RESUMEN
An understanding of the diagnostic performance of interferon-gamma release assays (in terms of parameters such as specificity, positive predictive value, and likelihood ratio) is important in the diagnosis of active tuberculosis in elderly Japanese patients because the high proportion of a prior history of tuberculosis among these patients can lead to misleading results. To elucidate the diagnostic performance of such assays, we examined the results of the QuantiFERON-TB Gold In-Tube test (QFT-GIT) in 65 patients in a younger group, and 52 patients in an elderly group who were suspected of having active tuberculosis and who had received the QFT-GIT. The median ages of the younger patients and elderly patients were 54 and 78 years, respectively. Among patients with active tuberculosis, the number of those with positive results on the QFT-GIT was similar in the two age groups [15 out of 17 (88.2 %) in the younger patients compared with 7 out of 8 (87.5 %) in the elderly patients]. However, in the patients with other diseases the number of those with positive QFT-GIT results was significantly different in the younger and elderly groups, being 6.3 and 27.3 %, respectively (P = 0.01). Although the sensitivity, negative predictive value, and negative likelihood ratio were similar in the two groups, the specificity, positive predictive value, and positive likelihood ratio were significantly lower in the elderly patients, being 72.7% (95 % confidence interval [CI], 57.2-85.0) compared with 93.8% (95 % CI, 82.8-98.7), 36.8% (95 % CI, 16.3-61.6) compared with 83.8% (95 % CI, 58.6-96.4), and 3.21 (95 % CI, 1.85-5.56) compared with 14.12 (95 % CI, 4.66-42.81), respectively. In the elderly patients with positive results on the test, a comparison of clinical data between those with active tuberculosis and those with other diseases demonstrated that the only clinical parameter showing a significant difference between these two groups was the radiological finding of small nodules in the patients with active tuberculosis (P < 0.01). The QFT-GIT may be less accurate in elderly patients, and radiological findings can be helpful in the clinical evaluation of patients with positive results on the test.
