SETTING: Nine university-affiliated chest clinics in Australia. OBJECTIVE: To evaluate the sensitivity of a whole blood human gamma-interferon assay (HGIA, QuantiFERON-TB) for specific T lymphocyte responses and Tuberculin skin testing (TST) for the detection of Mycobacterium tuberculosis infection in subjects with culture-proven M. tuberculosis disease (TBCP). DESIGN: Prospective testing of 129 patients with recent TBCP and 100 patients with non-tuberculosis lung disease (NTBLD). RESULTS: Using a defined level of specific IFN-gamma production and TST 10mm as positive cut-offs, the sensitivity of HGIA was 81% compared to 89% for TST (p=0.06). When positive responses in both TST and HGIA were combined, 96% of TB patients were detected. For the NTBLD group, 43% of whom were born overseas, 73% were negative for both the HGIA and TST. Prior immunization with M. bovis (bacille Calmette-Guerin) (BCG) or the type of TB had no effect on the sensitivities of the assays. For those treated for <2 months, the sensitivities for both assays were 84%, but for those treated for >2 months the sensitivity of TST (90%) tended to be higher than for HGIA (81%) (p=0.07). The distribution of TST results in TB patients showed a broad peak between 10 and 25 mm, while the results in the HGIA were bimodal in both TB and NTBLD patients. CONCLUSION: HGIA may prove an alternative to skin testing for detecting M. tuberculosis infection in certain settings.
SETTING: Nine university-affiliated chest clinics in Australia. OBJECTIVE: To evaluate the sensitivity of a whole blood human gamma-interferon assay (HGIA, QuantiFERON-TB) for specific T lymphocyte responses and Tuberculin skin testing (TST) for the detection of Mycobacterium tuberculosisinfection in subjects with culture-proven M. tuberculosis disease (TBCP). DESIGN: Prospective testing of 129 patients with recent TBCP and 100 patients with non-tuberculosis lung disease (NTBLD). RESULTS: Using a defined level of specific IFN-gamma production and TST 10mm as positive cut-offs, the sensitivity of HGIA was 81% compared to 89% for TST (p=0.06). When positive responses in both TST and HGIA were combined, 96% of TBpatients were detected. For the NTBLD group, 43% of whom were born overseas, 73% were negative for both the HGIA and TST. Prior immunization with M. bovis (bacille Calmette-Guerin) (BCG) or the type of TB had no effect on the sensitivities of the assays. For those treated for <2 months, the sensitivities for both assays were 84%, but for those treated for >2 months the sensitivity of TST (90%) tended to be higher than for HGIA (81%) (p=0.07). The distribution of TST results in TBpatients showed a broad peak between 10 and 25 mm, while the results in the HGIA were bimodal in both TB and NTBLD patients. CONCLUSION: HGIA may prove an alternative to skin testing for detecting M. tuberculosis infection in certain settings.
Authors: Hiroshi Nakaoka; Lovett Lawson; S Bertel Squire; Brian Coulter; Pernille Ravn; Inger Brock; C Anthony Hart; Luis E Cuevas Journal: Emerg Infect Dis Date: 2006-09 Impact factor: 6.883