Interferon-gamma release assays versus tuberculin skin testing in patients with rheumatoid arthritis.

The aim of this study was to analyze the results of interferon-gamma release assays (IGRAs) and tuberculin skin tests (TST) performed to detect latent tuberculosis infection (LTBI) in patients with rheumatoid arthritis (RA). Interferon-γ release assays and TST test results were summarized and systematically reviewed. Four hundred and five RA patients and 339 controls that underwent IGRA and/or TST were identified in seven studies. Five studies were case-control studies and two were cross-sectional studies. Among RA patients, the IGRA positivity rate was 31.6% (89/282; range 11.4%-44.6%), and the TST positivity rate was 23.0% (78/339; range from 14.60% to 45%). Concordance rates ranged from 40% to 76% and discordance rates from 24% to 29.7%. Agreement between IGRAs and TST in RA was poor (69.6%, k = 0.33, 95% CI 0.188-0.478). The IGRA positivity rate was 31.0% in RA and 40.0% in controls, which was not significant (relative risk [RR] 0.802, 95% CI 0.629-1.023, P = 0.075). The TST positivity rate was 24.7% in RA and 50.5% in controls, and this difference was not significant (RR 0.680, 95% CI 0.331-1.339, P = 0.295). Positivity rates of IGRA and TST were 31.6 and 23.0%, respectively, in RA patients. Agreement between IGRA and TST results in RA was poor. Our data suggest that both IGRA and TST are needed to detect LTBI in RA.