Abstract: TH-PO364
Measured Dialysate Sodium Deviates Frequently and Sometimes Substantially from Ordered Dialysate Sodium in Acute Care Hemodialysis
Session Information
- Dialysis: Dialysate and Clearance
October 25, 2018 | Location: Exhibit Hall, San Diego Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Dialysis
- 701 Dialysis: Hemodialysis and Frequent Dialysis
Authors
- Sheth, Vishwa C., University of California Davis, Sacramento, California, United States
- Bang, Heejung, University of California Davis, Sacramento, California, United States
- Don, Burl R., University of California Davis, Sacramento, California, United States
- Paracuelles, Vincent, University of California Davis, Sacramento, California, United States
- Craig, Maureen, University of California Davis, Sacramento, California, United States
- Chin, Andrew I., University of California Davis, Sacramento, California, United States
Group or Team Name
- University of California, Davis
Background
It is unknown if measured dialysate sodium (meas DNa) differs from the ordered DNa (ord DNa) in the acute care setting where hemodialysis (HD) dialysate is created at the point of care via individual acid and bicarbonate concentrates
Methods
We examined HD treatments in an acute care setting where meas DNa was obtained at HD start. An absolute difference between meas DNa and ord DNa >2 mEq/L was considered out of range. Factors in a simple/univariate logistic regression model with random intercept for each dialysis machine included: machine factors (ID#, model, hours of use) and dialysate factors (ord DNa, additives in acid or bicarb jug, dialysate flow rate)
Results
There were 1230 HD treatments on 21 unique Fresenius machines in a 3-mo period. After exclusions (machine ID# unknown, machine <20 HD's), 1196 treatments were analyzed. The difference between meas DNa and ord DNa was close to normally distributed but skewed towards meas DNa <ord DNa; meas DNa was out of range in 46.8% (n=560) of the cases (Fig 1). In the logistic regression analysis, there was a significant effect of machine ID# clustering, with an ICC of 0.20-0.25, on out of range DNa. Other factors (machine model, hours of use, dialysate flow, ord DNa, additives in acid or bicarb jug) were not associated with DNa out of range, when machine clustering was taken into account. In 227 treatments where the meas DNa was out of range (91.5% were <ord DNa), the nurse made a manual DNa adjustment based on the difference between meas DNa and ord DNa. A subsequent meas DNa level was checked during HD. After this adjustment, the meas DNa was within range of the originally ord DNa in 176 (77.5%) of the treatments.
Conclusion
Measured DNa was >2 mEq/L off from ordered DNa in about half of the treatments, with a bias towards a lower level. There appears to be a difference between dialysis machines, with some having a higher % of treatments with out of range DNa, despite consistent and rigorous machine maintenance and testing
Funding
- Clinical Revenue Support