摘要
Current severity scoring systems (PSI and CURB-65) have limitations in risk stratification for elderly patients with community-acquired pneumonia (CAP). Given the complex immune responses in elderly populations, dynamic biomarker monitoring may provide additional prognostic value. This study evaluates whether integrating early cytokine dynamics with traditional severity scores improves mortality prediction in elderly CAP patients. This prospective observational study included 81 CAP patients aged ≥ 65 years. Multiple cytokines were measured at admission and within 48 h. Traditional severity scores (PSI and CURB-65) were calculated at baseline. Patients were categorized into survival (n = 67) and mortality (n = 14) groups. The predictive value of cytokine dynamics alone and in combination with severity scores was assessed using ROC curve analysis. Among measured cytokines, IL-6 demonstrated significant prognostic value. The mortality group showed an 88% increase in IL-6 levels within 48 h, contrasting with a 49% decrease in survivors (p = 0.040). While individual PSI (AUC = 0.6631) and CURB-65 (AUC = 0.6231) showed modest discrimination, integration with IL-6 dynamics significantly improved predictive accuracy (PSI + IL-6: AUC = 0.7676, p = 0.0017; CURB-65 + IL-6: AUC = 0.7564, p = 0.0027). Early dynamic monitoring of cytokines, particularly IL-6, significantly enhances the prognostic accuracy of traditional severity scores in elderly CAP patients. This pilot study suggests that this integrated approach provides a more precise risk stratification tool, potentially enabling more personalized clinical decision-making. Larger multicenter studies are warranted to validate these findings and establish standardized cutoff values for clinical implementation. © The Author(s) 2025.
