Assessment of risk scores to predict mortality of COVID-19 patients admitted to the intensive care unit

Abstract

Objectives: To assess the ABC2-SPH score in predicting COVID-19 in-hospital mortality, during intensive care unit (ICU) admission, and to compare its performance with other scores (SOFA, SAPS-3, NEWS2, 4C Mortality Score, SOARS, CURB-65, modified CHA2DS2-VASc, and a novel severity score). Materials and methods: Consecutive patients (≥ 18 years) with laboratory-confirmed COVID-19 admitted to ICUs of 25 hospitals, located in 17 Brazilian cities, from October 2020 to March 2022, were included. Overall performance of the scores was evaluated using the Brier score. ABC2-SPH was used as the reference score, and comparisons between ABC2-SPH and the other scores were performed by using the Bonferroni method of correction. The primary outcome was in-hospital mortality. Results: ABC2-SPH had an area under the curve of 0.716 (95% CI 0.693–0.738), significantly higher than CURB-65, SOFA, NEWS2, SOARS, and modified CHA2DS2-VASc scores. There was no statistically significant difference between ABC2-SPH and SAPS-3, 4C Mortality Score, and the novel severity score. Conclusion: ABC2-SPH was superior to other risk scores, but it still did not demonstrate an excellent predictive ability for mortality in critically ill COVID-19 patients. Our results indicate the need to develop a new score, for this subset of patients.