AI Summary & Key Takeaways
AI InsightThese findings should be interpreted cautiously and require validation in independent cohorts.
📌 3 Key Takeaways
Abstract
Aims: Accurate prediction of mortality in critically ill patients is important for clinical decision-making and resource allocation in intensive care units (ICUs). This study aimed to evaluate factors associated with ICU mortality, with particular emphasis on APACHE II score and lactate-to-albumin ratio (LAR). Methods: This single-center retrospective cohort study reviewed adult patients admitted to a tertiary-care ICU between February 2021 and February 2024. Patients with ICU stays of less than 24 hours, incomplete clinical or laboratory data, or repeated admissions were excluded. Demographic characteristics, severity scores, laboratory parameters, and organ support requirements were obtained from electronic medical records. Logistic regression and receiver operating characteristic analyses were performed. Results: Of 2,358 screened ICU admissions, 2,051 patients were included in the final analysis, and 313 (15.3%) died during ICU stay. Non-survivors had significantly higher APACHE II and SOFA scores, lactate, LAR, and creatinine levels, and lower albumin levels (all p<0.001). In multivariable analysis, APACHE II score (adjusted OR: 1.122, 95% CI: 1.100-1.145), LAR (adjusted OR: 4.843, 95% CI: 3.574-6.562), creatinine level (adjusted OR: 1.840, 95% CI: 1.510-2.243), mechanical ventilation, vasopressor use, and renal replacement therapy were associated with ICU mortality. APACHE II demonstrated the highest discriminative performance (AUC=0.764), followed by LAR (AUC=0.750). Conclusion: In this single-center retrospective cohort, APACHE II score and LAR were associated with ICU mortality. APACHE II showed the highest discriminative performance, while LAR may serve as a simple adjunctive laboratory-based marker. These findings should be interpreted cautiously and require validation in independent cohorts.