Prevalence of neuropsychiatric disorders

Table 1 illustrates the range and prevalence of neuropsychiatric sequelae observed after stroke. Very similar problems, by and large with similar prevalence rates, are observed after traumatic brain injury. However, traumatic brain injury is associated with higher rates of bipolar affective disorder, and particularly of rapid-cycling bipolar disorder with a periodicity of days rather than weeks.

Clinical presentations

Aetiology

There has been a long-standing debate as to whether there is an association between the location of the brain lesion (particularly anterior, left-hemisphere lesions) and post-stroke depression. A systematic review suggested that there is no evidence to support such a relationship (Reference Carson, MacHale and AllenCarson et al, 2000). However, the methodology of this review, particularly the exclusion of some studies, has been criticised. A subsequent meta-analysis re-ignited the debate and suggested that proximity of the lesion to the left frontal pole predicts depressive illness (Reference Narushima, Kosier and RobinsonNarushima et al, 2003). This was further supported by another appraisal from Finland (Reference Vataja, Leppavuori and PohjasvaaraVataja et al, 2004).

Most traumatic brain injury is caused by closed head injury and is often complicated by alcohol. The vulnerability to contusions of areas of the brain that are involved in social behaviour, cognition and regulation of mood (medial orbitofrontal and anterior temporal lobes) (Reference Tranel, Bechara, Damasio and GazzanigaTranel et al, 2000) partly explains why the neuropsychiatric consequences of traumatic brain injury usually outstrip the neurological sequelae as predictors of outcome. The best predictor of outcome is the duration of post-traumatic amnesia, the period from injury to the return of continuous day-to-day memory. People with post-traumatic amnesia that lasts longer than 1 month are unlikely to return to work.

Treatment

It is important to be vigilant in identifying post-stroke depression, and treatment of depressive symptoms should always be considered (Box 1). The fact that they are ‘understandable’ or are symptoms that may be a direct consequence of brain injury (e.g. apathy) should not prevent a trial of an antidepressant. It has been suggested, on the basis of limited evidence, that the D₂-receptor agonist bromocriptine might have therapeutic benefit in the management of apathy.

There is very limited evidence from systematic reviews regarding the efficacy of antidepressants in treating post-stroke depression (Reference Turner-Stokes and HassanTurner-Stokes & Hassan, 2002). Although there is the suggestion that tricyclic antidepressant drugs are more effective, selective serotonin reuptake inhibitors (SSRIs), having fewer side-effects, are a reasonable first choice (Box 2).

There is increasing interest in the early prescription of antidepressants after stroke as a preventive strategy for mood disorder. However, a review of treatment trials found no prophylactic effect of antidepressants compared with placebo (Reference Anderson, Hackett and HouseAnderson et al, 2004). It has been suggested that treatment of post-stroke depression in the first month when compared with the third month benefits functional outcome at up to 2 years (Reference Narushima and RobinsonNarushima & Robinson, 2003).

If all else fails, electroconvulsive therapy should be considered. This has been shown to be safe for the treatment of depression after both stroke and traumatic brain injury (Reference Currier, Murray and WelchCurrier et al, 1992).

Antidepressants are useful in treating emotionalism (lability of mood, emotional incontinence), regardless of the presence of depression.

Several psychotropic drugs have been used in the management of agitation and/or aggression in people with acquired brain injury, without clear evidence of their efficacy. A review (Reference Fleminger, Oliver and GreenwoodFleminger et al, 2003a ) reported that beta-blockers have the best evidence supporting their efficacy in treating these behavioural symptoms. None the less, in view of the large doses used in these studies, which were associated with significant side-effects, they should be prescribed with caution and, ideally, with the patient’s consent. Carbamazepine is commonly used as a first-choice medication in this setting, despite the lack of research evidence for its efficacy. Sodium valproate may be used as an alternative. Response to medication is usually seen early, within the first 6 weeks, and it is suggested that this is an adequate period during which clinical benefit should be expected before switching to an alternative treatment. There is no evidence that aggression is different from agitation in terms of its response to medication (Reference FlemingerFleminger, 2003).

Olanzapine or quetiapine are reasonable antipsychotics for psychosis associated with traumatic brain injury, because they have fewer extrapyramidal side-effects.

There is little evidence for psychological interventions for post-stroke depression. Of those used in case studies and series reviewed by Reference Kneebone and DunmoreKneebone & Dunmore (2000), cognitive–behavioural therapy (CBT) showed the most promise. However, a small randomised controlled trial reported subsequent to this found CBT to be ineffective (Reference Lincoln and FlannaghanLincoln & Flannaghan, 2003). Similarly, for other acquired brain injuries, there is only scarce evidence supporting such psychological strategies (Reference Khan-Bourne and BrownKhan-Bourne & Brown, 2003). Early educational interventions after traumatic brain injury may prevent later symptoms.

Prevalence of neuropsychiatric disorders

The neuropsychiatric presentations of this multi-focal demyelinating disorder are varied and include mood, anxiety, cognitive and psychotic disorders. It has been suggested by several studies that the prevalence of depression in multiple sclerosis is higher than in control groups with different neurological illnesses. However, methodological problems have been cited, particularly with regard to clinician masking and diagnostic criteria. A lifetime prevalence for depressive symptoms of 40–50% is generally accepted (Reference Siegert and AbernethySiegert & Abernethy, 2005). Psychotic illness in multiple sclerosis is most commonly observed in the context of treatment with steroids, where it is most often an affective psychosis, although schizophreniform psychoses are also seen.

Cognitive impairment in multiple sclerosis has been estimated to have a prevalence of up to 50% in community samples.

Clinical presentations

Some depressive symptoms are probably more common than others in multiple sclerosis. Of these, fatigue and cognitive impairment have received the most attention. Generally, older studies have suggested that neither of these is correlated with depression in the disorder. However, more recent studies have supported positive correlations, suggesting that mental fatigue (Reference Schreurs, de Ridder and BensingSchreurs et al, 2002) and impaired effortful information processing (Reference Arnett, Rao and GrafmanArnett et al, 1997) in particular are features of the depression. Suicide rates in people with multiple sclerosis have been shown to be twice the mean rate seen in the adjusted population (Reference Stenager, Stenager and Koch-HenricksenStenager et al, 1992) and are associated with living alone, alcohol misuse and depressive disorder.

Cognitive deficits arise early in the course of multiple sclerosis and often before the diagnosis has been made. The course of cognitive decline appears to be slow in the majority of cases, although risk factors for a more rapid decline include disease progression, age and worsening of physical disability. Particular impairment is seen in verbal fluency, comprehension, naming and executive dysfunction, as well as memory.

Aetiology

Beta-interferon has been implicated as a cause of depression, although the evidence for this is contentious. Neuroimaging studies of lesion location have been conducted in individuals with multiple sclerosis and depression or psychosis. The most robust findings in depression implicate more hyperintense lesions in left inferior medial frontal regions and greater atrophy of left anterior temporal regions (Reference Feinstein, Roy and LobaughFeinstein et al, 2004). It has been suggested that location of temporal lobe lesion may correlate with psychotic illness (Reference Feinstein, du Boulay and RonFeinstein et al, 1992).

Treatment

Desipramine has been shown to be more effective than placebo for depression in multiple sclerosis, although anticholinergic side-effects may preclude its use. The SSRIs are a good first-line choice although they can cause sexual dysfunction.

Limited evidence exists regarding the best choice of antipsychotic for treatment of psychosis in multiple sclerosis, although an atypical antipsychotic is a sensible choice.

People with multiple sclerosis treated in an open trial of donepezil showed a reduction in cognitive impairment (Reference Greene, Tariot and WishartGreene et al, 2000).

Individuals with fatigue may benefit from approaches used for chronic fatigue syndrome. Cognitive–behavioural therapy for depression is likely to be useful and it has been found to have similar outcomes to treatment with an SSRI (Reference Mohr, Boudewyn and GoodkinMohr et al, 2001).

Prevalence of neuropsychiatric disorders

Neuropsychiatric disorder has been estimated to occur in 6% of people with epilepsy, rising to 10–20% in those with temporal lobe and/or refractory epilepsy. Non-epileptic seizures, which come under the rubric of conversion disorders or dissociative states, are an important differential diagnosis of epilepsy. It has been estimated that between 9 and 50% of patients referred to specialist epilepsy centres have non-epileptic episodes (Reference Francis and BakerFrancis & Baker, 1999). The most common neuropsychiatric disorders associated with epilepsy and their prevalence are summarised in Table 2.

Clinical presentations

In all presentations in epilepsy, it is important to distinguish between symptoms that are associated with seizures and those that are independent of seizures. Perhaps most importantly, differentiation of the relationship of neuropsychiatric presentations to the seizure, i.e. whether they are pre-ictal, post-ictal or inter-ictal, is paramount in guiding treatment planning, as this may influence whether the clinician manipulates anti-epileptics, or adds psychiatric medications or does both (see below).

Aetiology

Combinations of biological and psychosocial factors are likely to play a role in the aetiology of neuropsychiatric disorders in epilepsy. In depressive disorders, these factors include anti-epileptic-induced effects such as folate deficiency and frontal lobe dysfunction. In psychoses, hippocampal damage and chronic inhibitory neurophysiological changes may be involved. The hypothesis of forced normalisation, where the abnormal EEG is normalised by anti-epileptic drugs, remains popular. This is based on the observation that the EEG recordings of patients normalised during psychotic episodes and that normalisation induced by anti-epileptics may be an aetiological factor for psychoses. Forced normalisation has been reported in both temporal lobe and generalised epilepsies. A paranoid psychosis has been its most frequent manifestation and it has been observed following the use of various medications, including phenytoin, carbamazepine, ethosuximide and levetiracetam. Forced normalisation may not be restricted to psychoses in epilepsy, but may also present as episodes of major depression.

The social stigma associated with epilepsy is a key psychosocial risk. A critical review of studies of the aetiology of non-epileptic seizures reported that a history of abuse or trauma was indentified in etween 15 and 40% more people who experienced seizures than in control groups (Reference Fiszman, Alves-Leon and NunesFiszman et al, 2004).

Treatment

It is important to identify the relationship of the psychiatric disturbance to ictal events, the role of anti-epileptic drugs in the aetiology of symptoms and the importance of psychosocial factors. For symptoms related directly to the seizure itself, optimal control of seizures is the treatment of choice.

When considering psychotropic medication for people with epilepsy there is always the concern that the medication will decrease the seizure threshold. However, a retrospective study has suggested that psychotropic medication does not increase mean seizure frequency (Reference Gross, Devinsky and WestbrookGross et al, 2000). It is important to start psychotropic medication at a low dose and increase it slowly. The SSRIs may be less likely to reduce the seizure threshold than tricyclic antidepressants. Because there may be an interaction between psychotropics and anticonvulsants (for example fluoxetine may increase carbamazepine levels), levels of anticonvulsant should be closely monitored.

It should also be borne in mind that anti-epileptic drugs, often also used routinely to stabilise mood, can be associated with adverse neuropsychiatric reactions. Valproate-induced hyperammonaemic encephalopathy, which presents with impaired consciousness and focal neurological signs, may well go unrecognised if somnolence, decreased motor activity and increased lethargy are misinterpreted as a therapeutic response in patients with mania. Additional symptoms may include subtle personality change, confusion, vomiting and hyperventilation. These clinical features of hyperammonaemia are very variable, can be episodic and are often difficult to detect early. Being aware of a history of similar symptoms and a family history is useful. Serum ammonia levels should be measured and expert advice sought (Reference Hawkes, Thomas and JurewiczHawkes et al, 2001).

Haloperidol and sulpiride may be less epileptogenic than other antipsychotics, and therefore the treatment of choice for epileptic psychoses. There is limited clinical experience and few systematic studies of atypical antipsychotics in epilepsy and they should be used with care. Clozapine is particularly epileptogenic and, as always, a balance of risks and benefits must be considered, discussed with the patient and carefully documented when commencing atypical antipsychotic treatment in epilepsy.

Recent work suggests that CBT may be effective in the treatment of non-epileptic seizures (Reference Goldstein, Deale and Mitchell-O'MalleyGoldstein et al, 2004).

Prevalence of neuropsychiatric disorders

The common neuropsychiatric presentations and their prevalence rates are outlined in Table 3.

Clinical presentations

Aetiology

It is probable that both biological and psychological factors explain depression in Parkinson’s disease:

1. • biological – loss of monoaminergic neurons; hypometabolism in caudate, inferior orbitofrontal and medial frontal regions on positron emission tomography

2. • psychological – chronic disabling illness, psychosocial stress.

As previously stated, psychoses in Parkinson’s disease are often related to recent increases in antiparkinsonian medications. History of psychiatric illness, personality disorder or a family history of addiction disorders may predispose patients to hedonistic homoeostatic dysregulation.

Treatment

The evidence base for treatment of depression in Parkinson’s disease has been examined in a literature review and found to be limited (Reference Ghazi-Noori, Chung and DeaneGhazi-Noori et al, 2004). Tricyclic antidepressants have been shown to be effective in depression in Parkinson’s disease and may also improve motor symptoms by virtue of their anticholinergic activity (Reference Rascol, Goetz and KollerRascol et al, 2002). However, their side-effects, including cognitive impairment, may limit their use. Although SSRIs generally have fewer side-effects, the evidence that they work is not as strong, and some have extrapyramidal effects that exacerbate motor symptoms.

In psychosis, consideration should be given to decreasing or stopping antiparkinsonian drugs. Antipsychotic medications must be used with caution because of their extrapyramidal side-effects. There is good evidence that low-dose (<100 mg/day) clozapine can improve psychosis without worsening parkinsonism (Parkinson Study Group, 1999). Its use in patients with ischaemic heart disease, particularly elderly people, must be monitored carefully, as it can cause agranulocytosis and tachycardia. There is more limited evidence for the use of other atypicals, cholinesterase inhibitors (Reference Bergman and LernerBergman & Lerner, 2002) and odansetron to treat psychotic symptoms.

There is no current evidence supporting the use of cholinesterase inhibitors in dementia in Parkinson’s disease, although a randomised placebo-controlled trial has shown benefit with rivastigmine in Lewy body dementia (Reference McKeith, Del Ser and SpanoMcKeith et al, 2000).

Psychological treatments for depression in Parkinson’s disease have not been evaluated fully.

Prevalence of neuropsychiatric disorders

Estimates of the prevalence of psychiatric symptoms at the first presentation of Huntington’s disease are in the region of 30%. By and large, the neuropsychiatric presentations and their prevalence mirror those seen in Parkinson’s disease. In 3–6% of cases a schizophreniform psychosis is the first presentation of Huntington’s disease.

Clinical presentation

There has been significant interest in cognitive abnormalities in gene carriers identified by predictive testing prior to the full manifestation of the disease. The results have, unfortunately, been contradictory. Dispute remains as to whether there are cognitive differences between carriers and non-carriers, as well as to whether cognitive symptoms precede motor symptoms or vice versa (Reference Witjes-Ané, Vegter-van der Vlis and van VugtWitjes-Ané et al, 2003). Since George Huntington’s original description of the disorder, it has been well recognised that there is an increased rate of suicide in this group (Reference Schoenfeld, Myers and CupplesSchoenfeld et al, 1984).

Aetiology

Attempts (albeit with flawed methodologies) have been made to link the psychiatric signs and symptoms of Huntington’s disease to the length of the CAG repeat sequence, but no relationship has been found (Reference Weigell-Weber, Scmid and SpiegelWeigell-Weber et al, 1996).

Treatment

There are no controlled trials establishing the benefit of antidepressants for depression in Huntington’s disease. Atypical antipsychotics are preferable for psychoses, so as to avoid exacerbation of motor problems.

Clinical presentation

Delirium, often called an acute confusional state, is characterised by a disturbance of conscious level. The patient is obtunded (drowsy) or highly distractable. Attention and concentration are impaired (e.g. as demonstrated by poor performance on a digit span test). Mental state may fluctuate. The individual is neither alert nor oriented, and is likely to be agitated and frightened. Psychotic symptoms with hallucinations, often visual, and fleeting delusions may be elicited. Delirium may also present as a hypoactive withdrawn state akin to stupor.

Aetiology

A diagnosis of delirium always suggests an underlying physical or biological aetiology, and numerous physical problems, including drugs and drug withdrawal, may be aetiological factors.

Agitation is often present in delirium. If the patient has been treated with antipsychotics an important differential diagnosis of the agitation is akathisia. Poor sleep, pain, constipation and systemic illness may play a role.

Treatment

Management consists of making the patient safe and then seeking the underlying biological cause as well as treating the condition symptomatically. Nursing should take place in a well-lit side room with consistent staff. The environment should be calm and allow undisturbed sleep (Reference Inouye, Bogardus and CharpentierInouye et al, 1999).

Some patients will settle with reassurance and explanation. Relatives may be able to help. Haloperidol and lorazepam may be used to produce rapid sedation to manage agitation and aggression. The patient should be placed on regular nursing observations, with monitoring of respiration and neurological state. If sedation is required for more than 1 or 2 days, consider atypical antipsychotics such as olanzapine or quetiapine, which are less likely to produce extrapyramidal side-effects. If antipsychotics are not tolerated or cause intolerable side-effects, valproate, carbamazepine and beta-blockers may be helpful alternatives. Avoid drug combinations that may increase agitation and aggression by increasing confusion.

Narcolepsy

REM sleep behaviour disorder