| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
SPECIAL ARTICLE |
Department of Epidemiology, Graduate School of Public Health, University of Pittsburgh, Pennsylvania.
Address correspondence to Lewis Kuller, MD, DrPH, University of Pittsburgh, Epidemiology, Graduate School of Public Health, Pittsburgh, PA 15213. E-mail: kullerl{at}edc.pitt.edu
A
The incidence and prevalence of dementia are increasing. There is an urgent need to develop a preventive strategy. The identification of modifiable risk factors must therefore be a high priority. Newer imaging techniques provide an opportunity to identify subclinical manifestations of "dementias" that can be limited to the risk factors and subsequent clinical disease. The contribution of vascular disease to dementia and Alzheimer's disease (AD) should be a high priority as it offers a potential preventive strategy. Study designs need to be modified, including a greater emphasis on geographic variations in AD and dementia based on imaging studies, longitudinal studies of successful aging without cardiovascular disease (CVD) or AD, gene–environment interactions, and studies of unique populations that may be at lower risk because of specific lifestyles. Primary prevention trials for vascular disease should include a dementia component. Most, if not all, studies should include newer imaging studies.
Imaging of the brain, especially structural and functional magnetic resonance imaging (MRI) and positive emission tomography (PET) scanning, have provided an excellent opportunity to study both global and focal brain characteristics and metabolism in epidemiological and clinical studies (2). These studies as well as the ongoing postmortem evaluations in longitudinal studies offer an excellent opportunity to enhance dementia research (3). Recent longitudinal studies using imaging techniques have improved our ability to measure the determinants and, to a limited extent, the changes in the brain which occur prior to clinical diagnosis of dementia. Pathological changes characteristic of Alzheimer's disease (AD) are found prior to clinical diagnosis of AD (4). Extensive vascular disease can be identified in the brain increasing with age and found in both healthy persons and patients with dementia (5–9).
The need to identify the specific risk factors for dementia and interrelationship with genetic host susceptibility is critical to develop preventive strategies. It is extremely unlikely that we will be able to substantially reduce the morbidity and disability associated with dementia or AD without having a preventive, especially primary prevention, strategy (10).
Specific environmental and lifestyle risk factors that increase or decrease incidence of specific types of dementia have been reported (1). The most consistent observations have been the risk of dementia with lower levels of education, the effects of aging, and possibly higher risk associated with selected measures of inflammation.
Many studies have suggested that increasing physical activity (11) and socialization activities may reduce risk of dementia and AD. Alcohol consumption has also been associated with a reduced risk of dementia in some studies (12). Elevated systolic blood pressure (SBP), especially midlife SBP has been a consistent strong predictor of dementia, including AD (13–17). There is also evidence that a low cholesterol, saturated fat, or total caloric intake, especially beginning at early ages, may be related to a lower risk of AD.
The initial evidence of a strong association of apolipoprotein-E4 (ApoE4) with dementia began a major thrust in understanding the genetics and host susceptibility, especially as related to AD. More recent studies, however, have failed to substantially enhance the genetics of AD or other dementias, except in a few high-risk families.
There are probably at least three major interrelated hypotheses for the etiology of dementias. First, most dementia is due to AD, an amyloidosis, a protein-folding disorder (18). The disease results from changes in normal soluble proteins to insoluble fibrils. The pathology is due to specifically plaques consisting of amyloid ß 1-40 (Aß40) and 1-42 (Aß42), which form high fibrillar extracellular content, injury to neurons, increased intracellular neurofibrillary tangles due to hyperphosphorylated tau, and neuronal injury and loss ultimately leading to the cognitive changes consistent with AD. The cause of the disease is likely to be a combination of genetics and exposure to unknown environmental risk factors [such as high levels of oxidative stress (19) in the brain or injury secondary to vascular disease and inflammation] or to an undetected agent (such as a virus).
The development of new methods of quantifying Aß in the brain in vivo with PET scanning may provide an approach to quantify the extent of Aß in AD patients and controls and also to predict dementia based on the dose of Aß in the brain (20).
Lipoprotein metabolism in the brain likely plays an important role in the production of amyloid precursor protein (APP) and in Aß deposition (21,22). In animal models of AD, hypercholesterolemia accelerates the development of amyloid pathology in the brain. There is suggestive evidence that a high-fat and high-cholesterol diet, especially at younger ages, may increase risk of AD (22). Altered regulation of the low-density lipoprotein receptor (LRP) in the brain may be associated with risk of AD (23–25). Statin therapy has been reported in several experimental models to effect both the APP processing and microglia activity and risk of dementia (26,27).
Levels of Aß40 and Aß42 identified in the plasma are much lower than in spinal fluid and are inconsistently related to the risk of dementia (28,29). The major clearance of Aß is probably through the kidney (30–32). Higher levels of both Aß40 and Aß42 are found in older individuals with even moderately abnormal renal function. Aß levels are also related to measures of vascular disease, and higher levels of Aß in the plasma are a predictor of age-specific total mortality (32). It is possible that clearance of Aß in the central nervous system to peripheral blood and excretion by the kidney is a determinant of amyloid deposition and plaques in the brain. Measures of Aß turnover rather than static levels and especially in relationship to renal function may be an important measure of the increased accumulation of Aß in the brain.
A second hypothesis is that aging is associated with a loss of both white and gray matter neurons and the development of dementia (33–35). The amount of brain loss is quantified in part by increasing size of the ventricles. Increase in ventricle size with aging has been documented repeatedly both by computed tomography (CT) and MRI. The greater size of the ventricles is a risk factor for dementia (35). The reason for the loss of gray or white matter is not known. Small-vessel vascular disease in the white matter has been considered a risk factor for gray matter neuronal loss (36,37). Increased oxidative stress with aging, inflammation, or loss of neurotrophic factors may contribute to global brain atrophy. Increased exercise, cognitive training, antioxidants, and anti-inflammatory agents could reverse this global brain atrophy. It will be important to determine whether the rates of decline in gray matter vary among populations and within populations by specific risk factors and genetic (host susceptibility) attributes.
The third hypothesis states that vascular disease in the brain, especially small-vessel disease due primarily to hypertension, results in increasing white matter abnormalities as measured by white-matter lesion load and secondarily to loss of neurons in the gray matter. Vascular disease is also believed to contribute to focal neuronal loss and possibly to the increasing amount of Aß and the development of fibrillar plaques in AD (38–45).
We and others have described the strong association of dementia with vascular disease in the brain as measured by MRI, high white-matter lesion load, and infarcts 
3 mm (46,47). Epidemiological and clinical studies have demonstrated that there is a high prevalence of white-matter abnormalities among populations (48). White-matter abnormalities are strongly related to hypertension, brain infarcts, and vascular disease outside of the brain, including increasing carotid intima media thickness (IMT) and stenosis (49,50). The importance of these vascular abnormalities, especially in association with AD pathology, remains controversial (51,52).
High white matter grade (WMG) increases with age and is strongly related to increased ventricle size, a possible measure of brain atrophy. Both high white-matter lesion load and ventricle size are independent risk factors for dementia, including AD as well as "vascular dementia" (46). AD patients have more severe stenosis and atherosclerotic pathology in the Circle of Willis. Elevated blood pressure (BP) is a primary determinant of the vascular lesions in the Circle of Willis (53).
Midlife hypertension is a risk factor for subsequent dementia. There is, however, little evidence that antihypertensive therapy, except for one clinical trial (SYS-Euro) (54–57), reduces risk of dementia or prevents cognitive decline. Studies have not, however, included MRI measurement at baseline and have focused on older individuals who already may have extensive vascular disease in the brain and secondary neuronal loss and would be less likely to benefit from antihypertensive therapy. Also, it is possible that specific antihypertensive drugs may have a more beneficial effect than others.
There is probably enough evidence to focus a major emphasis on the vascular hypothesis as a primary determinant of AD and vascular-related dementias. Small-vessel vascular disease in the brain secondary to elevated BP and vascular stiffness may lead to ischemic injury due to inflammation and oxidative stress and a greater production and accumulation of amyloidgenic Aß, neurofibrillary tangles, and neuronal death (58). The effects of hypertension on the kidney and decreased renal function may reduce the clearance of Aß through the kidney and further increase brain amyloid plaques. Furthermore, the effects of BP and decreased renal function may also slow the clearance of other toxic chemicals from the brain which may be related to the development of AD and dementia. Also, susceptibility to vascular injury (i.e., inflammation) probably has a major genetic component (59). Prevention of increasing SBP with age by diet and drug therapy could, therefore, substantially reduce the risk of AD and overall dementia. The vascular hypothesis is particularly appealing because there are treatments available for vascular disease which could have major effects on the risk of dementia.
A potential interrelated hypothesis is that vascular disease in the brain is associated with increased inflammation and that the inflammatory changes contribute to an increase in both APP production and increasing Aß plaques as well as to changes in the confirmation of the Aß leading to fibrillar plaques and neuronal loss. Therefore, anti-inflammatory agents alone or in combination with antihypertensive therapy may reduce the risk of AD (60). Similarly, increased oxidative stress secondary to the small-vessel disease could modulate the adverse effect of hypertension and small-vessel disease. To date, antioxidants have not been beneficial in the treatment of AD.
The critical question, therefore, is how to develop research for the future that will test these evolving hypotheses. The following steps need to be taken:
First, clinical trials are the gold standard for testing etiological hypotheses. Further evaluation of the effects of antihypertensive agents on cognition and incidence dementia, but also on the morphological changes in the brain, are badly needed. These studies must include good cognitive measurements but also brain morphology and perhaps functional studies. Any new trials involving antihypertensive agents should probably include a major effort to study brain morphology, function, amyloid deposition, as well as cognitive function.
A second approach would be to evaluate pathophysiological and structural changes on MRI and in middle-aged individuals, including determinants of localized cerebral blood flow in relation to structural abnormalities in the brain and the effects of antihypertensive therapy or progression of white-matter lesion load and global and focal gray matter loss. These studies should be linked to neuropsychological evaluation to detect subtle changes in cognitive function.
Third, careful studies of geographic variations in the incidence of dementia that include brain imaging are badly needed (61,62). A key hypothesis will be to determine whether populations with lower prevalence of elevated BP (i.e., lower prevalence of obesity, lower salt intake) have decreased incidence of dementia and differences on MRI and other imaging techniques that would be consistent with lower prevalence of both vascular changes in the brain and AD pathology.
The current geographic variation studies in AD are primarily limited to the estimation of prevalence across populations. Such studies have generated some interesting observations, especially the apparent low prevalence in India (63), China (62), and Africa (64). A much more important and fundamental question is whether there are substantial variations in brain morphology, deposition of amyloid, or functional changes as measured by MRI and PET (65,66) across defined geographic populations with significantly different lifestyles, BP levels, and vascular stiffness.
Fourth, epidemiological studies also have noted that the prevalence of dementia increases dramatically with age but that there are many individuals 85 years old or older who have relatively normal cognitive function. These individuals provide a unique opportunity to study some of the factors, that is, BP, and absence of vascular disease outside of the brain. Is the absence of dementia or cognitive decline a function of general "good" aging or unique to the brain? For example, a small number, approximately 25% (mostly women), of the population have very little subclinical vascular disease or even clinical cardiovascular disease (CVD), at least up until age 80–90. There is some suggestion that such individuals also have low white-matter lesion load and small ventricular size (67–69). We and others have shown that individuals with low white-matter lesion load and small ventricular size have increased longevity. It is possible that lower SBP and less vascular stiffness over their lifetime has contributed to less vascular disease in the brain and periphery, better renal function, greater longevity, and better cognition. Do these individuals have less focal brain atrophy and fewer amyloid depositions or do they have extensive vascular disease and changes consistent with AD pathology yet are cognitively intact and have high brain reserve? Such studies would clearly require collaboration across sites and good measures of BP, vascular function, measures of aging, CVD, brain imaging, and functional status. These studies should also be linked with genetic analysis to link gene–environment interrelationships (70).
Summary
We must move beyond traditional descriptive studies. There is little need for new questionnaires or continued focus on classification of types of dementia based on clinical characteristics. Clinical and epidemiological research in dementia must move to analytical study designs that test specific hypotheses of etiology, identification of genes, risk factors, environmental interactions, and clinical trials that focus on prevention. Is there a major single agent that causes AD? Is AD pathology secondary in most individuals without major gene abnormalities to small-vessel disease, ischemia, oxidative stress, and inflammation? Is there an increase of toxic chemicals in brain with aging and possibly reduced renal clearance due to aging, vascular disease, and diabetes? Do individuals who have a low-calorie, low-fat, and low-salt diet (i.e., plant-based diets) have low risk of dementia, AD, and vascular disease? Is there a yet unidentified virus that causes inflammation, increased amyloid deposition, and dementia? Is brain aging part of systemic aging and, if so, can the aging process be slowed or "prevented" by specific lifestyle modifications? The hypertension, vascular hypothesis is most appealing because of good epidemiological and clinical support, availability of therapies, and reasonably good measurement tools. If the hypothesis was correct, that vascular disease in the brain secondary to elevated BP is a major player for most dementia including AD, then the potential implications for prevention of dementia are far greater than any of the current approaches.
The testing of etiological hypotheses of AD and dementia with a focus on prevention bringing together good epidemiological study designs and evolving imaging and metabolic studies are needed. Longitudinal studies, including good imaging studies and testing of specific well-defined hypotheses that lead to clinical trials, must be the direction of future dementia research in humans. Clinical trials will be required to test hypotheses. They will present interesting challenges given that drug treatments for hypertension and hyperlipidemia are established for treating CVD. Placebo-controlled trials of antihypertensive or lipid-lowering drugs may be difficult to do, especially among older individuals.
*NOTE
Please see the Editors' Note in the March 2006 issue of the Journal (p. 259) for a description of the article type "Green Banana."
F
Decision Editor: Luigi Ferrucci, MD, PhD
Received January 29, 2006
Accepted April 2, 2006
References
This article has been cited by other articles:
 |
|
 |
|
  H. H. Dodge, Y. Kita, H. Takechi, T. Hayakawa, M. Ganguli, and H. Ueshima Healthy Cognitive Aging and Leisure Activities Among the Oldest Old in Japan: Takashima Study J. Gerontol. A Biol. Sci. Med. Sci., November 1, 2008; 63(11): 1193 - 1200. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
