Journal Logo

Toxicology Rounds

Ingestion of Potassium Chloride Crystals Causes Hyperkalemia and Hemorrhagic Gastritis

Mosely, Dan S. MD; Osborne, Benjamin MD

Author Information

    Acute hyperkalemia resulting from toxic ingestions1,2,3 is only reported infrequently in the literature, and cases resulting from ingestion of potassium salts in water softeners in children have not been reported. Acute gastritis as a result of caustic ingestion is more common in the pediatric population, most commonly resulting from ingestion of strong acids or bases. Few cases of caustic ingestions from potassium chloride have been reported, however.4–6 As water softeners are increasingly used in the United States, the incidence for this potentially life-threatening toxic ingestion may increase.

    Case Report

    The patient, an 8-year-old African American girl, presented to the emergency department with abdominal pain. According to the child's mother, she was found in the water shed near their house approximately three hours before with a handful of salt crystals from their water softener in her hands. The child initially denied ingestion, and seemed active and playful so the mother did not feel there was any danger. Over the next few hours, however, the patient began to complain of increasing abdominal discomfort and vomited several times. The mother then called the hospital advice nurse, and following that discussion, brought her to the ED.

    Acute potassium chloride ingestion can result in gastrointestinal hemorrhage with scarring, gastric outlet obstruction, and metabolic acidosis

    On presentation, the child was continuing to complain only of abdominal discomfort. The mother told the triage nurse that the patient had possibly ingested a toxic substance, but the amount was unknown. She was immediately placed on a monitor. The initial rhythm was noted as seen in Figure 1. A quick history from the mother revealed that the child had been playing with the salts from the water softener, and a presumptive diagnosis of hyperkalemia was made.

    Figure 1
    Figure 1:
    Initial rhythm of 8-year-old who ingested potassium chloride crystals.

    Vital signs were blood pressure 116/36 mm Hg, pulse 94 bpm, respirations 24 bpm, temperature (oral) 97.5°F, and oxygen saturation 97% on room air. On initial exam, the child was alert, quiet, and intermittently complaining of abdominal discomfort. The remainder of the physical exam was unremarkable with the exception of mild epigastric tenderness to palpation. A rectal exam revealed guaiac-negative stool.

    An intravenous line was placed, and blood was sent for venous blood gas, chemistries, complete blood count, and type and screen. Calcium chloride 0.2 cc/kg (3 cc) was given via slow push. Dextrose (D50) 0.5 amp (12.5 g) and insulin 0.1 U/Kg (1.5 units) were given. She was concomitantly placed on continuous albuterol by nebulizer. The QRS on the EKG initially narrowed but then widened again. A second dose of calcium chloride (3 cc) was given, which resulted in immediate narrowing of the QRS complex (see Figure 2). The patient was then given sodium bicarbonate 1 mEq/Kg (15 mEq) as well as Lasix 0.5 mg/Kg (8 mg) intravenously. Finally, she was given kayexalate (15 g per rectum). The patient was transferred to the pediatric intensive care unit for further management including emergent dialysis.

    Figure 2
    Figure 2:
    After a second dose of calcium chloride (3 cc), the QRS complex immediately narrowed.

    Lab results included a venous blood gas pH of 7.18, PCO2 41 mmHg, and PO2 49 mmHg. Chemistries were sodium 143 mEq/L, potassium 10.3 mEq/L, chloride 112 mEq/L, bicarbonate 19 mEq/L, blood urea nitrogen 17 mg/dL, creatinine 0.8, glucose 249 mg/dL calcium 13.1 mEq/L, magnesium 2.5 mEq/L, and phosphorus 2.1 mEq/L. Complete blood count was notable for a white blood cell count of 18.9/mm3, hematocrit 42.7%, and platelets 300,000.

    Following transfer to the PICU, the patient was started on dialysis, which normalized the blood potassium level by the next morning. After arrival, however, the child began to vomit bright red blood. An emergent upper gastrointestinal endoscopy was performed which showed extensive mucosal scarring of the antrum and pylorus of the stomach (see Figure 3). The hematemasis was not hemodynamically significant, however, and the patient did not require transfusion. She was observed in the PICU throughout the remainder of the hospital course without complications, and was discharged on hospital day 12. Over the next several weeks, she developed increasing difficulty eating secondary to scarring of the stomach, which led to the development of a gastric outlet obstruction. This was verified by an upper gastrointestinal series with small bowel follow through. The patient underwent a pylorotomy four weeks after initial presentation.

    Figure 3
    Figure 3:
    An emergent upper gastrointestinal endoscopy revealed extensive mucosal scarring of the antrum and pylorus of the stomach.

    Clinical Consequences

    For various reasons, water softeners are being used with increasing frequency throughout the United States. For many, it is the desire to improve the taste of the water they drink; others seek the benefit of calcium or magnesium. It is the concentration of these minerals that defines the “hardness” of the water in which they are contained.7,8

    Many companies will offer potassium chloride as a “healthy alternative” to sodium chloride to reduce the amount of sodium left in the softened water.7,9 In at least one published case, the use of this type of exchange resin has possibly led to hyperkalemia in an elderly patient with renal failure and diabetes who was concomitantly using calcium channel blockers.10,11 Whether the potassium salt in the water softener was the direct cause of this patient's presentation is unclear, but given the fact that any exogenous potassium in patients with these comorbid conditions can lead to hyperkalemia, it is probable that this consumption exacerbated the problem. With the exception of patients who must closely monitor their sodium intake,9 there does not appear to be any true advantage to using potassium-based water softener.

    Acute potassium chloride ingestion has the obvious concern of causing hyperkalemia. However, there are other clinical consequences of an acute potassium ingestion. This case demonstrates several of these, including gastrointestinal hemorrhage with long-term scarring and gastric outlet obstruction, metabolic acidosis, and other laboratory abnormalities.

    Potassium is the major intracellular cation; only a very small fraction of total body potassium is in the intravascular space. Although hyperkalemia, defined as a serum potassium level greater than 5.5 mEq/L, is relatively common in the adult population, it is seen infrequently in pediatrics. Moderate or severe hyperkalemia (6.1 to 6.9 mEq/L and > =7.0 mEq/L) may lead to grave consequences because of its effect on membrane potentials, particularly of cardiac cells. Generally, the more acute the onset, the more deleterious the effect hyperkalemia has on the cell membranes. This makes hyperkalemia due to toxic ingestion of particular concern.12

    The primary metabolism and excretion of potassium takes place in the kidneys, and hyperkalemia is most often seen in acute and chronic renal failure where the ability of the kidneys to secrete or excrete potassium is impaired. Of note, this generally will not occur until moderate to severe renal failure is present because the kidneys have the ability to handle normal potassium loads. This balance is easily tipped in these patients with seemingly small increases in dietary potassium or with the use of drugs that impair potassium metabolism.

    As water softeners are increasingly used, the incidence for acute hyperkalemia may increase

    The goals in the treatment of hyperkalemia include protecting the heart from the toxic effects of potassium, acutely shifting potassium from the intravascular to the intracellular space, and eliminating potassium from the body. Parenteral infusion of calcium chloride is the primary means of myocardial stabilization and its effect on the myocardium is rapid and impressive, as seen in this case. There are numerous methods of encouraging a transcellular shift of potassium, including infusion of insulin/glucose and sodium bicarbonate and use of beta agonists (such as the nebulized albuterol used in this case).

    These effects are relatively short-lived, and are a temporizing measure pending the final phase of treatment. The definitive treatment for severe hyperkalemia is elimination of the potassium ion from the body. Ion exchange resins (such as sodium polysterene sulfonate) act via direct cation exchange in the gut, but work relatively slowly. It can be given orally and rectally and should be instituted concomitantly with the rapid acting treatments noted above. In cases of severe hyperkalemia, dialysis is the definitive method for removing potassium ions from the blood.

    Gastrointestinal bleeding is an alarming problem in children. The total blood volume of a child is relatively small and can deplete rapidly, while resuscitation efforts are hindered by difficult venous access. The mortality in this population is low, however, because they are capable of tolerating relatively high levels of total body blood loss compared with the adult population. This experience contrasts sharply with significant mortality in elderly patients despite aggressive management.13 Gastric ulceration and erosions may be seen in children receiving NSAIDs, such as aspirin, ibuprofen, naproxen, and ketorolac.14,15,16 Additionally, gastric erosions and bleeding as a result of acute potassium overdose has been well described in the literature.4,5,6 Not previously described, however, is development of gastric outlet obstruction after acute potassium ingestion. This complication has been described in cases of other cation ingestions, such as iron.

    Several metabolic and laboratory abnormalities were noted in this case. The most serious metabolic abnormality was an anion gap acidosis because it was seen in the presence of severe hyperaklemia. An acidosis in this setting will increase the amount of extracellular potassium, which will lead to further instability of the myocardial cells. The cause of this acidosis is most likely multifactorial. One possible contributing factor was a result of the H+-K+ ATP-ase present in the distal tubule of the kidney. It is possible that the critically high levels of potassium in the renal tubule led to a selective retention of hydrogen ions in favor of the potassium ion, leading to an increased intravascular level of hydrogen ions. This acidosis was likely further exacerbated by mild dehydration secondary to gastrointestinal losses from vomiting.

    Life-Threatening Conditions

    What is clear from this case is that acute ingestion of potassium salts can lead to at least two acutely life-threatening conditions: acute, severe hyperkalemia and acute hemorrhagic gastritis, both of which require prompt recognition and treatment. Although hematemasis was not seen in the ED in this case, it occurred very shortly after transfer to the PICU, and easily could have started during the ED presentation.

    Most notably is the progression to gastric outlet obstruction and the presence of a metabolic acidosis. This case demonstrates a constellation of signs and symptoms unique within the literature, and warrants further study to establish the clinical effects and causes of an acute potassium chloride ingestion. This includes investigation into the safety of potassium-based water softeners, the amount of potassium chloride needed to be clinically detrimental, and potential forms of gastric decontamination in acute potassium overdose. Although relatively rare, it is imperative that the emergency physician be familiar with the consequences of this toxic ingestion, and prepared to manage them appropriately.


    1. Wetli CV, Davis JH. Fatal hyperkalemia from accidental overdose of potassium chloride: Case report. JAMA. 1978;240(13):1339.
    2. Illingworth RN, Proudfoot AT. Rapid poisoning with slow-release potassium: Case report. Brit Med J 1980;281(6238):485.
    3. Bedford PD. Acute potassium intoxication. Lancet. 1964:268.
    4. Ward C, Hamid S, Dow J. Gastric complications of massive slow-K overdose. Brit J Surg 1987;74:490.
    5. Warr OS, Nash JP. Jejunal ulceration: Report of a case apparently associated with potassium gluconate. JAMA. 1967;199(3):185.
    6. McMahon FG, Ryan JR, et al. Upper gastrointestinal lesions after potassium chloride supplements: A controlled clinical trial. Lancet 1982;2(8307):1059.
    7. Das G. You and your drinking water: health implications for the use of cation exchange water softeners. J Clin Pharm 1988;28(8):683.
    8. Full CA, Wefel JS. Water softener influence on anions and cations. Iowa Dent J 1983;69(4):37.
    9. Yarows SA, Fusilier WE, Weder AB. Sodium concentration of water from softeners. Arch Intern Med 1997;157(2):218.
    10. Graves JW. Hyperkalemia due to a potassium-based water softener. New Engl J Med 1998;339(24):1790.
    11. Gosselin JA. Hyperkalemia due to a potassium-based water softener. New Engl J Med 1998;339(24):1790.
    12. Mandal AK. Hypokalemia and hyperkalemia. Med Clin North Am 1997;81(3):611.
    13. Fox VL. Gastrointestinal bleeding in infancy and childhood: High-risk, underappreciated, obscure, or preventable causes of gastrointestinal bleeding. Gastroenterol Clin North Am 2000;29(1):37.
    14. LiVoti G, Acierno C, Tulone V, et al: Relationship between upper gastrointestinal bleeding and nonsteroidal anti-inflammatory drugs in children. Pediatr Surg Int 1997;12:264.
    15. Matsubara T, Mason W, Kashani IA, et al: Gastrointestinal hemorrhage complicating aspirin therapy in acute Kawasaki disease. J Pediatr 1996;128:701.
    16. Alcaraz A, Lopez-Herce J, Serina C, et al. Gatrointestinal bleeding following ketorolac administration in a pediatric patient. J Pediatr Gastroenterol Nutr 1996;23:479.
    © 2003 Lippincott Williams & Wilkins, Inc.