Predictive value of serial evaluation of the Sequential Organ Failure Assessment (SOFA) score for intensive care unit mortality in critically ill patients with COVID-19: a retrospective cohort study

Abstract

Background:The Sequential Organ Failure Assessment (SOFA) score has been developed to score the severity of organ dysfunction in critically ill sepsis patients and has been proven to have a high predictive value for intensive care unit (ICU) mortality in severely ill patients. Our goal was to evaluate the prognostic value of the SOFA score as well as trends in SOFA score for ICU mortality in COVID-19 patients. Methods: All consecutive patients with confirmed COVID-19 pneumonia admitted to the ICU between March 13th, 2020, and October 17th, 2020 were included in this retrospective cohort study. The worst SOFA score was evaluated daily. Multiple logistic regression models were used to evaluate the predictive value of SOFA in ICU mortality. Results: 103 patients were included in this study. 30 patients (29%) died during their ICU stay and 73 (71%) patients were discharged alive. The ICU admission SOFA score was 5.2 +/- 3.3 in ICU non-survivors vs. 4.3 +/- 2.9 in ICU survivors (P = 0.15). The maximum SOFA score in ICU non-survivors was 11.7 +/- 4.7 vs. 7.4 +/- 4.3 in ICU survivors. SOFA scores increased the first week in both survivors and non-survivors, but the increase was less pronounced in survivors. In the multiple logistic regression models, neither admission SOFA score nor combination with delta SOFA in the first 48 hours was statistically significantly related to ICU mortality. Only the maximum SOFA score remained significant (OR = 1.23, 95% CI: 1.11-1.37, P < 0.001) in the multiple logistic models with an AUC of 0.91. Conclusions: Evaluation of SOFA scores in the first 48 hours after ICU admission is not a good prognostic indicator in COVID-19 patients. Only the maximum SOFA score was predictive for ICU mortality.