This site is intended for health professionals only

At the heart of general practice since 1960

Lithium toxicity presenting as delirium in an older patient

How does lithium toxicity present?

What steps can be taken to prevent it?

How should it be treated?

How does lithium toxicity present?

What steps can be taken to prevent it?

How should it be treated?

Around 1 in 400 adults suffer from bipolar disorder at some point during their lives although it is unusual for the condition to develop after the age of 40.1 The prevalence of bipolar disorder in the elderly is reported to be 1%.2

Medical management with lithium, valproate or olanzapine is recommended as first-line therapy by NICE.3 Careful monitoring of lithium levels is essential, in particular to prevent toxicity3 as lithium has a low therapeutic index.4 Older patients are especially vulnerable, with one study reporting that over a period of 9 years 4% of elderly patients on lithium were hospitalised because of lithium toxicity.5

Case report

A 79-year-old woman who lived independently alone was admitted to a medicine for the elderly ward, by her GP, with a short history of confusion. She rarely attended the practice. During the preceding week her mental state had deteriorated markedly. On a house call the night before admission she had been found to be delirious, fevered and dysphasic.

She subsequently fell overnight, sustaining a minor head injury. Her son found her in the morning unable to walk without assistance.

Her past medical history included bipolar disorder, hypothyroidism, and mild cognitive impairment. (The QOF indicators for mental health and hypothyroidism are listed in tables 1 and 2 respectively attached.) Neuropsychological testing in 2006 had shown disturbances in memory, new learning, language and visuospatial functioning. Current medications were lithium carbonate 500mg nocte and levothyroxine 150mcg mane. Her GP suspected that her compliance with levothyroxine was poor as a recent TSH of 38.95mU/l had been recorded (normal range 0.4 - 4.0mU/l). The patient attended the local psychiatric day hospital for monitoring of lithium and levels one week prior to admission had been subtherapeutic at 0.36mmol/l (normal range 0.4-0.8mmol/l).

On admission the patient was unable to give a history or engage in a mini mental state examination. She was pyrexial, tachycardic at 100bpm, normotensive, had dry mucous membranes, and a small laceration on her left temple caused by her fall. Neurological examination showed her to be drowsy (Glasgow Coma Scale 12) but agitated with a severe predominantly expressive dysphasia. Marked twitching in her legs and hands was noted.

The initial differential diagnosis was: infection; new stroke disease or subdural haematoma. A CT brain scan showed longstanding atrophic and ischaemic change. Chest X-ray was clear, urinalysis was negative, and ECG showed sinus tachycardia. Blood results were unremarkable with the exception of an elevated lithium level of 1.93mmol/l.

The patient received a single session of haemodialysis, which reduced the lithium to within the normal range. Over the next few days the pyrexia, twitching and dysphasia gradually resolved and she returned to her previous level of function.

The cause of the patient's lithium toxicity was initially unclear. There was no evidence of concurrent illness and her biochemical parameters were normal. She had not intended to overdose. Since her only other medication was levothyroxine and she had not been taking any OTC medications it was thought unlikely to be a drug interaction. It was therefore suspected that her lithium dose had been increased in response to a sub-therapeutic lithium level of 0.36mmol/l the week before. However, the lithium clinic said that this was not the case. She had not had any dose adjustments recently but, as poor compliance had been suspected, she had been issued with a compliance device the preceding week. The resultant good compliance following the introduction of the device appears to have led to lithium toxicity.

Lithium toxicity

Lithium toxicity is often non-specific in presentation, and symptoms may be clouded by comorbidity in elderly patients. It may result from excessive lithium intake, either intentional or accidental, or reduced lithium excretion. The latter encompasses a range of conditions including dehydration, low sodium levels, renal disease or the addition of interacting medications.6

Severity scores

The severity of lithium toxicity is graded into mild, moderate and severe by clinical features, rather than serum lithium levels (see table 3, attached).7

Regular use of lithium results in its accumulation within body tissues. Subsequently acute-on-chronic and chronic lithium toxicity is associated with an extent of baseline tissue saturation.8 This is a consequence of progressive renal impairment secondary to long-term lithium use and leads to prolongation of the lithium half-life. Interestingly, Waring et al reported that patients with acute-on-chronic and chronic lithium toxicity had higher poisoning severity scores than those with acute toxicity, despite all the patients having similar serum lithium concentrations.8 Poisoning severity scores grade poisoning into mild, moderate, severe and fatal, based upon the pattern of clinical features discussed earlier, rather than serum lithium levels.8

Treatment

The National Poisons Information Service is a valuable on-line resource for physicians and should be consulted in a case of poisoning.9 Haemodialysis remains the cornerstone of treatment for severe lithium toxicity and should be considered in all patients with marked neurological features.4,7,9 Other supportive measures should also be considered such as ensuring hydration, correcting electrolyte imbalance and controlling convulsions with iv diazepam.9

The onset of symptoms of toxicity may be delayed and following treatment clinical improvement takes longer than the corresponding reduction of lithium concentration in the blood.7 In suspected acute-on-chronic cases lithium levels should be measured immediately, at 6 hours post dialysis, and then repeated every 6-12 hours.9 The level may initially rebound following haemodialysis because of baseline tissue saturation and subsequently further treatment may be required.

Compliance aids

Our patient presented with delirium caused by lithium toxicity which appears to be associated with the introduction of a compliance aid.

Despite their widespread use in the community there is little evidence or guidance for the effectiveness of compliance devices.10, 11

Multi-compartment compliance devices come in a variety of types, based on the concept of using separate compartments for different times of the day. 10, 12

There are some potential complications of compliance aids but again the evidence is sparse:10, 13

•they are not useful for intentional non-compliance
•there are few data on the stability of medications within the compartments
•the introduction of compliance aids may lead to reduced awareness of the patient's own medications.11

However, in carefully selected individuals such devices may improve compliance.14 For example, these devices may prove useful for patients on complex drug regimens and particularly for patients taking drugs such as lithium that have a narrow therapeutic window.

The benefit for patients with significant cognitive impairment is doubtful and each patient should be assessed individually for suitability.15

Once patients have started to use a compliance aid they should be regularly assessed to ensure that the device is still appropriate for their needs.13,16 The impact of better adherence with prescribed medications needs to be observed. Improved compliance with such devices may be comparable with that resulting from supervised administration of medicines during hospital admission. This frequently results in the emergence of unwanted side-effects from those medications with which the patient had previously been non-compliant. For example, hypotension is often seen in inpatients who may have been prescribed antihypertensives or diuretics in the community, but were not actually taking them as intended.

Table 1 Table 2 Table 4 Useful information

National Poisons Information Service
www.toxbase.org

Authors

Dr Angela Wilkinson
MBChB MRCP
specialist registrar in medicine for the elderly

Dr Anna Gavine
BMSc(Hons) MBChB
foundation doctor

Lithium toxicity presenting as delirium in an older patient

Rate this article  (5 average user rating)

Click to rate

  • 1 star out of 5
  • 2 stars out of 5
  • 3 stars out of 5
  • 4 stars out of 5
  • 5 stars out of 5

0 out of 5 stars

Have your say