CHOLESTEROL AND CORONARY HEART-DISEASE - PREDICTING RISKS IN MEN BY CHANGES IN LEVELS AND RATIOS

Background: Little is known about the relative ability of different me asures of change in cholesterol to discriminate coronary heart disease risk. We evaluated this ability for changes in low-density lipoprotei n (LDL) cholesterol, high-density lipoprotein (HDL) cholesterol, LDL/H DL ratios, and total cholesterol/HDL ratios. Methods: We predicted ris ks of coronary heart disease using data from 3641 men in the Lipid Res earch Clinics Coronary Primary Prevention Trial. Treating these patien ts as a cohort, we estimated risks associated with changes in choleste rol levels independent of the patients' randomization group. Results: Changes in LDL and HDL cholesterol when used in combination were each significant predictors of coronary heart disease risk (odds ratios [OR ] for 10% increases, 1.15 and 0.84, respectively; P < 0.001). Changes in LDL/HDL and total cholesterol/HDL ratios had similar discriminating ability (OR for 10% increases, 1.17 and 1.21, respectively; P < 0.000 1), In the best discriminating models, changes in ratios added informa tion about risks to changes in LDL cholesterol, although changes in LD L cholesterol levels failed to add information to changes in ratios. C onclusions: Changes in total cholesterol/HDL and LDL/HDL ratios were b etter predictors of risk for coronary heart disease than were changes in LDL cholesterol levels alone. When assessed as percentage changes a veraged during the first two months of intervention, they were among t he best discriminators of risk. Clinicians selecting treatments for in tervention should include among their considerations the treatment's e ffect on both LDL and HDL cholesterol rather than their effects on LDL cholesterol levels alone.