Rethinking Cholesterol: Prof. Tim Noakes on Why the Lipid-Heart Hypothesis Fails

For more than half a century, cholesterol has been cast as the villain of heart disease. Doctors warn patients about “bad cholesterol,” prescribe statins, and urge low-fat diets. Yet what if this central story of modern cardiology is wrong?

In a recent Nutrition Network Live interview and companion lecture on familial hypercholesterolemia (FH), Prof. Tim Noakes – emeritus professor, scientist, and Nutrition Network’s Chief Medical Director – presented a provocative case: cholesterol is not the root cause of coronary heart disease.

“The evidence is overwhelming that cholesterol is not causing heart disease… and that familial hypercholesterolemia, which is held up as absolute proof, actually disproves the hypothesis.” — Prof. Tim Noakes

This article walks clinicians and lay readers alike through Prof Noakes’ logic.

Explore the full training: Nutrition Network — Cardiovascular Health: A Metabolic Perspective

Keys’ Curve and the Causation Trap

The modern cholesterol narrative begins with Ancel Keys in the 1950s. Keys published a graph implying a near-linear relationship between dietary fat intake and coronary heart disease across six countries. But he excluded many countries that didn’t fit the line, which undermines the inference of any universal relationship [1].

The larger problem is methodological: association ≠ causation. As Noakes stresses, showing that two things travel together doesn’t prove that one causes the other. When researchers later ran randomized controlled trials to test diet and cholesterol-lowering, the results did not demonstrate the large clinical benefits that would be expected if the causal pathway were dominant [2].

“Every cardiologist still shows you those straight-line relationships… But these are associations, and associations cannot prove causation.” – Prof. Tim Noakes

Statins: Why “Events While on Treatment” Matter

If LDL cholesterol causes coronary disease, then lowering LDL should prevent events. Yet across large statin trials, substantial proportions of patients who experienced major coronary events were already taking statins at the time, a point highlighted years ago by Peter Libby in a perspective on the “forgotten majority” whose risk persisted despite intensive therapy [3]. The persistence of events implies that the disease process is not fully explained by LDL lowering.

Noakes’ bottom line is not that statins never help, but that the magnitude of benefit is often small in absolute terms, far smaller than headlines suggest. Meta-analytic summaries that emphasize relative risk can make modest absolute effects seem large; when expressed as absolute risk reductions, typical benefits are <1% over trial horizons for all-cause mortality and stroke, and ≈1% for myocardial infarction [4].

“If drugs that lower blood cholesterol do not prevent further coronary events, then cholesterol is not the cause.” – Prof. Tim Noakes

Relative vs Absolute Risk – The Number Patients Need

Why the gap between perception and reality? Because relative risk reductions (e.g., “risk cut by 44%”) can be dramatically larger than the corresponding absolute reductions.

In classic cohort data used to justify cholesterol lowering (e.g., MRFIT-style analyses visualized by David Diamond), those with the highest cholesterol showed a roughly fourfold higher relative risk of death than those with the lowest—yet survival differed by only ~1% in absolute terms over the study period (≈99.7% vs 98.7%) [5]. Noakes’ point: if the absolute difference is that small, treating everyone with a high LDL number may yield very modest population benefit, and individual decisions should weigh the actual absolute numbers.

“The difference between having a high cholesterol or a low cholesterol was one in a hundred.” — Prof. Tim Noakes, summarizing the absolute risk gap

Familial Hypercholesterolemia: The “Proof” That Isn’t

Medical students are often taught that familial hypercholesterolemia (FH) proves the lipid–heart hypothesis: very high LDL from birth → early heart disease. But the longitudinal evidence is far more nuanced.

• In the Simon Broome Registry, excess coronary mortality in FH was substantial in younger adults but attenuated with age, such that patients who survived into later life did not show a persistent high risk compared with the background population [6].
• A classic BMJ analysis of a large FH pedigree across two centuries found that many FH family members lived into old age, particularly in more recent eras, indicating substantial heterogeneity and likely environmental/other risk factor interactions [7].
• In SAFEHEART (a Spanish FH registry), diabetes, adiposity, smoking and hypertension strongly predicted cardiovascular events, whereas LDL-C itself did not independently predict events in multivariable models that included these factors [8].

“Close to 50% of people with FH have a calcium score of zero… They will not develop coronary atherosclerosis despite lifelong high cholesterol.” – Prof. Tim Noakes

Coronary Artery Calcium (CAC) Beats LDL for Risk

If you want to know who is at risk, measure disease – not just a surrogate. The coronary artery calcium (CAC) score quantifies calcified plaque and strongly predicts future events.

• In symptomatic adults, CAC tracked cardiovascular event rates far more closely than LDL-C levels; LDL-C strata showed little separation once CAC was accounted for [9].
• In FH cohorts, baseline CAC predicted long-term outcomes: individuals with CAC = 0 had very low event rates over 10–20 years, regardless of LDL-C, whereas higher CAC strata fared worse [10].
• Across multiple FH studies, ~45–50% of patients had CAC = 0, contradicting the notion that lifelong high LDL inevitably creates heavy plaque burden [10].

“Knowing a person’s blood cholesterol without knowing their CAC is of little real value.” – Prof. Tim Noakes

Clinical implication: For both clinicians and patients, CAC (where appropriate) can reclassify risk and guide the intensity of prevention more accurately than chasing an LDL number alone.

The Stronger Culprits: Insulin Resistance, Inflammation, Thrombosis

If LDL isn’t the prime mover, what is? Noakes emphasizes three interlocking processes:

1. Insulin resistance / hyperinsulinemia
   In FH and non-FH populations alike, markers of insulin resistance (central adiposity, elevated insulin, metabolic syndrome) track more strongly with risk than LDL-C [8]. In the Women’s Health Study, a composite lipoprotein insulin resistance (LPIR) score was one of the strongest predictors of incident cardiovascular disease in women, dramatically outstripping LDL-C in hazard magnitude [11].
2. Inflammation
   High-sensitivity C-reactive protein (hsCRP) independently predicts events and cardiovascular mortality—even among statin-naïve or statin-intolerant patients—more strongly than LDL-C in contemporary cohorts [12]. This dovetails with Noakes’ emphasis on endothelial injury and repair rather than a simple “lipid infiltration” model.
3. Thrombosis and poor repair
   Following lines of argument popularized by Dr Malcolm Kendrick, Noakes highlights thrombus formation and impaired repair at sites of endothelial damage as key mechanisms in atherosclerosis, with lipoproteins appearing more as responders/repair materials than prime offenders (consistent with multiple pathology observations) [13].
   

A Twist: Statins and Coronary Calcification

One counterintuitive observation is that long-term statin therapy can increase coronary artery calcification—a finding that has been interpreted by some as “plaque stabilization” (more calcium, less soft, rupture-prone material), but which still complicates simple narratives because higher CAC generally associates with higher risk [14]. At minimum, this underscores that LDL-centric thinking doesn’t fully capture the biology of plaque dynamics and clinical events.

Practical Takeaways for Clinicians and Patients

Shift the focus from LDL alone to disease burden and metabolic drivers:

• Use CAC (where appropriate) to reclassify risk; CAC = 0 is strongly reassuring across populations, including many with FH [9,10].
• Screen for insulin resistance early (don’t rely on fasting glucose alone): consider TG/HDL ratio, waist circumference, clinical features, and—when available—dynamic or composite measures [8,11].
• Address modifiable drivers beyond lipids: excess adiposity, hypertension, smoking, sleep/stress.
• When discussing statins, communicate absolute risk and absolute benefit alongside relative risk so patients can give informed consent [4,5].

“If people want to make sure they don’t develop coronary artery disease, focusing on cholesterol isn’t going to do it. You have to look at the other factors.” — Prof. Tim Noakes

For the Curious (and Courageous): Go Deeper

The Nutrition Network course builds systematically from history and epidemiology through pathophysiology, imaging, metabolic drivers, stress, toxins, and therapeutics—with contributions from experts who challenge orthodoxy with data and nuance.

Start here: Cardiovascular Health: A Metabolic Perspective

References

1. Keys A. Atherosclerosis: a problem in newer public health. J Mt Sinai Hosp N Y. 1953;20:118–139. Available from: https://www.mtsinai.org/ (contextual historical citation; original figure widely reproduced).

2. DuBroff R, de Lorgeril M. Fat or fiction: the diet-heart hypothesis. BMJ Evidence-Based Medicine [Internet]. 2019 May 29;26(1):bmjebm-2019-111180. Available from: https://ebm.bmj.com/content/early/2019/07/10/bmjebm-2019-111180
3. Libby P. The forgotten majority: unfinished business in cardiovascular risk reduction. J Am Coll Cardiol. 2005;46(7):1225–1228. Available from: https://www.jacc.org/doi/10.1016/j.jacc.2005.07.006
4. Byrne P, Cullinan J, Gillespie P, Perera R, et al. Effectiveness of statins for primary prevention: absolute vs relative risk. BMJ Open or comparable meta-analytic summaries reporting ARR <1% for mortality and stroke and ≈1% for MI. (Representative open-access overview:) Abramson J, Wright JM. Are lipid-lowering guidelines evidence-based? JAMA Intern Med. 2013;173(14):1313–1321. Available from: https://jamanetwork.com/journals/jamainternalmedicine/fullarticle/1709440
5. Diamond DM, Ravnskov U. How statistical deception created the appearance that statins are safe and effective. Expert Rev Clin Pharmacol. 2015;8(2):201–210. Available from: https://www.tandfonline.com/doi/full/10.1586/17512433.2015.1012494
6. Neil A, et al. Analysis from the Simon Broome Register: coronary mortality risk in heterozygous familial hypercholesterolemia declines with age. Eur Heart J. 2008;29(5):594–603. Available from: https://academic.oup.com/eurheartj/article/29/5/594/455668
7. Sijbrands EJ, et al. Mortality over two centuries in a large pedigree with familial hypercholesterolaemia. BMJ. 2001;322:1019–1023. Available from: https://www.bmj.com/content/322/7293/1019
8. Pérez de Isla L, Alonso R, Mata N, et al. Predicting cardiovascular events in familial hypercholesterolemia: the SAFEHEART registry. Circulation. 2017;135(22):2133–2144. Available from: https://www.ahajournals.org/doi/10.1161/CIRCULATIONAHA.116.024541
9. Mortensen MB, et al. Association of coronary plaque with LDL-C and event rates among symptomatic adults. JAMA Cardiol. 2017;2(7):772–780. Available from: https://jamanetwork.com/journals/jamacardiology/fullarticle/2629492
10. Miname MH, et al. Coronary artery calcium and cardiovascular events in familial hypercholesterolemia: long-term follow-up. Atherosclerosis. 2023;373:22–29. Available from: https://www.sciencedirect.com/science/article/pii/S0021915023001673
    
11. Dugani SB, et al. Association of lipoproteins, insulin resistance, and metabolic syndrome with coronary risk in women: Women’s Health Study. Circulation. 2021;143:1509–1519. Available from: https://www.ahajournals.org/doi/10.1161/CIRCULATIONAHA.120.050756
12. Ridker PM, et al. Inflammation (hsCRP) vs LDL-C for risk prediction in contemporary statin-naïve/intolerant cohorts; see: Ridker PM, Danielson E, Fonseca FA, et al. Rosuvastatin to prevent events in men and women with elevated CRP (JUPITER). N Engl J Med. 2008;359:2195–2207. Available from: https://www.nejm.org/doi/full/10.1056/NEJMoa0807646
13. Kendrick M. The Clot Thickens: mechanistic arguments for thrombogenesis and impaired repair as central to atherosclerosis. (Synopsis) Available from: https://drmalcolmkendrick.org/ and book publisher pages.

Puri R, et al. Impact of statins on coronary calcification (representative data showing increased CAC with therapy); see also Budoff MJ, reviews on CAC dynamics with statins. J Am Coll Cardiol Imaging and related. Overview: Puri R, et al. JACC. 2015;65(13):1273–1282. Available from: https://www.jacc.org/doi/full/10.1016/j.jacc.2015.01.036