Lithium: Updated Human Knowledge Using an Evidence-Based Approach
- First Online:
- Cite this article as:
- Grandjean, E.M. & Aubry, JM. CNS Drugs (2009) 23: 397. doi:10.2165/00023210-200923050-00004
- 885 Downloads
Lithium use in mental diseases has changed over the years but remains a cornerstone of treatment in bipolar disorders. In two companion papers, we have reviewed existing (and especially recent) data on lithium efficacy and updated basic knowledge regarding the practical fundamentals of lithium therapy. The present paper reviews safety data on lithium available to date.
Gastrointestinal pain or discomfort, diarrhoea, tremor, polyuria, nocturnal urination, weight gain, oedema, flattening of affect and exacerbation of psoriasis are typical complaints of patients receiving long-term lithium therapy. Renal involvement results in a reduced urinary concentrating capacity, expressed as obligate polyuria, with secondary thirst. With long-term therapy, this may result in nephrogenic diabetes insipidus. In addition, glomerular filtration rate falls slightly in about 20% of patients. The view that only a few patients receiving long-term lithium are at increased risk of glomerular impairment and progressive renal insufficiency should be regarded with caution. The risk is increased in case of concomitant diseases or medications.
Lithium treatment may inhibit thyroid hormone release and induce goitre. Consequently, the prevalence of both overt and subclinical hypothyroidism is increased, with circulating thyroid auto-antibodies frequently being found. Much less commonly, thyrotoxicosis may also develop in association with lithium therapy.
Long-term lithium treatment may also be associated with persistent hyperparathyroidism and hypercalcaemia, as well as with hypermagnesaemia. Overweight of up to 4–10 kg is found in approximately 30% of lithium-treated patients.
Most neurological manifestations are benign, for example, the fine postural and/or action tremor present in 4–20% of patients. This is increased by high caffeine consumption and concomitant use of other psychotropic agents. A number of rare, potentially serious neurological adverse effects have been reported, including extrapyramidal symptoms, ‘pseudotumour cerebri’ or occasionally cerebellar symptoms. Severe neurological sequelae are exceptional.
Cognitive disturbances are often mentioned as a lithium-related adverse effect. The few controlled studies do show a statistically significant negative effect of lithium on memory, vigilance, reaction time and tracking. There are frequent reports of mild effects of lithium on cognition at therapeutic serum concentrations.
A number of deaths associated with lithium treatment have been reported. The most serious issue is that of non-accidental overdose, i.e. either long-term overdosage or acute overdose on long-term treatment. Progressive renal insufficiency, an exceptional complication of long-term lithium therapy, may also have a fatal outcome.
In relation to pregnancy, lithium salts are rated as category D (positive evidence of risk). Therefore, prescription of lithium should be avoided during the first trimester of pregnancy unless the benefit to the mother exceeds the risk to the fetus.
Although lithium transfer into breast milk is well established, the long-term fate of babies breast-fed by mothers receiving lithium therapy is unknown. Whether lithium therapy is safe in breast-feeding women is controversial. Although there is no absolute contraindication, it is known that the kidney is particularly sensitive to lithium just after birth.
Intoxication in patients on long-term treatment with lithium in the absence of history of acute ingestion is not rare. Contributing factors include change in daily dose, long-term high dosage, kidney disease or drug interaction. In suspected cases, serum concentrations should be obtained early and repeatedly. In addition to supportive measures, haemodialysis is the treatment of choice for severe cases.
Thorough knowledge of the limitations and drawbacks of lithium therapy is mandatory for its optimal use, especially at a time when its risk/benefit profile needs to be compared accurately with that of antiepileptic drugs and other mood stabilizing medications.