Management of Hypokalemia in Intensive Care Patients

A standardized hypokalemia treatment algorithm for critically ill patients was developed through a two-round expert panel process. Recommendations regarding the diagnosis, treatment, and monitoring of hypokalemia were evaluated using structured electronic questionnaires. For each recommendation, the Content Validity Ratio (CVR) was calculated as (Ne - N/2)/(N/2), where Ne was the number of panelists rating the recommendation as necessary and N was the total number of participating panelists. The minimum acceptable CVR was determined according to the total number of panelists participating in the corresponding survey round. Recommendations with a calculated CVR exceeding the applicable minimum CVR value were considered necessary and were included in the hypokalemia management algorithm.

The percentage of patients with documented hypokalemia, defined as a serum potassium concentration <3.5 mEq/L, who received potassium replacement therapy during the ICU stay. Receipt of potassium replacement therapy was assessed using electronic health records and medication administration records. The numerator was the number of patients with documented hypokalemia who received potassium replacement therapy, and the denominator was the total number of patients with documented hypokalemia. The outcome was expressed as a percentage (%).

The time elapsed between identification of hypokalemia, defined as a serum potassium concentration <3.5 mEq/L, and administration of the first dose of potassium replacement therapy. The time of serum potassium measurement and the time of potassium administration were obtained from laboratory records and medication administration records. The outcome was expressed in hours.

The percentage of patients who achieved normokalemia after initiation of potassium replacement therapy. Normokalemia was defined as a serum potassium concentration of 3.5-5.0 mEq/L and was assessed using laboratory serum potassium measurements. The numerator was the number of treated patients who achieved normokalemia, and the denominator was the total number of patients who received potassium replacement therapy. The outcome was expressed as a percentage (%).

The time elapsed between initiation of potassium replacement therapy and the first subsequent serum potassium measurement within the normokalemic range of 3.5-5.0 mEq/L. Potassium administration times and serum potassium concentrations were obtained from medication administration records and laboratory records. The outcome was expressed in hours.

The duration for which serum potassium remained within the normokalemic range of 3.5-5.0 mEq/L after achievement of normokalemia following potassium replacement therapy. Serum potassium concentrations were obtained from laboratory records. Duration was calculated from the first documented normokalemic serum potassium measurement to the first subsequent measurement outside the normokalemic range, ICU discharge, or death, whichever occurred first. The outcome was expressed in hours.

The percentage of patients who developed a new hypokalemia episode after previously achieving normokalemia. Normokalemia was defined as a serum potassium concentration of 3.5-5.0 mEq/L, and recurrent hypokalemia was defined as a subsequent serum potassium concentration <3.5 mEq/L after documented achievement of normokalemia. Consecutive serum potassium measurements <3.5 mEq/L without an intervening normokalemic measurement were considered part of the same hypokalemia episode and were not classified as recurrence. The numerator was the number of patients who developed at least one recurrent hypokalemia episode, and the denominator was the number of patients who achieved normokalemia. The outcome was expressed as a percentage (%).

The percentage of patients who developed hyperkalemia during or after potassium replacement therapy. Hyperkalemia was defined as a serum potassium concentration >5.0 mEq/L and was assessed using laboratory serum potassium measurements. The numerator was the number of treated patients who developed hyperkalemia, and the denominator was the total number of patients who received potassium replacement therapy. The outcome was expressed as a percentage (%).

The percentage of patients whose last available serum potassium concentration before ICU discharge was within the normokalemic range of 3.5-5.0 mEq/L. Serum potassium concentrations were obtained from laboratory records. The numerator was the number of discharged patients whose last available serum potassium concentration was within the normokalemic range, and the denominator was the total number of patients discharged from the ICU with an available serum potassium measurement. The outcome was expressed as a percentage (%).

The duration of ICU hospitalization was calculated from the date and time of ICU admission to the date and time of ICU discharge or death using electronic health records. The outcome was expressed in days.

The number of hypokalemia episodes per patient during the ICU stay was determined using laboratory serum potassium measurements. Hypokalemia was defined as a serum potassium concentration <3.5 mEq/L. A hypokalemia episode began with the first serum potassium measurement <3.5 mEq/L and continued until normokalemia, defined as a serum potassium concentration of 3.5-5.0 mEq/L, was achieved. Consecutive serum potassium measurements <3.5 mEq/L before achievement of normokalemia were considered part of the same episode and were not counted as separate episodes. A new hypokalemia episode was counted when serum potassium subsequently decreased to <3.5 mEq/L after documented achievement of normokalemia. The outcome was expressed as the number of hypokalemia episodes per patient.

The number of hyperkalemia episodes per patient during the ICU stay was determined using laboratory serum potassium measurements. Hyperkalemia was defined as a serum potassium concentration >5.0 mEq/L. A hyperkalemia episode began with the first serum potassium measurement >5.0 mEq/L and continued until serum potassium returned to the normokalemic range of 3.5-5.0 mEq/L. Consecutive serum potassium measurements >5.0 mEq/L before achievement of normokalemia were considered part of the same episode and were not counted as separate episodes. A new hyperkalemia episode was counted when serum potassium subsequently increased to >5.0 mEq/L after documented achievement of normokalemia. The outcome was expressed as the number of hyperkalemia episodes per patient.

The change in serum potassium concentration was calculated as the first serum potassium concentration measured after potassium replacement therapy minus the serum potassium concentration measured immediately before initiation of potassium replacement therapy. Serum potassium concentrations were obtained from laboratory records. The outcome was expressed in mEq/L.

The percentage of patients with mild hypokalemia, defined as a serum potassium concentration of 3.0-3.4 mEq/L, or moderate hypokalemia, defined as 2.5-2.9 mEq/L, who underwent serum potassium measurement at least once daily while hypokalemia persisted. Monitoring was assessed using laboratory records. The numerator was the number of eligible patients who received serum potassium monitoring at least once daily, and the denominator was the total number of patients with mild-to-moderate hypokalemia. The outcome was expressed as a percentage (%).

The percentage of patients with severe hypokalemia, defined as a serum potassium concentration <2.5 mEq/L, who underwent repeat serum potassium measurement 6 hours after intravenous potassium administration. Serum potassium measurements and intravenous potassium administration times were assessed using laboratory records and medication administration records. The numerator was the number of eligible patients who underwent the specified 6-hour serum potassium reassessment, and the denominator was the total number of patients with severe hypokalemia who received intravenous potassium. The outcome was expressed as a percentage (%).

The percentage of patients with hypokalemia who underwent measurement of the prespecified laboratory parameters during the diagnosis and management of hypokalemia. The parameters included urea, serum sodium, serum magnesium, and serum glucose. Monitoring was assessed using laboratory records. The numerator was the number of patients with documented measurement of all prespecified laboratory parameters, and the denominator was the total number of patients with documented hypokalemia. The outcome was expressed as a percentage (%).

The percentage of patients with severe hypokalemia, defined as a serum potassium concentration <2.5 mEq/L, or hypokalemia persisting despite potassium replacement therapy who underwent arterial blood gas analysis. Arterial blood gas testing was assessed using laboratory records. The numerator was the number of eligible patients who underwent arterial blood gas analysis, and the denominator was the total number of patients meeting either indication for arterial blood gas assessment. The outcome was expressed as a percentage (%).

The percentage of patients with severe hypokalemia, defined as a serum potassium concentration <2.5 mEq/L, or symptomatic hypokalemia who underwent at least one electrocardiogram (ECG). ECG assessment was determined using electronic health records and ECG records. The numerator was the number of eligible patients who underwent at least one ECG, and the denominator was the total number of patients with severe or symptomatic hypokalemia. The outcome was expressed as a percentage (%).

The percentage of patients receiving intravenous potassium chloride (KCl) at an infusion rate >20 mmol/hour who underwent continuous ECG monitoring during the infusion. KCl infusion rates and continuous ECG monitoring were assessed using medication administration records and electronic health records. The numerator was the number of eligible patients who received continuous ECG monitoring, and the denominator was the total number of patients receiving intravenous KCl at an infusion rate >20 mmol/hour. The outcome was expressed as a percentage (%).