Preliminary study to identify severe from moderate cases of COVID-19 using NLR&RDW-SD combination parameter

Abstract

Objectives: Investigate the characteristics and rules of hematology changes in patients with COVID-19, and explore the possibility to identify moderate and severe patients using conventional hematology parameters or combined parameters. Methods:The clinical data of 45 moderate and severe type patients with SARS-CoV-2 infections in Jingzhou Central Hospital from January 23 to February 13, 2020 were collected. The epidemiological indexes, clinical symptoms and laboratory test results of the patients were . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) : medRxiv preprint retrospectively analyzed. Those parameters with significant differences between the two groups were analyzed, and the combination parameters with best diagnostic performance were selected using the LDA method. Results:Of the 45 patients with COVID-19 (35 moderate and 10 severe cases), 23 were male and 22 female, aged 16-62 years. The most common clinical symptoms were fever (89%) and dry cough (60%). As the disease progressed, WBC, Neu#, NLR, PLR, RDW-CV and RDW-SD parameters in the severe group were significantly higher than that in the moderate group (P<0.05); meanwhile, Lym#, Eos#, HFC%, RBC, HGB and HCT parameters in the severe group were significantly lower than that in the moderate group (P<0.05). For NLR, the AUC, the best cut-off value, the sensitivity and the specificity were 0.890, 13.39, 83.3% and 82.4% respectively, and for PLR , the AUC, the best cut-off, the sensitivity and the specificity were 0.842, 267.03, 83.3% and 74.0% respectively.The combined parameter NLR&RDW-SD had the best diagnostic efficiency (AUC was 0.938) and when the cut-off value was 1.046, the sensitivity and the specificity were 90.0% and 84.7% respectively, followed by the fitting parameter NLR&RDW-CV (AUC = 0.923). When the cut-off value was 0.62, the sensitivity and the specificity for distinguishing severe type from moderate cases of COVID-19 were 90.0% and 82.4% respectively. Conclusions:The combined parameter NLR&RDW-SD is the best hematology index and can help clinicians to predict the severity of COVID-19 patients, and it can be used as a useful indicator to help prevent and control the epidemic.

Comparative analysis of laboratory indexes of severe and non-severe patients infected with COVID-19

“…The decrease of hemoglobin and PLT were significantly associated with the severity of the disease [33,34]. In addition, the combined parameter NLR&RDW-SD can help clinically to predict the severity of COVID-19 patients [35]. In conclusion, blood routine examination is of great value in the diagnosis and prognosis of COVID-19.…”

Predicting COVID-19 disease progression and patient outcomes based on temporal deep learning

“…LDA [14] NB [15] k-NN [16] LR [17,18] Cox [19] [17] RF [17] XGBoost [1] BPNN [17,19,20], GRNN [21], RBFNN [21], PNN [21] Basic CNN [22], Transfer CNN [23], GDCNN [24], COVID-Net [25], COVNet [26] /…”

“…We have summarized them in Table 1. For example, in most method of [27] and in [1,[14][15][16][17][18][19], authors used non-deep learning methods, such as k-NN, LR, Cox, SVM and DT to classify CT/X-ray images and predict the outcomes of COVID-19 patients. However, in terms of prediction accuracy, nondeep learning is not as good as deep learning methods.…”

Effects of COVID-19 on Respiratory System

“…The incubation period of this disease is 2-14 days, however, in most cases this period may take 4-5 days. It was stated that 80% of the patients had the disease with mild to moderate severity, 13.8% had severe illness and 6.1% had critically ill (respiratory failure, septic shock and/or multi-organ dysfunction or failure) (18). The most commonly observed symptoms are fever, fatigue, and dry cough.…”