Cardiovascular Disease

Reaven et al.'s 1993 controlled feeding study in the American Journal of Clinical Nutrition assigned participants to either a high-linoleic-acid diet (rich in soybean and safflower oil) or a high-oleic-acid diet (rich in olive oil) for five weeks each. LDL particles from the high-linoleic group showed 3.5 times greater susceptibility to copper-mediated oxidation compared with the high-oleic group. The linoleic acid content of LDL particles directly predicted their oxidative vulnerability.

This is because linoleic acid, the primary fatty acid in soybean, sunflower, corn, cottonseed and canola oils, contains two double bonds that are highly susceptible to free radical attack. When incorporated into LDL particles, linoleic acid makes the particle structurally more vulnerable to oxidation. Oxidised LDL is now recognised as the atherogenic particle: it is oxidised LDL, not native LDL, that is taken up by macrophage scavenger receptors to form the foam cells that constitute atherosclerotic plaque. Steinberg et al. (1989) established this oxidative modification hypothesis in the New England Journal of Medicine.

UK dietary guidelines recommend replacing saturated fat with unsaturated fat to reduce cardiovascular risk. This recommendation does not distinguish between monounsaturated fats (olive oil, animal fat) and polyunsaturated omega-6 fats (seed oils). The Reaven data show that the specific type of unsaturated fat matters: replacing saturated fat with omega-6 linoleic acid increases LDL oxidative susceptibility, the very mechanism now understood to drive atherosclerosis. UK seed oil consumption increased approximately tenfold between 1960 and 2000.

In 1953, Ancel Keys published a graph in the Journal of Mount Sinai Hospital plotting dietary fat consumption against coronary heart disease mortality across six countries. The correlation appeared striking. Keys had data from twenty two countries but selected six. When Yerushalmy and Hilleboe published a reanalysis in 1957 using all twenty two countries, the correlation disappeared. The published six had been chosen to fit the hypothesis.

Keys became one of the most influential nutrition scientists of the twentieth century. He sat on the nutrition committee of the American Heart Association and shaped the AHA's 1961 guidance recommending reduction of saturated fat. The guidance reached clinical practice globally within a decade. The Seven Countries Study, which Keys led from 1958, has since been criticised for country selection bias and inconsistent methodology. The dietary fat hypothesis it produced became embedded in public health guidance for fifty years before the PURE study (2017) and subsequent meta-analyses substantially revised the consensus.

Coronary heart disease mortality in England and Wales fell by seventy two per cent between 1981 and 2013. A 2016 analysis by Bhatnagar and colleagues in the British Medical Journal used the IMPACT model to attribute the decline. Approximately fifty eight per cent was attributable to falls in major risk factors, particularly reductions in smoking prevalence, blood pressure and total cholesterol. Approximately forty two per cent was attributable to medical and surgical treatments, including statins, antihypertensives, acute coronary interventions and cardiac rehabilitation.

The finding matters for two reasons. First, the largest component of the decline, namely changes in population risk factors, preceded widespread statin use. Smoking rates began falling in the 1970s; CHD mortality was already declining before statins reached large patient populations in the 1990s. Second, the IMPACT model attributes only around twenty per cent of the total decline to lipid-lowering treatments, of which statins are one component. The framing of statins as the primary driver of reduced heart disease mortality is not supported by the attribution data.

From 1961 to 2015, US and UK dietary guidelines advised limiting dietary cholesterol to 300mg per day. An egg contains approximately 186mg. The restriction discouraged consumption of eggs, liver and shellfish, the most nutrient-dense foods available, for fifty four years. The 2015 US Dietary Guidelines Advisory Committee removed the cap, stating that dietary cholesterol is "not a nutrient of concern for overconsumption" and that evidence for the 300mg limit was insufficient.

The biological basis for removing the restriction had been established decades earlier. Fernandez (2010) reviewed the evidence and found that dietary cholesterol has minimal impact on blood cholesterol in approximately two thirds of the population, classified as hypo-responders. In the remaining third, both LDL and HDL rise proportionally, leaving the LDL:HDL ratio unchanged. Soliman (2018) confirmed that the original studies conflating dietary and blood cholesterol failed to distinguish oxidised cholesterol in processed foods from the intact cholesterol in whole foods.

In July 2014 NICE published Clinical Guideline 181, reducing the ten-year cardiovascular risk threshold for statin prescribing in primary prevention from twenty per cent to ten per cent. A Robson and colleagues analysis published in the British Medical Journal estimated this would make eleven point eight million previously ineligible UK adults eligible for statin therapy, more than doubling the potential treatment population at a stroke of guideline revision.

The guideline committee included members with declared interests in pharmaceutical companies manufacturing statins, including Pfizer, AstraZeneca and Merck. NICE declared these interests manageable rather than conflicting. The ten-year risk calculators used to determine eligibility, including QRISK2, have been criticised for overestimating risk in primary prevention populations, meaning individuals at genuinely low risk are classified as eligible. The Cochrane Collaboration's review of primary prevention statins (Taylor 2013) found no all-cause mortality reduction in populations without pre-existing cardiovascular disease.

Naci and Ioannidis's 2013 meta-analysis in the BMJ combined data from 305 randomised controlled trials encompassing 339,274 participants and compared exercise interventions with drug interventions for all-cause mortality across four conditions: coronary heart disease, stroke, heart failure and prediabetes. For coronary heart disease secondary prevention, exercise and statins produced statistically indistinguishable mortality benefits. For stroke rehabilitation, exercise was more effective than drug therapy. For heart failure, diuretics outperformed exercise. For prediabetes, exercise outperformed drugs.

This is because exercise modulates cardiovascular risk through multiple simultaneous mechanisms: it improves endothelial function, reduces arterial stiffness, lowers resting heart rate, increases high-density lipoprotein, reduces visceral adiposity, improves insulin sensitivity and reduces systemic inflammation. Statins primarily lower LDL cholesterol via HMG-CoA reductase inhibition, with a modest anti-inflammatory effect. The pleiotropic benefits of exercise address more upstream risk factors than any single pharmaceutical agent.

NICE guideline CG181 recommends statin therapy for primary prevention in adults with a ten-year cardiovascular risk of ten per cent or greater, a threshold that includes approximately twelve million UK adults. The same guideline recommends lifestyle modification but provides no framework for exercise prescription, no exercise dose guidance and no mechanism for monitoring exercise compliance. In 2023, NHS England spent approximately £620 million on statin prescriptions and £7 million on exercise referral schemes.

The Cholesterol Treatment Trialists’ Collaboration (2010) meta-analysed 26 randomised trials comprising 170,000 participants and reported that statins reduce major vascular events by 22 per cent per 1.0 mmol/L LDL reduction. In absolute terms, for primary prevention patients with no prior cardiovascular event, the number needed to treat for five years is 104 to prevent one non-fatal heart attack, and 154 to prevent one stroke.

This is because relative risk reduction (22 per cent) sounds impressive, but the baseline risk in healthy individuals is low. A five-year absolute risk reduction of approximately 1 per cent means that for every 100 people taking statins daily for five years, 99 receive no cardiovascular benefit but are exposed to the drug’s adverse effects: myopathy in 5 to 10 per cent, new-onset diabetes in 9 per cent (Sattar et al. 2010), cognitive impairment and liver enzyme elevation.

Despite this marginal benefit in primary prevention, UK guidelines recommend statins for anyone with a ten-year cardiovascular risk above 10 per cent, which includes most men over sixty. The global statin market generates over $15 billion annually. The drugs are effective for secondary prevention (post-heart attack), but the mass prescription to healthy populations is a pharmaceutical success story that has little to do with preventing death and everything to do with creating permanent customers.

In 1989 Daniel Steinberg and colleagues published a landmark review in the New England Journal of Medicine demonstrating that it is oxidised LDL, not LDL per se, that drives atherosclerosis. Macrophages in the arterial wall express scavenger receptors that bind oxidised LDL but not native LDL. Once engulfed, the macrophage cannot expel the cholesterol and becomes a foam cell. Foam cell accumulation is the structural foundation of atherosclerotic plaque. Witztum and Steinberg confirmed the mechanism in 1991 with further evidence of immune response to oxidised LDL epitopes.

The clinical implication is significant. Interventions that lower LDL without reducing oxidative stress may not reduce plaque progression in proportion to the LDL reduction. The substrates most susceptible to oxidation are polyunsaturated fatty acids, including the linoleic acid that constitutes up to fifty per cent of seed oil fatty acid composition. Reaven and colleagues (1993) demonstrated a correlation coefficient of 0.89 between linoleic acid content of LDL phospholipids and LDL oxidation susceptibility.

Kearns et al.'s 2016 investigation in JAMA Internal Medicine uncovered internal Sugar Research Foundation (SRF) documents showing that the trade group paid three Harvard scientists (including D. Mark Hegsted, who later became head of USDA nutrition policy) approximately $50,000 in today's money to publish a 1967 review in the New England Journal of Medicine. The review selectively evaluated the evidence and concluded that dietary fat, not sugar, was the primary dietary cause of coronary heart disease. The SRF's involvement was not disclosed.

This is because the sugar industry faced an existential commercial threat from Yudkin's 1964 Lancet paper linking sugar consumption to heart disease. The SRF's internal documents, obtained from university archives, show a deliberate strategy to "ichieve publication of our research results in a major medical journal" (sic) to shift the scientific consensus. Hegsted's 1967 NEJM review is cited in the genealogy of every subsequent low-fat dietary guideline. The first US Dietary Goals in 1977 recommended reducing fat and cholesterol but placed no limit on sugar.

The 1967 review shaped the dietary guidance that would reduce fat intake and increase carbohydrate consumption across the developed world for fifty years. UK heart disease mortality peaked in 1979 and has fallen since, but type 2 diabetes, obesity, NAFLD and metabolic syndrome have all risen dramatically during the same low-fat era. The sugar industry's intervention was not revealed until 2016. By that time, the low-fat consensus had been the foundation of NHS nutritional guidance for nearly four decades.

Kearns et al. (2016, JAMA Internal Medicine) discovered internal documents from the Sugar Research Foundation (now the Sugar Association) showing that in 1967, the industry paid three Harvard scientists the equivalent of $50,000 (adjusted for inflation) to publish a review in the New England Journal of Medicine that downplayed the role of sugar in coronary heart disease and shifted blame to dietary fat. The review, authored by Hegsted, McGandy and Stare, was published without disclosure of the industry funding and became foundational to subsequent dietary policy.

This is because the Sugar Research Foundation had commissioned the review specifically to counter emerging evidence linking sucrose consumption to coronary heart disease. The internal documents show that the foundation selected the studies to include, reviewed drafts of the manuscript and provided feedback before publication. The resulting paper concluded that there was "no doubt" that reducing dietary fat and cholesterol was the only intervention needed to prevent heart disease, while characterising the evidence against sugar as inconclusive.

The consequences of this single paper shaped 50 years of nutritional policy. The 1977 US Dietary Goals and 1980 Dietary Guidelines for Americans recommended reducing fat intake and made no restriction on sugar. Food manufacturers removed fat from products and replaced it with sugar to maintain palatability. Between 1970 and 2000, per capita sugar consumption in the US increased from 55kg to 68kg per year while fat consumption decreased. During this same period, heart disease remained the leading cause of death and obesity rates tripled. The fat hypothesis, built on an industry-funded foundation, directed public health policy away from the actual dietary cause of metabolic disease for two generations.

St-Pierre et al. (2005) followed 2,072 men for thirteen years and found that small dense LDL particle concentration was the strongest lipoprotein predictor of ischaemic heart disease, with a threefold higher risk in the top versus bottom tertile. Total LDL cholesterol was not independently predictive after accounting for particle size. Men with high total LDL but predominantly large buoyant particles had no elevated risk.

This is because LDL particles exist on a spectrum from large and buoyant (pattern A) to small and dense (pattern B). Small dense LDL penetrates the arterial wall more easily, is more susceptible to oxidation and has a longer plasma half-life. What drives the shift from pattern A to pattern B is not dietary fat intake but rather high carbohydrate consumption, insulin resistance and seed oil-driven inflammation. A diet high in animal fats tends to produce large buoyant LDL, while a diet high in refined carbohydrates and seed oils produces small dense LDL.

The total cholesterol number that drives statin prescriptions is a crude aggregate that conflates protective large LDL with atherogenic small LDL. A patient with total cholesterol of 6.5 mmol/L composed predominantly of large buoyant particles may have lower cardiovascular risk than a patient with 4.5 mmol/L of predominantly small dense particles. The standard lipid panel does not distinguish between the two.

The Homocysteine Studies Collaboration's 2002 meta-analysis in JAMA pooled thirty prospective studies with 5,073 coronary heart disease events. Each 5 µmol/L increment in plasma homocysteine was associated with a 1.70-fold increased risk of coronary heart disease (95% CI 1.50 to 1.93). For stroke, the corresponding risk was 1.65. These associations were independent of traditional risk factors including cholesterol, blood pressure and smoking status. The relationship was dose-dependent with no apparent threshold.

This is because homocysteine damages vascular endothelium through oxidative stress, promotes smooth muscle cell proliferation, impairs nitric oxide-mediated vasodilation and activates the coagulation cascade. Elevated homocysteine typically results from inadequate intake of folate, vitamin B6 and vitamin B12, which serve as cofactors in homocysteine metabolism. The MTHFR C677T polymorphism, carried by approximately 10 per cent of the UK population in homozygous form, impairs folate metabolism and produces chronically elevated homocysteine.

Standard NHS cardiovascular risk assessment uses the QRISK3 algorithm, which does not include homocysteine. A plasma homocysteine test costs approximately £15 and is available from every NHS hospital laboratory but is not included in routine health checks. The standard treatment is B-vitamin supplementation (folate, B6, B12), which costs approximately £2 per month and reliably reduces homocysteine by 25 to 30 per cent. The QRISK3 score determines statin eligibility for twelve million UK adults without measuring a modifiable risk factor with a thirty-study evidence base.