Abstract: FR-PO654
Hyperkalemia and Progression of CKD
Session Information
- Fluid and Electrolytes: Clinical - Potassium, Sodium, Water
November 08, 2019 | Location: Exhibit Hall, Walter E. Washington Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Fluid and Electrolytes
- 902 Fluid and Electrolytes: Clinical
Authors
- Davis, Jill, AstraZeneca, Newark, Delaware, United States
- Done, Nicolae, Analysis Group, Inc., Boston, Massachusetts, United States
- Chamberlain, Christina X., Analysis Group, Inc., Boston, Massachusetts, United States
- Israni, Rubeen K., AstraZeneca, Newark, Delaware, United States
- Anzalone, Deborah A., AstraZeneca, Newark, Delaware, United States
- Betts, Keith, Analysis Group Inc., Los Angeles, California, United States
- Mu, Fan, Analysis Group, Inc., Boston, Massachusetts, United States
- Curhan, Gary C., Channing Division of Network Medicine, Brigham and Women's Hospital, Boston, Massachusetts, United States
- Billmyer, Emma, Analysis Group, Inc., Boston, Massachusetts, United States
- Szerlip, Harold M., Baylor University Medical Center, Dallas, Texas, United States
- Uwaifo, Gabriel I., Ochsner Medical Foundation, Slidell, Louisiana, United States
- Fonseca, Vivian A., Tulane University Medical Center, New Orleans, Louisiana, United States
- Wu, Eric, Analysis Group, Inc., Boston, Massachusetts, United States
Background
To compare rates of chronic kidney disease (CKD) progression between CKD patients with and without hyperkalemia (HK).
Methods
Adult patients with CKD stage 3-4, with or without HK, were selected from electronic medical records from the US Research Action for Health Network (2012-2018). CKD stage was identified by diagnosis codes or estimated glomerular filtration rate (eGFR). HK was defined as having ≥1 serum K+ lab value >5.0 mEq/L and confirmed by another HK event. Index dates were defined as 30 days after the first K+ lab >5.0 mEq/L for HK patients and after a randomly selected K+ lab ≤5.0 mEq/L for patients without HK. Patients were required to have ≥1 eGFR value in both the baseline (6 months pre-index date) and study period (up to 5 yrs post-index date). Two outcomes were evaluated separately: 1) progression to CKD stage 5; 2) ≥10 unit decline in eGFR. Kaplan-Meier analyses and multivariable Cox models adjusted for baseline eGFR, demographic characteristics, relevant comorbidities, and treatment use were performed. Sensitivity analyses were performed adjusting for albumin-creatinine ratio (ACR), stratifying by baseline CKD stage, and excluding patients with baseline acute kidney injury (AKI).
Results
Patients with HK (n=9,220) vs. those without HK (n=36,764) (mean age 72.3 vs. 72.7 yrs) had higher rates of CKD stage 4 (33.9% vs. 9.7%), heart failure (31.7% vs. 15.0%), and AKI (33.8% vs. 11.5%), and a higher ACR value (393.9 vs. 154.4 mg/g). In the 5-yr study period, 16.7% of HK patients and 3.5% of those without HK progressed to CKD stage 5, and 31.7% and 17.0%, respectively, had an eGFR decline ≥10 units (both log-rank p<0.001). In Cox models, patients with vs. without HK had a statistically significant higher risk of CKD progression to stage 5 (adjusted hazard ratio [aHR] 2.20; 95% confidence interval [CI], 2.02-2.38) and eGFR decline (2.40; 2.28-2.52) (both p<0.001). Cox model results were consistent when adjusting for ACR, stratifying by baseline CKD stage, or excluding AKI patients.
Conclusion
Even after adjusting for relevant comorbidities and treatments, HK was significantly associated with higher risk of progression to CKD stage 5 and eGFR decline among patients with CKD stage 3-4. Associations were robust in all sensitivity analyses.
Funding
- Commercial Support –