Avaliação do uso de corticosteroides no tratamento de pacientes hospitalizados por Covid-19 e seus efeitos sobre a Covid longa

Abstract

The COVID-19 disease, caused by the SARS-COV-2 virus, was responsible for infecting a large part of the world's population, characterizing a pandemic and resulting in a high number of deaths. Currently, the characteristics of the disease, its risk factors and severity are well established and, with vaccination, cases and deaths are controlled. During the pandemic, corticosteroids were widely used, due to their connection with the control of inflammation, establishing a relationship with reduced mortality. In addition to the consequences experienced during the disease outbreak, other situations today put the physical and mental health of the population at risk, such as long COVID, characterized by delayed symptoms in patients post-acute infection. Its characteristics, control and treatment have been the subject of investigation. Therefore, given the relationship between corticosteroids and the disease, this work aims to evaluate their effects on long COVID in hospitalized patients. In this retrospective cohort, patients admitted to a university hospital with positive RT-PCR tests and aged ≥18 years were evaluated from April 2020 to April 2021. As an outcome, we considered hospital discharge and death. Data were collected regarding sociodemographic and clinical characteristics and treatment with corticosteroids. Long COVID was assessed through post-discharge hospital returns. Statistical analysis was performed using the SPSS program. 420 patients were evaluated, with a prevalence of men (55%), aged ≥60 years (57%) and only 89 (21%) had no previous illnesses. Of these, 96 (23%) died, of which 73 (76%) were serious. Among the survivors, 104 (32%) returned to the hospital after discharge and the sociodemographic characteristics were similar to the data for the acute illness. The use of corticosteroids was able to reduce the number of deaths, with statistical significance, in the analysis referring to the general population, but did not maintain statistical relevance when adjusted for the severity of the cases. Regarding returns, the use of corticosteroids had no statistical significance in the surviving population and the result remained adjusted for severity. However, when analyzed by return rates, the use of corticosteroids represented a significant reduction in this rate by 35.6% (95%CI 6.8 to 55.6) RR = 0.644 (95%CI 0.444 to 0.932; p = 0.020). However, when adjusted for the severity of the cases, this effect did not reach statistical significance (p=0.212). For the analysis of unfavorable outcomes (death/return), despite the 10.2% reduction, the use of corticosteroids did not reach statistical significance (RR = 0.898, 95%CI 0.704 to 1.146; p=0.386), however, a 30% reduction when compared to larger populations. We concluded that the use of corticosteroids was able to reduce mortality and post-discharge hospital return rates in relation to the general population. Regarding the severity of the disease, such findings cannot be considered, and this can be explained by the low sample power.