Association between change in high density lipoprotein cholesterol and cardiovascular disease morbidity and mortality: systematic review and meta-regression analysis.

Description

In this meta-regression analysis, the aim was to investigate the association between treatment-induced changes in HDL cholesterol levels and total death, coronary heart disease (CHD) death and CHD events (CHD deaths and non-fatal myocardial infarction) after adjustments for LDL cholesterol. This analysis considered 108 randomized trials, which involved 299,310 individuals. After adjusting for LDL cholesterol, no association was found between treatment-induced changes in HDL cholesterol and risk ratios for CHD deaths, CHD events of total mortality. In all trials included, changes in HDL cholesterol explained less than 1% of the variability of the outcomes. Changes in the LDL cholesterol/HDL cholesterol ratio provided similar results, as did changes in LDL cholesterol alone. Moreover, for a 10 mg/dl (0.26 mmol/l) reduction in LDL cholesterol, the relative risk reduction was 7.2% (p=0.001) for CHD deaths, 7.1% (p<0.001) for CHD events and 4.4% (p=0.002) for total deaths, after adjusting for changes in HDL cholesterol and drug class. These results once again support LDL cholesterol reduction as the primary goal in lipid management. Although they recognized in their editorial that the study by Briel et al. supports the notion that LDL cholesterol should be the primary target of lipid modifying treatments, Ghali and Rodondi raised the point that this does not mean that clinicians should not try to increase HDL cholesterol in their patients through non-drug interventions such as physical activity and smoking cessation. They also identified some major limitations of the meta-regression analysis, such as the use of aggregated study data rather than individual patient data and the analytical constraints posed by the modest change and variability in mean HDL cholesterol levels in the examined studies.