TL;DR
This study provides a detailed description of chest CT findings in patients infected with SARS-CoV-2, identifying common patterns like ground-glass opacities and peripheral lesions. These CT features are crucial for early diagnosis and monitoring of COVID-19, especially when clinical symptoms are ambiguous or exposure history is unclear.
What Problem Does This Paper Address?
The emergence of SARS-CoV-2 (the virus causing COVID-19) in late 2019 led to a global health crisis. While clinical and epidemiological features of the disease were being documented, there was a significant need for understanding its characteristic manifestations on medical imaging, specifically chest CT scans. Early and accurate diagnosis is critical for patient management and public health interventions. This paper aims to fill this gap by systematically describing the CT findings in a cohort of confirmed SARS-CoV-2 infected patients, providing a visual reference for radiologists and clinicians.
Methods
The researchers conducted a retrospective review of chest CT scans from 67 patients hospitalized with confirmed SARS-CoV-2 infection. The CT data was collected between January 23rd and February 10th, 2020. A standardized approach was used for image analysis, with two senior radiologists independently reviewing the scans. A third radiologist was consulted for any discrepancies. The analysis focused on identifying and categorizing various CT features, including:
- Lesion distribution: Peripheral vs. central, and upper/middle vs. lower lobes.
- Lesion characteristics: Ground-glass opacities (GGO), consolidation, air bronchogram, reversed halo sign, interlobular septal thickening, pleural thickening/effusion, air cavities, bronchial wall thickening, tree-in-bud pattern, and intrathoracic lymph node enlargement.
- Number of lesions: Single vs. multiple.
The study also collected demographic, clinical, and laboratory data to correlate with the imaging findings.
Datasets
The study utilized a dataset of 67 chest CT scans from patients diagnosed with SARS-CoV-2 infection at Ruian People's Hospital. The data was collected between January 23rd, 2020, and February 10th, 2020. This dataset is a newly collected cohort for the purpose of this study and is not a publicly released benchmark.
Key Insights and Ideas
- CT as a Diagnostic Tool: The study strongly suggests that chest CT plays a vital role in the early diagnosis of COVID-19, especially when exposure history is unclear or nucleic acid tests are pending.
- Typical vs. Atypical Findings: The research clearly delineates common CT patterns (ground-glass opacities, peripheral and multiple lesions, lower lobe involvement) from less frequent ones, helping to refine diagnostic criteria.
- Early Stage Manifestations: The average time interval of 4.6 days between symptom onset and CT examination indicates that these findings represent the early stages of the disease.
- Progression and Recovery: The observation of some patients showing progression of CT findings during treatment, followed by recovery, highlights the dynamic nature of the disease on imaging and the potential for CT to monitor treatment response.
- Distinguishing Features: While many viral pneumonias share some CT features, the specific combination and prevalence of findings in SARS-CoV-2 patients are important for differentiation.
Key Results
- Prevalence of Abnormalities: Out of 67 patients, 54 (80.6%) showed positive CT findings, while 13 (19.4%) had negative CT scans.
- Most Common Findings:
- Ground-glass opacities (GGO): 42/54 patients (77.8%)
- Peripheral lesions: 50/54 patients (92.6%)
- Multiple lesions: 46/54 patients (85.2%)
- Lower lobe involvement: 42/54 patients (77.8%)
- Less Common Findings: Air bronchogram (33.3%), pleural thickening or effusion (25.9%), consolidation (22.2%), interlobular septal thickening (20.4%), reversed halo sign (16.7%).
- Rare Findings: Tree-in-bud pattern (1.9%), bronchial wall thickening (5.6%), intrathoracic lymph node enlargement (3.7%).
- Clinical Correlation: Fever (86.6%) and cough (77.6%) were the most common clinical symptoms, aligning with the imaging findings.
Limitations and Future Work
- Small Cohort Size: The study acknowledges that the patient cohort was small, and only a limited number of severe cases were included.
- Lack of Sequential CT: Due to the urgency of reporting findings, sequential CT studies were not performed.
- Future Directions: The authors plan to collect data from a larger patient population in future studies to further evaluate longitudinal changes in CT features during treatment. They also aim to develop a predictive model for clinical outcomes based on CT imaging.
Summary updated