Immune-mediated diseases (IMID) and coronavirus disease 2019 (COVID-19): Results from an observational retrospective multicenter study

ABSTRACT

Background: The novel coronavirus emerged in 2019 in Wuhan has caused a global pandemic of coronavirus disease (COVID-19). Immune-mediated diseases (IMID), as inflammatory arthritis or inflammatory bowel disease (IBD), have some special implications due to their pathogenesis and treatments. Some treatments employed in IMID are now being used in the treatment of severe COVID-19. There still exists controversy about IMID behavior and its complications. 1, 2 Objectives: To assess COVID-19 severity in IMID patients and its prognosis predictors. Methods: An observational retrospective multicenter study was performed in two Spanish Hospitals (Clinical University Hospital in Santiago De Compostela and Gregorio Marañón Hospital). Patients were selected if they were diagnosed of an IMID (rheumatoid arthritis, psoriatic arthritis, espondyloarthritis, ulcerative colitis and Crohn's disease) and had COVID-19 infection between February and April 2020. Demographic, clinical, analytical and treatment data were collected. Logistic regression was used to evaluate potential predictors. Stata 15.1 was used to perform statistical analysis. Results: 91 patients were recruited. 55 suffered from a rheumatic disease and 36 suffer IBD. Baseline characteristics are shown in Table 1. Univariable analysis reached age, comorbidity, female gender, flu vaccine, arthropathy, basal classical synthetic anti-rheumatic drugs (csDMARD), pneumonia and basal C-reactive protein (CRP) as potential predictors of non-severe (absence of death, respiratory insufficiency, intensive care unit admission or sepsis) COVID-19 disease (p < 0.2). After multivariable analysis, only female gender (OR 4.60 [CI95% 1.00, 21.2] p=0.050), lower age (OR 0.94 [CI95% 0.88, 1.00] p=0.042) and lower basal levels of CRP (OR 0.87 [CI95% 0.77, 0.97] p=0.010) were predictors for non-severe disease (p < 0.005). Mean time of healing (symptoms solved in outpatient and hospital discharge in admitted) from COVID-19 was 13.8 days (SD 16.3). Univariable analysis showed arthropathy, COVID-19 symptomatic and basal glucocorticoids (GC) dose as potential predictors of higher time-to-healing from COVID-19 disease (p < 0.2). After multivariable analysis, only lower GC basal dose predicts higher time-tohealing (OR -1.83 [CI95% -2.81, -0.84] p=0.001). 11 patients deceased because of COVID-19. Univariable analysis reached age, basal csDMARD, pneumonia and basal CRP as potential predictors of COVID-19 mortality (p < 0.2). After multivariable analysis, only higher age was a predictor for mortality (OR 1.14 [CI95% 1.04,1.25] p=0.006). Conclusion: IMID patients showed similar predictors of mortality than general population involving COVID-19. Immune-modulating agents did not seem to overshadow the prognosis of COVID-19 infection. Female gender, lower age and lower basal CRP could select a cohort of “good” prognosis patients with mild COVID-19 disease as well higher age points out the worst prognosis. Even that, each case should be individiualized.