Prognostic Factors for Predicting COVID-19 Severity and Mortality

Abstract

Background: COVID-19 has become a serious health problem worldwide. Objectives: The current study investigated the prognostic factors associated with demographical parameters, clinical and vital signs, and laboratory results for predicting severity and mortality in patients infected with COVID-19. Methods: This retrospective analysis was conducted on the medical records of 372 COVID-19-positive patients hospitalized at the Khatam al-Anbiya Hospital, Shoushtar, Iran, from Sep 2020 to Sep 2021. The association of demographic parameters, clinical and vital signs, and laboratory results with severity and patients' outcomes (survival/mortality) was studied. The patients were divided into the non-severe group (n = 275) and the severe group (n = 97). COVID-19 disease severity was determined based on the severity of pulmonary involvement using CT chest images. The collected data were analyzed using IBM SPSS software for Windows (version 18). Logistic regression analysis was employed using the Forward LR method to predict COVID-19 severity and mortality. Results: The rates of mortality and the severe form of the disease were 87.1% (n = 324) and 12.9% (n = 48), respectively. A prognostic value was observed in predicting COVID-19 severity and mortality for some clinical and vital signs (diabetes (P < 0.001, P = 0.019), hypertension (P = 0.024, P = 0.012), pulmonary diseases (P = 0.038, P < 0.001), and anosmia (P = 0.043, P = 0.044) and paraclinical parameters (FBS (P = 0.014, P = 0.045), BUN (P = 0.045, 0.001), Cr (P = 0.027, P = 0.047), Neut (P = 0.002, P = 0.005), and SpO2 (P = 0.014, P = 0.001)). Cardiovascular disorders (P = 0.037), fever (P = 0.008), and dyspnea (P = 0.020) were also effective at predicting disease-related mortality. Multiple logistic regression analyses showed that diabetes disease, the place of residence, PCO2, and BUN with R2 = 0.18, and age, pulmonary diseases, and BUN with R2 = 0.21 were involved in predicting the severity and mortality, respectively. Conclusions: It seems that in addition to the BUN, diabetes and pulmonary diseases play a more significant role in predicting the severity and mortality due to COVID-19, respectively.