Kidney Week

Abstract: FR-PO0480

Large In Silico Studies Support Use of Automatic Ultrafiltration Control During Hemodialysis to Target Blood Volume Range Associated with Improved Survival

Session Information

• Dialysis: Hemodiafiltration, Ultrafiltration, Profiling, and Interdialytic Symptoms
  November 07, 2025 | Location: Exhibit Hall, Convention Center
  Abstract Time: 10:00 AM - 12:00 PM

Category: Dialysis

• 801 Dialysis: Hemodialysis and Frequent Dialysis

Authors

• Fuertinger, Doris H., Renal Research Institute, New York, New York, United States

Doris H. Fuertinger, MS, PhD

• Ho, Kevin, Fresenius Medical Care Holdings Inc, Waltham, Massachusetts, United States

Kevin Ho, MD, FASN

• Meigel, Felix J., Renal Research Institute, New York, New York, United States

Felix J. Meigel, PhD

• Wang, Aiyuan, Fresenius Medical Care Holdings Inc, Waltham, Massachusetts, United States

Aiyuan Wang, RD

• Law, Perry N, Fresenius Medical Care Holdings Inc, Waltham, Massachusetts, United States

Perry N Law

• Wang, Fei, Fresenius Medical Care Holdings Inc, Waltham, Massachusetts, United States

Fei Wang

• Yueh, Sheng-Han, Renal Research Institute, New York, New York, United States

Sheng-Han Yueh, MS, MSc

• Zhang, Hanjie, Renal Research Institute, New York, New York, United States

Hanjie Zhang, PhD

Background

Adaptive Ultrafiltration (aUF) is a physiologically closed-loop controller which automatically adjusts UF rate based on relative blood volume (RBV) measurements during hemodialysis (HD) to steer a patient’s RBV trajectory into a target RBV range associated with improved survival¹. In silico studies were performed to simulate use of aUF in database-derived cohorts of HD patient-avatars to evaluate RBV target attainment.

Methods

Cross-sectional study: An in silico population was generated (N = 50,000) from randomly selected treatment data in U.S. HD patients (data from ApolloDialDB). Plasma refill rates (PRR) were simulated using estimated initial blood volumes², and PRR profiles were created to be consistent with reported HD treatment RBV data^1,3,4.
Longitudinal study: A second in-silico population was generated based on data from 12,999 U.S. HD patients all using RBV monitoring with 24 treatments per patient within a 4-month period. The longitudinal in silico study consisted of 12 Non-Adaptive (Non-aUF) HD treatments followed by 12 aUF treatments.
The mean % of RBV measurements within the RBV target range limits (TRLs) in aUF and Non-aUF treatments were compared in each study.

Results

In the cross-sectional study, the % RBV measurements within the RBV TRLs was significantly greater in treatments employing aUF than in treatments without aUF, 47.9 % vs 32.9 %, P<0.001, respectively (Fig 1a). In the longitudinal study, the mean % ± SD of RBV measurements within the TRLs was significantly increased in aUF treatments, 55.49 ± 16.83 % (aUF) vs 35.32 ± 21.88 % (Non-aUF), P<0.001, respectively (Fig 1b).

Conclusion

Use of Adaptive UF in two large in silico population studies of HD patients significantly increased the % of RBV measurements within the RBV TRLs in treatments employing aUF compared to treatments without aUF.

References:
1. Preciado P et al. NDT. 2019; 2. Nadler SB et al. Surgery. 1962; 3. Wang H et al. JASN. 2020; 4. Koomans HA et al. KI. 1984

Funding

• Commercial Support – Fresenius Medical Care

Digital Object Identifier (DOI)

• doi: 10.1681/ASN.20252bskj175