AUTHOR=Tian Kai-Xi , Dong Jie-Xuan , Du Juan , Huang Xu-Bing , Liu Yi 

TITLE=Development and validation of a lactate and procalcitonin-based nomogram model for predicting prognosis in ICU sepsis patients

JOURNAL=Frontiers in Disaster and Emergency Medicine

VOLUME=Volume 3 - 2025

YEAR=2025

URL=https://www.frontiersin.org/journals/disaster-and-emergency-medicine/articles/10.3389/femer.2025.1612093

DOI=10.3389/femer.2025.1612093

ISSN=2813-7302

ABSTRACT=ObjectiveA nomogram model for predicting the prognosis of patients with sepsis in the intensive care unit (ICU) was developed based on lactate and procalcitonin (PCT) and externally validated.MethodsFrom the anonymous medical database of the First Affiliated Hospital of Tianjin University of Traditional Chinese Medicine, 55 patients with sepsis admitted to the ICU of the First Affiliated Hospital of Tianjin University of Traditional Chinese Medicine were retrospectively collected. Demographic data, admission leukocyte count, lactate, PCT, and other clinical parameters were collected. Based on 28-day outcomes, patients were stratified into survival and mortality groups. Independent prognostic risk factors were identified In review through multivariate logistic regression analysis, and a nomogram prediction model was subsequently developed and internally validated. From the anonymous medical database of the First Affiliated Hospital of Tianjin University of Traditional Chinese Medicine, an additional 45 ICU sepsis patients were collected to form an independent cohort for external validation.ResultsThe cohort comprised 55 patients (37 male, 18 female), with 41 survivors and 14 non-survivors. No statistically significant differences were observed between survival and mortality groups in sex, age, White Blood Cell (WBC) Counts on Admission, Underlying Diseases, admission lactate,72-h lactate levels, 72-h lactate clearance rate, 24-h PCT levels, 24-h PCT clearance rate, or 72-h PCT levels (all P > 0.05). Significant differences were identified in 24-h lactate levels, 24-h lactate clearance rate, admission PCT levels, and 72-h PCT clearance rate (P < 0.05). Multivariate analysis confirmed 24-h lactate clearance rate and 72-h PCT clearance rate as independent prognostic risk factors for sepsis outcomes. The prognostic value of 24-h lactate clearance rate and 72-h PCT clearance rate in ICU sepsis patients was evaluated. The area under the ROC curve (AUC) for the 24-h lactate clearance rate was 0.709 (P = 0.02). For the 72-h PCT clearance rate, the AUC was 0.908 (P = 0.039). The combined model yielded an AUC of 0.920 (P = 0.036). Internal validation of the nomogram model incorporating both 24-h lactate clearance In review rate and 72-h PCT clearance rate demonstrated an AUC of 0.9012453, with a sensitivity of 0.539507 and specificity of 0.8985287. The calibration curve analysis shows that there is a certain correlation between the prediction probability of the model and the observation results, indicating that it has a certain prediction accuracy and reliability. Decision curve analysis (DCA) demonstrated some clinical net benefit across a range of threshold probabilities, supporting the model's utility in sepsis management. External validation of the nomogram model (incorporating 24-h lactate clearance rate and 72-h PCT clearance rate) using an independent cohort showed an AUC of 0.681, with sensitivity 0.510 and specificity 0.742.ConclusionBoth 24-h lactate clearance rate and 72-h PCT clearance rate may serve as independent prognostic risk factors for predicting outcomes in ICU sepsis patients. The combination of two independent prognostic risk factors may demonstrate superior predictive value for the prognosis of ICU sepsis patients compared to individual factors alone. A nomogram prediction model integrating these two independent risk factors may exhibits enhanced accuracy, reliability, and clinical utility. However, due to the limited sample size (n = 55) and single-center cohort design, these findings require validation through multicenter prospective studies with larger cohorts to strengthen the evidence level and confirm In review generalizability.