Portal de Periódicos da CAPES

Indeterminate Interferon-γ Release Assay Results in Children

2010; Lippincott Williams & Wilkins; Volume: 29; Issue: 3 Linguagem: Inglês

10.1097/inf.0b013e3181c4822f

1532-0987

T. G. Connell, Marc Tebruegge, Nicole Ritz, Penelope A. Bryant, David Leslie, Nigel Curtis,

Pneumonia and Respiratory Infections

To the Editors: We read with interest the recent report by Haustein et al entitled "The likelihood of an indeterminate test result from a whole-blood interferon-γ release assay for the diagnosis of Mycobacterium tuberculosis infection in children correlates with age and immune status."1 This report adds to recent publications that question the performance of current interferon-γ (IFN-γ) release assays (IGRA) for the diagnosis of tuberculosis (TB) in routine pediatric clinical practice.2–6 The retrospective study by Haustein et al in 237 children highlights the high proportion of indeterminate test results obtained with the QuantiFERON-TB (QFT) Gold In-Tube assay (35% of the study population). Notably, indeterminate test results were over-represented in children younger than 5 years of age, and those with immunodeficiencies or medical conditions associated with immunosuppression. Importantly, these groups represent children most at risk for disease progression after exposure to M. tuberculosis. Although current Australian guidelines do not recommend the routine use of IGRA in children with suspected TB disease or latent TB infection, a QFT assay (either the second generation Gold assay or third generation Gold In-Tube assay) is often requested by clinicians as an adjunctive test. In light of the findings by Haustein et al we reviewed the results of all QFT assays from children attending the Royal Children's Hospital Melbourne over a 5-year period (November 2003 to November 2008) to specifically determine the influence of age on assay performance. The population of children tested in our setting is somewhat different from that studied by Haustein et al as the prevalence of HIV infection in Australia in children is low (estimated prevalence: 2/100,000 children aged between 0 and 18 years). All samples were analyzed by the Victorian Infectious Diseases Reference Laboratory that has extensive experience in performing QFT assays, processing more than 4000 assays per year. All assays were done according to manufacturer's instructions. Results from 875 QFT assays (449 [51%] QFT Gold and 426 [49%] QFT Gold In-Tube) performed on 783 children (median age: 9.1 years, range: 25 days to 18.0 years) were available for analysis. Of the 875 QFT assays, 113 (13%) were indeterminate. Of these, the majority (89/113 [79%]) failed as a result of an inadequate response in the positive (phytohemagglutinin (PHA) mitogen) control, the remainder (24/113 [21%]) being deemed indeterminate as a result of a high IFN-γ concentration in the negative (nil). An additional 14 samples had an inadequate PHA mitogen control response but the assays were classified as determinate on the basis of a positive response to M. tuberculosis antigen stimulation. In assessing the influence of age on assay performance we found striking parallels to the data presented by Haustein et al. Specifically, in our population the median age of children with an inadequate PHA mitogen control response was lower than those with an adequate PHA mitogen control response (5.4 years vs. 9.4 years, P < 0.0001). Correlating the magnitude of PHA mitogen control response with age in the QFT Gold In-Tube assay is problematic as, according to manufacturer's instructions, the upper limit of detection for the assay is 15 IU/mL. Censored data (Fig. 1A) show that the magnitude of the PHA mitogen control response correlated positively with age (Spearman correlation r = 0.18, P < 0.0001). A positive correlation was also found when the data was not censored. The median PHA mitogen control response was significantly lower in children younger than 5 years of age than in children older than 5 years (Fig. 1B). Finally, children younger than 5 years of age were more likely to have an indeterminate assay result than children older than 5 years (Fig. 1C).FIGURE 1.: A, Relationship between PHA mitogen (positive) control response (censored at 15 IU/mL) and age with fitted linear regression line; (B) PHA mitogen control responses stratified by age (bars represent median values; P values were calculated by Mann-Whitney U test); (C) Proportion of indeterminate QFT assay results (ie, assays with failed PHA mitogen control response and/or high nil (negative) control combined) stratified by age (χ2 test).Our data adds considerable weight to the suggestion by Haustein et al that the performance of QFT is compromised in young children. We concur with Haustein et al that the most likely explanation for the high rate of failed PHA mitogen control responses relates to the functionally immature immune system of infants and young children. Reduced IFN-γ production in response to mitogens in this age group compared with adults is well documented.7 It is therefore crucial that age-specific cut-offs are established and incorporated into assay interpretation guidelines. We have previously raised our concerns about the use of QFT and other IGRA for the diagnosis of TB in children that are shared by other experts in this field.2,3,5,8–10 In addition to the high proportion of indeterminate assay results, significant discordance between the results of tuberculin skin testing and IGRA in children has been reported.9,11 Exploring the immunology underlying this discordance may help in the development of more accurate and reliable immunodiagnostic tests for the diagnosis of TB in children. In addition, long-term follow-up studies are needed to determine the true predictive value of IGRA for the development of active TB disease. We believe that until such data are available, IGRA should not be used as replacement tests for the tuberculin skin testing in children. Thomas G. Connell, MRCPI Marc Tebruegge, MRCPCH, MD Nicole Ritz, MD Penelope A. Bryant, MRCPCH, MD, PhD Department of Paediatrics The University of Melbourne Infectious Diseases Unit Department of General Medicine Murdoch Children's Research Institute The Royal Children's Hospital Melbourne David Leslie, FRCPA The Victorian Infectious Diseases Reference Laboratory North Melbourne, Australia Nigel Curtis, FRCPCH, PhD Department of Paediatrics The University of Melbourne Infectious Diseases Unit Department of General Medicine Murdoch Children's Research Institute The Royal Children's Hospital Melbourne Parkville, Australia