YIELD OF SERIAL TESTING FOR TUBERCULOSIS EXPOSURE IN THE STUDY OF A PROSPECTIVE ADULT RESEARCH COHORT WITH INFLAMMATORY BOWEL DISEASES (SPARC IBD)

Abstract

Abstract BACKGROUND More than 2 million Americans have Inflammatory Bowel Diseases (IBD), including Crohn’s Disease or Ulcerative Colitis. Although tuberculosis is uncommon in the United States, patients with IBD are at an increased risk for active tuberculosis infection due to the immunosuppressant therapies used to treat IBD. Due to this risk, testing for latent tuberculosis is recommended prior to and intermittently throughout administration of most IBD medications. The utility of this practice in a country with a low incidence and prevalence of tuberculosis is unknown. We aimed to determine the proportion of initial and subsequent tuberculosis tests that are positive, negative, or indeterminate. METHODS Data were acquired from SPARC IBD, a component of the Crohn’s & Colitis Foundation’s IBD Plexus research platform. SPARC IBD is a prospective cohort study conducted at 18 centers throughout the USA and includes more than 4,000 patients. Data are collected from SPARC IBD case report forms and the IBD Smartform, diagnoses, test results and prescriptions within the linked electronic medical records. Patients with one or more interferon gamma release assay (IGRA) test for latent tuberculosis were included in this study. The outcome investigated was a positive, negative, or indeterminate lab result. The proportion of positive and indeterminate tests were computed with 95% confidence intervals (CI). Fisher’s exact test was used to detect a change in positive test proportions following a negative test compared to initial lab results. RESULTS A total of 687 patients with one or more IGRA test were identified. The median age was 37 (IQR 27-50) with 54% female patients. Of the initial lab results, 644 tests were negative (93.7%, 95% CI, 91.7, 95.4), 37 were indeterminate (5.4%, 95% CI, 3.8, 7.3), and six were positive (0.9%, 95% CI, 0.3, 1.9). Following an initial negative result, 269 patients received a second test, of which five were positive (1.9%, 95% CI, 0.6, 4.3). Following two consecutive negative results, 142 patients received an additional IGRA test, of which one was positive (0.7%, 95% CI, 0.02, 3.9). The proportion of positive tests was not significantly lower following one negative (p = 0.20) or two consecutive negative tests (p > 0.99). POTENTIAL LIMITATIONS Many patients only had one test for latent tuberculosis, and it is unclear if the initial IGRA labs in our dataset were their first latent tuberculosis tests. CONCLUSIONS Based on the findings of this study, there appears to be value in continued serial testing following a negative result even in the United States, despite the overall low incidence of tuberculosis.