In general, in response to infectious pathogens and vaccines, the T cell response of young children may demonstrate decreased magnitude and slower kinetics than that observed in older children and adults, and it is more likely to represent TH2-type T cell immunity 
. More specifically regarding Mtb antigens, Kampmann and colleagues showed that IFN-γ production in response to Mtb antigens was lower in young children (<5 years old) than in older children (5–15 years old) 
. However in this cohort of household contacts of culture-positive pulmonary tuberculosis patients from a TB endemic area, we observed that young children mounted robust IFN-γ responses generally comparable to adults. Our study provides the first direct comparison of adults and children from the same household contact setting. Moreover, adequate numbers of individuals of various ages were enrolled to allow statistically robust comparisons. Therefore, we conclude that in these healthy young children from a TB endemic area, who do not develop active TB following household exposure to Mtb, young age itself did not hamper the development of adult-like IFN-γ responses to Mtb antigens.
Prompt, accurate identification of infants and young children with Mtb infection is an important priority as they are at higher risk of progression to disease and of developing more severe disease than older children and adults 
. Until recently, diagnosis of LTBI has relied on the tuberculin skin test (TST). However, development of interferon-gamma (IFN-γ) release assays (IGRAs) offer new ways to make the diagnosis of LTBI. One type of IGRA is the whole blood assay, which detects Mtb-antigen induced IFN-γ expression by T cells in whole blood using an ELISA (QuantiFERON®-TB Gold and QuantiFERON®- Gold IT [Cellestis, Corp]). A second type of IGRA is based on the ELISPOT assay which detects the frequency of IFN-γ secreting cells in response to Mtb antigens (T-spot.TB
test [Oxford Immunotec, Inc.]). Although these assays are technical and expensive, there is interest in developing them further for use in resource-limited settings where TB is endemic. The usefulness of IGRAs in making the diagnosis of Mtb infection in children is not yet firmly established and whether or not age itself, due to a decreased capacity to produce IFN-γ, decreases the sensitivity of IGRAs remains controversial. For example, the QuantiFERON®-TB demonstrated poor sensitivity for diagnosis of LTBI in Australian children, and negative results correlated with young age 
. Furthermore, a study of IGRAs in a diverse clinical population showed that indeterminate results of the QuantiFERON Gold test were disproportionately high in children less than 5 years of age 
. However, one study of South African children, found that ELISPOT was more sensitive than TST in diagnosing Mtb infection with children <3 years of with probable or confirmed TB 
and a study of European children showed that both QuantiFERON Gold and T-spot.TB
were as sensitive as TST in detecting Mtb infection in children with culture confirmed TB 
. One implication of our results for the potential utility of IGRAs in children is that the results of this large study of household contacts from a TB endemic, where the immune responses of children of all ages and adults were directly compared, indicates that IFN-γ responses in healthy individuals were equivalent and robust, and hence the sensitivity of IGRAs should not be compromised by young age alone.
IFN-γ responses in young children, as in adults, correlated with the TST and known risk factors for infection. Our results are consistent with those from child household contact studies using an ELISPOT assay 
or a whole blood assay 
where IFN-γ responses were also correlated with TST results and risk factors for infection. The concordance rates between commercial IGRAs and TST have been studied in several studies of children and vary widely from concordance rates of 26% to 95% (κ coefficients
0.08–0.73) depending upon the clinical setting 
. We found a concordance rate (72%) which falls within this broad range. Moreover, in our study, we directly compare concordance between young children and adults in the household contact setting and found them to be similar.
Our analysis of children enrolled in the first two years of the first household contact study 
as well as the current analysis of children enrolled over both studies presented here suggest that, in a setting of high exposure to Mtb within an endemic community, prior BCG vaccination does not affect TST results. Moreover, we conclude that prior BCG vaccination does not affect IFN-γ results. This is different than the results of Soysal et al 
who reported that their in-house IFN-γ ELISPOT assay was negative more often in BCG immunized than in unimmunized children. It is possible that differences in methodology could account for the discrepancy in the results between these two studies in that in the Soysal et al study, an ELISPOT assay using short-term overnight incubation and Mtb specific antigens, ESAT-6 and CFP-10 were used while in our study a longer term (5–7 day) whole blood assay with Mtb culture filtrate antigens were used.
The conclusions of our study of healthy young Ugandan household contacts should not be extrapolated to young children immunocompromised by conditions other than age such as HIV. Future studies, including those using commercially available IGRAs, should continue to elucidate the interaction between the young host, Mtb and distinct immunocompromising conditions. In addition, while young children are capable of mounting strong IFN-γ responses to Mtb antigens, it remains to be determined whether or not difference in other aspects of the T cell response to Mtb could predispose children to development of active TB, such as excessive Th2, Th17, and/or T regulatory responses and/or deficient CD8+ T cell responses. Alternatively, perhaps differences in innate, rather than adaptive immunity could be responsible for these differences in disease susceptibility, all areas which will require addition study.