Abstract
In this study we aimed to evaluate the ability of IMPROVE and IMPROVE-DD scores in predicting in-hospital mortality in patients with severe COVID-19. This prospective observational study included adult patients with severe COVID-19 within 12 h from admission. We recorded patients' demographic and laboratory data, Charlson comorbidity index (CCI), SpO
2
at room air, acute physiology and chronic health evaluation II (APACHE II), IMPROVE score and IMPROVE-DD score. In-hospital mortality and incidence of clinical worsening (the need for invasive mechanical ventilation, vasopressors, renal replacement therapy) were recorded. Our outcomes included the ability of the IMPROVE and IMPROVE-DD to predict in-hospital mortality and clinical worsening using the area under receiver operating characteristic curve (AUC) analysis. Multivariate analysis was used to detect independent risk factors for the study outcomes. Eighty-nine patients were available for the final analysis. The IMPROVE and IMPROVE-DD score showed the highest ability for predicting in-hospital mortality (AUC [95% confidence intervals {CI}] 0.96 [0.90–0.99] and 0.96 [0.90–0.99], respectively) in comparison to other risk stratification tools (APACHE II, CCI, SpO
2
). The AUC (95% CI) for IMPROVE and IMPROVE-DD to predict clinical worsening were 0.80 (0.70–0.88) and 0.79 (0.69–0.87), respectively. Using multivariate analysis, IMPROVE-DD and SpO
2
were the only predictors for in-hospital mortality and clinical worsening. In patients with severe COVID-19, high IMPROVE and IMOROVE-DD scores showed excellent ability to predict in-hospital mortality and clinical worsening. Independent risk factors for in-hospital mortality and clinical worsening were IMPROVE-DD and SpO
2
.