Performance evaluation of an in-house interferon-gamma release assay in the diagnosis and follow-up of active tuberculosis disease

Abstract number: P1839

Alagna L., Fortis C., Galli L., Scarpellini P.

Objectives: To evaluate the performance of the interferon-gamma release assays (IGRA) for the diagnosis of the active tuberculosis disease (TD) and to describe the changes of the IGRA during the follow-up.

Methods: 238 patients (pts)[M = 140, F = 90] have been tested with the IGRA because suspected of active TD.187 out of 238 were tested with the tubercoline skin test (TST). Pts were classified according to the American Thoracic Society (ATS) classification. The home-made IGRA detects the number of specific g interferon producing T cells by means of an enzyme-linked immunospot assay, using a restricted pool of synthetic highly immunogenic peptides derived from ESAT-6 and CFP-10 proteins. For the evaluation of the IGRA follow-up in pts with active TD (15 pts), we considered this timeline: T0=mean of the IGRA values up to one week after the start of the therapy, T1=second and third months after the start of the therapy and T2=end of the therapy. The TST and the clinical classification were taken as reference when compared to IGRA. McNemar test and k statistics were calculated. Linear regression was applied.

Results: Among pts with active TD (ATS3) no agreement was detected between the IGRA and TST (k = 0.0767, 95% CI [0.126–0.280]); sensibility (SS)=84%, specificity (SP)=24%, Negative Predictive Value (NPV)=50% and Positive Predictive Value (PPV)=61%. There was a high number of pts with positive IGRA and negative TST [23/30 (77%), p = 0.0013]. We analysed the performance of both tests in relation to the clinical classification (ATS3 vs non ATS3): TST, SS=59%, SP=70%, PPV=55% and NPV=72% and no agreement was found (k = 0.278, 95% CI [0.139–0.418]). IGRA, SS=82%, SP=62%, PPV=54% and NPV=86%; agreement was fair (k = 0.4015 [0.293–0.509]). During follow up, significant decrease of IGRA quantitative values was observed [median slope = -2.98 spot forming cells per million lymphocytes per week, p = 0.0215 (SFCML)]. 13 out of 15 pts show a decrease of values (median -4.23 SFCML, p = 0.021).

Conclusions: Our results show a different performance of TST and the IGRA in pts with active TD. In particular, given the high SS, IGRA has a better performance than TST. Regarding the follow up, the decline in the IGRA values during the therapy is significant; further studies can help to evaluate if this decline could predict a favourable outcome.

Session Details

Date: 10/04/2010
Time: 00:00-00:00
Session name: Abstracts 20th European Congress of Clinical Microbiology and Infectious Diseases
Location: Vienna, Austria, 10 - 13 April 2010
Presentation type:
Back to top