Combined Xpert MTB/RIF and urine LAM for rapid tuberculosis detection and TB/HIV co-infection predictors in Amhara, Ethiopia: multicenter retrospective study

Abstract

Tuberculosis, a preventable and curable disease caused by Mycobacterium tuberculosis , remains a major global health threat, especially alongside HIV/AIDS. This study evaluates the diagnostic concordance of Xpert MTB/RIF and urine LAM for rapid TB detection and explores TB-HIV co-infection factors in specialized hospitals. This multicenter, hospital-based retrospective cross-sectional study was conducted in three specialized hospitals. Data were extracted from patient records of HIV-positive individuals who underwent tuberculosis (TB) screening using Xpert MTB/RIF on sputum specimens and urine lipoarabinomannan (LAM) testing on urine specimens. Statistical analyses were performed using SPSS version 23.0. Continuous clinical variables, including CD4 + T-lymphocyte counts and plasma viral loads, were analyzed, and concordance between the two diagnostic methods was assessed. To identify factors independently associated with TB/HIV co-infection, both bi- variable and multivariable logistic regression analyses were performed. Among 366 participants, 63 tested positive for tuberculosis by urine LAM and 62 Xpert MTB/RIF (sputum), demonstrating excellent concordance (k = 0.92; P < 0.001). Logistic regression identified significant TB risk in HIV patients aged 25–44 (AOR = 11, 95% CI 1.4–83.3; P < 0.05) and 45–64 (AOR = 10, 95% CI 1.4–76.9; P < 0.05.). Virological failure is also significantly associated with tuberculosis, despite higher case proportions in undetectable viral loads (AOR = 2.5, 95% CI 1.3–4.5; P < 0.01). The study demonstrated strong concordance between diagnostic methods. Combining both tests enhances TB detection, but a single test may be sufficient in resource-limited settings. Higher TB/HIV co-infection association has been observed in individuals aged 25–44 and those with virological treatment failure.