Aim: The study was carried out on elderly nursing home residents in Taiwan. We assessed whether the serial QuantiFERON-TB Gold (QFT-G) assay and serial tuberculin skin test (TST) were reliable tools to predict or exclude the development of active tuberculosis (TB).
Methods: This prospective observational cohort study involved non-bacillus Calmette-Guérin-vaccinated 259 elderly nursing home residents free of active TB at baseline. Of these, 147 were eligible for follow up. Participants underwent serial QFT-G and TST at baseline and 2-year follow up, and were monitored for active TB over 5 years. Agreement between QFT-G and TST, incidence rate ratio, positive predictive value, and negative predictive value for progression to active TB were measured.
Results: During 5-year follow up, three participants developed active TB. The agreement between these two tests was 54.13% (ĸ = 0.167, P = 0.001). The incidence rate ratio was 15.8 (P = 0.016) for the QFT-G-conversion group compared with the TST-positive group at baseline. Positive predictive value for QFT-G conversion groups was 25%. Negative predictive value was 100% for the TST-negative group at baseline.
Conclusion: In the elderly nursing home residents, QFT-G conversion is a more reliable tool to predict the development of active TB. Meanwhile, TST is a valuable tool for predicting the chance of not developing active TB.
Keywords: Mycobacterium tuberculosis; Quantiferon TB-Gold assay; elderly; nursing home residents; tuberculin skin test.
© 2014 Japan Geriatrics Society.