Increased cardiovascular morbidity and mortality has been documented in several chronic inflammatory rheumatic diseases. In addition to the traditional cardiovascular risk factors, disease activity itself may contribute to this increase.

Tumour necrosis factor (TNF)-α, a pivotal cytokine in chronic inflammation, also affects lipid metabolism, insulin resistance and endothelial function. Therapeutic use of TNF-α blockers reduces inflammation and modifies the lipoprotein spectrum of patients. A significant increase in high-density lipoprotein (HDL) cholesterol levels has been documented after short-term infliximab therapy.1^–5 However, this beneficial effect seems to be transient and more prolonged use of TNF-α blockers may lead to an increase in total cholesterol and low-density lipoprotein (LDL) cholesterol, inducing a more “atherogenic” phenotype.6 7

The majority of clinical studies have looked at the use of infliximab. Data regarding the effect of other anti-TNF-α agents on lipid levels is scarce.

We compared the changes in lipid levels after 1 year of treatment with infliximab and etanercept, in a small group of patients with rheumatoid arthritis (RA), ankylosing spondylitis (AS) and psoriatic arthritis (PsA).

The study included 65 consecutive patients (36 females) with a mean age of 47.4 (SD 13.3) years, 30 patients with RA, 29 with AS and six with PsA. Forty-four patients were treated with infliximab and 21 with etanercept. At baseline, 23 (35.4%) had increased total cholesterol levels, 17 (26.2%) had increased LDL levels, seven (10.8%) had increased triglycerides and nine (13.8%) had low HDL. Thirty-three patients were on low-dose steroids and six were taking statins. There were no significant differences in lipid levels between the infliximab and the etanercept group.

After 1 year of treatment, 71.4% of patients with AS had a good response (reduction of BASDAI ⩾50%) and 73% of RA and PsA patients had a moderate/good DAS28 response. Despite reduced inflammatory activity and reduction in steroid dosage patients developed a more “atherogenic” lipid profile. There was an increase, in the whole group, in total cholesterol (p<0.0001) and LDL levels (p = 0.02), without significant changes in HDL or triglycerides, which is in agreement with previously published studies.6 7 When analysed separately, we found an elevation in total cholesterol (p<0.001) and LDL (p = 0.003) in the infliximab group, while in the etanercept group there was a significant increase in HDL (p<0.05) and no significant changes in total cholesterol or LDL (table 1). Despite the small number of patients, the differences between infliximab and etanercept could not be attributed to a distinct clinical response evaluated by EULAR and ASAS (ASsessment in Ankylosing Spondylitis working group) response criteria, disease activity level, steroid use or changes in body mass index, which were similar in the two groups.

Both etanercept and infliximab block TNF-α but lymphotoxin-α blockage is exclusive to etanercept. The distinct mechanism of action is a possible explanation for the improved lipid profile observed in the etanercept group. Indeed, lymphotoxin-α seems to be more “pro-atherogenic” than TNF-α and in animal models, the absence of this cytokine was associated with a threefold decrease in atherosclerotic lesions size, lower levels of total cholesterol and higher levels of HDL, compared with the TNF-α-deficient mice that exhibit more atherosclerosis.8

To our knowledge, a different effect of infliximab and etanercept on lipid levels has not been previously reported in the literature. Our findings suggest that etanercept use is associated with a less atherogenic lipid profile and this might have implications in drug selection based on an individual patient’s cardiovascular risk. Further investigation is necessary to address whether cardiovascular morbidity and mortality is different with these two anti-TNF-α drugs.