Pinpointing the Cause of Non-Alzheimer Dementia

November 1, 2006

Many physicians, including psychiatrists, may shy away from seeing elderly patients with symptoms of dementia because they imagine that there are a large number of alternative diagnoses and that differential diagnosis is complicated. In fact, however, the number of possible diagnoses in most situations is relatively small and the diagnosis of dementia in older patients is certainly feasible in primary care psychiatry.

Many physicians, including psychiatrists, may shy away from seeing elderly patients with symptoms of dementia because they imagine that there are a large number of alternative diagnoses and that differential diagnosis is complicated. In fact, however, the number of possible diagnoses in most situations is relatively small and the diagnosis of dementia in older patients is certainly feasible in primary care psychiatry.

In order to diagnose specific causes of dementia, the physician must first be certain that dementia is the syndrome that best describes the patient's symptoms. Although depression, delirium, psychosis, aphasia, and mild cognitive impairment share some features with dementia, each is a distinct syndrome with its own differential diagnosis. It is important to consider all of these syndromes when evaluating a patient with suspected dementia.

Alzheimer disease (AD) is the most common dementia; it constitutes 60% to 80% of all dementias. AD begins insidiously and typically evolves over years. Its core cognitive symptom is short-term memory loss, but other abnormalities of judgment, abstract reasoning, spatial orientation, word finding, and personality are common.

The clinical diagnosis of AD is based on the presence of a cognitive disorder in which short-term memory loss is predominant. As of 2006, there are no routinely used biomarkers for AD. Cerebrospinal fluid markers, such as reduced levels of amyloid beta peptide or increased levels of tau peptide, have modest accuracy, but because of the lack of an evidence base in autopsy-confirmed cases, the markers are of uncertain clinical value. The use of Pittsburgh compound-B (PIB) in positron emission tomography holds considerable promise as a means of establishing whether a patient has the amyloid beta peptide pathology of AD.1 Validation of PIB imaging in dementia is currently being investigated in Alzheimer centers worldwide. How ever, it is unlikely that PIB imaging will be available for general clinical use for a few years.

Vascular dementia and Lewy body disease are the next most common forms of dementia. They are about equally prevalent, each accounting for roughly 15% of all diagnosed dementias (Figure). Beyond AD, vascular dementia, and Lewy body disease, the other dementias are much less common. These and other dementias are de scribed in more detail below.

Vascular dementia
The effects of cerebrovascular disease on cerebral function cause vascular dementia.2 The incidence and prevalence of vascular dementia mirror AD in that this condition becomes increasingly common with advanced age. A number of cerebrovascular mechanisms can lead to a cerebral injury, including large-vessel infarctions, multiple lacunar infarctions, extensive subcortical and periventricular white matter disease, and microvascular changes. These tissue injuries are usually due to atherosclerotic disease or amyloid angiopathy. Autoimmune mechanisms are a far less likely cause of vascular dementia.

The full range of the clinical ex pression of vascular dementia is not completely understood. The most well-known syndrome is that in which cognitive impairment occurs within 3 months of a clinically recognized stroke and there is evidence of infarcts in cognitively relevant cerebral regions. Usually, there are neurologic signs or symptoms that are consistent with a cerebrovascular cause, such as hemiparesis or hemianopia.

 

Sometimes, a patient who presents with a cognitive disorder that clearly follows stroke may have a nondiagnostic imaging study. A more challenging clinical situation arises in patients who have no clear-cut clinical history of stroke but who have imaging studies showing multiple bilateral large- or small-vessel infarcts. In each of these instances, there is almost certainly a pathologically relevant cerebrovascular process.

There is currently no specific cog nitive profile of vascular dementia. However, if there were a typical clinically recognized syndrome of vascular dementia, it would be that of a person with profound executive and attentional deficits and slightly less impaired short-term memory. Patients with vascular dementia are often apathetic and lack initiative, although these features are not specific to the syndrome.

There is increasing awareness of the etiologic overlap of AD and vascular dementia. Both share risk factors such as hypertension, diabetes, and hyperhomocysteinemia. The neuropathologic overlap of the 2 syndromes is also con siderable. Many patients with otherwise typical Alzheimer pathology also have cerebrovascular disease. Similarly, many patients with significant amounts of cerebrovascular pathology also have some Alzheimer pathology.

Lewy body disease
Lewy body disease is a common pathologic condition in late-life dementia. There are 3 subtypes of Lewy body disease, including typical Parkinson disease and 2 forms with dementia. Parkinson disease is typically characterized by rest tremor, bradykinesia, gait and postural disturbances, masked facies, and rigidity. The dementing disorder in which Parkinson disease precedes cognitive impairment by more than a year is referred to as Parkinson disease dementia. The third form is defined by the presence of cognitive impairment preceding or occurring simultaneously with parkinsonism. This latter condition is sometimes called dementia with Lewy bodies.3

Dementia with Lewy bodies has been recognized clinically only in the past 10 to 15 years. Patients with dementia with Lewy bodies usually have mild Parkinson disease or a history of unexplained falls and postural disturbances. They often have a disorder of arousal characterized by excessive daytime sleeping, disrupted sleep at night in the form of dream enactment behavior (rapid eye movement sleep behavior disorder), prominent visual hallucinations, and marked fluctuations in cognition and level of alertness. There is an increased rate of depression as well. Patients with dementia with Lewy bodies frequently seek psychiatric consultations because of the combination of all these neuropsychiatric symptoms. Prominent visual hallucinations may be an important symptom that distinguishes dementia with Lewy bodies from AD.

Frontotemporal lobar degenerations
Frontotemporal lobar degenerations (FTLDs) constitute about 5% of diagnosed cases of dementia. FTLD--in contrast to AD, vascular dementia, and Lewy body disease--has a peak age of incidence between 50 and 70. There are 2 principal syndromic variants of FTLD.4 One is a disorder with prominent personality changes and aberrant behaviors. This disorder is referred to as frontotemporal dementia. In addition to changes in social conduct and interpersonal relationships, patients with frontotemporal dementia have cognitive disorders in the domain of executive function; that is, they show poor concentration, poor mental agility, marked mental inflexibility, and poor reasoning and judgment. In a typical situation with the FTLDs, social con duct or language disorders predominate over anterograde amnesia ("short-term memory loss") and enable FTLDs to be distinguished from AD. There often are characteristic imaging findings as well. Many patients who have FTLDs demonstrate focal frontal or anterior temporal lobe atrophy.

The other syndromic variant of FTLD is progressive aphasia. Several subtypes have been identified, and the clinical syndromes range from prominent expressive-language deficits (progressive nonfluent aphasia) to the disorder known as semantic dementia, in which speech is preserved, but loss of access to vocabulary occurs. Both behavioral and aphasic forms of FTLD can be associated with amyotrophic lateral sclerosis.

The neuropathology of FTLD has undergone a very rapid transformation since the late 1990s. One subtype of FTLD is associated with accumulation of the microtubule-associated protein tau. Disorders of tau include Pick disease, progressive supranuclear palsy, and corticobasal degeneration. Some FTLD patients with tau pathology have mutations in the tau gene. Pick disease was the only recognized subtype of FTLD until about 20 years ago. Since then, the spectrum of FTLD has broadened. A second major group of FTLD disorders lack abnormalities of tau protein. Patients with these types of FTLD disorder usually have accumulations of another ubiquitinated protein that has not yet been identified. Baker and colleagues5 reported multiple families with FTLD and progranulin mutations. Progranulin mutations are already recognized as the most common genetic cause of FTLD, accounting for about 10% of all FTLD cases.

Normal pressure hydrocephalus
The diagnosis of normal pressure hydrocephalus (NPH) has bedeviled neu rologists for nearly 4 decades. Once thought to be common, NPH is now considered by most students of the epidemiology of dementia to be an exceedingly rare cause of dementia.6 When radiologists see large ventricles, they often diagnose hydrocephalus. However, radiographic hydrocephalus is not equivalent to a diagnosis of NPH-ventricular enlargement is most commonly due to AD, not NPH.

The most reliable marker of NPH is the characteristic "magnetic gait" in which patients have great difficulty in lifting their feet off the ground while upright. This gait disorder looks like a very severe shuffling gait. (As a consequence, NPH enters into diagnostic consideration when dementia and gait disorder occur together. Usually, Lewy body disease or another neurodegenerative disorder is the cause rather than NPH.)

There is no consensus on the best way to identify patients who are likely to benefit from placement of a ventriculoperitoneal catheter. Lumbar puncture with removal of a large volume of cerebrospinal fluid (roughly 30 mL) together with prelumbar and postlumbar puncture observations of gait is perhaps the most reliable and practical method. Clear-cut improvement with high-volume cerebrospinal fluid removal is essential for neurosurgical referral.

Creutzfeldt-Jakob disease
Creutzfeldt-Jakob disease (CJD) is a very rare dementing illness: the incidence is about 1 case per million per year. Patients with CJD present with changes in personality or depressive symptoms simultaneously with cognitive changes and a variety of motor changes, such as cerebellar ataxia or parkinsonism. The clinical presentation of CJD is that of a dementing illness that evolves over weeks or months. Cerebrospinal fluid markers and MRI scans are very helpful in confirming the diagnosis. Electroencephalography is also helpful but not as sensitive as the changes on diffusion imaging in magnetic resonance.

Other disorders that can cause dementia
In the course of evaluating patients with dementia using findings from thorough histories and physical examinations, other conditions might suggest themselves. For example, if a patient with symptoms of dementia complained of a new or different headache or had neurologic signs such as hemiparesis, hemianopia, or cranial nerve abnormalities, some consideration should be given to space-occupying brain lesions (eg, brain tumors or subdural hema tomas) that could be diagnosed from results of a brain imaging study. On rare occasions, brain tumors or hema tomas could produce the gradual onset of a syndrome of cognitive impairment. However, both of these are likely to produce other, more typical features. Moreover, patients with either of these conditions are more likely to present with symptoms that evolve over a matter of weeks rather than years.

Chronic meningitides should also be considered in patients who appear to have dementia who also complain of headache or who have "focal" neurologic signs. Certain fungal infections or tuberculosis could also produce a syndrome of cognitive impairment together with headache or focal neurologic signs.

A review of medications is essential when evaluating patients with dementia. While most drug intoxications fall into the syndromic domain of delirium, and, in fact, are short-duration events, it is easy to imagine how symptoms caused by chronic marginal overdosing with digoxin, anticonvulsants, corticosteroids, narcotics, sedatives, antidepressants, or antipsychotics could evolve over many months and virtually mimic those of dementia.

The American Academy of Neurology has recommended a modest battery of blood tests for dementia diagnosis. However, none of the specific tests-complete blood cell count; electrolyte, calcium, vitamin B12, and thyroid hormone levels; and tests of liver and renal function-are necessarily linked to a specific disease causing dementia.7 Hepatic and renal failure are associated with cognitive impairment, but impairment for both is more likely to be in the syndromic domain of delirium. The same can be said for thyroid functions and vitamin B12 level. Many clinicians may recognize the alliterative names of myxedema madness and megaloblastic madness, but these are not dementias. In the vast majority of instances, something like vitamin B12 deficiency would cause a cognitive disorder of subacute onset with delusional thinking, altered attention, and confusion (ie, delirium). A paraneoplastic syndrome and a primary autoimmune disorder of the brain are both rare, but in the setting of a history of cancer or a subacute onset, these disorders should be screened for with a panel of autoimmune antibody tests. Otherwise, there are really no metabolic disorders that cause true dementia in the elderly.

Other neurologic diseases must be considered, but once again it is highly likely that there will be some characteristic clue in the history or physical examination findings. For example, Huntington disease is an autosomal disorder occurring in middle-aged persons that presents with a very distinctive movement disorder with chorea, athetosis, dystonia, dysarthria, and in coordination. Multiple sclerosis, on the other hand, is a sporadic disease with a wide range of age at onset. Multiple sclerosis is almost always associated with neurologic signs and symptoms out side the cognitive and behavioral domains.

The differential diagnosis of dementia can be effectively carried out in primary psychiatric practice; a careful history and examination to identify cognitive and behavioral issues should be coupled with sensitivity to other medical, psychiatric, and neurologic conditions or disorders. A valid mental status examination is critical. Referral to a qualified neuropsychologist may be needed if findings from the bedside mental status examination are equivocal or otherwise not definitive. Psychiatrists should be comfortable with the key elements of the neurologic examination that are necessary to diagnose Parkinson disease or focal neurologic signs in patients with cerebrovascular disease.

Dr Knopman is professor in the department of neurology and Mayo Alzheimer Research Center at the Mayo Clinic College of Medicine in Rochester, Minn. Dr Knopman was a one-time consultant to General Electric Healthcare and GlaxoSmithKline. He previously served on an advisory board for Myriad Pharmaceuticals and on a data safety monitoring board for Neurochem Pharmaceuticals and Sanofi-Aventis Pharma ceuticals, and he is an investigator in a clinical trial for Elan Pharmaceuticals.

References:

References1. Klunk WE, Engler H, Nordberg A, et al. Imaging brain amyloid in Alzheimer's disease with Pittsburgh Compound-B. Ann Neurol. 2004;55:306-319.
2. Knopman DS. Dementia and cerebrovascular disease. Mayo Clin Proc. 2006;81:223-230.
3. McKeith IG, Dickson DW, Lowe J, et al. Diagnosis and management of dementia with Lewy bodies: third report of the DLB Consortium. Neurology. 2005;65:1863-1872.
4. Kertesz A, McMonagle P, Blair M, et al. The evolution and pathology of frontotemporal dementia. Brain. 2005;128:1996-2005.
5. Baker M, Mackenzie IR, Pickering-Brown SM, et al. Mutations in progranulin cause tau-negative fronto temporal dementia linked to chromosome 17. Nature. 2006;442:916-919.
6. Vanneste J, Augustijn P, Dirven C, et al. Shunting normal-pressure hydrocephalus: do the benefits outweigh the risks? A multicenter study and literature review. Neurology. 1992;42:54-59.
7. Knopman DS, DeKosky ST, Cummings JL, et al. Practice parameter: diagnosis of dementia (an evidence-based review). Report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology. 2001;56:1143-1153.