Aims To describe interferon-γ release assay (IGRA) and IGRA/tuberculin skin test (TST) discordance and their associated factors in patients with uveitis.

Methods Prospective study of 102 consecutive new patients presenting with uveitis to a tertiary referral centre. All patients underwent a complete ocular and systemic examination at presentation, with QuantiFERON-Gold in-tube (QFT), T-SPOT.TB and TST performed. Discordant results of TST/QFT, TST/T-SPOT.TB and QFT/T-SPOT.TB were analysed. Binary logistic regression was used to identify factors associated with discordant results.

Results There was fair agreement between TST and T-SPOT.TB (κ=0.371, p<0.001); moderate agreement between TST and QFT (κ=0.471, p<0.001). Both IGRAs showed good agreement (κ=0.668, p<0.001). The discordance rate between the TST/QFT was 26.5%, while TST/T-SPOT.TB was 31.4%. The discordance rate between the IGRAs was 16.7%. There was a significant association between TST/IGRA discordance with increasing age (OR 0.95, 95% CI 0.92 to 0.99; p=0.022), less likely with panuveitis (OR 0.18, 95% CI 0.04 to 0.86; p=0.032), and with Indian ethnicity (OR 14.07, 95% CI 1.45 to 137.03; p=0.023).

Conclusions Discordance of IGRA and TST results is common and requires further study. In our study in patients with uveitis, we found that ethnicity, age, and type of uveitis were some significant factors associated with TST/IGRA discordance, which may need to be taken into consideration when interpreting these results.