What is Dementia?
Dementia is an organic brain syndrome which results in global cognitive impairments. Dementia can occur as a result of a variety of neurological diseases. Some of the more well known dementing diseases include Alzheimer’s disease (AD), multi-infarct dementia (MID), and Huntington’s disease (HD). Throughout this essay the emphasis will be placed on AD (also known as dementia of the Alzheimer’s type, and primary degenerative dementia), because statistically it is the most significant dementing disease occurring in over 50% of demented patients (see epidemiology).
The clinical picture in dementia is very similar to delirium, except for the course. Delirium is an acute transitory disorder. By contrast Dementia is a long term progressive disorder (with the exception of the reversible dementias). The course of AD can range anywhere from 1.5 to 15 years with an average of about 8.1 years (Terry , 1988). AD is usually divided into three stages mild, moderate, and severe. Throughout these stages a specific sequence of cognitive deterioration is observed (Lezak, 1993). The mild stage begins with memory, attention, speed dependent activities, and abstract reasoning dysfunction. Also mild language impairments begin to surface. In the moderate stage, language deficits such as aphasia and apraxia become prominent. Dysfluency, paraphasias, and bizzare word combinations are common midstage speech defects. In the severe stage the patient is gradually reduced to a vegetative state. Speech becomes nonfluent, repetitive, and largely non-communicative. Auditory comprehension is exceedingly limited, with many patients displaying partial or complete mutism. Late in the course of the disease many neuropsychological functions can no longer be measured. Also primitive reflexes such as grasp and suck emerge. Death usually results from a disease such as pneumonia which overwhelms the limited vegetative functions of the patient.
Dementia is commonly differentiated along two dimensions: age and cortical level. The first dimension, age, distinguishes between senile and presenile dementia. Senile dementia is used to describe patients who become demented after the age of 65, whereas presenile dementia applies to patients who become demented prior to that age. Late onset AD (LOAD) also known as senile dementia Alzheimer’s type (SDAT) is the predominant cause of senile dementia. Early onset AD (EOAD) is the most frequent cause of presenile dementia, but HD, Pick’s disease and Creutzfeldt-Jakob disease though not as frequent are also important causes in presenile dementia.
The second dimension, cortical level, differentiates between cortical and subcortical dementia. Cortical dementia is used to describe dementia which results from brain lesions at the cortical level, whereas subcortical dementia describes dementia resulting from subcortical brain lesions. AD and Pick’s disease are the best known examples of cortical dementia; whereas HD, Parkinson’s disease (PD), and progressive supranuclear palsy (PSP) are good examples of subcortical dementia (Mayke, 1994). Dementia with both cortical and subcortical features is also possible, in that case the term mixed dementia is used. MID is a common example of mixed dementia.
Historical developments in dementia
The use of the term dementia dates back to Roman times. The Latin word demens did not originally have the specific connotation that it does today. It meant ‘being out of one’s mind’ and, as such, was a general term for insanity (Pitt, 1987). It was the encylopedist Celsus who first used the word dementia in his De re medicina, published around AD 30. A century later the Cappadocian physician Aretaeus first described senile dementia with the word dotage (i.e., “The dotage which is the calamity of old age…dotage commencing with old age never intermits, but accompanies the patient until death.”). Curiously, dementia was mentioned in most systems of psychiatric classification throughout pre-modern times, though the precise meaning of the word is often unclear (Pitt, 1987).
It can be argued that the origins of the scientific study of dementia date back to the early nineteenth century. The initial steps were undertaken by the great French psychiatrist Pinel at the beginning of that century. Pinel’s observations led him to the conclusion that the term dementia should be applied in relation to the “progressive mental changes seen in some idiots” (Pitt,3). Furthermore, Pinel thought that dementia was a distinct abnormal entity, and thus he used the term dementia to designate one of the five classes of mental derangement. However, by applying the term dementia to ‘idiots’, Pinel failed to differentiate between dementia and mental subnormality. This was accomplished by Pinel’s student Esquirol in his 1838 textbook Mental maladies-A treatise on insanity. Esquirol summed up the difference between the demented and the mentally handicapped in the following epigram: “The dement is a man deprived of the possessions he once enjoyed, he is a rich man who has become poor. But the defective has been penniless and wretched all his life” (Mahendra, 10). Furthermore, Esquirol was also instrumental in the popularization of the term senile dementia. Remarkably, his description of senile dementia is very similar to our present day definition. Interestingly, in 1845 Griesinger proposed that senile dementia was due to a disease of the cerebral arteries, a faulty view which persisted until Alzheimer’s time.
Much of today’s basic knowledge about dementia was accumulated throughout the second half of the nineteenth century, and the first decade of the twentieth century. 1872 saw Huntington present a paper called “On chorea”, in which he discussed a typical case of what is now known as Huntington’s disease. Twenty years later in 1892 two significant events occurred. First Pick in a paper called “On the relation between aphasia and senile brain atrophy” described the case of August H. a 71 year old patient with senile dementia. Although the case is not typical of our present day conception of the disease Pick was given credit for discovering a new disease. The other more significant event in 1892 was Blocq and Mariensco’s description of scattered silver staining plaques in the cortex of senile patients. These plaques were subsequently named senile plaques (SP) by Simchowitz in 1911.
The year 1894 saw Alzheimer’s first major contribution , a differentiation between senile and vascular (arteriosclerotic) dementia. Alzheimer described the specific changes observed in arteriosclerotic atrophy of the brain, which resemble what we might call vascular dementia. In 1898 another milestone occurred when Binswanger introduced the term presenile dementia. Thus by the twentieth century significant changes were taking place in our understanding of dementia. The nineteenth century view that there was only one mental disease-insanity-and that dementia was its terminal stage was dispelled by Kraepelin in the 6th edition of his textbook Psychiatrie, published in 1899 (Pitt, 4). Kraepelin separated dementia praecox (a concept he proposed in 1898 in relation to Schizophrenia) from the other dementias (paralytica and organic), and Senile dementia was included under another category called involution psychosis (Pitt, 4).
In 1907 Alzheimer published his landmark case “A unique illness involving the cerebral cortex” in which he described a fifty-five year old demented woman. The case was very unusual for two reasons its clinical course, and the discovery of a striking microscopic lesion in the woman’s brain (Beach, 1987). The clinical course was unusual because of the young age of the patient and the rapidity of degeneration (the patient died within four and a half years of symptom onset). At autopsy neuropathological findings were even more unusual. One quarter to one third of cerebral cortical neurons had disappeared, and many of the remaining neurons contained thick, coiled masses of fibers within their cytoplasm (Beach, 1987). Alzheimer speculated that a chemical change had occurred in the neurofibrils. Thus Alzheimer described for the first time neurofibrillary tangles (NFT), which togther with SP are considered to be the neuropathologocal halmarks of AD (See Appendix 1 for Alzheimer’s original drawing of NFT). Alzheimer concluded that he discovered a unique entity separate from senile dementia as it was known at that time. However, it was not until 1910 when Kraepelin discussed the condition in the 8th edition of his textbook Psychiatrie that AD gained official recognition.
The second decade of the twentieth century witnessed the end of the golden period in dementia research (this only lasted until the 1960’s when a renaissance occurred). U’Ren cites two reasons as the principal causes (Pitt, 6). First the rise of Freud’s Psychodynamic theory caused American psychiatry to swerve in the direction of psychological explanations. Second Kraepelin’s descriptions and classifications seemed to leave little room for therapeutic efforts or optimism.
Notwithstanding, several key contributions have been made in the ‘Dark Ages’ of dementia research. In 1920 Creutzfeldt, and in 1921 Jakob, described cases of dementia with pyramidal and extrapyramidal signs. Although it is now thought that only Jakob’s case was typical of the disease the Creutzfeldt-Jakob disease (CJD) was given to the world. The year 1936 saw an important change with regards to the diagnosis of AD. Before 1936 it was common practice to provide a diagnosis based on both clinical and pathological characteristics. However, when it became clear that many non-demented people had some senile plaques and neurofibrillary tangles, Jervis and Soltz advised that only clinical criteria would suffice for a diagnosis of AD (Mahendra, 14). In 1948 Jervis published his landmark paper called “Early senile dementia in Mongoloid idiocy.” Jervis described three individuals with Down’s syndrome (DS), aged 37, 42 and 47 years, each of whom had shown a profound emotional and intellectual deterioration in the last few years of life. At autopsy, all were found to have SP and two also displayed NFT (Beach, 39). This was the first demonstration of NFT in DS and the first full clinical and pathological correlation supporting an Alzheimer-like syndrome in DS (Beach, 39).
Research in dementia began to revive in the early sixties. New causes of the dementia syndrome have been recognized including, depression, which in the form of psuedodementia may mimic dementia (Kiloh, 1961), progressive supranuclear palsy (Steele et al, 1964) and normal pressure hydrocephalus (Adams et al, 1965) , (cited in Pitt, 6). Prior to the 1960’s dementia was still viewed as a chronic, irreversible and untreatable condition (Mahendra, 14). Accordingly, in the 1960s, several writers in Europe called for a revision of the concept and emphasized that irreversibility should not be viewed as an essential feature of dementia. Another important change that took place in the 1960’s concerned epidemiology. Prior to the sixties arteriosclerosis was thought to be the predominant cause of dementia, whereas AD was thought to be rare (Pitt, 12). However, arteriosclerosis was decisively challenged as the prime cause of dementia by several reports between 1960 and 1970 (i.e.,Tomlinson, Blessed, and Roth, 1968 and 1970). These reports demonstrated that arteriosclerosis was greatly overestimated as a cause of dementia, and that the majority of patients dying with dementia in fact showed the characteristic plaques and tangles of AD. Furthermore, Katzman, in 1976 argued that because of similarity in the clinical picture and the identical nature of the histopatholgy, distinctions between AD and senile dementia were arbitrary and no longer useful (Pitt, 12). Thus when it was understood that AD and senile dementia are similar, it was clear that AD is a common illness.
In the mid-1970’s two important contributions were made. First, Butler in his 1975 book Why survive? Being old in America criticized the widespread notion that senility was a normal part of aging. Butler argued that, senility, was a result of brain disease or depression and was potentially treatable. The extension of this view was that senility was abnormal, and that its usual causes were diseases, not just aging (Pitt, 1987). Second, three different labs (Bowen et al, 1976; Davies & Maloney, 1976; and Perry et al, reported low levels of choline acetyltransferase, the marker enzyme for acetylcholine 1977) (ACh), in the brains of patients who died from AD. ACh deficiency has since been the target of most therapeutic efforts in AD (see treatment).
Throughout the 1980’s and 1990’s two trends emerged. First, with regards to diagnosis, criteria have been made stricter. Classification systems like the Diagnostic and Statistical Manual have evolved towards a more precise and comprehensive definition of dementia. Moreover, neuoroimaging techniques are becoming more and more standard, allowing in some cases for a more accurate diagnosis. Second, the past fifteen years have witnessed a substantial growth in genetically based research. For instance one of the genes involved in AD, the amyloid precursor protein (APP), has been localized to a specific segment of chromosome 21 (see risk factors).
Dementia is known as the quiet epidemic, but it affects a significant proportion of our population. In 1989 the Canadian consensus conference on the assessment of dementia reported that Canada had about 250,000 cases of dementia (which at the time comprised about 1% of the population), with 25,000 new cases occurring annually (Clarfield, 1989). Jorm et al. (1988) project that until the year 2025 Canada will experience a growth in the prevalence of dementia, more rapid than the rise if the number of elderly aged over 65. The majority of dementia cases are attributable to AD, vascular dementias, or a combination of these (Table 1). In the past there were hopes that up to 40% of dementias had reversible causes. However, recent reports (Clarfield, 1988; Barry and Moskowitz, 1988) suggest that the true incidence of reversible dementias is at the most 11% and is probably far lower, with drugs, metabolic causes and depression accounting for about two thirds of the cases (Clarfield, 1989).
Overall, there are no significant gender differences in prevalence and incidence rates for dementia as a whole. However, for AD, there is an increased prevalnce in females. Jorm et al. (1987) estimate a female to male AD prevalence ratio of 1.6. Ethnically there seem to be important differences in both prevalence and subtype of dementia. Prevalence wise, Heyman et al. (1991) found that out of a random sample of 4116 16% of African Americans had dementia compared to only 3.1% of Caucasians. The same study also found that mixed and MID were more likely to occur in African Americans (26% of dementias in African Americans compared to 14% in Caucasians). Moreover, in both Europe and North America most studies point to AD as the most common dementing illness; whereas in Asia (especially Japan) MID predominates (Morris, 1994). The observed high rate of stroke in Japan is consistent with a high MID rate. Possibly the higher level of stress in Japan leads to more strokes and therefore a higher incidence of MID.
Table 1. Etiology of Progressive Dementia and Approximate Incidence
senile dementia of the Alzheimer type 50%
Multi-infract dementia 10-15%
Mixed SDAT and MID 10-15%
Alcoholic-nutritional dementia 5-10%
Normal pressure hydrocephalus 5%
Life Expectancy and Mortality Estimates
The following summary is based on Terry’s (1988) review of the Wang (1978) and Barclay et al. (1985) studies. The Wang study examined senile dementia (mean age of onset 71.3 years) and presenile dementia (mean age of onset 53.8 years) survival rates during the 1960s. Senile dementia patients survived on the average 6.0 years, close to half of the expected survival rate (11.1 years) of similarly aged non demented people. Presenile demented patients survived slightly longer an average of 6.9 years, against an expected survival of 22.3 years. The Barclay et al. Studies examined survival rates in AD and MID patients in the 1980s. The mean survival rates for AD and multi-infarct dementia were 8.1 and 6.7 years respectively. Interestingly, the survival rate of demented women on the whole is significantly higher than that of men. Terry (1988) suggests that the lower survival rate of demented men is due to a higher incidence of MID in men.
Age is the biggest risk factor for developing dementia. According to a model proposed by Jorm et al. (1987) a doubling of the prevalence rate occurs every 5.1 years. (1987) For the elderly population aged 65 and above the prevalence of dementia is estimated at about 10%. Whereas in the very elderly it can reach up to 40% (Clarfield, 1989).
Genetic factors are important in some dementing diseases. In HD an autosomal dominant gene on chromosome 4 is directly responsible for the disease. The genetic evidence in AD is less conclusive. On the one hand there are studies (i.e., Breitner et al. , 1988) which have reported a cumulative risk of AD among relatives of patients approaching 50%, thus implying an autosomal dominant mode of transmission (Morris, 1994). But, on the other hand, genetically transmitted diseases should be concordant in monozygotic twins, this does not appear to be the case in AD. For instance both Creasey et al. (1989) and Kumar et al. (1991) have reported three pairs of monozygotic twins who were discordant. Whereas Nee et al. (1987) only found a 41% concordance rate for AD in 17 monozygotic twins.
Farrer et al. (1990) suggest that AD appears as an autosomal dominant in families in which the average age of onset among kindreds is under 58. Supporting evidence for this comes from studies which have linked EOAD with DS (Lezak, 1993). Individuals who are afflicted with Down syndrome and who survive to age 40 almost invariably develop Alzheimer like dementia. During the intermediate and terminal stages of DS the individual suffers from recent memory loss, apraxia, temporal disorientation, and mutism, all of which are also common in AD (Morris, 1994). Thus it is not surprising that four studies have found an increased risk for AD with late maternal age (Morris, 1994). The increased risk of AD to patients born to mothers over 40 is consistent with Down syndrome risk curve (Rocca et al. , 1991).
Both EOAD and DS have been localized to chromosome 21. However, chromosome 21 does not appear to be a very good genetic marker for EOAD (Green book, 104). Recent studies have shown that a defect in chromosome 14 is more likely to be associated with EOAD, but the specific gene(s) have not yet been isolated (Green book, 104). Evidence for genetic predisposition to LOAD has only emerged over the last two years. It is now known that a gene which codes for a lipoprotein called ApolipoproteinE (APOE) in chromosome 19 is involved (Green book, 101). APOE is linked to the type 4 allele (e4). It has now been proven that an increase risk for dementia is dependent on a strong chemical binding between the main ingredient of SP, the Beta amyloid protein, and the APOE-e4 (Green book, 102). Table 2 summarizes the genetic findings that have been made thus far in EOAD and in LOAD.
Table 2. Alzheimer’s Disease Genetics
Risk factor for developing AD
Not yet isolated
Not yet identified Marker
Other Risk Factors
Corsellis and Brierly (1959) [as cited by Graves & Kukull, 1994] have shown that dementia similar to that seen in AD may occur following a single head injury. In addition, dementia puglistica, (the so called ‘punch-drunk syndrome’) develops in some boxers. Lower education has also been associated with dementia. Animal studies demonstrate a positive relation between environmental stimulation and dendritic growth. It is also known that dendritic growth in humans continues throughout life. Possibly lower education is related to a lack of mental exercise, which could delay the onset of significant cognitive decline (Graves & Kukull, 1994).
Aluminum (Al) has been implicated as a possible neurotoxin, but the evidence is inconclusive (Carson & Butcher, 1992)). Proponents of the Al neurotoxin hypothesis argue that Al has been shown to accumulate in neurons with neurofibrillary degeneration, and that aluminosilicates accumulate in senile plaques. Critics argue that the abnormal accumulation of Al is an effect, not a cause, of brain degeneration. Another controversial risk factor is depression. Four studies have reported a statistically significant association between a history of depression and AD (Graves & Kukull, 1994). The controversy revolves around the idea that depression is possibly an early manifestation of AD.
There is some research suggesting that individuals with a weakened immune system may be more susceptible to develop AD. Heyman et al. (1984) [as cited by Graves & Kukull, 1994] have found an increased risk of AD associated with thyroid disease in women. However, their findings have not been replicated. Interestingly, there is some evidence to suggest that smoking can have a protective effect from AD. For instance, Duijn and Hofman (1991) [as cited by Graves & Kukull, 1994] have found a negative correlation between smoking and AD in a study involving 198 individuals.
For each dementing disease a specific neuropathological pattern is observed. However, due to the limited scope of this essay the discussion will be limited to the most important dementing disease, AD.
Several changes are observed at the gross neuropathological level in AD (Mirra & Gearing, 1994). Cortical atrophy is generally observed in the frontal, temporal, and parietal cortex. Sectioning of the brain reveals variable enlargement of the lateral and third ventricles. Disproportionate enlargement of the temporal horn of the lateral ventricle is commonly encountered, with concomitant atrophy of the entorhinal cortex, amygdala, and hippocampus.
At the microscopic level the two most distinguishing neuropathological features are senile plaques (SP) and Neurofibrillary tangles (NFT).
There are two types of SP, neuritic and diffuse, both plaques share antigenic determinants with the Beta amyloid 4 protein. Neuritic plaques can be distinguished by their abnormally thickened neurites ( i.e., axons or dendrites) arranged around a central core of amyloid (Mirra & Gearing, 1994). By contrast the diffuse plaques lack the thickened neurites and the amyloid core seen in the neuritic plaques (Mirra & Gearing, 1994). Plaques of both types are found in varying degrees in the neocortex, entorhinal cortex, hippocampus, and in the amygdala. SP also occur in the brains of healthy people. It is only when they exceed a certain critical number that AD emerges.
NFT are intraneuronal structures which occupy the cell body of the neuron. Usually NFT coexist with SP in the neocortex, but they may be absent there in up to 30% of AD patients (Mirra & Gearing, 1994). However, NFT are consistently found in the entorhinal cortex, hippocampus, amygdala, nucleus basalis of Meynert, and dorsal raphe nucleus (Mirra & Gearing, 1994). It is thought that the major antigenic component in NFT is the protein tau.
Neuronal loss is directly related to the degree of synaptic density, which has been found to be crucial in determining the severity of cognitive decline. It is greatest in the temporal lobes, but is also significant in the frontal and parietal lobes (Lezak, 1995). The strongest correlation with a global measure of dementia is the loss of functional synapses in the midfrontal and lower parietal areas which surround the temporal lobes (Lezak, 1995). The effect of this pattern of neuronal cortical loss is twofold. First it disconnects the temporal lobe structures from the rest of the cerebral cortex. This accounts for the prominence of memory impairments (Lezak, 1995). Second this pattern also disconnects the prefrontal structures from the parietal ones. This accounts for the compromised capacity for attentional tasks (Lezak, 1995). It is thought that besides the effects of SP and NFT, neuronal loss is chiefly related to the depletion of the neurotransmitter acetylcholine (see treatment).
In a study conducted by Brun and Gustafson (1978) [as cited by Cummings, 1988], the regional distribution of SP, NFT, and neuronal loss, was examined in AD patients. The results indicated that the most severely affected areas were the medial temporal and the temporo-parieto-occipital junction region (see figure 1). Two positron emission tomography (PET) studies by Benson et al. (1983) and Foster et al. (1984) [as cited by Cummings, 1988] have confirmed this pattern.