While renal replacement therapy (RRT) allows for precise fluid management as well as addressing electrolyte imbalances and the removal of other necessary compounds, its early initiation has not shown benefit in the general critically ill population. Moreover, the effects of early RRT initiation specifically in patients on venoarterial extracorporeal membrane oxygenation (VA-ECMO) also remain unclear. This retrospective study investigated adult patients who underwent VA-ECMO between April 2018 and March 2022 and used the clone-censor-weight method to emulate a hypothetical target trial and compare two groups: patients who initiated RRT within 2 days of VA-ECMO initiation (Early) and those who did not (Late). The primary outcomes were 28-day and 90-day hospital mortality analyzed by Cox proportional hazards models and the secondary outcome was 90-day RRT dependence by pooled logistic regression models. Inverse probability censoring weights were applied to adjust the models. A total of 2,513 VA-ECMO patients were cloned into both groups. The 28-day and 90-day mortalities were lower in the Early group (HR 0.59 [95% CI 0.53-0.68] and 0.67 [0.61-0.75]). However, the early group experienced greater RRT dependence at 90 days than the late group (OR 2.58 [1.94-3.46]). In conclusion, early initiation of RRT (within 2 days of VA-ECMO) was associated with lower hospital mortality but with a higher likelihood of 90-day RRT dependence in adult patients on VA-ECMO.
Keywords: Renal replacement therapy; Target trial emulation; Venoarterial extracorporeal membrane oxygenation.
© 2025. The Author(s).