Clinical Risk Factors for Mortality of Hospitalized Patients with COVID-19: Systematic Review and Meta-Analysis

Guiling Xiang,Liang Xie,Zhihong Chen,Shengyu Hao,Cuiping Fu,Qiao Wu,Xuhui Liu,Shanqun Li
DOI: https://doi.org/10.21037/apm-20-1278
2021-01-01
Annals of Palliative Medicine
Abstract:Background: New evidence from retrospective cohort studies on risk of death from COVID-19 infection became available.We aimed to systematically review the clinical risk factors for fatal outcome of COVID-19. Methods:We performed meta-analysis, using PubMed, EMBASE and Cochrane databases from December 1 2019 to June 10 2020.The meta-analysis summarized clinical, laboratory, radiological features, and complications of non-survivors with confirmed COVID-19.In addition, a fixed-or random-effects model was adopted based on the heterogeneity among studies.We also used funnel-plot with Egger's tests to screen potential publication bias.Results: In total, twenty studies with 15,408 COVID-19 cases were included in our meta-analysis.Male, current smoking, and older age were associated with in-hospital death.Patients aged 60 years or over had the highest pooled ORs [OR 4.94 (2.89, 8.44)].Non-survivors were more likely to have diabetes, hypertension, cardiovascular disease (CVD), respiratory disease, or chronic kidney disease (CKD).Respiratory disease had the highest pooled ORs [OR 2.55 (2.14, 3.05)].Dyspnea [OR 3.31 (1.78, 6.16);I 2 : 83%] and fatigue [OR 1.36 (1.07, 1.73);I 2 : 0%] were associated with increased risk of death.Increased white blood cell count, decreased lymphocyte and platelet counts, were also associated with increased risk of death.Biomarkers of coagulation function, inflammation, liver and kidney function, cardiac and muscle injury were also elevated in nonsurvivors.Conclusions: Male, current smoking patients aged 60 years or over might face a greater risk of in-hospital death and the comorbidities such as diabetes, hypertension, CVD, respiratory disease, and CKD could also influence the prognosis of the COVID-19.Clinical feature such as dyspnea and fatigue could imply the exacerbation and even death.Our findings highlighted early markers of mortality which were beneficial to identify fatal COVID-19.
What problem does this paper attempt to address?