A meta-analysis was just published in the Journal of the American Medical Association, Association of LDL Cholesterol, Non–HDL Cholesterol, and Apolipoprotein B Levels With Risk of Cardiovascular Events Among Patients Treated With Statins, March 28, 2012, Boekholdt et al. 307 (12): 1302 — JAMA, that suggests setting statin dosage by the non-HDL cholesterol, rather than LDL, as is more commonly done.
They looked at a number of studies and tried to determine the effect (in terms of increase in cardiovascular “events” per standard deviation change) for three measures: LDL cholesterol, non-HDL cholesterol, and apolipoprotein B. They found that non-HDL cholesterol was the best predictor of risk (hazard ratio of 1.16 per standard deviation, compared to 1.13 for LDL cholesterol and 1.14 for apolipoprotein B). That means for each standard deviation of increase in non-HDL cholesterol, there is a 16% higher chance of a major cardiovascular event (which includes myocardial infarctions, strokes, and unstable angina).
I looked at my own LDL and non-HDL levels from my most recent blood tests and saw that I was near the average of the top quartile for both, making my hazard ratio (relative to the bottom quartile) 1.26 based on LDL and 1.42 based on non-HDL. That means I have about a 42% higher chance of a stroke or heart attack than someone with the same age, gender, body mass index, … whose non-HDL levels were in the bottom quartile.
What I don’t know is whether more aggressive statin therapy is warranted in my case. How does the increased risk of liver or muscle damage from higher doses of statins compare to decreased risk of heart attack or stroke? Other than the inherited high cholesterol levels, the only risk factors I have are being male and over 55—I get a fair amount of aerobic exercise, don’t smoke, am not overweight, and don’t have high blood pressure. According to online cardiac-risk factor calculators, my risk of a heart attack (or death from coronary heart disease) in the next 10 years is 8%–12% (depending on which calculator I use and what it considers). Small changes in cholesterol (about what is achievable based on my 20 years of statin usage) can move me more solidly into the 8% risk category, but not really reduce the risk lower than that. If the calculations were based on non-HDL cholesterol, they might assess my risk slightly higher.