Background: Acute kidney injury (AKI) is a common and serious complication in critically ill patients with non-kidney organ dysfunction. Early prediction of AKI is crucial for timely intervention and improved outcomes. This study aimed to identify readily available non-renal predictors of AKI and to develop an exploratory prediction model in a specific cohort of critically ill patients with COVID-19-related septic shock requiring mechanical ventilation.
Materials and methods: This was a single-center, observational, retrospective cohort study conducted in the respiratory ICU of Hospital H+ Querétaro between April and December 2020. The study included 42 mechanically ventilated patients with septic shock secondary to SARS-CoV-2 infection and non-kidney organ dysfunction. AKI was defined using the KDIGO criteria. Trend analysis, bivariate and multivariate linear regression, were used to identify predictors of AKI and severe AKI.
Results: AKI occurred in 23 (54.8%) patients, with 6 (14.3%) developing severe AKI. Trend analysis revealed differences in norepinephrine dose, hemoglobin, and lactate trends between groups. A simplified logistic regression model, validated internally with bootstrapping to prevent overfitting, identified a protective trend associated with higher hemoglobin levels on admission. Quantitative analysis of a forecasting model for daily renal function showed moderate predictive accuracy.
Conclusions: This study identified several readily available non-kidney organ dysfunction variables that can predict AKI and its severity in critically ill patients with COVID-19-related septic shock. These findings may help in the early identification of at-risk patients and facilitate timely interventions to potentially improve outcomes. Further validation in larger and more diverse populations is warranted.