Objectives: This study sought to assess the contribution of low-density lipoprotein (LDL)-dependent and LDL-independent effects of LDL-lowering therapies to changes in C-reactive protein (CRP) in healthy or stable subjects.
Background: Correlations of change in LDL and CRP in individuals are lowered by their measurement variability. By using average changes in LDL and CRP in study groups, meta-analysis reduces this variability to better assess their correlation.
Methods: A systematic search for randomized placebo-controlled trials reporting change in LDL and CRP with LDL-lowering interventions retrieved 23 studies with 57 groups treated with a variety of statins, nonstatin drugs, or other regimens. Meta-analysis techniques assessed the relationships between average mean differences (placebo - treatment) in change in CRP and LDL.
Results: The overall reduction in CRP was 28% (95% confidence interval 26% to 30%). Significantly greater CRP reduction occurred in statin and statin-ezetimibe interventions, interventions using 80 mg/day of statins, and with greater LDL lowering. Meta-regression analysis showed a strong correlation between the change in LDL and CRP (r = 0.80, p < 0.001). Statin therapies had no significant effect on CRP after adjusting for the change in LDL. In a multivariate model applied to a range of LDL reduction typically seen with statins (20% to 60%), 89% to 98% of CRP change was related to LDL lowering and 2% to 11% was related to non-LDL effects of statins.
Conclusions: In clinical practice, most of the anti-inflammatory effect of LDL-lowering therapies is related to the magnitude of change in LDL. The potential non-LDL effects of statins on inflammation are much smaller in magnitude.