The brain is the most cholesterol-dense organ in the body. Much of that cholesterol is packed into myelin—the fatty, insulation-like sheath that wraps around neurons and allows electrical signals to travel rapidly and efficiently. Although the brain accounts for only about 2% of total body mass, it contains roughly 20–25% of the body’s total cholesterol.1
Given those numbers, it’s entirely reasonable that people worry about the effect of lowering cholesterol on their brain health: If cholesterol is so concentrated and critical in the brain, wouldn’t reducing it, such as with lipid-lowering therapies, risk impairing brain function?
While this concern makes intuitive sense, we already have some compelling insight into cholesterol synthesis to suggest it is not a major problem. Because circulating cholesterol does not cross the blood brain barrier, any cholesterol used by the brain must be synthesized within the brain itself. In other words, the cholesterol measured on a lipid panel is largely irrelevant to maintaining the structural cholesterol in neurons and myelin.
This local synthesis is reassuring for brain health, but it doesn’t necessarily eliminate all uncertainties. Clinically, some patients report symptoms such as brain fog from use of statins. While these reports are rare (and statistics on the exact prevalence of such symptoms are highly variable), they do occur. Subclinical negative effects compounded through decades could, the hypothesis goes, impact longer-term risk for outcomes like dementia. Despite our confidence that the brain synthesizes sufficient cholesterol on its own under normal circumstances, this remains a question worth testing: How do circulating lipids affect dementia risk?
Observational studies have attempted to answer this question by comparing people who take lipid-lowering medications to those who do not. Most studies find a neutral or beneficial effect of lowering atherogenic lipoproteins on dementia risk, but observational studies by their very nature have limitations. The lack of random assignment means we can’t make any conclusions about causation and, on average, people prescribed lipid-lowering medications are more likely to be taking other medications or have comorbidities that may obscure the true effects. Even after statistical adjustment, the sheer number of potential variables makes it challenging to determine whether lowering cholesterol is causally responsible for any observed change in dementia risk.
To move beyond this limitation, a recent study by Nordestgaard and colleagues used a powerful genetic approach called Mendelian randomization (MR) to ask a cleaner question: What happens to dementia risk in people who are genetically predisposed to lower lifelong levels of atherogenic lipoproteins?2
Rather than examining medication use over a few years, this approach leverages natural genetic variation to approximate the effect of lowering exposure to atherogenic lipoproteins, namely LDL, over an entire lifetime. By assessing lifelong biological exposure based on a randomly assigned, innate factor, rather than comparatively short-term drug interventions, this study allows us to make more definitive claims about the effect of cholesterol levels on dementia risk.
Before diving into what they found, it’s worth briefly reviewing how Mendelian randomization works, and why it offers a uniquely strong framework for evaluating causal questions like this one.
