[The praise of uncertainty: a systematic living review to evaluate the efficacy and safety of drug treatments for patients with covid-19.]
Cruciani F., Amato L., De Crescenzo F., Mitrova Z., Saulle R., Vecchi S., Davoli M.
BACKGROUND: SARS-CoV-2 is a coronavirus that causes a disease which can leads to a severe form of fatal pneumonia. At december 2020 in Italy, more than 2 million people have contracted the virus and 78,755 people have died. The scientific community is studying and testing numerous compounds that can be effective and safe for treating people with covid-19. AIM: To synthesize and evaluate the quality of evidence of efficacy and safety for the treatment. The available evidence is summarized in a living systematic review, a review that is constantly updated on the basis of the results of the new clinical studies. METHODS: A bibliographic search is launched weekly on the electronic databases and on the main clinical trial registers. Two researchers independently select the articles and assess the quality of the studies using the criteria developed by the Cochrane Collaboration, the certainty of the overall quality of the evidence is assessed using the GRADE criteria. RESULTS: At 31/12/2020, 101 randomized controlled studies were included that consider 72 different comparisons and include a total of 55,281 patients. 37 drugs are tested with respect to the standard treatment, 6 are evaluated against placebo and finally 29 compare different drugs with each other. By selecting studies that evaluate the efficacy and safety of a drug compared to standard treatment, which include at least 2 studies and which have low to high certainty of evidence, results show that corticosteroids, remdesivir, favipiravir, immunoglobulins, colchicine, and umbilical cord mesenchymal stem cell infusion could reduce overall mortality. No differences for the risk of any adverse events are observed between convalescent plasma and remdesivir compared to standard treatment. Remdesivir probably reduces the risk of serious adverse events; a similar effect, although less strong, is also noted for tocilizumab and the lopinavir-ritonavir combination. In contrast, hydroxychloroquine, corticosteroids and convalescent plasma transfusion are associated with safety concerns with respect to the risk of serious adverse events. CONCLUSIONS: The 101 studies included consider 72 comparisons and numerous outcomes, the results often coming from single studies and of small dimensions, and for 61% with a very low certainty of evidence, are difficult to summarize and the final result is to increase the uncertainty rather than providing useful information to the clinic and research. From all the work carried out it seems to us that the pandemic has highlighted the many shadows of scientific literature as tool to improve knowledge.