OBJECTIVES : A number of publications have disputed the benefits and risks of hydroxychloroquine (HCQ) treatment in COVID-19 patients. Recently, a now-retracted, high-impact factor publication based on data that has been questioned in the media, claimed that HCQ treatment increased frequency of ventricular arrhythmia and mortality risk in hospitalized COVID-19 patients. Our study aims to investigate the risks of HCQ treatment using real-world patient data.

METHODS : Real-world data composed of electronic medical records from approximately 54 million patients in Americas and APAC region (TriNetX Dataworks Network) were analyzed using TriNetX Analytics. COVID-19 patients were defined with the ICD-10 code U07.1 or a record of a positive SARS-CoV-2 RNA test result, recorded between December 20, 2019 and May 1, 2020. Cohorts were further defined by a hospitalization encounter within 1 week from the first diagnosis/positive test (index event), a record of HCQ treatment with or without macrolide (treatment groups), or no record of antimalarial treatment (control group). Patients who started HCQ therapy while on mechanical ventilation were excluded. Cohorts were balanced using propensity score matching and adjusted for 34 confounders. Recorded death was used to define mortality outcome, and ICD-10 code I49 was used as an outcome for ventricular arrhythmia. The observation period was set between 1 and 60-days post index event.

RESULTS : In the balanced cohorts of 485 patients each, use of HCQ with a macrolide, compared to the untreated group was neither associated with higher risk of mortality: risk ratio, RR (95% Cl) 1.061 (0.776,1.449), nor with cardiac arrhythmia RR 1 (0.42,2.381),. Similar results were observed in the group treated with HCQ alone compared to control: mortality RR 1.015 (0.74,1.392) and cardiac arrhythmia RR 1 (0.454,2.204)

CONCLUSIONS : This real-world analysis does not support claims that use of hydroxychloroquine in hospitalized COVID-19 patients is associated with a higher risk of mortality or cardiac arrhythmia.