Abstract: TH-PO286
The Timing of Renal Replacement Therapy in Patients with Venoarterial Extracorporeal Membrane Oxygenation (VA-ECMO): A Nationwide Observational Cohort Study in Japan
Session Information
- Hemodialysis: Volume, Metabolic Complications, Clinical Outcomes
November 02, 2023 | Location: Exhibit Hall, Pennsylvania Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Dialysis
- 801 Dialysis: Hemodialysis and Frequent Dialysis
Authors
- Takeuchi, Tomonori, The University of Alabama at Birmingham, Birmingham, Alabama, United States
- Kubo, Toshihiro, Tokyo Ika Shika Daigaku Daigakin Ishigaku Sogo Kenkyuka, Bunkyo-ku, Tokyo, Japan
- Inoue, Norihiko, Tokyo Ika Shika Daigaku Daigakin Ishigaku Sogo Kenkyuka, Bunkyo-ku, Tokyo, Japan
- Tolwani, Ashita J., The University of Alabama at Birmingham, Birmingham, Alabama, United States
- Wille, Keith M., The University of Alabama at Birmingham, Birmingham, Alabama, United States
- Wakabayashi, Kenji, Tokyo Ika Shika Daigaku Daigakin Ishigaku Sogo Kenkyuka, Bunkyo-ku, Tokyo, Japan
- Neyra, Javier A., The University of Alabama at Birmingham, Birmingham, Alabama, United States
- Fushimi, Kiyohide, Tokyo Ika Shika Daigaku Daigakin Ishigaku Sogo Kenkyuka, Bunkyo-ku, Tokyo, Japan
Background
Acute kidney injury is a common complication in patients on VA ECMO, thus fluid management plays a crucial role during VA ECMO. Renal replacement therapy (RRT) enables precise fluid management, but early initiation of RRT has not shown benefits in general critically ill populations. We hypothesized that early initiation of tight fluid control with RRT would be beneficial for VA ECMO patients. In this study, we aimed to determine the association of early initiation of RRT with clinical outcomes in patients on VA ECMO, using the Diagnosis Procedure Combination database, a nationwide claims database in Japan.
Methods
This is a cohort study of adult patients who underwent VA ECMO during hospitalization between 4/2018 and 3/2022. We excluded patients who initiated RRT before starting VA ECMO and patients with end-stage kidney disease. We used propensity-score-based inverse probability weighting (IPW) to balance the baseline factors and to compare outcomes between two groups: patients who initiated RRT within 48h of VA ECMO initiation (Early RRT group) and those who did not (Late RRT group). The primary outcome was in-hospital mortality and the secondary outcome was RRT dependence at discharge in survivors.
Results
Of 1,181 VA ECMO patients, 336 were in the Early and 845 were in the Late RRT groups. After IPW, the clinical factors between the groups were well balanced, including the prevalence of cardiovascular disease (96.7 vs 97.8%), the total SOFA scores on the day of VA ECMO initiation (10.4 vs 10.4), and its renal component (0.9 vs 0.9). The median time to initiate RRT from the start of VA ECMO was 0 (IQR 0–1) days in the Early and 5 (IQR 2–11) days in the Late RRT groups. Early RRT initiation was associated with increased in-hospital mortality (66.5 vs 53.6%; OR 1.7, 95%CI 1.3 – 2.3) and increased RRT dependence at discharge in survivors (12.7 vs 2.0%; OR 5.6, 95%CI 2.2 – 14.0).
Conclusion
In adult patients with VA ECMO, early RRT initiation was associated with increased in-hospital mortality and RRT dependence at discharge in survivors, suggesting that early RRT initiation may not provide advantages for this population. Further investigation is needed to explore causal inference and identify specific subpopulations that may benefit from early initiation of RRT.