Background: The role of chronic use of hydroxychloroquine (HCQ) in rheumatic disease (RD) patients during the SARS-CoV-2 pandemic is still subject of discussion.

Objectives: To compare the occurrence of COVID-19 and its outcomes between RD patients on HCQ use with individuals from the same household not taking the drug during community viral transmission in an observational prospective multicenter study in Brazil.

Methods: Participants were enrolled and monitored through 24-week (From March 29 ^th to Sep 30 ^th , 2020) regularly scheduled phone calls performed by trained medical professionals. Epidemiological and demographic data, as well as RD disease activity status and current treatment data, specific information about COVID-19, hospitalization, need for intensive care, and death was recorded in both groups and stored in the Research Electronic Data Capture (REDCap) database. COVID-19 was defined according to the Brazilian Ministry of Health (BMH) criteria. The statistical analysis was performed using IBM-SPSS v.20.0 software. Group comparisons were made using the Man-Whitney, Chi-Square and Fisher Exact Test, as well as multivariate regression models adjusted to confounders. Survival curves were performed using Kaplan-Meier analysis.

Results: A total of 10,427 participants mean age (SD) of 44.04 (14.98) years were enrolled, including 6004 (57.6%) rheumatic disease patients, of whom 70.8% had systemic lupus erythematosus (SLE), 6.7% rheumatoid arthritis (RA), 4% primary Sjögren’s syndrome (pSS), 1.8% mixed connective tissue disease (DMTC), 1% systemic sclerosis (SSc) and others (15.9), including overlap syndromes. In total, 1,132 (10.8%) participants fulfilled criteria for COVID-19, being 6.7% RD patients and 4.1% controls (p=0.002). A recent influenza vaccination had a protective role (p<0.001). Moderate and severe COVID-19 included the need for hospitalization, intensive care, mechanical ventilation or death. Infection severity was not different between groups (p=0.391) ( Table 1 ). After adjustments for multiple confounders, the main risk factors significantly associated with COVID-19 were higher education level (OR=1.29 95%CI 1.05-1.59), being healthcare professionals (OR=1.91; 95%CI 1.45-2.53), presence of two comorbidities (OR=1.31; 95%CI 1.01-1.66) and three or more comorbidities associated (OR=1.69; 95%CI 1.23-2.32). Interestingly, age >=65 years (OR=0.20; 95%CI 0.11-0.34) was negatively associated. Regarding RD, the risk factors associated with COVID-19 diagnosys were SLE (OR= 2.37; 95%CI 1.92-293), SSc (OR=2.25; 95%CI 1.05-4.83) and rituximab use (OR=1.92; 95%CI 1.13-3.26). In addition, age >=65 years (OR=5.47; 95%CI 1.7-19.4) and heart disease (OR=2.60; 95%CI 1.06-6.38) were associated with hospitalization. Seven female RD patients died, six with SLE and one with pSS, and the presence of two or more comorbidities were associated with higher mortality rate.

Conclusion: Chronic HCQ use did not prevent COVID-19 in RD compared to their household cohabitants. Health care profession, presence of comorbidities LES, SSc and rituximab were identified as main risk factors for COVID-19 and aging and heart disease as higher risk for hospitalization. Our data suggest these outcomes could be considered to manage them in clinical practice.

Acknowledgements: To the Brazilian Society of Rheumatology for technical support and rapid nationwide mobilization.

To all the 395 interviewers (medical students and physicians) who collaborated in the study and the participants

To CNPq (Number 403442/2020-6)

Disclosure of Interests: None declared