Abstract
BACKGROUND: Tumour necrosis factor alpha blockers in rheumatoid arthritis are known to increase the risk of serious infections defined as life-threatening, requiring hospitalisation or intravenous antibiotics. Recently, new biological agents have become available. Their safety is an important issue. PURPOSE: To assess if biological agents, ie rituximab, abatacept and anakinra increase the risk of serious infections in patients with rheumatoid arthritis in published randomised controlled trials. DATA SOURCE: A systematic review of the literature using PUBMED, EMBASE, Cochrane library and abstracts databases (American College of Rheumatology and European League Against Rheumatism annual meetings) was performed up to October 2007. This search was completed with data from the Food and Drug Administration, the European Agency for the Evaluation of Medicinal Products and manufacturers. DATA EXTRACTION: Three fixed-effect meta-analyses were performed to compare serious infection rates between each biological agent and placebo. Pooled odds ratios (ORs) were calculated, using the Mantel-Haenszel method with a continuity correction. DATA SYNTHESIS: Twelve randomised controlled trials with data concerning serious infections were analysed (three for rituximab, five for abatacept and four for anakinra). They included 745 patients, 1960 patients, 2062 patients and 2112 patients treated by rituximab, abatacept, anakinra and placebo respectively. The overall pooled ORs did not reveal a statistically significant increased risk of serious infection for abatacept and rituximab; this risk was increased for high doses of anakinra (> or =100 mg daily) versus low dose and placebo (ORs = 9.63 (95% CI, 1.31 to 70.91) and 3.40 (95% CI, 1.11 to 10.46) respectively). CONCLUSIONS: These meta-analyses did not reveal a significant increase in the risk of serious infections during rituximab or abatacept treatments in patients with rheumatoid arthritis; however, high doses of anakinra may increase this risk, especially when patients have comorbidity factors. Large studies must be performed to confirm this safety profile in daily practice.