Why LDL-C Misleads
The number on every lipid panel — but not the right one
When your doctor mentions "cholesterol," they almost always mean LDL-C (LDL cholesterol). It's measured by default, it's what risk calculators plug in, and it's what statins lower. But it's not what clogs your arteries. Particles do.
LDL-C measures cargo, not trucks
LDL-C tells you how much cholesterol is being carried, not how many particles are carrying it. Two people with identical LDL-C of 120 mg/dL can have very different particle counts — and the one with more particles has more opportunities to deposit plaque. The particle count is what matters.
Discordance is common
Up to 50% of people with LDL-C under 100 still have elevated particle counts and hidden cardiovascular risk. This is why metabolic-syndrome patients can have "normal" cholesterol yet still develop heart disease — their particle count tells the real story.
Where the Friedewald formula breaks
Most labs don't even *measure* LDL-C — they calculate it using the **Friedewald formula** (LDL-C = Total Chol − HDL-C − Triglycerides/5). When triglycerides are above ~200 mg/dL, that formula systematically *underestimates* LDL-C, so people with metabolic dysfunction routinely look healthier on paper than they are. Direct-LDL or NMR-measured LDL-P fixes this. Your "good" LDL-C of 88 might actually be 115.
Small dense LDL — the worst sub-population
LDL particles aren't uniform. **Small, dense LDL (sdLDL)** penetrates the artery wall more easily, sits there longer, and oxidises faster than the large, buoyant variety. Two people with identical LDL-C of 130 can have radically different risk if one is mostly large-buoyant and the other is mostly sdLDL. NMR LipoProfile or VAP testing distinguishes them; standard panels do not.
Select every metric that outperforms LDL-C for predicting CVD events:
Pick the metric most tightly correlated with cardiovascular event rates across the evidence base.
The HDL paradox
Conventional wisdom: "good cholesterol, more is better." Reality: above ~80 mg/dL, **higher HDL-C is associated with HIGHER mortality**, not lower. Mendelian-randomisation studies on genetic variants that raise HDL-C found *no reduction* in CVD events. The CETP-inhibitor drug trials (torcetrapib, dalcetrapib) raised HDL-C dramatically and either failed or harmed. HDL function — how well it pulls cholesterol out of artery walls — matters; the static number on a lab does not.
What the recent guidelines actually say
The 2019 European Society of Cardiology / European Atherosclerosis Society dyslipidemia guideline recommends measuring **ApoB to refine risk assessment in anyone with elevated triglycerides, diabetes, obesity, metabolic syndrome, or very low LDL-C**. The 2018 ACC/AHA US guideline lists ApoB ≥ 130 mg/dL as a "risk-enhancing factor" that pushes you toward statin therapy. The science has moved; clinic practice still trails by years.
Key Takeaway
LDL-C is the first number every doctor orders and the first number every guideline sets targets on. But it measures cargo, not trucks — and on most labs it's a *calculated* estimate that breaks under metabolic dysfunction. The modern move is ApoB (particle count) plus Lp(a) — the pair that actually predicts heart attacks.
References
- Discordance between LDL-C and LDL-P — why particle count wins
- LDL particle number vs cholesterol content — MESA study
- Limitations of Friedewald-calculated LDL-C in hypertriglyceridemia
- Small dense LDL particles and CVD risk — meta-analysis
- Mendelian randomisation of HDL-C and myocardial infarction
- Extreme HDL-C associated with increased mortality — CANHEART study
- 2019 ESC/EAS dyslipidaemia guidelines — ApoB recommendations