Practice Patterns and Outcomes of Potassium Repletion Thresholds during Critical Illness

Area of Science:

• Critical Care Medicine
• Nephrology
• Clinical Pharmacy

Background:

• Potassium repletion is a frequent intervention for critically ill patients.
• Current practices and outcomes associated with various intensive care unit (ICU) potassium repletion strategies remain unclear.

Purpose of the Study:

• To delineate potassium repletion practices in adult ICU patients.
• To compare the efficacy of different potassium repletion strategies.
• To evaluate the effectiveness and safety of specific potassium repletion thresholds on patient outcomes.

Main Methods:

• Retrospective analysis of the PINC AI Healthcare Database (2016-2022) involving critically ill adults admitted to an ICU.
• ICUs were categorized by potassium repletion patterns: threshold-based or probabilistic.
• Outcomes were compared between patients in different ICU types using hierarchical regression models and regression discontinuity analyses for specific thresholds (3.5 and 4.0 mEq/L).

Main Results:

• The study included 190,490 patients across 88 ICUs; 35.0% received potassium.
• Potassium repletion rates were comparable between threshold-based (33.5%) and probabilistic (36.4%) strategy ICUs, with no significant difference in adjusted risk (aOR, 1.09; 95% CI, 0.76-1.57).
• Regression discontinuity analyses showed that crossing potassium thresholds of 3.5 mEq/L or 4.0 mEq/L increased repletion rates by 36-39% but did not impact other patient outcomes.

Conclusions:

• Potassium repletion is prevalent in critically ill patients, frequently occurring within the 3.5-4.0 mEq/L range.
• The adoption of threshold-based repletion strategies in ICUs is not linked to clinically significant variations in patient outcomes.
• Current repletion practices may not require adjustment based on specific serum potassium thresholds.