Arthritis & Rheumatism, Volume 60,
October 2009 Abstract Supplement

The 2009 ACR/ARHP Annual Scientific Meeting
Philadelphia October 16-21, 2009.

Usefulness of in Vitro Interferon- Release Assays (IGRAS) for Diagnosis of Latent Tuberculosis Infection in Rheumatic Patients Scheduled for Anti-TNF- Treatment

Mateo1,  Lourdes, Minguez2,  Sonia, Latorre1,  Irene, Dominguez1,  José, Molto3,  Anna, Garcia2,  Emma, Grados1,  Dolors

Hospital Germans Trias i Pujol, Badalona, Spain
Hospital Universitari Germans Trias i Pujol, Badalona, Spain
M.D., Resident., Badalona, Spain


Detection of latent tuberculosis infection (LTBI) is mandatory in rheumatic patients candidates for treatment with anti-TNF-a agents. Although there is not a "gold standard" for diagnosis of LTBI, tuberculin skin test (TST) is the standardized method worldwide. Spanish Society of Rheumatology recommends the performance of a second-step TST, 1–2 weeks after the first one, and a chest radiograph. However, the existence of false-positive (BCG vaccination, nontuberculous mycobacterias) and false-negative (anergy, immunosuppression) TST results, has led to the development of new in vitro techniques in order to improve the sensitivity and specificity in LTBI detection.

Our aim is determining the usefulness of blood interferon-g release assays (IGRAS) and to compare them with TST in patients with different inflammatory rheumathic diseases scheduled for anti-TNF treatment.


The study included individuals with inflammatory rheumatism staring an anti-TNF-a agent. All patients underwent 2 TST, a chest radiograph and 2 IGRAS: One assay, (QuantiFERON GOLD in tube) measures IFN-g concentration in supernatant by enzyme-linked immunosorbent assay in response to specific antigens (ESAT-6, CFP-10 and TB7.7); the other enumerates IFN-g-secreting T cells by enzyme-linked immunospot (T-SPOT.TB). We performed a descriptive statistical study and calculated concordance between TST/T-SPOT.TB, TST/QFT and T-SPOT.TB/QFT by Cohen's kappa test, with k value > 0.75 representing excellent agreement, 0.4–0.75 good-moderate and <0.40 poor agreement.


We included 23 patients (10 males/13 females) with a mean age of 52.4 ± 13.2 years and different rheumatic diseases: 7 rheumatoid arthritis (RA), 4 ankylosing spondylitis, 5 psoriatic arthritis, 3 seronegative arthritis, 2 systemic lupus erythematous, 1 undifferentiated spondyloarthropaty and 1 SAPHO. Average time of disease evolution was 9.6 ± 9 years. All patients with RA had seropositive and erosive forms. Prior to anti-TNF-a treatment, 16 patients had received 2 DMARD (range 0–5) and 10 (43.5%) systemic glucocorticoids. 3/23 (13%) patients displayed positive TST. Only one had positive second-step TST after being negative the first one. T-SPOT.TB was positive in 7/23 (30.4%) and QFT in 5/23 (21.7%). QFT was undetermined in 3 cases (13%), while none for T-SPOT.TB. Level of agreement (kappa test) between TST/T-SPOT.TB was 0.51; between TST/QFT 0.46 and between T-SPOT.TB/QFT 0.79. Five patients (21.7%) received secondary prophylaxis for tuberculosis, 2 of them with negative TST and positive IGRAS.


1) IFN-g release assays are more accurate methods for diagnosis of latent tuberculosis infection than tuberculin skin test in rheumatic patients, even using a second-step TST. 2) Correlation between IGRAS and TST is moderate in immunocompromised individuals. 3) T-SPOT.TB seems to have greater sensitivity and lower rate of indeterminate results than QFT.

To cite this abstract, please use the following information:
Mateo, Lourdes, Minguez, Sonia, Latorre, Irene, Dominguez, José, Molto, Anna, Garcia, Emma, et al; Usefulness of in Vitro Interferon- Release Assays (IGRAS) for Diagnosis of Latent Tuberculosis Infection in Rheumatic Patients Scheduled for Anti-TNF- Treatment [abstract]. Arthritis Rheum 2009;60 Suppl 10 :996
DOI: 10.1002/art.26073

Abstract Supplement

Meeting Menu