The case was a familiar one: a man in his early forties, LDL above 5, father dead at 53, grandfather dead at 48. He’d already had one event and was back within twelve months with a left main lesion. Nothing about him was unusual — which is exactly the point. Most weeks in general practice contain a version of this patient, usually before the first event rather than after it.
What follows are my notes from the lipid session, reorganised around the questions that actually come up in the consulting room.
The target keeps moving, and it moves in one direction
Thirty years ago the LDL goal for everyone was under 3.4 mmol/L. Twenty years ago it was under 2.6, with 1.8 for high-risk groups. The 2025 European Society of Cardiology update sets under 1.4 for anyone who has had an event, and under 1.0 for the highest-risk patients — those with recurrent events. The 2023 Australian guidelines land in much the same place.
There’s a caveat worth remembering, because it catches people out. The target isn’t only an absolute number; it’s also a greater than 50% reduction from baseline. A patient whose index LDL was 2.0 and who now sits at 1.6 has technically crept under nobody’s line — they need to be under 1.0, not settling at 1.6 because it “looks close enough”.
What is a biologically normal cholesterol?
This is the question that reframes everything, and it has the same shape as the equivalent question about BMI or blood pressure: what the population does and what the species is built for have drifted apart.
Contemporary hunter-gatherer and subsistence populations — the Hadza, the Inuit, sub-Saharan groups studied across decades — have total cholesterols lower than our average LDL. So does essentially every other mammal on earth. The average Australian total cholesterol sits around 5.4; the populations above correspond to LDLs under 1.8.
Modern imaging says something consistent. Put patients on lipid-lowering therapy and image the vessel directly, and plaque regression begins to appear around 1.8 mmol/L — but it becomes the norm below 1.4. That was the finding of the GLAGOV study using intravascular ultrasound during angiography to measure plaque volume, with substantial Australian enrolment. The guideline number isn’t arbitrary; it’s roughly where the plaque starts going backwards.
The primary prevention data is older than most of our registrars
WOSCOPS, the Scandinavian and ASCOT work, the Air Force trial — all 1990s vintage. All showing the same thing: get LDL under 1.8 and the probability of an event becomes very low. Population genetics agrees. People born with LDLs in the ones are unlikely to have a cardiovascular event in their lifetime.
Calcium scoring makes the point vividly in the age group we most often see. In patients aged 40–54, an LDL under 1.4 corresponded to essentially no positive calcium scores. As LDL climbs, the proportion with detectable plaque climbs with it — and that tracks directly onto mortality.
Pooling more than thirty years of statin trials, natural-history studies and genetic data gives you the same straight line every time, extrapolating cleanly toward zero: lower is better. The line I use with patients is that each 1 mmol/L reduction in LDL buys roughly a 20% relative reduction in cardiovascular events over five years and around a 10% reduction in all-cause mortality. There are not many things in medicine proven across such large datasets over such long horizons. Blood pressure is about the only comparable example.
None of which means LDL is the whole story — sleep, smoking, exercise, genetics and the rest all matter. But LDL is causative. It is the thing that builds the plaque.
Cholesterol-years
Humans are among very few mammals that routinely live past fifty. An elephant dies somewhere between fifty and seventy, usually of starvation when its teeth wear out — a dental problem, not a cardiovascular one.
That longevity is why the concept of cumulative exposure — cholesterol-years — matters so much:
| Genotype | Typical LDL | Age at first event |
|---|---|---|
| Homozygous FH | >10 | 20s–30s |
| Heterozygous FH | >5 | 40s–50s |
| Average LDL | ~3 | 70s |
| Genetically low LDL | ~1 | Often never |
Heterozygous familial hypercholesterolaemia affects around 1 in 250 people. It persists in the population precisely because it doesn’t interfere with reproduction — it kills between 50 and 100. A faulty LDL receptor, inherited, entirely modifiable.
Which is the conversation to have with the man above. His father died at 53. His grandfather at 48. The Australian risk calculator put his five-year risk at 3–5%, and for a 43-year-old with kids at school that is already high — but the number that matters is the lifetime risk, which is close to 100%. Short-term risk low, lifetime risk near-certain, is a very common and very misleading combination.
The framing I keep coming back to: the goal is to work into your nineties and shake hands the week before you die. That’s what healthy ageing looks like. Not bypass surgery in your forties.
The prevention paradox
We are all good at identifying the extreme patients — the heavy smokers, the cholesterol of 9. But the majority of events occur in the low and intermediate risk groups, simply because that is where most people are. Plaque starts depositing in our thirties; fatty streaks were being found in the aortas of young soldiers at autopsy during the Korean War.
Once someone crosses the threshold into an atherosclerotic event, the prognosis is permanently altered. Around 20% of MI survivors are back within a year, and the recurrent event is more likely to be fatal. Australian data over the last decade puts twelve-month mortality after a non-STEMI at about 5%.
And yet roughly half of these patients never reach target. Worth noting that the PBS threshold of 1.8 is an economic criterion, not a clinical one — best practice is 1.4, and lower again for the most comorbid patients.
Lifestyle: right answer, wrong lever
A strict low-fat diet lowers LDL by 10–20%. For a patient sitting at 5.5, that’s a move to about 4.5. The magnitude of change we actually need is from 4 down into the ones, and no diet achieves that.
This is not an argument against lifestyle — it’s an argument about which outcome you’re buying with it. Diet and exercise produce plaque remodelling and anti-inflammatory effects that sit above and beyond LDL lowering:
- Moderate-intensity exercise, 300 minutes a week — roughly halves risk over five years. Moderate means breathless enough that you couldn’t hold this conversation comfortably.
- High-intensity, 75 minutes a week — single words only, or none.
- Mediterranean diet — a 30–50% risk reduction over five years.
So: lifestyle for the plaque biology, drugs for the number.
The pharmacological ladder
| Agent | Additional LDL reduction |
|---|---|
| Moderate–high intensity statin | ~50% |
| Ezetimibe (added) | ~25% |
| Evolocumab (PCSK9 mAb) | >60% |
| Inclisiran (siRNA) | ~50% |
The mechanisms all converge on the same receptor. Statins reduce hepatic cholesterol synthesis, so the liver expresses more LDL receptors. Ezetimibe interrupts enterohepatic circulation, with the same downstream effect. PCSK9 has no function other than causing lysosomal degradation of the LDL receptor — remove it and receptor expression rises. Evolocumab is a monoclonal antibody against the molecule; inclisiran is an interfering RNA that switches off its production.
For a patient starting at 5.5, the arithmetic is unforgiving: high-intensity statin gets you to roughly 2.8, adding ezetimibe to about 1.9 — still enough for plaque progression. Reaching 0.8 needs all three tiers.
Statin safety, and the nocebo problem
A recent Lancet analysis reviewed the reported adverse effects of statins against the half-million patients enrolled in major randomised trials, and identified four genuine side effects. Brain fog and most of the rest of the internet’s catalogue did not survive contact with the data — the paper argues for stripping much of it out of product information.
What’s real:
- Diabetes — around 1 in 1000, related to statin dose, not to LDL lowering
- Muscle symptoms — around 5%, against a similar placebo rate
- Myopathy or rhabdomyolysis — well under 1 in 10,000
- LFT derangement — around 1 in 1000, highest with atorvastatin, usually solved by substitution
The SAMSON trial is the one I’d most like patients to know about. Statin-intolerant patients received, in blinded rotation, one month of statin, one month of placebo, and one month of nothing, logging symptoms daily by phone across a year. Patients on placebo reported the same burden of unwellness as patients on statin. The patterns are fascinating: some people feel unwell every day regardless of what they’re taking; some rechallenge successfully and discover they were never intolerant; some feel dreadful on placebo, recover when it stops, feel identically dreadful on statin.
How I counsel it up front: 5% of people get muscle aches on a statin, and 5% get them on placebo, and you can’t tell them apart at the individual level. But a systemic drug doesn’t cause left elbow pain or a sore finger. What we’re looking for is the hit-by-a-bus feeling in the big muscle groups — shoulders, quads. And then I hand it over: stop it and see. If you improve, it might be the statin. Restart at a lower dose or a different brand and see what happens. Try it at night. Patients experimenting on themselves with permission tend to end up on treatment more often than patients who feel unheard.
It takes an enormous amount of energy to refute a claim that took none to produce — we’ve watched it with vaccines, with paracetamol, and we watch it weekly with statins. It’s exhausting. But bludgeoning a patient with data rarely helps them; being the person who takes the symptom seriously usually does.
Where PCSK9 inhibitors sit now
FOURIER established evolocumab in secondary prevention: a 15% relative risk reduction, and with longer follow-up, an absolute reduction in cardiovascular death of about 1.1%. The relative effect is smaller than the per-mmol rule of thumb would predict, because these patients have already crossed the threshold — stent, thrombosis, inflammation, residual risk that LDL lowering can’t undo.
On cognition, the EBBINGHAUS substudy of FOURIER put nearly 2000 patients through detailed cognitive testing over 19 months, with a subset followed for years; mean achieved LDL was 0.9, and there was no cognitive signal — including in those between 0 and 0.5. (For the “you need LDL for your brain” argument: LDL is around 1 at birth, when the brain is doubling in size, and from about age 40 the brain shrinks regardless.)
The newer development is VESALIUS-CV, presented at AHA 2025 and published in the NEJM — the first PCSK9 outcomes trial in a population without prior MI or stroke. 12,257 patients with atherosclerosis or high-risk diabetes, all on optimised lipid-lowering therapy, randomised to evolocumab or placebo and followed a median of 4.6 years. The primary composite of coronary death, MI or ischaemic stroke fell by 25%. The longer follow-up compared with FOURIER is the likely reason a mortality signal emerged here where it hadn’t before.
Practically, in Australia: evolocumab is a fortnightly self-administered pen, about fifteen seconds, refrigerated but stable at room temperature for up to 30 days, available under streamlined authority. The PBS criteria require LDL above 1.8 with a prior cardiovascular event and specialist involvement — a specialist letter stating the patient is appropriate for a PCSK9 inhibitor should satisfy that. Self-funded, it runs around $3,500 a year, which some patients will happily pay, particularly those who would rather inject fortnightly than swallow something daily.
What I’d actually do on Monday
- Start a low-dose statin, build trust, uptitrate. Co-prescribe ezetimibe early — cheap, generic, stays in the gut, minimal side effects, and it turbocharges the statin.
- Recheck lipids at four to six weeks. Statins have achieved their effect well before then. The PBS clock says three months, but earlier bloods keep patients engaged and let you titrate.
- Build the follow-up scaffolding deliberately. The classic failure mode is the patient who sees you without bloods, gets bloods and never returns, sees the cardiologist where nothing is repeated, and drifts. A phone call at four weeks fixes most of this.
- If LDL remains above 1.8 with a prior event, they need a PCSK9 inhibitor. Say so in writing.
- Remember adherence is the quiet variable: only about 60% of patients take a prescribed tablet 80% of the time, and patients are more reliable with antihypertensives than with statins. A drug not taken has no effect size.
The patients who most need this conversation are the ones sitting in front of us in their forties with a family history and a reassuringly low five-year risk score. The writing is often on the wall well before the first event, and unlike almost anything else in medicine, this one is genuinely modifiable.
Source: notes from a cardiology education meeting. The presenting cardiologist disclosed that the meeting was industry-sponsored, and PCSK9 inhibitor therapy featured prominently. Figures quoted are as presented, cross-checked against published trial data where possible. Educational content for clinicians — not a substitute for current guidelines or individual clinical judgement.