AUTHOR=Tang Guoxing , Luo Ying , Lu Feng , Li Wei , Liu Xiongcheng , Nan Yucen , Ren Yufei , Liao Xiaofei , Wu Song , Jin Hai , Zomaya Albert Y. , Sun Ziyong 

TITLE=Prediction of Sepsis in COVID-19 Using Laboratory Indicators

JOURNAL=Frontiers in Cellular and Infection Microbiology

VOLUME=Volume 10 - 2020

YEAR=2021

URL=https://www.frontiersin.org/journals/cellular-and-infection-microbiology/articles/10.3389/fcimb.2020.586054

DOI=10.3389/fcimb.2020.586054

ISSN=2235-2988

ABSTRACT=Background The outbreak of coronavirus disease 2019 (COVID-19) has become a global public health concern. Many inpatients with COVID-19 have shown clinical symptoms related to sepsis, which will aggravate the deterioration of patients’ condition. We aim to diagnose Viral Sepsis Caused by SARS-CoV-2 (VSCS-2) by analyzing laboratory test data of patients with COVID-19 and establish an early predictive model for sepsis risk.
Methods This study retrospectively investigated laboratory test data of 2453 patients with COVID-19 from electronic health records. Extreme gradient boosting (XGBoost) was employed to build four models with different feature subsets of a total of 69 collected indicators. Meanwhile, the Shapley Additive exPlanation (SHAP) method was adopted to interpret predictive results and analyze the feature importance of risk factors.
Findings Model for classifying CSVS-2with seven coagulation function indicators achieved the area under the receiver operating characteristic curve (AUC) 0.9213 (95% CI, 89.94-94.31%), sensitivity 97.17% (95% CI, 94.97-98.46%), and specificity 82.05%(95% CI, 77.24-86.06%). Model for identifying COVID-19 coagulation disorders with eight features provided an average of 3.68()4.60 days for early warning prediction. The model showedan AUC of 0.9298 (95% CI, 86.91-99.04%), sensitivity of 82.22% (95% CI, 67.41-91.49%), and specificity of 84.00% (95% CI, 63.08-94.75%).
Interpretation We found abnormality of the coagulation function was related to the occurrence of sepsis and the other routine laboratory test represented by inflammatory factors had a moderate predictive value on coagulopathy. This finding suggested that the early warning of CSVS-2patients could be achieved by our established model to improve the patient’s prognosis and reduce the mortality.