Cardiovascular disease in systemic lupus erythematosus

Abstract

Premature cardiovascular disease (CVD) is a major cause of morbidity and mortality in systemic lupus erythematosus (SLE). This is an important clinical problem and a main objective of this study was to identify risk factors and underlying mechanisms. Since both atherosclerosis and SLE are chronic inflammatory conditions, SLE can also be regarded as a "human model" for studies of inflammatory components in CVD, not least in women, who make up 90% of SLE cases.

A cohort comprising 208 SLE patients is the base of these studies. From this cohort a nested case control study was designed. Twenty-six women (52±8.2 yrs) with SLE and a history of CVD (SLE cases) were compared to 26 age-matched women with SLE but without manifest CVD (SLE controls) and 26 age-matched population-based control women (population controls). Common carotid intima- media thickness (IMT) was measured by Bmode ultrasound as a surrogate measure of atherosclerosis. Echocardiography was performed to assess valvular abnormalities. A monoclonal antibody (E06) to oxidized low density lipoprotein (oxLDL) was used to determine oxidation epitopes on LDL.

This is the first study to show that SLE cases had enhanced atherosclerosis, as measured by increased IMT, in comparison to population controls while the IMT of SLE controls did not differ from that of population controls. SLE cases were, as compared to both control groups, distinguished by higher degree of inflammation: raised levels of acute phase reactants (alpha-1-antitrypsin, C-reactive protein, orosomucoid and erythrocyte sedimentation rate) and soluble TNF-alpha; dyslipidemia with raised triglycerides (TG) and lipoprotein (a), and decreased high density cholesterol (HDL); raised plasma concentrations of circulating oxLDL, and autoantibodies to epitopes of oxLDL (aoxLDL); lupus anticoagulant (LAC) and hyperhomocysteinemia. The cumulative prednisolone dose was higher in the SLE case group. Disease duration, smoking, BMI or diabetes mellitus did not differ between the groups.

Valvular abnormalities were essentially confined to the SLE case group with a strong correlation to hyperhomocysteinemia and high TG, two risk factors with known potential to induce endothelial dysfunction and damage. Levels of TNF-alpha and soluble TNF-alpha receptors correlated with high TG/low HDL. This was first noticed among SLE cases and later confirmed in the large cohort. In other settings, TNF-alpha is known to cause similar patterns of dyslipoproteinemia through known mechanisms. Dyslipoproteinemia, especially high TG, and activity in the TNF-alpha system also correlated to higher disease activity and cumulative disease damage. Elevations of TG and TNF-alpha activity were more pronounced in a subgroup of patients with CVD and renal manifestations. Thus, they were markers for more severe SLE. High levels of LAC were also associated with CVD. AoxLDL are closely related to aPL, and were elevated in SLE in general. We also measured oxidation epitopes per LDL particle and found that they were enhanced in SLE in general, but in particular in SLE patients with CVD and renal manifestations. These data support a role for immune responses to "neo-self" antigens, generated during oxidation, in SLE related CVD.

In conclusion, we here demonstrate that atherosclerosis is enhanced in SLE patients with CVD. We have identified a set of traditional and non-traditional risk factors, which are important in SLE- related CVD. Several of these risk factors, like raised TNF-alpha levels and closely associated dyslipoproteinemia, correlate with disease activity. TNF-alpha may be a major factor in SLE-related CVD acting both by causing dyslipidemia and by promoting atherosclerosis-related inflammation. Enhanced expression of oxidatively modified "neoantigens" on LDL may also be of importance for premature CVD and a trigger for autoimmunity to phospholipids (aPL). Thus, the majority of CVD risk factors identified in these studies are more or less strongly associated with SLE per se.