Background Latent TB reactivation is a cause of concern in patients with SLE, since the underlying immune dysregulation and concurrent immunosuppressive use may influence the sensitivity of both the Tuberculin Skin Test (TST) and the newer interferon-Y release assays (IGRAs), such as the QuantiFERON-TB Gold (QFT-G) test. Screening for latent tuberculosis infection (LTBI) with TST has shown low sensitivity when used in SLE, with increased false negative results. The applicability of IGRAs in SLE patients has also not been fully explored.
Objectives Our study aims at comparing the performance of TST and QFT-G testing as a screening strategy for the detection of LTBI in a high-risk, inner city, SLE patient population.
Methods A retrospective chart review study was conducted in SLE patients at an outpatient rheumatology clinic of an urban teaching hospital serving a large immigrant population. 174 SLE patients were identified who satisfied the 2012 SLICC SLE diagnostic criteria. The patients had been screened for LTBI at initial evaluation using TST (positive if >/ 5mm), QFT-G, or both tests. QFT-G was interpreted as positive, negative, or indeterminate. Patients were diagnosed as having LTBI if TST and/or QFT-G were positive and were treated with 9 months of Isoniazid/B6, after ruling out active pulmonary tuberculosis (TB) by chest radiography. Findings from this study were compared to a previously presented study in RA where similar methodology had been used.
Results Of the 174 SLE patients who were studied, 73(42.0%) had documented LTBI screening. 45(25.9%) had been screened with TST, 28(16.1%) with QFT-G, and 15(8.62%) with both tests. Out of the 45 screened with TST, 8(17.7%) were positive, while of the 28 patients screened with QFT-G, 7(25%) had positive, 4(14.28%) indeterminate, and 17(60.7%) negative results. 2/7 QFT-G positive patients had a concurrent TST test done which was negative. When comparing patients who had both tests done, 1/15 tested positive and 10/15 tested negative for both tests, while 2/15 patient tested positive for TST but negative for QFT-G. The agreement between TST and QFT-G was 73.33% with a kappa value of 0.167(CI-95% -0.401 to 0.734) versus a kappa of 0.306(CI-95% 0.136-0.477) in the RA study. LTBI was diagnosed in 8% of the SLE patients in comparison to 38% in RA.
Conclusions TST and QFT-G tests have poor correlation with each other and LTBI diagnosis can be missed in SLE patients when using each test alone. A safer, comprehensive approach for LTBI detection may be the combined use of TST and IGRAs, especially since potent biologic agents are being added in the SLE therapeutic armamentarium.
Disclosure of Interest: None Declared