Potassium depletion in eating disorders can be fatal. As people with disordered eating become aware of the risks, they may seek on their own, or obtain on physician recommendation, oral potassium supplements. Such supplements are endorsed on the internet and are widely used, either with medical endorsement or spontaneously by patients as they become aware of actual or potential deficiencies. Such patients may not disclose their worry or their self-medication to physicians. Possible consequences include hypo- or hyperkalemia, potentially fatal. In this paper, these dangers are illustrated with composite cases. Recommendations are made.

• Potassium depletion

• Eating disorders

• Supplements

Harper G, Epstein D. Dangers of Oral Potassium in Eating Disorders. Ann Pediatr Child Health 2022; 10(6): 1286.

Among the risks associated with anorexia nervosa (AN) and bulimia nervosa (BN) is hypokalemia, potentially fatal. Oral potassium chloride supplements are often used to manage this condition. But the medical literature says little about the risks of such practice. Those risks, including fatal or near-fatal outcomes, are illustrated here.

Hypokalemia, secondary to vomiting or laxative abuse, is reported in up to 19.7% of outpatients with eating disorders [1]. Potassium is depleted in vomiting through loss of gastric hydrochloric acid with secondary metabolic alkalosis and compensatory renal secretion of potassium (“When H+ is low, K+ must go”) and through emetic loss of sodium ion with secondary volume constriction and compensatory aldosterone-mediated potassium excretion [2]. With laxatives, potassium is lost through renal compensation for intestinal losses of water and sodium.

In one review of adults with chronic hypokalemia, patient behavior was implicated in 14 of 27 patients: 5 abused diuretics, 8 had surreptitious vomiting, and 1 abused laxatives [3]. When vomiting or laxative abuse is concealed, the cause of potassium deficiency may not be identified. Sometimes, exhaustive laboratory evaluation of hypokalemia may yield clinical evidence of an eating disorder, while the patient continues to deny that possibility [4].

In other conditions where potassium is lost, oral potassium supplements are often prescribed. For patients with or without eating disorders, such supplements are often mentioned in on-line resources, usually with encouragement to seek medical care (see Box). In contrast, peer-reviewed articles on potassium deficiency and its treatment [5,6], say less of the use of supplements or the associated dangers. For example, in an interactive case in the New England Journal of Medicine, a patient with chronic AN has repeated crises in which serum [K+] falls as low as 1.9 mEq/L. But the discussion does not consider the possibility that concealed use of oral potassium supplements inbetween medical checkups may have masked the degree of underlying nutritional compromise that became apparent in crises [7].

The following composite scenarios illustrate the dangers.

A patient in her mid-20s maintains a low weight (37 kg), partly by taking up to 56 “Dulcolax” per night. Hypokalemia as low as 2.0 mEq/L is managed with “K-Lor” supplements. The patient is believed to use the “K-Lor” responsibly. But when gastroenteritis occurs, the patient takes vomiting, fatigue and weakness for signs of hypokalemia and increases her “K-Lor” dose to 120 mEq/day. Shaky, nauseated and diaphoretic, she is seen in an emergency ward where the initial impression is of hypokalemia, and she is nearly given intravenous potassium before an electrocardiogram indicates hyperkalemia with [K+] of 8 mEq/L.

A teenager with AN has vomiting, often concealed, and chronic hypokalemia. Advised of the dangers of potassium supplements, her parents bring to the physician what they believe to be all her pills. The patient denies having any more. For several weeks the patient’s electrolytes are normal. When her [K+] drops below 2.5 mEq/L, the patient acknowledges that, before clinic visits, she had taken hidden potassium, “So they wouldn’t know how much I was vomiting.”

A teenager with AN has sporadic hypokalemia treated with oral potassium. She vomits openly. Her parents see her gorging on water before medical checkups, then rushing to the bathroom after being weighed, but they do not report this. “We felt she deserved some help to pass the doctor’s test,” they explain. The pediatrician notes potassium of 3.0 mEq/L and prescribes potassium supplements, 50 mEq daily. The patient’s weight drops another 2.5 kg. [K+] is still low. Either potassium is not being taken as prescribed or the patient needs a higher dose. The doctor calls the parents, warns them of the dangers, and advises them to double the potassium dose. The daughter, refusing to call the doctor, continues vomiting and rushing about. The next day, following an argument, she goes to her room where she is found moribund 20 minutes later. Postmortem serum potassium is 3.5 mEq/L, probably elevated by cellular lysis.

The use of a potentially lethal drug to treat a potentially fatal condition requires risk-benefit analysis. The possible benefit is normalization of potassium deficiency with decreased risk of arrhythmia. But this requires that

• the patient’s ongoing losses are stable, or, if changing, can be ascertained promptly;

• the physician has accurate information about the patient’s metabolic deficiency;

• the patient’s requirement for replacement can be accurately calculated; and

• the medicine is taken as prescribed.

These requirements are often not met.

First, patients often do not take prescribed medication. Second, confused patients may be unaware of how much they are vomiting or purging. Third, vomiting and purging vary over time, as in Scenario One. Fourth, the patient may use prescribed potassium to manipulate her serum level before laboratory tests, as in Scenario Two. Finally, patients or parents may conceal and misrepresent, as in Scenario Three.

The dangers of prescribing oral potassium include:

1. Psychological Effects of Prescribing. Patients afraid of being controlled can experience the prescription of medicine “to set things right” as a threat. Panic and vomiting increase. Increased vomiting makes the potassium deficit worse.

2. Misuse of Medication by the Patient, as in Scenario One. Prescribed potassium becomes part of the patient’s management of her eating disorder. Misuse of the prescribed medicine can precipitate life-threatening hyperkalemia.

3. Clouded Biochemical Assessment. The physician prescribing oral potassium supplements cannot know whether a given serum level reflects recent oral doses not yet distributed in the body or is a more equilibrated reflection of body stores - “hypokalemia and potassium deficiency are not synonymous.” The problem is even greater when deliberate self-dosing by the patient occurs, as in Scenario Two.

4. Abetting Family Aversion. Families struggling with exhaustion and ambivalence may take a prescription as license to disregard the patient’s high-risk behavior, thinking that medications will solve the problem. Aversion is the most dangerous form that exhaustion and rage can take [8].

5. Creating a False Sense of Physician Security. Physicians prescribing medications to patients with eating disorders must consider the possible effects of the prescription on the patient and the family, including the possibility of a miscalculated or forgotten dose, of self-medication with or without suicidal intent, and of an unrecognized problem in the relationship with the prescriber [9]. Physicians who overestimate their alliance with the patient may miss signs that more vigorous intervention, such as hospitalization, is indicated.

Reflecting awareness of these dangers, access to oral potassium may be regulated, as in the United Kingdom (not available at all). In the United States, access is highly variable, with prescription and monitoring by a physician sometimes advocated, but not consistently enforced.

1) Those caring for persons with AN and BN do well to keep in mind that misrepresentation or frank lying may be as much a part of eating disorders as it is of substance abuse.

2) Let patients know that you understand that they may be afraid or uncertain as to what to say about what’s going on with them and want to hear whatever they are able to share.

3) Given the risks associated with disordered potassium, getting the accurate story is even more important than in other conditions.

4) The collateral data needed, in addition to what the patient can provide, may come from family members, peers, or more frequent medical visits.

5) Remember, and advise patient and others, that a point-in time laboratory test does not provide a complete picture of the patient’s metabolic/nutritional state.

6) Serious health risks may make it necessary to eliminate over-the-counter potassium supplement sales, as in the United Kingdom.

The use of oral potassium supplements in the patient with potassium deficiency secondary to AN or BN carries dangers of both hypo- and hyperkalemia. Medical action based on inaccurate information about the patient should not be allowed to increase the risk of death in these serious disorders.

Contributions from Thrassos Calligas MD and the late Leon Eisenberg MD are acknowledged.

1. Miller KK, Grinspoon SK, Ciampa J, Hier J, Herzog D, Klibanski A. Medical findings in outpatients with anorexia nervosa. Arch Intern Med. 2005; 165: 561-566.
2. Winston AP. The clinical biochemistry of anorexia nervosa. Ann Clin Biochem. 2012; 49: 132-143.
3. Gladziwa U, Schwarz R, Gitter AH, Bijman J, Seyberth H, Beck F, et al. Chronic hypokalaemia of adults:   Gitelman’s syndrome is frequent but classical Bartter’s syndrome is rare. Nephrology, Dialysis, Transplantation. 1995; 10: 1607-1613.
4. Hughes C, Koppanarayana S, Watson M, Nipah R, Laing I. Hypokalemia: A Curious Case in a Young Woman. J Appl Lab Med. 2020; 5: 802-807.
5. Seidler T, Jacobshagen C, Bauer M, Hasenfuss G, Waeschle RM. Distribution of potassium levels on admission for CPR – severe hypokaelemia with dysmorphophobic eating disorders. Resuscitation. 2011; 82: 535-537.
6. Pepin J, Shields C. Advances in diagnosis and management of hypokalemic and hyperkalemic emergencies. Emerg Med Pract. 2012; 14: 1-17.
7. Sacks Z, Vaidya A, Sharma N, Gottlieb B. A patient found unresponsive. Interactive Medical Case. N Engl J Med. 2012; 367: e36.
8. Maltsberger JT, Buie DH. Countertransference hate in the treatment of suicidal patients. Arch Gen Psychiatry. 1974; 30: 625-633.
9. Mintz DL, Flynn DF. How (not what) to prescribe: nonpharmacologic aspects of psychopharmacology. Psychiatr Clin North Am. 2012; 35:143-163.