Reading a cholesterol result: LDL, HDL, triglycerides and the number nobody mentions

Cholesterol does not dissolve in blood. It travels inside lipoproteins — particles with a fatty core and a protein shell — and essentially everything interesting about a lipid panel concerns those particles rather than the cholesterol itself.

What the standard panel reports

Total cholesterol is the sum of the cholesterol carried in all particle types. Because it combines things that push in opposite directions, it is the least informative number on the page and the one patients most often quote.

LDL cholesterol is the cholesterol carried in low-density lipoproteins, the particles that deposit cholesterol in arterial walls. The causal role of LDL in atherosclerosis is about as well established as anything in cardiovascular medicine — supported by randomised trials of multiple drug classes with different mechanisms, and by Mendelian randomisation studies showing that people with genetically lower lifetime LDL have substantially less cardiovascular disease.

Note that most laboratories do not measure LDL directly; they calculate it, traditionally with the Friedewald equation, which becomes unreliable when triglycerides are high or LDL is very low.

HDL cholesterol is associated with lower risk observationally. But raising it pharmacologically has repeatedly failed to reduce events — the CETP inhibitor trials and niacin trials are the clearest examples — and Mendelian randomisation does not support a causal protective role. HDL is better understood as a marker of metabolic health than as a lever to pull. Very high HDL is, oddly, associated with increased risk in some cohorts.

Triglycerides rise with insulin resistance, alcohol, excess refined carbohydrate and some medications. They are a useful signal of metabolic dysfunction.

The number worth asking for

Every atherogenic particle — LDL, VLDL, IDL, lipoprotein(a) — carries exactly one molecule of apolipoprotein B. So measuring ApoB counts the particles rather than the cholesterol inside them.

That distinction matters because particles vary in how much cholesterol they carry. Two people with identical LDL cholesterol can have quite different particle counts, and it is the number of particles entering the artery wall that drives atherosclerosis. Where LDL cholesterol and ApoB disagree — which happens especially in people with high triglycerides, diabetes or metabolic syndrome — ApoB is the better predictor of risk.

Non-HDL cholesterol — total minus HDL — captures most of the same information, requires no extra test, and does not need fasting. It is calculated from every standard panel and reported by very few of them.

Lipoprotein(a)

Lp(a) is an LDL-like particle with an additional protein attached. Levels are around 90 per cent genetically determined, established in childhood, and essentially fixed for life. High Lp(a) is an independent causal risk factor for atherosclerotic disease and for aortic stenosis.

It is not part of a routine panel in most places, and it should probably be measured once in anyone being assessed for cardiovascular risk — particularly with a family history of early heart disease that the standard numbers do not explain. Specific therapies are in late-stage trials; for now, a high level is a reason to manage everything else more aggressively.

Do you need to fast?

Increasingly, no. Non-fasting lipid panels are now accepted in many guidelines, including in Europe and Canada. Triglycerides run somewhat higher after eating and LDL calculation is less reliable, but for risk assessment a non-fasting sample is adequate — and it is far more representative, since people spend most of the day in a fed state.

What the target should be

There is no single number, because targets depend on absolute risk. The principle running through modern guidance is that lower is better, and that the intensity of treatment should match the risk of the person.

Broadly, in secondary prevention or very high risk, European guidance targets LDL below 1.4 mmol/L (55 mg/dL). In high risk, below 1.8 mmol/L (70 mg/dL). In moderate risk, below 2.6 mmol/L (100 mg/dL). American guidance frames it more in terms of percentage reduction and risk-enhancing factors than absolute targets.

What matters more than the exact threshold is that the number is interpreted alongside your overall risk, not in isolation. An LDL of 3.5 mmol/L means something very different in a 55-year-old non-smoker with normal blood pressure than in a 70-year-old diabetic with prior stenting.

What actually moves the numbers

Diet. Replacing saturated fat with unsaturated fat lowers LDL — this holds up well in controlled feeding studies. Soluble fibre, particularly from oats, barley and psyllium, lowers it modestly. Plant sterols and stanols lower it by around 8 to 10 per cent. Dietary cholesterol itself has a smaller effect than was assumed for decades, which is why egg guidance relaxed.

Weight loss and exercise lower triglycerides substantially and raise HDL modestly. Their effect on LDL is smaller than people expect.

Statins reduce LDL by 30 to 55 per cent depending on agent and dose. Ezetimibe adds a further 15 to 20 per cent. PCSK9 inhibitors add 50 to 60 per cent on top of a statin, and are reserved for high-risk patients not reaching target. Bempedoic acid is an option for those genuinely intolerant of statins.

The reasonable posture

Get non-HDL from the panel you already have. Get Lp(a) once. If you have metabolic syndrome or diabetes, push for ApoB. And treat the number as one input into an estimate of absolute risk over ten years, rather than as a score to be improved for its own sake.