Performance of Interferon‐gamma Release Assays for Tuberculosis Screening in Pediatric Inflammatory Bowel Disease

To the Editor: We appreciate the letter to the editor authored by Drs. Joob and Wiwanitkit comparing interferon-gamma release assays (IGRAs) to the tuberculin skin test (TST). The best measure of a screening tool is its sensitivity, and the IGRA sensitivity in our study was 83%. Its negative predictive value, moreover, was 99.9%. Multiple pediatric studies with a larger number of patients positive for tuberculosis have demonstrated IGRAs to have a higher sensitivity than TST (1,2). TSTs suffer from the subjectivity of the individual interpreting the skin reaction. Patients who are immunosuppressed and children under 4 years old are less likely to have a robust reaction to TST, and the traditional criteria of 15 mm induration can have a sensitivity of less than 70% (3). The positive predictive value of IGRs is not reflective of its utility as a screening test given the relatively low prevalence of LTBI. Our 4 false-positive results over a 7-year period were easily clarified by repeat IGRA testing that was negative. TST can have a high false-positive rate in patients who have received the Bacille Calmette-Guérin vaccine, and patients who have been infected with nontuberculous mycobacterial infections. Importantly, we found that prior TST results are inconsistently and poorly documented in the results section of electronic health records (EHRs) compared with IGRA, and if found in notes rarely specify the size of induration. This is significant in the era of electronic health records whereby retrieving results is an important feature and results need to be retrieved by treating providers, infusion centers, case managers, and nurses. Although IGRA serology testing has a slightly higher laboratory cost, TST has higher costs associated with nursing administration and reading. They require 2 clinic visits, assuming patient compliance, and sometimes 4 visits to increase sensitivity for older infections via "2-step testing." There are additional unmeasured costs to the patient through missed time at school or work, transportation, etc. In a recent study, IGRAs were found to be more cost-effective than TST in healthcare workers (4). We suggest that screening IBD patients is similar to screening healthcare workers, and many of the previous studies evaluating cost effectiveness used immigrants or close contacts to known TB patients, which are much different patient populations than IBD patients. We recognize that IGRA testing is not readily accessible in all clinics throughout the world. In these cases, TST can be an effective screening tool, and should be used for routine screening in pediatric patients with IBD receiving anti-tumor necrosis factor (TNF) alpha therapy. However, we suggest that if IGRA testing is available it should be considered as a replacement to TST because of its cost-effectiveness and ease of retrieving in the electronic health records.