Cholesterol is a structural and signalling molecule used throughout the body. It is not evenly distributed: the brain contains a disproportionately large amount, much of it associated with myelin and cell membranes.

Because cholesterol cannot dissolve in blood, it travels inside lipoproteins such as LDL and HDL. LDL and HDL are therefore transport particles, not cholesterol itself.

And LDL is not one uniform particle. Size, density, residence time and damage matter. Small dense LDL is more strongly associated with arterial disease, while LDL that becomes oxidised — damaged by oxidative stress — can enter the inflammatory process that produces arterial plaque. Routine cholesterol tests do not directly measure this oxidised LDL burden.

This distinction becomes particularly relevant in nutritional ketosis.

A higher-carbohydrate, higher-insulin metabolism tends towards fuel storage. Insulin is an anabolic hormone: it promotes storage of excess carbohydrate as glycogen and fat. When glucose stores are full, the liver can convert excess glucose into fat.

Importantly, some of the fat produced from carbohydrate is saturated fat. Human body fat itself contains a substantial saturated-fat component.

A ketogenic metabolism is different. Carbohydrate and insulin requirements are lower, while fatty acids and ketones are actively being used as fuels — part of what I call multifuelling.

The lipid profiles differ too. Higher-carbohydrate metabolism tends towards higher triglycerides, lower HDL and more small dense LDL; ketogenic metabolism tends towards lower triglycerides, higher HDL and larger LDL particles.

Much of our cardiovascular evidence has been gathered in the former — a higher-carbohydrate, higher-insulin, fat-storage state — and then extrapolated to the latter.

Arteries are also affected by smoking, high blood pressure, chronic inflammation, lack of exercise, poor sleep, stress, high glucose and high insulin.

For someone with type 1 diabetes, chronic glucose exposure deserves particular attention. Duration of diabetes and chronic hyperglycaemia are among the strongest predictors of complications, and the estimate quoted in Shifting Gears is that each year spent above an HbA1c of 58 mmol/mol is associated with around 100 days of life lost.

That doesn’t mean cholesterol should be ignored. A substantial LDL rise — particularly with familial risk or other cardiovascular factors — deserves investigation.

But neither should cholesterol be considered in isolation from glucose, insulin, blood pressure, smoking, inflammation, activity, sleep and the metabolic state in which the lipids are circulating.

The physiology of cholesterol, lipoproteins and cardiovascular risk is explored in greater detail in Shifting Gears.