Three syndromes complicate the diagnosis of dementia owing to similar symptoms of the presenting conditions:

Depression, delirium, and drug reactions are potentially reversible conditions, therefore it is important to investigate the differential diagnosis.

Subjective memory complaints, problems in word finding and naming and reduced concentration are common in elderly people, but may indicate depression (see Depression and dementia below). Pseudodementia secondary to depression is a common differential diagnosis and neuropsychological tests often fail to separate dementia and depression [Hofman, 2000]. To complicate this even further, depression is a relatively frequent complication of dementia.

Loss of hearing and visual impairments, which are common with increasing age, may affect cognitive performance and should be considered in the differential diagnosis.

Depression in the elderly may only be detected in about 25% of cases [Cole, 1996]. There are a number of reasons to explain why depression is missed in this segment of the population (see Table 7).

Despite the low rates of detection of depression in the elderly, differentiating depression and dementia poses additional problems for the clinician because depression in those over 40 years of age can cause some degree of cognitive change [Emery, 1992]. In some situations it may be unclear whether a patient has a clinical depression, dementia or both. Prevalence rates of depression complicating dementia reported worldwide vary from 0% to 86% [Ballard, 1996; Allen, 1995] and this extreme variability is due to difficulties in diagnosing depression in dementia (see Table 5).

The differentiating features of depression and dementia are shown in Table 8.

In mild dementia, depression presents in a similar way to that observed in the nondementing elderly patient [Draper, 1999]. Although both depressive and cognitive symptoms are present, the extent to which they are caused by dementia or depression is unclear due to the overlap in the symptoms of these conditions.

Depression with significant cognitive impairment in the elderly is termed depressive pseudodementia. The clinical features of this particular syndrome are shown in Table 8. The treatment regimen for depressive pseudodementia is similar to any other major depression and maintenance antidepressant therapy is strongly recommended; however, the course of the mood disorder has high rates of relapse [Sachdev, 1990; Stoudemire, 1993] and there are variable outcomes. Although the cognitive impairment may diminish with pharmacotherapy, it may not fully resolve.

Sometimes the residual impairment is age-related and benign, whereas at other times it may be due to comorbid cerebral disease, such as stroke or Parkinson’s disease.

A neuropsychological evaluation often helps with diagnosis once depression has been resolved. This is because reduced symptoms of depression may alter the test scores of the neuropsychological examination providing a more accurate account of dementia.

Although the outcome of elderly patients presenting with depression and significant cognitive impairment is variable, studies indicate that approximately 25–50% of patients aged 60 years and over, with depressive pseudodementia, develop clinical dementia within three to five years [Sachdev, 1990; Copeland, 1992; Alexopoulos 1993].

The presence of depression appearing first in later life has been suggested to be a prodrome of later onset dementia, accounting for up to 10% of dementia cases [Visser, 2000; Linka, 2000]. Depression has been shown to often precede the dementia syndrome, particularly in vascular dementia, dementia with Lewy bodies and other subcortical dementias [Ballard, 1996].

Diagnosing depression in dementia is difficult (see Tables 8 and 9) but the clinician should be sensitive to the fact that depression may develop at any stage of dementia, though perhaps less often in late stages. The most frequently reported symptoms of depression in established dementia include dysphoria, loss of interest and psychomotor changes [Ballard, 1996]; and thoughts of death and suicidal ideation are usually part of a depressive syndrome warranting treatment [Draper, 1998].

Sometimes patients with prominent medial frontal lobe dementia present symptoms of apathy, aspontaneity, lack of initiative and general slowness. Carers or partners often regard the patient as being mildly depressed or uncooperative. Clinical examination may reveal that the depressed affect is shallow and unsustained; observation and interview in a structured environment will highlight the patient’s capacity or inability to respond to his/her surroundings. If there is further doubt, a trial of antidepressants and/or a neuropsychological profile may be required.

Family and professional carers of patients should be educated about changes in mood and behaviour that are associated with depression. For example, a relatively sudden decline in function, over three to four weeks, should raise concern. However, a sudden change may also be due to delirium [Draper, 1999] (see Delirium and dementia) in addition to other conditions besides depression.

In moderate to severe dementia, clinicians should note any persistent incongruent symptoms that have a predominantly affective quality, such as hopelessness, helplessness, worthlessness, sadness, panic, anxiety, mood-congruent delusions, suicidality and emotional lability. Behavioural (agitation, aggression, vocal disruption) and motivational disturbances (loss of energy, loss of interests, refusal to eat) are often the presenting symptoms, but there are usually some accompanying depressive symptoms [Draper, 1999]. Antidepressant therapy may be warranted if there is any doubt in the presentation of symptoms.

Diagnostic aids can improve diagnosis of depression in dementia. For example, the Cornell Scale for Depression in Dementia [Alexopoulos, 1988] utilizes best available information from the patient interview and carer report – scores of ten or more suggest depression. However, the use of self-report scales, such as the Geriatric Depression Scale [Yesavage, 1983], without collateral information may be problematic when there is moderate to severe cognitive impairment. Misdiagnosis can occur in both directions – transient states of dysphoria can be labelled as a depressive syndrome due to inaccurate longitudinal reporting or a persistent depressive syndrome may be minimized by the patient’s poor recall of its effects [Draper, 1999].

The management of depression in dementia is variable, owing to the variety of clinical syndromes and subsyndromes of depression. Partners and carers should be informed of specific prognosis, including the degree of improvement that is likely to occur with treatment, if it is known. For example, the worst prognosis is for depression associated with vascular dementia [Ballard, 1996].

An algorithm for the management of depression and cognitive impairment is shown in Figure 7; an algorithm for the management of depression in previously diagnosed dementia is shown in Figure 8.

There are a number of both nonpharmacological and pharmacological treatments for the management of depression and cognitive impairment. Table 10 highlights the options available to the clinician. Elements of psychological treatments may be carried out, but trained mental health professionals are required to carry out the full treatment programmes. For some psychological therapies, only anecdotal reports are available regarding their effectiveness in the treatment of depression in dementia.

Antidepressant therapy is indicated in those who meet criteria for major depression, those with atypical depression or with prominent symptoms of anxiety or panic, and those whose mood disorder persists despite adequate trials of nonpharmacological therapies.

The choice of the antidepressant should be based on the effectiveness and tolerability for an elderly patient, however there are advantages and disadvantages for certain antidepressants. For example, limited evidence suggests that selective serotonin reuptake inhibitors (SSRIs), such as sertraline, citalopram and fluvoxamine, are effective in reducing depressive symptoms and improving associated behavioural disturbances including food refusal, agitation, irritability and restlessness in patients with dementia [Volicer, 1994; Nyth, 1992; Nyth, 1990; Taragano, 1997].

Sertraline and citalopram may be the SSRIs of choice in the elderly patient with dementia and depression, as they have few adverse effects on liver metabolism [Mitchell, 1997], but the elderly patient may not tolerate SSRIs as well as younger adults. Furthermore, the advantages of SSRIs may only be useful in mild to moderate depression, as it is uncertain whether they are effective in severe and melancholic depression, whereas tricyclic antidepressants (TCAs) or electroconvulsive therapy (ECT) may be more beneficial [Danish University Antidepressant Group, 1990; Roose, 1994]. This type of depression also often requires hospitalization due to psychosis, self-neglect, suicidality or agitation. TCAs may be effective but have anticholinergic side effects that include confusion, prostatism, orthostatic hypotension, sedation and dry mouth [Taragano, 1997; Reifler, 1989]. For a number of years, the TCA of choice has been nortriptyline due to better tolerability; while it has the least anticholinergic side effects of all TCAs, these still occur. See Table 10 for other antidepressants that should be considered in the treatment of depression in dementia.

Other evidence [Roth, 1996] supports the use of moclobemide – a reversible inhibitor of monoamine oxidase – as it is effective in dementia; whereas, venlafaxine – a serotonin and noradrenaline reuptake inhibitor – may be more effective than the SSRI paroxetine in severe depression [Poirier, 1999].

Delirium is characterized by acute widespread disturbance of higher cerebral function in response to a variety of systemic and neurological insults. The possible causes of delirium are shown in Table 11.

Delirium is a common clinical condition but the wide range of possible causes and the coexistence of underlying systemic illnesses can easily obscure its contribution to the patient’s presentation. Although delirium is not restricted to elderly or demented patients, its incidence increases with advancing age in the presence of cognitive impairment.

Delirium is diagnosed by a set of criteria defined by the American Psychiatric Association (APA) [DSM-IV, 1994]. In addition to these criteria (see Table 12) a disturbed sleep-wake cycle together with agitation or depression may be apparent.

Two broad clinical subtypes of delirium are identified in the elderly patient, sometimes referred to as ‘loud’ and ‘quiet’ delirium, respectively [O’Keeffe 1999].

It is hypothesized that these variants of delirium are caused by disruption of the neurotransmitter systems [Itil, 1966]. Some patients may exhibit features of both subtypes leading to further confusion about the nature and aetiology of the illness.

The most characteristic and consistent abnormality in delirium is disturbance in attention, which results in a state of extreme distractibility and inability to perform tasks that require any degree of concentration (such as reciting the days of the week backwards or repeating strings of digits), in addition to temporal and spatial disorientation. Also, the same brain areas responsible for general arousal and for maintaining coherent thought processes may be affected in delirium. Clouding of conciousness, a cardinal feature of delirium, may be missed if the patient is having a lucid interval. Therefore, other aspects of cognition may be affected resulting in rambling and incoherent speech, delusional ideas, misidentification of people and surroundings and even frank hallucinations.

Other typical features of delirium include drowsiness during the day and increased restlessness and confusion at night, and may give rise to reverse sleep–wake cycle, nocturnal wandering and other troublesome behaviours.

Patients presenting with delirium are unable to provide a reliable clinical history, therefore details of any illnesses are best obtained from family members to elucidate any change in the patient’s behaviour, the nature and severity of the change and any fluctuations in mental state. Structured interviews such as the Confusion Assessment Method [Inouye, 1990], the Confusion Rating Scale [Williams, 1988] and the Delirium Rating Scale [Trzepacz, 1988] have been developed to establish the extent of confusion associated with behavioural and mental change.

A thorough drug history should be obtained in all cases of delirium, as a wide variety of drugs (see Table 11), particularly those with anticholinergic activity, are capable of inducing an acute and easily reversible confusional state. People with dementia are susceptible to delirium from even a minor stressor, such as severe constipation or influenza.

The clinical examination of the patient presenting with symptoms of delirium should include general behavioural observations, such as appearance, alertness and mood and specific tests of attentional function. These may include the:

In the presence of severe cognitive impairment formal evaluation may not be feasible, but in these patients signs of rambling attention, distractibility and agitation alternating with drowsiness are usually obvious.

Initial clinical investigations for patients presenting with delirium are shown in Table 13. If diagnosis remains uncertain and there is no evidence of drug precipitation or infection, especially urinary tract infection, then a CT scan should be performed to exclude subdural haematoma or focal infection. If the patient is pyrexic and no evidence of urine, respiratory or blood infection can be found, then cerebrospinal fluid should be examined for evidence of meningitis or encephalitis. Lumbar puncture should never be undertaken in an acutely confused patient without a prior CT scan.

Delirium may be confused with other conditions in patients with abnormalities in cognitive function and mood, arousal, behaviour, thought and spoken language content. These conditions include progressive dementing illnesses, such as Alzheimer’s disease [McKhann, 1984], syndromes of fluent receptive dysphasia and psychotic illnesses. Table 14 shows the conditions related to dementia that complicate differential diagnosis in delirium.

In dementia, the cognitive performance is variable but the sensorium is clear with no dramatic fluctuation in the level of awareness. In delirium, the characteristic feature is fluctuation in the level of awareness due to disruption in attentional processing.

With regard to language, patients with dementia typically show fluent but empty spontaneous speech and marked anomia (the inability to name objects or pictures); whereas patients with delirium produce a muddled series of sounds, only some of which may be recognized as words.

In visuospatial function, using simple line drawings, patients with dementia produce systematically distorted results, whereas chaotic scribbling may be observed in patients with delirium.

Furthermore, in memory and learning tasks patients in the early stages of Alzheimer-type dementia can usually acquire information but forget it rapidly [Gauthier, 1997], whereas an inconsistent pattern of immediate repetition is typical of patients suffering from delirium.

The diagnosis of a cause for the delirium, for example, etabolic derangement, and its correction, may help with the differential diagnosis. The EEG can be helpful; while slowing of background rhythm is common to both delirium and dementia, low voltage fast activity is typical of delirium tremens and spikes/polyspikes frontocentrally may indicate hypnosedative drug withdrawal and tricyclic or phenothiazine intoxication [Trezpacz & Wise, 1997].

Dementia with Lewy bodies has a cognitive profile which is complex but typically presents with deficits in overall speed of processing, attention and arousal, together with prominent visuospatial dysfunction. The clinical features observed may include fluctuations in the level of attention and alertness, visual hallucinations and systematized delusions, together with features of mild parkinsonism and exaggerated intolerance of neuroleptics. These features, with the exception of parkinsonism – which may not be clinically detectable in the early stages – have much in common with delirium.

Dementia with Lewy bodies is estimated to account for as many as 20% of cases of dementia in the elderly [Galton, 1999], hence it is potentially the second most common cause of dementia in this age group. It is likely that an appreciable number of elderly patients presenting with delirium, for which no underlying aetiology is found will therefore follow a progressive course rather than recovering.

Although both dementia and delirium may occur in the same patient concurrently, both are common in elderly patients and may reasonably be considered to be unconnected events. Evidence for infection, metabolic or biochemical disturbances and other causes of delirium should be investigated at presentation of symptoms just as enthusiastically in the patient with established dementia as the patient with acute cognitive deterioration who has previously been well.

A practical problem facing clinicians is the onset of altered behaviour (agitation and withdrawal) and hallucinations in a patient with established dementia. This behavioural change – indicating delirium – could be a result of the natural history of the dementia or it may occur from an unrelated illness. Although this problem is not readily solved, recognition of the existence of dementia subtypes (such as DLB) that are associated with periods of acute confusion, together with a systematic search for evidence of acute underlying illnesses and drug precipitants, will provide the necessary framework for reaching an appropriate clinical decision.

A number of options are available for treating delirium. Both nonpharmacological and pharmacological therapies are shown in Table 15. As delirium is a secondary phenomenon – a symptom of systemic or cerebral functional derangement – reversal can be achieved, in many cases by treatment of the underlying illness.

Specific psychoactive drugs, especially those with a large anticholinergic load, such as tricyclic antidepressants and traditional antipsychotics, should be avoided in delirium as they can make the clinical assessment more complicated. However,different classes of drugs may be required to control difficult or potentially dangerous behavioural disturbances and distressing symptoms such as hallucinations, paranoid delusions and anxiety.

The high-potency neuroleptic, haloperidol, is considered the first-line treatment for delirium and is licensed for intravenous use by the US Food and Drug Administration, however, it is given rarely by this route in Europe. The APA guidelines [APA, 1999] for the treatment of delirium suggest a starting dose of 1–2 mg every two to four hours, as needed, but for elderly patients the recommended dosage is as low as 0.25–0.5 mg every four hours, as needed. Higher doses may be required for severely agitated patients.

Further guidance on the use of haloperidol as a first-line treatment for delirium in patients with dementia comes from the Expert Consensus Guidelines on the management of agitation in dementia [Alexopoulos 1998]. Atypical neuroleptics are recommended as second-line treatments while low-potency neuroleptics, such as chlorpromazine, are third-line choices.

For patients with diagnosed DLB, administration of any neuroleptic should be avoided [McKeith, 1996] as they may cause neuroleptic sensitivity characterized by extreme drowsiness, stiffness and a dramatic deterioration in cognitive state.

Creutzfeldt-Jakob disease (CJD) is a rapidly progressive, neurodegenerative disorder causing dramatic neuromuscular symptoms, profound dementia and death. CJD may be difficult to distinguish from AD when the symptoms progress at an atypically slow pace. A new variant of CJD (vCJD) has been linked to bovine spongiform encephalopathy (‘mad cow’ disease) and has heightened awareness of prion protein disorders or transmissible spongiform encephalopathies (TSEs).

These diseases take part of their name from a striking and common neuropathological feature – spongiform (‘sponge-like’) degeneration of the brain. Symptoms may take months or years to appear in the recipient host, otherwise, TSEs either develop spontaneously or, rarely, arise through genetic mutation and therefore occur as familial disorders. Regardless of their origin, all TSEs progress over a period of months, inevitably leading to the death of the affected individual.

Recent neuropsychological profiling of patients with vCJD [Kapur, 2003] suggests a combined cortical and subcortical dementia, with impaired performance being particularly prominent on tests of memory, executive function, speed of attention and visuoperceptual reasoning. Cognitive impairment may represent one of the earliest features of vCJD and it is possible that, at least in some cases, neuropsychological deficits precede the onset of psychiatric or neurological symptoms.

Creutzfeldt-Jakob disease affects multiple brain areas, which causes multifocal deficits that involve movement, cognition and psychiatric status.

Thorough neurologic, cognitive and psychiatric examinations are necessary for observing its clinical features. Recent advances in neuroimaging techniques have allowed researchers and clinicians to discover imaging patterns that distinguish CJD from other neurologic diseases [Martindale, 2003]. Diagnostic criteria for CJD have been proposed and are listed in Table 16.

Alcohol consumption has been associated with complex changes in cerebral vasculature and structure in elderly adults. Patients with severe chronic alcoholism show a high prevalence of emotional distress, such as anxiety and depression, and neuropsychological impairments, such as executive deficits.

Cognitive impairment is frequently observed in patients with alcohol misuse or alcohol addiction. Multiple cognitive functions are reduced in patients with alcoholism – frontal lobe functions, such as planning, abstract thinking, set shifting or continuous performance are most frequently affected. Alcohol amnestic syndrome, alcohol dementia and the Wernicke–Korsakow Syndrome constitute distinct entities in alcohol related cognitive disorders [Marksteiner 2002].

Currently, there are no established treatment options for alcohol-induced cognitive impairment. Alcohol abstinence is the most important step, but psychosocial interventions are essential to support the patients in their daily activities.

Interestingly, although alcohol abuse is associated with increased prevalence of cognitive dysfunction among the elderly, daily alcohol consumption of less than 20 g for women and 50 g or less for men might be associated with a decreased probability of cognitive impairment [Zuccala, 2001]. This suggests that light to moderate alcohol drinking might protect against dementia and Alzheimer’s disease among old people [Huang, 2002; Truelson, 2002].

PD begins insidiously and usually affects one side of the body before spreading to involve the other side. The dysfunction of the striato frontal circuits that occurs in PD results in cognitive and behavioural problems, as well as motor impairment. Clinical symptoms include depression, lack of motivation, passivity and dementia. Cognitive deficits also occur, which progress with advancing illness, sometimes to subcortical dementia [Girotti, 2003].

As the disease progresses, even motor symptoms become intractable to therapy and as no proven means of slowing progression have yet been found, the condition has a poor prognosis.

Pathological examination shows loss of neuromelanin-containing monoamine neurons, particularly dopamine neurons in the substantia nigra pars compacta. A pathologic hallmark is the presence of cytoplasmic eosinophilic inclusions or Lewy bodies in monoamine neurons. The content deficits of the dopamine neurons in the nigrostriatal neurons accounts for many of the motor symptoms, which can be ameliorated by dopamine neuron replacement therapy or levodopa therapy.

Hypertension is shown to be a risk factor for cognitive impairment and dementia, particularly vascular dementia, in addition to cerebrovascular morbidity and mortality.

Evidence suggests that the use of antihypertensives in the elderly can protect against vascular dementia and reduce the incidence of dementia in elderly patients with isolated systolic hypertension. The Vascular Dementia Project of the European multicentre Syst–Eur trial [Seux, 1999] studied the influence of antihypertensive therapy on the incidence of dementia in elderly patients with isolated systolic hypertension. The results of this study indicated that antihypertensive treatment initiated with nitrendipine (dihydropyridine calcium antagonist) can reduce the incidence of dementia in patients over the age of 60 years with isolated systolic hypertension.

In a placebo controlled, double-blind, randomized trial [Frishman, 2002] involving 2418 patients aged 60 years or older with isolated systolic hypertension, active treatment based on nitrendipine with the addition of enalapril, hydrochlorothiazide or both, significantly reduced not only stroke and cardiovascular complications but also the incidence of vascular dementia and Alzheimer’s disease.

Trials of antihypertensive treatments are ongoing to confirm this important finding in protecting against dementia.