A case of severe hyperkalemia: fast, safe and effective treatment is required
Article Outline
- 1. Cardiac toxicity of hyperkalemia
- 2. Evidence for therapeutic intervention
- 3. Returning to the case
- 4. Conclusion
- 5. Acknowledgment
- References
- Copyright
We read with great interest the article by Carvalhana et al titled “Management of Severe Hyperkalemia Without Hemodialysis: Case Report and Literature Review” [1]. Carvalhana et al describe a case of a 59-year-old diabetic man presenting with cardiac arrhythmia secondary to serum potassium of 10.4 mEq/L. The patient was on an angiotensin-converting enzyme inhibitor, had taken a single dose of a nonsteroidal anti-inflammatory drug, and had evidence of volume depletion. The creatinine was 108 μmol/L and urine output was present. Management consisted of 1 dose of calcium chloride, 10 U of insulin, intravenous sodium bicarbonate, calcium resonium, and volume expansion with saline. Although the initial management strategy had been to initiate hemodialysis, the patient was ultimately managed conservatively on account of intact urine output and hemodynamic stability. This case illustrates many of the myths and misconceptions regarding the cardiac toxicities and management of hyperkalemia. The purpose of this commentary was to outline the cardiac effects of hyperkalemia, discuss the evidence behind therapeutic interventions for its treatment, and, lastly, highlight the potential dangers of conservative management posed to the patient described in the case by Carvalhana et al.
1. Cardiac toxicity of hyperkalemia
Cardiac arrhythmias induced by hyperkalemia are unpredictable [2]. Classic teaching suggests the electrocardiographic changes seen in hyperkalemia follow a step-wise, dose-dependant manner. In reality, many patients show rapid changes in their electrocardiograms when exposed to hyperkalemia [3]. The risk of arrhythmia increases with potassium concentrations above 6.5 mEq/L [4], and even small elevations in serum potassium above this level may lead to rapid progression from peaked T waves to ventricular fibrillation. This implies that the longer the patient is exposed to high levels of potassium, the greater the risk of sudden deterioration. There are multiple case reports of severe hyperkalemia presenting with minimal or no electrocardiogram (ECG) changes, thereby highlighting the insensitivity of ECG for assessing the severity of hyperkalemia [5].
2. Evidence for therapeutic intervention
Intravenous calcium chloride stabilizes the myocardium and increases the arrhythmogenic threshold. It has a rapid onset of action (3-5 minutes) and lasts up to 1 hour [6]. Insulin decreases serum potassium by inducing intracellular shift. The magnitude of the fall ranges from 0.5 to 1.0 mEq/L [7] and is dose dependent with a greater shift occurring using 20 U of regular insulin vs 10 U [8]. β Agonists also increase intracellular shift and act synergistically with insulin. Up to 30% of patients are nonresponders to β-agonist therapy and caution should be exercised when used in patients with underlying coronary artery disease [9]. Bicarbonate therapy traditionally has been administered as either a continuous infusion or a bolus dose of 1 ampule (50 mEq) of sodium bicarbonate. Unfortunately, bicarbonate therapy has failed to show significant, predictable reductions in serum potassium under experimental conditions [10], [11]. Small studies in hemodialysis patients evaluating bicarbonate show variable benefit in the reduction in serum potassium. This benefit was only seen in the presence of significant metabolic acidosis (bicarbonate <22 mEq/L) and it is unclear whether this can be extrapolated to the general population [11]. Cation exchange resins (such as calcium resonium and sodium polystyrene sulfonate) have variable effectiveness with a slow onset of action (peak effects 4-6 hours) which limits their use for acutely lowering serum potassium [12]. It is unclear whether they are effective when administered without concurrent laxative therapy [13].
These are temporizing measures and if the serum potassium fails to show immediate reductions (as in the case where the serum potassium actually rose), emergent hemodialysis should be administered. Currently, there are no evidence-based guidelines for when to initiate dialysis in patients with hyperkalemia. However, delays in nephrology consultation regarding the initiation of dialysis have been shown to increase mortality and length of intensive care unit stay [14]. Acute hemodialysis has a very low complication rate and is a safe, effective procedure for the treatment of hyperkalemia. Under ideal conditions, hemodialysis can reduce serum potassium by 1 to 1.5 mEq/L per hour [15]. The major complication is related to central venous catheterization, namely, hemorrhage and/or pneumothorax. Ultimately, the decision on initiation of hemodialysis should be made on an individualized basis. Careful consideration of the physiology and clinical effects of hyperkalemia should guide the decision.
3. Returning to the case
In the case presented by Carvalhana et al, the patient may have been placed at undue risk and a strong argument can be made that hemodialysis therapy would have been a safer management option despite the good outcome the patient ultimately had. Firstly, the patient initially presented with a sign wave pattern on the ECG which can rapidly progress to ventricular fibrillation [3]. Secondly, the serum potassium was greater than 6 mEq/L for 7.5 hours (and >10 mEq/L for 2 hours) thus placing the patient at high risk for arrhythmia. Thirdly, intravenous calcium chloride is normally given with the end point being normalization of the ECG changes. However, in this case the patient was bradycardic, a situation which can worsen with calcium administration and ought to limit its use [16]. If the patient has a life-threatening arrhythmia, defibrillation is often unsuccessful until the serum potassium is reduced [6]. Fourthly, the serum potassium failed to reduce after insulin and bicarbonate were administered.
Cortical collecting duct secretion of potassium is largely dependent on distal sodium delivery and a functional tubule to secrete potassium. This was likely impaired in the described scenario despite the apparently normal appearing urine output. In the absence of urinary potassium measurement one cannot be certain of adequate potassium secretion. Also, this patient likely had some degree of underlying renal impairment and although prerenal azotemia was most likely, an element of acute tubular necrosis could not be ruled out. The patient's creatinine posthydration was 84 mol/L and assuming he weighed 60 kg this corresponds to a creatinine clearance of 67 mL/min. This is a 40% reduction in glomerular filtration rate. Furthermore, potassium secretion is also dependent on the action of aldosterone, and this patient has a variety of reasons for being hypoaldosteronic: he was a chronic user of an angiotensin-converting enzyme inhibitor and had used a nonsteroidal anti-inflammatory drug before hospitalization, both of which suppress the action of aldosterone and have long half lives of activity. Based on a reduced glomerular filtration rate and the hypoaldosteronism, the urinary potassium secretion would be expected to be low. For example, a urine K concentration of 16 mEq/L with a urine flow rate of 500 mL/h would result in a K excretion of only 8 mEq/h. This rate of K excretion would not safely lower serum potassium rapidly enough.
Although this patient had a good outcome without dialysis, appreciating the limitations of conservative management options and the physiology of hyperkalemia in an individual patient may have pointed toward more aggressive reduction of the potassium concentration using hemodialysis. An additional point deserves consideration when contemplating conservative management of hyperkalemia. In the absence of clinical trials evaluating hemodialysis vs potassium shifting and diuresis for correcting severe hyperkalemia, isolated case reports of successful conservative management must be viewed with caution. Unsuccessful attempts to treat hyperkalemia without dialysis are unlikely to be published (representing publication bias [17]). This is based on the assumption that an adverse outcome would have suggested that dialysis had been indicated and standard of care neglected. Given the likely bias of under- or even nonreporting of failed conservative management, it is unknown how many adverse outcomes exist for every successful case of corrected severe hyperkalemia without dialysis. Unfortunately, publication bias can make medical management of severe hyperkalemia appear safer than it is.
4. Conclusion
Hyperkalemia is a very treatable life-threatening disorder. The cardiac arrhythmias are often unpredictable and prolonging exposure poses needless risk to patients. Overall, timely hemodialysis represents a safer and effective management strategy for severe hyperkalemia, especially in cases with suspected renal impairment and reduced aldosterone activity.
5. Acknowledgment
We thank Dr Kamel S. Kamel for his critical review and thoughtful discussion of the manuscript.
References
- . Management of severe hyperkalemia without hemodialysis: case report and literature review. J Crit Care. 2006;21:316–321
- . Hyperkalemia revisited. Tex Heart Inst J. 2006;33:40–47
- . Electrocardiographic manifestations of hyperkalemia. Am J Emerg Med. 2000;18:721–729
- . Hyperkalemia in hospitalized patients. Arch Intern Med. 1998;158:917–924
- . Severe hyperkalemia with normal electrocardiogram. Int J Clin Pract. 2001;55:486–487
- . Potassium disorders—clinical spectrum and emergency management. J Resuscitation. 2006;70:10–25
- . Emergency interventions for hyperkalemia. The Cochrane Library. 2006;(2):1–61
- . Effect of various therapeutic approaches on plasma potassium and major regulating factors in terminal renal failure. Am J Med. 1988;85:507–512
- . Controversial issues in the treatment of hyperkalemia. Nephrol Dial Trans. 2003;18:2215–2218
- . Combined effect of bicarbonate and insulin with glucose in acute therapy of hyperkalemia in end-stage renal disease. Nephron. 1996;72:476–482
- . Effect of bicarbonate administration on plasma potassium in dialysis patients: interactions with insulin and albuterol. Am J Kidney Dis. 1996;28:508–514
- . Hyperkalemia: a potential silent killer. J Am Soc Nephrol. 1998;9:1535–1543
- . Laxative-induced diarrhea: a continuing medical problem. BMJ. 1974;1:537–541
- . Nephrology consultation in acute renal failure: does timing matter?. Am J Med. 2002;113:456–461
- Effect of dialysate sodium concentration on interdialytic increase in potassium. J Am Soc Nephrol. 2000;11:2337–2343
- . Severe bradyarrhythmias following calcium pre-treatment. Am Heart J. 1991;121(6 Pt 1):1812–1813
- Cochrane Collaboration open learning material for reviewers, version 1.1. November 2002. In: Anderson P, Green S, editors. http://www.cochrane-net.org/openlearning/.
PII: S0883-9441(07)00088-3
doi:10.1016/j.jcrc.2007.06.004
© 2008 Elsevier Inc. All rights reserved.
