Abstract: TH-PO040
Association of Serum Potassium Level at the Initiation of Renal Replacement Therapy with All-Cause Mortality Among Burn Intensive Care Unit Patients with AKI
Session Information
- AKI: Epidemiology, Risk Factors, Prevention
October 25, 2018 | Location: Exhibit Hall, San Diego Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Acute Kidney Injury
- 101 AKI: Epidemiology, Risk Factors, and Prevention
Authors
- Hsiung, Jui-Ting, University of California, Irvine, Orange, California, United States
- Kalantar-Zadeh, Kamyar, University of California, Irvine, Orange, California, United States
- Streja, Elani, University of California, Irvine, Orange, California, United States
- Moradi, Hamid, University of California, Irvine, Orange, California, United States
- Joe, Victor C., University of California, Irvine , Orange, California, United States
- Rizzo, Julie A., US Army Institute of Surgical Research, Fort Sam Houston, Texas, United States
- Stewart, Ian J., David Grant Medical Center, Travis Air Force Base, California, United States
- Chung, Kevin, Brooke Army Medical Center, Fort Sam Houston, Texas, United States
Background
Burn injury is associated with serious complications including acute kidney injury (AKI) requiring renal replacement therapy (RRT). There is little data on the association between baseline serum potassium (K) and outcomes in burn patients initiated on RRT. Therefore, we evaluated the association of K at the RRT initiation with survival in a cohort of burn patients.
Methods
We examined 170 burn patients enrolled in a multicenter observational study in eight burn centers across the United States. We used Cox regression models with adjustments for covariates including severity of illness, laboratory values, urine output, total burn surface area, and vasopressor requirement to evaluate the association of K level at RRT initiation with in-hospital all-cause mortality. We further explored the association between K slope, from the initiation to 48 hours after beginning of RRT, and mortality, using the same multilevel Cox models. Patients without either baseline or 48 hours K measurements were excluded from the slope analysis.
Results
The cohort was comprised of 169 AKI patients who required RRT and had a baseline K. 20% of patients were female with a mean±SD age of 51±17 years. Hypokalemia was associated with a higher risk of subsequent mortality. 150 patients were included in the slope analysis. There was a linear association between increasing K slope and higher mortality based on the result of restricted cubic splines.
Conclusion
While baseline hypokalemia at the time of RRT initiation was associated with higher risk of mortality, increasing K at 48 hours was also associated with worse outcomes in patients with burn-associated AKI. This finding may be due to inadequate dialysis, ongoing tissue destruction, or both. Future investigations are needed to examine the mechanisms which can explain these observations.