TB Screening Before Anti-TNF-alpha Therapy

By Carol A. Kemper, MD, FACP

Dr. Kemper is Clinical Associate Professor of Medicine, Stanford University, Division of Infectious Diseases, Santa Clara Valley Medical Center. Peer reviewer Connie T. Price, MD, is Assistant Professor, University of Colorado School of Medicine.

Dr. Kemper does research for GSK Pharmaceuticals, Abbott Laboratories, and Merck. Dr. Price reports no financial relationships relevant to this field of study.

This article originally appeared in the January 2010 issue of Infectious Disease Alert.

Screening for latent TB infection (LTBI) seems to have only gotten that much more complicated and, increasingly, infectious disease experts are being asked to interpret the newer tests results, especially in persons with inflammatory disorders or underlying immune suppression. In addition to the different gamma-interferon-based (GIB) assays available on the market, newer versions of outdated tests have been introduced. While GIB assays are more sensitive and specific for LTBI than TST, especially in previously BCG-vaccinated persons, only a minority of healthy individuals with LTBI remain at risk for reactivation TB. What is really needed is a diagnostic test that can tell you who those people are.

On the other hand, GIB results have reportedly waned in some patients with prior TB exposure, suggesting that some patients at risk for reactivation TB may be missed with currently available screening tools. Immunosuppression, such as the use of corticosteroids, not only increases the risk of falsely-negative TST, but increases the risk of indeterminate GIB assay results.

A retrospective analysis examined 50 patients with psoriasis who were screened for LTBI using TST, T-SPOT.TB, and chest radiographs in advance of treatment with an anti-TNF-alpha agent.1 Risk factors for LTBI, including prior TB exposure, abnormal chest radiograph suggestive of prior granulomatous disease, or residence in a country endemic for TB were compared to TST and T-SPOT.TB results. Ninety percent of the patients reported prior BCG vaccination; 20% had come from a country endemic for TB, 22% had had prior exposure to someone with TB, and 8% had abnormal chest radiographs.

The agreement between the TST and T-SPOT.TB tests, using kappa-quantified statistics, was poor (K =.3). Test results were discordant in 14 persons (28%) (both tests were negative in 28 persons and positive in 8). Twelve persons had a positive TST (defined as > 5 mm), a negative T-SPOT.TB test, and a normal chest radiograph, while two persons had a negative TST, a positive T-SPOT.TB test, and an abnormal chest radiograph. The T-SPOT.TB test strongly correlated with risk factors for LTBI compared with the TST.

Ten of 12 patients with a positive TST/negative T-SPOT.TB and a normal chest radiograph received no treatment for LTBI. This group received anti-TNF therapy with a mean follow-up of 76 weeks, with no evidence of reactivation TB. In contrast, one of 10 patients with a +T-SPOT.TB test developed miliary TB while receiving treatment for LTBI and anti-TNF-alpha therapy. Remarkably, 68% of the patients had received prior immunosuppressive therapy, which, while not further specified, raised the possibility of selection of patients at lower risk for developing reactivation TB.

Similar data were observed using the first-generation Quantiferon TB Gold test compared with TST in a group of 302 patients with inflammatory diseases (ie., rheumatoid arthritis, ankylosing spondylitis, and psoriatic arthritis) who were candidates for anti-TNF-alpha therapy.2 Sixty percent of the patients were female, with an average age of 50 years. In those with available histories, 152 of 200 reported prior BCG vaccination (76%), 9% reported prior exposure to TB, 8% were from countries endemic for TB, 4% had abnormal chest radiographs consistent with previous granulomatous infection, and 3% had received prior TB treatment. In all, 69 (29%) of the patients had one or more risk factor for LTBI. Using Danish guidelines, 45/241 (19%) patients had a positive TST (> 6 mm, or > 12 mm for those with prior BCG vaccination), while using U.S. Guidelines, 66 (28%) had a positive TST. In comparison, the Quantiferon was positive in 7%, negative in 88%, and indeterminate in 5%. Higher CD4 counts in patients correlated with a higher level of interferon-gamma production and greater likelihood of indeterminate quantiferon test results. In addition, corticosteroid therapy increased the number of indeterminate test results, while decreasing the sensitivity of the TST.

Using kappa-quantified statistics results, the agreement between the two tests was only K = .2. The Quantiferon TB Gold was significantly associated with risk factors for TB (RR 4.7, p = .002), especially prior residence in a TB endemic area (RR 7.8, p > .0001). Interestingly, 18 of 45 had +TST negative Quantiferon TB Gold test. Of those 18 with a positive Quantiferon results, only nine (50%) had a positive TST. Thus, the authors postulate that by using the TST alone significantly more patients would have received unnecessary treatment for LTBI, especially based on U.S. guidelines, but 50% of those at risk for reactivation LTBI would have been missed. Based on the lower sensitivity of the TST in this group compared with the Quantiferon TB Gold, and despite the discordant results, the authors advocate that both tests should be taken into consideration when screening for LTBI in patients with inflammatory diseases.


  1. Laffitte E., et al. Tuberculosis screening in patients with psoriasis before antitumor necrosis factor therapy: Comparison of an interferon-gamma release assay vs tuberculin skin test. Brit J Dermatol. 2009;1-4.
  2. Soborg B, et al. Comparison of screening procedures for Mycobacterium tuberculosis infection among patients with inflammatory diseases. J Rheumatol. 2009;36:1-9.