Low high-density lipoprotein cholesterol and response to simvastatin therapy in Scandinavian Simvastatin Survival Study (4S).

Response to Simvastatin Therapy in Scandinavian Simvastatin Survival Study (4S) To the Editor: Ballantyne et al1 compared the coronary heart disease (CHD) event rate and the response to simvastatin therapy in a subgroup of approximately 500 individuals from the Scandinavian Simvastatin Survival Study (4S) study with “isolated increased LDL-C” with a similarly-sized subgroup that had not only an increased LDL cholesterol (LDL-C), but also the lowest levels of HDL cholesterol (HDL-C) and highest levels of triglycerides (“the lipid triad”) in 4S. The group with the lipid triad was found to have a higher CHD event rate and a greater reduction in CHD events with simvastatin therapy than those with an isolated increased LDL-C. Those patients with isolated increased LDL-C were said to have “normal triglyceride and HDL-C levels.” Although individuals with isolated increased LDL-C did have highly favorable triglycerides with a mean (SD) of 80 13 mg/dL, this group had not only “normal” levels of HDL-C, but also a high enough HDL-C level at a mean of 63 11 mg/dL to have a “protective” effect from HDL-C. Indeed, in the GISSI-Prevenzione report2 of a European cohort followed for 4 years after myocardial infarction (which seems to be the only published predictive algorithm for secondary prevention to date), there was a relative 3% reduction of CHD death for each mg/dL increment of HDL-C that was greater than 55 mg/dL in men (with an even greater reduction of CHD death in women). With CHD, an HDL-C value greater than 55 mg/dL is distinctly unusual. In the GISSI report, which examined 11 000 men and women with CHD, only 11% of the men had HDL-C 55 mg/dL. To more fairly judge the contribution of a low HDL-C to the risk associated with a high LDL-C would require that a group with HDL-C in a range closer to the median range of HDL-C, with CHD of about 40 to 45 mg/dL,2 be compared with a lower HDL-C. It is possible that the risk and response to statin therapy in the 4S low HDL-C subgroup is so profoundly influenced by the high LDL-C (which is characteristic of the 4S patient population in general) that no differences would be apparent if a comparison were to be made between the 4S subgroup with a low HDL-C level and a HDL-C subgroup with an average HDL-C rather than a subgroup with an unusually high and protective level of HDL-C.