Assessing the Severity of COVID‐19 Lung Injury in Rheumatic Diseases Versus the General Population Using Deep Learning–Derived Chest Radiograph Scores
Authors' conclusion
Rheumatic disease patients with COVID-19 had more severe radiographic lung involvement than comparators.
Study context
- Population
- COVID-19 patients with ≥1 chest radiograph within ±2 weeks of diagnosis: rheumatic disease patients and matched general population comparators
- Participants
- 70 vs 463
- Exposure
- Rheumatic disease patients with COVID-19 vs matched general population comparators
- Rheumatic diseasen = 70
- Comparatorsn = 463reference
- Outcome
- Maximum pulmonary x-ray severity (PXS) score (deep learning–derived) at the 10th to 90th percentiles
Methodological detailsmatched comparators, quantile regression (unadjusted and adjusted for age, Charlson comorbidity index and ILD), Cox regression
- Matched comparators
- Quantile regression (unadjusted and adjusted for age, Charlson comorbidity index and ILD)
- Cox regression
Main finding
As reported in the abstract
Maximum PXS score, 90th percentile
Rheumatic disease
10.2
90th percentile
Comparators
9.2
90th percentile
Adjusted P = 0.03
Observational data: associations, not effects.
Further findings
Severe PXS score (>9)
P = 0.02
| Outcome | HR (95% CI) | |
|---|---|---|
| Mechanical ventilation in rheumatic disease patients, PXS >9vs ≤9 | 24.1 (6.7–86.9) | |
| Death in rheumatic disease patients, PXS >9vs ≤9 | 8.2 (0.7–90.4) | |
| 0.050.21520Hazard ratio (log scale) |
Original publication
Assessing the Severity of COVID‐19 Lung Injury in Rheumatic Diseases Versus the General Population Using Deep Learning–Derived Chest Radiograph Scores. Arthritis Care Res (Hoboken). Published 21 October 2022. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →