Front Med (Lausanne). 2020 Jul 3;7:324. doi: 10.3389/fmed.2020.00324. eCollection 2020.
Aim: Clinical findings indicated that a fraction of coronavirus disease 2019 (COVID-19) patients did not show fever. However, the difference between the clinical characteristics of fevered and non-fevered patients is still unclear. The aim of the present study was to describe the clinical characteristics of these patients and analyze the predictors for severe events of adult fevered COVID-19 patients. Methods: Clinical and laboratory data of fevered and non-fevered COVID-19 patients in Changsha, China, were collected and analyzed. Logistic regression analysis and Receiver Operating Characteristic Curve (ROC Curve) analysis were adopted to analyze risk factors and evaluate the effectiveness of the predictors for severe events in adult fevered COVID-19 patients. Results: Of the 230 adult COVD-19 patients in this study, 175 patients (76.1%) had fever and 55 patients (23.9%) did not have fever. Compared with non-fevered patients, the fevered patients showed a lower lymphocyte proportion (P = 0.000) and lymphocyte count (P = 0.000) as well as higher levels of C-reactive protein (CRP) (P = 0.000) and erythrocyte sedimentation rate (P = 0.000). The proportion of severe cases was significantly elevated in adult fevered patients (P = 0.000). Compared to non-severe fevered patients, severe fevered patients showed a lower lymphocyte count (P = 0.000), a lower lymphocyte proportion (P = 0.000), and higher levels of CRP (P = 0.000). As determined by the multivariate analysis, CRP (OR 1.026, P = 0.018) and lymphocyte proportion (OR 0.924, P = 0.009) were significantly associated with the risk of developing severe events in fevered adult COVID-19 patients. Furthermore, ROC Curve analysis revealed that the area under the curve (AUC) for CRP combined with lymphocyte proportion to diagnose severe events in fevered adult COVID-19 patients was 0.874 (95% CI 0.820-0.927). Conclusions: Adult fevered COVID-19 patients were more likely to progress into severe cases, while CRP and lymphocyte proportion were effective predictors for developing severe events in these patients.