ASN's Mission

ASN leads the fight to prevent, treat, and cure kidney diseases throughout the world by educating health professionals and scientists, advancing research and innovation, communicating new knowledge, and advocating for the highest quality care for patients.

learn more

Contact ASN

1401 H St, NW, Ste 900, Washington, DC 20005

email@asn-online.org

202-640-4660

The Latest on Twitter

Kidney Week

Abstract: FR-PO450

Development and Validation of a Predictive Model for Delivered Dose in Continuous Renal Replacement Therapy

Session Information

Category: Dialysis

  • 701 Dialysis: Hemodialysis and Frequent Dialysis

Authors

  • Galindo, Pablo E., Instituto Nacional de Ciencias Medicas y Nutricion Salvador Zubirán, Mexico City, CIUDAD DE MEXICO, Mexico
  • Martinez-Rueda, Armando Jezael, Instituto Nacional de Ciencias Medicas y Nutricion Salvador Zubirán, Mexico City, CIUDAD DE MEXICO, Mexico
  • Correa-Rotter, Ricardo, Instituto Nacional de Ciencias Medicas y Nutricion Salvador Zubirán, Mexico City, CIUDAD DE MEXICO, Mexico
  • Vega, Olynka, Instituto Nacional de Ciencias Medicas y Nutricion Salvador Zubirán, Mexico City, CIUDAD DE MEXICO, Mexico
Background

Continuous renal replacement therapy (CRRT) has become one of the most relevant therapies for acute kidney injury and critical patients. Since the beginning there has been controversy about delivering, measuring, and defining ideal dose. Measuring effluent volume related to body weight has been the method of choice for the last years. It’s well known that prescribed dose and delivered dose differ significantly, mainly because of down time, the effect of pre-dilution replacement, and the dialyzer capacity to saturate the effluent. Current practice recommends prescribing 25% more dose to achieve the desired goal. A tool that could account for the three main factors that lower the dose and could predict de delivered dose, could be of great help for the daily prescription of CRRT. The objective of this study is to validate the results of the proposed model in our Institution.

Methods

We developed an app-based model programed in Xcode® for IOS, with a prescription step method, and patient based format, that considers: down time, pre-dilution replacement, and effluent saturation. The model is given a desired therapy, and then calculates the simulated delivered dose. For validation we evaluated since March 2019, 5 treatments of CRRT in CVVHDF mode, and run 15 dose evaluations by measuring simultaneously: BUN pre-filter pre-dilution, BUN pre-filter post-dilution, and UN of the effluent solution. We then compared the delivered dose with the results of the predictive model.

Results

We found good correlation compared to the delivered dose. We conducted a Pearson correlation that showed: r =0.91, 95% CI= (0.74 - 0.97), R2=0.82, and a P value of <0.0001. To evaluate agreement we conducted a Bland Altman plot we demonstrated that 100% of results where between the 11% error: (Bias= -2, SD of Bias 4.2,95% Limits of agreement (-11– 5.5).

Conclusion

Even though the sample was small, the delivered dose predictor computes accurate results as compared to the delivered dose. The App based tool can help predict de delivered dose given to a patient, it can be of great utility to simulate different therapies with different prescriptions, and compare different results. At present more patients and treatments are being continuously evaluated to strengthen even further the results.