SETTING: Hong Kong Chest Clinics. OBJECTIVES AND METHODS: To conduct a prospective study investigating the role of a whole-blood interferon-gamma release assay, QuantiFERON®-TB Gold In-Tube (QFT-GIT), in the diagnosis of smear-negative tuberculosis (TB). The QFT-GIT result was compared with the final confirmed diagnosis after 12 months. RESULTS: Of 262 smear-negative subjects, 188 had active TB, 167 (88.8%) of whom were QFT-GIT-positive; 74 had inactive/non-TB, 30 (40.5%) of whom were QFT-GIT-negative. The positive (PPV) and negative predictive values for active TB were respectively 79.1% and 58.8%. For this target group with high TB prevalence (71.8%), a positive test increased the chance of active disease by only 7.3%. Despite a positive likelihood ratio (LR) of 1.49, the negative LR was 0.28, making the diagnosis of active TB much less likely after a negative test. Although sensitivity and specificity showed no difference across different age groups, the PPV decreased (P < 0.001) with increasing age, likely reflecting the increased prevalence of competing diagnoses. CONCLUSION: In an area with a high prevalence of latent TB infection, a positive QFT-GIT test does not add much to confirm the diagnosis of smear-negative TB, while a negative test indicates a need for further investigation.
SETTING: Hong Kong Chest Clinics. OBJECTIVES AND METHODS: To conduct a prospective study investigating the role of a whole-blood interferon-gamma release assay, QuantiFERON®-TB Gold In-Tube (QFT-GIT), in the diagnosis of smear-negative tuberculosis (TB). The QFT-GIT result was compared with the final confirmed diagnosis after 12 months. RESULTS: Of 262 smear-negative subjects, 188 had active TB, 167 (88.8%) of whom were QFT-GIT-positive; 74 had inactive/non-TB, 30 (40.5%) of whom were QFT-GIT-negative. The positive (PPV) and negative predictive values for active TB were respectively 79.1% and 58.8%. For this target group with high TB prevalence (71.8%), a positive test increased the chance of active disease by only 7.3%. Despite a positive likelihood ratio (LR) of 1.49, the negative LR was 0.28, making the diagnosis of active TB much less likely after a negative test. Although sensitivity and specificity showed no difference across different age groups, the PPV decreased (P < 0.001) with increasing age, likely reflecting the increased prevalence of competing diagnoses. CONCLUSION: In an area with a high prevalence of latent TB infection, a positive QFT-GIT test does not add much to confirm the diagnosis of smear-negative TB, while a negative test indicates a need for further investigation.