2026 Critical Care Congress Skip to main content
2026 Critical Care Congress Main banner
Final Program


Advance Program
Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.
Session Icons
Thought Leader Session
Thought Leader Session
Guidelines Session
Guidelines Session
Aging Adult Track
Aging Adult Track
Additional Registration Fee
Additional Registration Fee

Research Snapshot Theater: Renal, Adult I

Multicenter Study on Continuous Renal Replacement Therapy Timing and Intensity in ICU Patients

Tuesday, March 24, 2026
10:00 AM - 11:00 AM Central Time
Location: Connections Central - RST 15

  • Shan Guleria, MD (he/him/his)

    Disclosure(s): No relevant financial relationship(s) to disclose.

  • KC

    Kaveri Chhikara, MS

    Data Scientist
    University of Chicago

    Disclosure information not submitted.

  • ZL

    Zewei Liao, MS

    Data Scientist
    UChicago Medicine Center for Care and Discovery (CCD) - Hyde Park

    Disclosure information not submitted.

  • JK

    Jay Koyner, MD

    Dr.
    University of Chicago Medical Center

    Disclosure information not submitted.

  • AB

    Anna Barker, MD

    Clinical Instructor
    University of Michigan Hospital and Health Systems, Michigan

    Disclosure information not submitted.

  • WP

    William Parker, MD

    Associate Professor
    University of Chicago

    Disclosure information not submitted.

Introduction: Epidemiologic data on continuous renal replacement therapy (CRRT) treatment practices for severe acute kidney injury (AKI) are limited. In particular, thresholds for CRRT initiation and intensity of dosing are poorly described.


Methods: We conducted a retrospective cohort study of adult patients admitted to two health systems between 2018-2024 and the MIMIC-IV database. Using data standardized into the Common Longitudinal ICU data Format (CLIF), we identified all adult patients requiring CRRT, excluding patients with preexisting end-stage renal disease and those who died within six hours of starting CRRT. We extracted standard demographic and granular ICU clinical data in CLIF format, including CRRT mode and dosing, vasopressor dosing, standard ICU labs, and respiratory support data. Large language models were used to help with coding; all authors reviewed all output.


Results: The study cohort consisted of 7490 admissions (median age 60-64 years, 38.7% Female, 25.7% Black, 4.8% Hispanic, median SOFA score of 8-10). The severity of metabolic derangements at initiation ranged widely between sites. The site with the lowest median bicarbonate at initiation in mmol/L (16.8 vs 20.0-20.5) also had the highest median creatinine in mg/dL (3.9 vs. 2.7-2.8) and median potassium in mmol/L (4.7 vs 4.4-4.5). This same site had the highest rate of invasive mechanical ventilation (73% vs 62%-66%) at the time of CRRT initiation. Potentially as a result of this, arterial pH did not vary significantly between sites. Institutional modality preferences differed markedly, with each site implementing predominantly either CVVHD or CVVHDF. CRRT clearance intensity was highest for the site with the latest initiation (median dialysate flow rate of 3.1 L/hr vs 1.9-2.1 L/hr). The site with the latest dialysis initiation had the highest mortality at 24 hours (13.5% vs 7.8%) and 72 hours (31% vs 18%).


Conclusions: The data suggest heterogeneous CRRT practices regarding thresholds for initiation, modality, and dosing. Future directions include applying machine learning approaches for estimating individualised treatment effects and target trial emulation methods for optimal CRRT initiation and dosing strategies.