Clinical and Etiological Studies
on Dementia of Alzheimer Type
and Multiinfarct Dementia
AKADEMISK AVHANDLING
som med vederbörligt tillstånd av Rektorsämbetet vid Umeå universitet för avläggande av medicine doktorsexamen kommer att offentligen försvaras i hörsal C, Samhällsvetarhuset, fredagen den 29 april, 1983 kl 09.00
av
Gösta Bucht
ABSTRACT
CLINICAL AND ETIOLOGICAL STUDIES ON DEMENTIA OF ALZHEIMER TYPE AND MULTIINFARCT DEMENTIA,
by
G. Bucht, Department of Internal Medicine and Geriatric Medicine, University of Umeå, S-901 87 Umeå, Sweden
1. Clinical studies. Clinical diagnosis of dementia has been made largely on the basis of clinical findings supported by approp riate radiological and laboratory investigations. A minority of patients have treatable or reversible underlying causes for their dementing syndrome. It is important to distinguish be tween the two main forms of dementia Alzheimer's disease, se nile dementia of Alzheimer type (AD/SDAT) and MID so that advantage can be taken of any future progress in treatments. In the clinical study significant differences between several diagnostic procedures were found between patients with AD/SDAT and MID. Blood pressure was significantly lower in the AD/SDAT group and focal neurological signs were seen in 70% of the MID patients but only in 6% of patients with AD/SDAT. Electrocar diogram was normal in all patients with AD/SDAT but patholo gical in 75% of the MID patients. Electroencephalogram showed generalized slow frequencies in 79% of the AD/SDAT patients and localized changes in 65% of the MID patients. Computerized tomography showed a significantly greater dilation of the ven tricular system in MID patients compared to AD/SDAT patients and controls. Monoamine metabolites in the cerebrospinal fluid were lower in AD/SDAT patients and normal in MID patients. Psychopathological signs were found to be more variable and more pronounced in the AD/SDAT group compared with MID patients.
2. Etiological studies. Immunoglobulin and albumin were found changed in serum and CSF of both AD/SDAT and MID, indicating a more active immune response in MID and a less dense cerebrospi nal fluid barrier in both MID and AD/SDAT. There appears to be a consumption of IgG in the central nervous system in patients with AD/SDAT.
Abnormal chromosomes appearing as acentric fragments, i.e. without visible centromeres, were found in 90% of patients with AD/SDAT, 30% of patients with MID, and not at all in the con trol group. Increased aneuploidy was also seen both in patients with MID and AD/SDAT.
Diabetes mellitus in old age and AD/SDAT do not seem to ’ coex ist. Furthermore, patients with AD/SDAT have changed carbohyd rate metabolism with decreased fasting blood sugar concentra tions, increased glucose tolerance and higher concentration of insulin during an oral glucose tolerance test.
Key words: dementia of Alzheimer type - multiinfarct dementia - clinical studies - diagnosis - blood cerebrospinal fluid barrier function - cytogenetic changes - blood glucose - insulin secretion
New Series No 97 ISSN 0346 - 6612
From the Departments of Internal Medicine and Geriatric Medicine University of Umeå, Sweden
Clinical and Etiological Studies
on Dementia of Alzheimer Type
and Multiinfarct Dementia
by
Gösta Bucht
Umeå University Umeå 1983
2 CONTENTS ORIGINAL PAPERS.... INTRODUCTION... Clinical aspects Etiological aspects... MATERIALS AND METHODS... Clinical studies... Etiological studies... RESULTS AND DISCUSSION... Clinical studies.... ... Etiological studies... GENERAL SUMMARY AND CONCLUSIONS ACKNOWLEDGEMENTS... REFERENCES... 4 5 6 7 10 10 11 13 13 20 30 33 34
Rut
4
Original Papers
'The present thesis is based on the following papers:
I Bucht, G., Adolfsson,R. &.Winblad,B.
DEMENTIA OF ALZHEIMER TYPE AND MULTIINFARCT DEMENTIA - A CLINICAL DESCRIPTION AND DIAGNOSTIC PROBLEMS
Submitted to J. Am. Geriatr. Soc.
II Bucht, G. & Adolfsson, R.
THE COMPREHENSIVE PSYCHOPATHOLOGICAL RATING SCALE (CPRS) IN PATIENTS' WITH DEMENTIA OF ALZHEIMER TYPE AND MULTIINFARCT DEMENTIA
Submitted to Acta Psych. Scand.
III Alafuzoff, I., Adolfsson, R., Bucht, G., & Winblad, B.
ALBUMIN AND IMMUNOGLOBULIN IN PLASMA AND CEREBROSPINAL FLUID, AND BLOOD CEREBROSPINAL FLUID BARRIER FUNCTION IN PATIENTS WITH DEMENTIA OF ALZHEIMER TYPE AND MULTIINFARCT DEMENTIA J. Neurol. Sciences. Accepted for publication.
IV Nordenson, I., Beckman, G., Adolfsson, R., Bucht, G. & Win-blad B.
CYTOGENETIC CHANGES IN PATIENTS WITH SENILE DEMENTIA Age and Ageing. Accepted for publication.
V Bucht, G., Adolfsson, R., Lithner, F. & Winblad, B.
CHANGES OF BLOOD GLUCOSE AND INSULIN SECRETION IN PATIENTS WITH DEMENTIA OF ALZHEIMER TYPE
Acta Med. Scand. Accepted for publication.
These papers will be referred to in the text with the roman nume rals.
INTRODUCTION
The essential feature of dementia is a deterioration of mental functions caused by diffuse cerebral dysfunction. The causes of dementia are numerous, and comprehensive lists have been published by Slaby & Wyatt (1974) and Haase (1977).
To be considered demented the intellectual disabilities must be of a sufficient severity to interfere with social and occupational functioning. The deficit is multifaceted and involves memory, judgement, abstract thought, and a variety of high cortical func tions, as well as changes in personality and behaviour. The state of consciousness must not be clouded (DSM-III, Am. Psychiat. Ass. 1980). Neuropathologists have recognized two main forms«of organic dementias, Alzheimer's disease and senile dementia of Alzheimer type (AD/SDAT) and dementia due to cerebral softenings, probably secondary to cerebral arteriosclerosis (Alzheimer 1898, Fischer 1968, Tomlinson et al 1970, Jellinger 1976). Clinically, AD/SDAT is# defined as a dementia with an insidious.onset and an uniformly progressive v deteriorating course (Wells 1977, Liston 1979a, 1979b). The essential clinical features of* vascular dementi'a, "multiinfarct, dementia", are a stepwise deterioration in intellec tual functioning and a patchy distribution of mental deficits. Focal neurological signs and symptoms may also be present (Alzhei mer 1898, Rothschild 1941, Roth 1955, Ehrentheil 1957, Slater & Roth 1959, Hachinski et al 1974). In some instances both MID and AD/SDAT may co-exist with clinical features of both diseases, so called "mixed". Neuropathological studies of brains from demented patients show that AD/SDAT i.e. brains with neurofibrillary tang les and senile plaques account for approximately 50% of all demen tias and contribute to another 20%. Based on the amount of cereb ral softening present, MID probably accounts for approximately 17% and contributes to another group of about the same size (Tomlinson 1980). Thus, AD/SDAT and MID, alone or together, accounts for be tween 80 and 85% of all cases of dementia. The co-existence of AD/SDAT and MID, "mixed", is well established and occurs in app roximately 15% of the cases. Approximately 15% of the autopsy
6
cases have other specific causes of dementia or show no changes at all (Corsellis 1962, Tomlinson et al 1970, Jellinger 1976).
Clinical aspects
The clinical diagnosis of dementia and a judgement regarding the type of disorder causing the dementia have been made largely on the basis of clinical findings supported by appropriate radiologi cal and laboratory investigations (Hutton 1980, Roth 1981). In recent years patients with apparent dementing illnesses have ten ded to receive a closer medical scrutiny. This is partly due to a number of reports in the medical literature suggesting that a sig nificant minority of such patients have treatable or even rever sible underlying causes for their dementing syndrome, such as de pression, nutritional - metabolic - endocrine diseases, normal pressure' hydrocephalus and space occupying lesions within the skull (Marsden & Harrison 1972, Freemon 1976, Smith et al 1976, Katzman 1977, Victoratos et al 1977, Hutton 1982, Portera-Sanchez et al 1982). It is also important to distinguish between different forms of dementia so that advantage can be taken of any future progress in treatments. Recent research has been able to establish some of the etiological features of AD/SDAT; the deficiencies of neurotransmitters in the brain and cerebrospinal fluid being the most striking findings (Gottfries et al 1965, Pope et al 1966, Gottfries et al 1969, Davies & Maloney 1976, Perry et al 1977, Adolfsson et al 1979, Mann et al 1981). MID-patients might improve if underlying factors such as hypertension, arteriosclerosis of cervical arteries, and cardiac diseases are treated correctly (Hachinski et al 1974, Bousser 1977, Hutton 1980, Birkett & Raskin 1982).
The diagnostic accuracy of the clinical diagnosis AD/SDAT varies between different studies. Affective disorders are the most common forms of psychiatric disabilities misdiagnosed as dementia (Roth
1955, Ron et al 1979). Todorov et al (1975) found in a retrospec
tive study with histopathological verification of the diagnosis a significant coefficient of agreement between the clinical and
anatomical diagnosis but the different classes of dementia were only partially separable (50% overlap). Furthermore, there are clinical cases of apparent dementia where no distinctive patholo gical changes are found at biopsy or autopsy (Hughes et al 1973). In 1898 Alzheimer described the main diagnostic features of vascu lar dementia and it's pathological anatomical correlate. Other authors have described the clinical differences between vascular dementia and AD/SDAT (Rothschild 1942, Roth 1955, Ehrentheil 1957, Slater & Roth 1969). Hachinski et al (1975) converted the symptoms of vascular dementia into an "ischemic score", which has been pro ven to be useful in differentiating MID from AD/SDAT in a post mortem study (Rosen et al 1980). Several clinical studies have tried to differentiate MID and AD/SDAT by using a single or a few different investigative methods and most of them find group diffe rences, but no singlTe diagnostic tool has so far been proven .to able to separate MID and AD/SDAT in the single case (Rothschild 1942, Gordon & Sim 1967, Radue et al 1978, Roberts et al 1978, Harrison et al 1979, Hontela et al 1979, Striano et al 1981, Soi- ninen et al 1982). Few prospective studies, however, have been performed using a wide range of different investigative methods in an attempt to differentiate between MID and AD/SDAT.
Etiological aspects
The typical histopathological alterations in AD/SDAT are the seni le plaques and the neurofibrillary tangles (Alzheimer 1907). The senile plaques are composed of degenerative neuronal terminals, amyloid and reactive microglial cells, macrophages and astrocytes (Wisniewski & Terry 1973) and in the end stage an amyloid mass bordered by some glial cells (Terry & Wisniewski 1970). The neu rofibrillary tangles consist of twisted tubules, usually in a pair, therefore referred to as paired helical filaments (Kidd 1963, Wisniewski et al 1976, Wisniewski & Soifer 1979). The paired helical filaments are assembled in bundles in the perykarion crea ting the neurofibrillary tangle. The paired helical filaments are composed of a protein, similar to the -tubulin of normal neurotu buli. The presence of amyloid in the senile plaques and in the
8
vessel walls suggests that immunological factors may be involved in the development in AD/SDAT (Glenner 1978). Further, brain spe cific antibodies have been found to be increased in the serum of the aged and particularly in patients with AD/SDAT (Nandy 1978).
The development of histopathological changes in the brains of patients with Down's syndrome who have an extra chromosome 21, has led to a search for chromosomal abnormalities in AD/SDAT. Chromo somal aberrations have been reported (Nielsen et al 1968, Bergener & Jungklass 1970, Nordenson et al 1980), but others have not found these changes (Mark & Brun 1973, Brun et al 1978, Sulkava et al 1979). Ward et al (1979) reported aneuploidy in both sporadic and in familial cases of AD/SDAT, especially in the latter. There is also an increased risk of developing AD/SDAT among first order relatives of about 4 times than in the general population (Slater & Cowie 1971, Heston & Mastri 1977).
A possibility of a slow virus etiology for AD/SDAT, as in Jacob- Creutzfeldt's disease, has also been proposed. Suggestive of a viral etiology is the appearance of senile plaques and amyloid depositions similar to those in AD/SDAT in experimental scrapie virus infections (Wisniewski et al 1975). Transmission attempts have been made using material from cases with AD/SDAT on experi mental animals. In familial cases of AD/SDAT this resulted in a spongioform encephalopathy but without senile plaques and neuro fibrillary tangles (Gibbs & Gajdusek 1978).
Environmental agents have also been suspected to be of etiological importance, especially metals, such as aluminium. Experimentally,
aluminium produces neurofibrillary tangles (Klatzo et al 1965, Crapper & Dalton 1973), although of a different type than those seen in AD/SDAT. Increased amounts of aluminium have also been found in the cortex from patients with Alzheimer's disease (Crap per et al 1976).
No satisfactory theory for the causation of AD/SDAT has so far been presented. Etiological alternatives, viral, immunological and toxic environmental agents, may very well interact with hereditary factors to produce the disease (Crapper & deBoni 1978).
Disturbances in neurotransmitter concentration and function have repeatedly been reported the last ten years. Catecholamine distur bances with reduced concentrations of dopamine and noradrenaline and increased monoamine oxidase activity have been found in diffe rent parts of the brains from demented patients (Gottfries et al 1965, Adolfsson et al 1979, 1980, Mann et al 1980). The choliner gic system has been shown to have a specific relation to memory and cognitive function (Drachman 1977). Davies and Maloney (1976) found that cholineacetyl transferase (ChAT) is reduced in AD/SDAT, , while no change in postsynaptic receptors was observed (White ét al 1977) and no reduction of ChAT was found in normal aging (Bowen et al 1979). The reduction of ChAT activity in AD/SDAT also corre lates with the number of senile plaques and the degree of dementia (Perry et al 1978, Sims et al 1981).
The etiology of MID is thought to be the occurrence of multiple small or large cerebral infarcts (Fischer 1968, Worm-Petersen & Pakkenberg 1968, Hachinski et al 1974). Most authors are of the opinion that the occurrence of dementia due to cerebral infarction is well correlated with intra- and extracerebral arteriosclerosis (Worm-Petersen & Pakkenberg 1968, Bousser 1977, Birkett & Raskin 1982). Hypertension and heart disease are also believed to play an important part in the development of MID (Hachinski et al 1974). Cerebral softenings, however, are not infrequently present in sub jects in whome no evidence of clinical dementia have been found
(Tomlinson et al 1968, deReuck et al 1982). Tomlinson et al (1970) found that the ischemic destruction of cerebral tissue must be of a particular amount before evidence of dementia develops. They found that all patients with more than 100 ml of destroyed brain tissue were demented.
10
MATERIALS AND METHODS
Clinical studies (I, II)
Forty-one patients with mild.to moderate dementia, all living at home, were subjected to a clinical examination. The patients were divided, on anamnesis and presentation on admission, into two groups, one with a presumed diagnosis of AD/SDAT and the other with MID using.the diagnostic criteria stated in the DSM-III (Am. Psychiat. Ass. 1980). Two patients were excluded after the first examination due to concomitant diseases. The AD/SDAT group inclu ded 19 patients (9 men and 10 women; mean age 63 yrs, range 53-73 yrs; duration of illness 3.2±1.0 yrs).' The MID group included 20 patients, (11 men and 9 women; mean age 71 yrs, range 58-84 yrs; duration of illness 3.2±2.2 yrs). The selection of control persons was done through a letter to every person above 60 yrs of age in the community of Umeå requesting their participation in the study. After a telephone interview, 31 healthy elderly normals were selected (16 men and 15 women; mean age 67 yrs, range 60-72 yrs). The patients and the control persons were investigated while staying 2-7 days at the Department of Internal Medicine, Universi ty hospital, Umeå. A thorough clinical and psychiatric examination was performed including extensive laboratory tests (see I). Lumbal puncture was performed and monoamine metabolites were determined
in the ceretyospinal fluid (CSF). Radiology included a-chest X-ray and computed tomography of the brain (CT-scan). An electrocardio gram (ECG) and an electroencephalogram (EEG) were also performed. In study II, 18 of the patients with the diagnosis AD/SDAT and the
20 patients with the diagnosis MID were rated on the Comprehensive Psychopathological Rating Scale (CPRS) (Åsberg et al 1978). The
CPRS consists of a range of items describing psychiatric symptoms (reported by the patients)-and signs (observed items). A subscale consisting of items measuring common symptoms in patients with dementia was constructed.
Etiological studies (III, IV, V)
Albumin and immunoglobulin in plasma and cerëbrospinal fluid (III) Twenty patients witli AD/SDAT (9 men and 11 women; mean age 65 yrs, range 53-79 yrs) and 23 patients with MID (12 men and 11 women; mean age 71 yrs, range 55-85 yrs) were investigated. All patients were mildly to moderately demented. Sixteen healthy persons (12 men and 4 women; mean age 69 yrs, range 62-74 yrs) without signs of dementia or immunological, diseases served as controls. Blòod and CSF sampling were performed under standardized conditions (see III).
The concentrations of the major classes of'circulating immunoglo bulins (IgA, IgM, IgG) and albumin were determined by rocket immu- noelectrophoresis (Laurell 1972). Transudation of IgG through the blood-CSF barrier and local synthesis of IgG in the central ner vous system were calculated as described by Tibbling et al (1977) and Schuller & Sagar (1981). Blood-CSF barrier function was gra phically demonstrated by Reiber’s evaluation graph for protein profile in the central nervous system (Reiber 1980).
Cytogenetic changes in patients with senile dementia (IV)
Nineteen patients with a diagnosis AD/SDAT (10 men and 9 women; mean age 64 yrs, range 52-84 yrs), 17 patients with the diagnosis MID (9 men and 8 women; mean age 71 yrs, range 58-83 yrs) and ele ven institutionalized patients with Down’s syndrome (3 men and 8 women; mean age 30 yrs, range 18-65 yrs) were included. As cont rols, 20 individuals (12 men and 8 women; mean age 66 yrs, range 49-85 yrs) were chosen randomly from an ongoing health screenirfg program of the elderly. The patients with dementia were all mildly to moderately demented and all living at home at the time of the investigation. Blood sampling and the cytogenetical procedure were performed under standardized conditions (see. IV). The slides were stained in Giemsa (10%), G-banded (Wang & Fedoroff 1972)‘and C-banded (Sumner 1972). Screening for chromosome breakage was per formed on Giemsa-stained (non-banded slides).
Blood glucose and insulin secretion in patients with dementia (V) A. Retrospective study of records of patients with dementia
The prevalence of diabetes mellitus in different types of dementia was studied in a retrospective ßtudy of all records of patients with a dementia diagnosis hospitalized at Umedalen's hospital, Umeå, between 1971 and 1980. 839 records were studied and classi fied aß either AD/SDAT, MID or confusional states. The diagnostic criteria used were those stated in the DSM-III (Am. Psychiat. Ass. 1980). The diagnosis of overt diabetes mellitus was established on the grounds suggested by the National Diabetes Data Group (1979).
B. Metabolic study of the different patient groups
Five different groups of patients and controls were investigated with a fasting blood sugar and an oral glucose tolerance test (OGTT). Methods for blood glucose determination and calculation of the areas under the OGTT curves were performed as described by Lithner (1974). The first group consisted of 45 patients with the diagnosis AD/SDAT (22 men and 23 women; mean age 70 yrs, range 52-85 yrs), the second group consisted of 66 patients with the diagnosis cerebrovascular disease (CVD) without dementia (34 men and 32 women; mean age 70 yrs, range 43-87 yrs), the third group consisted of 15 patients with MID (5 men and 10 women; mean age 75 yrs, range 61-83 yrs), the fourth group consisted of 30 unselected nondiabetic hospitalized patients (15 men and 15 women; mean age 72 yrs, range 58-84 yrs) and the fifth group of 31 healthy elderly non-hospitalized persons (16 men and 15 women; mean age 68 yrs, range 61-73 yrs). The first four groups were all on the same diet and had approximately the same amount of activity and were all hospitalized at the Department of Internal Medicine, Umeå Univer sity during the investigation. The fifth group was not hospitali
sed, their diet habits were not controlled and they were also phy sically more active than the patients in the first four groups. The mean weight did not differ between the different patient groups.
AD/SDAT (3 men and 7 women; mean age 71 yrs, range 61-80 yrs) and in 10 healthy elderly persons (5 men and 5 women; mean age 70 yrs, range 67-72 yrs). Insulin levels in serum were determined with -a radioimmune assay method (Jalow & Berson 1969).
Malabsorption was also investigated through an A-vitamin load and a xylose-absorption test in patients with AD/SDAT.
RESULTS AND DISCUSSION
Clinical studies (I, II)
The differential diagnosis between AD/SDAT and MID is a common clinical problem (Birkett 1972, Hughes et al 1973, Todorov et al 1975 and Ron et al 1979). Several large retrospective but few prospective studies on the differential diagnosis and the outcome of treatment of dementia have been performed. The combined studies suggest a treatable cause for the dementing illness in approxi mately 15% of the patients (Marsden & Harrison 1972, Freemon 1976, Smith et al 1976, Katzman 1977, Victoratos et al 1977, Hutton 1982, and Portera-Sanchez et al 1982). In our study we selected mildly to moderately demented patients with either AD/SDAT or MID, based on the course of the illness and presentation on admission (Table 1).
Table 1 Background data on the three studied groups
AD/SDAT MID Controls
n 19 20 31 Male/female 9/10 11/9 16/15 Age (yrs) 63.0±6.0***° 71.4±7.8 67±3.5* Height (crn) 159±11 167±12 168±8 Weight (kg) 67±12 72±10 71±11 Duration of illn./yrs 3.2±1.0 3.2±2.2
*** p <0.001, * p <0.05, MID compared with AD/SDAT and controls ° p <0.05, controls compared with AD/SDAT
The anamnestre data showed striking differences between the two investigated groups with dementia. Sixty-three % of the AD/SDAT patients were free of other diseases, while 65% of the MID patients had cardiovascular diseases. The occurrence of depression earlier in their lives was also more common among MID patients (30%) than in AD/SDAT patients (5%).
The physical examination of patients with dementia is important in establishing the cause of the dementia (Rothschild 1941, Roth 1955, Ehrentheil 1957, Wells 1977). Our investigation revealed significantly lower systolic blood pressure in the AD/SDAT patients compared with both MID and controls and the diastolic blood pressure was significantly lower in the AD/SDAT patients compared with the MID patients. This difference has also been established by other authors (Ladurner et al 1981, Portera-Sanchez et al 1982). For the differentiation of MID from AD/SDAT Hachinski et al (1975) constructed an "ischemic score", which, in a post mortem study has been shown to be useful in distinguishing these two disorders (Rosen et al 1980). Our two groups of demented patients differed significantly in this respect, the AD/SDAT group having a significantly lower score than the MID group (Table 2).
Table 2 Ph^ical data on the three studied groups
AD/SDAT MID Controls
Blood pressure (mm Hg) Systolic 133±16*** 158±15 152±21 Diastolic 77±19** 93 ±10 87±11 Focal neurological^signs N/total* ** 1/19 14/20°o° 1/31 Ischemic score* ,0.7 ±1.1 9.9±2.6 -3 2 Heart (rei.Voi. cm /m ) 423±36 488±135 456±89 * After Hachinski et al (1975)
*** p<0.00ï, AD/SDAT compared with MID and controls ** p^O.Ol, AD/SDAT compared with MID
That the neurological examination in most cases differentiates MID from AD/SDAT has consistently been reported (Harrison et al 1979, Ladurner et al 1981, Sluss et al 1982).
Routine laboratory examinations showed that all laboratory data were within the normal range . in all investigated groups of patients. The primary aim of routine laboratory tests is to exclu de potentially remediable causes to the dementia (Marsden & Harri son 1972, Freemon 1976, Wells 1977, Gershon & Herman 1982).
ECG with signs of coronary heart disease or ^cardiac hypertrophy was exclusively found among the demented patients with MID (Table 3). Other authors have also stated that the occurrence of coronary heart disease or previous myocardial infarction are indicative of MID (Rothschild 1942,. Hontela & Schwartz 1979).
Table 3 ECG findings
AD/SDAT MID Controls
n (%) n (%) n (%)
Total 19 20 30
Normal 19 (100) 5 (25) 22 (74)
Arrythmias 0 3 (15) 4 (13)
Cardiosclerosis/hypertrophy 0 12 (60) 4 (13)
An EEG, with localized slow wave activity and assymetric findings, is more common in MID, patients, while patients with AD/SDAT show more homogenous generalized slow wave activities over both hemis pheres (Gordon & Sim 1967, Roberts et al 1978, Striano et al 1981, Soininen et al 1982b). In our study generalized slow frequencies were found in 79% of AD/SDAT patients and localized slow frequen cies arid abnormalities in 65% of the MID patients (Table 4). There is, however, an overlap between the EEG findings in different
16
forms of dementia, which diminishes the diagnostic value in indi vidual patients.
Table 4 EEG findings
AD/SDAT n {%) MID n (%) Total 19 20 Normal 2 (11) 4 (20)
Generalized slow frequencies 15 (79) 3 (15)
Localized slow frequencies 1 (5) 2 (10)
Generalized and local slow frequencies 1 (5) 8 (40)
Local abnormalities 0 3 (15)
When visible infarcts or an irregular and dilated ventricular sys tem is seen on the CT-scan this technique is of diagnostic value in identifying MID (Radue et al 1978, Roberts et al 1978). Our findings is that the MID patients have a significantly greater dilatation of the ventricular system than both controls and AD/SDAT. The cortical atrophy did not differ significantly between the three groups and does not seem to be of diagnostic help, at least, not in patients with mild to moderate dementia as in our study (Table 5).
Table 5 CT-scan of the brain
AD/SDAT MID Controls
Total (n) 17 20 28
Cortical atrophy (0-3) 1.2±0.9 0.910.8 0.710.7 Ventricular dilatation (0-3) 1.110.9 2.210.7*** 0.410.6
Visible infarcts (n) 1 4 0
Contrary to our findings, Soininen et al (1982) found a more pro nounced asymmetrical ventricular dilatation in a group of AD/SDAT patients. Wells (1977) rightly stated that the CT-scan alone must not be relied upon when trying to differentiate between different causes of dementia.
A routine lumbal puncture and analysis of blood cells, proteins and glucose is of no value in differentiating MID from AD/SDAT, but is important in that it aids discovery of other causes of dementia. Monoamine metabolites in the CSF have been extensively investigated in AD/SDAT. Lower concentrations of homovanillic acid (HVA) and 5-hydroxyindoleacetic acid (5-HIAA) have been found (Gottfries et al 19jo9, Mölsä 1980). In our investigation AD/SDAT patients had significantly lower concentrations of HVA than cont rols and a lower HVA than the MID patients. 5-HIAA was lower ip AD/SDAT patients compared with MID and controls, although' not significantly. MID patients did not differ from controls (Table 6). The overlap between the groups was however great, which limits this test as a diagnostic tool.
Table 6 CSF-monoaminemetabolites AD/SDAT 10 M±SD MID 15 M±SD Controls 22 M±SD HVA (nmol/1) 159±98*+ 227±93 242±101 5-HIAA (nmol/1) 98±62 132±41 125±42 HMPG (nmol/1) 57±16 59 ±14 61±11
* p<0.05, AD/SDAT compared with controls + p<0.1, AD/SDAT compared with MID
Rating of the demented patients on the CPRS showed significant differences in the psychopathology between the two groups (AD/SDAT and MID). The AD/SDAT group has a more variable psycho
18
pathology than the patients with MID. In table 7 the proportion of patients in each group (AD/SDAT and MID) with a score above 1 are listed. The psychopathological signs that differed most betwen the groups were "long-term memory", "specific speech defects", "diso rientation", "perplexity" and "distractability". We could, howe ver, not confirm that mood disorders and emotional lability symp toms, 'which most authors consider to be a frequent accompanying symptom in patients with MID, were specifically related to the MID group (Rothschild, 1941, Roth 1955, Ehrentheil 1957, Birkett 1972)
(Table 7).
Table 7 Proportion (%) of patients with AD/SDAT and MID with a score above 1 on the CPRS* variables
Observed items AD/SDAT MID
Apparent sadness 11.1 0 n.s.
Elated mood 16.7 5.0 n. s.
Labile emotional response 22.2 10.0 n.s.
Sleepiness 5.6 5.0 n.s
Distractability 27.8 0 p<0.02
Perplexity 50.0 10.0 p<0.01
Disorientation 89.0 50.0 p<0.01
Specific speech defects 44.5 5.0 p<0.01
Perseveration 27.8 5.0 n.s.
Overactivity 5.6 0 n.s.
Slowness of movement 33.3 10.0 n.s.
Agitation 5.6 0 n.s.
Global rating of illness 100.0 100.0 n.s.
Recent memory 94.5 75.0 n.s.
Long-term memory ■ 89.0 25.0 p<0.001
* From Åsberg et al (1978)
Mood disorders are regarded as a specific complication of cerebro vascular diseases as such (Marshall et al 1977) and not necessarly
associated with dementia. In fact, dementia as a symptom after stroke is unusual (Tomlinson et al 1968, deReuck et al 1982). Furthermore, the evaluation of cognition after stroke is diffi cult, while patients wrongly could be classified as demented. Thus, one explanation for the discrepancy between earlier observa tions and the present study could be a predominance of stroke patients, without dementia in earlier studies. Psychopathological signs like "distractability", "perplexity", "disorientation" and "specific speech defects" determine the quality of the verbal con tact with the patient. That these symptoms are less pronounced in MID patients could partly explain that they generally are con sidered to have at least a partial or good contact with their sur roundings and a better preservation of their personality. Rothschild 1941, Ehrentheil 1957, Birkett 1972). Further, the significantly better "long-term memory" observed in MID patients compared with AD/SDAT patients probably contributes to the better relative preservation of personality observed in MID patients.
Most authors agree that the best way to separate AD/SDAT from MID %is by the course of the illness, associated arteriosclerotic diseases and a neurological and neurophysiological investigation (Ehrentheil 1957, Hachinski et al 1974, Ladurner et al 1981, Reis berg & Ferris 1982). Gustafson and Nilsson (1982), however, used a dementia rating scale mainly based on the psychopathology of Alzheimer's and Pick's disease but their results still indicate that it is possible to distinguish MID from the primary degenera tive dementias using this scale. Most other authors try to sepa rate AD/SDAT from MID with psychometric tests. Significant diffe rences are found between the two diseases and AD/SDAT patients performed significantly and consistently lower than MID patients, although the type of cognitive deficits is similar for the two disorders e.g. focal or diffuse (Perez et al 1975, Ladurner et al 1981).
In studies I and II we have chiefly used routine clinical methods in investigating the patients and controls. Most of these methods
20
show group differences between AD/SDAT, MID and controls. The investigations, however, show an overlap between the different forms of dementia and the controls and no one method can be used alone as a diagnostic tool. To differentiate between different kinds of dementias, and to find underlying treatable causes for a dementing illness, it is of vital importance to perform a thorough investigation of every patient with dementia symptoms.
Etiological studies (III, IV, V)
Albumin and immunoglobulin in plasma and cerebrospinal fluid (III) Patients with MID had significantly higher concentrations of plas ma IgG (p<0.001) and a slightly elevated IgA compared with AD/SDAT patients and the controls. CSF-Alb was significantly higher in the MID group compared with the controls (p<0.01) and slightly higher compared to AD/SDAT patients. AD/SDAT patients had significantly lower concentrations of CSF-IgG than the MID patients (p <0.05). The CSF/plasmà IgG ratio was lower for the AD/SDAT group and the CSF plasma Alb ratio was higher for the MID group compared with the other two groups (Table,8).
Q
Table 8 CSF/Plasma ratios for IgG and Alb x 10° and CSF IgG index (CSF/plasma IgG ratio):(CSF/Plasma Alb ratio)
- ; Ö
Group Number* * IgG ratio x 10 Alb ratio x 10 CSF IgG index
AD/SDAT 20 1.9 + 1.7 9.8 + 3.5+ 0.20 + 0.23* MID 15 2.3 ± 1.8 14.2 + 6.5° 0 0.18 + 0.13° 0 Controls 16 2.9 ± 2.2 7.7 + 3.4 0.34 ± 0.23
+ p< 0.05, AD/SDAT compared with MID patients * p< 0.05, AD/SDAT compared with controls oo p< 0.01, MID compared with controls
« *5
JJk-nAk
I «G-ratii
^ • %
Figure 1. Evaluation graph for protein profile of cerebrospinal fluid according to Reiber (1980)
IgG ratio = CSF/plasma IgG x 10 Alb-ratio = CSF/plasma Alb x 10
a. Explanation tö figure. 1. Normal range; 2. B-CSF-B dysfunction with proportionally increased IgG; 3. B-CSF-B dysfunction with disproportionally increased IgG; 4. B-CSF-B,dysfunction with addi tional increased IgG locally produced in the CNS; 5. range for values with a locally in CNS synthetized IgG; 0. biologically irrelevant range.
b. Controls c. MID d. $DAT • Males G Females
22
Both patients with MID and AD/SDAT fell into an area which Reiber (1980) designates as ”biologically irrelevant", while controls fell into the area for the normal range (Figure 1). Transudation of IgG through the blood cerebrospinal fluid barrier was higher in the MID group compared both with the controls (p< 0.001) and the AD/SDAT patients (p< 0.01). The AD/SDAT patients showed a signifi cantly higher transudation when compared to controls (p< 0.01). No signs of local synthesis were found in either type of dementia.
The high plasma IgG levels in the MID group indicate a more active immune response in this group. Whether this is due to specific brain reactive antibodies (Nandy 1978) or a secondary manifesta tion due to vascular damage in the body remains to be established (Beregi et al 1978).. Elevated concentrations of CSF-Alb in the MID and AD/SDAT groups and an increased CSF/plasma Alb ratio in the MID group indicate either a less dense blood cerebrospinal fluid barrier function or a lower CSF turnover of albumin in the CNS. The high transudation values found in MID patients point towards a disturbed barrier function as ' being the underlying mechanism, which seems reasonable concerning the vascular background of this disease (Fischer 1968, Worm-Petersen & Pakkenberg 1968, Bousser 1977). The low CSF/plasma IgG ratio and CSF-IgG index in the AD/SDAT group along with a moderately higher transudation value compared with controls may indicate a consumption of IgG in the CNS. This binding could be explained by the findings of IgG in amyloid in senile plaques and vessel walls (Tomlinson et al 1970, Ishii & Haga 1976, Torack & Lynch 1981). Whether this binding is specific, e.g. involving brain reactive antibodies (Nandy 1978) or nonspecific, remains to be investigated. Both groups of demented patients fell into the area "biologically irrelevant" (Ganroth & -Laurell 1974, Reiber 1980). It appears that this pattern is speci
fic for demented patients and could be explained with an increased transudation of albumin and an antibody binding in the CNS.
tiç _changes _in jDatients „with „senile „dementia _ (_IV )_
no deviations from normal banded karyotypes were found in the patients with AD/SDAT (Mark & Brun 1973, Brun et al 1978, White et al 1981). An increased frequency, however, of structurally abnor mal chromosomes was observed in patients with AD/SDAT compared with both MID patients and controls (Table 9). All patients with AD/SDAT, except two, had at least one out of 100 cells carrying one or more abnormal chromosomes compared with 5 out 17 with MID and 2 out of 11 patients with Down’s syndrome and none out of 21 controls. The abnormal chromosomes looked like acentric fragments, i.e. there were no visible centromeres (Figure 2).
UU
-
m
-
h
Ìirai
C 6-12
IH fli n n«fi îiw sü j
D 13-15 E 16-18
il ft ö rt
a
a
X...x<
XK Ärt
su
F 19-20 G 21-22 A A.... t» flFig 2. Cell from a male SDAT-patient showing chromosomes without visible centromeres (arrows).
24 o o o 2 < O X OO <c o. o
2
a» oo 00 OJ O O z 4-» O 0) E cn to to _ O ro l- OJ X s- O JÛ CQ 00 <c ■O ro to E O ro l- CU O a: X o to CL ro cn cn ro o o o r-* o oo oo co <o o\ ro o cn cn oo oj «— «3“ cn cn o «3- n*» m ro *3- evi +i +i ro cvj co oj +i +1 00 00 vo O'» r-wo I-* * * * o o o o +1 cn lo o cn ro lo «o- «3: o o o o +1 +1 co »— ro evi *3- cvj *3-o *3-o o o +1 +i in cn evi ro +i lo cn o o +i VO in ro. CVJ +1 o ro cn cn o o +i «53- CVJ cn *-«r— CVJ +1 cn oo oo cvj oo cvj cn vo ■*— i— ro cvj 1— CVJ «— CVJ CVJ +1 +1 +1 Q +-> »—i l.0 Vf- (D u z: -0 O CL c >> ftJ >> ro 4-> M- ro C T- 1/1 4-> i- ^ •r- J-J - to. c <u cn ■r CN -<U E •fJ <U -r- c • E *r- H r- E II £ 0J 0) c =J <L> C 0 o jc —* 2T -a ^ Q * < Q oo o LO o o V/ a. * * p < 0 . 0 0 1 , D o w n ' s s y n d r o m e a n d c o n t r o l s c o m p a r e d t o S D A TInformative metaphases indicated premature centromere division .of whole chromosomes with a random chromosome involved. Similar chro mosome abnormalities in cells from patients with senile dementia have been reported by other authors (Bergener & Jungklass 1970, Martin et al 1981) but others have been unable to confirm this observation (Nielsen et al 1968, Mark & Brun 1973, Brun et al 1978, White et al 1981).
The chromosomal anomaly was most frequent in patients with AD/SDÀT but was also observed in some of the patients with multiinfarct dementia and in a low frequency in Down's syndrome, but were not seen in the controls. This might imply that this abberation is pathognomic for some types of dementia although not for AD/SDAT.
The degree of aneuploidy was elevated in the patints with AD/SDAT (9.8% cells with aneuploidy) and MID (9.8% cells with aneuploidy) compared with controls (6.5% cells with aneuploidy), although cells with acentric fragments were excluded. Other authors have also reported increased aneuploidy in patients with AD/SDAT (Ward et als 1979, Nordenson et al 1980). No deviations, however, from normal diploid chromosome number could be found by other authors (Nielsen et al 1968, Mark & Brun 1973 and Martin et al 1981). This may be due to the relatively low number of patients studied or the relatively higher percentage of very old individuals included as controls in these studies.
The mechanism behind premature division of the centromeres is not known. An attractive hypothesis is that premature centromere divi sion is connected with "hetero-chromatinization" of the affected chromosomes i.e. the inactive part of the chromosome being larger than normal. Increased amounts of "heterochromatin" have been found in neuronal and glial cells from patients with AD/SDAT (Crapper et al 1979, deBoni & Crapper-MacLachlan 1980). However, it remains to be shown that "heterochromatinization" is a general phenomenon in AD/SDAT and that it also occurs in other tissues. AD/SDAT has, however, been suggested to be a systemic disease with
26
changes also in other tissues (Adolfsson et al 1980, Carlsson et al 1980, Bucht et al 1983). Our findings of structural chromosomal cfianges in lymphocytes in patients with AD/SDAT strengthen this suggestion. The predominant neuronal lesion in AD/SDAT is the neu rofibrillary tangles which consist of paired helical filaments. If AD/SDAT is a generalized disease the disorganized microtubules might be present in other tissues outside the central nervous sys tem. Microtubules or other fibrillary elements are responsible during cell division for accurate chromatid separation and distri bution of chromosomes to the daughter cells. Thus, the increased frequency of fragments and the neurofibrillary tangles in AD/SDAT might be associated phenomena, both related to impairment of neu- rofibrills and microtubules.
Blood glucose and insulin secretion in patients with dementia (V)
The retrospective study of records of patients with dementia shows that 457 had MID, 317 AD/SDAT and 65 confusional states. Sixty- three of these patients were also found to have overt diabetes mellitus. None of the patients with diabetes mellitus were found in the group of patients with AD/SDAT, 55 were found in the group of patients with MID and 5 in the group of patients with confusio nal states. These findings suggest that AD/SDAT and diabetes mel litus in old age may not co-exist.
Lower fasting blood sugar was found in patients with AD/SDAT com pared with the other four groups and significantly lower compared with patients with CVD and hospitalized controls. The mean area under the OGTT-curve was smaller in the group of patients with AD/SDAT compared with all other groups and significantly smaller compared with the group of patients with CVD and the hospitalized controls (‘Table 10). The basal level of insulin before starting the OGTT was the same in a group consisting of patients with AD/SDAT and healthy elderly controls. During the OGTT patients with AD/SDAT had higher levels of insulin (Figure 3).
T A B L E 1 0 F a s t i n g b l o o d s u g a r a n d t h e a r e a u n d e r n e a t h t h e 0 6 T T c u r v e s i n p a t i e n t s w i t h S D A T , C V D a n d M I D a n d i n c o n t r o l p a t i é n t s . T h e s t a t i s t i c a l c o m p a r i s o n b e t w e e n S D A T a n d t h e o t h e r s t u d i e d g r o u p s w a s p e r f o r m e d w i t h S c h e f f é ' s m e t h o d o f p o s t e r i o r c o n t r a s t . o o i— r— O
e
CD 4 -Q OO CD S- CT> fO < <1) >, O i-cr> >> •a cn •»-VD o +l 00 00 +l 00 cn <£> V£> 00 +l <r ■ U*1 o N/ Ql v__ .—. rO i- Ol •r- <U fO CO >> 4-> Ql O O C >» *3 Q +J ►—« -0 -C i_ CD 4-> O > cj S CD C_> CD E i/l o V- ^ ini*■—■ TO O O CD i_ CO ' r0 > -D CD > fO i— c/i CD C-) z E O CD c •<- «O r-CD -r- S- c/1 •r- 4-> 4-> O «— «—* \/ \/ <— <D Jd ro ■r~ C — i- 03 "O •r- jz CD CD 4-> QJ Q. -M E *— +— i— C M l- i/l r- E CO C i- CD < <c CD r— CD 3 CD O O O O O 0 uo <t O O 2: -o X O X \ 1/1 C/1 l/l T3 O r-O Y S D A T < C28 taoNa/t 100 90 60* 70 60 50 40 30 20 10 P. C. 30 60 90 Miwt«i
Fig 3. Insulin levels in serum in arbitrary units during the OGTT in 10 patients with senile dementia of Alzheimer type (SDAT) and 10 age matched healthy elderly persons (C ). The difference in
eld
insulin level is statistically significant at 90 .minutes (p<0.05).
From this study it seems that the patients with AD/SDAT have a changed carbohydrate metabolism with decreased fasting blood glu cose and an increased glucose tolerance, i.e. smaller areas under the OGTT curve. The difference in blood sugar concentration and the area underneath the OGTT curve are not significantly changed when AD/SDAT patients were compared with healthy elderly normal persons. An explanation for this might be that the healthy elderly persons were not hospitalized, were on a different diet and were physically more active than the other four groups, conditions which all contribute to an increased glucose tolerance. No signs of malabsorption could be found when patients with AD/SDAT were examined with different malabsorption tests.
We have no satisfactory explanation for the elevated insulin
levels in patients with AD/SDAT, but a decreased insulin sensiti vity of the cells in patients with AD/SDAT, leading to a compensa tory increase in insulin levels could explain this. Also, the neu rotransmitters are of importance in the regulation of insulin and glucagon release, both by the sympathetic and parasympathetic ner vous system (Woods & Porte 1974, Bereiter et al 1981, Steffens 1981). Cholinergic and catecholaminergic substances also play an important role in the modulation of insulin and glucagori release in the pancreas (Feldman & Lebowitz 1970, Woods & Porte 1974, Lindström & Sehlin 1983) and a deficit of those might explain the elevated insulin levels we found in patients with AD/SDAT.
On the other hand, different neurotransmitters are serfsitive to hypoglycemia and hypoxia (Davis & Carlson 1973, Gibson & Blass 1976, Hedner 1977, He^dher & Lundborg 1977, Blass & Gibson 1978). If an error in the metabolism leading to hypoglycemia is the pri mary defect in patients with AD/SDAT* this could be the explana tion for the impaired synthesis of acetylcholine.
It has recently been discovered that patients with AD/SDAT have low levels of somatostatin in their brains (Davis et al 1980, Ros-' sor et al 1980). Somatostatin is a potent inhibitor of insulin secretion (Alberti et al 1973) and a deficiency of somatostatin in the pancreas in patients with AD/SDAT could explain our findings of high insulin levels.
If our findings of low blood sugar and high serum insulin concent rations in patiets with AD/SDAT are a primary defect causing the transmitter deficiencies, or, perhaps, more likely aré due to the known transmitter deficiencies in AD/SDAT remains to be estab lished. Whatever the cause of the hypoglycemia is in AD/SDAT, the already low concentrations of brain neurotransmitters are further negatively influenced by the lack of an important source of energy or perhaps a lack of metabolites necessary for the production of neurotransmitters.
30
GENERAL SUMMARY AND CONCLUSIONS
I. Clinical studies
Significant differences between several diagnostic procedures were found in a prospective study comparing patients with Alzheimer's disease and senile dementia of Alzheimer type (AD/SDAT) and mul tiinfarct dementia (MID). The majority of the patients with AD/SDAT were free of other diseases, while the majority of MID patients had cardiovascular disease. Blood préssure was signifi cantly lower in the AD/SDAT group and focal neurological signs were seen in 70% of the MID patients but only in 6% of patients with AD/SDAT. All AD/SDAT patients had a normal electrocardiogram (ECG) while 75% of the MID patients had a pathological ECG. Elec troencephalogram (EEG) showed generalized slow frequencies in 79% of the AD/SDAT patients and localized slow frequencies or abnorma lities were seen in 65% of the MID patients. Computerized tomo graphy showed visible infarcts in only 25% of the MID patients, but this group had a significantly greater dilation of the ventri cular system than patients with AD/SDAT. Monoamine metabolites in the cerebrospinal fluid were lower in the AD/SDAT group than in the MID and control groups. MID patients had normal concentra tions. Rating of psychopathological signs showed that AD/SDAT patients have a more variable psychopathology than patients with MID. MID patients had less pronounced psychopathological signs, especially those which make a verbal and personal contact with the patient possible, possibly explaining the generally accepted opi nion that patients with MID have a better preservation of their personalities.
On the basis of these findings, the following laboratory investi gations are proposed in patients with dementia in order to detect underlying treatable diseases and differentiate between AD/SDAT and MID.
I Somatic status examination including a thorough neurologiçal assessment.
II Mental status examination, including a rating of psychopatho- logical signs.
III Laboratory investigations Complete blood count Electrolytes
Liver function tests. Thyroid function tests Vitamin B -j
2
anc* folic acid Fasting blood sugar Urine analysis Electrocardiogram ElectroencephalogramThese investigative methods will in most instances reveal the correct diagnosis. In some instances, when the diagnosis is unclear, this list should be completed with cerebrospinal fluid investigations, computerized tomography of the brain or other spe cific investigations.
In the future other more refined investigative methods might be come available and enable a further increase in diagnostic accura cy. Some examples of these might be improved cytogenetic procedu res, albumin and immunoglobulin determination in serum and CSF, based upon established differences between different dementia disorders.
II Etiological studies
1. Study of immunoglobulin and albumin in patients with AD/SDAT and MID indicates a more active immune response in MID patients and a less dense blood cerebrospinal fluid barrier in both patients with MID and AD/SDAT. A consumption of IgG in the Cen tral nervous system seems to be present in patients with AD/SDAT, possibly explaining the findings of IgG and amyloid in senile pla ques and vessel walls. The differences in IgG and albumin in CSF
32
and serum in patients with MID and AD/SDAT could possibly also be of diagnostic value.
2. Abnormal chromosomes, appearing as acentric fragments, i.e. chromosome^ without visible centromere, were found in 90% of patients, with AD/SDAT, 30% of patients with MID, 20% of patients with Down's syndrome, but not at all in the control group. Infor mative metaphases indicated premature centromere division as the probable cause. Increased aneuploidy was seen both in patients with MID and AD/SDAT.
The appearance of acentric fragments in patients with dementia could be of diagnostic value in differentiating the dementias, and especially in differentiating organic dementia from other disea ses with dementia symptoms.
3. Diabetes mellitus in old age and AD/SDAT do not seem to coex ist. Furthermore, patients with AD/SDAT have a changed carbohydra te metabolism with decreased fasting blood sugar concentration, increased glucose tolerance and higher concentration of insulin during an oral glucose tolerance test. These findings could either be an effect of the known transmitter deficiencies in AD/SDAT or, on the other hand, be an explanation for the transmitter deficien cies, because hypoglycemia leads to a decrease in the production of most neurotransmitters<
ACKNOWLEDGEMENTS
I am greatly indebted to all who have contributed in different ways to this study and I wish to express my severe gratitude to the following:
My family, for their outstanding support and patience
My friend and fellow-student, professor Bengt Winblad, who has been my instructor, for his continuous support, stimulating dis cussions, constructive criticism and cooperation
Professor Per-Olof Wester for placing time and facilities at my disposal and for his personal support and interest
Med Dr Rolf Adolfsson, who throughout these studies has been my close coworker. His constructive criticism and creative spirit is greatly appreciated
Professor Frank Bergman, for placing the facilities of his insti tution at my disposal
Docent Ingrid Nordensson, professor Gunhild Beckman, docent Folke Lithner and med kand Irina Alafuzoff for their stimulating colla boration and co-authorship
Mrs Karin Gladh for excellent secreterial support, first-class typing and never failing humour
Mrs Ann-Britt Lundström for her technical assistance and skilful drawing of the figures
Photographer Bengt Carfors for excellent assistance
Fil lie Robert Elston for perfect linguistic revision
Docent Lars von Knorring, fil lic Lennarth Nyström, docent Stefan Marklund, med kand Per Nyberg and psychologist Gun Kajsajuricti for helpful criticism and support in different parts of the study
Dr Alf Myrstener, Walter Rapp and Tage Lundström for continuing support and for placing time at my disposal
The entire staff at the Department of Internal Medicine, Umeå uni versity, Ward 1 and 3, and the Department of Geriatric Psychiatry, Umedalen's sjukhus for their never failing cooperation
Last, but not least, I wish to express my sincere gratitude towards patients and their relatives who made this study possible
34
REFERENCES
Adolfsson, R., Gottfries, C.G., Roos, B.E. & Winblad, B. Changes in the brain catecholamines in patients with dementia of Alzheimer type. Brit. J. Psychiat. 135:216-223, 1979.
Adolfsson, R., Gottfries C.G., Oreland, L., Wiberg, Å. & Winblad, B. Increased activity of brain and platelet monoamine oxidase in dementia of Alzheimer type. Life Sci. 27:1029-1034, 1980.
Alberti, K.G.M.M., Christensen, E., Iversen, J., Seyer-Hansen, K., Christensen, N.J., Hansen, A.P. & Örskov, H. Inhibition of insulin secretion by somatostatin. Lancet 11:1299, 1973.
Alzheimer, A. Neuere Arbeiten über die Dementia senilis und die auf atheromatöser Gefässerkrankung basierenden Gehirnkrankheiten. Monatschr. Psychiat. Neurol. 3:101-115, 1898.
Alzheimer, A. Über eine eigenartige Erkrankung der Hirnrinde. Allg. Zeitschr. Psychiat. 64:146-148. 1907.
American Psychiatric Association. Diagnostic and Statistical Man- nual of Mental Disorders, 3rd ed. Washington DC, APA, 1980, pp 101-162.
Åsberg, M., Perris, C., Schalling, D. & Sedvall, G. The CPRS- Development and applications of a psychiatric rating scale. Acta Psychiat. Scand. Suppl 271, 1978.
Beregi, E., Lengyel, E. & Biro, S. Autoantibodies in aged indivi duals. Akt. Gerontol. 8:77-80, 1978.
Bereiter, D.A., Rohner-Jeanrenaud, F., Berthoud, H.R. & Jeanre- naud, B. CNS modulation of pancreatic endocrine function. Multiple modes of expression. Diabetologia 20:417-425, 1981.
Bergener, M. & Jungklass, F.K. Genetische Befunde bei Morbus Alz heimer und Seniler Demenz. Gerontol. Clin. 12:71-75, 1970.
Birkett, D.P. The psychiatric differentiation of senility and arteriosclerosis. Brit. J. Psychiat. 120:321-325, 1972.
Birkett, D.P. & Raskin, A. Arteriosclerosis, infarcts and demen tia. J. Am. Geriatr. Soc. 30:261-265, 1982.
Blass, J.P. & Gibson, G.E. Cholinergic systems and disorders of carbohydrate catabolism. In: Jenden, D.J. (ed.) Cholinergic mechanism and psychopharmacology, Plenum Press, New York, 1978, pp 791-803.
deBoni, U. & Crapper-McLachlan, D.R. Senile dementia and Alzhei mer's disease: A current view. Life Sci. 27:1-14, 1980.
Bousser, M.G. Les conceptions actuelles de démences artériopathi- que. L'Encéphale 3:357-372, 1977.
Bowen, D.M., Spillane, J.A., Curzon, G., Meier-Ruge, W., White, P., Goodhardt, M.J., Iwangoff, P. & Davison, A.N. Accelerated aging or selective neuronal loss as an important cause of demen tia? Lancet 1:11-14, 1979.
Brun, A., Gustafson, L. & Mitelman, F. Normal chromosome banding pattern in Alzheimer's disease. Gerontology 24:369-372, 1978.
Bucht, G., Adolfsson, R., Lithner, F. & Winblad, B. Changes of blood glucose and insulin secretion in patients with senile demen tia of Alzheimer type. Acta. Med. Scand. Acc. for publication 1982.
Carlson, A., Adolfsson, R., Aquilonius, S.M., Gottfries, C.G., Oreland, L., Svennerholm, L. & Winblad, B. Biogenic amines.in human brain in normal aging, senile dementia and chronic alcoho
36
lism. In: Goldstein, M. et al. (eds.) Ergot compounds and brain fucntion: Neuroendocrine and neuropsychiatrie aspects, Plenum Press, New York, 1980, pp 295-304.
Corsellis, J.A.N. Mental illness and the ageing brain. Oxford Uni versity Press, Oxford, 1962.
Crapper, DrR. & Dalton, A.J. Aluminium induced neurofibrillary degeneration, brain electrical activity and alterations in acqui sition and retention. Physiol. Behav. 10:935-945, 1973.
Crapper, D.R., Krishnan, S.S. & Quittkat, S. Aluminium, neurofib rillary degeneration and Alzheimer's disease. Brain 99:67-80, 1976.
Crapper, D.R. & deBoni, U. Brain aging and Alzheimer's disease. Can. Psychiat. Assoc. J. 4:229-233, 1978.
Crapper, D.R., Quittkat, S. $ deBoni, U. Altered chromatin confor mation in Alzheimer's disease. Brain 102:483-495, 1979.
Davies, P. & Maloney, A.J.F. Selective loss of cholinergic neuro nes in Alzheimer's disease. Lancet 11:1403, 1976.
Davies, P., Katzman, R. & Terry, R.D. Reduced somatostatin-like immunoreactivity in cerebral cortex from cases of Alzheimer's disease and Alzheimer senile dementia. Nature 288:279-280, 1980.
Davis, J.N. & Carlson, A. The effect of hypoxia on monoamine synthesis levels and metabolism in rat brain. J. Neurochem. 21:783-790, 1973.
Drachman, D.A. Memory and cognitive function in man: Does the cho linergic system have a specific role? Neurology 27:783-790, 1977.
affective disorders in aged patients. Geriatrics 12:426-432, 1957.
Feldman, J.M. & Lebowitz, H.E. Serotonininhibition of in vitro insulin release from golden hamster pancreas. Endocrinology 86:66-70, 1970.
Fisher, C.M. Dementia and cerebral vascular disease. In: Toole, J.F., Siekert, R.G. & Wishnant, J.P. (eds.). Cerebral vascular disease: 6th Princeton Conference, New York, 1968, pp 232-236.
Freemon, F.R. Evaluation of patients with progressive intellectual deterioration. Arch. Neurol. 33:658-659.
Ganroth, K. & Laurell, C.B. Measurement of IgG and albumin content of CSF, and its interpretation. Clin. Chem. 20:571-573, 1974.
Gershon, S. & Herman, S.P. The differential diagnosis of dementia. J. Am. Geriatr. Soc. 30:S58-66. Ì982.
Gibbs, C.J. & Gajdusek, D.C. Subacute spongioform virus encephalo pathies. The transmissible virus dementias. In: Katzman, R., Ter ry, R.D. & Bick, K.L. (eds.) Alzheimer's disease: Senile dementia and related disorders (Aging voi 7), Raven Press, New York, 1978, pp 559-575.
Gibson, G.E. & Blass, J.P. Impaired synthesis of acetylcholine in brain accompanying mild hypoxia and hypoglycemia. J. Neurochem. i 27:37-42, 1976.
Glenner, G.G. Current knowledge of amyloid deposits as applied to senile plaques and congophilic angiopathy. Ip: Katzman, R., Terry, R.D. & Bick, K.L. (eds.) Alzheimer's disease: Senile dementia and related disorders (Aging voi 7), Raven Press, New York, 1978, pp 493-501.
38
Gordon, E.B. & Sim, M. The EEG in presenile dementia. J. Neurol. Neurosurg. Psychiat. 30:285-291, 1967.
Gottfries, C.G., Rosengren, A.M. & Rosengren, E. The occurrence of homovanillic acid in human brain. Acta Pharmacol. Tox. 23:36-40, 1965.
Gottfries, C.G., Gottfries, I. & Roos, B.E. Homovanillic acid and 5-Hydroxyindoleacetic acid in the cerebrospinal fluid of patients with senile dementia, pi'esenile dementia and Parkinson's disease. J. Neurochem. 16:1341-1345, 1969.
Gustafson, L. & Nilsson, L. Differential diagnosis of presenile dementia on clinical grounds. Acta Psychiat. Scand. 65:194-209, 1982.
Haase, G.R. Diseases presenting as dementia. In: Wells, C.E. (ed.). Dementia, F.A. Davis Co., Philadelphia, 1977, pp 27-68.
Hachinski, V.C., Lassen, N.A. & Marshall, J. Multiinfarct demen tia, a cause of mental deterioration in the elderly. Lancet 11:207-209, 1974.
Hachinski, V.C., Iliff, L.D., Zilhka, E., duBoulay, G.H., McAllis ter, V.L., Marshall, J., Ross Russell, R.W. & Symon, L. Cerebral blood flow in dementia. Arch. Neurol. 32:632-637, 1975.
Harrison, M.J.G., Thomas, D.S., duBoulay, G.H. & Marshall, J. Mul ti-infarct 'dementia. J. Neurol. Sci. 40:97-103, 1979.
Hedner, T., ‘Lundborg, P. & Engel, J. Effect of hypoxia on monoami ne synthesis in brains from developing rats. Biol. Neonate. 31:122-126, 1977.
Heston, L.L. & Mastri, A.R. The genetics of Alzheimer's disease. Arch. Gen. Psychiat. 34:976-981, 1977.
Hontela, S. & Schwartz, G. Myocardial infarction in the differen tial diagnosis of dementias in the elderly. J. Am. Geriatr. Soc. 27:104-106, 1979.
Hughes, C.P., Myers, F.K., Smith, K. & Torack, R.M. Nosologic problems in dementia. A clinical and pathologic study of 11 cases. Neurology 23:344-351, 1973.
Hqtton, J.T. Clinical nosology of the dementing illness, In: Maletta, G.J. & Pirozzolo, F.J. (eds.) Advances in neurogeronto logy, Praeger, New York, 1980, pp 149-174.
Hutton, J.T. Results of clinical assessment for the dementia syn drome: Implications for^epidemiologic studies. In: Mortimer, J.A. & Schuman, L.M. (eds.) The epidemiology of dementia. Oxford Uni versity Press, New York, 1981, pp 62-69.
Ishii, T. & Haga, S. Immuno-electron microscopic localization of immunoglobulins in amyloid fibrils-of senile plaques. Acta Neuro- pathol. (Berlin) 36:243-249, 1976.
Jellinger, K. Neuropathological aspects of dementias resulting from abnormal blood and cerebrospinal fluid dynamics. Acta Neurol. Belg. 76:83-102, 1976.
Katzman, R. Normal pressure hydrocephalus, In: Wells, C.E. (ed.) Dementia, F.A. Davis Co., Philadelphia, 1977, pp 69-92.
Kidd, M. Paired helical filaments in electron microscopy of Alz heimer's disease. Nature 197:192-193, Ì963.
Klatzo, I., Wisniewski, H.M. & Streicher, E. Experimental produc tion of neurofibrillary degeneration. J. Neuropathol. Exp. Neurol. 24:187-199. 1965.
40
Klinische Unterscheidungskriterien bei vaskulärer (Multiinfarkt) und primär degenerativer Demenz (Alzheimer). Nervenarzt
52:401-404, 1981.
Laureil, C.B. Electroimmunoassay. Scand. J. Clin. Lab. Invest. 29: 21-37, 1972.
Lindström, P. & Sehlin, J. Opposite effects of 5-hydroxytryptophan and 5-hydroxytryptamin on the function of microdissected ob/ob mouse pancreatic islets. Diabetologia 24:52-57, 1983.
Liston, E.H. The clinical phenomenology of presenile dementia. A critical review of the literature. J. Nerv. Ment. Dis. 167:329- 336, 1979 a.
Liston, E.H. Clinical findings in presenile dementia. A report of 50 cases. J. Nerv. Ment. Dis. 167:337-342, 1979 b.
Lithner, F. Cutaneous erythema, with or without necrosis, locali zed to the legs and feet - a lesion in elderly diabetics. Acta Med. Scand. 196:333-342, 1974.
Mann, D.M.A., Lincoln, J., Yates, P.O., Stamp, J.E. & Toper, S. Changes in the monoamine containing neurones of the human CNS in senile dementia. Brit. J. Psychiat. 136:533-541, 1980.
Mark, J. & Brun, A. Chromosomal deviations in Alzheimer's disease compared to those in senescence and senile dementia. Gerontol. Clin. 15:253-258, 1973.
Marsden, C.D. & Harrison, M.J.G. Outcome of investigation of patients with presenile dementia. Brit. Med. J. 2:249-252, 1972.
Marshal, F., Folstein, R.M.& McHugh, P.R. Mood disorder as a spe cific complication of stroke. J. Neurol. Neurosurg. Psychiat. 40:1018-1020, 1977.
Martin, J.M., Kellett, J.M-. & Kal..., J. Aneuplöidy in cultured human lymphocytes: II. A comparison between senescence and demen tia. Age and Ageing 10:24-28. 1981.
Mölsä, P. Dementia. A clinical study in the Finnish population. Academic dissertation. Medical Faculty, University of Turku, Tur ku, 1980.
Nandy, K. Brain-reactive antibodies in aging and senile dementia. In: Katzman, R., Terry, R.D. & Bick, K.L. (eds.) Alzheimer’s disease: Senile dementia and related disorders (Aging voi 7), Raven Press, New York, 1978, pp 503-512.
National Diabetes .Data Group. Classification and diagnosis of dia betes mellitus and other categories of glucose intolerance. Diabetes 28:1039-1057, 1979.
Nielsen, J., Jensen, L., Lindhardt, H., Stöttrup, L. & Sönder- gaard, A. Chromosomes in senile dementia. Br. J. Psychiat. 114:303-309, 1968.
Nordenson, I., Adolfsson, R., Beckman, G., Bucht, G & Winblad, B. Chromosomal abnormality in dementia of Alzheimer type. Lancet 1:481-482, 1980.
Perez, F.I., Rivera, V.M., Meyer, J.S., Gay, J.R.A., Taylor, R.L. & Mathew, N.T. Analysis of intellectual and cognitive performance in patients with multi-infarct dementia, vertebrobasilar insuffi ciency with dementia, and Alzheimer's disease. J. Neurol. Neuro- surg. Psychiat. 38:533-540, 1975.
Perry, E.K., Perry, R.H., Blessed, G. & Tomlinson, B.E. Necropsy evidence of central cholinergic deficits in senile dementia. Lan cet 1:189, 1977.
42
Perry, E.K., Tomlinson, B.E., Blessed, G., Bergman, K., Gibson, P.H. ÿ Perry, R.H. Correlation of cholinergic abnormalities with senile plaques and mental test scores in senile dementia. Brit. Med. J. 2:1457-1459, 1978.
Pope, A., Hess, H.H. & Lewin, E. Microchemical pathology of the cerebral cortex in presenile dementias. Trans. Am. Neurol. Ass. 89:15-16, 1964.
Portera-Sanchez, A., del Ser, T., Bermejo, F. & Arredondo, J.M. Clinical diagnosis of senile dementia of Alzheimer type and vascu lar dementia. In: Terry, R.D., Bolis, C.I. & Toffano, G. (eds.) Neural aging and its implications in human neurological pathology, Raven Press, New York, 1982, pp 169-188.
Radue, E.W., duBotllay, G.H., Harrison, M.J.G. & Thomas, D.J. Com parison of angiographic and CT findings between patients with mul tiinfarct dementia and those with dementia due to primary neuronal degeneration. Neuroradiology 16:113-115, 1978.
Reiber, H. The discrimination between different blood-CSF barrier dysfunctions and inflammatory reactions of the CNS by a recent evaluation graph for the protein profile of CSF. J. Neurol. 224:89-99, 1980.
Reisberg, B. & Ferris, S.H. Diagnosis and assessment of the older patient. Hosp.. Com. Psychiat. 33:104-110, 1982.
deReuck, J., Sieben, G., deCoster, W. & vand der Eecken, H. Demen tia and confusional state in patients with cerebral infarcts: A clinical study. Eur. Neurol. 21:94-97, 1982.
Roberts, M.A., McGeorge, A.P. & Caird, F.I. Electroencephalography and computerized tomography in vascular and non-vascular dementia in old age. J. Neurol. Neurosurg. Psychiat. 41:903-906, 1978.
