Biomarcadores de injúria tubular aguda como preditores da mortalidade na COVID – 19

Abstract

COVID-19 is an infectious disease caused by a new type of coronavirus called SARS-CoV-2, which has a high capacity to infect humans and high transmissibility. Acute kidney injury (AKI) is a common complication in severe patients with COVID-19, affecting about 50% of patients in intensive care units (ICU), in addition, kidney injury is a risk factor for mortality. The use of more sensitive and specific urinary biomarkers can be useful in the early detection of kidney injury and can be used as predictors for different events. This work was carried out with the objective of evaluating the role of urinary biomarkers in predicting the mortality of critically ill patients with COVID-19 in the ICU. A prospective study was carried out with participants affected by COVID-19, admitted to the ICU of the Instituto Doutor José Frota (IJF) in Fortaleza, from June 2020 to April 2021 (1st and 2nd wave of COVID-19). Blood and urine samples were collected within 24 hours of admission to the ICU and the participants' medical records were monitored to verify changes in laboratory parameters. AKI was defined according to the KDIGO criteria. The urinary biomarkers MCP-1, Nephrin, NGAL and KIM-1 were quantified by ELISA and the predictive values for mortality of each biomarker were analyzed using ROC curves. Additionally, analyses were performed to estimate the chance of survival at 2 months using regression models. The average age of participants was 57 ± 16 years. Around 69% of participants developed AKI during their hospital stay and 38% died. Nephrin and MCP-1 biomarkers did not show significant differences between survivors and death groups. The biomarkers proteinuria (0.728; p = 0.004), NGAL (0.750; p = 0.002) and KIM-1 (0.749; p = 0.002) showed good values of ROC curves to predict death and in the combined analysis proteinuria*KIM- 1*NGAL was found to provide the best result (0.810; p < 0.001). In Cox regression models, proteinuria/creatinine ratio (>0.90), urinary KIM-1 (>1.8 ng/mg-Cr) and urinary NGAL (>118.8 ng/mg-Cr) were associated with a lower chance of survival at 2 months in the univariate analysis; however, in the multivariate model, after adjustments, only urinary NGAL remained associated, presenting a hazard ratio = 5.666 (95% CI: 1.761–18.227). Tubular renal damage was observed according to changes in the levels of biomarkers of acute tubular injury. KIM-1, NGAL and proteinuria correlated with death, and NGAL was an independent predictor for this outcome.