Alzheimer's Disease: What Is Known About Delaying Or Preventing Its Onset And Progression?
by William R. Ware, PhDAlzheimer's disease (AD) is one of a number of forms of dementia. According to the commonly used definition, dementia requires an irreversible, global impairment of cognitive skills including memory. The impairment must be great enough to interfere with the activities of daily living (1). Common characteristics of dementia include severe memory loss, inability to formulate abstract thoughts, confusion, problems with concentrating, difficulty carrying out both routine and complex tasks, personality changes and paranoid or bizarre behavior (2). These characteristics are the result of abnormal brain processes, not age. Up until about 1975 the term Alzheimer's disease or Alzheimer's dementia was reserved for the rare, early onset (40-60 years old) cases of cognitive loss, now known to generally result from a strong genetic predisposition. In fact, the demented patient cared for by Alois Alzheimer in 1901 in Germany was only 51, died in 1905, and the pathology exhibited by her brain was subsequently described and termed "Alzheimer's disease." Beyond the age of 65 loss of cognitive ability was traditionally viewed as a more or less inevitable consequence of aging (1). When it was recognized that presenile and many senile dementias shared a common pathology, dementias in later life finally became a recognized "disease." While conventional wisdom held that problems caused by normal aging were unavoidable, a true disease theoretically should be subject to both treatment and prevention. This had the highly significant consequence of opening the door for research funding (3). Considering AD as a disease had and still has strong political implications, and in the last decade there has been a virtual explosion in both fundamental and clinical research.
Harrison's Manual of Medicine, 15th Edition, lists five common causes of dementia and nine additional causes among adults. Thus the physician, when confronted with a patient presenting with signs of dementia, clearly faces a daunting task of differential diagnosis, made especially critical since some causes of dementia are treatable and even curable. It is in fact easy to argue that sooner rather than later specialized assessment from a neurologist or a specialist in neuropsychiatry, or both, is highly desirable, not only to avoid the calamity of leaving untreated a curable mental disease but also to identify as early as possible the cause of the dementia in order that therapeutic and delaying tactics can be initiated, even if the underlying disease is in fact judged incurable. A complete medical work-up is also needed to rule out potentially treatable causes of dementia such as encephalopathy, intracranial mass (e.g. tumor), infections that cause dementia, endocrine problems such as an under active thyroid, over-medication, metabolic problems, etc. Depression and delirium can mimic dementia, and are also treatable (2).
AD and vascular dementia (VaD) and a mixture of the two (mixed dementia) appear to be the most common forms, and this review will be restricted to these. The prevalence of dementia in community dwelling individuals over the age of 65 is estimated to be from 6-8%, whereas for those in nursing homes and chronic care facilities, the number can be as high as 75%. Prevalence of AD doubles approximately every 5 years after age 60. It is estimated that at least 30% of the US population over 85 has AD. While disease progression can be delayed, and in a few cases to be discussed below, dramatic reversals have been achieved, the normal course is a steady decline in almost all aspects associated with the "quality of life." For example, the inability to find the way home may be followed by the inability to fine some room in a home occupied for, perhaps, decades. At this stage there are frequently serious disruptive psychological problems as the individual understandably fails to cope with the disability. Resultant behavioral problems may require aggressive intervention with behavior modifying drugs in order for the individual to continue to live at home. Profound depression is also very common. Eventually there is the loss of control of bodily functions, especially urinary control, the loss of the ability to recognize family members, and finally total incapacitation. The average time from diagnosis of AD to death depends on age. Researchers from Johns Hopkins University found that patients diagnosed with AD in their 60s or 70s have a median remaining lifespan of 7-10 years, whereas those diagnosed in their 90s have a median remaining lifespan of only approximately 3 years (4). Studies suggest that dementia is an underlying but not necessarily primary cause of death. In fact, some studies find that the immediate causes of death are similar in demented and non-demented hospitalized elderly (5). However, individuals with VaD have an increased risk of death due to cardiovascular disease and in particular heart failure when compared to patients with AD (5).
It is common for individuals to seek out professional advice before the symptoms of AD or dementia are present. Generally the complaint relates to memory problems that have become bothersome or have caused as spouse or family member to suggest evaluation. This is covered below under the heading "Mild Cognitive Impairment."
The term Alzheimer's disease, while widely used, has a number of different connotations and modern usage has attempted to clarify the exact meaning. Thus today it is common to see the following definitions (6):
Because the neuropathological or histopathological manifestations of AD can occur without producing dementia, it has been suggested (1) that the terms "Alzheimer's disease" and "Alzheimer's dementia" should be used, the former referring to the presence of the neuropathologic characteristics and the latter requiring both the neuropathology and dementia. Both terms obviously require post mortem evidence.
Criteria for vascular dementia are also categorized by "definite" and "probable" as follows (6):
It should be noted that neuroimaging (MRI, PET, or single photon emission CT) is becoming a particular valuable tool for the differential diagnosis of dementia (7, 8). Clinical features also allow to some extent the differentiation of VaD and AD (6). VaD can evolve in a stepwise fashion following transient ischemic attacks that are really mini-strokes. They may be accompanied by various frequently vague and transitory symptoms such as light-headedness, dizziness, visual disturbance, headache, or mild tingling of the extremities (9). Also, the development of atherosclerosis over many years can slowly decrease the blood supply to the brain with associated cell damage and cell death due to oxygen and glucose deprivation, resulting eventually in VaD. The dementia can also occur after a sudden, paralyzing stroke. In some cases, post-stroke dementia can be prevented by appropriate treatment (9)
As the population age distribution shifts to older individuals, it is clear that a huge personal and financial burden is looming. While there is a high level of interest among the general public regarding the results and promise of anti-aging research, life extension will be a mixed blessing unless the problem of dementia is solved. It has been estimated that AD afflicts about 15 million people worldwide with about 4 million in the US alone (10). In the absence of a cure or a successful preventative protocol, it is estimated that 14 million Americans will develop AD by 2050 (11). It thus comes as no surprise that one of the most active areas of medical research today, both from the clinical and basic science perspective, involves finding the causes and devising treatments for AD. A literature search using the key word "Alzheimer" brings up about 3000 medical journal references just for the year 2002!
WHAT IS THE CAUSE OF ALZHEIMER'S DISEASE?
The conventional wisdom regarding the cause of AD focuses on what is termed the Amyloid Cascade Hypothesis (ACH). In this model, something goes wrong with the metabolism associated with so-called amyloid precursor protein (APP), and the fragments derived from it. These fragments, called amyloid beta or A-beta for short, then exhibit an abnormal tendency to aggregate yielding the so-called senile plaques that are seen in AD brains at autopsy. At the same time, something goes wrong with another protein called tau, which leads to the neurofibrillary tangles also seen in the AD brain.
The neuron looks something like an octopus with branches (dendrites) protruding from the central body, one of which, called the axon, is much longer than the others. The axon makes contact and communicates with other neurons through contact (synapses) to the dendrite of another neuron. The AD tangles form in the axon, blocking the flow of nutrients to the nerve endings and interrupting communications inside the cell and eventually killing it. The amyloid plaque that forms at the axon-dendrite contact points inhibits communication and induces an inflammatory reaction, and eventually killing healthy synapses and cells around the plaque. Thus the ACH regards A-beta and the plaques as neurotoxic and responsible in part for the death of neurons and the loss of cognitive ability. In the earliest clinical phase, AD is characterized by an almost pure memory impairment, which is thought to be due to synaptic dysfunction caused by A-beta(14). What actually triggers these problems with APP, A-beta and tau does not seem very clear, nor is there proof that these plaques and tangles directly cause the clinical symptoms of the disease. There appears to be general agreement however, that the dementia is due to the cumulative effect of the death of neurons and the failure of signaling pathways. Some believe that oxidative stress produced by an excess of highly reactive free radicals may play a role in the initiation process. Other theories view inflammation as a trigger or factor, and as well, restricted cerebral blood flow is a frequently cited causative factor. Toxins are also implicated. There are those who believe that both neurotoxicity and neuroprotective process are at work with tau and A-beta and that A-beta may actually be neuroprotective in some of its forms and at some stage in it production (15).
The ACH has led to potential therapies. A vaccine for A-beta which causes the break-up of senile plaques and the clearance of A-beta in mice has been developed by a pharmaceutical company. Human trials were recently halted due to the incidence of a serious inflammatory disorder of the central nervous system observed in a few participants in the trial, a halt that came before any changes in clinical symptoms could be evaluated with statistical significance. If the vaccine had reduced the plaque burden and resulted in clinical improvement, this would have gone a long way toward establishing the merits of the ACH. But this is not to be, at least in the foreseeable future. Pharmaceutical companies are also working on inhibitors for the enzymes that chop off the A-beta fragments from the APP, but nobody knows what the overall effects of this type of inhibition might be, since these enzymes have other actions in human biochemistry. The Amyloid Cascade Hypothesis has had a big impact on the direction of therapeutic research, in fact it has dominated it. If the hypothesis is flawed or seriously incomplete, as some think, then considerable time and vast sums of money have been spent, both in academic and pharmaceutical settings, going down only one road, so to speak, at the expense of considering and developing therapeutic alternatives not based on the ACH.
In the view of some neuroscientists the causes and the mechanism of progression of AD are much more complex than is implied by the Amyloid Cascade Hypothesis. Considerable attention is now focused on the significance of the large variety of vascular abnormalities that are also frequently present in the AD brain. In addition, there is the fundamental question of the relationship between the ultimate causes of AD and the many types of cerebral vascular disease. In fact, it has been recognized for some time that AD and vascular disease share many epidemiologic risk factors (16, 17). In a paper recently published in the journal Stroke, de la Torre (18) provides an interesting list of reported risk factors for AD from epidemiologic studies on elderly subjects. Included are atherosclerosis, stroke, diabetes, smoking, high cholesterol, cardiac disease, high serum viscosity, thrombogenic factors, high serum homocysteine, hypertension and hypotension, high fibrinogen levels, and transient ischemic attacks, all of which point to the involvement of vascular disorders. His thesis is that AD, when not caused by a strong genetic predisposition (the early onset type of AD), is initiated by vascular factors that precede the neurodegenerative process. This conclusion he claims is consistent with most of the basic and clinicopathologic data reported for AD, and is not inconsistent with other findings which indicate neurodegenerative processes as the cause of this disorder (18, 19). It has in fact been proposed recently that VaD is the most common form of dementia in the elderly (20), but some studies do not find this to be the case. A detailed discussion of vascular risk factors for AD from an epidemiologic perspective can also be found in a recent review by Breteler (21).
While clinical diagnostic protocols attempt to differentiate AD and VaD (6), and pure AD requires by definition the absence of vascular disease deemed serious enough to cause dementia, it seems clear (17) that in fact not only do AD and VaD share many of the same clinical signs and symptoms, but that the pure forms may represent two extremes of a continuum of pathology. If one looks at the prevalence of common pathological lesions in AD and VaD, many are frequently seen together at significant levels of incidence. Studies suggesting that the biochemistry that leads to plaque formation can be induced in part by vascular disease further justifies the continuum notion by extending it to factors that may be closely related to the ultimate causes (18, 19). de La Torre (22) actually would like to see AD classified as a vascular disorder. He considers that this would likely improve the chances of finding useful treatments because clinical studies could focus on more realistic and relevant pathologic targets than at present.
A recent article in The New England Journal of Medicine (23) underscores the connection between vascular disease and dementia. The study deals with the risk of dementia and cognitive decline associated with so-called silent brain infarcts, which in this context were mini-strokes that went unnoticed by the individual. The effects were seen with MRI imaging, identified if present at baseline, and followed for a number of years with repeated imaging. Neuropsychological assessment was used to establish the presence of dementia and cognitive decline. A significant finding was that silent brain infarcts increased not only the risk of dementia, but also that in this study the majority of cases that occurred were of the Alzheimer subtype. In an accompanying editorial in the same issue, Blass and Ratan point out that such studies provide guidance in intervention because recommendations can be based on the large body of well documented evidence concerning the prevention of vascular disease (24).
Further evidence for the view that the pathological basis for AD has many unresolved issues was recently presented by Vagnucci and Li in the journal Lancet (13). They present a fascinating table that summarizes the relative risk reduction of AD found in high-risk populations resulting from the use of a number of different drugs, including NSAIDs, steroids, histamine H2 blockers (such as Pepcid AC, or Zantac), calcium channel blockers, and lipid lowering agents (statins). Relative risk reductions as low as 0.14 were found for H2 blockers, with most relative risk numbers well below 0.5. These are impressive risk reductions by any standard, and yet the drugs obviously had quite different primary actions, as defined by the indications normally calling for their use (abnormal blood lipid profile, hypertension, heart burn, inflammation from arthritis, etc.). Vagnucci and Li suggest that the connecting factor for the action of this set of drugs, while not excluding other actions, is antiangiogenic activity (inhibition of neovascularization). In their view, AD in part involves the activation of large populations of endothelial cells by angiogenesis due to inflammation and oxygen deprivation. These cells are thought to be involved in the secretion of precursor substrate for A-beta and a neurotoxic peptide that selectively kills neurons. They advance the hypothesis that the drugs listed above have antiangiogenesis actions which inhibit this mechanism of neurotoxicity. Considerable evidence is presented for this hypothesis, which may eventually stimulate drug design and clinical trials. The point, however, is how little is really know about the details of what appear to be the highly complex pathological phenomena that ultimately cause AD.
MILD COGNITIVE IMPAIRMENT - AN EARLY WARNING SIGNAL?
There are a number of causes of (MCI) other than a pre-Alzheimer disorder. These include depression, mini-strokes, alcoholism, vitamin deficiency, low thyroid levels, overmedication, and sleep problems (27). According to Dr. Majid Fotuhi of Harvard Medical School, the majority of people who have memory problems and fear that they have AD may be experiencing the symptoms of depression and are in fact surprised to learn that depression can be viewed as a brain disease (27). Depression is treatable and patients can regain both their memory and a good outlook on life. Just as in the case of AD, it is critical to exclude these other causes of MCI when evaluating a patient with a memory complaint.
In recent years growing attention has been given to the possibility that memory problems of the elderly, both self-reported and observed by spouses or relatives, may be a useful and reliable predictor of the future risk of developing dementia, including AD. The central problem with studies directed at this important question has to do with the definition and diagnosis of MCI. The perception of memory problems differs considerably from individual to individual, and may not always be consistent with that noticed by an observer. In addition, some studies include in the definition of MCI one other aspect of cognitive impairment, so on the one hand there is the so called amnesic MCI, which is just a memory abnormality, and on the other a combination of amnesic MCI and an additional cognitive deficit. An important aspect of all definitions of MCI is the absence of dementia. Sometimes, the assessment is made with a battery of tests, sometimes just with a history and questions. A distinguishing feature of MCI as compared to age-associated memory impairment (normal) is that individuals with the latter do not get worse rapidly - they frequently complain of the same degree of memory problems for a number of years. However, this is not clear cut, and some individuals thought to have MCI actually get better or do not progress to AD.
Evidence from a number of studies does indeed suggest that the presence of MCI significantly increases the probability of developing AD. The increased risk depends on just how MCI is defined, and how accurate the definition matches the patient. It is unclear whether amnesic MCI or MCI with an added cognitive impairment has a greater risk of progressing to AD, although the evidence is in favor of a greater risk for the latter. Recent studies indicate that persons with MCI have as high as a 50% probability of developing AD sometime in the 4 years following the initial diagnosis (26). One study found an annual conversion rate of 8.3%, which is about 38% over four years (28). A study organized by the Mayo Clinic found an annual rate of conversion of 10-12% per year for amnesic MCI patients characterized by a memory complaint abnormal for their age, but with normal activities of daily living, normal general cognitive function, and the absence of dementia (29). In normal subjects one expects less that 2% conversion per year for the group in question. While studies yielding smaller conversion rates than 8-12% per year can be found, it is nevertheless clear that MCI, even when defined to involve just a memory problem, indicates a potential risk that is significant.
Studies have been reported where individuals classified as having MCI but no dementia have for some reason died and their brains have been examined. There are also cases where the brains of individuals free of any signs of MCI have been examined for evidence of AD pathology. What is perhaps surprising is that there were a number of individuals in these two groups where the pathology of AD was clearly and abundantly evident, and yet they showed no signs of dementia (30, 31). In other words, the pathology of AD can presumably develop, in some cases extensively, while the individual remains asymptomatic or develops only MCI. This may be related to the rather short interval between patients presenting with MCI and the development of clear clinical evidence for probable AD, at least for those in which this is indeed the sequence of events (32). In other words, the AD pathology appears to develop over an extended period prior to the MCI diagnosis. Also, in other studies of groups classified as having MCI, a significant percentage have, after several years, reverted to normal, for no apparent reason, even though progression to AD was seen in other members of the cohort (28, 33). Some researchers (32) like to view these early stages as a continuum, where at the beginning there can be AD pathology, or perhaps vascular pathology, or both, and later on MCI can develop, followed by the development of the clinical symptoms of AD. But up to some point the process or processes may either slow, halt, or even reverse without intervention. This broad view is very important because symptoms of MCI, especially amnesic MCI, need not necessarily mean that AD is inevitable.
AGE. There is no doubt that after 65 the risk of AD, VaD or mixed dementia increases dramatically. Advancing age should provoke concern regarding prevention.
GENETICS. A risk factor that obviously cannot be changed. There are in fact a number of genetic mutations that appear to favor AD, and in particular early onset AD. The genetic characteristic described by having the APOE allele e4 (an allele is any one of a series of two or more different genes that may occupy the same locus on a specific chromosome) is implicated in sporadic or late onset AD. Knowledge of the presence of genetic risk factors might lead one to be more aggressive in taking preventive measures. However, current medical practice (27) does not recommend genetic testing. There can be serious psychological issues involved as well as insurance problems (especially in the US). In some cases, an indication of genetic risk is obtained from the presence of early onset dementia in a family.
HEAD INJURY. Well characterized among boxers, where the end result is known as Dementia Pugilistica. Head injury with and without the loss of consciousness has in general been found to contribute to the development of AD (34, 35), with the loss of consciousness associated with the highest risk. There is some evidence that head injuries in young men are associated with AD and other dementias in later life.
OXIDATIVE STRESS, VITAMIN DEFICIENCY, POOR DIET, AND LOW ANTIOXIDENT LEVELS. A number of studies suggest that oxidative stress may play a role in the pathogenesis of AD (36). Lesions are found in AD brains that are typically associated with exposure to free radicals (37). Free radicals, which are highly reactive molecules, inflict damage on cell components and alter chemical components and reactions of biochemical systems, frequently adversely. Also higher intake or higher serum concentrations of some antioxidants are related to reduced risk of AD or cognitive impairment. Antioxidants studied include vitamin-E, vitamin-C, vitamin-A and those found in red wine (38). Martin and Mayer have recently reviewed studies concerning vitamin E and C and the reduction of the risk of AD (39). This will be discussed in more detail below. The situation with the B vitamins (see below for more details) is more complex because of the relationship with serum levels of homocysteine, an independent risk factor for AD (40). As one ages, the ability to utilize the vitamin content of food may decrease. The nutritional status of the elderly (and others, for that matter) is also frequently poor due to an inadequate diet, which is a common problem among the elderly (41). Related to both of these factors are the commonly observed low levels of antioxidants in the elderly which have been associated with increased risk of dementia (42).
ATHEROSCLEROSIS, EXISTING CARDIOVASCULAR DISEASE, TENDENCY FOR THROMBOSIS. One of the arguments for a vascular component in the etiology of AD and MCI is the epidemiologic observation that the presence of cardiovascular disease (CVD) is a risk factor (18). Also, the presence of atherosclerosis is not only a risk factor for CVD but also for AD (43). Thus individuals having or being at risk of having CVD should be particular aggressive in attempting to decrease or eliminate the CVD risk factors. Factors include hypertension, diabetes, metabolic syndrome (also called Syndrome X), being overweight, having a bad blood lipid profile, and having high C-reactive protein levels.
INSULIN RESISTANCE AND TYPE 2 DIABETES. Both are well known risk factors for CVD and thus, if one believes the connection between AD and vascular disease, they automatically become risk factors for AD and VaD. In addition, there is considerable evidence concerning the effect of insulin and glucose levels on brain health, and there is also independent evidence, based on a number of studies, some with very large cohorts, that type 2 diabetes is a risk factor for cognitive impairment, AD and VaD (44-49). Impaired glucose tolerance is also implicated in poor memory performance in the elderly. (48). Chronically low levels of blood glucose can result in permanent damage to brain cells (19). Chronically high levels of blood glucose imply diabetes.
SERUM CHOLESTEROL. Cholesterol has become the focus of intense interest in connection with MCI and AD, partly due to the recognition that the APOE allele e4, a known AD risk factor, is involved in cholesterol metabolism (50), and partly because of the epidemiologic evidence that cholesterol lowering drugs decrease the risk of AD (51, 52). Fundamental questions regarding the pathogenesis of AD include how nontoxic, soluble A-beta is converted into its toxic, aggregated form and how the protein tau is hyperphosphorylated to form neurofibrillary tangles. Growing evidence suggests that altered neuronal cholesterol metabolism may be involved in these pathological processes (50). There is also evidence suggesting that the development of late-life AD or MCI is associated with mid-life hypercholesterolemia (high cholesterol levels) (53, 54). The biochemical and pathological details are far from understood, and while the conventional wisdom holds that cholesterol does not cross the blood-brain barrier, there remains the observation that high serum cholesterol is a risk factor for AD (55). In fact, it is thought that brain cholesterol is made in the brain and not derived from the peripheral circulation. Not only statins but other cholesterol lowering drugs appear to decrease the risk of AD, and there are large differences among statins themselves as to their ability to penetrate the blood-brain-barrier (55). Thus, in spite of what appears to be a very primitive understanding of the role and biochemical mechanisms linking cholesterol and dementia, it would appear wise to treat elevated serum cholesterol levels as a significant risk factor.
ELEVATED HOMOCYSTEINE. Again, we see the connection between CVD and AD illustrated in the observation that elevated homocysteine levels are an independent risk factor for AD and dementia (40). This risk factor is closely related to vitamin B12 and folic acid deficiencies. It can be argued that every adult should have their homocysteine blood level measured, and this appears especially true for the elderly. For example, a level greater than 14 micromol/L nearly doubles the risk of AD in the elderly (40). An elevated level calls for aggressive treatment with B vitamins.
VITAMIN B 12 AND FOLIC ACID (FOLATE)DEFICIENCY. Over the years, a number of studies, but not all have indicated that deficiencies in vitamin-B12, vitamin-B6 and folic acid are related to elevated risk of dementia or AD (56-59). Several recent studies appear to confirm this. Wang et al (60) found in a longitudinal study that the risk increased by two-fold when a serum cut-off of B12 < 150 pmol/L and folate < 10 nmol/L was compared to normal levels of both vitamins. An interesting three-fold increase in risk was found for a subgroup with good baseline cognition, and a similar pattern was seen with higher cut-off of 250 pmol/L and 12 nmol/L. Similar results were found by Clarke et al (61) in a case-control study of total serum homocysteine, B 12 and folate and AD. For folate and B 12 serum levels the odds ratios obtained by comparing the upper with the lower third of the population were 3.3 and 4.3, respectively. Snowdon et al (62) found in the Nun Study that low serum folate was strongly associated with atrophy of the cerebral cortex, particularly in persons with AD lesions seen at autopsy. These results, when considered with a number of other studies (see (60) for references), suggest a strong link to AD risk from low B 12 and folate and high homocysteine, the latter, as discussed above, being also an independent risk factor for AD. However, mechanistic details are far from clear since there are a number of potential ways in which low levels of vitamin B 12 could, for example, lead to decreased cognition and AD (63), and in fact, there may be multiple pathogenic mechanisms at work (40, 64). Thus monitoring the serum B12 and folate concentration in the elderly appears important (60).
DEFICIENCY IN OMEGA-3 ESSENTIAL FATTY ACIDS. Recent studies (65, 66) have provided evidence that a deficiency in the omega-3 essential fatty acids may be linked to the risk of developing AD and other dementias. Correlations were found between risk and plasma levels of these fatty acids. One omega-3 fatty acid in particular, docosahexaenoic acid (DHA), appeared critical, and in another study a low level was not only characteristic of AD patients but also those with cognitive impairment (65). Tully et al used cholesteryl ester-docosahexaenoic acid levels to examine the relationship between DHA levels and AD and found similar results (67). In the Rotterdam Study it was found that low levels of fish consumption, a good measure of omega-3 fatty acid intake, was a risk factor for subsequent development of AD in the elderly (68).
In a recent article in Neurobiology of Aging, Yehuda et al (69) describe their recent work on the role of polyunsaturated fatty acids in restoring the aging neuronal membrane. They found that low omega-3 along with high omega-6 levels of fatty acids played a role in free radical damage in the brain, resulting in decreased fluidity of the neuronal membrane, and suggesting that dietary changes could perhaps overcome this problem. In their opinion, the dietary ratio omega-6 to omega-3 should be about 4, which is much lower than that found in most North Americans diets.
HYPOTENSION. A reduction in the blood supply to the brain (hypoperfusion) can adversely effect the supply of both oxygen and nutrients to the brain and lead to permanent damage. Closely related to hypoperfusion is hypotension (abnormally low blood pressure). The association of cerebral hypoperfusion and AD is well established, and recent observations suggest reduced cerebral blood flow is rather global in AD, without significant variation between different brain regions (70). In AD it has also been found that cerebral hypoprofusion is associated with cortical watershed microinfarcts (occlusion in the area where the blood supply from cerebral arteries overlaps at the extreme periphery of the vascular bed) which appear to further aggravate the degenerative process and impact adversely on the risk of dementia (70). Thus monitoring blood pressure and identifying arterial hypotension is very important. In older patients with heart failure, systolic hypotension is also associated with cognitive impairment (71).
HYPERTENSION. Studies concerning the connection between hypertension and dementia are inconsistent (72-74). The reasons appear complex (75) and may be associated with differences in study design, subject selection, etc. Nevertheless, there are a number of studies that find a connection between hypertension and impaired cognition in a broad range of areas (72). In addition, studies have shown a link between midlife hypertension and subsequent development of cognitive impairment (72, 76). It is also hard to ignore the results of intervention trials (77, 78) with antihypertensive drugs, where reductions in the incidence of dementia were observed. In one large, recent study (77), the reduction in the incidence of dementia was found to be over 50%! In this study, patients at entry had systolic pressure between 160 and 219 mm Hg, with diastolic below 95. Treatment consisted of a calcium channel blocker with the possible addition of an ACE inhibitor and/or a diuretic, and the follow-up lasted 2 years (blinded) and almost 4 years overall. The decreases in blood pressure, however, were not particularly large. The authors comment that calcium channel blockers may provide better protection against stroke than diuretics or beta-blockers, an effect seen in other studies. Again, we see a connection between the risk of AD and vascular disease. However, as pointed out above, these medications may have actions related to the risk of dementia that are independent of their effect on hypertension. Obviously, much remains to be clarified.
ALCOHOL ABUSE. The risk curve for the relationship between alcohol consumption and dementia appears to be J-shaped, with many studies indicating a protective effect of moderate consumption, slight risk with abstention, and a high risk associated with alcohol abuse (79-81). In the Copenhagen City Heart Study (82) red wine was found to be the effective drink. The same J-shaped relationship and protective effect is seen in connection with cardiovascular disease (83). Most studies indicate the optimum level of consumption to be one drink (e.g. a glass of wine) a day for women and two for men, but this action can naturally only be recommended if there is no history or risk of alcoholism (83).
ATRIAL FIBRILLATION. Atrial fibrillation is thought to be a vascular risk factor for dementia (84) and may impact the aging brain by sustained hypoperfusion (impaired blood supply). Ott et al (85) using data from the Rotterdam Study, found a significant positive association with both dementia and impaired cognitive function and the presence of atrial fibrillation (risk ratios of 2.3 and 1.7 respectively). The strongest association was with AD rather than VaD. Sabatini et al (86) report an even higher risk ratio of 3.4. Atrial fibrillation has also been identified as an independent determinant of low cognitive function (87).
LEADING DULL LIFE AND/OR A LOW LEVEL OF EDUCATION. This is called the "brain reserve hypothesis" and assumes that the more highly developed and active the brain, the more it is able to tolerate the degenerative processes that occur during aging and the development of MCI, AD and VaD. For example, three recent studies found that in elderly subjects cognitively stimulating activities resulted in a reduced risk of AD or an improvement in cognitive abilities (88-90). Friedland et al (91) also found similar results but comment that it is not clear if mental inactivity is a risk factor or a reflection of very early subclinical effects of AD. There is some evidence that the nature of an individual's lifetime principal occupation is related to the risk of AD (92). In this study, non-manual work was compared with manual or "goods production" work, with the latter yielding an increased relative risk of 1.6. In the famous Nun Study (93), Snowdon and associates found that low linguistic ability in early life (late-teens) correlated strongly with neuropathological assessment of AD at autopsy. They suggest that low linguistic ability may reflect suboptimal neurological and cognitive development, which might then increase the susceptibility to AD pathology in late life. While not all studies confirm the brain reserve hypothesis, and some studies lead to only modest associations (94), it can be argued that in order to 'play it safe" it may be wise to keep the brain as active as possible throughout life and in particular in later life.
ALUMINUM. The hypothesis that aluminum is associated with the risk of AD was proposed over 35 years ago and to this day remains highly controversial. The possible connection between aluminum in drinking water and AD has been the subject of a number of studies with inconsistent results. Recently Gauthier et al (95) proposed that one reason for the many conflicting results is that not enough attention has been paid to the chemical form of the dissolved aluminum. They found that there was a correlation between AD risk and monomeric organic aluminum but not total aluminum in drinking water. However, in a study published in 2000, Rondeau et al found a relative risk of almost 2 for drinking water containing greater than 0.1 mg/L (96). Newman (97) has advanced the hypothesis that an unfavorable ratio of essential fatty acids to total saturated fat plus trans fats may have an adverse effect on the permeability of the blood-brain-barrier, with the result that there can be excessive aluminum accumulation. There is considerable literature on proposed mechanisms of neurotoxicity of aluminum in the brain based on animal and cell culture studies (96, 98).
EMOTIONAL STRESS. The connection between stress and AD appears to properly belong in the relatively new field of psychoneuroimmunology. Dr. Gabor Mate has in fact just published a book, "When The Body Says No", which provides an extensive discussion of the connection between stress and disease, including AD, and a general introduction to psychoneuroimmunology (99). The mechanisms involved appear to be somewhat speculative, but essentially involve the action of stress on the immune system which generates inflammation with an inappropriate production of interlukin-6 and other proinflammatory cytokines. When this is chronic rather than a short-term "fight or flight" reaction, problems can arise. This chronic immune system dysfunction results in processes in the brain that may ultimately result in damage. It is significant that many markers of these processes are seen in the AD brain (100, 101). While some scientists like to refer to this as an autoimmune process, it has been suggested that a more appropriate term is autotoxicity. The proposed role of autotoxicity in AD has recently been reviewed by McGeer and McGeer (101).
MERCURY. Mercury has been suggested in the etiology of AD, and some individuals have had their mercury amalgams removed. Mercury is of course a well known neurotoxin, but the toxicity appears to depend on exposure and the chemical nature of the mercury. Normal environmental exposure is thought to result in blood levels well below those required for chronic or acute neurotoxicity (102). There appears to be a positive correlation between blood mercury levels and the risk of AD (102), and it is also clear that dental amalgams emit measurable quantities of mercury vapor (103). However, in a landmark paper Saxe et al (104) found no correlation between the number or size of dental amalgams and the mercury load of the brain nor was there any correlation between the incidence of very carefully diagnosed and documented AD and brain mercury levels. The authors conclude that their results do not support the hypothesis that low levels of mercury are a pathogenic factor in AD or that dental amalgams are a major health risk for AD. Nevertheless, there are a number of unanswered questions regarding the connection of mercury and AD, especially since there is an association between AD and serum mercury.
PESTICIDES AND HERBICIDES. Pesticides and herbicides have been implicated but the data is sparse and difficult to collect (105). In a study of a large cohort of Canadians, it was found that an increased risk of dementia was associated with residing in a rural area, as was occupational exposure to pesticides or fertilizers, which was suggested as the rural connection (106). Even the connection between trace pesticides or herbicides and cancer is controversial (107), but it is hard to argue against the wisdom of limiting exposure to any toxic chemical, just on general principles.
TREATMENTS PRESENTLY APPROVED BY THE FDA
AD is at present incurable, and thus only the treatment of symptoms, both cognitive and behavioral, represents current medical practice (2, 27). The cholinesterase inhibitors are the only class of drug that is currently approved (FDA) to treat AD and cognitive impairment. Motivation comes from the observation that a deficit of acetylcholine occurs in brains of AD patients. Acetylcholine is one of the principal chemical messengers in the body. After it is released into the space between nerves, it is quickly broken down by the enzyme cholinesterase. The cholinesterase inhibitors decrease the activity of this enzyme and are therefore thought to restore to some extent the concentration of acetylcholine in the AD brain.
The use of cholinesterase inhibitors in clinical practice has recently been reviewed by Cummings (108), and as well is discussed in recent books on AD (2, 27). The three commonly used drugs are donepezil (Aricept), rivastigmine (Exelon), and glantamine (Reminyl). They produce reproducible effects in patients with mild-to-moderate AD, with drug-placebo differences seen on global and cognitive measures. Their use is also associated with improvements in activities of daily living and behavior, a delay in nursing home admission, and there is some evidence that they remain effective over several years (108). They do not stop the progression of AD, but they delay the symptoms of cognitive decline. It is considered a success if a patient takes one of these drugs and after two years there has been no further decline (27). One of the most noticeable effects is on behavioral problems and attention (108). Side effects of cholinesterase inhibitors include nausea, vomiting, diarrhea and anorexia. Cholinesterase inhibitors have also been found to provide modest improvements in patients with VaD (108).
Antipsychotic, antianxiety and antidepressant drugs are frequently prescribed to treat behavioral symptoms. These drugs can increase the quality of life for both the patient and caregiver, but they also can have a number of side effects (2).
Some physicians use large doses of vitamin E along with cholinesterase inhibitors, and this appears to be accepted as standard medical practice (6). The justification is primarily from a paper in the April 24, 1997 issue of the New England Journal of Medicine, which reported that patients with moderate AD treated with high doses of vitamin E (2000 IU per day of dl-a-tocopherol) experienced on average a 7-month delay in the progression of the disease. This antioxidant delayed the loss of the ability to perform daily activities, the necessity of moving into a nursing home, and the progression to severe dementia, but cognition itself did not improve (109). Large doses of vitamin E should only be taken under a doctor's supervision because of possible side effects such as bleeding and GI problems (2). Studies are currently ongoing regarding the question of the use of cholinesterase inhibitors and/or vitamin E in the treatment of MCI, and the results are awaited with considerable interest.
OTHER APPROACHES TO DECREASING THE RISK OF VaD, AD AND MCI AND SLOWING THE PROGRESSION OF THESE DISEASES
This is perhaps the most exciting area currently under investigation because the question of risk reduction, i.e. primary prevention, is addressed.
REDUCING THE RISK OF VASCULAR DISEASE
Since vascular disease is implicated in the development of dementia, perhaps even starting in mid-life, reducing the associated risk factors should be given high priority, quite apart from the benefits associated with reducing the risk of heart attack or stroke (27). One obvious approach is to assess, with the assistance of one's physician, the status of the various CVD risk factors. These should include cholesterol and triglyceride levels, i.e. a complete blood lipid profile, blood pressure, weight, homocysteine levels, presence of insulin resistance or diabetes, smoking, and perhaps C-reactive protein and fibrinogen levels. While the direct connection between smoking and AD is controversial, it is nevertheless a risk factor for CVD and cancer. Action in general is indicated if any of the CVD risk factors are outside the normal range. If there is any question at all regarding insulin resistance, a glucose tolerance test seems indicated since fasting glucose alone does not reveal the whole picture. The most common glucose tolerance test involves a fasting glucose followed by a glucose challenge (a drink). Serum glucose is then measured two hours later.
There is an understandable reluctance on the part of many individuals to take prescription drugs unless it is absolutely necessary, and there appears to be great merit in a plan which first attempts to normalize CVD risk factors by aggressive lifestyle changes such as diet and exercise. Since being overweight or obese is very common in the developed world, the question of diet will be important for many individuals. Unfortunately, this requires confronting the controversy concerning low-carb vs. high-carb diets and the question of the role of dietary fat (83). Space does not permit a discussion of this topic, but the reader is referred to Walter Willett's new book (83) and the text and references of the three-part article in the IHN newsletters of November 2002 to January 2003, which is concerned with the relation of diet, and in particular dietary fat, and CVD.
SUPPLEMENTS AND OVER-THE-COUNTER DRUGS
VITAMINS E AND C. Vitamin-E is the only supplement (at 2000 IU/day) used by mainstream medicine to treat the progression of AD. However, there has been considerable research regarding the role of vitamin E in reducing the risk of AD and VaD in individuals initially free of dementia (39). The results are not consistent. Recent studies have had as endpoints AD, VaD and cognitive function, with most studies using only one of these outcomes. Both the use of supplements and the intake of dietary vitamin-E have been studied, and as well, the concurrent use of vitamin-C. In some studies vitamin E from foods was found effective, with no added influence from supplements. This is reminiscent of studies concerned with vitamin E and CHD which also have been inconsistent and where some studies found vitamin E from food rather than supplement sources to be protective.
Vitamin E consists of a number of forms, the most important of which appear to be alpha-tocopherol and gamma-tocopherol. In the North American diet, about 70% of the vitamin E from food sources is said to be gamma-tocopherol, supplements are generally d-alpha-tocopherol, or the synthetic "dl" form. One mg of natural vitamin E is equivalent to approximately 1.5 IU of d-alpha-tocopherol, whereas 1 mg of the synthetic form is approximately equivalent to 1 IU of dl-alpha-tocopherol. The alpha-tocopherol form is regarded as having much higher bioactivity as compared to gamma-tocopherol, although this may be an artifact of the measurement of activity. In addition, supplementation with alpha-tocopherol suppresses both plasma and tissue gamma-tocopherol. The alpha-form and gamma-form also do not have identical biological actions, and the gamma-form is thought to be a more potent anti-inflammatory agent than the alpha-form (110). The point of all these details is that food sources are different than supplements and that vitamin E supplements interact with and suppress the gamma-form from food. The net result is the potential for great confusion and inconsistent results unless these factors are taken into account. However, whether this is the reason why some studies find protective effects with food sources alone is still unclear.
The amount of vitamin E available from food in typically in the range of 5-15 IU which is significantly less than what is generally taken via specific supplements, i.e. 100-800 IU. It is thus fascinating that two recent studies (42, 111) found that vitamin E from food, especially at the high end of >15 IU/day, to be significantly protective for AD, with no added effect of supplements, although these studies did not rule out a role of supplements alone producing a positive result. On the other hand, Grodstein et al in a very recent study from Harvard (112), which was based on a cohort from the Nurses Study, found that the use of specific vitamin E supplements along with specific vitamin C supplements, both at high doses (E > 600 mg/day, C > 750 mg/day), for greater than 10 years was related to modest cognitive benefits in older women (70-79 y of age). For those taking vitamin E alone, the benefits appeared to be weaker and there was no evidence of a trend with duration of use. They also found little support for the effect of vitamin C supplementation alone on cognitive function. Morris et al (113) recently reported that when the endpoint was cognitive decline with age, vitamin E intake from foods or supplements was protective, but there was no evidence of an association with vitamin C. Studies of vitamin E or vitamin E plus vitamin C status in the plasma or cerebral spinal fluid have also found inverse correlations with the risk of loss of cognitive function (114, 115). Also, Masaki et al (116) found that supplementation with vitamin E and C provided protection against both VaD and cognitive decline, but E alone was not effective.
A number of other studies could be quoted, and while the majority comes out in favor of vitamin E, some find no statistically significant effect. However, it would appear that there is sufficient evidence in connection with reducing the risk of cognitive decline, VaD or AD to justify either supplementation with both E and C, or making sure that ones diet is rich in both. The problem is that it is not easy to get large amounts of E from food (83). Food sources include vegetable oils, whole grains, green leafy vegetables and nuts.
VITAMIN B12, FOLIC ACID AND VITAMIN B6. As discussed above, high serum levels of homocysteine as well as low levels of vitamin B12 and folic acid (folate) appear to be significant risk factors for AD, VaD, and cognitive decline. Thus, obtaining serum levels of homocysteine and B 12 as part of a risk assessment or physical exam would appear to be a reasonable step toward prevention. High levels of homocysteine are treated effectively with B 12, B 6 and folic acid. However, the interpretation of serum B 12 levels is not as simple. There appears to be a poor correlation between serum levels and tissue levels, so that serum levels can fail to reveal a deficiency (117). Also, studies that relate serum levels to neuropsychiatric abnormalities or AD have used a variety of cut-off values which go well into the conventional, normal serum range, which suggests that the conventional "normal" cut-off is too low. A realistic cut-off of 300 pg/ml has been suggested (117). In addition, anemia that can be caused by a B 12 deficiency does not in general coexist with the neuropsychiatric abnormalities or AD, and thus the absence of anemia is not relevant in this context, but its presence is, since it is well known that anemia increases the risk of dementia (117).
High homocysteine levels provide a fairly satisfactory marker for low B 12 and folate. Thus there are thus two simple actions indicated. First, if homocysteine levels are high, it seems reasonable that they should be treated aggressively until brought to normal. While many physicians would measure serum B 12 levels in patients presenting with cognitive or memory complaints, there is a problem when the results are "low normal," as well as a problem with above mentioned issue with tissue vs. serum levels. It can be argued that since it is cheap, simple and safe, supplementation with B 12 is indicated in these cases simply to see if it produces an improvement in clinical symptoms. In this context, a recent study by Abyad (117) is interesting in that he found the reversal of clinical symptoms among a group with cognitive dysfunction was most pronounced if they were treated early, with patients returning to normal if treated within three to six months, and poor results when treatment was initiated a year after the appearance of symptoms. This may explain why studies examining the reversal of dementia or cognitive impairment with B 12 therapy frequently yield poor results (118).
Both oral and intramuscular administration has been shown to increase serum levels (119). It would seem that the oral sublingual approach is the easiest to implement, although injectable B 12 is available without prescription in some jurisdictions. With the elderly, there can be a serious problem of absorption from oral supplementation, and in fact this may be one of the reasons for the deficiency in the first place. A typical oral dose of 2 mg/day (not sublingual) was found to be as effective as 1 mg intramuscularly per month (119). Some physicians start with more frequent injections. Multivitamins generally contain only very small amounts of B 12. Sublingual oral administration circumvents some of the absorption problems. Food sources of vitamin B 12 include milk products, meat, poultry, fish and spinach. Sources of folate include green leafy vegetables, beans and peas, grain products, tomatoes, oranges, beets, soybeans, fish and eggs.
OMEGA-3 ESSENTIAL FATTY ACIDS AND FISH OIL. The link between AD, VaD and CVD has been discussed. Epidemiologic studies suggest that omega-3 fatty acids reduce the risk of CVD by decreasing the risk of arrhythmias and thrombosis, decreasing triglyceride levels and the rate of growth of atherosclerotic plaque, improving endothelial function, and reducing inflammatory responses (120, 121). The American Heart Association (AHA) (120) recommends that all adults eat fish (in particular fatty fish) at least two times a week. For patients with high levels of triglycerides, the AHA recommends eicosapentaenoic and docosahexaenoic acids (EPA and DHA) supplements, e.g. fish oil, of two to four grams a day. Patients taking more than three grams of these fatty acids from supplements should do so only under the care of a physician. Dietary actions to reduce the risk of CVD have been discussed in detail in many publications (83, 122, 123).
There have been limited peer reviewed reports of intervention studies. Yehuda et al (124) used a mixture of omega-3 and omega-6 fatty acids in a ratio of 1:4 in a 4-week study of 100 AD patients and found improvements in mood, cooperation, appetite, sleep, short term memory and the ability to navigate in the home. Terano et al found a decrease in the severity of AD when subjects were supplemented for one year with 0.72 g DHA/day (125). There is also a case study reported that implicated Omega-3 fatty acids in the reversal of AD. In 1990 an Australian physician, Dr Robert Peers, reported (126) observing a remarkable reversal of AD in a nursing home patient. The patient went from being restless and destructive and unable to dress himself to being calmer, regaining weight and was again able to dress himself. It appeared the reason for this totally unexpected improvement was that fish was served every week, which provided omega-3 fatty acids that had been almost totally absent in the patient's diet for at least five years before he was diagnosed with AD.
There is now additional evidence for the merits of omega-3 fatty acids and in particular fish oil for the treatment of late AD. In the recent book The Omega Rx Zone (122), Barry Sears describes the work of Dr. Dan Ward, who owns and operates an extended care and rehabilitation center in Florida. Along with Dr. Sears he developed a protocol for treating terminal AD patients with very high doses of pharmaceutical grade fish oil - 25 g/day. Twenty-five grams per day of fish oil is indeed a very high dose, and it is important that the oil be free of dangerous contaminants. The fish oil was combined with a diet based on Sears' Zone principles (moderate amounts of complex carbohydrates, low fat protein and monosaturated fat) and fed using shakes that these patients could consume. The results were sensational. Some patients went from being bed ridden and totally unable to take care of themselves to eating normally, playing cards, recognizing friends and family, and even going home to lead more or less normal lives. Ward and Sears observed that the improvement was fish oil dose dependent, indicating that the omega-3 supplementation was probably responsible for the remarkable reversal of what is universally an irreversible decline until death. They also observed a positive, synergistic effect of the altered diet. This is of course the sort of anecdotal evidence that is held is great contempt by those dedicated to the principles of evidence based medicine, but it stretches the imagination to ascribe these results to a placebo effect and suggest that perhaps a proper double blind, placebo controlled study would yield different results. After all, these patients were barely aware of their surroundings! It would be like suggesting that animal studies be placebo controlled! The study of Yehuda et al (69) discussed above provides one possible mechanism for these observations. It remains to be seen if an increase in neuronal membrane fluidity and a restoring of the aging neuronal membrane could be responsible for the action of fish oil in the reversal of late-stage AD.
Thus following the AHA recommendations concerning omega-3 essential fatty acids for cardiovascular health should also have a positive effect on brain health. Food sources of omega-3 oils include fish, walnuts, flaxseeds, canola oil and so-called omega-3 eggs, where the chickens have been fed flaxseed. Fish oil provides the omega-3 fatty acids EPA and DHA, both of which can be made in the body from non-fish sources. However, the ability to synthesize these fatty acids from a-linolenic acid, the omega-3 fatty acid found in, for example, flax seed and nuts, may decline with age. Thus there may be a significant advantage associated with getting the required substances either from fish or fish oil. The pharmaceutical grade of fish oil presumably has much lower levels of impurities than usually found in the ordinary variety.
GINKGO BILOBA. An herbal medicine which appears to enhance blood circulation and has antioxidant properties. Ginkgo has been used extensively in China and Europe for treating a wide range of conditions including memory and concentration problems, confusion, depression, anxiety, tinnitus and headache as well as AD. The mechanisms of action are thought to involve increasing blood supply by dilating blood vessels, reducing blood viscosity, modifying neurotransmitter systems and reducing the levels of oxygen free radicals (127). There have been a number of recent studies, mostly double-blind, placebo-controlled and randomized, which examined the effectiveness of ginkgo on memory problems and delaying the onset of AD. Birks et al (127) have made a detailed and comprehensive review of all the studies that met their standards for inclusion. When clinical global improvement was used as a criterion, benefits from the use of ginkgo were found. Benefits to cognition were also found, as well as benefits associated with activities of daily living and mood and emotional function. They found no significant differences between ginkgo and placebo in the proportion of individuals experiencing adverse effects. Not included in this review was a study reported in 2003 (128) that examined the relationship between the use of ginkgo and other cerebral and peripheral vasotherapeutics and the onset of cognitive impairment. The results obtained after a 7-year follow-up indicated these treatments resulted in a significant drop in the risk of developing AD in elderly women. A recent paper in the JAMA reported negative results for a memory enhancement trial with normal, healthy subjects (129). This study has been criticized on a number of counts (see (130) and subsequent letters to the editor).
The cellular and molecular mechanisms of the neuroprotective actions of ginkgo still remain largely unproved. One interesting aspect was recently presented in the Proceedings of the National Academy of Science (131). This study indicated that ginkgo has an inhibitory effect of A-beta aggregation and as well influences an enzyme involved in the cell death signaling process, both of which could account for a neuroprotective effect.
Ginkgo preparations are of variable potency, and dose dependent studies are limited. Typical doses used in studies range from 40 to over 200mg per day (127). Ginkgo has been shown to increase the risk of bleeding in some people, and it can interact with aspirin, warfarin, and other medications. Thus it appears prudent to use it only under the supervision of a physician (27), especially when large doses are contemplated.
NON-STEROIDAL ANTI-INFLAMMATORY DRUGS. It has been known for some time that anti-inflammatory agents appear to provide protection from AD. This observation is directly related to the theory that AD is in part an inflammatory disease (132). This important question has been exhaustively reviewed by the Neuroinflammation Working Group (133), who provide strong arguments for the consideration of AD as an inflammatory disease, even though the relative contribution of inflammation in comparison to other mechanisms of neurodegeneration still remains unclear (132). Non-steroidal anti-inflammatory drugs have received considerable attention in epidemiologic studies related to prevention or delaying onset or progression of AD. Already by 1996, 17 studies had been published and reviewed (134) and a review in 2001 (135) discusses a total of 22 studies. The most recent review was published in 2003 (136). Typically, odds ratios in the range of 0.4 to 0.6 are found for risk reduction associated with NSAID use, although the study based on the Rotterdam Study cohort (12) found an amazing relative risk of developing AD of 0.2 for those taking NSAIDs for more than 24 months. No effect was observed with aspirin, in agreement with most but not all studies, but aspirin was positively associated with the risk of VaD. These studies involve conventional NSAIDs such as ibuprofen, naproxen, and diclofenac, which inhibit both COX-1 and COX-2 activity. It is significant that NSAIDs appear to only offer protection up to a point several years prior to the appearance of diagnosable dementia. Thus their utility appears to be in primary prevention in the early or latent stages of the disease. This is consistent with the fact that to date there are no trials showing a significant slowing of AD progression in groups of patients treated with an anti-inflammatory drug (135). Also, NSAIDs use does not appear to influence the risk of VaD (135). The new COX-2 inhibitors Celebrex and Vioxx are currently being tested for its effectiveness is primary and secondary prevention. It will be several years before the results are available.
Thus the critical question--should an individual start taking NSAIDs to lower the risk of developing AD? While this is no doubt a matter of opinion, Dr. Majid Fotuhi, Harvard professor and Neurology Consultant with the Alzheimer's Disease Research Center, Johns Hopkins Hospital answers "not yet" (27). If one has arthritis and takes NSAIDs anyway, then he points out that there is an added benefit. For anyone who does not need to take these drugs, and does not want to follow his recommendation of waiting for more trials, then consultation with a physician is suggested (27). It of course must be emphasized that this class of drug is well known to have the very serious side effect of inducing gastric bleeding, which is in fact quite common in the elderly taking NSAIDs, and can be serious or even life-threatening. Other side effects include ulcers and possible kidney problems (2). All this has to be balanced against impressive risk reductions in the context of primary prevention and the presence of 22+ studies already reported in the literature.
Unfortunately, it appears that there is very little data from long-term studies on low-dose intervention with NSAIDs where primary prevention of dementia is the endpoint. Nevertheless, there is some indication that for reducing the risk of AD, low doses are as effective as the high doses used to treat arthritic pain (135). Long-term studies of the use of low doses must still address the problem of side effects.
ESTROGEN. Studies regarding use of unopposed estrogen to reduce the risk of AD have been inconsistent and there is still no consensus. Under these circumstances, it has been recommended that women should probably not take estrogen for the sole purpose of delaying the onset of AD (2, 27). It is important to distinguish between unopposed estrogen treatment and so-called hormone replacement treatment (HRT) which involves both estrogen and progestin. Two recently reported studies (137, 138) have found that for postmenopausal women over the age of 65, HRT increased the incidence of dementia and had small adverse effect on global cognitive function. Effects on MCI did not differ between the treatment and placebo groups. These were large and significant studies, and the conclusion is that the risks of HRT outweigh the benefits, since there are also other negative (and recently much publicized) aspects of HRT associated with cardiovascular disease and cancer.
ALPHA-LIPOIC ACID & N-ACETYL-L-CYSTEINE (NAC). These two chemicals, available as supplements, have been suggested for use in primary prevention or treatment of AD because of their relationship to glutathione, a potent, highly important, and in fact essential endogenous antioxidant, and because they themselves have antioxidant properties as well as other potential actions. In fact, they are found to increase the levels of glutathione, something that taking oral glutathione directly cannot accomplish to a significant extent. It has been suggested (139) that NAC has potential in the prevention and treatment of age-related mitochondrial neurodegenerative disease. NAC has many other actions which are summarized by Atkins (140). a-lipoic acid is also a powerful antioxidant. Packer calls it a universal or superantioxidant and discusses its uses (141). He regards it as by far the best supplement for increasing glutathione levels, recommending 50 mg twice a day. It has been tested for slowing or arresting the decline of cognitive function in AD (142). Positive results were obtained but the study involved only nine patients. Berkson has presented the full and fascinating story of a-lipoic acid and its many uses in a recent book (143), although AD is not featured. Thus there do not appear to be sufficient studies to allow a judgment on the wisdom of supplementation with either of these two substances in the context of AD, although the reader might wish to review the general arguments provided in the books mentioned as to the potential benefits associated with these supplements. There do not appear to be any long-term studies addressing adverse effects, but a-lipoic acid is licensed in Germany for the treatment of diabetic neuropathy (142). NAC is generally accepted by mainstream medicine for the treatment of liver failure that may results from an acetaminophen (Tylenol) overdose (140).
OTHER ANTIOXIDANTS. If one accepts the theory that oxidative stress is an important risk factor for cognitive impairment, VaD and AD, then having a diet rich in antioxidants would obviously seem to be important. There are a number of antioxidants available from food or in supplemental form that are not included in the above discussion, such as Coenzyme Q 10, a large number of flavonoids and carotenoids, and selenium, which while not in itself an antioxidant, is an essential component in two important antioxidant related enzymes. Space does not permit a detailed discussion, and the reader is referred to Lester Packer's book The Antioxidant Miracle (141) for a comprehensive review and guidance on supplementation and doses. Packer has for many years been director of a laboratory devoted to antioxidant research at the University of California at Berkeley. His book contains a detailed discussion of many antioxidants and why they are important, even critical for general health. Attention is directed in particular to his theory regarding the so-called antioxidant network, consisting of vitamins C and E, coenzyme Q-10, a-lipoic acid and glutathione, which his research indicates involves critical synergism. Readers may also find the book on vita-nutrients by Atkins to be a valuable resource (140).
DIET. The risk factors discussed above suggest that a diet based on slowly digested carbohydrates, high fiber foods, whole grains, nuts, fruits and vegetables, fish and lean meat and red wine might provide protection against CVD and thus dementia (83, 144). Fruits, especially berries, and green, leafy vegetables provide a wide spectrum of antioxidants. Hu and Willett (123) have recently reviewed the most recent epidemiologic results concerning diet and the prevention of heart disease, which is relevant to the prevention of CVD in general and thus AD and VaD, and a general discussion is also available in Willett's recent book (83).
The possibility that AD is at least partly an inflammatory disease also raises diet issues. Barry Sears deals at length with this subject in The Omega Rx Zone (122). He emphasizes the importance of a proper dietary balance of the essential polyunsaturated fatty acids to achieve a non-inflammatory state. This means keeping the omega-3 and omega-6 fatty acids in a ratio that does not favor the production of inflammatory eicosanoids and the related inflammatory hormones. Yehuda (69, 124) has also raised this point. To accomplish this Sears recommends a diet that typically is 1/3 lean meat or fish, and 2/3 fruits and vegetables, with added monosaturated fat, for example from olive oil. To this diet he adds several grams daily of pharmaceutical grade fish oil, giving a diet that is rich in omega-3 essential fatty acids and has a good balance of these two essential fatty acids. Yehuda (69, 124) suggests that the omega-3 to omega-6 ratio should be in about 1 to 4, a value which is consistent with estimates of the ratio in the diets of Stone Age Man, whose genetic profile still by and large controls our metabolism today (122). The emphasis in Sears' program is also on slowly digested fruits and vegetables, since insulin control is also an issue. Bread, pasta, rice, potatoes, carrots and foods high in sugar or of high glycemic index are for example minimized, and trans-fat is avoided. Sears advances arguments that this diet also minimizes the risk of metabolic syndrome, a risk factor for CVD and thus AD, and may even reverse it. Interested readers should consult his book.
MENTAL ACTIVITY. Maintaining a high level of mental activity may be far from simple, especially if an individual's occupation is dull, repetitive, and offers little or no intellectual stimulation. The post-retirement period also can pose a serious problem, even for those who have had intellectually stimulating occupations, and this situation is aggravated by mandatory retirement. The retired mathematics professor who simply plays golf several times a week may view this as keeping active and a welcome respite from teaching, solving problems and perhaps doing research, but the potential decline in intellectual activity and stimulation could be dramatic. Activities that might help exercise the brain include playing contract or duplicate bridge, becoming active in a chess club, doing hard crossword puzzles, or learning to play a musical instrument, just to pick some obvious post retirement activities that offer more mental stimulation than watching TV or mowing the lawn. Taking courses in night school or in the adult education program of a university, or even online is also a possibility. The opportunity exists to learn a new language, or study some demanding topic such as logic, philosophy, mathematics, some aspects of computer science, etc. There are those who actually get a university degree or even a second degree after retirement just for the pleasure of doing it. Many academics have seen this personally and watched these individuals graduate at 70 or 75 or 80 years of age, clearly in possession of "all their marbles."
Presumably vascular damage is ongoing, probably from an early age, and therefore taking steps to reduce the risk of CVD is indicated for anyone, independent of the presence or absence of cognitive disease. It would appear sensible to give CVD prevention a very high priority if the concern is preventing cognitive impairment, AD or VaD.
MCI is not classified as a disease, and main stream medicine does not recognize any treatment (145). However, three classes of drugs and a few of supplements are currently undergoing large clinical trials to determine if they alleviate the memory problems or delay the progression of MCI to AD. Included are cholinesterase inhibitors, vitamin E, estrogen, COX-2 inhibitors, and ginkgo biloba. It will be several years before any results appear. These studies may fail to yield definitive results because it is necessary to demonstrate clinical improvement when only a moderate cognitive deficit is present, and as mentioned above, a considerable percentage of MCI patients remain stable or revert to normal without intervention. It may also turn out that effective preventive measures must be initiated years before clinical symptoms appear. This further complicates the design and execution of studies. The recommendation of mainstream medicine for those diagnosed with MCI is summed-up by the following quote (146): "Treatment (for MCI) should not be prescribed based on current speculation, but must await confirmation from well designed clinical trials." The impatient are advised to enroll in clinical trials (145), an option that must surely have very limited availability! This advice, while no doubt required under the rules of mainstream medicine, ignores supplements that appear of low risk and may offer benefit, but this is an area where everyone is "on their own." A number of the micro-nutrients discussed are already present in food, and modest increases brought about by supplementation may be associated with minimal risk. Larger doses should be taken only under medical supervision.
Studies addressing primary prevention of MCI, AD or VaD that combine a number of interventions, not just two such as we have seen in the case of vitamin C and E, have apparently never been undertaken and in fact may never occur. It has in fact recently been suggested that the use of multiple antioxidants may have considerable merit (36, 147). For example, if one starts in mid-life taking vitamins E, C, various other antioxidants such as flavonoids and carotenoids, and a-lipoic acid, the B vitamins, as well the omega-3 essential fatty acids EPA and DHA, all from modest supplementation in addition to what is obtained from food, and in addition eats a non-inflammatory heart-healthy diet and attempts to minimize all the CVD risk factors, and in addition perhaps takes low doses of NSAIDs, will this program decrease the risk of MCI or AD or VaD? It might be predicted on the basis of what is already known that the results could be sensational, but a study that covered this extensive a set of interventions, had a large cohort, was multi-center and lasted for 20-30 years, would seem totally unrealistic. Who is going to fund it since there are no prescription drugs involved? Besides, critics would say that if positive benefits were obtained, it would be impossible to determine which interventions were responsible. Also there would be complaints about the absence of blinding. Those accustomed to a more pragmatic way of viewing things might say, "So what if it works." However, it is possible that large, ongoing prospective studies may be able to address to some extent the question of multiple actions and lifestyle factors in connection with the risk of MCI, AD or VaD, but it remains to be seen if enough data and statistical power is present to provide guidance, and even if anyone is going to look for such answers.
Interventions that show some benefit for mild AD have been discussed above, but trials by and large have not been carried out that satisfy the requirements of evidence-based medicine. Thus, the only approved or tolerated treatments involve the cholinesterase inhibitors and high-dose vitamin E. At present there are no approved, effective treatments for advanced or end-stage AD aside from behavior modifying drugs. The merits of the intervention discussed above for end-stage AD, i.e. high doses of fish oil and a special diet, derive from anecdotal evidence described in a book many physicians would never read, either because they were unaware of its existence or just on general principles. Perhaps the Sears-Ward protocol will be subjected to a trial, although again there is nothing in it for the pharmaceutical industry. High or even moderate doses of fish oil might even be tried on mild or early AD.
Grounds for some optimism can be found in the huge amount of basic and clinical research currently ongoing which may result in new drugs based in part on a growing understanding of the fundamental causes and mechanisms of dementia, although the almost exclusive fixation on the Amyloid Cascade Hypothesis gives cause for concern. Modern imaging techniques also show great promise in providing both diagnosis and a much-needed means of evaluating potential therapies. Also, there is research on blood and urine markers which may lead to approved tests that could aid in differential diagnosis. However, as emphasized above, MCI, AD and VaD appear to be complex, multifactorial diseases which may start in mid-life, and one should not underestimate the challenges that lie ahead or expect cures or highly effective interventions to be just around the corner.