P487 A Study of Optimal Screening for Latent Tuberculosis in Patients with Inflammatory Bowel Disease
T. Al-Taweel*, M. Strohl, T. Bessissow, A. Bitton, E. Seidman, W. Afif
McGill University Health Center, Division of Gastroenterology, Montreal, Canada
Reactivation of latent tuberculosis infection (LTBI) in patients with IBD, secondary to treatment with anti-tumour necrosis factor (TNF)- α agents, can lead to serious and life-threatening illness. No gold standard exists for the detection of LTBI and there is conflicting evidence within the inflammatory bowel disease (IBD) literature regarding the best screening strategy for LTBI. The aims of our study were to 1) assess whether the addition of an interferon-gamma release assays (IGRA) to a tuberculin skin tests (TST) would improve the detection of LTBI, 2) evaluate the concordance of the TST and IGRA and 3) assess the impact of various demographic, diagnostic and treatment factors on the each test.
Consecutive IBD patients being considered for anti-TNF- α treatment underwent testing with a TST, IGRA (QuantiFERON Gold In-Tube) and chest x-ray (CXR). All patients completed a self-administered questionnaire. The association of both tests with demographic factors, LTBI risk factors (including CXR), BCG vaccination and immunomodulator (IM) therapy were evaluated using Fisher's exact test, and agreement between the TST and IGRA was evaluated using the kappa statistic.
One hundred and fifty-five consecutive IBD patients (131 Crohn's disease, 22 ulcerative colitis and 2 indeterminate colitis) underwent testing with both TST and IGRA. Six patients had indeterminate IGRA results and were excluded from the analysis. Twenty-four (18%) patients had a history of BCG vaccination. Twenty-eight patients (19%) had at least one risk factor for LTBI, including four (3%) with an abnormal CXR. One hundred and two patients (70%) were taking immunosuppressive therapy at time of inclusion. Nine patients were TST positive (6%) and 7 patients (5%) were IGRA positive. Concordance between TST and IGRA was 91.9% (but κ = 0.21), with only 2 patients being positive for both tests. Neither test was affected by age, gender or BCG vaccination. In bivariate analysis, the presence of risk factors for LTBI was found to positively influence TST results (OR 19.8, 3.9-102.1), but not IGRA. IGRA was negatively influenced by IM therapy (OR 0.06, 0.007-0.5), but not TST. Four of the five patients who were IGRA positive but TST negative were treated or advised to begin treatment for LTBI by a respirologist.
IGRA was negatively influenced by IM therapy, while the presence of risk factors for LTBI was found to positively influence TST results. There is fair agreement between positive TST and IGRA results. The addition of IGRA to the standard practice of TST and CXR increased the number of cases presumed to have LTBI and influenced management. Given these findings, dual testing with TST and IGRA should be considered in IBD patients.