Patients with rheumatic disease and COVID-19 had more severe radiographic lung involvement when compared with controls, according to a study published in Arthritis Care & Research.1
“Since rheumatic disease patients are more likely than the general population to have baseline chronic lung disease, it is unclear whether Pulmonary X-Ray Severity (PXS) scores may reliably predict risk of severe COVID-19 outcomes in this population,” investigators explained. “If so, deep learning-derived x-ray scores may be leveraged for use in larger studies evaluating the relationships between rheumatic disease, immunosuppression, imaging results, and COVID-19 outcomes.”
A comparative cohort study of patients at Massachusetts General Hospital and Brigham and Women’s Hospital was conducted between patients with rheumatic disease, newly diagnosed COVID-19, and ≥1 chest radiograph, and controls. Eligible patients were aged 18 years or older and were seen between January 31, 2020 and January 31, 2021. Matched controls were based on age (± 5 years), sex, hospital, and date of positive COVID-19 result (± 14 days). Clinical outcomes were obtained from the COVID-19 Data Mart via the Mass General Brigham (MGB) electronic heath records (EHR).
Both unadjusted and adjusted regression were used to compare the maximum PXS score at the 10th and 90th percentiles, with a score of >90 defined as severe PXS. The association of PXS >9 and mechanical ventilation and death was evaluated using Cox regression.
A total of 70 patients with rheumatic diseases were compared with 463 controls (mean age 62 years, with 69% females diagnosed with rheumatic disease and 68% in the control group). Of the 70 patients, 206 chest radiographs were analyzed within the initial 2-week period of COVID-19 diagnosis, compared with 936 radiographs in the control cohort.
While maximum PXS scores were comparable between both patients and controls in the 10th and 60th percentiles (50th percentile scores of 3.1 vs 3.2, respectively), they were significantly higher between the 70th and 90th percentiles, with a 90th percentile score of 10.2 for patients with rheumatic disease vs 9.2 for controls, adjusted p=0.03. PXS scores were still significantly higher in patients with rheumatic disease even after adjusting for age via the Charlson Comorbidity Index (CCI) (90thpercentile scores of 10.2 vs. 9.2, p=0.03).
Patients with rheumatic disease were more likely to have a PXS score >9 (20% vs 11%, p=0.02), were more likely to require mechanical ventilation (578.9 vs 2.7 events/1,000 months; HR 24.1 [95% CI: 6.7, 86.9]), and had a higher risk of death (7.5 vs 0.9 events/1,000 months; HR 8.2 [95% CI: 0.7, 90.4]).
Matching patients with controls to the site of both hospitals to account for differences in technique strengthened the study. Further, data regarding diagnosis and care was obtained via a data warehouse. However, chest radiographs were not available for all patients with rheumatic disease, which may have created bias toward patients with more severe disease. Additionally, generalizability may have been reduced as the study was conducted within the same geographic region and the sample size reduced the ability to ascertain the impact of confounders or mediators regarding rheumatic disease and PXS scores. The small sample size also limited subgroup analyses of patients with different rheumatic diseases and those who were receiving different immunosuppressive medications.
“Our study confirms that rheumatic disease patients are more susceptible to lung involvement from COVID-19 and corroborates previous findings suggesting increased risk for mechanical ventilation and mortality,” investigators concluded. “The higher risk of severe pulmonary involvement from COVID-19 provides additional support for COVID-19 vaccination and risk mitigation among rheumatic disease patients.”
Patel NJ, D'Silva KM, Li MD, et al. Assessing the Severity of COVID-19 Lung Injury in Rheumatic Diseases versus the General Population Using Deep Learning-Derived Chest Radiograph Scores [published online ahead of print, 2022 Mar 21]. Arthritis Care Res (Hoboken). 2022;10.1002/acr.24883. doi:10.1002/acr.24883