarians do, and that it accelerated atherosclerosis in mice. As the TMAO section above explains, TMAO looks like a possible accelerator rather than the engine, but this is a pathway that trimming fat and avoiding additives does nothing to change.

Heme iron, the form of iron found in meat, is another candidate. A 2015 meta-analysis of 13 cohort studies with 252,164 participants found that each extra milligram a day of heme iron was associated with a 7 percent higher risk of cardiovascular disease, while non-heme iron and total iron showed no association. That finding is contested: a published letter argued that numerical errors undermine it, and the original authors replied defending their analysis.

A 2021 viewpoint in the Journal of the American Heart Association argues that the long focus on cholesterol and saturated fat misses toxic metabolites that gut bacteria make from eggs and meat, such as TMAO, and that people with impaired kidney function are especially vulnerable. Its authors conclude that egg yolk and red meat should be avoided and meat intake limited. That is an argued position rather than new data, and one author consults for Amgen while another has received grants from pulse-industry groups. Taken together, the evidence supports part of that argument and contradicts the rest: processing and saturated fat make meat worse, but in the best-controlled trial, unprocessed meat with saturated fat held constant still raised LDL and ApoB compared with non-meat protein, and the strongest gains in both trials and cohorts come from replacing meat with legumes, nuts, and soy.

Is small, dense LDL just “damaged” LDL?

The lecture repeatedly calls small, dense LDL “damaged” LDL and treats large, buoyant LDL as the harmless kind. Small, dense LDL is a subclass defined by size and density, and several of the papers cited do not say what the lecture says they do. The study of LDL “band 4” never discusses oxidation or damage at all; the word appears only in the titles of papers in its reference list. The 15-year study of nearly 28,000 women that the lecture describes as showing particle count predicts heart disease “two and a half times” actually studied peripheral artery disease, the kind that narrows the arteries of the legs, with 110 cases; the hazard ratio for LDL particle count was 2.03, and the authors describe a lipoprotein pattern for leg artery disease that differs from coronary disease.

The Japanese study cited for a fivefold risk is reported accurately: 3,080 people followed for eight years, with a 5.41-fold higher risk of coronary disease in the top quarter of small, dense LDL cholesterol. But that estimate rests on 79 coronary cases, and its confidence interval ran from 2.12 to 13.82. A 2020 meta-analysis that included this same study, pooling 21 studies, 30,628 people, and 5,693 coronary events, found a pooled relative risk of 1.07 for the top versus bottom quarter of small, dense LDL cholesterol, with large differences between studies. Its authors call small, dense LDL a risk marker, not yet a proven treatment target. Picking the most dramatic study out of a pool whose combined answer is far smaller is the essence of card stacking.

The deeper problem is that nothing here shows large LDL particles are safe. The better-supported lesson points the other way: count all the particles. In the Copenhagen population, among 95,108 people, ApoB above what LDL cholesterol alone would predict was associated with higher risk of heart attack in a dose-dependent way across the entire range of LDL cholesterol. In an analysis of 2,794 young adults from the CARDIA study, ApoB measured in their mid-twenties predicted coronary calcium in midlife in a dose-response pattern. The lecture’s own best point, that particle number matters, leads straight to ApoB, not away from it.

Can low triglycerides and high HDL make high LDL safe?

The lecture proposes the ratio of triglycerides to HDL as a way to tell whether a lipid profile is atherogenic, meaning likely to build plaque, or not. The study behind that idea measured 97 people with type 2 diabetes and found that the ratio was the best standard-panel predictor of “pattern B,” a predominance of small particles, with direct measurement of small, dense LDL the most specific test. It predicts particle size in that group. It says nothing about oxidation, and nothing about whether a high number of particles is safe. The HDL section above explains why a high HDL cannot be read as a shield, and the KETO-CTA study later in this article followed people chosen specifically for low triglycerides, high HDL, and high LDL, and found soft plaque still advancing.

If something else raised my LDL, does that make it harmless?

The lecture lists reasons LDL can rise besides diet: insulin resistance, a diet heavy in dairy, low vitamin B12, and inflammatory signals. Some of that has support, though mostly from early or indirect evidence. The B12 finding comes from human fat cells grown in the laboratory alongside clinical associations. The inflammation finding comes from hamsters. The dairy evidence is a cross-sectional study of 272 women, which cannot show cause, and a four-week crossover trial in 47 overweight or obese adults, in which a high-dairy diet reduced insulin sensitivity compared with a diet high in lean red meat. At most, these explain why LDL rises in some people. They do not make the resulting particles harmless. Genetics tested that directly, as the genetics section above describes: people whose LDL is lower because of variants in the cholesterol absorption gene, the statin-target gene, or both get the same protection per unit of lower LDL, whichever route produced it. The particles do their damage however their number was set.

Isn’t the benefit of lowering LDL tiny for healthy people?

Another argument points to how small the absolute benefit looks in people at low risk, often summarized as needing to treat around 100 people to prevent one event. The arithmetic is real, and the most careful look at it comes from the Cholesterol Treatment Trialists’ 2012 analysis of individual data from 27 randomized trials. Among people whose five-year risk of a major vascular event was under 10 percent, each 1 mmol/L (39 mg/dL) of LDL lowering prevented about 11 events per 1,000 people over five years: 4.1 percent had an event with treatment against 5.2 percent without. The absolute number is small because few low-risk people have an event in any five-year window. The proportional benefit was not smaller, though. In the two lowest-risk groups it was at least as large as in everyone else (rate ratios of 0.62 and 0.69 per 1 mmol/L), and the authors concluded that the benefit “greatly exceeds any known hazards of statin therapy.” The analysis was funded by public research councils and heart and cancer charities.

The large trials publish their absolute numbers alongside the relative ones, and they are worth seeing. In VESALIUS-CV, among people with atherosclerosis or diabetes who had never had a heart attack or stroke, a heart attack, stroke, or coronary death occurred in 8.0 percent of the placebo group and 6.2 percent of the evolocumab group over five years. In Ez-PAVE, aiming for an LDL under 55 rather than under 70 lowered a combined measure of cardiovascular death, heart attack, stroke, procedures to restore blood flow, and hospitalization for unstable angina from 9.7 to 6.6 percent over three years. In ODYSSEY OUTCOMES, 49 patients needed to be treated for four years to prevent one major event, and the absolute benefit was larger in people who started with an LDL of 100 or more. Against benefits of this size, the Cholesterol Treatment Trialists estimated the excess of muscle disease at about half a case per 1,000 people treated for five years, with a possible excess of hemorrhagic stroke of a similar size. Relative and absolute risk answer different questions. The consistent relative effect, across drugs and genes, is part of what shows LDL is a cause; the absolute effect tells you how much lowering it matters for a particular person over a particular stretch of time, and it grows with that person’s baseline risk and with the years the lowering lasts.

A small five-year number in people who are not yet sick says little about a disease that builds for decades. The genetics section above shows that LDL lowered from birth buys roughly three times the protection of the same lowering started later in life, and the REACT study found plaque in more than half of apparently healthy adults. One writer on X, who describes himself as a former “lipid denialist” who changed his mind because of “my patients, my data, and the broader evidence,” put the error plainly: “the mistake was extrapolating a modest short-term absolute benefit in low-risk people into the conclusion that lifetime ApoB exposure was unimportant.”

Don’t statins cause terrible side effects?

Some people do get real side effects, but blinded trials show far fewer than most people believe, because expecting harm can produce symptoms on its own, a phenomenon called the nocebo effect. In the SAMSON trial, 60 people who had stopped statins because of side effects rotated month by month among a statin, an identical placebo, and no tablet at all, without knowing which was which. Ninety percent of the symptom burden they felt on the statin also appeared on placebo, and half of them went back on the medication. Across 19 large double-blind trials with 123,940 people, muscle pain or weakness was reported by 27.1 percent of those on a statin and 26.6 percent of those on placebo; in the first year the statin caused a small excess, about one complaint in 15, and the authors concluded that more than 90 percent of the muscle symptoms reported on statins were not caused by the drug. A 2026 analysis of the same trials checked the 66 other side effects listed on statin labels, beyond the established effects on muscles and on diabetes, and found only four more common on the drug: two kinds of abnormal liver tests, changes in urine composition, and fluid swelling. The questions about memory and dementia are covered in the drug-trial section above. All three studies were funded by heart charities and public research councils. None of this means a statin is right for everyone; it means fear of side effects is a poor reason, on its own, to rule one out.

Is cholesterol a weak risk factor compared with diabetes or smoking?

A different argument, made by Dr. Fung, grants that cholesterol is a risk factor but says it is a minor one, and it cites the American Heart Association’s own risk calculator, the PREVENT equations. In the table cited, each 1 mmol/L (39 mg/dL) of higher non-HDL cholesterol, the cholesterol carried in all the ApoB particles, came with a hazard ratio of 1.00 for total cardiovascular disease in women and 1.05 in men, against 1.65 and 1.58 for diabetes and 1.62 and 1.48 for smoking. Those numbers are quoted correctly. A video critique by Nicholas Verhoeven of the channel Physionic points to what sits in the next column.

PREVENT was built from 6,612,004 American adults aged 30 to 79 with no known cardiovascular disease, followed for an average of 4.8 years. Its “total cardiovascular disease” outcome combines two different conditions: atherosclerotic disease, meaning heart attacks and strokes caused by plaque, and heart failure. Cholesterol is not even among the inputs the model uses to predict heart failure. In the plaque-disease column of the same table, the hazard ratio for non-HDL cholesterol is 1.09 in women and 1.15 in men, and in the model’s simpler base version it is 1.12 and 1.17, all clearly above no effect. The 2026 ACC/AHA cholesterol guideline uses the plaque-disease version of these equations, citing the same hard atherosclerotic endpoint used in earlier guidelines. Reading the combined column blends a disease cholesterol drives with one it does not predict.

Even the plaque-disease numbers cannot rank risk factors the way the argument tries to. Diabetes is scored as yes or no, while cholesterol is scored per 39 mg/dL, and a person with high cholesterol sits several of those units above someone with low cholesterol. The follow-up averaged under five years in people who began without known disease, while plaque builds over decades, and the model has no input for how many years of high cholesterol someone had before the study began. It is also centered at age 55, and the effect of non-HDL cholesterol in the model shrinks with each decade of age, so the same cholesterol carries more weight at younger ages, as a disease of lifetime exposure would predict. Verhoeven adds that diabetes is a whole-body disease that often raises cholesterol along with blood sugar and inflammation, so comparing it with cholesterol alone is like comparing a truck with the truck lot it belongs to.

Nobody is arguing against treating diabetes, and the guidelines do not treat the two as rivals: the 2026 guideline sets its own LDL goals for people with diabetes, under 70 mg/dL, with non-HDL under 100, for adults aged 40 to 75 with multiple risk factors. And lowering LDL helps across the board. In the 26 randomized trials pooled by the Cholesterol Treatment Trialists, major vascular events fell by about a fifth per 1.0 mmol/L of LDL lowering in every subgroup examined, including 25 percent in people with no previous vascular disease. The useful question is not which risk factor looms largest, but whether lowering each one prevents events. For cholesterol, the trials have already answered.

What about “your body makes it,” “your brain needs it,” “it’s just repairing damage,” and “arteries but not veins”?

Three one-liners travel alongside these arguments, and Dr. Carvalho answers each with an analogy. Your liver makes cholesterol, so it cannot be bad? Your liver makes glucose too, and high blood glucose is still diabetes. Many vital molecules, sodium and potassium among them, are dangerous when their blood levels run too high. Your brain needs cholesterol? It makes its own behind the blood-brain barrier, as the first section of this article explains, and does not draw on the cholesterol in your blood. Plaque forms in arteries but not veins, so cholesterol cannot be the cause? High blood sugar bathes veins too, yet no one concludes diabetes is harmless to arteries. Plaque grows where the pressure is, and veins moved into arterial pressure during bypass surgery develop it, as our post on how a trapped particle becomes a plaque explains.

One more line holds that LDL is the body’s repair crew, rushing to patch damaged arteries, so lowering it is like blaming firefighters for the fire. The mechanism described throughout this article runs the other way: particles slip into the artery wall and get trapped first, and the inflammation and damage grow out of what those trapped particles become. Firefighters do not start fires, and the trapped particles are what this fire is made of.

The lecture ends by granting that LDL “can become a problem” once oxidized. The section later in this article on whether only oxidized LDL causes heart disease explains why that gets the order of events backward: particles are trapped first and oxidized afterward, and the antioxidant trials built on the opposite idea failed. None of the true facts in the lecture undo the genetic, trial, and imaging evidence climbed in the pyramid above. They only look like they do when the rest of the cards are left in the deck.

Why “But I Feel Great” Cannot Settle This

Can you feel plaque building in your arteries?

Here is the cruelest design flaw in human anatomy: the inside of your arteries cannot feel anything.

The endothelium has no pain receptors. You cannot feel an ApoB particle slipping through the lining. You cannot feel the electrostatic snap of a positively charged protein locking onto a negatively charged proteoglycan. You cannot feel a macrophage bloating into a foam cell. A plaque can grow silently for thirty or forty years without narrowing the artery enough to cause chest pain, and remember, the Bogalusa autopsies found the process already underway in teenagers. For a shocking number of people, the very first symptom of heart disease is the heart attack. That silence is why feeling great cannot tell you whether a diet is safe for your arteries, and it is where much of the low-carb and carnivore conversation lives. Our companion post on the topic tallies the long-term evidence for carnivore eating and reports nine human studies total, none of them a randomized trial or a long-running cohort, several of them case reports, one of them a Facebook survey; read what carnivore diet research actually exists, and why so many people are convinced anyway.

Is erectile dysfunction an early warning sign of heart disease?

There is one early warning worth knowing about, though, at least for men: because the arteries feeding the penis are narrower than the coronary arteries, the same systemic plaque process tends to show up there first, and erectile dysfunction symptoms precede coronary symptoms in roughly 58 to 67 percent of patients. It is less a separate problem than the same disease knocking on an earlier door.

Why does losing weight make a bad diet look good?

This is why the “I feel amazing” testimonial, the backbone of every diet movement, is scientifically worthless as a safety claim. And there is a specific illusion worth naming. When someone drops weight on a meat-heavy, low-carb diet, shrinking fat tissue genuinely lowers systemic inflammatory signals like IL-6 and CRP. Blood sugar steadies. Energy climbs. Every dashboard light the person can feel turns green. Meanwhile, if that diet is loaded with saturated fat and cholesterol, LDL receptors stay suppressed, ApoB particles stay elevated, and the silent trapping inside the arterial wall continues every minute, felt by no one. Systemic inflammation and local arterial inflammation are two different fires, and losing weight only douses one of them. Tools like the Dietary Inflammatory Index consistently score the components of whole plant foods, fiber and polyphenols among them, as anti-inflammatory and saturated fat as pro-inflammatory; processed and high-heat-cooked animal products add advanced glycation end-products and heme iron on top of that. This particular trap, weight loss flattering every marker you can see while the ones you cannot see quietly worsen, is worked through in detail in a companion piece on why a whole-food plant-based diet is the easiest way to reach and keep a healthy weight, which also makes the case that the same mechanism taking the weight off is the one lowering your ApoB, rather than the two working against each other.

What did the KETO-CTA study actually find?

The low-carb world is currently running this exact experiment on itself. Lean, athletic keto and carnivore followers, so-called lean mass hyper-responders, often post LDL levels far above 200 while nearly everything else on the panel looks clean, and argue the LDL must be harmless in their context. The KETO-CTA study’s own entry requirements show what that “everything else” means in practice: at least two years on the diet, LDL of 160 or under before starting it, then LDL past 190 or up by half since starting, alongside HDL of 60 or higher, triglycerides of 80 or under, hemoglobin A1c under 6.0 percent, fasting glucose under 110, and C-reactive protein under 2. That reading has some real support behind it. The researchers cite an existing observational study in which a coronary calcium score of zero was common even among people with LDL of 190 or higher, and usually came with no detectable plaque on a CT scan.

A CT imaging study of 100 such people on ketogenic diets, the KETO-CTA study, was funded by the Citizen Science Foundation, the nonprofit of coauthor David Feldman, and its participants averaged an LDL of 242. They were recruited through social media, and the study’s primary outcome was the change in noncalcified plaque volume, so plaque was the point of the study from the start. The results did not settle the question in their favor. Over a single year, coronary plaque advanced in this group, with the amount of soft plaque already present at the start being the strongest predictor of how much more accumulated. The study’s first published paper, titled to argue that plaque predicts plaque while ApoB does not, said so plainly in its own abstract: a median noncalcified plaque volume rise of 18.9 cubic millimeters over one year, which the paper’s body and figure legend translate to a 43 percent relative change. The raw number was there. What was not there, in any prominent form, was that relative framing, and other physicians said so publicly before the authors supplied it themselves in a later reply, calling its absence “a sincere oversight, not intentional selective reporting.” The retraction came only after that public back-and-forth, not before it. The increase was also far from uniform: the interquartile range ran from 9.3 to 47.0 cubic millimeters, and the mean change of 31.5 cubic millimeters carried a standard deviation just as large, meaning some participants’ plaque barely moved while others progressed rapidly. The authors made this point themselves in their reply to critics, cautioning readers against treating the group as though everyone in it carries the same risk.

What that paper got wrong, and the reason it was retracted in 2026, is a more interesting story than a simple measurement error. Other physicians published a letter raising concerns about selective reporting, the statistics, and the short one-year window. Separately, and more consequentially, most of the authors had deliberately stayed blinded to the raw data until after publication, a safeguard meant to protect the analysis from bias. Once the anonymized data reached the study’s funder, the Citizen Science Foundation, and its founder David Feldman, a closer look turned up anomalies suggesting the outside vendor that read the scans, Cleerly, had not kept its own analysts blinded to which scan was the baseline and which was the one-year follow-up, exactly the kind of lapse that can distort a subtle change. It also came out that coauthor James Earls was Cleerly’s chief medical officer and held equity in the company; the published paper disclosed his Cleerly affiliation but not the equity stake, which the journal has confirmed he reported to them only after the paper was accepted, too late to appear in print. Earls is no longer listed as a coauthor on the follow-up work, and Cleerly’s own leadership page no longer lists him as chief medical officer. The retraction notice states that the errors identified were too significant to fix with a correction.

When the results were re-reported in a 2026 preprint, the scans were reanalyzed with the study’s pre-specified method, a software system called QAngio, read by blinded expert readers, and a second company, HeartFlow, ran an independent, blinded confirmatory analysis. The progression was still there, slightly smaller in relative terms: a median noncalcified plaque increase of 5.6 cubic millimeters, a 37 percent relative rise against 43 percent in the original analysis, and the HeartFlow analysis produced comparable results. Worth stating plainly, because it cuts against the direction a critic might assume: fixing the blinding problem did not manufacture a plaque increase that was not there before. Because the original and new analyses used different software, their absolute volumes, 18.9 and 5.6 cubic millimeters, are not a like-for-like comparison, and the authors note that relative percentages run high when baseline plaque is as low as it was here. If the unblinded reads distorted anything, they appear to have exaggerated progression, not hidden it. Some participants did improve: 15 percent showed regression in the QAngio analysis and 35 percent in the HeartFlow analysis, and seven people showed regression by both methods. Both versions agree that plaque advanced across the group, and the authors describe the overall progression as modest and uneven.

One detail explains, about as well as anything can, how a team of physicians could publish measurements they could not vouch for without any of them personally misreading a scan: none of them read the scans themselves. The plaque measurements came from Cleerly’s software pipeline, and the study depended on that vendor keeping its analysts blinded to which scan was the baseline and which was the one-year follow-up. Coauthor Nicholas Norwitz has since published his own detailed account of what happened next. He says the funder asked Cleerly to redo the analysis properly blinded, and Cleerly declined. Separately, a small number of participants, acting through their own cardiologists and apparently without Cleerly’s knowledge, independently resubmitted their scans to the same company for a fresh read. Only 8 of the study’s 100 participants went through this path, so it is a small, self-selected group rather than a systematic check, not the main evidence the reanalysis rests on. Within that group of 8, Norwitz reports the average plaque change reversed entirely: a mean increase of 20.9 cubic millimeters in the original unblinded reads became a mean decrease of 5.1 cubic millimeters once blinding was restored. That is Norwitz’s account, from his own side of the dispute. Cleerly’s spokesperson declined to answer Retraction Watch’s questions about why the company had not blinded its original analysis, so there is no public account from the company itself, though in both cases the blinded re-read came out lower than Cleerly’s original unblinded read. The two differ on what remained: the new blinded analyses still found net progression across the group, while the eight re-read scans showed net regression, one more reason to treat the eight as an anecdote rather than a check.

Does a low calcium score mean you’re in the clear?

Here is the finding that matters most for anyone leaning on a calcium scan for reassurance. Fifty-seven percent of these same participants started the year with a coronary calcium score of zero. A year later, 54 percent still scored zero and 81 percent still scored under 100, both comfortably inside what doctors treat as the low-risk range. A calcium score only counts the calcified fraction of plaque. It has nothing to say about the soft plaque underneath, and soft plaque is exactly what grew, by that same 37 percent, in the same people, over the same year the scan kept saying they were fine. The much larger REACT study, described earlier, found the same blind spot in ordinary adults: among participants in their 30s who had coronary plaque, 41.8 percent of the men and 48.4 percent of the women had a calcium score of zero. The reassuring observational study the researchers cite was a snapshot; this study followed the same people forward and caught the calcium score standing still while the plaque it cannot see kept building.

The re-report also remains a non-peer-reviewed preprint with no comparison group, and not everyone considers the matter settled. One of the study’s most vocal critics, family medicine physician Brad Stanfield, told the retraction-tracking publication Retraction Watch that the retraction was warranted, but that presenting the reanalysis as confirming the original story looked “more advocacy than science” to him. A second critic, internist Michael Mindrum, said he saw real improvement in the new analysis but remained skeptical of the whole project. The authors, for their part, argue that their participants’ LDL spanned a wide enough range to test whether it predicted who progressed, and report that it did not. The counterargument is that the entire range sat far above normal, the group was small, and one year is short, so a null result here says little about whether LDL matters. It is a bit like finding that height fails to predict basketball skill in a room where everyone is already well over six feet. And in 2026, researchers from that same circle published the ezetimibe paper described earlier, actively studying how to bring the number down. A quick overhaul that clears the rest of the lipid panel is not the same thing as a diet proven safe for the arteries, and this study is the clearest demonstration of that gap: feeling great is a mood, a clean calcium score can miss what is actually growing underneath it, and ApoB was the one number on these panels that pointed toward what the scans then showed: a group with very high ApoB grew plaque. In a room where everyone’s ApoB is high, ApoB cannot sort the room, but it did say what would happen to it. Only one of those three predicts your arteries.

The Nitrate Paradox: Same Atoms, Opposite Fates

Are the nitrates in vegetables bad for you?

One oxygen atom separates nitrate from nitrite, yet in food they behave like different substances, because chemistry never acts alone. It acts inside a food matrix.

Dietary nitrate from leafy greens and beets is concentrated by your salivary glands, converted by oral bacteria into nitrite, and then transformed in the body into nitric oxide, the master signaling molecule that relaxes blood vessels, lowers blood pressure, and helps maintain the endothelial lining that keeps lipoproteins moving instead of sticking. Randomized trials of dietary nitrate show real, sustained blood pressure reductions. Crucially, plants deliver their nitrate bundled with vitamin C and polyphenols, and vitamin C in particular blocks nitrite from forming N-nitroso compounds, a protective effect shown first in the test tube and later in human studies.

Why are the nitrites in processed meat different?

Meat is the mirror image. It contains little beneficial nitrate but is often cured with added nitrite, which can react with amines to form N-nitroso compounds, a family of potent carcinogens. The heme iron in red and processed meat makes matters worse: feeding studies show it promotes the formation of these compounds inside the gut itself. This chemistry sits behind the IARC’s decision to classify processed meat as a Group 1 carcinogen. Same nitrogen, opposite outcomes, and the difference is the plant matrix wrapped around it. If the mirror-image worry is on your mind, that plant foods carry their own dangerous compounds, we handled that argument separately in our post on plant toxins and antinutrients.

What Plant Foods Actually Do, Mechanism by Mechanism

Everything in this article so far describes a trap: particles in the blood, charges on a protein, a mesh in the wall. Plant-predominant eating disarms that trap at nearly every stage, and the mechanisms are worth spelling out.

1. Do plants contain cholesterol, and what are phytosterols?

Plants contain essentially zero cholesterol. Every gram of plant food displacing animal food is dietary cholesterol that never enters the customs gate. Better still, plants carry phytosterols, cholesterol look-alikes that compete for space in the micelles and at the NPC1L1 gate, so cholesterol gets crowded out of absorption. At intakes of 0.6 to 3.3 grams a day, plant sterols and stanols cut LDL by roughly 6 to 12 percent in randomized trials.

2. How does fiber lower cholesterol?

Remember that your own bile pours roughly a gram of cholesterol and a load of bile acids into the gut daily, and most of it normally gets reabsorbed and recycled. Soluble fiber from oats, barley, beans, and psyllium binds those bile acids and escorts them out of the body. The liver must then pull cholesterol out of your bloodstream to manufacture replacement bile. This is not a folk theory: it is the exact working mechanism of an entire prescription drug class, the bile acid sequestrants. Oatmeal and beans are a food-strength version of the same mechanism as that drug, and this is the mechanism that reaches even the cholesterol your body made itself.

Ordinary portions of oats show the effect in randomized trials. In Beijing, 166 adults with high cholesterol ate either 100 grams of instant oat cereal or 100 grams of wheat noodles every day for six weeks, and total and LDL cholesterol fell significantly in the oat group compared with the noodle group. In a Thai crossover trial, 24 adults ate 70 grams of oatmeal a day, supplying 3 grams of beta-glucan, for four weeks and rice porridge for another four; on oats, LDL fell 10 percent from where it started, more than on rice. And in a Chinese trial of 210 adults comparing 80 grams of oats with 80 grams of rice daily for 45 days, total and non-HDL cholesterol fell more with oats than with rice, LDL fell 9.1 percent from baseline in the oat group, and the oat group’s gut bacteria shifted, with more of species such as Akkermansia muciniphila and Roseburia. All three trials had funding from PepsiCo or authors employed by it. Large cohorts point the same way: among 54,871 Danish adults aged 50 to 64 followed for 13.6 years, men and women eating the most whole grains had about a 25 percent lower risk of heart attack than those eating the least, and in men, each extra 25 grams of oatmeal a day went with a 16 percent lower risk; rye bread was linked to lower risk in both men and women, while wheat showed no significant association. In three Harvard cohorts of more than 205,000 American men and women followed for an average of 25.8 years, people who ate oatmeal at least twice a week had about a 21 percent lower risk of coronary heart disease than those who ate it less than once a month, and every whole-grain food studied except popcorn was linked to lower risk. The benefit from oatmeal appeared to level off at about half a serving a week, so a regular habit, not a large daily dose, accounted for most of it. Like all cohort findings, these are associations, adjusted for diet and lifestyle but not proof on their own; the trials above supply that.

Fiber may not be the whole story. A 2026 study in Nature Communications ran two small randomized trials in people with metabolic syndrome. In one, 17 people replaced their usual diet with three 100-gram oat meals a day for two days, and their total cholesterol fell 8 percent and their LDL 10 percent, though they also ate fewer calories, which the authors acknowledge. Gut bacteria convert phenolic compounds in oats into molecules such as ferulic acid and dihydroferulic acid that then appear in the blood, and the researchers found these compounds tracked the cholesterol drop, explaining about 19 percent of the variation in LDL, with laboratory experiments suggesting how they might act. In the companion six-week trial, though, replacing one meal a day with 80 grams of oats while keeping an otherwise usual Western diet left cholesterol largely stable, which the authors describe as a rather mild effect, a contrast with the 45-day Chinese trial above, which used the same 80 grams a day in people with mildly high cholesterol and found a benefit compared with rice. Nicholas Verhoeven of the channel Physionic, reviewing the study, said he was unimpressed with some of its data and disagreed with some of the researchers’ interpretations. The microbial pathway is a promising addition to the fiber story, not yet a replacement for it; the steadier evidence is the everyday-portion trials above. Our carbohydrates post goes deeper on oats, whole grains, and what the gut bacteria that ferment them actually produce.

3. Does swapping saturated fat for unsaturated fat lower LDL?

Replacing saturated fat with the unsaturated fats of nuts, seeds, olives, and avocados causes the liver to display more LDL receptors, the vacuum cleaners come back online, and LDL and ApoB fall. This is the most consistent finding in the entire controlled-feeding literature. Polyunsaturated fats also signal through liver receptors like PPAR-alpha to reshape the particles the liver ships out, and larger, more buoyant particles appear somewhat less prone to retention in the wall, though the headline effect is simpler: fewer ApoB particles in circulation, period. Two of the loudest objections to this swap, that seed oils are inflammatory and that coconut oil is a health food, each get a full evidence-pyramid treatment of their own: saturated fat, seed oils, and heart disease and coconut oil versus animal fat.

Do you need to eat fat to absorb antioxidants?

Before we get to what these compounds do inside a lipoprotein, there is a step that has to happen first, and it explains why a plate of undressed vegetables is not the same as a plate of vegetables eaten with a little fat. Carotenoids are locked inside plant cells and are stubbornly insoluble in water, so they reach your bloodstream by exactly the route this article already walked you through for cholesterol. Fat in the meal triggers bile release, the bile and fat assemble into mixed micelles, the micelles ferry the freed carotenoids to the intestinal wall, and the cells pull them in using the very same transporter family that handles cholesterol, chiefly SR-B1, with CD36 and NPC1L1 also implicated. From there they are packed into chylomicrons, ride the lymphatic system up through the thoracic duct, and only then make their way into LDL and HDL. No fat, no micelles, no delivery.

This is not a subtle effect. In one controlled crossover study, salads eaten with fat-free dressing produced almost no measurable carotenoid absorption at all, essentially a flat line, while the same salads with full-fat dressing produced a large rise that kept climbing as the fat went up. In another, funded by the California Avocado Commission, adding avocado to a salad multiplied beta-carotene absorption more than fifteenfold, roughly quintupled lutein, and, importantly for the whole-food argument, the avocado fruit worked just as well as extracted avocado oil. Cooking helps too, by breaking down the cell walls that trap these compounds: lycopene absorption from tomato paste ran two and a half to nearly four times higher than from the same amount of fresh tomato. Heat and fat together, as in a simmered sauce with a little olive oil, are better than either alone.

The practical amount is small. A modest amount of fat in a meal, a spoonful of tahini, a quarter of an avocado, or a small handful of walnuts, substantially increases absorption, and in a dose-ranging trial absorption kept rising steadily as fat increased from zero to eight grams. Two honest limits on this advice. It applies to the fat-soluble compounds, so the carotenoids, including the provitamin A ones, vitamin E, and vitamin K from greens, and not to water-soluble ones like vitamin C and the anthocyanins, which do not need fat at all. And responses vary a lot between people, with some individuals absorbing very little regardless. It is also the one real caveat on very-low-fat plant-based eating, and it is easily solved without reaching for refined oil or animal products, since whole-food fats do the job. Our post on animal fat, plant fat, and your arteries takes up the same absorption question from the side of the fat itself, and our anti-inflammatory diet post follows these antioxidants after absorption, to the reactive oxygen species they neutralize and the organs they reach.

4. How do antioxidants protect LDL particles?

So that is how the cargo gets loaded. Fat-soluble antioxidants from plants, carotenoids like lycopene and beta-carotene, ride inside the lipid core of LDL particles, alongside the body’s own coenzyme Q10, while water-soluble compounds like vitamin C and anthocyanins patrol the surrounding plasma. They sacrifice themselves to neutralize free radicals before those radicals can oxidize the particle’s fats or warp its ApoB protein.

Here is why that matters for the trap specifically. The initial ionic handshake between ApoB and a proteoglycan is reversible: a native, healthy particle that gets snagged can let go and drift back into circulation unharmed. What makes retention permanent is what happens next. Stuck in the subendothelial space, cut off from the plasma’s antioxidant supply, a snagged particle begins to oxidize, and oxidized particles aggregate, fuse, bind to the matrix through additional mechanisms, and get devoured by macrophages, none of which can be undone. Worse, the oxidation products inflame the wall and push it to manufacture even more proteoglycan mesh, so every particle that turns toxic strengthens the trap waiting for the next one. A particle that arrives loaded with carotenoids and vitamin E carries its own fire extinguisher into the one compartment where the bloodstream cannot help it. In laboratory oxidation tests such particles resist conversion longer, and the retention model predicts that a particle that resists conversion stays detachable and starves that feedback loop. Whether that is what protects people who eat this way has been tested only with single antioxidants in capsules, which, as the next section shows, did not work. One precision note for the careful reader: heavy oxidation does not tighten the original ionic bond itself (it actually degrades the positively charged lysines), so the honest statement is not that oxidation strengthens the handshake, but that oxidation converts a temporary snag into permanent capture through aggregation, immune uptake, and wall remodeling.

Is it only oxidized LDL that causes heart disease?

A popular argument takes the mechanism in the previous paragraph and runs the other way with it: LDL itself is harmless, the claim goes, and only oxidized LDL causes heart disease, so the answer is antioxidants rather than lower LDL. We look at Palmer’s version of this argument in our post on whether vegans need DHA and EPA. Three lines of evidence show why it gets the order of events backward.

First, oxidation is a step in how LDL causes disease, not a rival cause. The European Atherosclerosis Society’s 2020 consensus statement on how LDL drives atherosclerosis lays out the sequence: LDL enters the artery wall and is retained there, “with ensuing pathways of modification,” and it is in the subendothelial matrix that retained particles become “susceptible to oxidation.” A particle has to be delivered and snagged before that local oxidation can begin, which fits the same conclusion the panel draws from the broader genetic evidence: that “the cumulative arterial burden of LDL-C drives the development and progression” of the disease. The panel also notes that LDL lingering longer in the bloodstream has more opportunity to be modified, including by oxidation, but that points the same way: more LDL, circulating longer, means more oxidized LDL. Oxidized LDL is not an alternative to LDL. It is what LDL becomes after too much of it gets caught. The European Atherosclerosis Society supported the panel’s meetings, and most of its authors disclose research funding or fees from drug companies, including makers of cholesterol-lowering drugs.

Second, the genetic evidence tracks how much LDL people carry, not how oxidized it is. The Mendelian randomization studies in Level 3 of the evidence pyramid compare people who inherited genes that set their LDL lower or higher from birth, and heart disease follows the LDL. The same consensus statement puts a number on it: in Copenhagen data analyzed this way, each 1 mmol/L (39 mg/dL) of genetically higher LDL cholesterol came with about twice the risk of heart attack. The genes used in that analysis are ones the panel classifies as setting the quantity of LDL, and quantity alone is enough to move risk.

Third, the direct test. If oxidation alone were the problem, supplementing vitamin E, the major antioxidant in the body’s fatty compartments, should prevent heart attacks. The HOPE trial tested exactly that: 9,541 people aged 55 or older with cardiovascular disease or diabetes were randomized to 400 IU a day of natural-source vitamin E or placebo for an average of 4.5 years. Heart attack, stroke, or cardiovascular death struck 16.2 percent of the vitamin E group and 15.5 percent of the placebo group, not a significant difference. An extension, HOPE-TOO, kept following participants for a median of 7 years and still found no cardiovascular benefit, though it did find a 13 percent higher risk of heart failure in the vitamin E group.

The fair criticisms are worth stating, and the most common one does not hold up. It is often said the pills may not have been absorbed, but the investigators checked blood levels in a randomly selected subsample, and vitamin E rose by roughly 70 percent on treatment while staying flat on placebo. The stronger criticisms come from the investigators themselves: they tested a single antioxidant rather than the combination that comes packaged in food, and they studied older, high-risk patients over a few years, so the trial cannot speak to lifelong intake. HOPE-TOO was run partly to answer the duration question, and seven years of treatment did not change the answer. The trials were funded by Canada’s Medical Research Council, the Heart and Stroke Foundation of Ontario, several drug companies, and the Natural Source Vitamin E Association; the HOPE-TOO report states that its sponsors had no role in the study’s design, analysis, or writing, and one member of its writing committee disclosed grant support from that association among many industry sources.

Put together, oxidized LDL is real, and it matters, which is why this article spends so long on the antioxidants that ride inside LDL particles. But it is downstream of LDL, not a substitute for it, and whatever antioxidant protection diet offers appears to come packaged in whole foods, since capsules have not delivered it.

Why do antioxidants need to be resupplied?

There is a detail in the discussion of the trap above worth slowing down for, because it explains why the arterial wall is such a dangerous neighborhood. When we say an antioxidant sacrifices itself, that is not a metaphor. It is an electron transfer. Vitamin E quenches a lipid radical by handing over an electron, and in doing so becomes a mild radical itself. It does not have to stay spent, because the body runs a regeneration relay: vitamin C restores the vitamin E, and the vitamin C is itself regenerated by several systems, some relying on glutathione and some drawing on NADPH through the enzyme thioredoxin reductase. Antioxidant protection is not a fixed tank that drains. It is a resupply line. One honest note for the careful reader: the vitamin C to vitamin E step is well demonstrated in the test tube, and researchers are still working out how much of it, and how much each regeneration route, operates in living tissue.

Now recall where a retained particle sits. Wedged in the subendothelial space, it is separated from that circulating relay. Its onboard vitamin E and carotenoids can absorb the first hits, but there is no ready resupply to regenerate them. The extinguishers discharge and nobody refills them, which fits the response-to-retention model exactly. Retention is not only a location problem. It is a supply-line problem.

This is also why coenzyme Q10 keeps turning up on both sides of the story. CoQ10 is a working component of the mitochondrial electron transport chain, ferrying electrons as part of energy production, and it is simultaneously a lipid antioxidant carried inside LDL particles, where it shields the particle’s fats from peroxidation. When LDL from supplemented volunteers was oxidized in the lab, lipid oxidation stayed very slow until 80 to 90 percent of the ubiquinol aboard had been consumed, then accelerated sharply. The same molecule is both a piece of the engine and a piece of the fire suppression.

Which reframes the dietary question. Reactive oxygen species are not a malfunction. They arise from ordinary electron leak at the transport chain, the unavoidable byproduct of running an oxygen-powered metabolism. You cannot stop generating them. What you decide, three times a day, is whether the food arriving alongside that metabolism brings replacement crew. Whole plant foods deliver carotenoids, tocopherols, vitamin C, and polyphenols packed in with their calories. Animal foods, as the companion piece on dietary fat documents, arrive with very little.

Do antioxidant supplements work as well as antioxidant foods?

Diet also rewrites the particle’s flammability from the inside. LDL whose core is built from oleic acid, the monounsaturated fat of olives, avocados, and most nuts, is measurably harder to oxidize in laboratory assays than LDL enriched in more chemically fragile fats. And notice what nature did with its most oxidizable fats: it packed the polyunsaturated oils of nuts and seeds inside the food supply’s richest sources of vitamin E, shipping the fuel and the fire suppression in the same shell. The matrix, again. Early laboratory work even suggests some polyphenols may interfere with the ApoB-proteoglycan binding step itself, though that line of research is young and should be labeled as such. And one caution the pyramid taught us: isolated antioxidant pills have not produced a benefit in trials, and the beta-carotene trials described earlier actually caused harm. Whatever protection these compounds offer appears to travel with the whole food, not the capsule.

5. Can diet stabilize existing plaque?

Polyphenols stimulate the endothelium to produce more nitric oxide, measurably improving arterial function in human trials, restoring the lining that controls how many particles get through in the first place. Plant-heavy patterns lower inflammatory markers like CRP, which points to less of the signaling that recruits new macrophages to existing plaques. No diet dissolves a calcified core, and no honest writer will tell you otherwise. But shrinking the soft lipid core and thickening the fibrous cap converts a rupture-prone plaque into a stable one, and rupture or erosion of the plaque surface, not narrowing, is what kills. The regression trials, from Ornish’s lifestyle program to the intensive LDL-lowering imaging studies, show arteries genuinely improving once particle counts fall far enough for long enough.

Put plainly: plant-forward eating lowers how many particles are circulating, changes what those particles carry, protects them from turning toxic, and repairs the wall they threatened. It is the only dietary pattern that addresses every stage of the trap at once, which is presumably why it keeps winning at every level of the evidence pyramid.

Where Honest Scientists Still Argue

No credible overview pretends the field has zero live debates, so here are the real ones, stated fairly.

Are eggs actually bad for you?

Dietary cholesterol by itself is the most modest lever in the chain. Its average effect on blood levels is smaller than saturated fat’s, individual responses vary enormously, and the egg literature in particular is genuinely mixed. The split is real and worth seeing: one large analysis of US cohorts found that more dietary cholesterol and more eggs tracked with modestly higher cardiovascular risk, while a set of three big prospective cohorts with a meta-analysis found no association between up to one egg a day and heart disease in Western populations. That kind of disagreement is why the AHA allows an egg a day for healthy people and why guidelines pivoted to food patterns instead of milligram caps. The strongest, least disputed claim was never “an egg will kill you.” It is that ApoB-containing particles cause atherosclerosis, and that the overall package of cholesterol-and-saturated-fat-rich animal foods raises them, while plant-predominant patterns lower them.

What about the PURE study and full-fat dairy?

Some observational findings genuinely complicate the simple story. The large PURE cohort, following more than 135,000 people across 18 countries, found higher saturated fat intake associated with modestly lower total mortality, and full-fat fermented dairy keeps turning up more neutral than its saturated fat content predicts. That is a real finding about food matrices rather than a rehabilitation of saturated fat, and a 12-week trial in 72 adults with metabolic syndrome echoed it: three daily servings of full-fat milk, yogurt, and cheese left LDL no different from a limited-dairy diet, though the study was funded largely by dairy industry groups. Notably, the same butter that sits inside those neutral total-dairy averages is associated with higher mortality when it is studied on its own, and swapping it for plant oils lowers LDL in trials. The best reading of the total evidence is that the replacement matters: swapping saturated fat for whole-food unsaturated fats and intact carbohydrates helps, swapping it for refined starch and sugar does not, which fits the Harvard replacement analyses, and the Women’s Health Initiative showed that cutting total fat without targeting the replacement achieves little.

Does keto work?

Low-carb diets produce real short-term wins. Weight loss, improved glycemic control, higher energy. Those results are genuine and worth acknowledging. The unresolved cost is long-term ApoB exposure in the people whose LDL climbs, and the early imaging data from the low-carb community’s own imaging study is, so far, not reassuring.

A 2026 trial in Cell Metabolism shows the short-term wins at their best, and also their limits. Researchers randomly assigned 55 adults with what they call metabolically unhealthy obesity, meaning obesity with prediabetes and a fatty liver, to a very-low-carbohydrate ketogenic, a Mediterranean, or a very-low-fat plant-forward diet, supplied all of their food, and matched the weight loss at about 10 percent. Weight loss improved insulin sensitivity in every group, but the ketogenic diet cut liver fat more, by 67 percent against 45 percent on the other two diets, and improved the liver’s insulin sensitivity and blood sugar more. LDL cholesterol and ApoB did not differ between the groups. The trial was funded by the National Institutes of Health. Its authors note that the sample was small and that they cannot tell whether the liver benefits depend on eating this way during weight loss. It is a trial of people who were already metabolically sick, losing weight under close supervision. It says nothing about a lean, healthy person eating ketogenically for years, which is exactly the group whose LDL tends to climb, as the KETO-CTA participants’ average LDL of 242 shows.

What could the evidence reviewers not conclude?

Honesty here means reporting what the same committees could not conclude. When the federal reviewers assessed dietary cholesterol directly, they found the evidence insufficient to determine its independent relationship with blood lipids, precisely because dietary cholesterol travels with saturated fat in real food and the two are hard to separate. They also could not grade higher-fat versus lower-fat dairy, could not draw a conclusion on ultra-processed foods and cardiovascular disease, and when they attempted the food-level saturated fat question as a whole, they found the overall body of evidence too thin for a verdict, even while individual comparisons within it graded strong. Those gaps are real. What they do not do is undercut the causal core, which rests on genetics, mechanism, and drug trials rather than on food-frequency questionnaires.

What is not seriously disputed?

The core: LDL and ApoB cause atherosclerotic disease, established by consilience across genetics, mechanism, epidemiology, and randomized trials; lowering them, by any of several mechanisms including diet, lowers events; and no major scientific body anywhere on Earth recommends eating more cholesterol or more saturated fat.

The Bottom Line

• Your body manufactures every milligram of cholesterol it needs. The dietary requirement is zero, per the National Academies.
• Heart disease is a particle-trapping disease. ApoB particles enter the artery wall, bind electrostatically to proteoglycans, oxidize, and build plaque. Block the trapping and far less disease develops.
• Cholesterol-and-saturated-fat-rich eating raises the number of trapped particles by suppressing LDL receptors and enriching remnants. The process is silent for decades because arteries cannot feel.
• Every tier of evidence, from Nobel-winning cell biology to genetic natural experiments to metabolic ward feeding studies to randomized trials to the consensus of essentially every major scientific body across multiple countries, points the same direction.
• Plant-predominant eating attacks the mechanism at all five stages: zero cholesterol in, bile drained out, receptors restored, particles shielded from oxidation, endothelium healed. Its results show up in everything from four-week LDL trials to multi-decade population outcomes.
• Feeling great is not a cardiovascular test. ApoB is. If you take one action from this article, ask for that number, and ask once for your lipoprotein(a), the one cholesterol particle diet cannot meaningfully lower.

Keep Reading

This post is the molecular core of a larger series that walks the same evidence pyramid through the dietary fat debate. If cholesterol was your way in, these take the neighboring threads further.

• Animal Fat, Plant Fat, and the Journey From Your Plate to Your Arteries follows a fatty meal from bile to chylomicron to liver, contrasts how the animal and plant kingdoms package their fat, and dismantles the “I feel great on keto” illusion that this post’s silent-damage section only introduces.
• How a Trapped Particle Becomes a Plaque picks up where both posts leave off, at the artery wall itself: how particles get caught, what oxidation, sugar, and iron do to them there, how foam cells and cholesterol crystals build a plaque, and what makes one rupture. It also explains how the particles made by your gut and your liver respond to diet, including what carnivore eating does to LDL, answers why plaque forms in arteries and not veins and whether fish get heart disease, and introduces the whole family of apolipoproteins, ApoE4 included.
• Carbohydrates follows fiber, whole grains, and oats from the plate to the gut bacteria that ferment them, and explains why intact carbohydrates lower cholesterol while refined sugars push it up.
• Saturated Fat, Seed Oils, and Heart Disease climbs every rung of the evidence hierarchy to answer whether replacing saturated fat with plant oils helps, and takes the “seed oils are poison” claim apart level by level.
• Coconut Oil vs. Animal Fat settles the saturated fat that keeps getting a health halo, and shows why “but it raises HDL” cannot rescue it, an argument this post’s HDL section sets up.
• Why a Whole-Food Plant-Based Diet Is the Easiest Way to Lose Weight takes on the mirror image of this post’s warning: almost any diet that removes weight will flatter the markers you can see, which is exactly why the silent ones matter.
• Plant-Based Nutrition is the wider overview, covering every nutrient and outcome rather than lipids alone, and the vitamin B12 question covers the one supplement this way of eating genuinely requires.
• The Adventist Health Studies and what the longest-living people eat go deeper on the population evidence this post summarizes in a paragraph.
• Our look at Hong Kong, meat, and longevity explains why the city that tops both the meat-supply and life-expectancy tables does not show that meat is healthy, and what the evidence credits for its long lives instead.
• Whether vegans need DHA and EPA works through the omega-3 question this post only touches in its DHA aside.
• Why your body converts nutrients asks why the body keeps its own conversion pathways for nutrients like vitamin A and DHA, and what eating the animal’s finished molecule skips.
• Suppose the Science Isn’t Settled takes the “nobody really knows what to eat” position seriously and follows it to where it actually leads.
• The Number You’re Being Told to Ignore, by Jay Motley, MD, is a recommended companion read: a shorter, plain-language case for why LDL and ApoB matter, how the most common counterarguments fall apart, and how a physician who has eaten whole-food plant-based for more than eight years applies it to his own numbers.

For readers who want to explore the primary research further, NutritionFacts.org maintains a free, ad-free library of evidence reviews.

Thanks to Jay Motley, MD, whose piece The Number You’re Being Told to Ignore pointed us to several of the studies added in a recent update to this article, including the research on statin side effects, LDL in older adults, early-life cholesterol exposure, and ApoB as a predictor of heart attacks.

This article is for education, not personal medical advice. If you take lipid-lowering medication or have existing heart disease, work with your clinician before making major dietary changes.

References

1. Brown MS, Goldstein JL. A receptor-mediated pathway for cholesterol homeostasis. Science. 1986;232:34-47. doi:10.1126/science.3513311
2. Dietschy JM, Turley SD. Cholesterol metabolism in the central nervous system during early development and in the mature animal. J Lipid Res. 2004;45:1375-1397. doi:10.1194/jlr.R400004-JLR200
3. Institute of Medicine. Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids. National Academies Press; 2005. Chapter 9, Cholesterol. nap.nationalacademies.org
4. Altmann SW, et al. Niemann-Pick C1 Like 1 protein is critical for intestinal cholesterol absorption. Science. 2004;303:1201-1204. doi:10.1126/science.1093131
5. Williams KJ, Tabas I. The response-to-retention hypothesis of early atherogenesis. Arterioscler Thromb Vasc Biol. 1995;15:551-561. doi:10.1161/01.atv.15.5.551
6. Skalen K, et al. Subendothelial retention of atherogenic lipoproteins in early atherosclerosis. Nature. 2002;417:750-754. doi:10.1038/nature00804
7. Steinberg D, et al. Beyond cholesterol: modifications of low-density lipoprotein that increase its atherogenicity. N Engl J Med. 1989;320:915-924. doi:10.1056/NEJM198904063201407
8. Duewell P, et al. NLRP3 inflammasomes are required for atherogenesis and activated by cholesterol crystals. Nature. 2010;464:1357-1361. doi:10.1038/nature08938
9. Cohen JC, Boerwinkle E, Mosley TH Jr, Hobbs HH. Sequence variations in PCSK9, low LDL, and protection against coronary heart disease. N Engl J Med. 2006;354:1264-1272. doi:10.1056/NEJMoa054013
10. Myocardial Infarction Genetics Consortium Investigators. Inactivating mutations in NPC1L1 and protection from coronary heart disease. N Engl J Med. 2014;371:2072-2082. doi:10.1056/NEJMoa1405386
11. Ference BA, et al. Effect of long-term exposure to lower low-density lipoprotein cholesterol beginning early in life on the risk of coronary heart disease: a Mendelian randomization analysis. J Am Coll Cardiol. 2012;60:2631-2639. doi:10.1016/j.jacc.2012.09.017
12. Ference BA, et al. Low-density lipoproteins cause atherosclerotic cardiovascular disease. 1. Evidence from genetic, epidemiologic, and clinical studies. A consensus statement from the European Atherosclerosis Society Consensus Panel. Eur Heart J. 2017;38:2459-2472. doi:10.1093/eurheartj/ehx144
13. Boren J, et al. Low-density lipoproteins cause atherosclerotic cardiovascular disease: pathophysiological, genetic, and therapeutic insights. A consensus statement from the European Atherosclerosis Society Consensus Panel. Eur Heart J. 2020;41:2313-2330. doi:10.1093/eurheartj/ehz962
14. Kannel WB, Dawber TR, Kagan A, Revotskie N, Stokes J. Factors of risk in the development of coronary heart disease: the Framingham Study. Ann Intern Med. 1961;55:33-50. doi:10.7326/0003-4819-55-1-33
15. Stamler J, Wentworth D, Neaton JD. Is the relationship between serum cholesterol and risk of premature death from coronary heart disease continuous and graded? Findings in 356,222 primary screenees of the Multiple Risk Factor Intervention Trial (MRFIT). JAMA. 1986;256:2823-2828. doi:10.1001/jama.1986.03380200061022
16. Kagan A, et al. Epidemiologic studies of coronary heart disease and stroke in Japanese men living in Japan, Hawaii and California. J Chronic Dis. 1974;27:345-364. doi:10.1016/0021-9681(74)90014-9
17. Kaplan H, et al. Coronary atherosclerosis in indigenous South American Tsimane: a cross-sectional cohort study. Lancet. 2017;389:1730-1739. doi:10.1016/S0140-6736(17)30752-3
18. Orlich MJ, et al. Vegetarian dietary patterns and mortality in Adventist Health Study 2. JAMA Intern Med. 2013;173:1230-1238. doi:10.1001/jamainternmed.2013.6473
19. Berenson GS, et al. Association between multiple cardiovascular risk factors and atherosclerosis in children and young adults. The Bogalusa Heart Study. N Engl J Med. 1998;338:1650-1656. doi:10.1056/NEJM199806043382302
20. Mensink RP, Zock PL, Kester AD, Katan MB. Effects of dietary fatty acids and carbohydrates on the ratio of serum total to HDL cholesterol and on serum lipids and apolipoproteins: a meta-analysis of 60 controlled trials. Am J Clin Nutr. 2003;77:1146-1155. doi:10.1093/ajcn/77.5.1146
21. Mensink RP; World Health Organization. Effects of Saturated Fatty Acids on Serum Lipids and Lipoproteins: A Systematic Review and Regression Analysis. Geneva: WHO; 2016. ISBN 978-92-4-156534-9. iris.who.int
22. Hegsted DM, McGandy RB, Myers ML, Stare FJ. Quantitative effects of dietary fat on serum cholesterol in man. Am J Clin Nutr. 1965;17:281-295. doi:10.1093/ajcn/17.5.281
23. Keys A, Anderson JT, Grande F. Serum cholesterol response to changes in the diet. Metabolism. 1965;14(7). PubMed search
24. Appel LJ, et al. A clinical trial of the effects of dietary patterns on blood pressure. DASH Collaborative Research Group. N Engl J Med. 1997;336:1117-1124. doi:10.1056/NEJM199704173361601
25. Jenkins DJ, et al. Effects of a dietary portfolio of cholesterol-lowering foods vs lovastatin on serum lipids and C-reactive protein. JAMA. 2003;290:502-510. doi:10.1001/jama.290.4.502
26. Ras RT, Geleijnse JM, Trautwein EA. LDL-cholesterol-lowering effect of plant sterols and stanols across different dose ranges: a meta-analysis of randomised controlled studies. Br J Nutr. 2014;112:214-219. doi:10.1017/S0007114514000750
27. Brown L, Rosner B, Willett WW, Sacks FM. Cholesterol-lowering effects of dietary fiber: a meta-analysis. Am J Clin Nutr. 1999;69:30-42. doi:10.1093/ajcn/69.1.30
28. Kapil V, Khambata RS, Robertson A, Caulfield MJ, Ahluwalia A. Dietary nitrate provides sustained blood pressure lowering in hypertensive patients: a randomized, phase 2, double-blind, placebo-controlled study. Hypertension. 2015;65:320-327. doi:10.1161/HYPERTENSIONAHA.114.04675
29. Scandinavian Simvastatin Survival Study Group. Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet. 1994;344:1383-1389. doi:10.1016/S0140-6736(94)90566-5
30. Cholesterol Treatment Trialists (CTT) Collaboration. Efficacy and safety of more intensive lowering of LDL cholesterol: a meta-analysis of data from 170,000 participants in 26 randomised trials. Lancet. 2010;376:1670-1681. doi:10.1016/S0140-6736(10)61350-5
31. Cannon CP, et al. Ezetimibe added to statin therapy after acute coronary syndromes (IMPROVE-IT). N Engl J Med. 2015;372:2387-2397. doi:10.1056/NEJMoa1410489
32. Sabatine MS, et al. Evolocumab and clinical outcomes in patients with cardiovascular disease (FOURIER). N Engl J Med. 2017;376:1713-1722. doi:10.1056/NEJMoa1615664
33. Leren P. The Oslo Diet-Heart Study: eleven-year report. Circulation. 1970;42:935-942. doi:10.1161/01.cir.42.5.935
34. Turpeinen O, et al. Dietary prevention of coronary heart disease: the Finnish Mental Hospital Study. Int J Epidemiol. 1979;8:99-118. doi:10.1093/ije/8.2.99
35. de Lorgeril M, et al. Mediterranean diet, traditional risk factors, and the rate of cardiovascular complications after myocardial infarction: final report of the Lyon Diet Heart Study. Circulation. 1999;99:779-785. doi:10.1161/01.cir.99.6.779
36. de Lorgeril M, Renaud S, Mamelle N, et al. Mediterranean alpha-linolenic acid-rich diet in secondary prevention of coronary heart disease. Lancet. 1994;343(8911):1454-1459. doi:10.1016/S0140-6736(94)92580-1
37. Estruch R, et al. Primary prevention of cardiovascular disease with a Mediterranean diet supplemented with extra-virgin olive oil or nuts. N Engl J Med. 2018;378:e34. doi:10.1056/NEJMoa1800389
38. Howard BV, et al. Low-fat dietary pattern and risk of cardiovascular disease: the Women’s Health Initiative Randomized Controlled Dietary Modification Trial. JAMA. 2006;295:655-666. doi:10.1001/jama.295.6.655
39. Ramsden CE, et al. Use of dietary linoleic acid for secondary prevention of coronary heart disease and death: evaluation of recovered data from the Sydney Diet Heart Study and updated meta-analysis. BMJ. 2013;346:e8707. doi:10.1136/bmj.e8707
40. Ramsden CE, et al. Re-evaluation of the traditional diet-heart hypothesis: analysis of recovered data from Minnesota Coronary Experiment (1968-73). BMJ. 2016;353:i1246. doi:10.1136/bmj.i1246
41. Alpha-Tocopherol, Beta Carotene Cancer Prevention Study Group. The effect of vitamin E and beta carotene on the incidence of lung cancer and other cancers in male smokers. N Engl J Med. 1994;330:1029-1035. doi:10.1056/NEJM199404143301501
42. Omenn GS, et al. Effects of a combination of beta carotene and vitamin A on lung cancer and cardiovascular disease (CARET). N Engl J Med. 1996;334:1150-1155. doi:10.1056/NEJM199605023341802
43. Ornish D, et al. Can lifestyle changes reverse coronary heart disease? The Lifestyle Heart Trial. Lancet. 1990;336:129-133. doi:10.1016/0140-6736(90)91656-U
44. Ornish D, et al. Intensive lifestyle changes for reversal of coronary heart disease. JAMA. 1998;280:2001-2007. doi:10.1001/jama.280.23.2001
45. Esselstyn CB Jr, Gendy G, Doyle J, Golubic M, Roizen MF. A way to reverse CAD? J Fam Pract. 2014;63:356-364b. PubMed 25198208
46. Nissen SE, et al. Effect of very high-intensity statin therapy on regression of coronary atherosclerosis: the ASTEROID trial. JAMA. 2006;295:1556-1565. doi:10.1001/jama.295.13.jpc60002
47. Nicholls SJ, et al. Effect of evolocumab on progression of coronary disease in statin-treated patients: the GLAGOV randomized clinical trial. JAMA. 2016;316:2373-2384. doi:10.1001/jama.2016.16951
48. Voight BF, et al. Plasma HDL cholesterol and risk of myocardial infarction: a mendelian randomisation study. Lancet. 2012;380:572-580. doi:10.1016/S0140-6736(12)60312-2
49. HPS3/TIMI55-REVEAL Collaborative Group. Effects of anacetrapib in patients with atherosclerotic vascular disease. N Engl J Med. 2017;377(13):1217-1227. doi:10.1056/NEJMoa1706444
50. Sacks FM, et al. Dietary fats and cardiovascular disease: a presidential advisory from the American Heart Association. Circulation. 2017;136:e1-e23. doi:10.1161/CIR.0000000000000510
51. Carson JAS, et al. Dietary cholesterol and cardiovascular risk: a science advisory from the American Heart Association. Circulation. 2020;141:e39-e53. doi:10.1161/CIR.0000000000000743
52. Grundy SM, et al. 2018 AHA/ACC multisociety guideline on the management of blood cholesterol. Circulation. 2019;139:e1082-e1143. doi:10.1161/CIR.0000000000000625
53. Mach F, et al. 2019 ESC/EAS guidelines for the management of dyslipidaemias: lipid modification to reduce cardiovascular risk. Eur Heart J. 2020;41:111-188. doi:10.1093/eurheartj/ehz455
54. World Health Organization. Saturated Fatty Acid and Trans-Fatty Acid Intake for Adults and Children: WHO Guideline. 2023.
55. Bouvard V, et al; International Agency for Research on Cancer Monograph Working Group. Carcinogenicity of consumption of red and processed meat. Lancet Oncol. 2015;16:1599-1600. doi:10.1016/S1470-2045(15)00444-1
56. Rock CL, et al. American Cancer Society guideline for diet and physical activity for cancer prevention. CA Cancer J Clin. 2020;70:245-271. doi:10.3322/caac.21591
57. Raj S, Guest NS, Landry MJ, Mangels AR, Pawlak R, Rozga M. Vegetarian dietary patterns for adults: a position paper of the Academy of Nutrition and Dietetics. J Acad Nutr Diet. 2025;125(6):831-846. doi:10.1016/j.jand.2025.02.002
58. American College of Lifestyle Medicine. Dietary Lifestyle Position Statement for Treatment and Potential Reversal of Disease.
59. US Department of Agriculture, US Department of Health and Human Services. Dietary Guidelines for Americans, 2025-2030. Released January 2026. dietaryguidelines.gov
60. Dietary Guidelines Advisory Committee. Scientific Report of the 2025 Dietary Guidelines Advisory Committee. 2024. doi:10.52570/DGAC2025
61. Willett W, et al. Food in the Anthropocene: the EAT-Lancet Commission on healthy diets from sustainable food systems. Lancet. 2019;393:447-492. doi:10.1016/S0140-6736(18)31788-4
62. Rockstrom J, Thilsted SH, Willett WC, et al. The EAT-Lancet Commission on healthy, sustainable, and just food systems. Lancet. 2025;406:1625-1700.
63. GBD 2017 Diet Collaborators (Afshin A, et al). Health effects of dietary risks in 195 countries, 1990-2017: a systematic analysis for the Global Burden of Disease Study. Lancet. 2019;393:1958-1972. doi:10.1016/S0140-6736(19)30041-8
64. Shivappa N, Steck SE, Hurley TG, Hussey JR, Hebert JR. Designing and developing a literature-derived, population-based dietary inflammatory index. Public Health Nutr. 2014;17:1689-1696. doi:10.1017/S1368980013002115
65. Hord NG, Tang Y, Bryan NS. Food sources of nitrates and nitrites: the physiologic context for potential health benefits. Am J Clin Nutr. 2009;90:1-10. doi:10.3945/ajcn.2008.27131
66. Gonzalez J, Wageman J, Buchanan L, Kalayjian T, Calkins M, Saner E, Soto-Mota A. Increased LDL-C reduction with ezetimibe in ketogenic diet-induced hypercholesterolemia. Lipids Health Dis. 2026. doi:10.1186/s12944-026-03034-w
67. Dehghan M, et al. Associations of fats and carbohydrate intake with cardiovascular disease and mortality in 18 countries from five continents (PURE): a prospective cohort study. Lancet. 2017;390:2050-2062. doi:10.1016/S0140-6736(17)32252-3
68. Health Canada. Canada’s Food Guide. 2019. food-guide.canada.ca
69. Kronenberg F, et al. Lipoprotein(a) in atherosclerotic cardiovascular disease and aortic stenosis: a European Atherosclerosis Society consensus statement. Eur Heart J. 2022;43:3925-3946. doi:10.1093/eurheartj/ehac361
70. Reyes-Soffer G, et al. Lipoprotein(a): a genetically determined, causal, and prevalent risk factor for atherosclerotic cardiovascular disease: a scientific statement from the American Heart Association. Arterioscler Thromb Vasc Biol. 2022;42:e48-e60. doi:10.1161/ATV.0000000000000147
71. Nissen SE, et al. Bempedoic acid and cardiovascular outcomes in statin-intolerant patients (CLEAR Outcomes). N Engl J Med. 2023;388:1353-1364. doi:10.1056/NEJMoa2215024
72. Zhong VW, et al. Associations of dietary cholesterol or egg consumption with incident cardiovascular disease and mortality. JAMA. 2019;321:1081-1095. doi:10.1001/jama.2019.1572
73. Drouin-Chartier JP, et al. Egg consumption and risk of cardiovascular disease: three large prospective US cohort studies, systematic review, and updated meta-analysis. BMJ. 2020;368:m513. doi:10.1136/bmj.m513
74. Varbo A, et al. Remnant cholesterol as a causal risk factor for ischemic heart disease. J Am Coll Cardiol. 2013;61:427-436. doi:10.1016/j.jacc.2012.08.1026
75. Nordestgaard BG. Triglyceride-rich lipoproteins and atherosclerotic cardiovascular disease: new insights from epidemiology, genetics, and biology. Circ Res. 2016;118:547-563. doi:10.1161/CIRCRESAHA.115.306249
76. Sniderman AD, et al. Apolipoprotein B particles and cardiovascular disease: a narrative review. JAMA Cardiol. 2019;4:1287-1295. doi:10.1001/jamacardio.2019.3780
77. Soto-Mota A, Norwitz NG, Manubolu VS, et al. RETRACTED: Longitudinal data from the KETO-CTA study: plaque predicts plaque, ApoB does not. JACC Adv. 2025;4:101686. (Retracted 2026 at the request of the authors and the editors.) doi:10.1016/j.jacadv.2025.101686
78. Editors of JACC: Advances. Expression of Concern: “Plaque Begets Plaque. ApoB Does Not: Longitudinal Data from The KETO-CTA Trial.” JACC Adv. 2026;5(3):102607. doi:10.1016/j.jacadv.2026.102607
79. Soto-Mota A, Norwitz NG, Manubolu VS, et al. Retraction: Longitudinal Data From the KETO-CTA Study: Plaque Predicts Plaque, ApoB Does Not. JACC Adv. 2026;5(5):102824. doi:10.1016/j.jacadv.2026.102824
80. Orrall A. Widely criticized keto diet study retracted. Retraction Watch. May 22, 2026. retractionwatch.com
81. Norwitz N. We Want to Retract Our Own Paper. StayCurious Metabolism (Substack). March 5, 2026. staycuriousmetabolism.substack.com
82. Soto-Mota A, Norwitz NG, Manubolu VS, Kinninger A, Wood TR, Earls J, Feldman D, Budoff M. Reply: The KETO CTA Study. JACC Adv. 2025;4(7):101862. doi:10.1016/j.jacadv.2025.101862
83. Budoff MJ, Kinninger A, Manubolu VS, Norwitz NG, Feldman D, Soto-Mota A. The impact of sustained LDL-C elevation on plaque changes: primary coronary plaque progression results from the Keto CTA study. medRxiv. Posted January 16, 2026. doi:10.64898/2026.01.15.26343955 (Preprint; not peer-reviewed.)
84. Zhang Y, Chadaideh KS, Li Y, et al. Butter and plant-based oils intake and mortality. JAMA Intern Med. 2025;185(5):549-560. doi:10.1001/jamainternmed.2025.0205
85. Hooper L, Martin N, Jimoh OF, Kirk C, Foster E, Abdelhamid AS. Reduction in saturated fat intake for cardiovascular disease. Cochrane Database Syst Rev. 2020;8:CD011737. doi:10.1002/14651858.CD011737.pub3
86. Anitschkow N, Chalatow S. On experimental cholesterin steatosis and its significance in the origin of some pathological processes (1913). English reprint: Arteriosclerosis. 1983;3:178-182. doi:10.1161/01.ATV.3.2.178
87. Dayton S, Pearce ML, Hashimoto S, Dixon WJ, Tomiyasu U. A controlled clinical trial of a diet high in unsaturated fat in preventing complications of atherosclerosis. Circulation. 1969;40(suppl II):II-1-II-63.
88. Astrup A, Magkos F, Bier DM, et al. Saturated fats and health: a reassessment and proposal for food-based recommendations. J Am Coll Cardiol. 2020;76:844-857. doi:10.1016/j.jacc.2020.05.077
89. O’Keefe JH Jr, Cordain L, Harris WH, Moe RM, Vogel R. Optimal low-density lipoprotein is 50 to 70 mg/dl: lower is better and physiologically normal. J Am Coll Cardiol. 2004;43(11):2142-2146. doi:10.1016/j.jacc.2004.03.046
90. Wang Z, Klipfell E, Bennett BJ, et al. Gut flora metabolism of phosphatidylcholine promotes cardiovascular disease. Nature. 2011;472(7341):57-63. doi:10.1038/nature09922
91. Koeth RA, Wang Z, Levison BS, et al. Intestinal microbiota metabolism of L-carnitine, a nutrient in red meat, promotes atherosclerosis. Nat Med. 2013;19(5):576-585. doi:10.1038/nm.3145
92. Heianza Y, Ma W, Manson JE, Rexrode KM, Qi L. Gut microbiota metabolites and risk of major adverse cardiovascular disease events and death: a systematic review and meta-analysis of prospective studies. J Am Heart Assoc. 2017;6(7):e004947. doi:10.1161/JAHA.116.004947
93. Schiattarella GG, Sannino A, Toscano E, et al. Gut microbe-generated metabolite trimethylamine-N-oxide as cardiovascular risk biomarker: a systematic review and dose-response meta-analysis. Eur Heart J. 2017;38(39):2948-2956. doi:10.1093/eurheartj/ehx342
94. Cho CE, Taesuwan S, Malysheva OV, et al. Trimethylamine-N-oxide (TMAO) response to animal source foods varies among healthy young men and is influenced by their gut microbiota composition: a randomized controlled trial. Mol Nutr Food Res. 2017;61(1):1600324. doi:10.1002/mnfr.201600324
95. Whitehead A, Beck EJ, Tosh S, Wolever TMS. Cholesterol-lowering effects of oat beta-glucan: a meta-analysis of randomized controlled trials. Am J Clin Nutr. 2014;100(6):1413-1421. doi:10.3945/ajcn.114.086108
96. Blanco Mejia S, Messina M, Li SS, et al. A meta-analysis of 46 studies identified by the FDA demonstrates that soy protein decreases circulating LDL and total cholesterol concentrations in adults. J Nutr. 2019;149(6):968-981. doi:10.1093/jn/nxz020
97. Hadi A, Askarpour M, Ziaei R, et al. Effect of flaxseed supplementation on lipid profile: an updated systematic review and dose-response meta-analysis of sixty-two randomized controlled trials. Pharmacol Res. 2020;152:104622. doi:10.1016/j.phrs.2019.104622
98. Schoeneck M, Iggman D. The effects of foods on LDL cholesterol levels: a systematic review of the accumulated evidence from systematic reviews and meta-analyses of randomized controlled trials. Nutr Metab Cardiovasc Dis. 2021;31(5):1325-1338. doi:10.1016/j.numecd.2020.12.032
99. Montorsi P, Ravagnani PM, Galli S, et al. The artery size hypothesis: a macrovascular link between erectile dysfunction and coronary artery disease. Am J Cardiol. 2005;96(12B):19M-23M. doi:10.1016/j.amjcard.2005.07.006
100. Dietary Guidelines Advisory Committee. Scientific Report of the 2015 Dietary Guidelines Advisory Committee. USDA and HHS; February 2015. Part D, Chapter 1. odphp.health.gov
101. US Department of Health and Human Services, US Department of Agriculture. 2015-2020 Dietary Guidelines for Americans. 8th ed. Released January 2016. Chapter 1. dietaryguidelines.gov
102. US Department of Agriculture, US Department of Health and Human Services. Dietary Guidelines for Americans, 2020-2025. 9th ed. December 2020. dietaryguidelines.gov
103. Eckel RH, Jakicic JM, Ard JD, et al. 2013 AHA/ACC guideline on lifestyle management to reduce cardiovascular risk. Circulation. 2014;129(25 Suppl 2):S76-S99. doi:10.1161/01.cir.0000437740.48606.d1
104. Vincent MJ, Allen B, Palacios OM, Haber LT, Maki KC. Meta-regression analysis of the effects of dietary cholesterol intake on LDL and HDL cholesterol. Am J Clin Nutr. 2019;109(1):7-16. doi:10.1093/ajcn/nqy273
105. American College of Cardiology. How do the 2025-2030 Dietary Guidelines for Americans measure up for cardiovascular health? Expert Analysis. January 27, 2026.
106. Dietary Guidelines Advisory Committee, Nutrition Evidence Systematic Review. Food Sources of Saturated Fat and Risk of Cardiovascular Disease: A Systematic Review. USDA; 2024. doi:10.52570/NESR.DGAC2025.SR24
107. Dietary Guidelines Advisory Committee, Nutrition Evidence Systematic Review. Dietary Patterns and Risk of Cardiovascular Disease: A Systematic Review. USDA; 2024. doi:10.52570/NESR.DGAC2025.SR13
108. Dietary Guidelines Advisory Committee, Nutrition Evidence Systematic Review. Types of Dietary Fat and Cardiovascular Disease: A Systematic Review. USDA; 2020. doi:10.52570/NESR.DGAC2020.SR0501
109. Satija A, Bhupathiraju SN, Spiegelman D, et al. Healthful and unhealthful plant-based diets and the risk of coronary heart disease in U.S. adults. J Am Coll Cardiol. 2017;70(4):411-422. doi:10.1016/j.jacc.2017.05.047
110. Kim H, Caulfield LE, Garcia-Larsen V, Steffen LM, Coresh J, Rebholz CM. Plant-based diets are associated with a lower risk of incident cardiovascular disease, cardiovascular disease mortality, and all-cause mortality in a general population of middle-aged adults. J Am Heart Assoc. 2019;8(16):e012865. doi:10.1161/JAHA.119.012865
111. Shan Z, Wang F, Li Y, et al. Healthy eating patterns and risk of total and cause-specific mortality. JAMA Intern Med. 2023;183(2):142-153. doi:10.1001/jamainternmed.2022.6117
112. Glenn AJ, Lo K, Jenkins DJA, et al. Relationship between a plant-based dietary portfolio and risk of cardiovascular disease: findings from the Women’s Health Initiative prospective cohort study. J Am Heart Assoc. 2021;10(16):e021515. doi:10.1161/JAHA.120.021515
113. Brown MJ, Ferruzzi MG, Nguyen ML, Cooper DA, Eldridge AL, Schwartz SJ, White WS. Carotenoid bioavailability is higher from salads ingested with full-fat than with fat-reduced salad dressings as measured with electrochemical detection. Am J Clin Nutr. 2004;80(2):396-403. doi:10.1093/ajcn/80.2.396
114. Unlu NZ, Bohn T, Clinton SK, Schwartz SJ. Carotenoid absorption from salad and salsa by humans is enhanced by the addition of avocado or avocado oil. J Nutr. 2005;135(3):431-436. doi:10.1093/jn/135.3.431
115. White WS, Zhou Y, Crane A, Dixon P, Quadt F, Flendrig LM. Modeling the dose effects of soybean oil in salad dressing on carotenoid and fat-soluble vitamin bioavailability in salad vegetables. Am J Clin Nutr. 2017;106(4):1041-1051. doi:10.3945/ajcn.117.153635
116. Gartner C, Stahl W, Sies H. Lycopene is more bioavailable from tomato paste than from fresh tomatoes. Am J Clin Nutr. 1997;66(1):116-122. doi:10.1093/ajcn/66.1.116
117. Reboul E. Mechanisms of carotenoid intestinal absorption: where do we stand? Nutrients. 2019;11(4):838. doi:10.3390/nu11040838
118. Goulinet S, Chapman MJ. Plasma LDL and HDL subspecies are heterogeneous in particle content of tocopherols and oxygenated and hydrocarbon carotenoids. Arterioscler Thromb Vasc Biol. 1997;17(4):786-796. doi:10.1161/01.ATV.17.4.786
119. Theobald HE, Chowienczyk PJ, Whittall R, Humphries SE, Sanders TAB. LDL cholesterol-raising effect of low-dose docosahexaenoic acid in middle-aged men and women. Am J Clin Nutr. 2004;79(4):558-563. doi:10.1093/ajcn/79.4.558
120. Bernstein AM, Ding EL, Willett WC, Rimm EB. A meta-analysis shows that docosahexaenoic acid from algal oil reduces serum triglycerides and increases HDL-cholesterol and LDL-cholesterol in persons without coronary heart disease. J Nutr. 2012;142(1):99-104. doi:10.3945/jn.111.148973
121. Buettner GR. The pecking order of free radicals and antioxidants: lipid peroxidation, alpha-tocopherol, and ascorbate. Arch Biochem Biophys. 1993;300(2):535-543. doi:10.1006/abbi.1993.1074
122. Mohr D, Bowry VW, Stocker R. Dietary supplementation with coenzyme Q10 results in increased levels of ubiquinol-10 within circulating lipoproteins and increased resistance of human low-density lipoprotein to the initiation of lipid peroxidation. Biochim Biophys Acta. 1992;1126(3):247-254. PubMed 1637852
123. Brand MD. Mitochondrial generation of superoxide and hydrogen peroxide as the source of mitochondrial redox signaling. Free Radic Biol Med. 2016;100:14-31. doi:10.1016/j.freeradbiomed.2016.04.001
124. Tabas I, Williams KJ, Boren J. Subendothelial lipoprotein retention as the initiating process in atherosclerosis: update and therapeutic implications. Circulation. 2007;116(16):1832-1844. doi:10.1161/CIRCULATIONAHA.106.676890
125. Linus Pauling Institute, Oregon State University. Vitamin E. Micronutrient Information Center. lpi.oregonstate.edu
126. Novartis. Novartis announces Lp(a)HORIZON Phase III topline results for pelacarsen in patients with elevated Lp(a) and established cardiovascular disease. Media release, Basel, September 4, 2026. novartis.com
127. Blumenthal RS, Morris PB, Gaudino M, et al. 2026 ACC/AHA/AACVPR/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA guideline on the management of dyslipidemia. Circulation. Published online March 13, 2026. doi:10.1161/CIR.0000000000001423
128. Tsimikas S, Gordts PLSM, Nora C, Yeang C, Witztum JL. Statin therapy increases lipoprotein(a) levels. Eur Heart J. 2020;41(24):2275-2284. doi:10.1093/eurheartj/ehz310
129. Robertson TL, Kato H, Rhoads GG, et al. Epidemiologic studies of coronary heart disease and stroke in Japanese men living in Japan, Hawaii and California: incidence of myocardial infarction and death from coronary heart disease. Am J Cardiol. 1977;39(2):239-243. doi:10.1016/S0002-9149(77)80197-5
130. Dayton S, Pearce ML, Goldman H, et al. Controlled trial of a diet high in unsaturated fat for prevention of atherosclerotic complications. Lancet. 1968;2(7577):1060-1062. doi:10.1016/S0140-6736(68)91531-6
131. Expert Panel on Integrated Guidelines for Cardiovascular Health and Risk Reduction in Children and Adolescents; National Heart, Lung, and Blood Institute. Summary report. Pediatrics. 2011;128(Suppl 5):S213-S256. doi:10.1542/peds.2009-2107C
132. Turley SD, Dietschy JM. The intestinal absorption of biliary and dietary cholesterol as a drug target for lowering the plasma cholesterol level. Prev Cardiol. 2003. doi:10.1111/j.1520-037X.2003.01691.x
133. Abifadel M, Varret M, Rabès JP, et al. Mutations in PCSK9 cause autosomal dominant hypercholesterolemia. Nat Genet. 2003;34(2):154-156. PubMed 12730697
134. Fernández-Friera L, Fuster V, López-Melgar B, et al. Normal LDL-cholesterol levels are associated with subclinical atherosclerosis in the absence of risk factors. J Am Coll Cardiol. 2017;70(24):2979-2991. doi:10.1016/j.jacc.2017.10.024
135. Sachdeva A, Cannon CP, Deedwania PC, et al. Lipid levels in patients hospitalized with coronary artery disease: an analysis of 136,905 hospitalizations in Get With The Guidelines. Am Heart J. 2009;157(1):111-117.e2. doi:10.1016/j.ahj.2008.08.010
136. Bohula EA, Marston NA, Bhatia AK, et al. Evolocumab in patients without a previous myocardial infarction or stroke (VESALIUS-CV). N Engl J Med. 2026;394:117-127. doi:10.1056/NEJMoa2514428
137. Lee YJ, Lee SJ, Kim JW, et al. Intensive LDL cholesterol targeting in atherosclerotic cardiovascular disease (Ez-PAVE). N Engl J Med. 2026;394:1365-1375. doi:10.1056/NEJMoa2600283
138. Giugliano RP, Mach F, Zavitz K, et al. Cognitive function in a randomized trial of evolocumab (EBBINGHAUS). N Engl J Med. 2017;377(7):633-643. doi:10.1056/NEJMoa1701131
139. Ternhamar T, et al. Association between timing of statin treatment after diabetes diagnosis and risk of dementia: a nationwide observational study. Lancet Reg Health Eur. 2026;69:101814. doi:10.1016/j.lanepe.2026.101814
140. Nielsen SF, Nordestgaard BG. Negative statin-related news stories decrease statin persistence and increase myocardial infarction and cardiovascular mortality: a nationwide prospective cohort study. Eur Heart J. 2016;37:908-916. doi:10.1093/eurheartj/ehv641
141. Vafai SB, Täubel J, Ashdown T, et al. In vivo base editing of PCSK9 with VERVE-102 for hypercholesterolemia. N Engl J Med. 2026. doi:10.1056/NEJMoa2601283
142. Delgado-Lista J, Alcala-Diaz JF, Torres-Peña JD, et al. Long-term secondary prevention of cardiovascular disease with a Mediterranean diet and a low-fat diet (CORDIOPREV): a randomised controlled trial. Lancet. 2022;399:1876-1885. doi:10.1016/S0140-6736(22)00122-2
143. Yeagle PL. Unraveling a mystery: why human cells require cholesterol. Sci Adv. 2022;8:eade5927. doi:10.1126/sciadv.ade5927
144. Ohkawa R, Low H, Mukhamedova N, et al. Cholesterol transport between red blood cells and lipoproteins contributes to cholesterol metabolism in blood. J Lipid Res. 2020;61(12):1577-1588. doi:10.1194/jlr.RA120000635
145. Fessler MB, Rose K, Zhang Y, Jaramillo R, Zeldin DC. Relationship between serum cholesterol and indices of erythrocytes and platelets in the US population. J Lipid Res. 2013;54:3177-3188. doi:10.1194/jlr.P037614
146. Mason P. The truth about high cholesterol (lecture). YouTube. youtube.com
147. Carvalho. “High cholesterol is healthy!” (Myth finally explained). YouTube. youtube.com
148. Ravnskov U, Diamond DM, Hama R, et al. Lack of an association or an inverse association between low-density-lipoprotein cholesterol and mortality in the elderly: a systematic review. BMJ Open. 2016;6:e010401. doi:10.1136/bmjopen-2015-010401
149. Johannesen CDL, Langsted A, Mortensen MB, Nordestgaard BG. Association between low density lipoprotein and all cause and cause specific mortality in Denmark: prospective cohort study. BMJ. 2020;371:m4266. doi:10.1136/bmj.m4266
150. Flegal KM, Kit BK, Orpana H, Graubard BI. Association of all-cause mortality with overweight and obesity using standard body mass index categories: a systematic review and meta-analysis. JAMA. 2013;309(1):71-82. doi:10.1001/jama.2012.113905
151. Ravindrarajah R, Hazra NC, Hamada S, et al. Systolic blood pressure trajectory, frailty, and all-cause mortality >80 years of age: cohort study using electronic health records. Circulation. 2017;135:2357-2368. doi:10.1161/CIRCULATIONAHA.116.026687
152. Aggarwal V, Schneider ALC, Selvin E. Low hemoglobin A1c in nondiabetic adults: an elevated risk state? Diabetes Care. 2012;35(10):2055-2060. doi:10.2337/dc11-2531
153. Schöttker B, Rathmann W, Herder C, et al. HbA1c levels in non-diabetic older adults: no J-shaped associations with primary cardiovascular events, cardiovascular and all-cause mortality after adjustment for confounders in a meta-analysis of individual participant data from six cohort studies. BMC Med. 2016;14:26. doi:10.1186/s12916-016-0570-1
154. Hirshowitz B, Brook JG, Kaufman T, Titelman U, Mahler D. 35 eggs per day in the treatment of severe burns. Br J Plast Surg. PubMed search
155. Kaufman T, et al. Effect of an egg-rich diet on plasma lipids and proteins in severely burned patients. Isr J Med Sci. 1978;14:736. PubMed search
156. Li MY, Chen JH, Chen C, Kang YN. Association between egg consumption and cholesterol concentration: a systematic review and meta-analysis of randomized controlled trials. Nutrients. 2020;12:1995. doi:10.3390/nu12071995
157. Weggemans RM, Zock PL, Katan MB. Dietary cholesterol from eggs increases the ratio of total cholesterol to high-density lipoprotein cholesterol in humans: a meta-analysis. Am J Clin Nutr. 2001;73:885-891. PubMed search
158. Khaw KT, Sharp SJ, Finikarides L, et al. Randomised trial of coconut oil, olive oil or butter on blood lipids and other cardiovascular risk factors in healthy men and women. BMJ Open. 2018;8:e020167. doi:10.1136/bmjopen-2017-020167
159. Schwingshackl L, Bogensberger B, Benčič A, Knüppel S, Boeing H, Hoffmann G. Effects of oils and solid fats on blood lipids: a systematic review and network meta-analysis. J Lipid Res. 2018;59:1771-1782. PubMed search
160. Wu D, Yang Q, Su B, et al. Low-density lipoprotein cholesterol 4: the notable risk factor of coronary artery disease development. Front Cardiovasc Med. 2021;8:619386. doi:10.3389/fcvm.2021.619386
161. Aday AW, Lawler PR, Cook NR, Ridker PM, Mora S, Pradhan AD. Lipoprotein particle profiles, standard lipids, and peripheral artery disease incidence. Circulation. 2018;138(21):2330-2341. doi:10.1161/CIRCULATIONAHA.118.035432
162. Higashioka M, Sakata S, Honda T, et al. Small dense low-density lipoprotein cholesterol and the risk of coronary heart disease in a Japanese community. J Atheroscler Thromb. 2020;27:669-682. doi:10.5551/jat.51961
163. Liou L, Kaptoge S. Association of small, dense LDL-cholesterol concentration and lipoprotein particle characteristics with coronary heart disease: a systematic review and meta-analysis. PLoS One. 2020;15(11):e0241993. doi:10.1371/journal.pone.0241993
164. Johannesen CDL, Mortensen MB, Langsted A, Nordestgaard BG. Excess apolipoprotein B and cardiovascular risk in women and men. J Am Coll Cardiol. 2024;83(23):2262-2273. doi:10.1016/j.jacc.2024.03.423
165. Wilkins JT, Li RC, Sniderman A, Chan C, Lloyd-Jones DM. Discordance between apolipoprotein B and LDL-cholesterol in young adults predicts coronary artery calcification: the CARDIA study. J Am Coll Cardiol. 2016. doi:10.1016/j.jacc.2015.10.055
166. King RI, Florkowski CM, Yeo J, Walmsley TA, Shand BI, Scott RS, George PM. What is the best predictor of the atherogenic LDL subclass phenotype ‘pattern B’ in patients with type 2 diabetes mellitus? Ann Clin Biochem. 2011;48:166-169. doi:10.1258/acb.2010.010185
167. Adaikalakoteswari A, Finer S, Voyias PD, et al. Vitamin B12 insufficiency induces cholesterol biosynthesis by limiting s-adenosylmethionine and modulating the methylation of SREBF1 and LDLR genes. Clin Epigenetics. 2015;7:14. doi:10.1186/s13148-015-0046-8
168. Qin B, Anderson RA, Adeli K. Tumor necrosis factor-alpha directly stimulates the overproduction of hepatic apolipoprotein B100-containing VLDL via impairment of hepatic insulin signaling. Am J Physiol Gastrointest Liver Physiol. 2008;294:G1120-G1129. doi:10.1152/ajpgi.00407.2007
169. Tucker LA, Erickson A, LeCheminant JD, Bailey BW. Dairy consumption and insulin resistance: the role of body fat, physical activity, and energy intake. J Diabetes Res. 2015;2015:206959. doi:10.1155/2015/206959
170. Turner KM, Keogh JB, Clifton PM. Red meat, dairy, and insulin sensitivity: a randomized crossover intervention study. Am J Clin Nutr. 2015;101:1173-1179. PubMed search
171. Ference BA, Majeed F, Penumetcha R, Flack JM, Brook RD. Effect of naturally random allocation to lower low-density lipoprotein cholesterol on the risk of coronary heart disease mediated by polymorphisms in NPC1L1, HMGCR, or both. J Am Coll Cardiol. 2015. doi:10.1016/j.jacc.2015.02.020
172. Li Y, Hruby A, Bernstein AM, et al. Saturated fat as compared with unsaturated fats and sources of carbohydrates in relation to risk of coronary heart disease: a prospective cohort study. J Am Coll Cardiol. 2015;66(14):1538-1548. doi:10.1016/j.jacc.2015.07.055
173. Steen JP, Klatt KC, Chang Y, et al. Effect of interventions aimed at reducing or modifying saturated fat intake on cholesterol, mortality, and major cardiovascular events: a risk stratified systematic review of randomized trials. Ann Intern Med. 2026;179(2):242-255. doi:10.7326/ANNALS-25-02229
174. Khan SS, Matsushita K, Sang Y, et al. Development and validation of the American Heart Association’s PREVENT equations. Circulation. 2024;149(6):430-449. doi:10.1161/CIRCULATIONAHA.123.067626
175. Verhoeven N (Physionic). Video critique of Dr. Fung’s argument that blood cholesterol is a minor risk factor. YouTube. youtube.com
176. Bundgaard H, García-Lunar I, Kofoed KF, et al. Prevalence of silent atherosclerosis across adult life (REACT). N Engl J Med. 2026. doi:10.1056/NEJMoa2609059
177. Individual variation in the distribution of apolipoprotein B levels across the spectrum of LDL-C or non-HDL-C levels. JAMA Cardiol. 2024;9(8):741-747. doi:10.1001/jamacardio.2024.1310
178. ApoB, LDL-C, and non-HDL-C as markers of cardiovascular risk. J Clin Lipidol. 2025;19(4):844-859. doi:10.1016/j.jacl.2025.05.024
179. Schwartz GG, Steg PG, Szarek M, et al. Alirocumab and cardiovascular outcomes after acute coronary syndrome (ODYSSEY OUTCOMES). N Engl J Med. 2018;379(22):2097-2107. doi:10.1056/NEJMoa1801174
180. Cholesterol Treatment Trialists’ (CTT) Collaborators. The effects of lowering LDL cholesterol with statin therapy in people at low risk of vascular disease: meta-analysis of individual data from 27 randomised trials. Lancet. 2012;380:581-590. doi:10.1016/S0140-6736(12)60367-5
181. @ironDPC. Post on X about changing his view of ApoB and lipid lowering. x.com
182. Mortensen MB, Nordestgaard BG. Elevated LDL cholesterol and increased risk of myocardial infarction and atherosclerotic cardiovascular disease in individuals aged 70-100 years: a contemporary primary prevention cohort. Lancet. 2020;396(10263):1644-1652. doi:10.1016/S0140-6736(20)32233-9
183. Gencer B, Marston NA, Im K, et al. Efficacy and safety of lowering LDL cholesterol in older patients: a systematic review and meta-analysis of randomised controlled trials. Lancet. 2020;396(10263):1637-1643. doi:10.1016/S0140-6736(20)32332-1
184. Domanski MJ, Tian X, et al. Time course of LDL cholesterol exposure and cardiovascular disease event risk. J Am Coll Cardiol. 2020;76(13):1507-1516. doi:10.1016/j.jacc.2020.07.059
185. Marston NA, Giugliano RP, Melloni GEM, et al. Association of apolipoprotein B-containing lipoproteins and risk of myocardial infarction in individuals with and without atherosclerosis: distinguishing between particle concentration, type, and content. JAMA Cardiol. 2022;7(3):250-256. doi:10.1001/jamacardio.2021.5083
186. Wood FA, Howard JP, Finegold JA, et al. N-of-1 trial of a statin, placebo, or no treatment to assess side effects (SAMSON). N Engl J Med. 2020;383:2182-2184. doi:10.1056/NEJMc2031173
187. Cholesterol Treatment Trialists’ Collaboration. Effect of statin therapy on muscle symptoms: an individual participant data meta-analysis of large-scale, randomised, double-blind trials. Lancet. 2022;400:832-845. doi:10.1016/S0140-6736(22)01545-8
188. Cholesterol Treatment Trialists’ (CTT) Collaboration. Assessment of adverse effects attributed to statin therapy in product labels: a meta-analysis of double-blind randomised controlled trials. Lancet. 2026. doi:10.1016/S0140-6736(25)01578-8
189. Motley J. The Number You’re Being Told to Ignore. Unscripted (Substack). October 1, 2026. jaymotley.substack.com
190. Zhang J, Li L, Song P, et al. Randomized controlled trial of oatmeal consumption versus noodle consumption on blood lipids of urban Chinese adults with hypercholesterolemia. Nutr J. 2012;11:54. PubMed search
191. Thongoun P, Pavadhgul P, Bumrungpert A, et al. Effect of oat consumption on lipid profiles in hypercholesterolemic adults. J Med Assoc Thai. 2013;96(Suppl 5). PubMed search
192. Xu D, Feng M, Chu Y, et al. The prebiotic effects of oats on blood lipids, gut microbiota, and short-chain fatty acids in mildly hypercholesterolemic subjects compared with rice: a randomized, controlled trial. Front Immunol. 2021;12:787797. doi:10.3389/fimmu.2021.787797
193. Helnæs A, Kyrø C, Andersen I, et al. Intake of whole grains is associated with lower risk of myocardial infarction: the Danish Diet, Cancer and Health Cohort. Am J Clin Nutr. 2016. doi:10.3945/ajcn.115.124271
194. Klümpen L, Mantri A, Philipps M, et al. Cholesterol-lowering effects of oats induced by microbially produced phenolic metabolites in metabolic syndrome: a randomized controlled trial. Nat Commun. 2026;17:598. doi:10.1038/s41467-026-68303-9
195. Verhoeven N (Physionic). Eating oats releases molecules that impact our heart. YouTube. youtube.com
196. Hu Y, Willett WC, Manson JE, Rosner B, Hu FB, Sun Q. Intake of whole grain foods and risk of coronary heart disease in US men and women. BMC Med. 2022;20:192. doi:10.1186/s12916-022-02396-z
197. Key TJ, et al. Consumption of meat, fish, dairy products, and eggs and risk of ischemic heart disease: a prospective study of 7198 incident cases among 409 885 participants in the pan-European EPIC cohort. Circulation. 2019;139:2835-2845. doi:10.1161/CIRCULATIONAHA.118.038813
198. Carter S, Hill AM, Yandell C, et al. Impact of dietary cholesterol from eggs and saturated fat on LDL cholesterol levels: a randomized cross-over study. Am J Clin Nutr. 2025;122:83-91. doi:10.1016/j.ajcnut.2025.05.001
199. Spence JD, Srichaikul K, Jenkins DJA. Cardiovascular harm from egg yolk and meat: more than just cholesterol and saturated fat. J Am Heart Assoc. 2021;10:e017066. doi:10.1161/JAHA.120.017066
200. Schmidt KA, Cromer G, Burhans MS, et al. Impact of low-fat and full-fat dairy foods on fasting lipid profile and blood pressure: exploratory endpoints of a randomized controlled trial. Am J Clin Nutr. 2021;114:882-892. PubMed search
201. Zhong VW, Van Horn L, Greenland P, et al. Associations of processed meat, unprocessed red meat, poultry, or fish intake with incident cardiovascular disease and all-cause mortality. JAMA Intern Med. 2020;180(4):503-512. doi:10.1001/jamainternmed.2019.6969
202. Al-Shaar L, Satija A, Wang DD, et al. Red meat intake and risk of coronary heart disease among US men: prospective cohort study. BMJ. 2020;371:m4141. doi:10.1136/bmj.m4141
203. Papier K, Knuppel A, Syam N, Jebb SA, Key TJ. Meat consumption and risk of ischemic heart disease: a systematic review and meta-analysis. Crit Rev Food Sci Nutr. 2023;63(3):426-437. doi:10.1080/10408398.2021.1949575
204. Shi W, Huang X, Schooling CM, Zhao JV. Red meat consumption, cardiovascular diseases, and diabetes: a systematic review and meta-analysis. Eur Heart J. 2023;44(28):2626-2635. doi:10.1093/eurheartj/ehad336
205. Micha R, Wallace SK, Mozaffarian D. Red and processed meat consumption and risk of incident coronary heart disease, stroke, and diabetes mellitus: a systematic review and meta-analysis. Circulation. 2010;121(21):2271-2283. doi:10.1161/CIRCULATIONAHA.109.924977
206. Iqbal R, Dehghan M, Mente A, et al. Associations of unprocessed and processed meat intake with mortality and cardiovascular disease in 21 countries (Prospective Urban Rural Epidemiology [PURE] Study): a prospective cohort study. Am J Clin Nutr. 2021;114:1049-1058. PubMed search
207. Red meat intake and the risk of cardiovascular diseases: a prospective cohort study in the Million Veteran Program. J Nutr. 2024;154:886-895. doi:10.1016/j.tjnut.2023.12.051
208. Lescinsky H, Afshin A, Ashbaugh C, et al. Health effects associated with consumption of unprocessed red meat: a Burden of Proof study. Nat Med. 2022;28:2075-2082. doi:10.1038/s41591-022-01968-z
209. Concerns about the Burden of Proof studies. Nat Med. 2023. doi:10.1038/s41591-023-02294-8
210. Zeraatkar D, Han MA, Guyatt GH, et al. Red and processed meat consumption and risk for all-cause mortality and cardiometabolic outcomes: a systematic review and meta-analysis of cohort studies. Ann Intern Med. 2019;171:703-710. doi:10.7326/M19-0655
211. Johnston BC, Zeraatkar D, Han MA, et al. Unprocessed red meat and processed meat consumption: dietary guideline recommendations from the Nutritional Recommendations (NutriRECS) Consortium. Ann Intern Med. 2019;171:756-764. doi:10.7326/M19-1621
212. Bergeron N, Chiu S, Williams PT, King SM, Krauss RM. Effects of red meat, white meat, and nonmeat protein sources on atherogenic lipoprotein measures in the context of low compared with high saturated fat intake: a randomized controlled trial. Am J Clin Nutr. 2019;110:24-33. PubMed search
213. Guasch-Ferré M, Satija A, Blondin SA, et al. Meta-analysis of randomized controlled trials of red meat consumption in comparison with various comparison diets on cardiovascular risk factors. Circulation. 2019;139:1828-1845. doi:10.1161/CIRCULATIONAHA.118.035225
214. Roussell MA, Hill AM, Gaugler TL, et al. Beef in an Optimal Lean Diet study: effects on lipids, lipoproteins, and apolipoproteins. Am J Clin Nutr. 2012;95:9-16. doi:10.3945/ajcn.111.016261
215. López-Moreno M, et al. Industry study sponsorship and conflicts of interest on the effect of unprocessed red meat on cardiovascular disease risk: a systematic review of clinical trials. Am J Clin Nutr. 2025;121:1246-1257. doi:10.1016/j.ajcnut.2025.02.030
216. Structural funding bias in red meat research: the elephant in the room (letter). Am J Clin Nutr. 2025;122:1134-1135. PMC12674065
217. Neuenschwander M, et al. Substitution of animal-based with plant-based foods on cardiometabolic health and all-cause mortality: a systematic review and meta-analysis of prospective studies. BMC Med. 2023;21:404. doi:10.1186/s12916-023-03093-1
218. Wang Z, Bergeron N, Levison BS, et al. Impact of chronic dietary red meat, white meat, or non-meat protein on trimethylamine N-oxide metabolism and renal excretion in healthy men and women. Eur Heart J. 2019;40:583-594. doi:10.1093/eurheartj/ehy799
219. Fang X, An P, Wang H, et al. Dietary intake of heme iron and risk of cardiovascular disease: a dose-response meta-analysis of prospective cohort studies. Nutr Metab Cardiovasc Dis. 2015;25:24-35. doi:10.1016/j.numecd.2014.09.002
220. Bitterman H. Heme iron and cardiovascular disease: required refinements (letter). Nutr Metab Cardiovasc Dis. 2020. doi:10.1016/j.numecd.2020.03.002
221. Dietary intake of heme iron is associated with increased cardiovascular disease risk: reply to Dr. Bitterman. Nutr Metab Cardiovasc Dis. 2020;30:1053-1055. doi:10.1016/j.numecd.2020.03.007
222. Alshahrani SM, Fraser GE, Sabaté J, et al. Red and processed meat and mortality in a low meat intake population. Nutrients. 2019;11:622. doi:10.3390/nu11030622
223. O’Connor LE, Paddon-Jones D, Wright AJ, Campbell WW. A Mediterranean-style eating pattern with lean, unprocessed red meat has cardiometabolic benefits for adults who are overweight or obese in a randomized, crossover, controlled feeding trial. Am J Clin Nutr. 2018;108:33-40. PubMed search
224. Liu Y, Liu F, Zhang L, et al. Association between low density lipoprotein cholesterol and all-cause mortality: results from the NHANES 1999-2014. Sci Rep. 2021;11:22111. doi:10.1038/s41598-021-01738-w
225. SPRINT Research Group. A randomized trial of intensive versus standard blood-pressure control. N Engl J Med. 2015;373(22):2103-2116. doi:10.1056/NEJMoa1511939
226. Lichtenstein AH, Khera A, Anderson CAM, et al. 2026 dietary guidance to improve cardiovascular health: a scientific statement from the American Heart Association. Circulation. 2026;153:e1285-e1295. doi:10.1161/CIR.0000000000001435
227. U.S. Department of Health and Human Services and U.S. Department of Agriculture. The Scientific Foundation for the Dietary Guidelines for Americans, 2025-2030. realfood.gov
228. Petersen MC, Smith GI, Farabi SS, et al. Effect of diet macronutrient content on the cardiometabolic response to weight loss: a randomized clinical trial. Cell Metab. 2026;38:2006-2017. doi:10.1016/j.cmet.2026.07.020
229. Hu X, Ren H, Cao Y. The association between trimethylamine N-oxide levels and ischemic stroke occurrence: a meta-analysis and Mendelian randomization study. BMC Neurol. 2023;23:413. doi:10.1186/s12883-023-03458-2
230. May JM, Mendiratta S, Hill KE, Burk RF. Reduction of dehydroascorbate to ascorbate by the selenoenzyme thioredoxin reductase. J Biol Chem. 1997;272(36):22607-22610. PubMed search
231. Cross AJ, Pollock JRA, Bingham SA. Haem, not protein or inorganic iron, is responsible for endogenous intestinal N-nitrosation arising from red meat. Cancer Res. 2003;63(10):2358-2360. PubMed search
232. Kuhnle GG, Story GW, Reda T, et al. Diet-induced endogenous formation of nitroso compounds in the GI tract. Free Radic Biol Med. 2007;43(7):1040-1047. doi:10.1016/j.freeradbiomed.2007.03.011
233. Mirvish SS, Wallcave L, Eagen M, Shubik P. Ascorbate-nitrite reaction: possible means of blocking the formation of carcinogenic N-nitroso compounds. Science. 1972;177(4043):65-68. doi:10.1126/science.177.4043.65
234. Vermeer ITM, Moonen EJC, Dallinga JW, Kleinjans JCS, van Maanen JMS. Effects of ascorbic acid and green tea on endogenous formation of N-nitrosodimethylamine and N-nitrosopiperidine in humans. Mutat Res. 1999;428:353-361. doi:10.1016/S1383-5742(99)00061-7
235. Physicians Committee for Responsible Medicine. Physicians sue government to withdraw Dietary Guidelines for Americans. News release, August 19, 2026. pcrm.org; complaint: pcrm.widen.net
236. California Healthline. USDA: judge rules in favor of PCRM on dietary guidelines. October 4, 2000 (summarizing the Washington Post). californiahealthline.org
237. American Diabetes Association Professional Practice Committee. 5. Facilitating positive health behaviors and well-being to improve health outcomes: Standards of Care in Diabetes 2026. Diabetes Care. 2026;49(Suppl 1):S89-S131. doi:10.2337/dc26-S005
238. Institute of Medicine. Dietary Reference Intakes: The Essential Guide to Nutrient Requirements. National Academies Press; 2006. Cholesterol chapter. nap.edu
239. Heart Outcomes Prevention Evaluation Study Investigators; Yusuf S, Dagenais G, Pogue J, Bosch J, Sleight P. Vitamin E supplementation and cardiovascular events in high-risk patients. N Engl J Med. 2000;342(3):154-160. doi:10.1056/NEJM200001203420302
240. HOPE and HOPE-TOO Trial Investigators; Lonn E, Bosch J, Yusuf S, et al. Effects of long-term vitamin E supplementation on cardiovascular events and cancer: a randomized controlled trial. JAMA. 2005;293(11):1338-1347. doi:10.1001/jama.293.11.1338

Last updated: Wednesday, October 7, 2026 at 7:44 PM EDT