Clinical Application of Combined CEACAM1_A and CEACAM1_B in Early Warning and Diagnosis of Sepsis

BACKGROUND: This retrospective observational study aimed to evaluate the predictive value of CEACAM1_A, CEACAM1_B, and their combined model in the early warning diagnosis of sepsis. Based on previously identified CEACAM1_A and CEACAM1_B, this study evaluated their potential in the early diagnosis of sepsis and explored the construction of predictive models that combine genetic markers with clinical indicators. METHODS: A total of 144 acutely infected patients admitted to Hefei Second People's Hospital between July 2023 and February 2025 were enrolled in this retrospective observational study. The patients were divided into sepsis and non-sepsis groups according to the Sepsis-3 standard. The sepsis group included 96 patients, and the non-sepsis group consisted of 48 patients. Baseline characteristics, biochemical parameters, and expression levels of the CEACAM1_A and CEACAM1_B genes were collected from both groups. Statistical analyses were performed via SPSS version 27.0 and the R programming language. For data distribution, intergroup comparisons were conducted via appropriate parametric and nonparametric tests. Univariate and multivariate logistic regression analyses were employed to identify independent predictors, and receiver operating characteristic (ROC) curve analysis was used to evaluate model performance. RESULTS: The study revealed significantly greater expression levels of CEACAM1_A and CEACAM1_B in the sepsis group than in the non-sepsis group (p < 0.001), with positive correlations with disease severity (correlation co-efficients of 1 and 0.992, respectively). Multivariate analysis revealed that CEACAM1_A (odds ratio [OR] = 1.001, 95% confidence interval [CI]: 1.001 - 1.006, p < 0.001), CEACAM1_B (OR = 1.001, 95% CI: 1.000 - 1.002, p < 0.001), lactate (OR = 4.154, 95% CI: 2.207 - 7.819, p < 0.001), and C-reactive protein (OR = 1.012, 95% CI: 1.004 - 1.020, p = 0.002) were independent risk factors for sepsis. The area under the curve (AUC) for the combined predictive model of CEACAM1_A and CEACAM1_B was 0.914, outperforming the other indicators. CONCLUSIONS: The combined application of these two biomarkers significantly improved the accuracy of early sepsis detection, potentially facilitating optimal resource allocation and improving patient outcomes.