Tuberculosis risk and anti-tumour necrosis factor agents in rheumatoid arthritis: a critical appraisal of national registry data.

Tuberculosis (TB) remains a major global health problem. In patients with rheumatoid arthritis (RA), the risk of reactivation of latent TB infection (LTBI) is increased and treatment with tumour necrosis factor (TNF) antagonists further increases this risk. However, interpretation of results describing comparative TB risk during therapy with different TNF antagonists is difficult. This is not only a result of different patient ethnic groups and background TB rates, but also because of differing methods of data acquisition. This paper offers a critical appraisal of registry data pertaining to RA patients treated with different anti-TNF agents, focusing on methodological approaches that may limit the generalizability of findings or invalidate the direct comparison of TB risk between different national registries. Underlying factors that can make data interpretation challenging are discussed, including differences in methods for TB diagnosis or data collection and reporting, as well as background TB risk. The introduction of special monitoring systems, such as prospective multinational registries, to strengthen surveillance and better quantify the extent of under-reporting is required, especially in countries where the background TB risk is high.