Performance of pneumonia severity index and CURB-65 in predicting 30-day mortality in patients with COVID-19

Abstract

OBJECTIVE: The aim of the study was to analyze the usefulness of CURB-65 and the pneumonia severity index (PSI) in predicting 30-day mortality in patients with COVID-19, and to identify other factors associated with higher mortality. METHODS: A retrospective study was performed in a pandemic hospital in Istanbul, Turkey, which included 681 laboratory-confirmed patients with COVID-19. Data on characteristics, vital signs, and laboratory parameters were recorded from electronic medical records. Receiver operating characteristic analysis was used to quantify the discriminatory abilities of the prognostic scales. Univariate and multivariate logistic regression analyses were performed to identify other predictors of mortality. RESULTS: Higher CRP levels were associated with an increased risk for mortality (OR: 1.015, 95% CI: 1.008-1.021; p < 0.001). The PSI performed significantly better than CURB-65 (AUC: 0.91, 95% CI: 0.88-0.93 vs AUC: 0.88, 95% CI: 0.85-0.90; p = 0.01), and the addition of CRP levels to PSI did not improve the performance of PSI in predicting mortality (AUC: 0.91, 95% CI: 0.88-0.93 vs AUC: 0.92, 95% CI: 0.89-0.94; p = 0.29). CONCLUSION: In a large group of hospitalized patients with COVID-19, we found that PSI performed better than CURB-65 in predicting mortality. Adding CRP levels to PSI did not improve the 30-day mortality prediction.