Skip to main content

Hydroxychloroquine and Azithromycin as a treatment of COVID-19: results of an open-label non-randomized clinical trial

Our take —

This study is the first reported prospective clinical study of HCQ among laboratory confirmed SARS-CoV-2 patients. The study was non-randomized and nearly a quarter of patients initially enrolled were not analyzed. The primary outcome, viral shedding, was inconsistently measured and defined, and levels of viral RNA differed between the groups at baseline. While viral shedding appeared lower in the intervention arm, confounding factors are present. The study did not formally assess clinical endpoints, and it is unclear whether there is any clinical benefit to the treatment.

Study design

Non-Randomized Trial

Study population and setting

This trial took place in March 2020. Inclusion criteria were : >12 years of age, PCR-confirmed SARS-CoV-2 infection; exclusion criteria were: known allergy to hydroxychloroquine, known contraindication to drug, pregnant or breast-feeding. Intervention patients were recruited from a single hospital in Marseille,France; control patients were recruited from same hospital and 3 other hospitals/facilities in Southern France. Participants were followed daily for 14 days. Intervention patients (n=26) received 600mg daily hydroxychloroquine (HCQ). Azithromycin was added to the treatment in 6 of the intervention arm cases. Control patients (n=16) consisted of those who refused treatment with HCQ or who did not meet inclusion criteria. Treatments, if any, were not disclosed for controls. . The primary outcome was viral RNA shedding defined as a PCR-positive nasopharyngeal swab at day 6 post inclusion. Daily PCRs were done.

Summary of Main Findings

6/26 patients in the intervention arm were lost to follow-up, including 3 individuals who were transferred to the ICU and 1 who died. The primary analysis was based on 20 individuals in the intervention arm and 16 in the control arm. 6 (17%) of the 36 participants included in the analysis were asymptomatic, 22 (61%) had upper respiratory tract symptoms, and 8 (22%) had lower respiratory tract symptoms. Mean age of intervention arm participants was 51 compared to 37 in the control arm. Intervention arm participants were more likely to have lower respiratory symptoms and be male compared to control arm participants. Participants who received HCQ + azithromycin had lower viral RNA loads at treatment initiation than the HCQ only and control groups. At day 6, there was no detectable viral shedding in individuals who received HCQ+azithromycin. Viral shedding was observed in 43% participants with HCQ only and 88% of control arm participants.

Study Strengths

First reported prospective clinical study on the effectiveness hydroxychloroquine in SARS-CoV-2 infected patients.


There were numerous limitations to this unblinded, non-randomized clinical trial. These included limited description of co-morbidities and demographics of the study population, substantial difference in age and severity of disease at study entry between arms, missing data on the primary study outcome, no statistical adjustment for differences between study arms, and no assessment of clinical endpoints. Additionally, the primary endpoint (viral shedding/clearance) was not consistently defined across patients and not assessed at the same frequency between arms. Levels of viral RNA at baseline were also lower in the one of the treatment groups, and sites other than the primary site did not perform daily PCR testing. The total length of follow-up was 14 days but only data through day 6 are shown. There were 6 individuals in the intervention arm excluded from primary analysis, 4 of whom had known relevant clinical outcomes (ICU transfer or death). Overall, there was limited data on clinical course or survival.

Value added

This study highlights that further research is needed to evaluate the clinical efficacy HCQ prior to general use.

This review was posted on: 4 May 2020