Predictors of Severity and Mortality Among Patients with Coronavirus Disease 2019: A Systematic Review and Meta-Analysis

Background: The COVID-19 pandemic has overwhelmed the global health systems, and it is imperative to understand how to effectively triage patients. Deeper understanding of predictors of disease severity and mortality is pivotal to effectively triage patients with COVID-19 to maximize the benefit of scarce intensive care unit resources, while minimizing the potential harm of outpatient management of ill patients. Methods: We performed a systematic review and meta-analysis of observational studies, assessing factors associated with severity and mortality among laboratory confirmed COVID-19 patients. We searched PubMed, Embase and WHO database for articles up to May 8, 2020. Randomized trials were excluded. Odds ratios (with 95% CI) and risk ratios (with 95% CI) were used to determine the association between the various demographic, clinical, laboratory and radiological factors and the development of severe disease or mortality. We performed meta-regression to determine the percentage change in the occurrence of the outcomes. Heterogeneity across studies were assessed using I2 and Tau2 statistics. Findings: Among 15680 articles obtained from the literature search, 109 articles were included in the analysis. Increasing age and male gender were associated with higher mortality rates and severe disease. The risk of mortality was higher in patients presenting with dyspnea (RR 2·55, 95% CI 1·88–2·46) and hemoptysis (RR 1·62, 95%CI 1·25–2·11). Co-morbidities such as diabetes (RR 1·59, 95%CI 1·41–1·78), hypertension (RR 1·90, 95%CI 1·69–2·15), cardiovascular diseases (RR 2·27, 95% CI 1·88–2·79) and chronic obstructive pulmonary disease (RR 2·29, 95% CI 1·90–2·75) were associated with a higher risk of death. In-hospital complications such as acute respiratory distress syndrome (ARDS), sepsis, shock and acute cardiac injury had adverse outcomes, with ARDS having the highest risk ratio (RR 20·19, 95% CI 10·87–37·52). Lung consolidation on computed tomography (CT) had significant association with death (RR 2·07, 95% CI 1·35–3·16). Congestive heart failure (OR 4·76, 95% CI 1·34–16·97) had greater odds of developing severe disease. Among the radiological features, hilar lymphadenopathy (OR 8·34, 95%CI 2·57–27·08), bilateral lung involvement (OR 4·86, 95%CI 3·19–7·39) and reticular pattern (OR 5·54, 95%CI 1·24–24·67) were more frequently seen in patients with severe disease. Patients with leukocytosis, lymphopenia, elevated C-reactive protein and D-dimer levels had higher odds of severe disease and greater risk of mortality. Interpretation: Our study identified several important predictors of disease severity and mortality among patients with COVID-19. Knowledge of these predictors might help in the prioritization and management of these patients. Funding: NoneDeclaration of Interests: The authors declare no competing interests.