How low should cholesterol be?

The evidence supports lower being better, with no identified threshold below which further reduction stops helping. Targets are therefore set by risk rather than by any notion of a normal value. European guidelines suggest LDL below 3.0 mmol/L for low-risk individuals, below 2.6 for moderate risk, below 1.8 with at least a 50% reduction for high risk, and below 1.4 for very high risk — with a stricter target of below 1.0 proposed for those who have a second event within two years. Very low LDL levels achieved in trials have not been associated with harm, including no convincing signal for cancer, cognitive impairment or haemorrhagic stroke at the levels studied. The other half of the equation is duration: starting earlier produces much larger lifetime benefit than the same reduction started late.

Are statin side effects real?

Muscle symptoms are genuinely experienced, but in the great majority of cases they are not caused by the statin. This was resolved by n-of-1 randomised trials in patients who had already stopped statins because of side effects: SAMSON alternated statin, identical placebo and empty months, and found that around ninety per cent of the symptom burden occurred equally during placebo months. StatinWISE reached the same conclusion. This is the nocebo effect — expectation of harm producing real symptoms — and it is not a suggestion that patients are inventing anything. The practical consequence is that most people who have stopped a statin can be successfully restarted, using a structured rechallenge, a lower dose, alternate-day dosing or a different statin. Genuine statin myopathy with a raised creatine kinase does occur but is rare, and rhabdomyolysis is very rare.

What is lipoprotein(a) and should I have it tested?

Lipoprotein(a) is an LDL-like particle carrying an additional protein that makes it both atherogenic and prothrombotic. Its level is determined almost entirely by genetics, remains essentially constant throughout life, and is not meaningfully changed by diet, exercise or statins. Around one in five people worldwide have an elevated level, which independently increases the risk of heart attack, stroke and calcific aortic valve disease. Because a single measurement is sufficient for life, guidelines increasingly recommend testing at least once in every adult — yet it is almost never done, leaving a large group of people whose risk is entirely invisible on standard lipid panels. There is currently no approved drug that specifically lowers it, so management means treating every other risk factor more aggressively; several targeted therapies are in phase 3 outcome trials.

What is familial hypercholesterolaemia?

Familial hypercholesterolaemia is an inherited condition, usually caused by a fault in the LDL receptor gene, that causes very high LDL cholesterol from birth. It affects around one in 250 people — much commoner than most people assume — and because the arteries are exposed to high cholesterol from childhood rather than from middle age, untreated risk of early coronary disease is increased roughly twenty-fold. Fewer than one in ten affected people are diagnosed in most countries. Clues include an LDL above 5 mmol/L, thickened tendons at the ankle or knuckles (tendon xanthomata, which are effectively diagnostic), a corneal arcus before the age of 45, and heart attacks in relatives at a young age. Because it is dominantly inherited, each first-degree relative has a fifty per cent chance of being affected, which makes family cascade testing one of the most cost-effective screening programmes in all of medicine.

What if statins are not enough or not tolerated?

Several proven options exist, and they can be combined. Ezetimibe blocks cholesterol absorption in the gut, adds around fifteen to twenty per cent further LDL reduction, and reduced events in the IMPROVE-IT trial; it is cheap, well tolerated and the logical first addition. PCSK9 inhibitors given by injection every two to four weeks lower LDL by a further fifty to sixty per cent and reduced events in large outcome trials. Inclisiran works on the same pathway using small interfering RNA but is given only twice a year after two loading doses, which removes daily adherence from the equation entirely. Bempedoic acid acts one step before the enzyme statins block and is not activated in muscle tissue, so it is specifically useful for people with genuine statin intolerance — it reduced cardiovascular events in the CLEAR Outcomes trial in exactly that population.

How does GMJ News cover cholesterol and lipid management?

GMJ News covers the causal evidence for LDL and the lower-for-longer principle, statin intensity and drug interactions, the SAMSON and StatinWISE nocebo evidence and structured rechallenge, lipoprotein(a) biology and once-in-a-lifetime measurement, familial hypercholesterolaemia and cascade testing, ezetimibe, PCSK9 inhibitors, inclisiran and bempedoic acid, ApoB and non-HDL cholesterol as superior risk markers, hypertriglyceridaemia and pancreatitis risk, cardiovascular risk calculators and their limitations in South Asian and inflammatory disease populations, and the instructive failure of the HDL-raising hypothesis. Relevant to PHIG's cardiovascular prevention and NCD programmes.

Home › Topics › Hyperlipidaemia and Cholesterol

Hyperlipidaemia and Cholesterol

GMJ News knowledge hub · last reviewed September 2026 · Georgian Medical Journal

Summary Statistics Glossary GMJ news FAQ Documents Organizations Research

Elevated low-density lipoprotein cholesterol is one of the most firmly established causal risk factors in all of medicine, supported by randomised trials, Mendelian randomisation and dose-response consistency across more than 200,000 trial participants, and the governing principle is now expressed simply as “lower for longer is better” — both the magnitude and the duration of LDL reduction determine cardiovascular benefit, with no threshold below which lowering ceases to help (WHO). Two clinical problems dominate practice: statin intolerance is very largely a nocebo phenomenon — the SAMSON and StatinWISE n-of-1 randomised trials showed that the great majority of muscle symptoms attributed to statins occur equally on placebo, meaning most patients who stopped can be successfully rechallenged; and lipoprotein(a) is an independent, genetically determined and largely treatment-resistant risk factor that should be measured once in every adult's lifetime, yet remains almost universally untested despite affecting around one in five people worldwide.

Key messages

Lower for longer is better — no threshold below which lowering stops helping

LDL cholesterol is causally related to atherosclerotic cardiovascular disease, established through randomised trials, Mendelian randomisation and consistent dose-response across more than 200,000 trial participants. The Cholesterol Treatment Trialists meta-analyses show that each 1 mmol/L reduction in LDL produces roughly a 22% proportional reduction in major vascular events per year of treatment, and the benefit continues at very low achieved levels with no identified threshold. Both magnitude and DURATION matter — which is why lifelong exposure from genetic variants produces far larger effects than the same reduction achieved late in life, and why early treatment in high-risk patients is disproportionately valuable.

Statin intolerance is very largely a nocebo effect — most patients can be rechallenged

Muscle symptoms are reported by 5-20% of statin users in observational studies, yet in blinded randomised trials the excess over placebo is minimal. The n-of-1 trials settled the question: SAMSON (NEJM 2020) alternated statin, placebo and no-tablet months in patients who had stopped statins for side effects and found that around 90% of the symptom burden occurred equally on placebo; StatinWISE reached the same conclusion. The clinical implication is not that patients are imagining symptoms — they are real — but that they are not caused by the drug in most cases, and structured rechallenge, dose reduction, alternate-day dosing or switching agent successfully returns the majority to therapy.

Lipoprotein(a) — measure once in every adult, and almost nobody does

Lp(a) is an LDL-like particle with apolipoprotein(a) attached, whose concentration is 80-90% genetically determined, essentially unaffected by diet, exercise or statins, and stable throughout life. Elevated Lp(a) affects roughly one in five people worldwide and independently increases risk of myocardial infarction, stroke and calcific aortic stenosis. Guidelines increasingly recommend measuring it at least ONCE in every adult's lifetime, since a single measurement suffices and identifies a large group whose risk is otherwise invisible. Management is currently intensification of all other modifiable risk factors, since no approved therapy specifically lowers Lp(a) — though several antisense and siRNA agents are in phase 3 outcome trials.

Familial hypercholesterolaemia — common, treatable, and massively underdiagnosed

Heterozygous FH affects approximately 1 in 250 people — far commoner than generally assumed — and causes lifelong LDL elevation with untreated risk of premature coronary disease increased around 20-fold. Fewer than 10% of cases are diagnosed in most countries. Clues: LDL above 5 mmol/L in adults or 4 mmol/L in children; tendon xanthomata (pathognomonic); corneal arcus before 45; and premature cardiovascular disease in first-degree relatives. Cascade testing of relatives is the single most cost-effective genetic screening programme in medicine, since each index case identifies several affected relatives who are otherwise asymptomatic until their first event.

Beyond statins — ezetimibe, PCSK9 inhibitors, inclisiran and bempedoic acid

When statins are insufficient or not tolerated, a sequence of proven options exists. Ezetimibe blocks intestinal cholesterol absorption, adds around 15-20% LDL reduction, and reduced events in IMPROVE-IT. PCSK9 monoclonal antibodies (evolocumab, alirocumab) lower LDL by around 50-60% on top of statin and reduced events in FOURIER and ODYSSEY OUTCOMES. Inclisiran is a small interfering RNA given twice yearly after loading — a dosing schedule that fundamentally changes adherence, since it is administered in clinic rather than depending on daily tablets. Bempedoic acid, which acts upstream of HMG-CoA reductase and is not activated in muscle, reduced events in CLEAR Outcomes specifically in statin-intolerant patients.

ApoB and non-HDL cholesterol are better risk markers than LDL alone

LDL cholesterol measures the CHOLESTEROL CONTENT of particles, whereas apolipoprotein B counts the NUMBER of atherogenic particles — one ApoB molecule per LDL, VLDL, IDL and Lp(a) particle. Since atherosclerosis is driven by particle number entering the arterial wall, ApoB is a superior risk marker, and this discordance matters most in exactly the patients where risk is most often underestimated: those with diabetes, metabolic syndrome, obesity or hypertriglyceridaemia, who characteristically have small dense LDL particles and therefore many particles carrying little cholesterol each. Non-HDL cholesterol (total minus HDL) is a close and free approximation, requires no fasting, and should be used routinely where ApoB is unavailable.

Key statistics

~22%

proportional reduction in major vascular events per 1 mmol/L LDL reduction per year

CTT/Lancet

SAMSON

around 90% of statin-attributed symptoms occurred equally on placebo in n-of-1 trials

NEJM 2020

1 in 5

people have elevated lipoprotein(a) — measure once in a lifetime; almost never tested

EAS/ESC

1 in 250

prevalence of heterozygous familial hypercholesterolaemia; under 10% diagnosed

EAS/WHO

50-60%

additional LDL reduction from PCSK9 inhibitors on top of maximal statin therapy

FOURIER/ODYSSEY

Twice yearly

inclisiran dosing after loading — changes adherence from a daily to a clinic-administered decision

ORION/NEJM

Lipid-lowering therapies — approximate LDL cholesterol reduction

Source: ESC/EAS. Reductions are approximate and additive when agents are combined.

Glossary of key terms

Statin intensity classification
PharmacologyHigh-intensity (LDL reduction 50% or more): atorvastatin 40-80mg, rosuvastatin 20-40mg. Moderate-intensity (30-49%): atorvastatin 10-20mg, rosuvastatin 5-10mg, simvastatin 20-40mg, pravastatin 40mg. Low-intensity (under 30%): simvastatin 10mg, pravastatin 10-20mg. Doubling the dose of any statin adds only around 6% further LDL reduction — the rule of six — which is why adding ezetimibe (a further 15-20%) is more effective than uptitrating a statin already at moderate dose. Simvastatin 80mg is no longer recommended because of myopathy risk. Interactions matter: simvastatin and lovastatin are heavily CYP3A4-dependent and should be avoided with clarithromycin, azole antifungals, ciclosporin, verapamil and diltiazem; rosuvastatin and pravastatin are less affected.

Lipoprotein(a) biology
LipidologyLp(a) consists of an LDL particle covalently bound to apolipoprotein(a), a protein with structural homology to plasminogen. It is atherogenic through its LDL component, prothrombotic through plasminogen-mimicry interfering with fibrinolysis, and pro-inflammatory through oxidised phospholipid carriage. Concentration is determined almost entirely by the LPA gene, particularly the number of kringle IV type 2 repeats, and varies substantially between ancestral groups — being on average higher in people of African ancestry. It is not lowered meaningfully by lifestyle or statins (which may slightly raise it), reduced around 20-25% by PCSK9 inhibitors, and around 30% by niacin without outcome benefit. Apheresis is used in extreme cases. Antisense oligonucleotide and siRNA therapies achieving 80-90% reduction are in cardiovascular outcome trials.

Cardiovascular risk assessment tools
PreventionRisk calculators estimate absolute event probability to guide treatment intensity: SCORE2 and SCORE2-OP in Europe (calibrated to four regional risk levels), the Pooled Cohort Equations and PREVENT in the United States, QRISK3 in the UK, and WHO/ISH charts for settings without local cohort data. All share limitations: they underestimate risk in South Asian populations and in those with chronic inflammatory disease, HIV, severe mental illness or chronic kidney disease; they do not incorporate Lp(a) or family history adequately in most versions; and they perform poorly at the extremes of age. Risk enhancers and coronary artery calcium scoring are used to reclassify patients near a treatment threshold — a zero calcium score is a powerful negative predictor allowing statin deferral in selected intermediate-risk individuals, while any calcium in a younger patient argues strongly for treatment.

Hypertriglyceridaemia
LipidologyTriglyceride elevation has two distinct clinical concerns. Moderate elevation (2-10 mmol/L) is a marker of remnant cholesterol and insulin resistance, associated with cardiovascular risk — managed by treating the cause (obesity, alcohol, poorly controlled diabetes, hypothyroidism, drugs) and by statin therapy directed at ApoB. Severe elevation (above approximately 10-11 mmol/L) becomes a pancreatitis risk requiring urgent reduction with fibrates, dietary fat restriction, alcohol cessation and glycaemic control. On specific therapies: fibrates reduce triglycerides substantially but have not consistently reduced cardiovascular events, and the PROMINENT trial of pemafibrate was neutral; icosapent ethyl reduced events in REDUCE-IT, though debate persists over the mineral oil comparator; and gemfibrozil should not be combined with statins because of rhabdomyolysis risk.

Cascade testing in familial hypercholesterolaemia
Genetics/Public healthFH is autosomal dominant, so each first-degree relative of an index case has a 50% chance of being affected — making systematic cascade testing extraordinarily efficient. Genetic testing identifies a causative variant in LDLR, APOB or PCSK9 in around 60-80% of clinically definite cases, and a negative genetic test does not exclude the clinical diagnosis. Screening approaches: cascade testing from index cases identified in lipid clinics; universal childhood lipid screening at 1-2 years or 9-11 years, followed by reverse cascade testing to parents, which has the advantage of identifying affected adults before their first event; and opportunistic case-finding from very high LDL results in laboratory databases. Children with FH benefit from statin initiation from around age 8-10, with long-term follow-up showing normalised cardiovascular outcomes into adulthood.

HDL cholesterol — the failure of a hypothesis
Evidence/HistoryLow HDL is robustly associated with cardiovascular risk observationally, which generated decades of effort to raise it therapeutically. Every approach has failed to improve outcomes: niacin (AIM-HIGH, HPS2-THRIVE, which showed harm), fibrates in unselected populations, and the CETP inhibitors — torcetrapib caused excess mortality, dalcetrapib and evacetrapib were neutral, and anacetrapib's modest benefit was attributed to LDL rather than HDL effects. Mendelian randomisation studies show that genetic variants raising HDL do not reduce cardiovascular risk, indicating that low HDL is a marker rather than a cause. This is one of the clearest cautionary examples in cardiovascular medicine of the difference between association and causation, and of why surrogate endpoints require outcome trials before they change practice.