RÉSUMÉ
OBJECTIVES@#To investigate the clinical characteristics and risk factors for severe events of coronavirus disease 2019 (COVID-19) in elderly patients.@*METHODS@#Retrospective analysis was performed on the clinical data of all elderly COVID- 19 patients treated in Changsha Public Health Treatment Center from January 17, 2020 to March 15, 2020, which included basic diseases, symptoms, test results, and other clinical characteristics, and prognostic indicators such as severity of illness, length of hospital stay, virus shedding time and mortality rate. The differences in clinical characteristics and prognostic indicators between elderly, middle-aged, and young COVID-19 patients were also analyzed. Logistic regression model was used to conduct univariate and multivariate analysis of risk factors for developing severe events in elderly COVID-19 patients; receiver operating characteristic (ROC) curve analysis was used to evaluate the prediction efficacy.@*RESULTS@#Of the 230 COVID-19 adult patients, 34 were young patients (14.8%), 136 were middle-aged patients (59.1%), and 60 were elderly (26.1%). Among the 60 elderly patients, 23 were male (38.3%) and 37 were female (61.7%), with a medium age of 66 years old. Common symptoms were fever (66.7%), cough (50.0%), and fatigue (41.7%). C reactive protein (CRP) was increased significantly. The proportion of severe cases was 31.7%, and mortality was 1.7%. The median length of hospitalization and median virus shedding time were 18.5 days and 21 days, respectively. Compared with the young and the middle-aged patients, the elderly had a higher proportion of hypertension, diabetes, and cardiovascular diseases, more common shortness of breath, higher proportions of pneumonia and severe cases (all <0.05), and the decreased lymphocyte count and lymphocyte percentage (both <0.05), as well as higher CRP and erythrocyte sedimentation rate (ESR) levels (both <0.05). Compared with non-severe cases, severe elderly patients demonstrated higher CRP and aspartate aminotransferase (AST) levels (all <0.05), the reduced lymphocyte count (<0.05), and the prolonged length of hospitalization and virus shedding duration (both <0.05). Univariate logistic regression analysis indicated that the lymphocytes proportion, CRP and AST levels were significantly correlated with the risk for developing severe events in elderly COVID-19 patients (all <0.05). Multivariate logistic regression found that severe events in elderly patients with COVID-19 were significantly correlated with CRP level (OR=1.041, =0.013). ROC curve analysis revealed that the area under the curve (AUC) for CRP to diagnose severe events in elderly COVID 19 patients was 0.851.@*CONCLUSIONS@#The proportion of severe cases in elderly COVID-19 patients is higher than that in young and middle-aged patients. CRP level has a good predictive value for the possibility of severe events in elderly COVID-19 patients.