TYPE: Late Breaking Abstract
TOPIC: Critical Care
PURPOSE: Determine the prognostic factors of severity and mortality in critically ill patients due to COVID-19 that allow a rapid identification upon admission to the ICU and better results with existing resources, using validated prognostic models
METHODS: An analytical observational study was conducted in COVID-19 critically ill patients in 22 ICUs in 8 Latin American countries. A multivariate analysis was performed and severity and mortality prediction models were estimated using StataCorp 15.0®
RESULTS: 539 patients were included in the study. Hospital mortality was 48.42%. Factors associated with severity were: SOFA Score (OR: 1.18 CI: 1.07-1.31 p=0.001), APACHE Score (OR: 1.08 95% CI: 1.03-1.14 p=0.002), tachypnea (OR: 1.39 95% CI: 1.31-1.47 p=<0.05) and PaFi behaved as a protective factor (OR: 0.98 95% CI: 0.98-0.99, p<0.05). Factors associated with mortality were: invasive mechanical ventilation, (OR: 5.75 95% CI 3.18-10.42, p = <0.05 ), septic shock (OR: 2.50 CI: 1.58-4.12, p =<0.05) APACHE score (OR: 1.08 95% CI 1.04-1.12, p=<0.05), creatinine (OR:1.18 95% CI 1.04-1.34, p=0.010) and older than 60 years (OR: 2.03 CI: 1.32-3.14p=0.001). The dissociation between SpO2-PaO2 has 45% less mortality and it is a protective factor (OR: 0.45 95% CI 0.25-0.82, p=0.009)
CONCLUSIONS: In our study, the validation of the clinical models of severity and mortality had diagnostic accuracy of 86.09% and 73.7% respectively and predictions can be made appropriately. Data processing and protocol adjustments served as the basis for the challenges of subsequent waves
CLINICAL IMPLICATIONS: International collaboration has been the main mitigation tool for this new pathology that still presents us with new challenges
DISCLOSURE: Nothing to declare.
KEYWORD: COVID-19