Is LDL cholesterol a cause of heart disease, not just a marker?
Claim attributed to Mainstream cardiology and lipidology, principally the European Atherosclerosis Society Consensus Panel (Ference et al.) and the Cholesterol Treatment Trialists' Collaboration. , This is the established scientific consensus, not a fringe or commercial position. A cholesterol-skeptic minority (Ravnskov and colleagues) contests it, mainly on LDL's weak or inverse link to all-cause mortality in the elderly.
Three independent lines of evidence (genetics, randomized trials across unrelated drug mechanisms, and large-scale epidemiology) converge on the same dose-dependent, cumulative relationship. This is one of the most robustly established causal claims in modern medicine.
Genetics, three unrelated drug mechanisms and epidemiology all point the same way: LDL drives plaque, so it is a cause, not just a number on a lab report.
What it’s supposed to target
- ApoB / LDL particles
- Arterial-wall retention
- Plaque formation (atherogenesis)
- Cumulative lifetime exposure
The mechanism is the response-to-retention model. Circulating apoB-containing particles (LDL chief among them) cross into the artery wall and get trapped by the connective-tissue matrix; trapped particles are oxidized and modified, draw in immune cells that engulf them to become foam cells, and over years build the lipid-laden plaque that narrows arteries and ruptures to cause heart attacks and strokes. The driver is the number of particles depositing over time, which is why apoB or LDL-particle count can matter more than the cholesterol cargo (LDL-C) they carry.
Crucially, the damage is cumulative: it is LDL concentration multiplied by years of exposure, like pack-years for smoking. That is why people with lifelong genetically low LDL are strongly protected, and why lowering LDL earlier and by any mechanism (statin, ezetimibe, PCSK9 inhibitor) cuts events in proportion to the absolute drop. Unlike most entries here, this is not a hopeful theory: the pathway, the genetics, and the trials all point the same way, which is what makes 'LDL is causal' settled rather than contested.
Mechanism is theory, not proof. A plausible pathway explains why something might work, not whether it does. The verdict rests on the evidence below, not the elegance of the theory.
What would have to be true
Lowering LDL by independent mechanisms must reduce events proportionally to the LDL drop. HOLDS: statins, ezetimibe and PCSK9i all do.
Lifelong genetically lower LDL must lower coronary risk independent of any drug. HOLDS: Mendelian randomization shows exactly this.
A coherent biology must explain how LDL particles initiate plaque. HOLDS: the response-to-retention model.
'Causal' would have to mean sole cause. DOES NOT HOLD, and need not: it means a major modifiable contributory cause in a multifactorial disease.
What the evidence actually shows
Genetics supplies the ethical equivalent of a lifelong randomized trial
A direct RCT raising LDL to test harm would be unethical and will never be run, but Mendelian randomization fills the gap. Variants in many unrelated genes (PCSK9, HMGCR, NPC1L1, LDLR) that lower LDL each lower coronary risk dose-dependently, with no heterogeneity by gene. Ference 2012 found a 54.5% CHD reduction per 1 mmol/L of lifelong lower LDL, roughly a 3-fold larger per-unit effect than statins started later, demonstrating that exposure is cumulative and causal. Because genetic instruments cannot be confounded by drug-company sponsorship, this line is independent of pharma.
Three unrelated drug mechanisms all scale benefit to the LDL drop
The EAS Consensus Panel concluded the evidence 'unequivocally establishes that LDL causes ASCVD.' Statins (CTT meta-analyses, ~21-22% fewer major vascular events per 1 mmol/L), ezetimibe (IMPROVE-IT, HR 0.94) and PCSK9 inhibition (FOURIER, HR 0.85) all cut events roughly proportional to the absolute LDL reduction. That three mechanistically distinct classes produce benefit scaled to LDL is strong evidence LDL itself is the causal node, not an off-target effect. ASCVD remains multifactorial: LDL is necessary-but-not-sufficient, the central driver, not the sole cause.
Studies, graded, and who paid
Mendelian randomization plus multi-mechanism RCTs plus epidemiology all agree on a dose-dependent effect.
Well supported by retention biology; apoB can diverge from LDL-C and is arguably the better measure.
Genuinely murky: reverse causation and competing risks confound the observational data.
| # | Study | Type | Size | Funding / COI | Key limitations |
|---|---|---|---|---|---|
| 1 | EAS Consensus Panel (Ference et al. 2017) | Consensus synthesis of genetic, epidemiologic and RCT evidence | >200 prospective studies (~2M), MR across 50+ genes, 30+ RCTs (~200,000) | Mixed EAS panel; authors report pharma ties, but the synthesized genetic evidence is industry-independent. Disclosures not fully captured. | Narrative synthesis, not a single pooled analysis; conclusions inherited from constituent studies. |
| 2 | Ference et al. 2012, JACC (Mendelian randomization) | MR meta-analysis | 312,321 participants | Independent Genetic-instrument design not subject to sponsorship confounding; no industry funding identified. | MR assumes variants act only via LDL; estimates lifelong, not short-term, exposure effects. |
| 3 | CTT Collaboration (Oxford CTSU statin meta-analyses) | Individual-participant-data meta-analysis of ~28 statin RCTs | ~175,000 participants | Mixed Academically led (Oxford CTSU; MRC, Cancer Research UK, British Heart Foundation); underlying statin RCTs were industry-funded. | Pools heterogeneous trials; underlying trials carry sponsor effect-size optimism. |
| 4 | FOURIER (Sabatine et al. 2017, evolocumab) | Randomized, double-blind, placebo-controlled trial | 27,564 patients with established ASCVD | Industry-funded Funded by Amgen, maker of evolocumab; sponsor-favorable result, flagged. Mechanistically independent of statins/ezetimibe. | Median follow-up only 2.2 years; LDL reduced 59% (92 to 30 mg/dL), HR 0.85 (0.79-0.92). |
| 5 | Ravnskov et al. 2016 BMJ Open, with Oxford CEBM rebuttal | Systematic review of observational cohorts (skeptic view) plus post-publication peer review | ~68,000 elderly across 19 cohorts | Independent Authored by cholesterol skeptics (THINCS); some authors hold anti-statin advocacy positions. Included to represent and rebut the dissent. | CEBM finds limited validity: statin-prescription bias, HDL-C confounding (only 3/19 adjusted), single-database search; all-cause-mortality endpoint cannot rebut ASCVD causality. |
The decisive feature is convergence: pharma-independent genetics and industry trials agree on the same dose-dependent magnitude, which is precisely why the causal claim is not a marketing artifact.
Unproven ≠ disproven
Net benefit of aggressive LDL lowering in the very elderly or frail is less certain than in middle age, where competing mortality and reverse causation complicate the data.
Where claim and evidence diverge
Almost nothing relevant is missing here; a trial deliberately raising LDL is unethical, but Mendelian randomization supplies the ethical equivalent via lifelong genetic allocation.
The money trail
The big outcome trials are industry-funded (Merck for ezetimibe, Amgen for FOURIER, statin makers behind the CTT trials).
But the Mendelian-randomization genetics and the academically led CTT meta-analytic work are not contingent on drug-company sponsorship, and genetic instruments cannot be confounded by it.
The honest read
LDL is a cause of heart disease, not merely a marker; this is settled science resting on a rare convergence of genetics, multi-mechanism trials and epidemiology.
'Causal' means a major modifiable contributory cause within a multifactorial disease, and apoB / particle number is the more precise target.
What would change this verdict
Mendelian-randomization studies stop showing dose-dependent coronary risk reduction with lifelong lower LDL across unrelated genes.
A novel LDL-lowering mechanism reduces LDL substantially yet fails to reduce ASCVD events proportionally.
Sources
- Ference BA, Ginsberg HN, Graham I, et al. Low-density lipoproteins cause atherosclerotic cardiovascular disease. 1. Evidence from genetic, epidemiologic, and clinical studies. A consensus statement from the European Atherosclerosis Society Consensus Panel. Eur Heart J. 2017;38(32):2459-2472.
- Ference BA, Yoo W, Alesh I, et al. Effect of long-term exposure to lower LDL-C beginning early in life on the risk of coronary heart disease: a Mendelian randomization analysis. J Am Coll Cardiol. 2012;60(25):2631-2639.
- Cholesterol Treatment Trialists' (CTT) Collaboration, University of Oxford CTSU. Individual-participant meta-analyses of statin trials.
- Sabatine MS, Giugliano RP, Keech AC, et al. Evolocumab and clinical outcomes in patients with cardiovascular disease (FOURIER). N Engl J Med. 2017;376(18):1713-1722.
- Oxford CEBM post-publication peer review of Ravnskov et al. 2016 (Lack of an association or an inverse association between LDL cholesterol and mortality in the elderly), BMJ Open.
- Cannon CP, Blazing MA, Giugliano RP, et al. Ezetimibe added to statin therapy after acute coronary syndromes (IMPROVE-IT). N Engl J Med. 2015;372(25):2387-2397. PMID 26039521.
People also ask
- Does LDL cholesterol actually cause heart disease or just predict it?
- It causes it. Three independent lines of evidence (genetics via Mendelian randomization, randomized trials across unrelated drug mechanisms, and large-scale epidemiology) converge on the same dose-dependent relationship. This makes LDL a major modifiable cause of atherosclerotic disease, not just a marker.
- What is Mendelian randomization and why does it prove LDL is causal?
- Mendelian randomization uses inherited gene variants as a lifelong natural experiment. Variants in unrelated genes (PCSK9, HMGCR, LDLR) that lower LDL each lower coronary risk dose-dependently. Because genetic instruments cannot be confounded by drug-company sponsorship, this line is independent of pharma.
- Is apoB a better measure than LDL cholesterol?
- Likely yes. The proximate agent appears to be apoB or LDL particle number, not the cholesterol cargo itself, supported by retention biology. ApoB can diverge from LDL-C and is arguably the more precise target, though this sub-claim is rated slightly less certain than the core causal finding.
- Does lowering LDL help very elderly or frail people live longer?
- This is genuinely uncertain. Net benefit of aggressive LDL lowering in the very elderly or frail is less established than in middle age. Reverse causation and competing mortality risks confound the observational data, so the all-cause mortality picture in this group remains murky.
Caveat is journalism, not medical advice. We check public claims against published evidence; we don’t diagnose, treat, or tell you what to take.