Pillar 01 · Heart

Heart disease starts decades before the diagnosis

Atherosclerosis is driven by ApoB (apolipoprotein B), a count of the atherogenic particles in your blood. The damage is cumulative: particle concentration multiplied by years of exposure, the area under a lifelong curve [Ference BA, 2017] [Zhang Y, 2021]. Measured against optimal targets instead of lab “normal,” most adults are already above the line, even when a standard panel and a 10-year risk score say they are fine. Two numbers and one decision show where you actually stand, and as of 2026, US guidelines have moved a long way toward this view [Blumenthal RS, 2026].

Lp(a): check it once

Lipoprotein(a), said “ell-pee-little-ay,” is an LDL-like particle whose level is set almost entirely by your genes: diet and exercise barely move it, and one measurement usually holds for life. About 1 in 5 people carry an elevated level, and almost none of them know it.

Risk: Borderline / grey zone

Lp(a) is largely genetic, measured ~once in a lifetime, not meaningfully lowered by statins or lifestyle (targeted drugs are in late-stage trials), and adds to ApoB-driven risk. Bands follow the EAS 2022 consensus; the 2026 US guideline uses a single "elevated" threshold of ≥50 mg/dL (≥125 nmol/L), with risk roughly doubled from ~100 mg/dL (~250 nmol/L).

Sources

That is the case for the test in one sentence: a single, cheap draw uncovers a hidden, largely genetic multiplier that should change how aggressively you manage everything else on this page, starting with your ApoB and blood pressure. The 2026 US guideline now agrees, recommending a one-time Lp(a) measurement for every adult [Blumenthal RS, 2026]. And because it is inherited, a high result is worth sharing: each of your first-degree relatives has roughly a coin-flip chance of carrying it too.

Drugs that lower Lp(a) are in the pipeline

Nothing you can do today meaningfully lowers Lp(a): statins do not (they can even nudge it up slightly), lifestyle does not, and today’s PCSK9-blocking antibodies trim it only by about a quarter [Kronenberg F, 2022]. That is changing. Several targeted drugs now in phase 3 trials cut Lp(a) by 80 to 95%, including lepodisiran, an injection given as little as once a year [Nissen SE, 2025], and muvalaplin, a daily pill. What none of them has shown yet is the only thing that matters, fewer heart attacks; the first outcomes verdict (the pelacarsen Lp(a)HORIZON trial) is due imminently. Until then, a high Lp(a) is managed indirectly: drive ApoB and blood pressure lower than you otherwise would.

ApoB: where you actually sit

Here is the uncomfortable part: “optimal” ApoB is far below what a standard panel treats as unremarkable. Physiologically normal LDL, the level seen in hunter-gatherers, healthy newborns, and other mammals that do not develop atherosclerosis, is only about 50 to 70 mg/dL [O'Keefe JH Jr, 2004], which maps to an ApoB well under the 65 mg/dL top of the “Optimal” band below. Typical adult levels run much higher, so most people who check, including plenty whose lab flagged nothing, land in “High” or “Very high.” ApoB is also a better measure of risk than LDL-C alone [Sniderman AD, 2019].

Tier: High

Sources

Two footnotes to the gauge. If your LDL-C is 190 mg/dL or higher, that is a different conversation, possible familial hypercholesterolemia (about 1 in 250 people), and it warrants a doctor now, whatever any risk score says [Grundy SM, 2019]. And if your triglycerides run high, that is a separate, mostly metabolic flag [Quispe R, 2021]: see Metabolism for the insulin-resistance side of that story.

Lowering ApoB earlier shrinks the area under the curve dramatically, which is why a 10-year risk score, which under-weights younger people, misses the point for prevention. Drag across the chart to read your cumulative exposure at any age. The dashed line, about 5,000 mg·years, is an illustrative marker of the accumulated plaque burden at which heart attacks typically begin to occur [Ference BA, 2018] [Ference BA, 2024]. Crossing it is not an appointment with an event, and staying under it is not immunity; risk rises continuously with the area under the curve. What the threshold makes visible is the one variable you control: the age at which you would cross it, and how far an earlier, lower ApoB pushes that age out.

Hover or drag across the chart to read the cumulative exposure at any age.

Cumulative exposure from birth to age 100: 7275 mg·years. Crosses the illustrative threshold at about age 62 — the age this model expects accumulated plaque to reach the level where heart attacks start occurring. The goal is to push it later.

ApoB or non-HDL-C are preferred; LDL-C is converted approximately and individual ApoB can differ (which is why measuring ApoB is best). Illustrative model: exposure accrues from birth; ApoB is assumed to rise linearly with age (adjust the per-decade rate), anchored to your current value. An intervention drops ApoB and holds it. Real trajectories vary. The dashed threshold (an LDL-C–derived ~5,000 mg·years) approximates the cumulative exposure, and with it the plaque burden, at which myocardial infarctions typically begin to occur. Crossing it is not an event date, and it is not a cliff: risk rises continuously with the area under the curve. The practical goal is moving your crossing age later, ideally past the right edge of the chart.

Sources
Why “normal” on a lab panel is already too high

Lab reference ranges describe the population, and the population is not fine. Even the 2026 guideline, a large step in the right direction, sets ApoB goals of 90, 70, or 55 mg/dL depending on risk tier [Blumenthal RS, 2026], while many panels do not flag ApoB until well above any of those. And because atherosclerosis is driven by cumulative exposure [Ference BA, 2018], a number a panel waves through keeps accruing damage every year it sits there.

Take two people a standard lab calls normal (illustrative, using the same cumulative-exposure model as the chart above against its 5,000 mg·years threshold):

  • An ApoB of 90 mg/dL, exactly at the guideline’s most lenient goal, crosses the lifetime-exposure threshold around age 56.
  • An ApoB of 110 mg/dL crosses it around age 45, about a decade earlier.

On this site’s prevention-optimal scale those same two numbers are already “High” and “Very high.” The gap between “goal met” and optimal is the case for preventive cardiology.

How low is too low?

There is no known floor. In FOURIER, driving LDL-C down to under 20 mg/dL kept cutting events with no threshold of benefit and no rise in adverse events [Giugliano RP, 2017]. That evidence is from high-risk secondary-prevention patients, so it argues for as low as reasonably achievable rather than a fixed target of 20 for everyone. The pattern now reaches earlier, too: in 2025, the PCSK9-blocking antibody evolocumab cut first-ever major events by 25% in high-risk people with no prior heart attack or stroke, at an achieved LDL-C around 45 mg/dL [Bohula EA, 2025]. Either way, the “Optimal” band above stays open-ended toward zero.

When should these first be checked?

Earlier than almost anyone does it. US consensus recommends a universal lipid screen at ages 9–11 and again at 17–21, mainly to catch familial hypercholesterolemia before decades of exposure accrue [National Lipid Association, 2015]. For adults the practical answer is: with your next routine panel. ApoB and Lp(a) are cheap add-ons, and a measurement in your 20s or 30s anchors your lifetime trajectory while there is still maximal time to bend it [Zhang Y, 2021].

Statins: the highest-leverage lever

For most people who check, a statin is worth a serious conversation, not a last resort. The benefit is roughly proportional to how much you lower ApoB and for how long, with no floor found yet [Giugliano RP, 2017], so almost everyone sits somewhere on that curve where lower is better. Statins are among the most-studied, safest drugs we have, and they are under-used relative to that benefit. The higher your ApoB or Lp(a), or the more other risk you carry, the stronger the case, and starting earlier cuts lifetime exposure instead of reacting to an event. Bring your numbers to your doctor.

For years that stance leaned further than the guidelines. In 2026 the guidelines caught up most of the way: the new ACC/AHA dyslipidemia guideline scores 30-year as well as 10-year risk, endorses “lower for longer,” and for the first time supports considering a statin from age 30 for people with high LDL-C or high 30-year risk [Blumenthal RS, 2026]. What is still “prevention-optimal” rather than official is treating optimal, not “goal met,” as the finish line.

Intensity LDL-C lowering Typical regimens When used
high ≥50% Atorvastatin 40–80 mg, Rosuvastatin 20–40 mg Clinical ASCVD, LDL-C ≥190 mg/dL, or high estimated risk; goal of maximal lowering.
moderate 30–49% Atorvastatin 10–20 mg, Rosuvastatin 5–10 mg, Simvastatin 20–40 mg, Pravastatin 40–80 mg Intermediate risk, or when high-intensity is not tolerated.
low <30% Pravastatin 10–20 mg, Simvastatin 10 mg When higher intensities are not tolerated.

The intensity ladder from the AHA/ACC guidelines. Since 2026, guidelines steer by numeric goals (LDL-C under 100, 70, or 55 mg/dL by risk tier) rather than intensity alone; the ladder is how you climb to them.

Sources

But don’t statins have side effects?

The most common serious reason to avoid a statin is genuine intolerance, and it is less common than its reputation. In blinded n-of-1 trials, people who had previously quit statins over side effects took a statin, a placebo, and nothing in random months without knowing which was which: about 90% of the symptom burden they blamed on the statin also appeared on placebo [Wood FA, 2020], and a separate series found no overall difference in muscle symptoms between statin and placebo months [Herrett E, 2021]. Half of the participants in the first trial were able to restart a statin afterward.

Two other worries, briefly. Statins do slightly raise the odds of a new type 2 diabetes diagnosis, roughly 9% in relative terms and almost entirely in people already near the diagnostic line [Sattar N, 2010]; if that is you, Metabolism is the other half of this conversation, and the cardiovascular benefit still wins on net. And on memory: a 2025 meta-analysis covering over 7 million people found statin users had lower dementia risk, not higher [Westphal Filho FL, 2025].

When side effects are real, the usual fix is a lower dose, a different statin, or a non-statin, such as ezetimibe or bempedoic acid (which cut events in a trial run specifically in statin-intolerant patients [Nissen SE, 2023]), rather than giving up on lowering ApoB. For most people the question is tolerability, not whether it would help.

Beyond statins: antibodies, silencers, and a one-time edit

Statins are the foundation because they are cheap, oral, and carry decades of safety data, but they are not the ceiling. PCSK9 inhibitors (evolocumab, alirocumab), antibody injections every two to four weeks, cut LDL-C by roughly another 60% on top of a statin and reduce events [Sabatine MS, 2017]; they are the drugs behind the very-low-LDL results above. Inclisiran silences the same gene with an injection every six months (after starter doses at the outset and month three) for about a 50% reduction [Ray KK, 2020], though its own outcome trials only read out around 2027. And the logical endpoint of “lower for longer” is now in human trials: a one-time gene edit that turns down PCSK9 in the liver cut LDL-C by up to 62%, durable so far out to 18 months of follow-up [Vafai SB, 2026]. That is lifetime exposure treated as the target, set once. It is years from routine use, and the practical escalation path today remains statin, then ezetimibe, then a PCSK9-class drug, with insurers gating each step.

The honest case against starting early

Steelmanned, it goes: statin trials run about five years, and over that horizon the absolute risk reductions are modest, roughly 0.8% for death and 1.3% for heart attack across trial populations [Byrne P, 2022], and smaller still for a low-risk 35-year-old. No trial has ever enrolled low-risk young adults for the forty years the prevention argument actually spans, and none ever will; the USPSTF accordingly rates the under-40 evidence “insufficient” [US Preventive Services Task Force, 2022].

The counter is that the benefit compounds. Across 60 trials, the risk reduction per unit of LDL lowering grows with each additional year of treatment [Burger PM, 2024], and people whose genes give them lifelong lower LDL get roughly three times the per-mmol protection that statins started in middle age deliver [Ference BA, 2012]. Starting early is a bet that decades of compounding beat decades of exposure. The 2026 guideline now partly endorses that bet, but it is still a bet, worth making with eyes open.

(At the other end of life the evidence gap is about to close: two definitive trials of statins in adults 70 and older, STAREE and PREVENTABLE, read out in 2026–27 [STAREE Investigators, 2026] [PREVENTABLE Investigators, 2026].)

On the fence? What a calcium score adds

A coronary artery calcium (CAC) scan is a quick, no-contrast CT that counts calcified plaque: direct evidence of the disease rather than an estimate of it. The 2026 guideline gives it a formal tie-breaker role when the statin decision is genuinely uncertain: a score of 0 can justify deferring, any score above 0 argues for treating to an LDL-C goal under 100 mg/dL, and higher scores argue for lower goals [Blumenthal RS, 2026]. There is now randomized evidence behind this: CAC-guided prevention slowed plaque progression in the CAUGHT-CAD trial [Nerlekar N, 2025].

One caveat, and it is this page’s thesis in miniature: calcium is a late signature. A zero in your 30s or 40s is expected and does not rule out the soft plaque that calcifies later, so a young zero is a reason to keep ApoB low, not a license to ignore it.

The rest is worth knowing, but the three things above are what move the needle, with one exception: if you smoke, quitting outranks everything else on this page.

Blood pressure

High blood pressure compounds ApoB-driven risk by accelerating arterial damage. “Normal” is under 120/80 [Whelton PK, 2018]. The 130/80 below is the diagnostic line where Stage 1 hypertension begins and the formal treatment goal for most adults; the prevention-optimal target is the 2025 guideline’s encouragement to get systolic as close to under 120 as you tolerate [Jones DW, 2025]. The same guideline ties pressure to the brain, with a Class 1 finding that keeping systolic under 130 protects against cognitive decline and dementia [Jones DW, 2025]; see Brain. Because office readings run high, judge yourself against home readings: rested, seated, arm supported at heart level, two readings morning and evening for a week, averaged.

Category: Normal

Zones follow the 2017 ACC/AHA categories; because the higher of the systolic and diastolic categories applies, the bands are L-shaped. Your reading is the dot. The dashed green box is the prevention-optimal target: the 2025 ACC/AHA guidance to get as close to under 120/80 as tolerated. Judge yourself against home readings (rested, seated, averaged); office readings run high.

Sources
The history rhymes

We have seen this arc before, with blood pressure. Life-insurance actuaries quantified that higher blood pressure predicted earlier death back in the 1920s, yet for decades mainstream medicine treated moderately elevated pressure as benign “essential” hypertension [Kotchen TA, 2011]. Only after landmark trials, and much later trials of lower targets like SPRINT [SPRINT Research Group, 2015], did guideline thresholds fall: the 2017 ACC/AHA guideline moved “high blood pressure” down to 130/80 [Whelton PK, 2018], and the 2025 guideline now urges getting as close to under 120/80 as tolerated [Jones DW, 2025]. ApoB looks like the same story, earlier in its arc.

Lifestyle: what it moves, and what it can’t

Lifestyle is not the main lever on ApoB; that is why the statin conversation exists. But the honest numbers are worth knowing, and two of the items below outrank everything pharmacological.

Your risk score, in context

Conventional scores estimate your next 10 years. The AHA PREVENT model (race-free) is the current standard, and since 2026 it is what the guideline itself runs on, with action bands starting far lower than most people expect (borderline begins at 3%) [Khan SS, 2024] [Blumenthal RS, 2026]. For ages 30–59 it also projects 30-year risk, the guideline’s answer to the complaint this page has always made: a 10-year window under-weights younger people.

Estimated 10-year ASCVD risk: 1.3% · Low

Estimated 30-year ASCVD risk: 8.7%

AHA PREVENT (race-free), the risk engine the 2026 guideline runs on. Its 10-year action bands: under 3% low, 3–5% borderline, 5–10% intermediate, 10%+ high. A 10-year window under-weights younger people, which is exactly what the 30-year figure and the lifetime exposure above are for.

Sources
What the calculators don’t see

Risk equations run on a handful of inputs, and several well-documented multipliers are not among them: a family history of premature heart disease, chronic inflammatory disease, and, for women, premature menopause (before 40) or a history of preeclampsia or gestational diabetes [Parikh NI, 2021]. Any of these argues for reading a borderline score as a higher one. The menopause transition itself also shifts LDL and its atherogenic subclasses adversely around the final menstrual period [El Khoudary SR, 2023], so a pre-menopause panel may no longer describe you: re-measure on the far side.