Biochemical Predictors of Acute Kidney Injury in Critically Ill COVID-19 Patients.

ABSTRACT

It is estimated that 50% of patients with coronavirus disease 2019 (COVID-19) have varying degrees of renal involvement. In this clinical biomarker development research, we examined in a retrospective study design the temporal changes in biochemical laboratory parameters in relation to the development of acute kidney injury (AKI). In a sample of 399 patients admitted from May 2020 to May 2021 to a tertiary health care intensive care unit (ICU), the incidence of AKI was 27.3%, and the median time to AKI was on 7th day of ICU admission. Most common etiology of AKI was kidney hypoperfusion. Within 72 h of developmental of low blood pressure, 63.76% developed AKI. The likelihood of AKI was higher in those with elevated serum ferritin, aspartate transaminase, and thrombocytopenia (low platelet count). A cutoff value of 750.3 ng/mL [area under the ROC curve (AUC) = 0.777] for serum ferritin, and 40.05 U/L for alanine aminotransferase (AUC = 0.677) 1 day before development of AKI displayed, respectively, a sensitivity of 76.2% and 64.3%, whereas the specificity was 69.5% and 64.1%, respectively, for these two biochemical predictors. A cutoff value of platelets (152.50 × 109/L [AUC = 0.75]) measured 4 days before development of AKI, displayed 83.3% sensitivity and 16.4% specificity. Taken together, our study thoroughly examined the temporal association of various clinical and laboratory parameters with AKI and prediction models were developed as per results of the time series data. These observations in a tertiary health care setting contribute to ongoing efforts for biomarker discovery and development using routine biochemical tests so as to forecast AKI in patients with COVID-19.