Pulmonary Embolism Severity Index Predicts Adverse Events in Hospitalized COVID-19 Patients: A Retrospective Observational Study

Abstract

ObjectivesPulmonary embolism is one of the leading causes of death in patients with COVID-19. Autopsy findings showed that the incidence of thromboembolic events was higher than clinically suspected. In this study, the authors investigated the relationship between pulmonary embolism severity index (PESI) and simplified PESI (sPESI) on admission to the hospital, as well as adverse events in hospitalized COVID-19 patients without clinically documented venous and/or pulmonary embolism. The adverse events investigated were the development of acute respiratory distress syndrome, the need for intensive care unit admission, invasive or noninvasive mechanical ventilation, and in-hospital mortality.DesignA retrospective and observational study.SettingTwo large-volume tertiary hospitals in the same city.ParticipantsA total of 720 hospitalized COVID-19 patients with a positive polymerase chain reaction were evaluated.InterventionsNone.Measurements and Main ResultsOf the study population, 48.6% (350) were women, and the median age was 66 years (19-96). The overall in-hospital mortality rate was 20.5%. In the multivariate logistic regression analysis, a significant relationship was found between the whole adverse events considered and PESI, as well as sPESI (p < 0.001). According to the results, sPESI ≥2 predicts in-hospital mortality with a sensitivity of 61.4% and specificity of 83.3% (area under the curve = 0.817, 95% confidence interval 0.787-0.845, p < 0.001). Similarly, PESI classes IV and V also were found as independent risk factors for in-hospital mortality (for PESI class IV, odds ratio = 2.81, p < 0.017; for PESI class V, odds ratio = 3.94, p < 0.001).ConclusionsPESI and sPESI scoring systems were both found to be associated with adverse events, and they can be used to predict in-hospital mortality in hospitalized COVID-19 patients without documented venous and/or pulmonary embolism.