The Predictive Effectiveness of Blood Biochemical Indexes for the Severity of COVID-19

OBJECTIVE: We aimed to explore the predictive effectiveness of blood biochemical indexes for COVID-19 severity. METHOD: We retrospectively analyzed the clinical data of COVID-19 patients who were cured and discharged from the Public Health Treatment Center of Changsha from January 30, 2020, to February 19, 2020. According to the clinical classification of the disease, the patients were divided into severe and nonsevere groups. General clinical data and underlying medical conditions were recorded through the electronic medical record (EMR) system. Laboratory examination results of the patients during their hospitalization were collected, including the first results for erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), peripheral blood lymphocyte ratio and count, and peripheral blood white blood cell (WBC) count. Univariate and multivariate logistic regression models were used to analyze the predictive effectiveness of blood biochemical indexes and other related factors for COVID-19 severity. RESULT: In all, 108 COVID-19 patients (median age: 43.9 years (range: 1–75); male patients: 56 (51.85%)) were enrolled, of whom 24 (22.22%) showed severe disease and 84 (77.78%) showed nonsevere disease, and two in 24 patients with severe disease developed into a critically severe type and died. Fever was the most common onset symptom (67.59%), followed by cough (48.15%) and fatigue (37.04%). Comorbidities were important factors affecting the severity of COVID-19, and among the patients with severe disease, the proportion with comorbidities was 70.83%, and the proportion without comorbidities was 29.17%. The intergroup difference was significant (P < 0.05). In patients with CRP levels (mg/L) of ≤8, >8–≤20, >20–≤40, and >40, the proportions of those with severe and nonsevere disease were 0 to 32, 7 to 19, 6 to 23, and 11 to 10, respectively; the intergroup difference was significant (P < 0.05). CONCLUSION: The presence or absence of comorbidities and CRP elevation were independent significant predictors of COVID-19 severity, and hypertension was found as the most common comorbidity in patients with severe disease.