Objective: Lower levels of low-density lipoprotein (LDL) cholesterol may be associated with increased cardiovascular (CV) risk in rheumatoid arthritis (RA). This study was undertaken to determine whether the complex relationship between levels of LDL and high-density lipoprotein (HDL) cholesterol and CV risk is different in RA patients as compared to non-RA controls.
Methods: Using data from a US health insurance plan (2003-2012), we conducted a cohort study that included patients with RA and non-RA control subjects matched with regard to age, sex, and index date. The nonlinearity of associations between lipid levels and incidence of major adverse CV events (MACE) was tested. We used multivariable Cox proportional hazards regression models to examine for an interaction between lipid levels and RA status in relation to the risk of MACE, after adjustment for CV risk factors.
Results: In total, 16,085 RA patients and 48,499 non-RA controls were studied. The mean age was 52.6 years and 78.6% were women. The relationship between LDL cholesterol levels and incidence of MACE was nonlinear and similar between RA patients and non-RA controls (P for interaction = 0.72). No significant increase in CV risk was observed between the lowest quintile of LDL cholesterol levels (≤91.0 mg/dl) and the second, third, or fourth quintiles, whereas the highest quintile (>190.0 mg/dl) conveyed a 40% increase in risk of MACE (hazard ratio [HR] 1.40, 95% confidence interval [95% CI] 1.17-1.68). The relationship between HDL cholesterol levels and incidence of MACE was also nonlinear and similar between RA patients and non-RA controls (P for interaction = 0.39). Compared to the lowest quintile of HDL cholesterol levels, each successive quintile was associated with a reduced risk of MACE (HR 0.45, 95% CI 0.48-0.72 for lowest quintile [≤43.0 mg/dl] versus highest quintile [>71.0 mg/dl]).
Conclusion: The complex relationship between LDL cholesterol levels, HDL cholesterol levels, and risk of MACE was nonlinear in RA patients and also not statistically significantly different from that in an age- and sex-matched non-RA cohort.
© 2015, American College of Rheumatology.