Accuracy of an interferon-γ release assay to detect active pulmonary and extra-pulmonary tuberculosis

OBJECTIVE: To examine the performance of an interferon-gamma (IFN-γ) release assay (QuantiFERON-TB 2G assay [QFT-G]) to detect Mycobacterium tuberculosis infection in a Japanese general hospital, for the diagnosis of active pulmonary tuberculosis (PTB) and extrapulmonary tuberculosis (EPTB). DESIGN: We prospectively examined the performance of QFT-G in 194 patients suspected of active TB. Diagnosis was confirmed by 1) positive M. tuberculosis cultures, or 2) clinical manifestations or laboratory or pathological findings consistent with active TB and response to specific therapy. RESULTS: Three patients with indeterminate QFT-G results were excluded. Among the remaining 191 patients, 77 had active TB. When the cut-off concentration of IFN-γ was set at 0.35 IU/ml, as recommended by the manufacturer, the assay was positive in 69 patients and negative in 122. The sensitivity of the assay was 76.6% in all patients, 74.5% in the 47 patients with PTB and 80.0% in the 30 patients with EPTB. The overall specificity of the assay was 91.2%. CONCLUSION: Although the specificity of the QFT-G to detect active TB was high and its sensitivity low, it was as accurate for the detection of active EPTB as for PTB when the 0.35 IU/ml INF-γ cut-off concentration was used.