In a patient receiving lithium for psychiatric disorder, the onset of other disease and its treatment seems to have destabilized lithium control.
CASE HISTORY
A man aged 67 was admitted after an episode of transient left leg weakness, lasting 20 minutes. He was known to have type 2 diabetes, treated by diet but poorly controlled (glycosylated haemoglobin 11.8% [normal 4.4-6.0]). There was a long history of bipolar affective disorder treated with lithium carbonate 400 mg twice daily and more recently low-dose venlafaxine. On admission his serum lithium was 0.35 mmol/L (therapeutic range 0.4-1.1). CT of the brain was normal. A transient ischaemic attack was diagnosed and he was discharged home on metformin, gliclazide, pravastatin, aspirin and Asasantin Retard (dipyridamole and aspirin) as well as his usual maintenance dose of lithium carbonate and venlafaxine.
Eight days later he returned, having experienced confusion, anorexia and generalized shaking for three days. Random blood glucose was 3.0 mmol/L, blood urea 10.0 mmol/L (2.5-6.6), serum creatinine 154 μmol/L (62-115). He was judged to have symptomatic hypoglycaemia and was treated with oral Lucozade and a maintenance infusion of 5% dextrose. The hypoglycaemic medications were stopped. Next day his blood glucose concentrations were in the normal range but his symptoms were worse. He was increasingly drowsy, obtunded, and abulic and he had developed a coarse tremor with myoclonic jerks of his arms and legs. Repeat CT and MRI of the brain were normal. Serum and urine toxicology were negative. Cerebrospinal fluid white cell count was normal, protein 0.73 g/L (0.15-0.45), glucose 6.4 mmol/L (blood glucose 8.5 mmol/L). Serum lithium was raised at 2.02 mmol/L. The electroencephalogram showed diffuse slowing throughout the brain with no epileptiform activity. Encephalopathy secondary to lithium intoxication was diagnosed and all his psychotropic medications were stopped. He was treated with isotonic saline as fluid replacement and with three sessions of haemodialysis with a blood flow of 170 mL/min. Lithium levels fell to undetectable. His tremor and myoclonic jerks stopped and over the following week his toxic encephalopathy resolved. Detailed psychiatric evaluation, and checks with his local pharmacy and general practitioner on the dose and quantity of his tablets, yielded no evidence of a deliberate or accidental overdose.
COMMENT
Lithium intoxication has various clinical manifestations but neurological symptoms predominate.1 The onset tends to be gradual, with initial confusion, then impaired consciousness leading to coma and occasionally death.2 Neurological effects include seizures, choreiform and parkinsonian movements, and cerebellar signs (dysarthria, ataxic gait, incoordination).1,2 The symptoms do not necessarily correlate with lithium concentrations: toxicity has been reported at therapeutic levels and negligible symptoms at high levels.3 As a general rule aggressive treatment (intravenous fluids, haemodialysis) should be given when serum lithium exceeds 2 mmol/L, but this decision should be governed more by the clinical features than by the spot lithium level.3
The reason why in our patient the lithium concentration rose so rapidly from the subtherapeutic to the toxic range over a week is not clear. The medications prescribed after his first admission have not been reported to interfere with lithium and the combination of lithium and venlafaxine seems well tolerated in patients with bipolar affective disorder. Dehydration might have interfered with renal clearance and so raised the serum lithium, but the patient was clinically envolaemic. Oakley et al.4 have reported on 28 patients with severe lithium neurotoxicity, 26 of whom had chronic poisoning and the remaining 2 acute-on-chronic poisoning. They identified three risk factors contributing independently to lithium toxicity—namely, nephrogenic diabetes insipidus, age >50 years and thyroid dysfunction. Subnormal endogenous creatinine clearance was also implicated.
Haemodialysis is the cornerstone of treatment for lithium toxicity: with its low atomic weight and negligible protein binding, lithium is one of the most readily dialysable toxins.5 Haemodialysis should be stopped when the lithium concentration falls to the therapeutic range but may need to be repeated since rebound can occur, with high concentrations, after a treatment session.
References
- 1.Ilagan MC, Carlson D, Madden JF. Lithium toxicity: 2 case reports. Del Med J 2002;74: 263-70 [PubMed] [Google Scholar]
- 2.Meltzer E, Steinlauf S. The clinical manifestations of lithium intoxication. Isr Med Assoc J 2002;4: 265-7 [PubMed] [Google Scholar]
- 3.Nagappan R, Parkin WG, Holdsworth SR. Acute lithium intoxication. Anaesth Intensive Care 2002;30: 90-2 [DOI] [PubMed] [Google Scholar]
- 4.Oakley PW, Whyte IM, Carter GL. Lithium toxicity: an iatrogenic problem in susceptible individuals. Aust N Z J Psychiatry 2001;35: 833-40 [DOI] [PubMed] [Google Scholar]
- 5.Fenves AZ, Emmett M, White MG. Lithium intoxication associated with acute renal failure. South Med J 1984;77: 1472-4 [DOI] [PubMed] [Google Scholar]
