Why heart attacks follow the count
On this page
When the big studies put ApoB, LDL and triglycerides side by side, ApoB is the one that stays tied to heart attacks. People born with genes that lower ApoB by 10 mg/dL had about 23% lower odds of heart disease, and it didn’t matter which fat the genes lowered.
What we’d do
Make ApoB your main number once you have it, and plan around it whenever it disagrees with your LDL. The 2026 US guideline is more careful than we are. It calls the test “reasonable” for people already on treatment and “may be reasonable” for everyone else. We’d ask for it with every cholesterol panel, since it costs little to run and it catches the people whose LDL looks fine while their particle count doesn’t.
If you remember three things
- The genes point at the count. Lowering ApoB through triglycerides or through LDL gave the same benefit per point.
- Head to head, ApoB wins. In 429,959 people, only ApoB stayed tied to heart attacks once the other lipids were in the same model.
- A normal LDL can hide it. Statin users with a high ApoB and a low LDL had about 1.5 times the heart attacks.
Why LDL got the job first
LDL has been the target for decades. It’s on every panel, nearly every cholesterol trial measured it, and statins were judged by how far they pushed it down. The 2026 guideline still calls LDL “the traditional lipid marker”. So when someone says ApoB is the better number, the real question is whether it adds anything, or whether it’s the same information under a new name.
The way to settle it is to find people whose ApoB and LDL point in different directions, and see which number their arteries follow.
The lifelong experiment nature ran
You can’t randomly assign people to have a high or low ApoB for 50 years. But genes do something close to that. Each person gets a random shuffle of their parents’ genes before birth, and some of those genes nudge cholesterol up or down for life.
In 2019 Ference and his colleagues used that shuffle on 654,783 people, including 91,129 with coronary heart disease, more than half of them in UK Biobank. They compared two groups of genes. Variants in LPL lower triglycerides, and variants in LDLR lower LDL. Both lower ApoB, by different routes.
Per 10 mg/dL lower ApoB, the odds of heart disease were 0.771 with the triglyceride genes and 0.773 with the LDL genes. The two routes gave the same protection for the same drop in the count. The authors concluded that the benefit of lowering triglycerides or LDL “may be proportional to the absolute change in ApoB”.
Like dents in a guardrail
Picture the guardrail on a busy highway. It gets dented by the trucks that hit it, and the number of dents follows the number of trucks, not what they were carrying. Each ApoB particle that gets into an artery wall is one more chance for plaque. Lower the count by any route and you lower the chances.
Head to head
Only ApoB stayed
429,959 people in UK Biobank and clinical trials, up to about 11 years of follow-up. Marston and colleagues, JAMA Cardiology, 2022
389,529 people not on cholesterol medicine and 40,430 with heart disease on statins. Looked at one at a time, triglycerides, LDL and ApoB were all tied to heart attacks. Put in the same model, “only apoB” was, in both groups. For each standard step up in ApoB (one standard deviation), the risk of a heart attack rose 27%.
It’s an observational study, so it shows which number tracks risk, not what lowering it does.
On statins, in Copenhagen
13,015 statin users, about 8 years. Johannesen and colleagues, JACC, 2021
People with ApoB above the group’s median but LDL below it had 1.49 times the heart attacks and 1.21 times the deaths of people with both below. A high LDL with a low ApoB wasn’t linked to more heart attacks or deaths.
Everyone was already on a statin, so it tells us most about treated people.
Starting young
2,794 adults aged 18 to 30, followed 25 years. CARDIA, Wilkins and colleagues, JACC, 2016
The higher their ApoB as young adults, the more likely they were to have calcium in their heart arteries in midlife. A high ApoB with a low LDL carried 1.55 times the odds.
Calcium is a sign of plaque, not a heart attack itself.
The authors of the UK Biobank study put it in one line. They found the risk of a heart attack was “best captured by the number of apoB-containing lipoproteins”.
What the 2026 guideline says
The guideline, published in March 2026 by the American College of Cardiology, the American Heart Association and nine other groups, says “ApoB predicts ASCVD risk more accurately than LDL-C in cases of disagreement, or discordance”. ASCVD means heart attacks, strokes and other artery disease. It then gives two recommendations.
- For adults on cholesterol medicine (class 2a, “reasonable”): measuring ApoB once their LDL or non-HDL goal is met, to decide whether to do more, above all with heart disease, diabetes, kidney and metabolic disease or high triglycerides.
- For adults not on medicine (class 2b, “may be reasonable”): measuring ApoB to sharpen their risk, help decide whether to start treatment, and look for inherited cholesterol problems.
It also lists an ApoB of 120 mg/dL or more among the things that raise risk, and 140 or more as severe high cholesterol. What should my ApoB be? puts those lines on one ruler.
The best case against it
ApoB and LDL correlate at 0.96 in UK Biobank. For most people, the skeptic says, a second test tells you what the first one already did. That’s true. The value is all in the minority where the two split, and the guideline says much of that split comes from the old formula labs use to work out LDL. With a newer formula, Martin/Hopkins, only about 2% of people with an LDL under 70 and about 1% under 100 had an ApoB over the goal.
We’d still test it. You can’t know whether you’re in the 2% without it, the people most likely to be there (high triglycerides, diabetes, extra weight, a statin) are the ones at most risk, and the National Lipid Association says the test is cheap to perform. A 2026 computer model from Northwestern University of 250,000 American adults found that treating to an ApoB under 78.7 prevented more heart attacks and strokes than treating to an LDL under 100. It’s a model, and nobody has run that comparison as a trial yet.
8 sources
- Ference BA and others, Association of triglyceride-lowering LPL variants and LDL-C-lowering LDLR variants with risk of coronary heart disease, JAMA, 2019pmc.ncbi.nlm.nih.gov
- Marston NA and others, Association of apolipoprotein B-containing lipoproteins and risk of myocardial infarction in individuals with and without atherosclerosis, JAMA Cardiology, 2022pmc.ncbi.nlm.nih.gov
- Johannesen CDL and others, Apolipoprotein B and non-HDL cholesterol better reflect residual risk than LDL cholesterol in statin-treated patients, JACC, 2021pubmed.ncbi.nlm.nih.gov
- Wilkins JT and others, Discordance between apolipoprotein B and LDL-cholesterol in young adults predicts coronary artery calcification: the CARDIA study, JACC, 2016pmc.ncbi.nlm.nih.gov
- Blumenthal RS, Morris PB, Gaudino M and others, 2026 ACC/AHA guideline on the management of dyslipidemia, JACC and Circulation, 13 March 2026doi.org
- The same guideline, full text, Circulation, 2026api.iscpcardio.org
- Soffer DE, Marston NA, Maki KC and others, 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, 2024pmc.ncbi.nlm.nih.gov
- Northwestern University, Little-used cholesterol test could prevent more heart attacks, strokes, news release, 8 April 2026news.northwestern.edu
