Can the Progression of COVID-19 Pneumonia be Predicted?

Abstract

Background: Coronavirus disease-2019 (COVID-19) remains a major cause of morbidity and mortality. There are many parameters affecting the progression of the disease. The purpose of the present study was to evaluate and compare the initial data of patients hospitalized with the diagnosis of COVID-19 pneumonia, who progressed during the hospitalization period, with other patients who recovered or remained stable, and to investigate the risk factors that can be used to predict the disease progression. Materials and Methods: Patients, who received inpatient treatment with the diagnosis of COVID-19 pneumonia, were included in the study retrospectively. Two groups were created from all patients according to their progression in hospital follow-ups: Group 1: Progression group and group 2: Recovery/stabilization group. If patients had clinical, laboratory and/or radiological deterioration or died during follow-up, these patients were included in the progression group. If patients recovered or remained stable, these patients were also included in the recovery/stabilization group. The demographic data, initial hemogram, biochemical parameters and radiological data of the patients were recorded. Results: It was determined in the univariate analysis that the age, smoking status, comorbidity, heart disease, chronic obstructive pulmonary disease, cancer, dyspnea, fever, leukocytosis, lymphopenia, elevated neutrophil-lymphocyte ratio (NLR), C-reactive protein, albumin, lactate dehydrogenase, ferritin, D-dimer, troponin-T, pro-B-type natriuretic peptide (pro-BNP) were risk factors predicting disease progression all p-values<0.05. In the multivariate logistic regression analysis, it was found that fever, NLR, and D-dimer could be used to predict the disease progression (p<0.05). In the ROC analysis, the sensitivity of NLR was 83.3%, specificity 57.5%, and cut-off >3.545 [area under curve (AUC)=0.752; p<0.001]; the sensitivity of pro-BNP was 71.8%, specificity 73.8%, and cut-off >332.8 (AUC=0.752; p<0.001), the sensitivity of troponin-T was 81.2%, specificity was 60.6%, and cut-off was >4.58 (AUC=0.730; p<0.001) in predicting progression. Conclusion: The identification of risk factors predicting progression is important in reducing morbidity and mortality rates. Fever, NLR, D-dimer troponin-T and pro-BNP are important parameters that can be used to predict progression.