Validation of the Acute Physiology and Chronic Health Evaluation (APACHE) II Score in COVID-19 Patients Admitted to the Intensive Care Unit in Times of Resource Scarcity

Introduction During the coronavirus disease 2019 (COVID-19) pandemic, a high number of patients needed to be admitted to the intensive care units (ICUs). Such a high demand led to periods where resources were insufficient and the triage of patients was needed. This study aims to evaluate the performance of the Acute Physiology and Chronic Health Evaluation (APACHE) II as a predictor of mortality in periods where triage protocols were implemented. Methods A single-center, longitudinal, retrospective cohort study was performed on patients admitted to the ICU between January 2020 and December 2021. Patients were divided into two periods: Period 1 (where patients needing ICU admission outnumbered the available resources) and Period 2 (where resources were adequate). The discriminative power of the APACHE II was checked using the receiver operating characteristic (ROC) curves. Calibration was accessed, and survival analysis was performed. Results Data from 428 patients were analyzed (229 in Period 1 and 199 in Period 2). The area under the ROC curve (AUROC) was 0.763 for Period 1 and 0.761 for Period 2, reflecting a good discriminative power. Logistic regression showed the APACHE II to be a significant predictor of mortality. The Hosmer-Lemeshow test demonstrated good calibration. The Youden index was determined, and a log-rank test showed a significantly lower survival for patients with higher APACHE II scores in both periods. Conclusions The APACHE II score is an effective tool in predicting mortality in patients with COVID-19 admitted to the ICU in a period where resource allocation and triage of patients are needed, paving a way for the future development of better and improved triage systems.