Biological agents, particularly anti-Tumour Necrosis Factor (TNF)-α agents, have emerged as an effective treatment in patients with chronic inflammatory diseases. An association between anti-TNF-α antibodies and reactivation of latent tuberculosis infection (LTBI) has been established. Appropriate screening for TB infection has become mandatory before starting a treatment based on TNF-α inhibition. The objective was to determine the usefulness of IFN-γ release assays in diagnosing LTBI in patients with inflammatory rheumatic diseases scheduled for anti-TNF-α treatment. The study included 53 individuals with inflammatory rheumatism. All patients had a TST, a chest radiograph, QuantiFERON Gold In-Tube (QFN-G-IT) and T-SPOT.TB. To investigate the influence of non-tuberculous mycobacteria (NTM) infections on non-BCG-vaccinated patients, with a positive TST result and both negative IFN-γ assays, we performed an ex vivo ELISPOT, stimulating the cells separately with NTM sensitins. TST was positive in 7 cases, T-SPOT.TB in 11 and QFN-G-IT in 9 cases. Agreement between TST and T-SPOT.TB and QFN-G-IT was 77.35% (κ = 0.33 and κ = 0.40, respectively), and between both in vitro tests, it was 83.01% (κ = 0.57). Of the three patients with positive TST and negative T-SPOT.TB and QFN-G-IT, one positive ELISPOT result was obtained after stimulation with NTM sensitins. Positive TST, T-SPOT.TB and QFN-G-IT results were not affected by the immunosuppressive therapies. IFN-γ release assays are useful methods for avoiding TST false-positive results, but in those patients with a high risk of developing active TB and in the absence of predictive value studies in this specific kind of population for knowing how safe is the use of IGRAs alone, the combined use of TST and IFN-γ tests should be recommended in order to increase the overall number of LTBI diagnoses.