Severity Scores in COVID-19 Pneumonia: a Multicenter, Retrospective, Cohort Study

Area of Science:

• Pulmonary Medicine
• Infectious Diseases
• Critical Care Medicine

Background:

• Assessing outcomes for hospitalized COVID-19 pneumonia patients remains a challenge.
• Early identification of high-risk individuals is crucial for timely intervention.

Purpose of the Study:

• To evaluate the effectiveness of existing pneumonia severity scores in predicting outcomes for COVID-19 pneumonia.
• To compare the performance of PSI, CURB-65, qSOFA, and MuLBSTA in identifying high-risk patients.

Main Methods:

• Utilized data from 10,238 patients with COVID-19 pneumonia from the SEMI-COVID-19 Network.
• Calculated PSI, CURB-65, qSOFA, and MuLBSTA scores on hospital admission.
• Assessed scores for predicting in-hospital mortality, ICU admission, and mechanical ventilation using AUROC, sensitivity, and specificity.

Main Results:

• PSI (AUROC 0.835) and CURB-65 (AUROC 0.825) demonstrated superior predictive performance compared to qSOFA (AUROC 0.728) and MuLBSTA (AUROC 0.715).
• PSI exhibited the highest sensitivity (84.1%), while qSOFA showed the highest specificity (95.7%).
• In-hospital mortality was 20.9% among the study cohort.

Conclusions:

• PSI and CURB-65 are more effective than qSOFA and MuLBSTA for predicting mortality in COVID-19 pneumonia.
• qSOFA, though simple to apply and highly specific, has limited sensitivity for identifying at-risk patients.