ApoB 164, LDL 112. The Number That Predicted His Father's Death at 57.

He came in on a Tuesday, mid-October. Forty-six years old, senior vice president at a logistics company, and he had printed out his lab results from Function Health. He sat across from me with a yellow highlighter and the kind of look that told me he was not sure whether to be angry or frightened.

“My LDL is 112,” he said. “That’s fine, right? That’s what my internist said. He told me I was a great patient and that I didn’t need another physical for three years.”

He slid the page across to me. His ApoB was 164 mg/dL.

I looked at the number. Then I looked at him. He had a father who had died at 57, sitting at a computer, no prior symptoms. His maternal grandfather had had heart attacks in his forties. He himself had never had a CAC score. His Lp(a), also on the Function Health panel, was 78 mg/dL.

“Your internist read the LDL,” I said. “He didn’t read this.”

I pointed to the ApoB.

“What does 164 mean?” he asked.

It means your arteries are being pelted by atherogenic particles at roughly twice the rate they should be. It means the genetic lottery your father lost, and your grandfather lost before him, is running in your blood right now, invisible to standard lipid panels, invisible to the seven-minute appointment that declared you fine. It means the number that predicts whether you will see your children graduate college is not the number your doctor looked at.

The question he asked next is the one this article answers: “If ApoB is the number that matters, why doesn’t anyone order it?”

The Mechanism

The standard lipid panel reports LDL-C: the estimated concentration of cholesterol inside LDL particles. It is calculated using the Friedewald equation, a formula developed in 1972 by Friedewald, Levy, and Fredrickson and published in Clinical Chemistry, not directly measured. It tells you how much cholesterol is in the LDL bucket. It does not tell you how many buckets there are.

ApoB, apolipoprotein B-100, is the structural protein on the outer surface of every atherogenic lipoprotein particle. One ApoB molecule per particle. Always. VLDL carries one. IDL carries one. LDL carries one. Each Lp(a) particle carries one. When a laboratory measures ApoB, it directly counts the total number of atherogenic particles in circulation. This is the distinction that governs atherosclerosis biology: it is particles, not cholesterol mass, that penetrate the arterial wall.

The mechanism of atherogenesis is particle-driven. ApoB-containing lipoproteins cross the arterial endothelium by transcytosis, a process that is proportional to the concentration of particles in circulation, not the cholesterol content of those particles. Once a particle enters the subendothelial space, it can be retained by binding to proteoglycans in the arterial wall matrix. Smaller, denser LDL particles, which characterize the metabolic syndrome phenotype, penetrate the endothelium more readily and are retained longer because their smaller size allows deeper diffusion into the arterial wall. Once retained, ApoB-containing particles are oxidized and taken up by macrophages via scavenger receptors. Those macrophages become foam cells. Foam cell accumulation is the early plaque. This is not a contentious area of cardiovascular biology: the particle retention hypothesis, formalized by Williams and Tabas in a series of papers beginning in 1995 in Arteriosclerosis, Thrombosis, and Vascular Biology, is the current mechanistic consensus.

The Friedewald equation has a specific failure mode that matters clinically: it systematically underestimates LDL-C, and therefore underestimates particle burden, when triglycerides are elevated. The equation assumes a fixed relationship between VLDL triglycerides and VLDL cholesterol. When triglycerides rise above approximately 150 mg/dL, that ratio shifts, the VLDL cholesterol is underestimated, and the calculated LDL-C is inflated relative to true LDL-C. The result is that in patients with metabolic syndrome, insulin resistance, or elevated triglycerides, the Friedewald-derived LDL-C looks better than the actual atherogenic particle burden. Martin et al., publishing in the Journal of the American College of Cardiology in 2013, developed the extended Martin-Hopkins equation to correct this, but the standard lipid panel in most laboratories still defaults to Friedewald.

In the metabolic syndrome phenotype, where elevated triglycerides shift LDL composition toward smaller, denser particles, a man can have an LDL of 112 while his particle count is running at a level that corresponds to an LDL of 165 or higher in terms of atherosclerotic burden. His LDL appears acceptable. His arterial wall is being exposed to accelerated plaque biology. The standard panel cannot detect this because it is measuring the weight of cholesterol cargo, not the number of delivery trucks.

The LDL-ApoB discordance exists in both directions. Some patients have an LDL of 140 with an ApoB of 85, meaning their particles are cholesterol-rich and fewer in number. Their LDL overestimates their particle burden. Statins reduce their ApoB effectively. Conversely, a man with LDL of 112 and ApoB of 164 has particles that are small and abundant. His LDL substantially underestimates his particle burden. The treatment strategy and the intensity both change. The direction of discordance determines whether the clinician is over-treating or under-treating based on LDL alone.

Lp(a), lipoprotein(a), adds a second layer of atherogenic mechanism that LDL-C cannot capture at all. Lp(a) is an LDL-like particle with an additional protein, apolipoprotein(a), disulfide-bonded to the ApoB. Apolipoprotein(a) has structural homology with plasminogen, the body’s primary clot-dissolving protein. Because it competes with plasminogen for fibrin binding sites, elevated Lp(a) simultaneously promotes plaque formation and impairs the fibrinolytic response that would otherwise limit thrombosis at a ruptured plaque. This dual mechanism, pro-atherogenic and pro-thrombotic, is why Lp(a) elevation is associated with both coronary artery disease and aortic valve calcification. Thanassoulis et al., writing in the New England Journal of Medicine in 2013, identified a variant in the LPA gene that raises Lp(a) and is independently associated with aortic valve calcification. Individuals with Lp(a) above 50 mg/dL carry approximately two to three times the rate of clinically significant aortic valve disease compared to those with normal Lp(a). This means elevated Lp(a) is not only a coronary risk marker; it is a reason to monitor the aortic valve with periodic echocardiography, a conversation most patients with elevated Lp(a) have never had.

What the Evidence Shows

The evidence that ApoB outperforms LDL-C as a predictor of cardiovascular events is not new and is not narrow. It spans multiple large cohorts, multiple geographic populations, and multiple study designs.

The MESA cohort analysis published in JAMA Cardiology in 2021 confirmed that ApoB predicted incident cardiovascular events significantly better than LDL-C across all subgroups, with the largest predictive advantage concentrated in the metabolic syndrome phenotype. This finding is important because it demonstrates that the patients most likely to be misclassified by LDL-C are precisely the patients most common in clinical practice: middle-aged men with abdominal adiposity, moderately elevated triglycerides, and LDL that looks acceptable.

The AMORIS study, published by Walldius et al. in The Lancet in 2001, followed 175,553 subjects and found that the ApoB/ApoA-I ratio predicted myocardial infarction more accurately than any other lipid measure including LDL-C, total cholesterol, and the total cholesterol/HDL ratio. The authors reported that the relationship between ApoB/ApoA-I and MI risk was continuous, graded, and present at ApoB levels that would not have triggered treatment under standard LDL-based guidelines. At 175,000 subjects, AMORIS is among the largest prospective lipid studies conducted.

The INTERHEART study, published by Yusuf et al. in The Lancet in 2004 and covering 52 countries and over 27,000 participants, found that the ApoB/ApoA-I ratio carried the strongest population-attributable risk of any lipid measure for myocardial infarction, across all regions, age groups, and sexes. The population-attributable risk for elevated ApoB/ApoA-I exceeded that of hypertension, diabetes, abdominal obesity, and current smoking when evaluated individually. This is not a marginal improvement in prediction; it represents the single strongest modifiable lipid risk signal in the largest global MI study conducted.

On Lp(a), the Copenhagen City Heart Study provided foundational outcome data. Kamstrup et al., publishing in the Journal of the American Medical Association in 2009, reported data from over 9,000 subjects followed for up to 10 years. Each doubling in Lp(a) concentration was associated with a 22 percent increase in MI risk and a 20 percent increase in ischemic heart disease risk, independent of LDL-C and standard risk factors. Critically, the relationship was causal: Mendelian randomization analyses using genetic variants that raise Lp(a) confirmed that the elevated Lp(a) itself drives risk, rather than being a marker correlated with some other causal variable. Lp(a) levels are approximately 80 to 90 percent genetically determined, do not respond meaningfully to lifestyle change, and are not reduced by standard statin therapy. They are determined largely at birth and remain stable throughout life after age 5.

Lp(a) is present at elevated levels (above 50 mg/dL or above 100 nmol/L) in approximately 20 percent of the global population. In populations of African heritage, the prevalence is substantially higher and the cardiovascular consequences are disproportionate. Novel agents specifically targeting Lp(a) are in late-stage clinical development. Pelacarsen, an antisense oligonucleotide targeting hepatic LPA mRNA, and olpasiran, a small interfering RNA, have both demonstrated 70 to 90 percent reductions in Lp(a) in Phase II and Phase III trials. The cardiovascular outcomes trial for pelacarsen (Lp(a) HORIZON) enrolled over 7,680 patients and results were reported in 2025. The Lp(a) that felt permanently untreatable as recently as 2022 is now the subject of two cardiovascular outcome trials with Phase III data.

The ESC/EAS treatment targets for ApoB are the clinical translation of this evidence:

For most adults in primary prevention without major risk factors, the ApoB target is below 100 mg/dL (LDL equivalent below 130 mg/dL).

For high-risk adults (diabetes with organ damage, multiple risk factors, stage 2-3 CKD), the ApoB target is below 80 mg/dL (LDL equivalent below 100 mg/dL).

For very high-risk adults (established ASCVD, familial hypercholesterolemia with established disease), the ApoB target is below 65 mg/dL (LDL equivalent below 70 mg/dL).

The 46-year-old man with ApoB 164 and a father who died at 57 sits functionally in the very-high-risk category despite having no established cardiovascular disease. The 2019 ACC/AHA risk-enhancing factors, published in JACC by Grundy et al., explicitly list premature ASCVD in a first-degree male relative before age 65 as a risk-enhancer that can move treatment thresholds downward. Combined with Lp(a) of 78 mg/dL, which exceeds the ACC/AHA risk-enhancing threshold of 50 mg/dL, this man’s functional risk category is substantially higher than his LDL-derived Framingham calculation would suggest. His ApoB target is below 65 mg/dL. He started at 164. That gap determines the treatment intensity.

The treatment pathway from ApoB 164 to target: high-intensity statin as first line produces approximately 40 to 50 percent ApoB reduction, bringing most patients from 164 to the 80 to 100 mg/dL range. Adding ezetimibe provides an additional 20 to 25 percent reduction. For patients remaining above target on statin plus ezetimibe, PCSK9 inhibitors (evolocumab or alirocumab) reduce ApoB by an additional 50 to 60 percent on top of existing therapy. Repeat ApoB measurement 6 to 8 weeks after initiating or changing therapy confirms whether target has been achieved.

What to Do This Week

1. Request ApoB at your next lab visit. Specify “ApoB (apolipoprotein B)” explicitly and confirm your physician received the result and reviewed it. Many electronic health record systems do not automatically flag an elevated ApoB if the LDL-C appears within the acceptable range. The test is widely available and costs approximately $15 to $30 at commercial laboratories.

2. Request Lp(a) if it has never been measured. It needs to be done once. Lp(a) levels are genetically determined and remain stable throughout adult life; a single measurement is informative for life. If you are of African heritage, this measurement is particularly important because prevalence of elevated Lp(a) is substantially higher in that population, and the cardiovascular consequences are disproportionate. Lp(a) above 50 mg/dL (or above 100 nmol/L) moves you into a category where all other modifiable risk factors require more aggressive management.

3. Calculate your ApoB-to-LDL discordance by dividing your ApoB (mg/dL) by your LDL plus 20. If this ratio exceeds 0.9, your particle count substantially exceeds what your LDL-C suggests. The larger the excess, the more critical ApoB becomes as your treatment target in place of LDL. An ApoB of 164 and an LDL of 112 gives a ratio of 1.24, well above the discordance threshold, which is why treating his LDL to a lower number with less intensive therapy would have left his particle burden largely unaddressed.

4. If you have a father, brother, or son who had a cardiac event before age 65, make sure your physician specifically evaluates whether your ApoB, Lp(a), and family history together place you in a higher risk category than your LDL-C alone would indicate. The 2019 ACC/AHA risk-enhancing factors, published by Grundy et al. in JACC, explicitly include premature ASCVD in a first-degree male relative before age 65 as a risk-enhancer that can lower the treatment threshold. Family history is not anecdote; it is phenotypic evidence of shared genetic risk.

5. Share your Lp(a) result with your first-degree relatives and tell them to get tested. Because Lp(a) is 80 to 90 percent heritable, each child of a parent with elevated Lp(a) has approximately a 50 percent chance of inheritance. A sibling of someone with elevated Lp(a) has elevated prior probability of carrying the same variants. Knowing your number protects not only you: it creates the clinical awareness that allows your children’s physicians to apply an appropriately low ApoB target when they are adults, and it gives your siblings the information they need to advocate for their own testing. A $30 blood test, done once, is one of the highest-yield preventive actions one generation can hand to the next.

The standard lipid panel was designed in the 1960s and 1970s for population-level epidemiology, when individual lipoprotein particle measurement was not technically or economically feasible at scale. It was never designed for individual-level risk precision. ApoB and Lp(a) are the measurements that close that gap: ApoB by counting every atherogenic particle rather than weighing their cholesterol cargo, and Lp(a) by capturing a genetically determined thrombogenic risk that the standard panel cannot see at all. The man sitting across the desk with a yellow highlighter and an LDL of 112 deserved both numbers from the first visit. The standard panel did not give them to him. That is the gap this article is about.