This site is intended for health professionals only

At the heart of general practice since 1960

The importance of early GP accounts this year

As the row over NICE proposals to ration Alzheimer's drugs drags on, Professor Robert Howard reminds GPs that atypical antipsychotics have a useful role, in spite of the Committee on Safety of Medicine's safety alert last year

Take-home points

 · Behavioural symptoms in dementia have complex aetiology

 · Drug treatment should be considered only after environmental, psychological and educational interventions have failed

 · All drug treatments carry a risk of sedation, worsening of cognitive decline, stroke and extrapyramidal side-effects and this should be explained (if appropriate) to patients and their carers

 · The best evidence is for treatment with low-dose atypical antipsychotic, eg risperidone 0.5mg once daily, increasing if necessary to twice daily

 · Many patients can stop treatment after a few weeks without worsening of symptoms

 · Use of benzodiazepines or agents that act solely through sedation is not supported by an evidence base and may be most dangerous

At some point during their illness, almost all patients with dementia will present with behaviours that are difficult for family or institutional carers to manage. Unsurprisingly, these symptoms are much more distressing to patients and their carers than cognitive deficits and are often a major factor in decision-making at the point of institutionalisation.

Identifying triggers for

difficult behaviour

The emergence of behavioural symptoms is usually a consequence of interactions between the dementia, the patient's physical and mental health and his or her physical and personal environment.

A man who has become aggressive towards his wife, for example, may be in pain or even harbour delusional beliefs about who she is.

His aggression might only be triggered by her efforts to dress him or take him to the lavatory ­ aspects of his personal care that he may not recognise that he needs help with.

To be successful, any management strategy must involve an initial assessment of the patient and a consideration of potential contributing factors.

If there is no clear physical explanation for the behaviour, might the patient be depressed or have a psychosis?

Is the patient within an environment where he feels comfortable and happy and are his carers capable of understanding his needs and communicating with him?

Often, those involved in the care of a person with dementia who has begun to exhibit a difficult-to-manage behaviour can be encouraged to take a step back from reacting to the behaviour.

Rather, they can be helped to think in terms of how modification of their own behaviour towards the patient might prevent or reduce expression of the problem.

Drug treatment

When behavioural disturbances have no clear situational or other trigger or occur in a setting where carers cannot be expected to cope with them, then drug treatment is indicated.

Choosing a drug for this situation has been complicated by the Committee on Safety of Medicines (CSM) message on the use of atypical antipsychotics in patients with dementia1.

In March 2004 the CSM advised that risperidone or olanzapine should not be used for the treatment of behavioural symptoms and that risperidone should only be used in the short-term and under specialist advice for acute psychotic conditions.

The CSM safety warning was issued in response to an apparent threefold increase in risk of cerebrovascular adverse events in dementia patients treated with risperidone and olanzapine compared with those who had received placebo.

Typically, during a 12-week trial, about 3 per cent of patients prescribed an atypical and 1 per cent randomised to receive placebo suffered a transient ischaemic attack or stroke.

The CSM warning only involved the atypicals, although the risk of cerebrovascular adverse events in dementia patients who are prescribed more traditional or typical antipsychotics is unknown.

However, at least one recent study has suggested that in a general elderly population the risk of stroke is no higher with prescription of atypicals rather than typicals3.

Further, the atypical antipsychotics risperidone and olanzapine have the best placebo-controlled evidence base for effectiveness against aggression, agitation and psychosis2. They are less likely to cause the common (extrapyramidal, vasomotor, anticholinergic or sedating) side-effects associated with typical antipsychotics.

The best evidence is for treatment with low-dose atypical antipsychotic, eg risperidone 0.5mg once daily, increasing if necessary to twice daily.

Use of benzodiazepines or agents that act solely through sedation is not supported by an evidence base and may be most dangerous.

Recently, the Faculty of the Psychiatry of Old Age of the Royal College of Psychiatrists has issued good practice guidance for the prescribing of atypicals in the management of behavioural and psychological symptoms in dementia4.

Its reasons for doing so are stated in the introduction:

'Since the CSM alert there have been reports of inappropriate interpretations of this guidance including groups of patients having their medication withdrawn en masse without considering the individual circumstances of the patients and/or switching patients to other medications which are likely to have more harmful side-effects.

'There have also been long-standing concerns about inappropriate use of antipsychotic medication in older people with dementia namely: drugs given for the wrong reason

(eg given for depression) or without any documented reason for the prescription, two or more antipsychotics prescribed at the same time, drugs given at too high a dose and for too long without reviewing the need or the dose.'

The guidance stresses the importance of an assessment of the risks and benefits of drug treatments in each case and recognises this will often mean that atypical antipsychotics continue to be the drugs of choice.

The guidelines also emphasise that drug treatment should be time-limited.

Once behaviour has settled it is generally possible to stop prescription of one of these drugs without causing difficulties5.

Many patients can stop treatment after a few weeks without worsening of symptoms.

'3T' approach for prescribing

The Faculty of the Psychiatry of Old Age4 3T approach for managing behavioural and psychiatric symptoms of dementia:

·Drug treatments should have a specific Target symptom

·The starting dose should be low and Titrated upwards

·Drug treatments should be Time limited

Royal College of Psychiatrists' guidance on atypical antipsychotics for treating specific symptoms

1 The atypical antipsychotics risperidone and olanzapine have the best evidence base for effectiveness compared to placebo for physical aggression, agitation and psychosis

2 The effect of atypical antipsychotics in these situations is not entirely attributable to sedation

3 Typical antipsychotics are effective with similar symptoms but have a weaker evidence base

4 Similar types of side-effects can occur with all antipsychotics but the severity and frequency of each side-effect is different in the two groups.

·At effective doses typical antipsychotics tend to have more side-effects which are more severe.

Typical antipsychotic side-effects are more likely to include extrapyramidal side-effects, tardive dyskinesia, anticholinergic side-effects (with possible acceleration of cognitive decline) and drowsiness (in higher doses).

·At their usual doses atypical antipsychotic side-effects are more likely to include weight gain, disrupt blood glucose control, hyperlipidaemia and CVAEs.

At higher doses sedation and extrapyramidal side-effects can occur as well.

Both classes of drugs can cause paradoxical agitation.

The choice between the two classes of drugs should be informed by these general side-effect profiles as well as any of the individual circumstances of the case.

5 A decision to start atypical antipsychotic drugs should be adequately documented and all the factors considered in making this decision should

be recorded.

A clear date to review the need for these drugs should also be noted.

Royal College of Psychiatrists' guidance on managing patients already on atypical antipsychotics

1 The decision to continue an atypical antipsychotic is best taken by clinicians on a case-by-case basis on the balance of potential risks and benefits in the same way that a decision is made for initiating drug treatment.

2 Long-term treatment with antipsychotics carries cumulative risks of cognitive decline, falls and other side-effects. The need for continuing treatment with antipsychotics should therefore always be reviewed.

A recent study suggests that antipsychotics could be withdrawn successfully in people who have been relatively free of behavioural symptoms for at least three months.

It is prudent to withdraw antipsychotic drug treatment cautiously and gradually unless there are specific and distressing side-effects from medication.

3 However, not everyone on atypical antipsyhotics should have their drug stopped or changed.

Behavourial and psychiatric symptoms of dementia (BPSD) can persist in the long-term and is often resistant to treatment.

Atypicals should be continued for:

·people with continuing BPSD

·where it is felt that severe adverse consequences may occur (or have occurred) if

they are discontinued, and

·where no alternative approaches are suitable.

4 The decision to continue these drugs should be documented and the factors considered in making this decision should

be recorded.4


1 Atypical antipsychotic drugs and stroke: Message from Professor Gordon Duff, chair, Committee on Safety of Medicines (CEM/CMO/2004/1).



2 De Deyn P et al. A randomised trial of risperidone, placebo and haloperidol for behavioural symptoms of dementia. Neurology 1999; 53:946-955

3 Herrman N et al. Atypical antipsychotics and risk of cerebrovascular accidents.

Am J Psychiatry 2004; 161:1113-1115

4 Faculty of the Psychiatry of Old Age. Atypical antipsychotics and BPSD. Prescribing update for old age psychiatrists.


5 Ballard C et al. A three- month randomised placebo-controlled neuroleptic discontinuation study in 100 people with dementia: The Neuropsychiatric Inventory median cut-off is a predictor of clinical outcome. J Clin Psychiatry 2004; 65:114-119

Two GP dilemmas

Nursing home wants more night sedation

Question: A patient with marked dementia in a nursing home is often aggressive and has a tendency to hallucinate or misinterpret visual images at night. He has a slightly stiff Parkinsonian gait and has fallen once or twice and staff find him very difficult to manage. The GP suspects the home is not run optimally and has made suggestions but feels they can do little more. The nursing staff are now requesting more night sedation.

Answer: This history is highly suggestive of dementia with Lewy bodies which would be an important diagnosis to make in this situation because such patients may have severe sensitivity reactions to neuroleptics.

This is one of the rare situations in which short-term treatment with a benzodiazepine would be sensible while you await result of specialist (old age psychiatry) referral on guidance for best management.

90-year-old is wandering in middle of night

Question: A patient of 90 lives alone but has her 70-year-old sons (who have health problems of their own) living nearby. She seems very vague whenever the GP sees her with little recall of recent events and forgets medication ­ she even forgets to look in the dosette box the GP provided. However, she scored a good MMSE score when visited by a CPN for the elderly from the mental health team and so was discharged back to the GP's care. Twice the police have been called when the patient has been found wandering in the nearby busy street late at night, confused, in night clothes asking for buses to a place where she lived 20 years ago. The patient refuses to consider moving.

Answer: Once again I'd have to recommend getting back to the old age psychiatry team. Very often patients with mild dementia can present very convincingly and score well above the cut-off of 24/30 on the MMSE that some people use to define the presence of dementia. But, late at night or at other times when orientating cues and carers are not around, the true extent of the underlying impairment is unmasked.

The old age psychiatry team would see her difficulties in a different light if they knew about the wandering.

The patient might do well on a cholinesterase inhibitor ­ certainly adding a sedating medication is only likely to make things worse ­ but the current draft of NICE guidance might mean that this treatment option was not available.

Robert Howard is professor of old age psychiatry and psychopathology at the Institute of Psychiatry in London

Rate this article 

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