Z Gastroenterol 2014; 52 - A16
DOI: 10.1055/s-0034-1376076

Is the tuberculin skin test alone accurate in moderate-to-severe BCG vaccinated patients with inflammatory bowel disease treated with immunosuppressives to test for latent tuberculosis?

K Gecse 1, Z Kürti 1, B Lovász 1, Á Szabó 1, M Mandel 1, A Gyurcsányi 1, K Kristóf 2, Z Végh 1, A Mohás 1, B Csákó 1, L Kiss 1, P Golovics 1, P Lakatos 1
  • 11st Department of Medicine, Semmelweis University, Budapest, Hungary
  • 2Institute of Medical Microbiology, Semmelweis University, Budapest, Hungary

Background and Aims: There are few data available on effect of immunomodulator/biological therapy on the accuracy of tuberculin skin test (TST, Mantoux skin test) and interferon-gamma release assay (IGRA) in BCG vaccinated immunosuppressed IBD patients. Our aim was to define the accuracy of the TST and IGRA tests in a BCG vaccinated referral IBD cohort treated with immunosuppressives and/or biologicals.

Patients and Methods: Data of 90 consecutive moderate-to-severe IBD (77 CD, 13 UC) patients were analyzed (male/female: 42/48, median age at diagnosis: 23.0; SD: 10.02 years, duration: 7.0; SD:6.1 years). Patients were treated with immunosuppressives (azathioprine, steroids) and/or anti-TNF therapy. Blood samples for IGRA were collected during routine laboratory testing parallel with TST. The result of TST was determined according to international guidelines. Both in- and outpatient records were collected and comprehensively reviewed.

Results: TST positivity rate was 25.8%, 23.6%, 14.6% or 13.5% with cut-off values of 5, 10, 15 and 20 mm. IGRA positivity rate was 8.1% with indeterminate result in 1.2%. The correlation between TST and IGRA was significant, with moderate-to-good kappa values if TST results were > 15 mm (kappa: 0.41 – 0.45, p < 0.001). If TST 15 or 20 mm is defined as TST positivity an additional 10.3% and 9% of IBD patients required a pulmonologists consultation. There was no association between the type and number of immunomodulators used or any disease phenotype characteristics and the TST or IGRA results. Importantly, smoking was identified as a risk factors for TST but not IGRA positivity (OR: 4.35, 4.15 and 4.92, p = 0.012, p = 0.029 and 0.018 for TSTcut-off 10, 15 and 20 mm).

Conclusions: The TST and IGRA are partly complimentary methods and accuracy is acceptable also in BCG vaccinated and immunosuppressed IBD patients. A TST of > 15 mm should be used as a cut-off to identify patients at risk for latent TB in these patients. Smoking is a risk factor for TST positivity.