Prebiologic Therapy Tuberculosis Screening Experience in a Pediatric Rheumatology Center: TST and IGRA Are Both Necessary

    loading  Checking for direct PDF access through Ovid


We read with interest the article by Lowenthal and the recent related reply letter by Tebruegge. Lowenthal et al1 compared the results of tuberculosis (TB) screening tools [tuberculin skin test (TST) and interferon-gamma release assay (IGRA)] in immigrant children before and after their arrival in California, USA. Because of the high number of TST+/IGRA− patients, the authors suggest for preimmigration screening IGRA instead of TST. More recently, Tebruegge et al2 reported that a substantial proportion of TST+/IGRA− discordant children are likely to be TB-infected on the basis of their cytokine biomarker profiles,3 thus, suggesting the usefulness of using both tests.
We also would like to provide our experience of a tertiary pediatric rheumatology center in screening our patients with juvenile idiopathic arthritis (JIA) for TB infection before starting anti–tumor necrosis factor-alpha (TNFα) treatment. Current rheumatologic treatment guidelines recommend screening for latent TB infection (LTBI) before starting anti-TNFα treatment; however, the most appropriate screening test remains debatable.4
Among 120 JIA patients screened between January 2013 and April 2016, using TST and IGRA (QuantiFERON-TB GOLD) combined, 6 patients (5%, 3 female and 3 male) had at least 1 positive TB screening methods, and they were diagnosed to have LTBI (Table, Supplemental Digital Content 1,, which illustrates epidemiology data and features of 6 children with LTBI among 120 JIA children screened before starting anti-TNFα treatment); thus, they received 3 months of treatment with isoniazid and rifampicin. All of them had negative chest radiographs. The median age was 8.5 years (range 2–17.5 years). A possible Mycobacterium tuberculosis exposure history has been detected in 2 cases. The biologic therapy was deferred 1 month after anti-TB chemoprophylaxis onset, and LTBI treatment was well tolerated. The median follow-up was 28 months, and during this period, no children developed active TB disease.
TST was positive in 5/6 patients; 3 of them were IGRA negative. IGRA was positive in 3/6 patients, 1 of them (TST negative) remaining IGRA positive when retested after 2 months. Therefore, only 2 patients had both TB tests positive. All patients with positive IGRA became negative after 3 months of anti-TB chemoprophylaxis.
The total concordance between TST and IGRA was 0.933; the Cohen kappa coefficient was 0.299, indicating only a modest agreement.
Although just 13% (16 subjects) of the entire cohort were not of Italian origin, almost all LTBI patients (5/6) were foreign (Fischer exact test, χ2 26.7, P < 0001).
In addition, 4 other patients had indeterminate IGRA with negative TST, but interferon-gamma release assay repeated after 1 month resulted negative. As previously reported in literature, the median age of these patients showed a lower age trend than that in children with positive IGRA (median age 5.4 years; range 2.17–7.5 vs 13 years; range 11.08–14.75); this was not unexpected because performance of IGRAs shows a lower efficiency in young children.5
Given the lack of evidence and in agreement with Tebruegge et al and other authors,6,7 we suggest performing both TST and IGRA before starting a biologic treatment to minimize the risk of developing an active TB infection while receiving biologic treatment. Particular attention should be paid to immigrants because in our data set, 31% (5/16) of them presented with LTBI.

Related Topics

    loading  Loading Related Articles