Guillermo Eduardo Cuéllar-Nevárez, José Rafael Linares Morales, Luis Arturo Camacho Silvas, Sandra Guadalupe Chavarría Hidalgo, Guadalupe Virginia Nevárez-Moorillón, Karla Elva Loza Solano
Tuberculosis (TB) remains a leading cause of morbidity and mortality among people living with human immunodeficiency virus (HIV), in whom immunosuppression often results in atypical clinical manifestations, paucibacillary disease, extrapulmonary involvement, and disseminated forms that complicate timely diagnosis. This narrative review summarizes current and emerging diagnostic approaches for TB-HIV, with emphasis on their diagnostic performance, clinical applicability, implementation challenges, and potential integration into future algorithms. Conventional methods, including sputum smear microscopy, mycobacterial culture, and chest radiography, remain relevant but have important limitations in people living with HIV (PLHIV), particularly in advanced immunosuppression. Rapid molecular tests, especially Xpert MTB/RIF and Xpert MTB/RIF Ultra, have improved early detection of Mycobacterium tuberculosis and rifampicin resistance, although diagnostic gaps persist in sputum-scarce and extrapulmonary disease. Non-sputum-based tests, particularly WHO-recommended urine LF-LAM and emerging next-generation assays such as FujiLAM II, offer important opportunities for severely ill patients and those with advanced HIV disease. Host-derived biomarkers, transcriptomic signatures, and artificial intelligence-assisted chest imaging may further strengthen screening, triage, and diagnostic prioritization. Integrated, context-adapted diagnostic algorithms combining microbiological, molecular, urinary, imaging, and host-response tools are needed to improve early TB detection and reduce mortality in resource-limited settings.