ABSTRACT
PURPOSE: Acute kidney injury (AKI) is a frequent complication among COVID-19 patients in the intensive care unit, but it is less frequently investigated in general internal medicine wards. We aimed to examine the incidence, the predictors of AKI, and AKI-associated mortality in a prospective cohort of non-ventilated COVID-19 patients. We aimed to describe the natural history of AKI by describing trajectories of urinary markers of hemodynamic, glomerular, and tubular injury. METHODS: 141 COVID-19 patients were enrolled to the study. AKI was defined according to KDIGO guidelines. Urine and renal function parameters were followed twice a week. Multivariate logistic regression was used to determine the predictors of AKI and mortality. Trajectories of urinary markers were described by unadjusted linear mixed models. RESULTS: 19.7% patients developed AKI. According to multiple logistic regression, higher urinary albumin-to-creatinine ratio (OR 1.48, 95% CI 1.04-2.12/1 mg/mmol) and lower serum albumin (OR 0.86, 95% CI 0.77-0.94/1 g/L) were independent predictors of AKI. Mortality was 42.8% in the AKI and 8.8% in the group free from AKI (p < 0.0001). According to multiple logistic regression, older age, lower albumin, and AKI (OR 3.9, 95% CI 1.24-12.21) remained independent predictors of mortality. Urinary protein-to-creatinine trajectories were diverging with decreasing values in those without incident AKI. CONCLUSION: We found high incidence of AKI and mortality among moderately severe, non-ventilated COVID-19 patients. Its development is predicted by higher albuminuria suggesting that the originally damaged renal structure may be more susceptible for virus-associated effects. No clear relationship was found with a prerenal mechanism, and the higher proteinuria during follow-up may point toward tubular damage.