This is a result which should rekindle the debates on clinical tests carried out using chloroquine derivatives. According to a relatively large ex-post study on hydroxychloroquine, this worldwide drug did not seem to improve the lot of American patients with Covid-19. Its effectiveness remains to be rigorously confirmed or denied.
The preliminary study, made public this Tuesday by its authors before being evaluated by the reading committee of a medical journal, covers 368 patients in the network of public hospitals for American veterans, and who are either dead or discharged from hospital before April 11.
Three groups of patients
The authors analyzed a posteriori the medical files of these patients, which they grouped into three sets in order to compare them: those treated with hydroxychloroquine alone (HC); those who have had the hydroxychloroquine-azithromycin cocktail (an antibiotic) promoted in particular by the French doctor Didier Raoult; and those who have never received hydroxychloroquine.
The proportion of patients who died was highest in the hydroxychloroquine alone group (28%), compared to the cocktail group (22%) and the group without HC (11%). However, this conclusion can be misleading because the group of patients who received only hydroxychloroquine was initially sicker and more at risk than the other two groups: it contained more smokers and people with diabetes or cardiovascular and pulmonary history.
The authors statistically corrected this initial imbalance, and observed that "the increased risk of mortality in the hydroxychloroquine-only group persisted". The specificity of the patients treated must also lead to caution on any generalization to an entire population. The patients studied were all men, mostly black, a population hardest hit by the epidemic in the United States. The median age was advanced: over 65 years.
No harmonized protocols
Hydroxychloroquine is one of the treatments used in multiple emergency countries in severe cases of Covid-19, but controversy exists over its impact. The ideal, scientifically, is to conduct a randomized clinical trial, in which groups of comparable patients would follow different treatments, at random.
Such large-scale trials are under way, notably the European Discovery trial, but their results are not yet known. In the meantime, doctors are experimenting with molecules and treatments. Researchers can look at the results a posteriori, but in the absence of harmonized protocols on duration, doses, times of intervention, the degree of severity of cases at the start, it is difficult to draw reliable conclusions.
