The standard lipid panel is not a wrong test. It is an incomplete one. It measures the amount of cholesterol being carried inside particles, not the number of particles doing the carrying — and it does not include, at all, an inherited particle called Lp(a) that roughly one in four or five people worldwide carries at a high level. In March 2026 a new American guideline on the management of dyslipidemia recommended, for the first time, that Lp(a) be measured at least once in every adult's lifetime, and that apoB be measured in specific groups. This article explains what the two numbers mean, why they are worth asking about, and why none of this makes your old lipid panel worthless.
This article is educational and is not a substitute for your doctor.
What your lipid panel actually measures — and what it does not
A lipid profile usually gives you four numbers: total cholesterol, "bad" LDL cholesterol, "good" HDL cholesterol, and triglycerides. These are useful numbers, and decades of prevention research rest on them. But understanding their limits starts with one question: what exactly is being measured?
LDL-C is a number about the cargo, not about the number of vehicles
Cholesterol does not float freely in blood. It travels inside carrier particles called lipoproteins. The "LDL-C" figure on your report measures the mass of cholesterol inside those particles, not how many of them there are.
Picture a street carrying ten tons of cement. It could be in ten large trucks, or forty small ones. Same weight, entirely different traffic. And in an artery, what lodges in the wall and starts a plaque is the particle, not the cargo inside it. So two people with identical LDL-C readings can be carrying very different particle counts — because one person's particles are smaller and poorer in cholesterol.
ApoB: the number that turns a weight into a count
This is where apolipoprotein B comes in. Its usefulness rests on something simple and elegant: every atherogenic particle carries exactly one apoB molecule. Cleveland Clinic puts it plainly — "Each Apo B molecule carries one of the bad lipoproteins" — which is why the test can give "an accurate count because of the one-to-one ratio."
In the words of the National Lipid Association's expert clinical consensus: "ApoB represents the total concentration of atherogenic lipoprotein particles in the circulation and more accurately reflects the atherogenic burden of lipoproteins when compared to low-density lipoprotein cholesterol (LDL-C)." ApoB, in other words, is not another kind of cholesterol. It is a counter.
What happens when you put both numbers into one equation?
The strongest evidence that the count is the causal unit comes from genetics rather than from comparing two patients. In a multivariable Mendelian randomisation analysis published in PLOS Medicine in 2020, apoB, LDL cholesterol and triglycerides were put into a single model to see which remained causally associated with coronary heart disease:
- ApoB: odds ratio 1.92 (95% CI 1.31–2.81), P < 0.001.
- LDL cholesterol: 0.85 (0.57–1.27), P = 0.44 — not statistically significant, and the direction of the estimate actually reversed.
- Triglycerides: 1.12 (1.02–1.23), P = 0.01 — weak.
The authors' conclusion, verbatim: "These findings suggest that apolipoprotein B is the predominant trait that accounts for the aetiological relationship of lipoprotein lipids with risk of CHD."
Honesty requires reading the limitation the same authors wrote in the next breath: "A limitation is that, owing to the nature of lipoprotein metabolism, measures related to the composition of lipoprotein particles are highly correlated, creating a challenge in making exclusive interpretations on causation of individual components." In plain terms: in most people these numbers move together, so they are hard to separate cleanly. The correct message is not "LDL-C is meaningless." It is: when the two numbers disagree, the count is the one carrying the information.
Before you order a new test: there is a third number already on your report, for free
This is the part usually left out of enthusiastic talk about apoB. That same consensus says: "ApoB and non-HDL-C stratify ASCVD risk more accurately than LDL-C," and also: "When there is discordance between LDL-C and apoB, or LDL-C and non-HDL-C, atherosclerotic cardiovascular disease risk generally aligns better with apoB or non-HDL-C."
Non-HDL cholesterol is a subtraction you can do right now: total cholesterol minus HDL. No extra order, no extra fee. So the nearest step to reading your panel better is not a new test — it is a number that has been sitting there unexamined. The 2026 guideline gives it targets alongside LDL-C: under 130 mg/dL at borderline and intermediate risk, and under 100 mg/dL at high risk.
So who actually needs apoB?
Here the new guideline was specific rather than sweeping. According to the American Heart Association's summary, "measuring apoB may be used to assess any residual ASCVD risk and guide treatment among people with cardiovascular-kidney-metabolic syndrome, Type 2 diabetes, high triglycerides or known cardiovascular disease who have reached their LDL-C and non-HDL-C goals," and apoB "may be a more accurate risk marker than LDL-C" in those groups.
Note that final condition. ApoB here is not a replacement for the ordinary panel; it is a question asked after the ordinary panel looks reassuring. Cleveland Clinic's published reference ranges are 66–133 mg/dL for males and 60–117 mg/dL for females, with results "higher than 130 mg/dL" indicating elevated risk. One small practical advantage: "You don't have to fast before Apo B testing" — because it is directly measured rather than calculated — though a 12-hour fast may still be required if a full lipid panel is drawn at the same time.
Lp(a): the number missing from your report because nobody ordered it
The second problem is of an entirely different kind. ApoB is a question of units. Lp(a) is a question of absence. Cleveland Clinic's sentence needs no interpretation: "It's not part of a standard lipid panel that checks the kinds of cholesterol everyone talks about." It requires a specific request from a clinician. And what is not ordered is not measured; what is not measured does not enter your risk estimate.
What is Lp(a)? A ball with a tail
A 2025 review in the Cleveland Clinic Journal of Medicine describes Lp(a) as "a lipoprotein particle comprising the LDL-like component apolipoprotein B covalently bound to a protein called apolipoprotein(a)," and likens it to "a ball (the apolipoprotein B) with a tail (the apolipoprotein[a])." The ball is what your panel knows about. The tail is what it does not — and that tail is what makes the particle stickier to the arterial wall and more closely tied to inflammation and to calcification of the aortic valve.
More importantly, your level is not a product of your habits. The European Atherosclerosis Society consensus statement says "Lp(a) concentration is predominantly (>90%) determined by genetic variability at the LPA locus." In Cleveland Clinic's plainer language for patients: "The genes you get from your biological parents decide your lipoprotein (a) level," and your childhood level "will most likely be the level you have for years to come."
The numbers: what do 125 and 250 nmol/L mean?
Societies converge on one practical threshold: 50 mg/dL, or 125 nmol/L. The 2025 review sets out three zones: below 30 mg/dL (75 nmol/L) is normal; 30–50 mg/dL (75–125 nmol/L) is a "gray zone"; and "≥50 mg/dL or 125 nmol/L" is elevated. Between 20% and 25% of the global population has elevated levels.
As for how much risk, the 2026 guideline summary puts it in a form you can carry in your head: "Values of Lp(a) >125 nmol/L and >250 nmol/L are associated with 1.4-fold and ≥2-fold increased ASCVD risk" over the long term. The European consensus adds a sentence worth pausing on: a level of 100 mg/dL (about 250 nmol/L) "approximately doubles the risk of ASCVD irrespective of baseline absolute risk." It does not cancel your other risk factors — it multiplies whatever you already have.
Why "once in a lifetime"? And what changed in 2026?
Because the number is genetically fixed, one good measurement is enough. That is what made the 2026 recommendation a departure in American practice: "Lp(a) should be measured at least once in adulthood," and "lifestyle changes minimally affect Lp(a) levels, so repeat testing is generally not needed." The American College of Cardiology's own summary calls this "the first time the ACC/AHA guideline has made universal screening and genetic cascade screening recommendations regarding Lp(a)."
Cascade screening is what makes a single test bigger than itself. The National Lipid Association notes that "cascade screening of first-degree relatives of patients with elevated Lp(a) can identify additional individuals at risk who require intervention." One test for you can be information for a whole family.
And if it is high and there is no drug for it? The lesson of 2026
This is the honest question, and in 2026 the answer became clearer — and less optimistic.
The trial everyone was waiting for, and it did not work
When that Cleveland Clinic Journal of Medicine review appeared in November 2025, it said phase 3 results were "expected in 2026." They arrived. On 4 September 2026 Novartis announced the topline results of Lp(a)HORIZON (NCT04023552): a phase III, randomized, double-blind, placebo-controlled outcomes trial in 8,323 patients with elevated Lp(a) (entry ≥70 mg/dL overall, with a ≥90 mg/dL subpopulation) and established cardiovascular disease. The drug, pelacarsen, did lower Lp(a). The primary endpoint — a composite of cardiovascular death, non-fatal myocardial infarction, non-fatal stroke and urgent coronary revascularization requiring hospitalization — was not met.
In the words of the company's chief medical officer: "Although lower Lp(a) levels were observed with pelacarsen, the findings did not demonstrate that this translated into reduced cardiovascular risk in the overall study population." He added: "These are not the results we hoped for, but they provide important evidence that advances scientific understanding of the relationship between Lp(a) lowering and cardiovascular outcomes."
It did not pass quietly. In TCTMD's coverage, cardiologists were reported to be "questioning Lp(a)'s role in prevention and management," with some arguing the result shows that "treating a non-causal marker of disease doesn't treat the disease."
The distinction that explains everything: a risk marker is not necessarily a drug target
This is the most important idea in the article, and it is a general one that will serve you when reading any medical news: a number being useful for prediction is one thing; lowering that number being useful is another. Two different jobs, requiring two different kinds of evidence.
The genetic and epidemiological evidence that high Lp(a) predicts risk is strong; the European consensus states that "epidemiologic and genetic studies involving hundreds of thousands of individuals strongly support a causal and continuous association between Lp(a) concentration and cardiovascular outcomes." But one large trial showed that lowering it, with this drug, by this much, in this population, did not reduce events. Researchers have several candidate explanations — the absolute magnitude of lowering, baseline levels, follow-up duration, and the fact that participants were already on intensive lipid and blood-pressure therapy. Full results have not yet been published, and other outcome trials are ongoing.
Practically: the value of knowing your Lp(a) today does not depend on a drug existing for it. Its value is that it rearranges what you do about everything else. The societies' advice in the absence of a specific therapy is explicit: "Patients with elevated Lp(a) should receive early, more-intensive risk factor management, including lifestyle modification and lipid-lowering drug therapy in high-risk individuals, primarily to reduce low-density lipoprotein cholesterol (LDL-C) levels." The number you cannot change makes the numbers you can change more important, not less.
And the news about a single injection?
On 28 August 2026, at the European Society of Cardiology congress in Munich, the first-in-human results of a long-acting gene-silencing therapy (Kylo-11) were presented and simultaneously published in The Lancet: 71 participants enrolled, 70 given a single injection or placebo, followed about 11 months, with Lp(a) reductions of roughly 95–97% at the highest doses sustained for close to a year. Put that in context: a phase 1 study in 70 people, whose result is a biomarker rather than a clinical outcome — and pelacarsen is precisely why that caution is warranted. Still, the lead author, who also co-wrote the 2025 review, said something that describes the reader's situation better than it describes the drug: "Most people with high Lp(a) don't know it."
The Saudi picture: a common number, and a test that is not ordered
Here the story becomes local — and the gap is not where we usually expect it.
On the prevalence side: a 2024 study in Diabetes, Metabolic Syndrome and Obesity measured Lp(a) in 361 people in Saudi Arabia (162 men, 199 women, aged 18–93) through private laboratories between 2015 and 2020. The 10th, 25th, 50th, 75th and 90th percentiles were 9, 20, 47, 93 and 153 nmol/L. Levels were elevated (≥75 nmol/L) in 28.4% of men and 30.6% of women, significantly higher in women (P < 0.01), and the authors concluded that levels "were comparable to those observed in other ethnic groups." This must be read at its true size: the authors themselves cited "the small sample size," a retrospective design, the "absence of clinical data," and "selection bias." A strong indicator, not a national survey.
On the ordering side: a 2025 study in the Journal of Clinical Lipidology surveyed 1,069 physicians across six Gulf countries (Saudi Arabia, Oman, the UAE, Qatar, Kuwait and Bahrain). The finding: "Only 252 (23.6%) participants reported having requested Lp(a) measurements." Mean knowledge score was 9.0 ± 3.5 out of 17, and "75% self-reported as being unaware of Lp(a) or 'neutral'." The leading reasons given for not testing were "lack of information (31.2%) and unavailability in hospitals (23.9%)." The authors' conclusion, verbatim: "Physicians in the Arabian Gulf region report limited basic and clinical knowledge of Lp(a), which could result in underestimation of cardiovascular risk. These findings, in the largest such study to date, are a call to action to increase awareness about Lp(a) and accessibility to testing."
Put the two halves together: a number elevated in roughly three of every ten people in a Saudi sample, and a test that only one in four physicians has ever ordered. The gap here is not patient awareness, not patient refusal, and not the cost of a drug — it is the order form itself, and whether the laboratory offers the test. Which is why the single practical instruction in this article is a question, not a decision: "Has my Lp(a) ever been measured, even once?" — addressed to your doctor, not to yourself.
When to see a doctor, and who should ask first
These are silent risk numbers, not symptoms. But some signs call for immediate medical care rather than an article:
- Chest pain, pressure or heaviness, especially on exertion or spreading to the arm, jaw or back.
- New or worsening shortness of breath, or breathlessness when lying flat.
- Sudden numbness or weakness on one side of the body, slurred speech, or sudden loss of vision — stroke signs requiring emergency care immediately.
- Fainting or severe dizziness, or palpitations with chest pain.
And the people with most reason to ask their doctor about these two numbers specifically:
- Anyone with a family history of premature heart disease, or a relative affected at a young age.
- Anyone with a first-degree relative known to have elevated Lp(a) — cascade screening.
- Anyone who has had a cardiovascular event, or recurrent events despite other risk factors being controlled.
- Anyone with familial hypercholesterolemia or very high LDL cholesterol.
- Anyone with type 2 diabetes, high triglycerides, or cardiovascular-kidney-metabolic syndrome — the groups the 2026 guideline names for apoB.
Remember that high Lp(a) can raise risk "even if your other cholesterol numbers are OK." That is not a reason to worry; it is a reason for one specific question at your next appointment. And any decision about medication — starting, stopping or changing a dose — belongs to your doctor, not to something you read.
Where does Bakery 8 stand in all this?
We will be candid, in a way that does not serve us.
First: no bread — not ours, not anyone's — lowers Lp(a). And the European consensus statement contains a sentence that would have made the best advertisement in this bakery's history: "Lifestyle interventions have minimal impact, but a low carbohydrate/high fat diet may decrease levels by 10–15%." We are declining to use it, and here is why. It sits inside a passage that opens by saying lifestyle has minimal impact. The 2025 review states that Lp(a) levels are "largely unaffected by lifestyle changes." A 10–15% shift in a number that is more than 90% genetically determined moves nobody from 250 nmol/L into safety. And most importantly, no outcomes trial has ever shown that lowering Lp(a) through diet reduces cardiovascular events — indeed the one large trial that lowered it with a potent drug did not reduce events. Using that sentence to sell bread would be exploiting a figure, not sharing health information.
Second, and harder: the dietary pattern our products serve may raise apoB in some people — the very number this article has argued is the more accurate one. In a meta-analysis published in the American Journal of Clinical Nutrition in 2024, covering 41 trials and 1,379 participants, LDL cholesterol on a low-carbohydrate diet rose by 41 mg/dL (95% CI 19.6–63.3) in people with a body mass index below 25; it "did not change" in those between 25 and under 35; and it fell by 7 mg/dL (95% CI −12.1 to −1.3) in those at 35 or above. The effect depends on who you are — lean or not.
And we will pass on the complexity as it is, not as we would like it. A review published in the European Journal of Preventive Cardiology in January 2026 says the question of whether such elevations in "lean-mass hyper-responders" carry the same risk "remains unresolved," and that "long-term outcome data for LMHRs are currently lacking." But it adds the sentence that must not be dropped: "uncertainty regarding the LMHR phenotype should not be interpreted as uncertainty regarding LDL causality, which remains a foundational principle." Its recommendation is neither dismissal nor alarm, but "shared decision-making between clinicians and patients," with apoB measurement and imaging where appropriate.
The conclusion that concerns us: if you follow a low-carbohydrate pattern — the pattern we sell to — you are not exempt from having your lipids measured. You may be among the people with most reason to measure them and show them to a doctor. "Healthy and delicious" describes what is in the loaf; it does not return a test result on your behalf. This is not the first time we have said our pattern has a nutritional weak spot — we said the same about magnesium, and about potassium and salt substitutes.
So what can we honestly still say? Something small and bounded. If bread or something sweet is going to be part of your day anyway, choosing a version with no added sugar and fewer carbohydrates reduces one metabolic load — triglycerides and post-meal blood sugar — and does nothing for Lp(a), and guarantees nothing for apoB. On that basis only, you are welcome to browse our low-carb, sugar-free breads or our desserts with no added sugar. That is all we claim, and nothing beyond it. For the wider picture, we have also written about the logic of "one reading is not a verdict" in home blood pressure monitoring, and about choosing dietary fats in an earlier article on healthy fats and cholesterol.
Frequently asked questions
Does this mean the lipid panel I already had was worthless?
No. The lipid panel remains the foundation of risk assessment, and the 2026 guideline's targets are written in LDL-C and non-HDL-C to begin with. The point is that it answers a narrower question than people assume: it measures cholesterol mass rather than particle number, and it does not include Lp(a) at all. Not wrong — incomplete in certain contexts.
Should I order apoB and Lp(a) for myself from a private lab?
It is better for both the order and the interpretation to go through your doctor, because the number only means something inside your full risk picture: age, family history, blood pressure, diabetes, smoking. That said, the 2026 recommendation on Lp(a) is broad enough — at least once for every adult — to make raising it at your next appointment entirely reasonable.
If Lp(a) is high and there is no drug for it, what is the point of knowing?
The point is that it changes how intensively you treat everything else: firmer LDL-C lowering, starting earlier, tighter control of blood pressure, glucose and smoking, and screening your first-degree relatives. The societies' advice without a specific therapy is "early, more-intensive risk factor management" — and those are measures available today, not someday.
Do I need to repeat the Lp(a) test every year?
Generally no. Because the level is more than 90% genetically determined and relatively stable, the 2026 guideline summary states that "lifestyle changes minimally affect Lp(a) levels, so repeat testing is generally not needed." One good measurement in a lifetime is the rule, with exceptions your doctor may judge.
Does rising cholesterol on a keto diet mean I should stop?
That is an individual medical decision, not a general rule. What the available evidence says: the rise is real and happens particularly in lean people (41 mg/dL on average in the meta-analysis); its long-term significance "remains unresolved"; and doubt about that phenotype "should not be interpreted as uncertainty regarding LDL causality." The recommended path is shared decision-making with your doctor, built on measurement — apoB included — rather than assumption.
References
- Joint guideline writing committee (ACC/AHA and multiple societies; committee chair Roger S. Blumenthal). "2026 Guideline on the Management of Dyslipidemia." Journal of the American College of Cardiology, published online 13 March 2026. doi:10.1016/j.jacc.2025.11.016 — accessed via American College of Cardiology and American Heart Association summaries.
- Kronenberg F, Mora S, Stroes ESG, et al. "Lipoprotein(a) in atherosclerotic cardiovascular disease and aortic stenosis: a European Atherosclerosis Society consensus statement." European Heart Journal 2022;43(39):3925–3946. doi:10.1093/eurheartj/ehac361.
- Koschinsky ML, Bajaj A, Boffa MB, et al. "A focused update to the 2019 NLA scientific statement on use of lipoprotein(a) in clinical practice." Journal of Clinical Lipidology 2024. doi:10.1016/j.jacl.2024.03.001.
- Soffer DE, Marston NA, Maki KC, et al. "Role of apolipoprotein B in the clinical management of cardiovascular risk in adults: An Expert Clinical Consensus from the National Lipid Association." Journal of Clinical Lipidology 2024;18(5):e647–e663. doi:10.1016/j.jacl.2024.08.013.
- Richardson TG, Sanderson E, Palmer TM, Ala-Korpela M, Ference BA, Davey Smith G, Holmes MV. "Evaluating the relationship between circulating lipoprotein lipids and apolipoproteins with risk of coronary heart disease: A multivariable Mendelian randomisation analysis." PLOS Medicine 2020;17(3):e1003062. doi:10.1371/journal.pmed.1003062.
- Al-Dalakta A, Cho LS, Sarraju A. "Lipoprotein(a) in clinical practice: What clinicians need to know." Cleveland Clinic Journal of Medicine 2025;92(11):679–685. doi:10.3949/ccjm.92a.25020.
- Novartis. Media release: Lp(a)HORIZON (NCT04023552) Phase III topline results for pelacarsen, 4 September 2026 — with the National Lipid Association's resource page on the topline results.
- TCTMD. Coverage of the Lp(a)HORIZON topline results, September 2026.
- Jawi M, Al Mahdi HB, Daghistani H, et al. "Lipoprotein(a): Levels and Reference Intervals Among People in Saudi Arabia." Diabetes, Metabolic Syndrome and Obesity 2024;17:2935–2942. doi:10.2147/DMSO.S465885.
- Alhabib RK, Batais MA, AbuDujain NM, et al. "Lipoprotein(a) knowledge, awareness, and clinical practice among physicians in the Arabian Gulf region." Journal of Clinical Lipidology 2025;19(6):1619–1627. doi:10.1016/j.jacl.2025.09.034.
- Soto-Mota A, Flores-Jurado Y, Norwitz NG, Feldman D, Pereira MA, Danaei G, Ludwig DS. "Increased low-density lipoprotein cholesterol on a low-carbohydrate diet in adults with normal but not high body weight: A meta-analysis." American Journal of Clinical Nutrition 2024;119(3):740–747. doi:10.1016/j.ajcnut.2024.01.009.
- Özlek B. "Reassessing LDL-cholesterol elevations in lean-mass hyper-responders: clinical significance and unresolved questions." European Journal of Preventive Cardiology 2026;zwag038. doi:10.1093/eurjpc/zwag038 (14 January 2026).
- Cleveland Clinic. "Apolipoprotein B (APOB) Test."
- Cleveland Clinic. "Who Needs a Test for Lipoprotein (a)?"
- Cleveland Clinic Newsroom. "First in Human Long-Acting Gene Silencing Therapy Significantly Lowers Heart Disease Risk Marker with One Injection," 28 August 2026 — presented at the European Society of Cardiology congress, Munich, and published in The Lancet.
Related keywords: lipid panel, cholesterol test, apoB, apolipoprotein B, Lp(a), lipoprotein(a), non-HDL cholesterol, LDL cholesterol, atherosclerosis, inherited cholesterol risk, cascade screening, 2026 dyslipidemia guideline.