Association between change in high density lipoprotein cholesterol and cardiovascular disease morbidity and mortality: systematic review and meta-regression analysis.
Level 1 - systematic review of randomized trials
Systematic review and meta-regression analysis of 108 randomized controlled trials
PubMed 19221140 · doi:10.1136/bmj.b92
What was done
A systematic review and trial-level meta-regression of randomized controlled trials was conducted using Medline, Embase, Central, CINAHL, and AMED databases up to October 2006. Eligible studies evaluated lipid-modifying interventions aimed at reducing cardiovascular risk, reported HDL cholesterol concentrations and clinical outcomes (CHD death, non-fatal myocardial infarction, or total death) separately by group, and had a minimum treatment and follow-up duration of six months. The analysis assessed whether treatment-induced changes in HDL cholesterol were associated with clinical risk ratios after adjusting for changes in LDL cholesterol and drug class.
What was found
The analysis included 108 randomized trials involving 299,310 participants. Across all models adjusted for LDL cholesterol changes, treatment-induced changes in HDL cholesterol showed no association with risk ratios for CHD deaths, CHD events, or total deaths, explaining less than 1% of the variability in clinical outcomes. The ratio of LDL to HDL cholesterol explained no more variability than LDL changes alone. In contrast, after adjusting for HDL changes and drug class, each 10 mg/dL (0.26 mmol/L) reduction in LDL cholesterol produced relative risk reductions of 7.2% (95% CI 3.1% to 11%; P=0.001) for CHD deaths, 7.1% (95% CI 4.5% to 9.8%; P<0.001) for CHD events, and 4.4% (95% CI 1.6% to 7.2%; P=0.002) for total deaths.
Why it matters
This study demonstrated that simply raising circulating HDL cholesterol levels is not an effective surrogate target for reducing cardiovascular morbidity or mortality, solidifying LDL cholesterol lowering as the primary therapeutic goal in lipid management.
Limits
The meta-regression relied on trial-level aggregate data rather than individual participant data, limiting the ability to assess patient-level modifiers or non-linear effects. The search concluded in October 2006, omitting subsequent trials of newer HDL-raising drug classes. The study only assessed circulating HDL cholesterol mass concentration rather than HDL particle function, subfractions, or reverse cholesterol transport metrics.
Cited by
- supports Therapeutic interventions designed to raise HDL cholesterol have failed to reduce heart disease risk, whereas interventions lowering LDL have consistently succeeded.