Kidney Week

Abstract: SA-PO310

Establishing a Standardized Central Venous Catheter Rate

Session Information

• Hemodialysis and Frequent Dialysis: Potpourri
  November 05, 2022 | Location: Exhibit Hall, Orange County Convention Center‚ West Building
  Abstract Time: 10:00 AM - 12:00 PM

Category: Dialysis

• 701 Dialysis: Hemodialysis and Frequent Dialysis

Authors

• Deluca, Thomas Pasquale, Fresenius Medical Care, Global Medical Office, Waltham, Massachusetts, United States

Thomas Pasquale Deluca,

• Monaghan, Caitlin, Fresenius Medical Care, Global Medical Office, Waltham, Massachusetts, United States

Caitlin Monaghan,

• Belmonte, Kathleen, Fresenius Medical Care, Global Medical Office, Waltham, Massachusetts, United States

Kathleen Belmonte,

• Larkin, John W., Fresenius Medical Care, Global Medical Office, Waltham, Massachusetts, United States

John W. Larkin,

• Usvyat, Len A., Fresenius Medical Care, Global Medical Office, Waltham, Massachusetts, United States

Len A. Usvyat,

• Hymes, Jeffrey L., Fresenius Medical Care, Global Medical Office, Waltham, Massachusetts, United States

Jeffrey L. Hymes,

• Maddux, Franklin W., Fresenius Medical Care AG & Co KGaA, Bad Homburg, Hessen, Germany

Franklin W. Maddux,

Background

High central venous catheter (CVC) rates negatively affect dialysis patients around the world, with ailments such as CR-BSIs and thrombosis (Sohail, EKIR 2021). The process for getting a patient ready for AV access can be complex, thus catheter rates at clinics may be higher than expected. This model will be used as a tool to identify clinics that are performing worse than expected with CVC rates and aid to improve them.

Methods

The analysis was conducted on a sample of 143,411 active hemodialysis dialysis patients as of March 1^st, 2022, with end stage kidney disease. Patients must have had a CVC for greater than 90 days to be included in the study. A stepwise logistic regression model was trained on the data, with outcome as CVC or not. The model used BMI, age, location (Rural or Urban), sex, vintage, comorbidities, and nursing home residence.

Results

Using a nationwide catheter prevalence threshold of 19% for classifying predictions as having a CVC or not, the model performance on a test dataset showed balanced accuracy=61% and area under the curve=0.68. Predicted probabilities for clinics of size greater than 25 patients, had similar predictions to the actual rates (Figure 1). The top 3 most important predictors are vintage, age, and BMI.

Conclusion

The standardized CVC rate model gives clinics a tool to identify patients that look like they may be a good candidate for AV access but currently have a CVC. This tool will allow direct intervention into the practices of clinics to reduce their catheter rates. The model is in early development and a pilot will be initiated to continue validation of the model.

Figure 1: Prediction of clinic’s catheter rates on the x axis vs their true catheter rates on the y axis. The oval highlights the clinics that are targeted for intervention.

Funding

• Commercial Support –