Sensitivity of tuberculosis infection diagnostic tests among different populations: a systematic review and meta-analysis

Background We aimed to estimate the sensitivity of skin and blood-based tuberculosis (TB) infection tests and assess variation in sensitivity by test type and population characteristics. Methods We conducted a meta-analysis of studies involving individuals with microbiologically confirmed TB disease (surrogate reference standard) who underwent TB infection testing. Searches were performed across databases up to February 24, 2025. Data were extracted using a standardized form and pooled using random-effects meta-analysis. Primary analyses focused on tests performed before or during TB disease treatment. Results From 103,031 records, 307 studies were included: 110 evaluated tuberculin skin test (TST), 185 QuantiFERON (QFT), 107 T-SPOT.TB, 10 antigen-based skin tests, and 5 TB-IGRA. Sensitivity was consistently lower among people with HIV. Among populations without vs. with HIV, respectively, TST ≥5mm sensitivity was 80.3% (95%CI: 73.7-85.5) and 53.6% (41.5-65.4), TST ≥10mm sensitivity was 75.1% (68.8-80.5) and 64.5% (36.8-85.0), antigen-based skin tests sensitivity was 77.7% (69.7-84) and 61.3% (39.6-79.3), QFT sensitivity was 82.6% (80.7-84.3) and 66.5% (59.8-72.6), and T-SPOT.TB sensitivity was 88.2% (86.2-90.0) and 69% (56.6-79.2). TB-IGRA was only evaluated in populations without HIV (sensitivity: 85.4%; 81.3-88.7). Sensitivity in other populations targeted for TB infection testing generally fell between estimates among people with and without HIV. In comparative analyses, TST ≥5mm and TST ≥10mm had comparable sensitivity to QFT, however T-SPOT.TB had superior sensitivity to both TST ≥5mm and QFT. Conclusion TB infection test sensitivity varies by test type and population. All tests showed reduced sensitivity in people with HIV. T-SPOT.TB demonstrated the highest sensitivity across populations.