Construction of a predictive model for the severity of sepsis based on APACHE Ⅱ score, IL-6, and T lymphocyte subsets

Abstract

Objective To construct a predictive model for the severity of sepsis using the Acute Physiology and Chronic Health EvaluationⅡ (APACHEⅡ) score, interleukin-6 (IL-6), and T lymphocyte subsets, and to evaluate the model's effectiveness. Methods A total of 225 patients with sepsis admitted to the Anqing Municipal Hospital from January 2021 to September 2023 were selected as the study subjects. IL-6, T lymphocyte subsets (CD4+, CD8+, CD4+/CD8+), C-reactive protein (CRP), procalcitonin (PCT), and APACHEⅡ scores were measured within 24 hours after diagnosis. Based on the Sepsis-3.0 diagnostic criteria, patients were divided into the sepsis group (109 cases) and the septic shock group (116 cases). Logistic regression analysis was used to select variables and construct a predictive model for the severity of sepsis. Calibration plots and decision curve analysis were employed to evaluate the model's fit and clinical value. Results The levels of CD4+ and CD4+/CD8+ in the septic shock group were lower than those in the sepsis group, while CRP, PCT, IL-6, and APACHEⅡ scores were higher in the septic shock group than in the sepsis group (P＜0.05). The constructed predictive model was as follows: ln［P/(1-P)］=0.999+0.054×APACHEⅡ score-0.054×CD4+-0.18×CD4+/CD8++0.001×IL-6. Calibration plots and decision curve analyses indicated that the model had good fit and clinical value. Conclusion The predictive model composed of CD4+, CD4+/CD8+, IL-6, and APACHEⅡ score can be used for early assessment of the severity of sepsis, providing assistance for clinical diagnosis and treatment.

Keywords

• interleukin-6， t lymphocyte subsets， acute physiology and chronic health evaluation score， sepsis， predictive model