Can the Chest X-ray Brixia Score predict COVID-19 disease severity and mortality in resource-limited settings? A retrospective cross-sectional study at a tertiary hospital in Northwestern, Tanzania

Résumé fourni par la source

Abstract Background COVID-19 primarily affects the lung, thus the chest x-ray (CXR) is the first line imaging modality of evaluating the disease. CXR scoring systems for quantifying the severity and progression of lung abnormalities in COVID-19 pneumonia have been introduced. However, a chest X-ray severity scoring system has never been implemented in Tanzania to assist in predicting COVID-19 patient outcomes. Therefore, this study was designed to determine whether the CXR Brixia Score can predict outcomes among patients with COVID-19 pneumonia. Materials and Methods A retrospective cross-sectional hospital-based study conducted on COVID-19 pneumonia patients confirmed by nasopharyngeal swab RT-PCR assay at Bugando Medical Center (BMC) using data retrieved from the hospital database. Data retrieved included socio-demographics and clinical information. Baseline CXRs were reviewed and assigned a Brixia score by two experienced Radiologists. Disease severity, length of the hospital stay and in-hospital mortality were recorded as outcomes and correlated with the Brixia scores by logistic regression analysis after adjusting for potential cofounders. A p-value of <0.05 was significant. Results A total of 220 patients were enrolled with a mean age of 59 (±12.7) years. A severe Brixia score had a significant likelihood of severe form of disease (aOR=4.8, 95%CI=2.35 - 13.67, p=0.001) and death (aOR=5.39, 95%CI=1.65 - 24.73, p=0.041). A severe Brixia score did not predict a longer hospital stay (cOR=2.29, 95%CI=0.49 - 9.72, p=0.293). Conclusion CXR Brixia Score can predict disease severity and in-hospital mortality particularly in resource-limited settings. It may assist in timely identifying patients in need of aggressive management, and thereby avert adverse outcomes including mortality.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.