The key role played by low-density lipoprotein (LDL) particles in the pathogenesis of coronary heart disease (CHD) is well accepted, as is the benefit of lowering LDL in high-risk patients. What remains controversial is whether we are using the best measure(s) of LDL to identify all individuals who would benefit from therapy. Many studies have shown that, at a given level of LDL cholesterol, individuals with predominantly small LDL particles (pattern B) experience greater CHD risk than those with larger-size LDL. However, it is not clear from this observation that small LDL particles are inherently more atherogenic than large ones because, at a given level of LDL cholesterol, individuals with small LDL have more LDL particles in total. The phenotype of small LDL particle size co-segregates with a cluster of metabolic factors, including elevated triglycerides and reduced HDL cholesterol, and in multivariate analyses has generally been found not to be independently associated with CHD risk. In contrast, LDL particle number measured by nuclear magnetic resonance has consistently been shown to be a strong, independent predictor of CHD.