Biomarkers of renal injury as early predictors of COVID-19 associated AKI: a prospective observational trial (BRICOAKI-trial)

ABSTRACT

Introduction: Early prediction by the use of serum and urinary biomarkers for the detection of acute kidney injury (AKI) may be very valuable to optimize the management and helps in improving the outcomes. This study aims to investigate whether daily measurement of urinary and plasma renal biomarkers have a role in earlier predicting COVID-19 associated AKI. Method(s) The study was conducted as a single-center, prospective, observational cohort study between August 2020 and December 2020 in hospitalized COVID-19 patients. A total of 65 moderate and severe COVID-19 positive adult (>= 18 years) patients were enrolled for this study. We measured serum creatinine, cystatin C, NGAL, KIM-1, Urine-Klotho, TIMP-2, IL-6 level, and urinary microalbumin/urinary creatinine on various days. The receiver operating characteristic curve (ROC) analysis was used to find the sensitivity and specificity of various markers to predict the incidence of AKI. Result(s) A total of 24 moderate and 41 severe COVID-19 patients were included. Out of which 47 patients developed (72.3%) acute kidney injury (AKI) over the course of COVID-19. Among these subjects, 18/47 (38.2%) developed severe AKI (KDIGO 2 + 3), and 5/47 (10.6%) required RRT. NGAL was found to be the best marker to predict the probability of AKI (Area under curve AUC of 0.713-0.786) with a sensitivity of 76-90% and specificity of 56-79% on different days of assessment from Day 1 to Day 7. IL-6 had moderate accuracy of prediction and cystatin C, KIM-1, Urine-Klotho, TIMP-2, IL-6 had poor accuracy for predicting the incidence of AKI. Conclusion(s) Urinary biomarkers like NGAL have good predictability for AKI.