All children with latent tuberculosis infection (LTBI) are recommended for treatment to prevent the risk of progression to disease.1,2 To screen for LTBI, quantiferon is increasingly used in children in low-prevalence settings.1,2 However, the quantiferon assay has high variability in adults: 20%–80% of borderline-positive quantiferon results (0.35–0.99 IU/mL) are negative when repeated.3,4 Adults with borderline-positive quantiferon have an extremely low risk of developing active tuberculosis (TB), suggesting false positive borderline quantiferon results.4,5 False positive results have high implications in children as treatment of LTBI includes potentially toxic drugs for 3–9 months.1,2 The significance of borderline-positive quantiferon in children is unexplored.
We present the results of a strategy with systematically repeating quantiferon in children with borderline-positive tests in a country with a low prevalence of TB.
MATERIALS AND METHODS
From March 2015, a follow-up quantiferon test was recommended for all children 0–17 years of age with borderline-positive quantiferon at the tertiary pediatric center, Copenhagen University Hospital, Denmark. This recommendation was made for all children regardless of the reason for testing. A borderline-positive result was defined in the range 0.35–0.99 IU/mL, as suggested by Jonsson et al.4 Children younger than 5 years with borderline-positive quantiferon were recommended both a quantiferon retest and a tuberculin skin test (TST). We prospectively followed all children with borderline quantiferon results in the period from March 2015 to February 2019. To ensure completeness, all quantiferon data from children, 0–17 years of age, at our center during the study period were retrospectively reviewed. Quantiferon-TB Gold Plus replaced Quantiferon-TB Gold from July 2017. The study was approved by the Data Protection Agency and waiver of informed consent was obtained (3-3013-2889/1).
Of a total of 647 quantiferon test taken, 27 (4%) were positive (>0.35 IU/mL), 51 (8%) were indeterminate and 569 (88%) were negative. Borderline-positive quantiferon accounted for 10 of 27 (37%) positive tests. All patients with borderline-positive quantiferon results had a confirmatory test performed. The 10 patients with borderline-positive results were tested due to TB exposure, adoption, immunosuppressive therapy or TB suspicious symptoms (Table 1). They were 1–13 years of age (median 8). Three children (30%) received methotrexate when the initial test was done. Nine of 10 (90%) with an initial borderline-positive quantiferon test had a negative follow-up sample after a median of 2 weeks (1–24) (case 1–9; Table 1). Children younger than 5 years had additionally negative TST. Patients with negative retest did not receive treatment for LTBI. One asymptomatic child, 11 years of age, with recent exposure remained quantiferon positive and received treatment for LTBI (case 10). At follow-up after a median 24 months (5–50), no patients had developed TB disease. None was lost to follow-up.
Of the 647 quantiferon results, 403 were from children screened routinely prior to initiation of biologic therapy. Of these, 5 (1.2%) did have borderline-positive tests (case 4–8) and none (0%) had quantiferon above 1 IU/mL.
Borderline-positive quantiferon was found in more than one-third of all positive tests. Ninety percent were negative when repeated. This points to high variability of interferon-γ response just above the cutoff level in children. The findings are consistent with studies in adults in whom 20%–80% of borderline-positive quantiferon results are negative when repeated.3,4 Test variability may be due to pre-analytic, technical and patient-related factors. True immunologic reversions are unlikely since reversions occurred after a median of 2 weeks in our cases, in contrast to the longer-lived TB-immune responses.
This is the first study evaluating quantiferon variability in children with borderline-positive results. All children with borderline-positive quantiferon and negative confirmatory tests, were regarded as having false positive results and did not receive treatment for LTBI. Children younger than 5 years additionally had negative TST. Further, 2 young children had long-term quantiferon follow-up after 6 months and 3 years, which remained negative, also pointing to false positive results. Children older than 5 years only had 1 confirmatory quantiferon test taken and no TST performed. They were regarded as having false positive results, in accordance with the newest guidelines from Infectious Diseases Society of America,1 but we cannot rule out the possibility of true positive borderline results followed by false negative confirmatory samples. However, if this was the situation for all our cases, the prevalence of LTBI would be unexpectedly high in ethnic Danish children screened routinely because of immunomodulating therapy based on the following assumptions: The incidence of LTBI infection in Danish children is not known, as no true gold standard for LTBI exists, but given an annual incidence of TB disease of 0.7 per 100,000 ethnic Danish children6 and assuming 5% of children with LTBI develop TB,7 the incidence of LTBI is expectedly 14 per 100,000 ethnic Danish children. In contrast, we found an incidence of positive quantiferon of 1000 per 100,000 Danes (4 of 403), that is, a 70-fold higher prevalence. Thus, the low a priori risk of LTBI in ethnic Danish children challenges the accuracy of the quantiferon test and suggests a low positive predictive value.
Although methotrexate has not been found to reduce antigen responses in adults,8 methotrexate treatment in 3 children may have resulted in false negative confirmatory quantiferon test, as a variety of other immunosuppressive agents has been found to reduce antigen responses in quantiferon assays, including corticosteroids and inhibitors of calcineurin and tumor necrosis factor-alpha.9,10 Careful interpretation of borderline-positive quantiferon is important in all children receiving immunosuppressive therapy due to their high risk of developing TB disease if misinterpreted. Preferably, these children should be screened before initiation of any immunosuppressive therapy.
Treating children for LTBI due to false positive results is not without consequences as it includes (1) daily medicine administration for 3–9 months, which may be a challenge in children, (2) outpatient consultations, (3) possible parental anxiety of a TB diagnosis and (4) frequent adverse effects such as nausea and vomiting. False positive results may also influence research in pediatric LTBI by inclusion of non-truly infected children. No pediatric guidelines include a “retesting zone” despite the significant test variability around the cutoff level.1,2 As suggested by centers for disease control and prevention and several others,4,11 a “retesting zone” could improve the use of quantiferon, particularly in low-prevalence settings. In children, 2 follow-up quantiferon tests, in addition to TST, could minimize the risk of misinterpretation of borderline-positive results. Still, children younger than 5 years with known recent exposure requires treatment, regardless of quantiferon result and not awaiting retesting, due to the high risk of disease progression before test conversion.1,2
The limitations of this study include (1) the small number of patients with borderline-positive quantiferon, (2) only one control quantiferon test in most children, (3) the potential influence of methotrexate on the quantiferon result in 3 patients and (4) the short follow-up of a median of 24 months. Therefore, this study is not robust enough to make firm conclusions of false positive quantiferon, only the presence of a high test variability in the borderline range and, in some children, a high suspicion of false positive results. Our population included primarily children without known TB exposure, hence the results cannot be generalized to populations from high-endemic settings. We suggest that borderline-positive quantiferon results in children from low-prevalence settings are repeated short- and long-term, in addition to evaluating TST, to further explore the hypothesis of high risk of false positive results.
1. Lewinsohn DM, Leonard MK, LoBue PA, et al. Official American Thoracic Society/Infectious Diseases Society of America/Centers for Disease Control and Prevention Clinical Practice Guidelines: Diagnosis of Tuberculosis
in Adults and Children. Clin Infect Dis. 2017;64:111–115.
3. Gamsky TE, Lum T, Hung-Fan M, et al. Cumulative false-positive QuantiFERON
-TB interferon-γ release assay results. Ann Am Thorac Soc. 2016;13:660–665.
4. Jonsson J, Westman A, Bruchfeld J, et al. A borderline range for Quantiferon
Gold In-Tube results. PLoS One. 2017;12:e0187313.
5. Winje BA, White R, Syre H, et al. Stratification by interferon-γ release assay level predicts risk of incident TB. Thorax. 2018 Apr 5. [Epub ahead of print].
6. Nordholm AC, Holm LL, Svensson E, et al. Tuberculosis
transmission in Danish children: a nationwide register-based study. Pediatr Infect Dis J. 2019;38:340–343.
7. Marais BJ, Gie RP, Schaaf HS, et al. The natural history of childhood intra-thoracic tuberculosis
: a critical review of literature from the pre-chemotherapy era. Int J Tuberc Lung Dis. 2004;8:392–402.
8. Arias-Guillén M, Sánchez Menéndez MM, Alperi M, et al. High rates of tuberculin skin test positivity due to methotrexate therapy: false positive results? Semin Arthritis Rheum. 2018;48:538–546.
9. Edwards A, Gao Y, Allan RN, et al. Corticosteroids and infliximab impair the performance of interferon-γ release assays used for diagnosis of latent tuberculosis
. Thorax. 2017;72:946–949.
10. Barton E, Gao Y, Ball D, et al. Calcineurin inhibitors and variation in the performance of interferon-gamma release assays used to detect TB infection. Ann Am Thorac Soc. 2019;16:771–775.
11. Mazurek GH, Jereb J, Vernon A, et al; IGRA Expert Committee; Centers for Disease Control and Prevention (CDC). Updated guidelines for using interferon gamma release assays to detect Mycobacterium tuberculosis
infection - United States, 2010. MMWR Recomm Rep. 2010;59RR-51–25.