Abstract
Aim: Lung injuries in patients with Coronavirus Disease 2019 (COVID-19) are often associated with severity scores. This study aimed to describe the relationship between clinical categorization and the severity of chest computed tomography (CT) scan features in a low-resource setting. This research adopted a retrospective, descriptive, and analytical study design to explore the data. Study Design: The study was carried out in the Intensive Care Unit (ICU) of the National COVID-19 Reference Hospital. Patients were classified into moderate and severe clinical forms, based on the World Health Organization (WHO) definitions of clinical syndromes associated with COVID-19. CT scans were categorized as moderate (≤50%) or severe (>50%) grades, according to the extent of lung injuries. The chi-square test or Fisher's exact test, along with logistic regression, were conducted using R software. Results: The study included 133 patients, with a mean age of 57.9±15.6 years and a sex ratio of 1.2. Comorbidities were present in 84.2% of patients, who presented with moderate (41.3%) and severe (58.7%) clinical forms. Lung lesions were categorized as moderate (45.1%) and severe (54.9%) grades. Clinical severity was associated with the extent of lung lesions on CT scans (p<0.001). Diabetes (p=0.01), low blood pressure (p=0.04), oxygen saturation levels below 85% (SpO2<85%; p=0.04), and respiratory distress (p=0.02) were associated with severe clinical forms. Obesity (p=0.01), SpO2<85% (p=0.04), and respiratory distress (p=0.02) were associated with high-grade findings of CT scans. Conclusions: Clinical severity in COVID-19 patients was associated with the severity of pulmonary CT scan findings. This clinical categorization could be useful in low-resource settings to guide the management of COVID-19 patients.