Background
The reported incidence of hyperkale­mia in hospitalized patients is between 1% and 10%.[1-3] The majority of cases are due to medications, including angiotensin-converting enzyme inhi­bi­tors (ACEIs), angiotensin II receptor blockers (ARBs), and diuretics such as spironolactone, which reduce the loss of potassium.[4,5] 

Severe hyperkalemia is a life-threatening condition that can result in muscle paralysis or fatal arrhythmias, and requires prompt reduction of serum potassium levels.[5] However, there is little consistency in the management of hyperkalemia among clinicians at various levels of training.[6] 

A review by the Cochrane Col­laboration concludes that the current body of evidence for the emergency management of hyperkalemia lacks a sufficient number of large randomized controlled trials.[6] Based on a small number of studies, the use of a beta-agonist as well as intravenous insulin and glucose appears to be effective in rapid reduction of hyperkalemia. Furthermore, the use of nebulized beta-agonists combined with IV insulin and glucose appears to be more effective than either therapy used alone. 

Results for bicarbonate therapy were found to be equivocal, and monotherapy with bicarbonate is not recommended. In the absence of gastrointestinal pathology, potassium-exchange resins were not found to be effective within 4 hours and should not be relied upon for rapid effects. If medical therapy fails or is only temporarily successful, as in the case of renal failure, dialysis should be considered. Anecdotal and animal data suggest that IV calcium is effective in treating arrhythmias.[6]

In 1998, Acker and colleagues designed a quality improvement (QI) study on the treatment of hyperkale­mia in hospitalized patients.[7] They found that during the observational phase of the study only 39% of cases were managed according to the hos­pital standard for treatment.

Even after the intervention phase, when laboratory personnel faxed a copy of the hospital management guideline to the ward for each case of hyperka­lemia, there was little improvement. Postintervention data showed that only 42% of cases were managed ac­cording to the standard, up slightly from 39% during the observational phase. The ef­fect of intervention may have been limited by the small number of faxed guidelines (38%) actually transmitted.

We designed a QI project targeting the management of acute hyperkale­mia on the general medicine wards at St. Paul’s Hospital, a tertiary care and teaching centre in Vancouver, British Columbia, Canada. Our goal was to in­crease the proportion of hyperka­lemia cases managed according to the best available evidence by 10%, and to reduce the cost incurred per case by 10%. 

During the observational phase, we collected data on how hyperkale­mia was being managed on all hospital wards. In the intervention phase, we distributed a guideline based on an evidence-based treatment algorithm to house staff (Figure [14]). We then de­termined whether the intervention resulted in an improved adherence to evidence-based protocols and a re­duction in cost incurred per case of hyperkalemia.

Methods
Study population
During the observational phase of the study, 23 March to 6 April 2010, 54% of hyperkalemia cases occurred on the general medicine wards. The intervention was thus focused on cases of acute hyperkalemia occurring in pa­tients admitted to general medicine wards. Patients in the intensive care unit and those undergoing renal re­placement therapy were excluded, as these patients are managed according to their own hyperkalemia protocols. For study purposes, a case of hyperkalemia was defined as serum potassium greater than 4.7 mmol/L.

Data collection
All cases of hyperkalemia that occur­red between 23 March and 17 October 2010 were identified by the Department of Pathology and Laboratory Medicine at St. Paul’s Hospital. Data pertaining to the management of hyperkalemia were obtained retrospectively through paper charts and electronic health records. 

Two independent members of the QI team determined whether each case of hyperkalemia was managed according to evidence-based protocols. Cost of care was calculated for each case of hyperkalemia, and included costs of medications, laboratory tests, and electrocardiograms.

Plan-do-study-act cycles
Two plan-do-study-act (PDSA) cycles were implemented during the data-collection period. The first PDSA cy­cle ran from 22 July to 8 August 2010. Prior to this, a pocket-sized guideline for the management of acute hyperkalemia incorporating best available evidence was created and distributed to the house staff on the general medicine wards, in­cluding the members of each clinical teaching unit—an attending physician, a senior resident in the second or third year of the internal medicine program, two junior residents in the first year of residency, and two third-year medical students. 

Prior to the second PDSA cycle, which ran from 22 September to 17 October 2010, the guideline was distributed to all internal medicine residents in order to reach house staff providing over­night coverage. In addition, posters describing the protocols were placed in the emergency department and general medicine wards. During the en­tire study period, house staff were unaware of the data collection and analysis.

Statistical analysis
We used run charts, a common method of plotting data over time to evaluate the success of improvement efforts in an objective way. Run charts are appropriate for quality improvement projects and have been shown to be effective in detecting signals in a wide range of health care applications.[8]

Results
During the observational phase (23 March to 6 April 2010), 114 cases of hyperkalemia occurred. Only 63% of cases were managed according to the best available evidence. The overall cost incurred per hyperkalemia case was $16.74. 

During the first PDSA cycle (22 July to 8 August 2010), 76 cases of hyperkalemia occurred. The average percentage of cases managed according to the best available evidence increased from 63% in the observational phase to 79%, achieving the absolute target increase of 10%. In addition, the overall cost incurred per hyperkalemia case decreased to $7.80, also achieving the absolute target reduction of 10%. However, there was still significant inconsistency in the management of hyperkalemia.

During the second PDSA cycle (22 September to 17 October 2010), 98 cases of hy­perkalemia occurred. The average percentage of cases managed according to the best available evidence increased substantially to 94%. In addition, the overall cost incurred per hyperkalemia case decreased to $7.51.

Conclusions
Providing internal medicine residents with a user-friendly guideline for the management of acute hyperkalemia resulted in a significant increase in the proportion of cases that were managed according to best available evidence. The use of evidence-based management also resulted in a reduction of treatment cost.

These results came at a minimal cost to our program. The only financial costs incurred involved printing the protocols and posters, further demonstrating that successful quality improvement projects can be implemented effectively with little to no cost.

The study had important limitations. First, the incidence of severe hyperkalemia (defined as a serum po­tassium level greater than 6.0 mmol/L) was low throughout the study, limiting our ability to assess the effect of the intervention for patients with hyperkalemia in this range. Second, since the house staff were unaware of our data collection and analysis, the recognition of hyperkalemia and the clinical reasoning process guiding intervention or nonintervention were poorly documented in patient charts.

Further quality improvement studies are needed to assess safety and sustainability outcomes of our intervention.

Competing interests
None declared.

This article has been peer reviewed.

Drs Rajan, Widmer, Kim, Dehghan, and Alsahafi are in their third year of residency training in the internal medicine program at the University of British Columbia. Dr Levin is head of the Division of Nephrology at UBC and a practising nephrologist at St. Paul’s Hospital. She is also a professor in the Department of Medicine at UBC, where she actively participates in resident mentoring, research, and education.