BackgroundWe performed a meta-analysis to evaluate the effect of anti–tumor necrosis factor (TNF) therapy on the frequency of extra–articular manifestations (EAMs) in patients with ankylosing spondylitis (AS).MethodsWe searched with the terms ‘ankylosing spondylitis’, ‘infliximab’, ‘etanercept’, ‘adalimumab’, ‘golimumab’, ‘certolizumab’, ‘TNF inhibitor/blocker/antagonists’ or ‘anti-TNF’ on MEDLINE, EMBASE and Cochrane Library for randomized controlled trials (RCTs) of ≥12 weeks with parallel or crossover design of TNF inhibitor versus placebo to treat uveitis, inflammatory bowel disease (IBD) and/or psoriasis of AS, published before February 2014.ResultsWe found 8 RCTs that fit our criteria. Anti–TNF therapy was associated with less uveitis than placebo in patients with AS (OR: 0.35, 95% CI: 0.15–0.81, P = 0.01). Subgroup analysis showed receptor fusion proteins were more efficacious for uveitis than placebo (OR: 0.30, 95% CI: 0.09–0.94, P = 0.04), but monoclonal antibodies were not (OR: 0.43, 95% CI: 0.12–1.49, P = 0.18). Anti–TNF therapy and placebo group did not significantly differ in treating IBD in AS patients (OR: 0.75, 95% CI: 0.25–2.29, P = 0.61). In subgroup analysis, neither monoclonal antibodies (OR: 0.45, 95% CI: 0.10–1.92, P = 0.28) nor receptor fusion proteins (OR: 1.52, 95% CI: 0.25–9.25, P = 0.65) significantly differed from placebo in treating IBD. We found no suitable reports on psoriasis.ConclusionsAnti–TNF therapy was preventive for flares or new onset of uveitis in AS patients, and might be an alternative for these patients. However, monoclonal anti–TNF antibodies and TNF receptor fusion proteins were not efficacious for IBD in AS patients.