Share this article on:

Deciphering Cases of Lithium Toxicity

Gussow, Leon MD

Toxicology Rounds

Dr. Gussow is a voluntary attending physician at the John H. Stroger Hospital of Cook County in Chicago (formerly Cook County Hospital), an assistant professor of emergency medicine at Rush Medical College, and a consultant to the Illinois Poison Center.

A 47-year-old man is brought to the emergency department after reportedly ingesting an overdose of lithium carbonate 300 mg capsules (total dose not known) four hours before presentation. The lithium had been prescribed for his wife, who had been on the drug for several years. The paramedics report that he had no previous medical history, and was not on any medications.



Acute-on-chronic toxicity cases have the poorest outcomes because recent ingestion is superimposed on an existing body load of the drug

On examination, the patient appeared tremulous, and was oriented only to person. His speech was slurred. He denied nausea, but was not vomiting. Vital signs were blood pressure 132/84 mm Hg; pulse 105 beats per minute; respiratory rate 18 breaths per minute; temperature 98.9°F.

Neurological examination revealed altered mental status and symmetric generalized tremors in the hands and arms. Laboratory results showed normal electrolytes and renal function. Glucose was 132 mg/dL. The serum lithium level was 5.4 mEq/L. (Therapeutic levels are 0.6-1.2 mEq/L.) The urine screen was negative for other drugs.

Back to Top | Article Outline

Rationale for Dialysis

Lithium, the third element on the periodic table, is the lightest and smallest metallic element. Like sodium and potassium, it has a valence of +1. Although it was not generally used to treat mania and bipolar disorders until the 1970s, its medicinal and toxic properties have long been appreciated. The soft drink 7-Up once contained lithium citrate. In fact, back in the 1920s, it was called Bib-Label Lithiated Lemon-Lime Soda. Later, lithium chloride was marketed as a salt substitute for those who were on low-sodium diets. Unfortunately, this resulted in a number of deaths from lithium poisoning until its use was halted by the Food & Drug Administration in 1949.

Although lithium is not absorbed well by activated charcoal and has no antidote, it is readily removed from the extracellular compartment (i.e., the blood) by hemodialysis. While various authors have suggested specific lithium levels for which hemodialysis should be instituted, no such protocol has been adequately tested and validated. It is important that the clinician understand several basic concepts about lithium toxicity so rational decisions about submitting a patient to dialysis can be made. These key concepts are:

  • ▪ Neurological manifestations of lithium toxicity are determined by the lithium level in the brain, not in the blood. (Table 1.)
  • ▪ After acute ingestion, the passage of lithium from the blood into the brain is delayed. This is why the therapeutic effect of lithium does not occur immediately, and up to 10 days may be needed for complete distribution.
  • ▪ Although severe lithium toxicity can cause death, this is usually related to respiratory failure caused by coma, and should respond to supportive treatment. It is unlikely that hemodialysis alone prevents such deaths.
  • ▪ Prolonged high levels of lithium in the brain can cause long-term irreversible neurological deficits, such as ataxia, tremor, movement disorders, and short-term memory loss. It is reasonable to think that timely hemodialysis could prevent some of these sequelae.
Table 1

Table 1

Because of lithium's delayed distribution, there are three distinct patterns of toxicity. It is important for the clinician to try to distinguish among them in each case. The patterns are:

  1. Acute toxicity: This occurs when someone not chronically on lithium takes an overdose. Although blood levels can be high, initially there is little drug in the central nervous system, and neurological signs and symptoms are delayed. As with acute ingestion of many other metals, gastrointestinal manifestations (nausea, vomiting, diarrhea) predominate at first.
  2. Chronic toxicity: This occurs when, because of various precipitating factors, the lithium level rises despite a patient's adherence to prescribed dosing. (Table 2.) Because the patient has been taking lithium long-term, the brain starts out with a certain lithium level, and even small increases in the blood lithium can cause neurological toxicity. Because there is no acute overdose, GI effects are usually absent.
  3. Acute-on-chronic toxicity: This is the worst of both worlds. When a patient on chronic lithium therapy takes an acute overdose, the tank is already full, so to speak, and even small increases in the blood lithium can raise brain lithium concentration to dangerous levels. This can produce both GI symptoms and rapid onset of neurological toxicity.
Table 2

Table 2

Back to Top | Article Outline

Chronic Toxicity

Our case did not clearly fit any of these three patterns. The absence of nausea and vomiting, along with the apparent rapid onset of mental status impairment, was not consistent with the report of acute ingestion, and looked more like chronic toxicity. There was no history, however, that the patient had been on lithium therapeutically. This disconnect between the apparent history and clinical pattern was potentially important because in many cases the decision to start hemodialysis is based, at least in part, on the pattern of toxicity.

Table 3

Table 3

Early in acute toxicity, most of the lithium is in the blood, readily available for removal by hemodialysis. Functioning kidneys, however, do a reasonable job of clearing lithium, and even some patients with relatively high levels can do well with hydration and observation alone. The goal is to keep the lithium load from distributing into the central nervous system. It is reasonable to think, though, that above a certain level, the kidneys cannot keep up, and dialysis will be necessary to prevent neurological toxicity. Although the exact level at which this occurs is not known and certainly will differ from patient to patient, many authors recommend that hemodialysis be strongly considered in acute lithium ingestion for levels of 4.0 mEq/L and above.

In chronic lithium toxicity, increased levels more closely correlate with clinical presentation. Hemodialysis, while clearing the blood of lithium, will not immediately have access to lithium in the brain, and may have to be prolonged or repeated. Many authors suggest that in chronic toxicity, hemodialysis be considered for lithium levels of 2.5 mEq/L or higher.

Cases of acute-on-chronic toxicity often have the poorest outcomes because the recent ingestion is superimposed on an existing body load of the drug. Because of the delay before lithium in the blood enters the CNS, there is a period of time when it can be effectively removed by hemodialysis. The threshold for starting dialysis should be low in these cases.

Whatever the pattern of lithium toxicity, patients with renal insufficiency or significant neurological manifestations should be dialyzed. (Table 2.) And one more crucial point: In marginal cases where the decision is to forego dialysis, it is essential to monitor the patient's clinical condition and serial lithium levels carefully. A decreasing level may not reflect adequate renal elimination but rather passage of lithium into the intracellular compartment and brain. If the lithium level is going down and the patient is improving clinically, things are heading in the right direction.

Because of his markedly altered mental status and because the clinical picture was consistent with acute-on-chronic toxicity, the patient received hemodialysis for six hours, immediately after which his lithium level was 2.4 mEq/L. Eight hours later, his lithium level had rebounded to 3.8 mEq/L, and an additional six-hour dialysis session was done. His symptoms resolved, and he recovered without long-term sequelae. We never determined why the clinical presentation did not match the history.

© 2005 Lippincott Williams & Wilkins, Inc.