Data-driven targets for predicting outcomes in patients with HIV infection and tuberculous meningitis in China

Objectives: For people living with HIV (PLWH) and tuberculous meningitis (TBM), current studies on risk stratification and poor prognosis lack a clear optimal threshold. To address this gap, this study aims to identify an optimal immunological threshold for risk stratification and explore predictors of poor prognosis in this population. Methods: T-lymphocyte count on admission. CD4 thresholds were determined using restricted cubic splines with knots at quintiles. Multivariable logistic regression of risk factors derived adjusted ORs with 95% CIs. Results: T-lymphocyte count <50 cells/µL and elevated blood urea nitrogen (BUN) as independent risk factors for adverse outcomes. Conclusions: T-cell count below 50 cells/µL defines a critical stratification threshold for clinical risk classification, with elevated BUN serving as an additional prognostic marker. These findings support the prioritisation of severely immunocompromised patients for targeted management and further mechanistic studies.