Objective: Treatment with tumour necrosis factor antagonists (anti-TNF) has been recognized as a risk factor for tuberculosis (TB) reactivation. Our aim was to evaluate risk of TB reactivation in rheumatologic and non-rheumatologic diseases treated with the same anti-TNF agents with and without concomitant therapies.
Methods: We searched for randomized controlled trials (RCTs) evaluating infliximab, adalimumab, and certolizumab in both rheumatologic and non-rheumatologic diseases until 2012. Results were calculated as pooled rates and/or pooled odd ratios (OR).
Results: Overall, 40 RCTs with a total of 14,683 patients (anti-TNF: 10,010; placebo: 4673) were included. TB reactivation was 0.26% (26/10,010) in the anti-TNF group and 0% (0/4673) in the control group, corresponding to an OR of 24.8 (95% CI 2.4-133). TB risk was higher when anti-TNF agents were combined with methotrexate or azathioprine as compared with either controls (24/4241 versus 0/4673; OR 54; 95% CI 5.3-88) or anti-TNF monotherapy (24/4241 versus 2/5769; OR 13.3; 95% CI 3.7-100). When anti-TNF was used as monotherapy, TB risk tended to be higher than placebo (2/5769 versus 0/4673; OR 4; 95% CI 0.2-15.7).
Conclusions: TB risk with anti-TNF agents appeared to be increased when these agents were used in combination with methotrexate or azathioprine as compared with monotherapy regimen. TB risk seemed to be higher than placebo, even when monotherapy is prescribed.
Keywords: Ankylosing spondylitis; anti-TNF; inflammatory bowel diseases; methotrexate; psoriasis; rheumatoid arthritis; tuberculosis.