Article Text

OP0066 Can We Really Identify Latent Tuberculosis? Comparison of the Performance of Tuberculin Skin Test and Interferon Gamma Release Assay before Anti-TNF Therapy: A Systematic Review and Meta-Analysis
  1. M. Locci1,
  2. F. Erny1,2,
  3. F. Barchechat-Flaisler1,
  4. J.-L. Leroux1,
  5. C. Gaujoux Viala1,2
  1. 1Rheumatology, CHU Nimes, Nimes
  2. 2EA 2415, Montpellier I University, Nîmes, France


Background Current guidelines recommend screening for latent tuberculosis infection prior the introduction of anti-tumor necrosis factor therapy. However, many patients are already taking immunosuppressive therapy, which can affect current diagnostic tests.

Objectives Due to the recent introduction of the interferon-gamma release assays (IGRA), we sought to assess their ability to detect latent tuberculosis infections in patients before biological treatment and the impact of immunosuppressive therapy on these new tests.

Methods MEDLINE, EMBASE and COCHRANE were searched (up to April 2013) to identify studies comparing the performance of interferon-gamma release assays (QuantiFERON QTF, T-SPOT.TB) to Tuberculin Skin Test (TST) in candidates for anti TNF treatment, with inflammatory bowel disease or rheumatic disease. The agreement between different tests was described by kappas. In each trial, the odds ratios (OR) were determined to assess the impact of treatment effect on the tests positivity. Where possible, pooled OR were calculated by meta-analysis methodology, using the Mantel-Haenszel method. In case of statistical heterogeneity,all meta-analyses were carried out using random-effects models.

Results On 533 studies, 45 studies (n=9226 patients) were included for analysis. The agreement between IDR/IGRA was poor. Kappas between QTF and TST were calculated in 18 studies, ranging from 0.03 to 0.52: poor in 15 studies (<0.4) and moderate in 3 studies (0.4-0.6). Kappas between T-SPOT and TST were calculated in 7 studies, ranging from 0.13 to 0.40: poor in 6 studies and moderate in 1 study. Moreover, the agreement between the two IGRA was moderate. Kappas ranged from 0.28 to 0.71 in 8 studies: poor in 1 study, moderate in 5 studies and good in only 2 studies.

Immunosuppressive therapy did not significantly influence positive QTF results (7 studies-pooled OR 0.86, 95% CI 0.64-1.15) and positive T-SPOT results (4 studies-pooled OR 0.89, IC 95% 0.63-1.26). In contrast, the corticotherapy significantly reduced the rate of T-SPOT + (2 studies- pooled OR: 0.45, IC 95% 0.23-0.89) and almost significantly reduced the rate of QTF + (3 studies- pooled OR:0.63, IC 95% 0.38-1.02).

Conclusions The screening of latent tuberculosis raises concerns, especially due to the lack of agreement between IGRA. The impact of immunosuppressive therapy on IGRA was not significant whereas corticotherapy significantly decreased the rate of IGRA positivity.

Disclosure of Interest None declared

DOI 10.1136/annrheumdis-2014-eular.5452

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.