ApoB and Lp(a): what a normal cholesterol panel misses

By Marcus Renshaw · · updated

For years my lipid panel came back unremarkable and I filed it as done. Total cholesterol fine, LDL-C inside the range, no flag. What I was not looking at was the number that actually tracks the thing clogging an artery: how many particles are carrying that cholesterol around. A panel can hand you a clean LDL-C and still hide the risk, and the two markers that close that gap both moved to the front of cardiovascular guidance in 2026.

A standard cholesterol panel measures the cholesterol carried inside your particles, not how many particles you have. ApoB counts every plaque-causing particle directly, so it catches risk a normal LDL-C misses. Lp(a) is a mostly genetic particle you measure once in a lifetime. In 2026 both were promoted to the front of cardiovascular risk assessment.

What does a standard cholesterol panel actually measure?

A basic lipid panel reports total cholesterol, HDL-C, triglycerides, and LDL-C. The one everyone anchors on, LDL-C, is the mass of cholesterol riding inside your LDL particles. It is usually calculated rather than measured directly, most often with the Friedewald equation, which drifts and breaks down when triglycerides run high, exactly the metabolic state where you most want an accurate read.

The deeper problem is what LDL-C cannot see. Cholesterol does not float loose in blood and it does not cause plaque on its own. It travels packed inside particles, and it is the particles that lodge in the artery wall. Two people can carry the identical LDL-C while one has a modest number of large, cholesterol-rich particles and the other has a swarm of small, cholesterol-poor ones. Same cholesterol mass, very different particle counts, and the second person is at higher risk while the panel calls them equal.

Why does ApoB predict risk better than LDL-C?

Every atherogenic particle, whether LDL, IDL, VLDL, or Lp(a), carries exactly one molecule of apolipoprotein B on its surface. Measure ApoB and you are counting those particles one for one, a direct headcount instead of a guess at the average cargo. That is why multivariable genetic analyses point at ApoB as the lipid measure that actually drives coronary artery disease: the particle number, not the cholesterol it happens to carry.

The gap between the two markers has a name, discordance, and it is where ApoB earns its keep. When LDL-C reads normal but ApoB reads high, the extra particles carry real, measurable risk that the cholesterol number missed. This is common in the people reading a blog like this: anyone with insulin resistance, high triglycerides, or a serious recomp in progress tends to run small, cholesterol-depleted LDL, which is precisely the pattern that makes LDL-C look better than the particle count justifies.

ApoB has two practical advantages on top of the biology. It is measured directly rather than calculated, so it does not fall apart at high triglycerides. And it does not require fasting, which removes one more source of noise from a marker you want to compare draw after draw.

What is Lp(a), and why do you only measure it once?

Lipoprotein(a) is an LDL-like particle with an extra protein bolted on, and it is nasty out of proportion to how little anyone hears about it. It is atherogenic, pro-inflammatory, and pro-clotting, and it drives both coronary disease and aortic valve narrowing.

The reason you test it once and never again is that your level is set by genetics. It is roughly ninety percent heritable, stable across your adult life, and it lands high in about one in five people. Diet does not move it. Exercise does not move it. Statins do not lower it and can nudge it up slightly. So a single measurement tells you your lifetime exposure, which is why the 2026 guideline recommends measuring it once in every adult and not repeating it. Drugs that lower Lp(a) directly are in late-stage trials as of 2026, cutting it 80 to 95 percent, but nothing is approved yet and the first trial testing whether that lowering prevents heart attacks has not reported, so for now the value of the test is knowing your number, not treating it.

Knowing it still changes decisions. A high Lp(a) is a standing reason to drive every other risk factor lower, ApoB included, because you cannot lower this one. And it reframes a family history of early heart attacks from bad luck into something you can actually name and account for.

What are the target numbers?

For Lp(a), the European consensus thresholds are the cleanest guide: under 75 nmol/L largely rules out Lp(a)-attributable risk, over 125 nmol/L rules it in, and the 75 to 125 band is a grey zone. Report it in nmol/L, not the older mg/dL mass units, since that is the current best-practice assay. The risk climbs continuously from there: the 2026 guideline puts Lp(a) over 125 nmol/L at about 1.4 times the risk and over 250 nmol/L at two times or more.

For ApoB, the guideline’s goal table puts a number against each LDL-C tier: under 90 mg/dL at the lowest tier, under 70 at higher risk, and under 55 if you already have cardiovascular disease. Read those as conditional rather than blanket targets. The under 90 appears in primary prevention only when triglycerides run 150 to 499 mg/dL, and in diabetes only without additional risk factors, and several of the higher-risk cells are written as optional. ApoB itself sits at Class 2a in the guideline, something to measure once you have already hit your LDL-C and non-HDL-C goals. As with any target, the number is a direction to move toward, not a line that certifies you are fine below it, which is the whole difference between a reference range and an optimal target.

How does this fit a self-experimenter’s panel?

Here is where the clinical guideline is the floor, not the answer. In the guideline, ApoB is framed as a tie-breaker, ordered mostly for people with diabetes, metabolic syndrome, or high triglycerides who have already hit their LDL-C goal. If you are optimizing for the next few decades rather than screening for disease, you do not wait for that narrow trigger. You put ApoB on the panel as your primary lipid marker and let it be the trend line you actually watch, because it is the honest particle count and LDL-C is a proxy for it.

The budget math is easy. ApoB is a cheap add to any draw, so there is no reason not to run it every time you pull lipids. Lp(a) is a single one-time test, on the order of a few hundred SEK, that you never repeat, which makes it one of the highest-value blood tests you will ever buy per krona. Order it once, record it, done.

Then account for your own stack, because it confounds these numbers like any others. If you run a GLP-1, a thyroid dose, or a SERM, your lipids shift, and high triglycerides quietly wreck the calculated LDL-C while leaving ApoB intact, which is one more reason to trust the particle count. A statin lowers ApoB but does nothing good for Lp(a). Omega-3 and nattokinase work on triglycerides and clotting, not on Lp(a) at all. None of that is visible if you only ever look at total cholesterol. And the standard caution holds: these markers inform decisions, they do not replace a clinician, and anything you change to move them is a conversation to have with one.

For me this is not theoretical. ApoB has been on my standing panel for a while, so I watch the particle count instead of the cholesterol mass. Lp(a) is the one test I keep deferring to the next budget cycle, which is exactly the mistake this article is arguing against, since it costs almost nothing and I only ever need to do it once.

This is the reason the bloodwork tracker in the Operating System bundle plots each marker against both its lab range and a target you set, and carries a marker like Lp(a) as a fixed one-time genetic value while ApoB gets a full trend across draws. A cholesterol panel read one row at a time hides all of this. The same markers, logged against their targets and watched over time, are what turn a lipid draw into a decision, the way the Year One framework treats every marker.

Add ApoB to your next blood draw, and if you have never measured Lp(a), order it this once. Those two lines tell you more about your cardiovascular risk than every total-cholesterol result you have ever filed away.

FAQ

Is ApoB better than an LDL-C or cholesterol test?

For risk, yes. LDL-C measures the cholesterol mass inside your LDL particles, while ApoB counts every atherogenic particle directly, since each one carries a single ApoB molecule. When the two disagree, the particle count predicts events better. ApoB is also measured directly rather than calculated, so it stays reliable when triglycerides are high, and it does not require fasting.

What is a good ApoB level?

The 2026 guideline pairs an ApoB number with each LDL-C tier: under 90 mg/dL at the lowest tier, under 70 at higher risk, and under 55 if you already have cardiovascular disease. They are conditional rather than blanket targets, since the under 90 applies to primary prevention only when triglycerides run 150 to 499 mg/dL, and to diabetes only without additional risk factors. Treat the number as a direction to move toward with your trend, not a threshold that certifies you are safe below it.

Why do you only test Lp(a) once?

Because it is about ninety percent genetic and stable across your adult life. Diet, exercise, and statins do not meaningfully lower it, so a single measurement captures your lifetime exposure. The 2026 guideline recommends measuring it once in every adult and not repeating it, unless a future Lp(a)-lowering drug changes that.

What Lp(a) level is high?

Under 75 nmol/L largely rules out Lp(a)-attributable risk and over 125 nmol/L rules it in, with a grey zone between. Risk rises continuously above that: roughly 1.4 times the average risk over 125 nmol/L and two times or more over 250 nmol/L. Ask for the result in nmol/L rather than the older mg/dL mass units.

Can you lower Lp(a)?

Not reliably with lifestyle or current drugs. Diet and exercise do not move it, and statins can raise it slightly. RNA-based therapies that lower it directly are in late-stage trials as of 2026 but are not yet approved. For now the point of testing is to know your number and to drive every other risk factor, ApoB included, lower to compensate.

Should I get ApoB and Lp(a) if my cholesterol is normal?

That is exactly when they are most useful. A normal LDL-C with a high ApoB is a common and quietly higher-risk pattern, especially with insulin resistance or high triglycerides. And Lp(a) is invisible on a standard panel entirely, so a clean cholesterol result tells you nothing about it. Both add information a basic panel cannot.

Do I need to fast before an ApoB test?

No. ApoB does not require fasting, which is one of its practical advantages over a calculated LDL-C. That makes it easier to compare across draws taken under slightly different conditions, since fasting state is one less variable moving your number between tests.

Sources