Gautret · International journal of antimicrobial agents 2020 · non-randomized open-label controlled trial · n=36

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

This paper has been RETRACTED. It must not be used as evidence.

Cited 4968 times in the scientific literature.

Level 3 - non-randomized controlled study

Non-randomized, open-label controlled clinical study (subsequently retracted)

PubMed 32205204 · doi:10.1016/j.ijantimicag.2020.105949 · record verified 2026-08-29

What was done

French patients with confirmed COVID-19 were enrolled in an open-label, non-randomized trial to receive 600 mg of hydroxychloroquine daily, with azithromycin added based on clinical presentation. Viral load in nasopharyngeal swabs was tested daily. Untreated patients from another center and patients who refused the protocol were used as negative controls. The primary endpoint was the presence or absence of viral carriage at Day 6 post-inclusion.

What was found

The study evaluated 36 patients: 6 asymptomatic, 22 with upper respiratory tract symptoms, and 8 with lower respiratory tract symptoms. Twenty patients received hydroxychloroquine. The authors reported a statistically significant reduction in viral carriage at Day 6 compared to controls, with azithromycin combination reported as significantly more efficient for viral elimination; specific numerical percentages, relative risks, or confidence intervals were not reported in the abstract.

Why it matters

This study heavily influenced early global COVID-19 treatment strategies, but its clinical conclusions are completely undermined by subsequent formal retraction.

Limits

The publication was retracted due to severe ethics and methodology violations, including unverified ethical approval timing, lack of informed consent for azithromycin, non-standardized PCR testing between centers, and lack of equipoise. Additional design weaknesses include open-label design, non-randomized allocation, confounding control group selection (protocol refusers and another hospital), dissent from multiple co-authors, and a very small sample size (20 treated cases).

Cited by