Outcomes of continuous kidney replacement therapy in pediatric patients: the impact of timing and predictors of mortality.

Background: While critically ill children receiving continuous kidney replacement therapy (CKRT) are at increased risk of mortality, few studies have examined the predictors of mortality in this population. This study aimed to evaluate CKRT outcomes, focusing on predictors of mortality.

Methods: This cohort study included children aged 0-14 years who received CKRT. It collected baseline, clinical, and laboratory data. Descriptive analyses were performed. Least absolute shrinkage and selection operator (LASSO) regression was used to select the best predictors of mortality. A multivariable logistic regression model was constructed and validated with 1000 bootstraps.

Results: This study included 113 children who received CKRT, of whom 83 (73.5%) survived and 30 (26.6%) died. Children admitted to the intensive care unit with a higher Pediatric Risk of Mortality III score, sepsis, longer intubation, or hypoalbuminemia (< 30 g/dL) were more likely to die. Multifactorial acute kidney injury was more common in those who died than in those who survived (83.3% vs. 31.3%, p < 0.01). As the exclusive indication for CKRT, fluid overload was more common in those who died than in those who survived (26.7% vs. 6.0%, p = 0.01). LASSO and multivariable regression models identified hemodynamic instability, as evidenced by inotropic support use, and abnormal coagulation, as evidenced by not using heparin anticoagulation, as independent predictors of mortality. Initiating CKRT late (> 48 h) was associated with mortality in the univariate but not the multivariate analysis.

Conclusions: Hemodynamic instability was the best predictor of mortality in critically ill children receiving CKRT.