Abstract
BACKGROUND: Active tuberculosis (ATB) in children and adolescents remains a major diagnostic challenge. Inflammation, nutrition, and immune status are closely linked to tuberculosis (TB) progression. We therefore proposed a novel biomarker, the logarithmic lymphocyte-albumin product (logLAP), integrating immune and nutritional indicators, and investigated its association with ATB in children and adolescents. METHODS: This retrospective study included 1,080 participants aged <18 years. Participants were classified into ATB (n = 904) and non-TB (n = 176) groups. LAP was calculated as lymphocyte × albumin, and logLAP was derived as log(LAP). Discriminatory ability was evaluated using receiver operating characteristic (ROC) curves. Logistic regression models were applied to assess the association between logLAP and ATB. Restricted cubic spline (RCS) analyses and stratified analyses were conducted to explore nonlinear relationships and subgroup differences. RESULTS: Compared with non-TB controls, ATB patients exhibited significantly lower logLAP levels (4.07 ± 0.55 vs. 4.39 ± 0.47, p < 0.001). In ROC analysis, LAP achieved the higher area under the curve (AUC = 0.6955) than neutrophil-lymphocyte ratio (NLR), neutrophil-albumin ratio (NAR), platelet-lymphocyte ratio (PLR), systemic immune-inflammation index (SII) and prognostic-nutritional index (PNI). The clinical diagnostic cut-off value for logLAP was 4.23, with a sensitivity of 60% and a specificity of 74%. Logistic regression models revealed the third tertile (Q3) of logLAP showed a strong inverse association with ATB (OR = 0.33, 95% CI: 0.18-0.59, p < 0.001) after full adjustment. RCS analyses with four knots revealed a significant non-linear relationship between logLAP and ATB (p for non-linearity = 0.021), confirming a threshold effect at logLAP = 3.48. Stratified analyses indicated consistent associations across most subgroups, while logLAP showed limited associations in children aged 0-7 years and those with severe immune dysfunction (CD4+/CD8+ T cell counts below the reference range). The AUC of logLAP demonstrated a clear advantage over traditional biomarkers; however, its diagnostic performance remained relatively limited. This suggests that logLAP is more suitable as a supplementary tool and should be used in conjunction with other clinical or laboratory indicators. CONCLUSION: As a biomarker integrating immune-nutritional status, logLAP exhibits a robust inverse association with ATB infection in children and adolescents aged ≥8 years with normal or moderately impaired immune function, supported by comprehensive analyses.