Re: “Analysis and Reporting of Tests for Tuberculosis Infection in Childhood Contacts of Infectious Source Cases”

To the Editors: We thank Drs. Kitai, Rea and Rose for their interest in our work.1,2 We agree that children referred after initial screening in the community must not be included when calculating rates of test positivity in various contact groups. As stated in the Materials and Methods section, as a footnote to Table 1, and shown in the Figure (Supplemental Digital Content 3, https://links.lww.com/INF/D401), we considered the 216 selectively referred children not evaluable in this regard. For the analysis of test concordance (κ coefficients) however, we utilized all children with known Calmette-Guérin Bacillus status having both tests (Supplemental Digital Content 4, https://links.lww.com/INF/D400). We considered referral bias as being relatively unimportant for this analysis. Admittedly, some of the tuberculin skin test (TST)-negative children who were not referred to our clinic may have been interferon-gamma release assay (IGRA)-positive if tested. However, TST-positive/IGRA-negative discordance is more common. Besides, a perhaps more important source of bias is the fact that IGRA confirmation of a positive TST was initially sought only when we considered infection uncertain. Counting all 254 children with both tests, we found a positive IGRA in 6/55 (11%) TST-negative children and a negative IGRA in 135/199 (68%) TST-positive children. We do point out that we recommend the confirmation of a negative IGRA with a TST in selected cases. There were 105 children exposed in schools or day care, who were not selectively referred but had all their tests at our clinic and were hence assessed as evaluable. Of these, 15 were included among the household contacts because of several hours of daily, close contact with the index case. The purpose of the present article was to compare results using the 2 tests. We are now preparing a report that includes the crosstabulations mentioned by Kitai, Rea and Rose. Rutger Bennet, MDSahar Nejat, MDMargareta Eriksson, MDAstrid Lindgren Children’s HospitalKarolinska University HospitalStockholm, Sweden