Key takeaways
- The 2026 cholesterol guideline from the American College of Cardiology, the American Heart Association and nine other groups recommends that every adult have lipoprotein(a), or Lp(a), measured at least once12.
- About 20% of people worldwide have a high Lp(a) level, which is mostly inherited and barely changes with diet or exercise268.
- In September 2026, a trial of 8,323 people found that the Lp(a)-lowering drug pelacarsen didn’t reduce a combined count of heart attacks, strokes, heart deaths and urgent artery procedures, so for now a high result mainly tells you to push harder on everything else19.
- A drug that doubled “good” HDL cholesterol in 30,449 adults with artery disease cut major coronary events by about 11 per 1,000 people over four years, by our math, a benefit that tracked its modest drop in non-HDL cholesterol rather than the HDL rise1112.
- Food lowers LDL by measurable but modest amounts: about 5 mg/dL for a daily serving of nuts and about 26 mg/dL for the combined “Portfolio” diet, the guideline reports2.
On September 4, 2026, the drug company Novartis announced results from a trial of 8,323 people who had heart disease and high levels of lipoprotein(a), an inherited particle that carries cholesterol9. Its drug, pelacarsen, did its job: it lowered Lp(a)9.
It did not lower the combined rate of heart deaths, heart attacks, strokes and urgent artery-opening procedures9. “These are not the results we hoped for,” said Shreeram Aradhye, Novartis’s chief medical officer10.
The pattern is familiar. In 2017, a drug that more than doubled HDL, the so-called good cholesterol, cut heart events only about as much as its small effect on LDL and similar particles predicted1112.
Both stories carry one lesson. The number on your lab sheet isn’t always the thing doing the damage. Plaque is built by particles that carry cholesterol into artery walls, and those particles can be counted.
Cholesterol rides in particles, and the particles do the damage
Fats don’t dissolve in blood, so the body packs cholesterol and triglycerides into protein-wrapped particles called lipoproteins. Kenneth Feingold of the University of California, San Francisco, explains them in Endotext, a medical textbook hosted by the National Institutes of Health5.
The main types, largest to smallest:
- Chylomicrons carry fat from a meal out of the gut5.
- VLDL (very-low-density lipoprotein) carries fat made in the liver5.
- Remnants are what’s left after the body pulls triglycerides out of those two. They’re richer in cholesterol, and some VLDL remnants become LDL5.
- LDL (low-density lipoprotein) is the main source of cholesterol buildup in arteries, according to MedlinePlus, the National Library of Medicine’s consumer site3.
- Lp(a) is an LDL-like particle with an extra protein, apolipoprotein(a), attached56.
- HDL (high-density lipoprotein) is the smallest. It helps carry cholesterol out of tissues and back to the liver35.
Each chylomicron, VLDL, LDL and Lp(a) particle carries a single molecule of a protein called apolipoprotein B, or apoB; chylomicrons use a shorter form5. Feingold’s chapter lists VLDL, remnants, LDL and Lp(a) as promoting artery disease, and HDL as protecting against it5.
That makes apoB a head count: one apoB, one potentially harmful particle. A cholesterol number tells you how much cholesterol those particles carry, which is related to how many there are but isn’t the same.
Your panel measures three things and calculates the rest
A lipid panel reports total cholesterol, LDL, HDL and triglycerides, and it may add two calculated lines, says the National Heart, Lung, and Blood Institute (NHLBI). Non-HDL cholesterol is total minus HDL. A “remnant” figure is total minus LDL and HDL4.
LDL itself is often calculated. The 2026 guideline prefers newer formulas, called Martin/Hopkins and Sampson/NIH, over the older Friedewald equation, which can’t be used once triglycerides reach 400 mg/dL2.
You may not need to skip breakfast. The guideline says a nonfasting panel works for a first check, for estimating risk and for deciding on treatment, with exceptions covered in our companion triglyceride article2. Some offices still ask for 8 to 12 hours without food, NHLBI notes, so follow your own instructions4.
Non-HDL deserves more attention than it gets. It bundles the cholesterol in LDL and VLDL, the particles MedlinePlus groups as the “bad” kinds, and the guideline recommends that labs report it23. MedlinePlus suggests most adults test every five years, and more often from middle age3.
Your lipid panel, line by line
We built this decoder from the sources cited. It covers adults, since children have different ranges3. Your clinician may set lower goals based on your risk; our article on primary and secondary prevention explains how.
| Line on the report | What it measures | Healthy level for most adults | What moves it |
|---|---|---|---|
| Total cholesterol | Cholesterol in all particles | Under 200 mg/dL3 | Diet, weight, genes, age3 |
| LDL | Cholesterol in LDL, usually calculated2 | Under 1003 | Saturated fat swaps, plant foods, weight, medicines23 |
| HDL | Cholesterol in HDL | 60 or more is best; under 40 for men or 50 for women is low3 | Smoking lowers it3 |
| Triglycerides | Fat carried mostly in VLDL and chylomicrons5 | Under 1503 | Added sugar, refined starch, alcohol, weight2 |
| Non-HDL | Total minus HDL4 | Under 13034 | The same things that move LDL and triglycerides |
| Remnant, if shown | Total minus LDL and HDL4 | No standard target in the sources we used | Mostly triglyceride habits; see our companion article |
| ApoB (separate test) | Number of apoB particles5 | No single cutoff in the AHA summary1 | The same things that move LDL |
| Lp(a) (separate test, usually once) | An inherited particle6 | Higher risk from 125 nmol/L, or 50 mg/dL12 | Mostly genes; little else2 |
About 25% of U.S. adults have an LDL of 130 or higher, the guideline says2. If yours is 160 or higher as a young adult, or early heart disease runs in your family, the guideline says drug treatment should be considered earlier in life1. Familial hypercholesterolemia, an inherited form of very high LDL, is covered in our family history article.
ApoB counts the particles for people whose numbers already look good
The guideline’s apoB advice is cautious. For adults on treatment who have reached their LDL and non-HDL goals, it gives apoB testing a moderate-strength recommendation to decide whether to step up treatment, especially with heart disease, type 2 diabetes or high triglycerides2. For adults not on treatment, its endorsement is weaker: apoB may help sharpen a risk estimate2.
The Heart Association’s summary says apoB is most useful when triglycerides are above 200, in diabetes, or when LDL is already below 701. Our reading: in those groups, a good-looking cholesterol number may not settle the question, and a direct particle count can break the tie.
MedlinePlus lists apoB as a test for people with certain conditions, not everyone3. The AHA summary gives no apoB cutoff, so ask what number your clinician is aiming for1.
Lp(a) is inherited, common and stubborn
About 20% of people worldwide have elevated Lp(a), according to an American Heart Association teaching guide8. Genes explain roughly 70% to 90% or more of the differences in Lp(a) between people, a 2022 AHA scientific statement found6. Levels are highest in people of African ancestry2.
Your level stays about the same from childhood to old age, NHLBI says4. The guideline adds that lifestyle barely changes it, which is why a single test is usually enough2. Statins don’t lower it and may raise it slightly8. The heart association’s guide says high Lp(a) does the most harm to the heart’s arteries and to the aortic valve, where it’s tied to calcium buildup8.
The guideline ties risk to levels measured against a typical value of about 20 nmol/L2:
- 125 nmol/L (50 mg/dL) or more: about 1.4 times the risk; roughly the top 20% of people2.
- 250 nmol/L (100 mg/dL) or more: about 2 times the risk; roughly the top 5%2.
- 430 nmol/L (180 mg/dL) or more: about 4 times the risk; roughly the top 1%2.
What do those multiples mean in people? This is a simplified illustration, by our arithmetic. If 50 of every 1,000 people with your other risk factors would have a heart attack or stroke within 10 years, 1.4 times that is about 70, and twice that is about 100. Real risk tools adjust differently, so ask your clinician for your own estimate.
Labs report Lp(a) in nmol/L or mg/dL, and the guideline prefers nmol/L tests, so check which unit your report uses2. You don’t need to fast2. If your level is high, the guideline recommends testing your parents, siblings and children too2.
A high result changes your plan. The guideline treats it as a reason for more intensive LDL lowering and early, careful control of blood pressure, blood sugar, smoking and other changeable risks12. Apheresis, which filters Lp(a) from the blood, is approved only for a narrow group, and levels rebound within about two weeks8.
A major Lp(a) drug trial came up empty, at least so far
Pelacarsen’s trial enrolled people with heart disease and an Lp(a) of at least 70 mg/dL9. Novartis gave no results yet for a planned group with levels of 90 or more, and full data are due at a medical meeting9. As of mid-October 2026, the company still hadn’t reported how large any effect on heart events was, Pharmacy Times noted13.
Patients were already well treated, with an average LDL around 66 mg/dL, a cardiologist noted in coverage by the news site TCTMD10. Many cardiologists are now questioning Lp(a)’s role in prevention, TCTMD reported10. The National Lipid Association, one of the guideline’s co-sponsors, still told clinicians to keep measuring it14.
Other Lp(a)-lowering drugs, from Amgen, Eli Lilly and Silence Therapeutics, are still being tested, and none has FDA approval10. The 2022 AHA statement had noted that outcome trials would be needed for final proof7.
Where does that leave you? A high Lp(a) still marks higher risk, which is why the guideline says to measure it once12. What’s missing is a medicine proven to cut that risk by targeting Lp(a) itself, so the advice centers on LDL and the other risks you can change17.
Drugs that raised “good” cholesterol didn’t protect hearts
HDL looked like the obvious target: if it hauls cholesterol away, more should be better. The trials said otherwise.
Pfizer halted torcetrapib, a drug that raised HDL, in 2006 after a large study found more deaths and heart events among people taking it12. Two later drugs in the same class, dalcetrapib and evacetrapib, showed no benefit12.
The fourth, anacetrapib, was tested in the REVEAL trial, published in the New England Journal of Medicine in 2017. It enrolled 30,449 adults with artery disease who were already on a strong statin11. By the trial’s midpoint, HDL was 104% higher on anacetrapib, while non-HDL cholesterol was only 18% lower11.
After a median of 4.1 years, major coronary events struck 10.8% of the anacetrapib group and 11.8% of the placebo group11. That’s about 11 fewer people per 1,000, by our math. A 17 mg/dL drop in non-HDL cholesterol would be expected to cut heart attacks and coronary deaths by about 10%, roughly what the trial saw, and Martin Landray of the University of Oxford, a co-principal investigator, called it unlikely that the HDL rise played a major role12. Merck sponsored the trial1112.
High-dose niacin raises HDL too, and our guide to the 13 vitamins explains why its heart trials also disappointed. One reason may be that the HDL number doesn’t show how well HDL does its cleanup job, Feingold cautions5.
So treat HDL as a clue, not a target. The 2026 guideline sets goals for LDL and non-HDL, not HDL12.
What food can change, in milligrams per deciliter
The 2026 guideline puts numbers on what diet does to LDL. These are average drops from studies it reviewed2:
- The Portfolio diet, built around nuts, soy foods, extra fiber and a margarine with added plant sterols: about 26 mg/dL2.
- A vegan diet, in a crossover trial of 62 adults with overweight: about 15 mg/dL, while the Mediterranean phase showed no significant drop2.
- Vegetarian or vegan versus omnivorous eating: about 12 mg/dL2.
- Plant protein in place of meat: about 8 mg/dL2.
- One serving of nuts a day: about 5 mg/dL2.
- Three daily servings of oatmeal: under 5 mg/dL2.
The guideline calls the link between swapping saturated fat for unsaturated fat and lower LDL strong, and our article on healthy fats covers that swap2. It adds that the swap lowers LDL more consistently than cutting dietary cholesterol2. MedlinePlus agrees that saturated fat is the main dietary problem, though dietary cholesterol also matters3.
What about pills? The guideline recommends against dietary supplements for lowering LDL or triglycerides, citing limited and inconsistent data2. Red yeast rice, often sold as a natural statin alternative, has its own quality problems, described in our article on supplement contamination.
How the main claims stack up:
| Claim | Evidence rating | What it rests on |
|---|---|---|
| Lowering LDL and other apoB particles cuts heart events | The basis of the 2026 guideline’s goals; REVEAL’s benefit tracked non-HDL lowering11112 | |
| Whole-food diet changes lower LDL | for modest drops | Studies reviewed in the guideline, about 5 to 26 mg/dL2 |
| High Lp(a) raises heart risk | AHA statement calls it causal; guideline risk tiers26 | |
| A drug that lowers Lp(a) prevents heart events | Pelacarsen’s outcome trial missed its goal; other drugs are still in trials910 | |
| Diet or exercise can meaningfully lower a high Lp(a) | Levels are mostly genetic and minimally changed by lifestyle268 | |
| Raising HDL with drugs, by itself, prevents heart attacks | CETP inhibitor trials, including one halted for deaths1112 | |
| Supplements lower LDL or triglycerides enough to recommend | Guideline advises against them2 |
Natural-health instincts hold up well here. The best-tested diet steps are whole foods: nuts, beans, oats and plant proteins. They work best stacked together, as the Portfolio diet shows2.
But respect for the body also means respecting what it inherits. Food can’t rewrite your Lp(a) or fully tame an inherited LDL problem. Knowing those numbers is itself a form of prevention, because it tells you where your effort will count most.
When to see a clinician. Make an appointment if your LDL is 160 or higher or close relatives had early heart disease; the guideline treats both as reasons to consider medicine earlier in life1. NHLBI suggests asking about Lp(a) testing if your family history is unknown4. If you take a cholesterol medicine, don’t stop or change it on your own; bring your questions to your prescriber. Our article on 911 symptoms lists the heart attack and stroke signs that can’t wait.
Questions readers ask
Do I have to fast before a cholesterol test?
Usually not. The 2026 guideline says a nonfasting panel works for most purposes2. People with very high triglycerides or certain family histories may need a fasting test; our triglyceride article explains when.
My HDL is high. Does that protect me?
Not reliably. Drugs that raised HDL didn’t prevent heart attacks, and one was halted after more deaths1112. Your LDL, non-HDL and Lp(a) tell your clinician more about risk1.
Can I lower Lp(a) naturally?
Not by much. Genes explain most of it, and the guideline says lifestyle barely changes it26. What you can do is lower your other risks, which is exactly what the guideline advises for people with high Lp(a)12.
The bottom line
The old “good” and “bad” labels miss the main point. Artery plaque is driven by apoB particles: LDL, remnants and Lp(a). Non-HDL cholesterol and, for some people, an apoB test track them better than HDL ever did.
Ask for an Lp(a) test once, because most of it is inherited and the result can change how hard you and your clinician push on everything else. Then let food do what it does well: a stack of whole-food changes can lower LDL meaningfully, alongside any treatment you need.
Related reading
About this article
This article was researched using the 2026 cholesterol guideline and its official summary, American Heart Association statements, NIH and National Library of Medicine resources, published trial reports, a company announcement and medical news coverage, all listed below. It was reviewed in October 2026. It is general education, not medical, legal or financial advice.
Sources
- American Heart Association. “Top Things to Know: 2026 Guideline on the Management of Dyslipidemia.” AHA Professional Heart Daily; guideline by Blumenthal RS, et al., published online March 13, 2026, in Circulation. professional.heart.org
- Blumenthal RS, Morris PB, et al. “2026 ACC/AHA/AACVPR/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA Guideline on the Management of Dyslipidemia.” Circulation, 2026; doi:10.1161/CIR.0000000000001423 (copy posted on a Universidad Nacional de Córdoba course site). fcm.aulavirtual.unc.edu.ar
- MedlinePlus. “Cholesterol Levels: What You Need to Know.” National Library of Medicine, updated October 7, 2026. medlineplus.gov
- National Heart, Lung, and Blood Institute. “Blood Cholesterol: Diagnosis.” NHLBI, updated April 18, 2024. nhlbi.nih.gov
- Feingold KR. “Introduction to Lipids and Lipoproteins.” Endotext, NCBI Bookshelf, updated January 14, 2024. ncbi.nlm.nih.gov
- Reyes-Soffer G, Ginsberg HN, Berglund L, et al. “Lipoprotein(a): A Genetically Determined, Causal, and Prevalent Risk Factor for Atherosclerotic Cardiovascular Disease: A Scientific Statement From the American Heart Association.” Arteriosclerosis, Thrombosis, and Vascular Biology, 2022;42(1):e48–e60 (Capital Region of Denmark research portal record). research.regionh.dk
- American Heart Association. “Top Things to Know: Lipoprotein(a): A Genetically Determined, Causal, and Prevalent Risk Factor for Atherosclerotic Cardiovascular Disease.” AHA Professional Heart Daily. u1-cd-professional.sc.heart.org
- American Heart Association. “The ABCs of Lp(a).” AHA quality improvement teaching slides, citing the 2022 AHA scientific statement. heart.org
- Novartis. “Novartis announces Lp(a)HORIZON Phase III topline results for pelacarsen in patients with elevated Lp(a) and established cardiovascular disease (CVD).” Company media release, September 4, 2026. novartis.com
- TCTMD. “Lp(a)HORIZON: Top-line Results Dash Hopes for Pelacarsen.” September 8, 2026. tctmd.com
- HPS3/TIMI55–REVEAL Collaborative Group (Bowman L, et al.). “Effects of Anacetrapib in Patients with Atherosclerotic Vascular Disease.” New England Journal of Medicine, 2017;377(13):1217–1227 (Duke University Scholars record). scholars.duke.edu
- TCTMD. “Anacetrapib Finds (Modest) Success Where All Other CETP Inhibitors Failed.” August 29, 2017. tctmd.com
- Pharmacy Times. News report on the pelacarsen (Lp(a)HORIZON) and ziltivekimab (ZEUS) trial results. October 16, 2026. pharmacytimes.com
- National Lipid Association. “Top-Line Results From the Phase 3 Lp(a)HORIZON Trial.” NLA, September 14, 2026. lipid.org