ABSTRACT
Background: An increased risk of severe COVID-19 outcomes may be seen in patients with autoimmune diseases on moderate to high daily doses of glucocorticoids, as well as in those with comorbidities. However, specific information about COVID-19 outcomes in SLE is scarce. Objectives: To determine the characteristics associated with severe COVID-19 outcomes in a multi-national cross-sectional registry of COVID-19 patients with SLE. Methods: SLE adult patients from a physician-reported registry of the COVID-19 GRA were studied. Variables collected at COVID-19 diagnosis included age, sex, race/ethnicity, region, comorbidities, disease activity, time period of COVID-19 diagnosis, glucocorticoid (GC) dose, and immunomodulatory therapy. Immunomodulatory therapy was categorized as: antimalarials only, no SLE therapy, traditional immunosuppressive (IS) drug monotherapy, biologics/targeted synthetic IS drug monotherapy, and biologic and traditional IS drug combination therapy. We used an ordinal COVID-19 severity outcome defined as: not hospitalized/hospitalized without supplementary oxygen;hospitalized with non-invasive ventilation;hospitalized with mechanical ventilation/extracorporeal membrane oxygenation;and death. An ordinal logistic regression model was constructed to assess the association between demographic characteristics, comorbidities, medications, disease activity and COVID-19 severity. This assumed that the relationship between each pair of outcome groups is of the same direction and magnitude. Results: Of 1069 SLE patients included, 1047 (89.6%) were female, with a mean age of 44.5 (SD: 14.1) years. Patient outcomes included 815 (78.8%) not hospitalized/hospitalized without supplementary oxygen;116 (11.2) hospitalized with non-invasive ventilation, 25 (2.4%) hospitalized with mechanical ventilation/ extracorporeal membrane oxygenation and 78 (7.5%) died. In a multivariate model (n=804), increased age [OR=1.03 (1.01, 1.04)], male sex [OR =1.93 (1.21, 3.08)], COVID-19 diagnosis between June 2020 and January 2021 (OR =1.87 (1.17, 3.00)), no IS drug use [OR =2.29 (1.34, 3.91)], chronic renal disease [OR =2.34 (1.48, 3.70)], cardiovascular disease [OR =1.93 (1.34, 3.91)] and moderate/ high disease activity [OR =2.24 (1.46, 3.43)] were associated with more severe COVID-19 outcomes. Compared with no use of GC, patients using GC had a higher odds of poor outcome: 0-5 mg/d, OR =1.98 (1.33, 2.96);5-10 mg/d, OR =2.88 (1.27, 6.56);>10 mg/d, OR =2.01 (1.26, 3.21) (Table 1). Conclusion: Increased age, male sex, glucocorticoid use, chronic renal disease, cardiovascular disease and moderate/high disease activity at time of COVID-19 diagnosis were associated with more severe COVID-19 outcomes in SLE. Potential limitations include possible selection bias (physician reporting), the cross-sectional nature of the data, and the assumptions underlying the outcomes modelling.
ABSTRACT
Background: Targeted DMARDs may dampen the inflammatory response in COVID-19, perhaps leading to a less severe clinical course. However, some DMARD targets may impair viral immune defenses. Due to sample size limitations, previous studies of DMARD use and COVID-19 outcomes have combined several heterogeneous rheumatic diseases and medications, investigating a single outcome (e.g., hospitalization). Objectives: To investigate the associations of baseline use of biologic or targeted synthetic (b/ts) DMARDs with a range of poor COVID-19 outcomes in rheumatoid arthritis (RA). Methods: We analyzed voluntarily reported cases of COVID-19 in patients with rheumatic diseases in the COVID-19 Global Rheumatology Alliance physician registry (March 12, 2020 -January 6, 2021). We investigated RA treated with b/ tsDMARD at the clinical onset of COVID-19 (baseline): abatacept (ABA), rituximab (RTX), Janus kinase inhibitors (JAK), interleukin-6 inhibitors (IL6i), or tumor necrosis factor inhibitors (TNFi). The outcome was an ordinal scale (1-4) for COVID-19 severity: 1) no hospitalization, 2) hospitalization without oxygen need, 3) hospitalization with any oxygen need or ventilation, or 4) death. Baseline covariates including age, sex, smoking, obesity, comorbidities (e.g., cardiovascular disease, cancer, interstitial lung disease [ILD]), concomitant non-biologic DMARD use, glucocorticoid use/ dose, RA disease activity, country, and calendar time were used to estimate propensity scores (PS) for b/tsDMARD. The primary analysis used PS matching to compare each drug class to TNFi. Ordinal logistic regression estimated ORs for the COVID-19 severity outcome. In a sensitivity analysis, we used traditional multivariable ordinal logistic regression adjusting for covariates without matching. Results: Of the 1,673 patients with RA on b/tsDMARDs at the onset of COVID-19, (mean age 56.7 years, 79.6% female) there were n=154 on ABA, n=224 on RTX, n=306 on JAK, n=180 on IL6i, and n=809 on TNFi. Overall, 498 (34.3%) were hospitalized and 112 (6.7%) died. Among all patients, 353 (25.3%) were ever smokers, 197 (11.8%) were obese, 462 (27.6%) were on glucocorticoids, 1,002 (59.8%) were on concomitant DMARDs, and 299 (21.7%) had moderate/ high RA disease activity. RTX users were more likely than TNFi users to have ILD (11.6% vs. 1.7%) and history of cancer (7.1% vs. 2.0%);JAK users were more likely than TNFi users to be obese (17.3% vs. 9.0%). After propensity score matching, RTX was strongly associated with greater odds of having a worse outcome compared to TNFi (OR 3.80, 95% CI 2.47, 5.85;Figure). Among RTX users, 42 (18.8%) died compared to 27 (3.3%) of TNFi users (Table). JAK use was also associated with greater odds of having a worse COVID-19 severity (OR 1.52, 95%CI 1.02, 2.28). ABA or IL6i use were not associated with COVID-19 severity compared to TNFi. Results were similar in the sensitivity analysis and after excluding cancer or ILD. Conclusion: In this large global registry of patients with RA and COVID-19, baseline use of RTX or JAK was associated with worse severity of COVID-19 compared to TNFi use. The very elevated odds for poor COVID-19 outcomes in RTX users highlights the urgent need for risk-mitigation strategies, such as the optimal timing of vaccination. The novel association of JAK with poor COVID-19 outcomes requires replication.
ABSTRACT
Background: Acute Respiratory Distress Syndrome (ARDS) is a life-threatening complication of COVID-19 and has been reported in approximately one-third of hospitalized patients with COVID-191. Risk factors associated with the development of ARDS include older age and diabetes2. However, little is known about factors associated with ARDS in the setting of COVID-19, in patients with rheumatic disease or those receiving immunosuppressive medications. Prediction algorithms using traditional regression methods perform poorly with rare outcomes, often yielding high specificity but very low sensitivity. Machine learning algorithms optimized for rare events are an alternative approach with potentially improved sensitivity for rare events, such as ARDS in COVID-19 among patients with rheumatic disease. Objectives: We aimed to develop a prediction model for ARDS in people with COVID-19 and pre-existing rheumatic disease using a series of machine learning algorithms and to identify risk factors associated with ARDS in this population. Methods: We used data from the COVID-19 Global Rheumatology Alliance (GRA) Registry from March 24 to Nov 1, 2020. ARDS diagnosis was indicated by the reporting clinician. Five machine learning algorithms optimized for rare events predicted ARDS using 42 variables covering patient demographics, rheumatic disease diagnoses, medications used at the time of COVID-19 diagnosis, and comorbidities. Model performance was assessed using accuracy, area under curve, sensitivity, specificity, positive predictive value, and negative predictive value. Adjusted odds ratios corresponding to the 10 most influential predictors from the best performing model were derived using hierarchical multivariate mixed-effects logistic regression that accounted for within-country correlations. Results: A total of 5,931 COVID-19 cases from 67 countries were included in the analysis. Mean (SD) age was 54.9 (16.0) years, 4,152 (70.0%) were female, and 2,399 (40.5%) were hospitalized. ARDS was reported in 388 (6.5% of total and 15.6% of hospitalized) cases. Statistically significant differences in the risk of ARDS were observed by demographics, diagnoses, medications, and comorbidities using unadjusted univariate comparisons (data not shown). Gradient boosting machine (GBM) had the highest sensitivity (0.81) and was considered the best performing model (Table 1). Hypertension, interstitial lung disease, kidney disease, diabetes, older age, glucocorticoids, and anti-CD20 monoclonal antibodies were associated with the development of ARDS while tumor necrosis factor inhibitors were associated with a protective effect (Figure 1). Conclusion: In this global cohort of patients with rheumatic disease, a machine learning model, GBM, predicted the onset of ARDS with 81% sensitivity using baseline information obtained at the time of COVID-19 diagnosis. These results identify patients who may be at higher risk of severe COVID-19 outcomes. Further studies are necessary to validate the proposed prediction model in external cohorts and to evaluate its clinical utility. Disclaimer: The views expressed here are those of the authors and participating members of the COVID-19 Global Rheumatology Alliance, and do not necessarily represent the views of the ACR, NIH, (UK) NHS, NIHR, or the department of Health.