Aim: To assess the likelihood of detecting latent tuberculosis infection [LTBI] by the positive conversion of a serial tuberculin skin test [TST] at 1 year in inflammatory bowel disease [IBD] patients with negative baseline two-step TST.
Methods: In this multicentre prospective cohort study, we evaluated rate and predictors of conversion of TST at 1 year in patients with negative baseline TST. We also evaluated management of patients who had a positive TST at baseline or a conversion at 1 year. In all patients we assessed TB cases occurring during follow-up.
Results: Of the 192 IBD patients receiving anti-tumour necrosis factor [TNF] and 220 IBD controls not receiving anti-TNF, 35 [8.5%, 95% CI 5.7-11.3] had positive conversion (median TST induration 13 mm, interquartile range [IQR] 9-16). Ten anti-TNF cohort patients [5.2%, 95% CI 2.5-9.5] versus 25 controls [11.4%, 95% CI 7.5-16.3] had TST conversion [p = 0.029]. In multivariate analysis, conversion was associated with smoking habit (odds ratio [OR] 2.19, 95% CI 1.08-3.97; p = 0.028). Anti-TNF-treated patients had a lower conversion rate [OR 0.41, 95% CI 0.20-0.83; p = 0.013]. The likelihood of conversion correlates with fewer immunosuppressive therapies between baseline TST and TST at 1 year [p = 0.042]. One case of active TB [isoniazid-resistant strain] occurred in a patient with positive baseline TST receiving anti-TNF [0.05 events/100 patient-years].
Conclusions: Serial TST at 1 year can detect LTBI in IBD patients receiving anti-TNF therapy with negative baseline TST. Serial TST seems to be advisable to reduce the risk of TB cases associated with inability to detect LTBI in pre-treatment screening.