Abstract
OBJECTIVE: Tuberculosis (TB) in patients with rheumatoid arthritis (RA) undergoing treatment with anti-tumor necrosis factor (TNF) agents is commonly the result of reactivation of latent TB infection (LTBI); detection and treatment of LTBI is essential before treatment with anti-TNF agents. We reported previously that the tuberculin skin test (TST) is inaccurate for diagnosis of LTBI in patients with RA. Here, we compare the prevalence of LTBI in RA patients and matched controls according to positive TST and QuantiFeron-TB Gold In-Tube version (QFT) results and determine their agreement. METHODS: A cross-sectional study of 101 RA patients and 93 controls was conducted in Lima, Perú, where the prevalence of LTBI in the general population has been estimated to be 68%. Blood was drawn for QFT assay followed by TST using 2-TU of RT 23 purified protein derivative. TST was deemed positive at >or= 5 mm for RA patients and >or= 10 mm for controls. RESULTS: There were no significant differences between RA patients and controls for age, sex, bacillus Calmette-Guérin vaccination, or history of or contact with TB. 88% of patients had active RA disease and 2 (1.9%) patients had indeterminate QFT results. The number of subjects testing positive with the QuantiFeron assay was comparable between patients and controls (44.6% vs 59.1%, respectively), whereas the TST detected significantly less LTBI among RA patients (26.7%) than controls (65.6%). Thus, the rate of LTBI in RA patients represented 75% and 41% of the rate in their controls using QFT or TST, respectively (p = 0.008). Poor agreement between TST and QFT was seen in RA patients, but in controls, good agreement was observed between these tests. CONCLUSION: In a TB-endemic population, the QuantiFeron-TB Gold In-Tube assay seemed to be a more accurate test for detection of LTBI in RA patients compared with the TST, and may potentially improve the targeting of prophylactic therapy before treatment with anti-TNF agents.