[Reproducibility of the QFT-G-IT assay in the contact examination in which a high positive rate was shown independent of contact time].

OBJECTIVE A contact examination conducted using the QuantiFERON-TB GOLD In-Tube (QFT-G-IT) assay shows a high positive rate independent of the closeness of contact. Among individuals with a short contact time of only 1.1 +/- 1.3 h with the index tuberculosis (TB) case, the QFT-G-IT assay positive rate was 13.9%, which was considered extraordinarily high (for reference, the proportion of previously infected individuals in the Japanese general population with the same age has been estimated to be approximately 7.3%). We retested the QFT-G-IT-positive contacts to examine the reproducibility of the QFT-G-IT assay and to confirm the reliability of the test results. METHOD Of the 216 participants who underwent the first examination, 33 who tested positive (23 close contacts and 10 casual contacts) were retested by an experienced technician at the same laboratory; care was taken to minimize possible causes of variations (method of shaking tubes, temperature, vibration during transportation of the specimens, etc.). In addition, the participants were tested using other interferon (IFN)-gamma release assays (IGRAs), namely, QuantiFERON-TB GOLD (QFT-G) and T-SPOT, in order to confirm the reliability of the QFT-G-IT assay. RESULT Among the 33 retested participants, only 7 (4 close contacts and 3 casual contacts) tested positive, and the remaining 26 participants (19 close contacts and 7 casual contacts) had discordant results. Out of the 11 participants (33.3%) in whom IFN-gamma levels varied between 0.35 and 0.50 IU/ml (just above the diagnostic cut-off) in the initial test, 10 (90.9%) tested negative, with IFN- gamma levels being less than 0.35 IU/ml in the retest. The results of the different IGRAs showed moderate to high agreement, with kappa values of 0.406 between QFT-G-IT and QFT-G and 0.604 between QFT-G-IT and T-SPOT. CONCLUSION The findings of this study showed that the QFT-G-IT assay was not reproducible and robust. Judging from the agreement between the different IGRAs, it seems that the results of the retest conducted using the QFT-G-IT assay had a higher reliability than those of the initial test. In the QFT-G-IT retest, reversion from positive to negative occurred mostly in the case of participants with initial measurements just above the diagnostic cut-off. Therefore, attention must be given not only to the dichotomous results (positive or negative) but also to the exact level of IFN-gamma production. The results of the QFT-G-IT assay may also be affected by various environmental factors. If the QFT-G-IT assay yields a high positive rate disproportionate to the closeness of contact, the result should be carefully interpreted, taking into account the unreliability of the QFT-G-IT assay as a possible cause of discordance. Further studies under various settings are needed to establish the reliability of IGRAs.