Factors Associated with the Initiation of Renal Replacement Therapy in Patients on VV-ECMO: A Case-Control Study.

Abstract

Acute Kidney Injury (AKI) is a common complication in patients with Acute Respiratory Distress Syndrome (ARDS) receiving VV-ECMO support, carrying a high risk of progression to Renal Replacement Therapy (RRT). Both AKI and RRT are linked to an increased risk of mortality. This study aims to evaluate the risk factors associated with the need for RRT in patients undergoing VV-ECMO. Methods: This is a retrospective case-control study involving patients on VV-ECMO therapy admitted to the intensive care unit (ICU) between 2019 and 2023. Patients on VV ECMO support, with or without RRT, were included and their severity scores and associated mortality were calculated. A multivariate logistic regression analysis was performed to assess the variable RRT using odds ratios (OR) with their corresponding confidence intervals (CI) for the outcome variables. Results: A total of 192 subjects were included, with a mortality rate of 39.6%. Of these, 68.7% were male, with an average ICU stay of 25.1 days and a need for RRT in 19.7% of cases. The multivariate analysis independently associated the use of vasopressors with RRT norepinephrine OR 5.61 (95% CI, 1.64-19.1) and vasopressin OR 4.64 (95% CI, 2.15-10.0)). An increase in creatinine levels before ECMO support is associated with an increased risk OR 2.21 (95% CI 1.54-3.18), and 24 h after ECMO support, the risk rises further adjusted odds ratio (AOR) 3.32 (95% IC 1.55-7.09). The accuracy of severity scores presented weak discrimination and similar behavior, except for DEOx for the primary outcome, with an AUC of 0.79 (95% CI, 0.72-0.87), and APACHE II with an AUC of 0.68 (95% CI, 0.59-0.78). Conclusions: The prediction of RRT in patients on VV-ECMO support was superior for DEOx, which is influenced by the use of vasopressors, creatinine levels, and platelet transfusion prior to cannulation. This could be useful for predicting early interventions in this patient population.