Original Author: Paul Perry, Ph.D, BCPP
Latest Revisers: Paul Perry, Ph.D, BCPP, Vicki Ellingrod,
Pharm.D., BCPP
Creation Date: 1996
Peer Review Status: Internally Peer Reviewed
Through 1978, approximately 123 cases of lithium intoxication have been reported since 1949 and most of them since 1963 (Hansen and Amdisen 1978). Acute overdosage results in a 25% mortality rate, while mortality in patients intoxicated during maintenance treatment with lithium is approximately 9%. Additionally persistent central nervous system or renal damage or both are estimated to occur in about 10% of patients (Hansen and Amdisen 1978). A more recent review of 213 cases reported through 1986, found that acute overdosage results in a 33% (26/80) neurologic morbidity rate (usually ataxia and tremor), while mortality in patients intoxicated during maintenance treatment with lithium is approximately 8% (15/193) (El-Mallakh 1986). In 1991 the American Association of Poison Control Centers reported 4149 cases of lithium exposures with 80% of the cases involving adults. Two thirds of all exposures were intentional and the mortality rate was less than 1%. Moderate to severe toxic reactions were seen in 15% (Krishel and Jackimczyk 1991).
PRESENTATION
It is important to recognize that the symptoms of lithium intoxication often develop gradually. The group of signs of impending intoxication include: vomiting, diarrhea, coarse tremor, sluggishness, sleepiness, vertigo, and dysarthria with slurred and indistinct speech. Patients ought to be educated as to the early signs of lithium intoxication so that they can differentiate these symptoms from an influenza virus and alert their physicians during the early stages of lithium intoxication. Signs of the fully developed intoxication include: coarse tremor, muscle twitching, muscular hypertonia with increased and sometimes asymmetrical deep tendon reflexes, seizures, attacks of hyperextension of the arms and legs, EEG changes, and a grayish hue of the skin. The patients sensorium may vary from confused to comatose. The intoxication may lead to a disturbance of fluid and electrolyte balance, fall in blood pressure, and possibly shock (Schou 1978).
In most patients, the optimum standard prophylactic lithium level varies between 0.45-1.00 mEq/L. However, the critical serum lithium concentration, that at which there is a danger of lithium accumulation, does not have a fixed value but may vary with the sodium intake and the factors that affect the minimum sodium requirement. Thus, it is particularly important for the clinician to be aware that inappropriate increases in the serum lithium concentration may be a harbinger of a pending lithium intoxication.
Lithium intoxication usually occurs at serum levels greater than 2 mEq/L (Thomsen and Schou 1975, Jefferson and Greist 1977, Fieve 1977). However, it is also possible to find toxic reactions to lithium within normal therapeutic serum levels. Shopsin et al (1976) reported six cases of lithium toxicity in 11 schizophrenics with a mean serum lithium level of 0.75 mEq/L and none exceeding a level of 1.25 mEq/L. Importantly, Rochford et al (1970) discovered that 37% of young adult psychiatric patients demonstrated neurological abnormalities such as organic brain syndrome, seizures, EPS, and/or EEG changes. However, only 5% of normal controls showed such changes while none occurred among patients with affective disorders. Tucker et al (1965) noted an abnormal EEG rate of 26% in depressives versus a 37% rate in schizophrenics. Regardless of the rates, it should not be surprising that the patients most likely to have neurological abnormalities such as schizophrenics or schizoaffectives are more likely to experience lithium toxicity at therapeutic levels.
ETIOLOGY
Lithium poisoning may be caused by (1) the intake of a single, large overdose with suicidal intent or (2) a reduction of the renal lithium clearance without a corresponding reduction of dosage. The renal lithium clearance may be reduced as a result of (1) kidney disease or (2) sodium deficiency and 3) water deprivation. The sodium deficiencies and water losses may be caused by (1) a low or no-added salt diet, (2) a weight reduction diet without the addition of extra salt and the reduced fluid intake (3) use of diuretic drugs, (4) extrarenal sodium and water loss, e.g., sweating from a fever, vomiting, diarrhea, etc. and (5) a rise of the serum lithium concentration above a certain patient-specific critical level (El-Mallakh 1986).
In a review of 100 cases of lithium intoxication from the literature, Hansen and Amdisen (1978) listed the following explanations for 75 intoxications: overdosage (16%), daily dose too high (35%), renal disease (9%), diuretics (8%), infectious disease (13%), gastroenteritis (8%), dehydration (13%), hot weather (1%), sodium-poor diet (8%), slimming diet (1%), anorexia (9%) , and depression with anorexia (4%). There is some overlap in these causes. In all the cases where information was available, either kidney disease or factors influencing fluid and electrolyte metabolism preceded the intoxication.
Of the fluid and electrolyte disturbances that lead to lithium intoxication it appears that dehydration may be the more important of the two. Thomsen and Olesen (1979) estimated the effect of water deprivation on lithium clearance in lithium-polyuric rats. During a 3 hour period of water deprivation, the rats lost water an amount corresponding to 10% of their total body weight. The lithium clearance fell to about 24% of the baseline value of the controls. In the animals who where only sodium chloride deprived, the lithium clearance only decreased to 77% of the baseline clearance value.
Lithium is not bound to plasma protein and is therefore filtered freely through the glomerular membrane. It is reabsorbed in the proximal tubules in the same percentage of the glomerular filtration rate as sodium and water (~75%), whereas it is neither reabsorbed nor secreted in the more distal tubular segments of the nephron. Thomsen and Olesen (1979) hypothesized that dehydration leads to volume contraction (decreased renal plasma flow) which further leads to an increase of the fractional proximal tubule reabsorption of sodium (> 75%) and a decrease of the fractional distal tubule reabsorption of sodium. An increase in the fractional proximal tubule reabsorption of sodium would lead to an increase in the fractional proximal tubule reabsorption of lithium and therefore a reduction in lithium clearance. However during volume contraction due to water and sodium deprivation, not all of the fall of the lithium clearance can be accounted for by decreased delivery of sodium, water, and lithium to the proximal tubule. Lithium is under these circumstances also reabsorbed to some extent more distally in the nephron. For this reason distal reabsorption of lithium may occur during the periods of water deprivation. Thus the decrease in lithium clearance in polyuric patients is a result of 1) reduction of renal plasma flow and of the glomerular filtration rate for sodium and lithium and 2) due to an increase of the amount of lithium being reabsorbed both proximally and distally in the nephron. The decrease of lithium clearance may lead to a rise of the serum lithium concentration and hence to intoxication.
LONG-TERM SEQUELAE
Most patients will completely recover after a lithium intoxication but for those with residual effects the sequelae will reflect cerebellar dysfunction. The most common neurologic residual signs are ataxic scanning speech, truncal ataxia, broad-base ataxic gait, incoordination of limb movements tremor of head and hands and nystagmus. Other patients may have signs of cognitive damage. These can include poor short-term memory, poor insight, limited comprehension and dementia. The majority of these signs may appear during the intoxication and persist for more than 2 months after the discontinuation of lithium. If signs continue for 6 months or longer they are often permanent (Groleau 1994).
TREATMENT
Hansen and Amdisen (1978) feel that according to their extensive experience that serum lithium levels of 1.5-2.5 mEq/L (mild, grade I) 12 hours after the last dose are usually accompanied by slight or moderate symptoms of intoxication. Twelve hour steady state levels of 2.5-3.5 mEq/L (moderate, grade II) are regarded as serious and 12 hour steady state levels > 3.5 mEq/L (severe, grade III) are to be considered as life-threatening. Based on El-Mallakh's review (1986) of 172 reported cases, mild lithium intoxications were characterized by drowsiness (% not calculated), hyperreflexia (44%), hypertonia (40%), fasiculations (27%), dysarthria (25%), ataxia (20%), and apathy (10%). Moderate intoxications were marked by the preceding symptoms plus impaired consciousness (71%), course tremor (45%), myoclonis (31%), lethargy (24%), choreoathetosis (7%), and paresis/paralysis (7%). Finally, the severe intoxications usually requiring hemodialysis presented with the preceding symptoms plus coma/stupor (80%), muscle twitching (48%), generalized or focal seizures (26%), and spasticity (9%). Neurologic sequelae occurred in 26/80 (33%) cases and in included ataxia 12/26 (46%), tremor 11/26 (42%), dyarthria 9/26 (35%), organic brain syndrome 6/26 (23%) and dysmetria 4/26 (15%). This latter group of patients had an average maximum serum lithium concentration of 3.51 mEq/l and were toxic for an average of 20 days. Of the patients who did not suffer neurologic sequelae, there maximum serum lithium concentration was 2.54 mEq/l while being intoxicated for an average of 11 days. Deaths occurred in 8% (15/193) of the cases. It was more likely to occur in patients who only received supportive treatment or in whom severe symptoms occurred (11%) while it was less likely to occur in patients who were treated with hemodialysis (5%). The cause of death was either circulatory collapse or respiratory failure. The mean serum lithium concentration in these patients was 4.24 mEq/l.
It is critical that patients are cognizant of the signs and symptoms of lithium intoxication so that the drug can be stopped at the first signs of an intoxication and the serum concentration decrease as quickly as possible. Hansen and Amdisen (1978) recommend fluid therapy or forced diuresis only in patients with early symptoms of lithium intoxication, normal renal function, and when it is certain that the lithium level has been elevated for only a few days and not above 2.5 mEq/L. During fluid therapy or forced diuresis, the serum lithium level should be determined at short intervals to assure that a lithium concentration of < 1.0 mEq/L will be reached within 30 hours after the start of treatment. If their criteria cannot be met, dialysis should be initiated.
|
TABLE 1: THERAPEUTIC APPROACHES TO LITHIUM INTOXICATION |
|||
|
TREATMENT |
(NO SEQUELAE) |
|
|
|
|
|
|
|
|
Support |
|
|
|
|
Saline Diuresis |
|
|
|
|
Hemodialysis |
|
|
|
|
Peritoneal Dialysis |
|
|
|
The treatment of lithium intoxication should be individualized. The aim of the treatment is to provide general supportive care while removing excessive lithium from the body as rapidly as possible. The severity of intoxication appears to be related to both serum concentration and the duration of exposure to the toxic level (Hansen and Amdisen 1978). Horowitz et al (1969) reported a case where a patient was exposed to a level of 8.2 mEq/L for only a brief duration without any neurological symptoms. Therefore, in the case of acute overdose, gastric lavage or induced emesis should be utilized to remove unabsorbed lithium. However, since lithium intoxication is usually a chronic problem rather than an acute one, emesis or gastric lavage may not be always worthwhile. If a patient shows signs and symptoms of lithium intoxication, lithium administration should be stopped and serum lithium concentrations should be determined. Supportive and corrective measures such as maintaining adequate blood pressure and prevention of pulmonary infection in comatose patients may be necessary. Thomsen et al (1975) suggested saline infusion, 1-2 liters within the first six hours to enhance the clearance of lithium from the body. However, they incorrectly pointed out that sodium chloride is only effective in a patient who has a reduced lithium clearance secondary to low-sodium diet or long-term diuretic therapy. Recently, Holstad et al (1988) demonstrated that in a group of normal subject volunteers administered lithium for a week, a 2 liter normal saline infusion increased lithium clearance to 40 ml/min (range = 28-53 ml/min). In patients with reduced lithium clearance due to glomerulonephritis or pyelonephritis, however, large doses of sodium chloride lead only to a slight rise of lithium clearance and may carry a risk of fluid overloading. Therefore, in patients with normal sodium balance, other methods must be used to increase the excretion of lithium. Diuretic drugs such as furosemide or the thiazides should not be used because they deplete sodium from the body. Other types of diuretics that act on the proximal tubules rather than the distal tubules, such as aminophylline, urea, acetazolamide or sodium carbonate, have been shown to increase lithium clearances by 30-60% (Thomsen and Schou 1968). However, this approach has been superseded by the use of dialysis which is found to be more effective (Jefferson and Griest 1978). Both peritoneal dialysis and hemodialysis are quite effective in removing lithium from the body. Thomsen et al (1975) estimated that peritoneal dialysis can result in an addition of about 15 ml/min to the renal clearance of lithium. A normal renal lithium clearance is 15-30 ml/min. Hemodialysis may account for an addition of as much as 50 ml/min to the lithium clearance. Current hemodialyser efficiency requires a 4-6 hour dialysis which is repeated if necessary. Serum lithium levels must be determined at short intervals after dialysis to ascertain that the level remains low. A serum lithium level of < 1.0 mEq/L 6-8 hours after dialysis is regarded as an endpoint to treatment (Hansen and Amdisen 1978). However, if the post-dialysis lithium level rebounds to a level greater than 1.0 mEq/l and the patient is symptomatic, a second dialysis treatment might be considered. Table 1 contrasts the efficiency of supportive care, saline infusion, and dialysis in the treatment of lithium intoxications (El-Mallakh 1986).
Perry et al (1984) examined the effect of chronic oral theophylline on lithium clearance. Theophylline concentrations ranging from 5.4-12.7 ug/ml demonstrated a mean increase of 30% in the lithium clearance. However, a significant linear correlation between the serum theophylline concentrations and the percent increase in lithium clearance predicted that theophylline levels between 15 and 20 mg/ml would produce lithium clearance increases between 62 and 107%. Thus intravenous theophylline infusions might be an alternative to dialysis in the treatment of lithium intoxication.
Holstad et al (1988) contrasted the effect of theophylline infusion to saline infusion on lithium clearance in a group of 10 normal subjects given lithium carbonated 1200 mg/d for one week. The 12 hour theophylline infusions significantly increased lithium clearances from 40-61 ml/min when compared to the 12 hours 2 liter normal saline infusions. The mean theophylline concentration required to produce this effect was 14 mg/ml (range = 11-19 mg/ml). However, the viability of this treatment remains to be demonstrated in patients who are intoxicated from lithium.
REFERENCES
El-Mallakh RS (1986). Acute lithium neurotoxicity. Psychiatric Dev 4:311-28.
Fieve RR (1977). Clinical use of lithium in affective disorders. Drugs 13:458-466.
Hansen HE, Amdisen A (1978). Lithium intoxication (Report of 23 cases and review of 100 cases from the literature). Quart J Med 47:123.
Holstad SG, Perry PJ, Kathol RG, et al (1988). The effects of theophylline infusion versus sodium bicarbonate infusion on lithium clearance in normal subjects. Psychiatry Res 25:203-11.
Groleau G (1994). Emerg Med Clinics North Am; 12: 511.
Horowitz LC, Fisher GU (1969). Acute lithium toxicity. N Engl J Med 281:1369.
Jefferson JW, Greist JH (1977). Lithium poisoning. In: Primer of Lithium Therapy, Baltimore: Williams and Wilkins Co, 201-6.
Krishel S, Jackimczyk K (1991). Cyclic antidepressants, lithium, and neuroleptic agents. Pharmacology and toxicology. Emerg Med Clinics North Am; 9: 53.
Perry PJ, Calloway RA, Cook BL et al (1984). Theophylline precipitated alterations of lithium clearance. Acta Psychiatrica Scand 69:528-37.
Rochford JM, Detre T, Tucker GJ, et al (1970). Neuropsychological impairments in functional psychiatric diseases. Arch Gen Psychiatry 22:114-119.
Schou M (1978). Preclinical and clinical pharmacology of lithium. In: Principles of Psychopharmacology. New York: Academic Press 343-56.
Shopsin G, Johnson G, Gershon S, et al (1976). Neurotoxicity with lithium: Differential drug responsiveness. Int Pharmacopsychiatry 5:170-182.
Thomsen K, Schou M (1968). Renal lithium excretion in man Am J Physiol 215:823-827.
Thomsen K, Schou M (1975). The treatment of lithium poisoning. In: Johnson FN ed., Lithium Research and Therapy. London: Academic Press, 227-36.
Thomsen K, Olesen OV (1979). The effect of water deprivation on lithium clearance and lithium excretion fraction in lithium-polyuric rats. J Pharmacol Exp Ther 209:327-9
Tucker GJ, Detre T, Harrow M, et al (1965). Behavior and symptoms of psychiatric patients and the EEG. Arch Gen Psychiatry 12:278-86.