Brixia Score for Predicting Mortality and Length of Stay in COVID-19 Confirmed Patients at the Hospital in Bandung

Abstract

On March 11, 2020, the World Health Organization declared the COVID-19 pandemic. This disease damages the lung and resulting mild to severe pneumonia. This study aimed to determine the value of the Brixia score for predicting mortality and length of stay of COVID-19 confirmed patients. The study design was case-control with secondary data from digital medical records of COVID-19 confirmed patients (December 2020 to February 2021). All patients' chest x-rays (CXR) were scored using the Brixia score. Logistic regression and the Spearman rank correlation test were used to identify mortality and length of stay predictors. There were 636 subjects included in this study, with the proportion of deceased patients (case group) being 20.3% (95% CI=17.33, 23.59%). Most CXR findings had signs of pneumonia (95.1%), including ground-glass opacities (GGOs) mixed with consolidation. The distribution of GGOs and consolidation were most frequent in the peripheral of survived patients (83.9%), while the deceased group had peripheral involvements mixed with medial (45.0%) and bilateral (22.2%). The mean Brixia score in the group of decease patients was significantly higher than the group of survived patients (11.95 vs 6.73, p=0.00). Brixia score had an OR of 1.14, 95% CI=1.07, 1.21 after adjusting by age, SpO2 level, and comorbid. The chance of dying was higher than 50% if the Brixia score reached to score of 15 (probability=49%, 95% CI=41, 56%). However, the Brixia score has no significant correlation with length of stay (rho=0.05, p=0.24). In conclusion, the CXR Brixia score can predict mortality, but it can not predict the length of stay of hospitalized COVID-19 confirmed patients.