Cholesterol levels not highly correlated with heart disease - Recent study indicates that the two most commonly used risk algorithms (based primarily on blood cholesterol levels) correlate poorly with heart disease. Lead author Dr. Kevin M. Johnson says the risk profiles based on Framingham score or the National Cholesterol Education Program (NCEP) Adult Treatment Panel III (ATP III) are "weak discriminator of the overall atherosclerotic plaque burden and may lead to over- or undertreatment of patients."
Johnson points out that the Framingham risk estimate is derived from epidemiologic observations. As other studies have shown, Framingham predicts the risk of a coronary event only 60% to 65% of the time. "....[T]here will be a lot of people who have a low Framingham risk who have a lot of atherosclerotic plaque, and a lot of people with high risk, by Framingham score, with no plaque,"
BINGO! Exactly right, and exactly what I've been saying to my docs for months now. Epidemiologic studies are next to worthless for clinical decisions. Epidemiologic studies should guide clinical , NOT clinical . People with low cholesterol, high cholesterol, and everything in-between have heart disease. If the model (risk profile) does not fit the observations, the model needs to be replaced with another that better matches observations. Promising models include how the molecular size of cholesterol impacts heart disease risk, Unified Theory of Heart Disease - by Linus Pauling, etc.
Will these results impact the current focus on statins, the main class of medications used to control cholesterol? Unlikely. Old models die hard--especially one so entrenched in the public psyche as the cholesterol model. I remember hearing a doc say that clinical treatment lags 10 years behind current research findings. So perhaps clinicians will adjust to this news in another decade. In the meantime, I'm fending for myself.
|The Protein Power Lifeplan |
by Michael R. Eades & Mary Dan Eades