1 - Clinical Implications of the Aging Process | Understanding Thin Client/Server Computing (Strategic Technology Series)

Editors: Kane, Robert L.; Ouslander, Joseph G.; Abrass, Itamar B.

Title: Essentials of Clinical Geriatrics, 5th Edition

> Table of Contents > Part I - The Aging Patient and Geriatric Assessment > Chapter 1 - Clinical Implications of the Aging Process

Chapter 1

Clinical Implications of the Aging Process

The care of older patients differs from that of younger patients for a number of reasons. Some of these can be traced to the changes that occur in the process of aging, some are caused by the plethora of diseases and disruptions that accompany seniority, and still others result from the way old people are treated. Several terms are used by gerontologists to describe the phenomena of aging (Bengston et al., 1999). The aged refers to populations who are characterized as having achieved a certain length of life or expected life span. Aging relates to the developmental process of growth and senescence over time. Age-related refers to how age is taken into account in health and social systems.

Perhaps one of the most intriguing challenges in medicine is to unravel the process of aging. Although we may be able to see pure aging in a cellular culture, it is very hard to visualize in the intact organism. Discussions about aging seem to imply accumulation of chronic diseases. How then does one separate the changes caused solely by aging from the sequelae of disease? Would a group of disease-free older persons be the appropriate models to help us understand the aging process? The prospect sounds uncomfortably like describing life on the basis of a colony of germ-free mice.

Nonetheless, the distinction between so-called normal aging and pathologic changes is critical to the care of older people. We wish to avoid both dismissing treatable pathology as simply a concomitant of old age and treating natural aging processes as though they were diseases. The latter is particularly dangerous because older adults are so vulnerable to iatrogenic effects.

There is growing appreciation that everyone does not age in the same way or at the same rate. The changing composition of today's older adults compared with that of a generation ago may actually reflect a bimodal shift wherein there are both more disabled people and more healthy older people. As described in a recent systematic review, several measures of old age disability and limitations

P.4

have shown improvements in the last decade (Freedman et al., 2002). Attention has become focused on the variations in aging, with great interest directed toward those described as aging successfully that is, showing the least decline in function with time (Rowe and Kahn, 1987).

CHANGES ASSOCIATED WITH NORMAL AGING

We have already noted the critical and difficult distinction a clinician must make to attribute a finding to either the expected course of aging or the result of pathologic changes. This distinction perplexes the researcher as well. We currently lack precise knowledge of what constitutes normal aging. Much of our information comes from cross-sectional studies, which compare findings from a group of younger persons with those from a group of older individuals. Such data may reflect differences other than simply the effects of age. The older group grew up in a different environment, perhaps with different diet and activities. They represent a cohort of survivors. We have come to appreciate that what we see in the older patient is largely a result of what is brought to old age. For example, the decrease in the frequency of osteoporosis today has been related to the observation that women entering the high-risk period (postmenopause) have stronger bones with thicker cortices.

Many of the changes associated with aging result from gradual loss. These losses may often begin in early adulthood, but thanks to the redundancy of most organ systems the decrement does not become functionally significant until the loss is fairly extensive.

Based on cross-sectional comparisons of groups at different ages, most organ systems seem to lose function at about 1 percent a year beginning around age 30 years. Other data suggest that the changes in people followed longitudinally are much less dramatic and certainly begin well after age 70 years.

In some organ systems, such as the kidney, a subgroup of persons appear to experience gradually declining function over time, whereas others' function remains constant. These findings suggest that the earlier theory of gradual loss must be reassessed as reflecting disease rather than aging.

Given a pattern of gradual deterioration whether from aging or disease or both we are best advised to think in terms of thresholds.

The loss of function does not become significant until it crosses a given level. Thus the functional performance of an organ in an older person depends on two principal factors: (1) the rate of deterioration and (2) the level of performance needed. It is not surprising then to learn that most older persons will have normal laboratory values. The critical difference in fact, the hallmark of aging lies not in the resting level of performance but in how the organ (or organism) adapts to external stress. For example, an older person may have a normal fasting blood sugar but be unable to handle a glucose load within the normal parameters for younger subjects.

P.5

The same pattern of decreased response to stress can be seen in the performance of other endocrine systems or the cardiovascular system. An older individual may have a normal resting pulse and cardiac output but be unable to achieve an adequate increase in either with exercise.

Sometimes the changes of aging work together to produce apparently normal resting values in other ways. For example, although both glomerular filtration and renal blood flow decrease with age, many elderly persons have normal serum creatinine levels because of the concomitant decreases in lean muscle mass and creatinine production.

Thus serum creatinine is not as good an indicator of renal function in the elderly as in younger persons. Because knowledge of kidney function is so critical in drug therapy, it is important to get some measure of this parameter. A useful formula for estimating creatinine clearance on the basis of serum creatinine values in the elderly has been developed (Cockcroft and Gault, 1976). (The actual formula is provided in Chap. 14.) Table 1-1 summarizes some of the pertinent changes that occur with aging. For many items, the changes begin in adulthood and proceed gradually; others may not manifest themselves until well into seniority. Readers interested in a more detailed discussion of the changes associated with aging should consult the several excellent reviews on the subject (Birren and Schaie, 2001; Masoro and Austad, 2001).

TABLE 1-1 CHANGES ASSOCIATED WITH AGING

ITEM	MORPHOLOGY	FUNCTION
Overall	Decreased height (vertebral compression and stooped posture secondary to increased kyphosis) Decreased weight (after age 80 in longitudinal studies) Increased fat-to-lean body mass ratio Decreased total body water
Skin	Increased wrinkling Atrophy of sweat glands
Cardiovascular system	Elongation and tortuosity of arteries, including aorta Increased intimal thickening of arteries Increased fibrosis of media of arteries Sclerosis of heart valves	Decreased cardiac output during exercise Decreased heart rate of arteries response to stress Decreased compliance of peripheral blood vessels
Kidney	Increased number of abnormal glomeruli Interstitial fibrosis	Decreased creatinine clearance Decreased renal blood flow Decreased maximum urine osmolality
Lung	Decreased elasticity Decreased activity of cilia	Decreased forced vital capacity and forced expiratory volume Decreased maximal oxygen uptake Decreased cough reflex
Gastrointestinal tract	Decreased hydrochloric acid Fewer taste buds	Slowed intestinal motility
Skeleton	Osteoarthritis Loss of bone structure
Eyes	Arcus senilis Decreased pupil size Growth of lens	Deceased accommodation Hyperopia Decreased acuity Decreased color sensitivity Decreased depth perception
Hearing	Degenerative changes of ossicles Increased obstruction of eustachian tube Atrophy of external auditory meatus Atrophy of cochlear hair cells Loss of auditory neurons	Decreased perception in high frequencies Decreased pitch discrimination
Immune system		Decreased T-cell activity
Nervous system	Decreased brain weight Decreased cortical cell count	Increased motor response Slower psychomotor performance Decreased intellectual performance Decreased complex learning Decreased hours of sleep Decreased hours of rapid eye movement (REM) sleep
Endocrine		Decreased triiodo-thyronine (T₃) Decreased free (unbound) testosterone Increased insulin Increased norepinephrine Increased parathormone Increased vasopressin

BIOLOGICAL AGING

It is now a commonly accepted notion that aging is a multifactorial process. Extended longevity is frequently associated with enhanced metabolic capacity and response to stress. The importance of genetics in the regulation of biological aging is demonstrated by the characteristic longevity of each animal species. However, heritability of life span accounts for 35 percent of its variance, whereas environmental factors account for >65 percent of the variance (Finch and Tanzi, 1997), and genes specifically selected to promote aging are unlikely to exist.

Several theories of aging have been promulgated and recently reviewed (Vijg and Wei, 1995; Kirkwood and Austad, 2000). These theories fall into either of two general categories: (1) accumulation of damage to informational molecules or (2) the regulation of specific genes (Table 1-2).

TABLE 1-2 MAJOR THEORIES ON AGING

THEORY	MECHANISMS	MANIFESTATIONS
Accumulation of damage to informational molecules	Spontaneous mutagenesis Failure in DNA repair systems	Copying errors
	Errors in DNA, RNA, and protein synthesis	Error catastrophe
	Superoxide radicals and loss of scavenging enzymes	Oxidative cellular damage
Regulation of specific genes	Appearance of specific protein(s)	Genetically programmed senescence

DNA undergoes continuous change both in response to exogenous agents and intrinsic processes. Stability is maintained by the double-strandedness of DNA and by specific repair enzymes. It has been proposed that somatic mutagenesis, either owing to greater susceptibility to mutagenesis or deficits in repair mechanisms, is a factor in biological aging. In fact, there is a positive correlation of species longevity with DNA repair enzymes. In humans, the spontaneous mutagenesis rate is not adequate to account for the number of changes that would be

P.6

P.7

necessary, and there is no evidence that a general failure in repair systems causes aging. However, limited maintenance and repair may lead to accumulation of somatic damage.

A related theory, the error catastrophe theory, proposes that errors occur in DNA, RNA, and protein synthesis, each augmenting the other and finally culminating in an error catastrophe. Translation was considered the most likely source for age-dependent errors because it was the final common pathway. However,

P.8

increased translational errors have not been found in either in vivo or in vitro aging. Amino acid substitutions do not increase with age, although some enzyme activities may be altered by changes in posttranslational modification, such as glycosylation.

The major by-products of oxidative metabolism include superoxide radicals that can react with DNA, RNA, proteins, and lipids, leading to cellular damage and aging. There are several scavenging enzymes and some small molecules, such as vitamins C and E, that protect the cell from oxidative damage. There is no significant loss of scavenging enzymes in aging, and vitamins C and E do not increase longevity in experimental animals. However, interest in this hypothesis persists, because overexpression of antioxidative enzymes retards the age-related accrual of oxidative damage and extends the maximum life span of transgenic fruit flies; moreover, caloric restriction lowers levels of oxidative stress and damage and extends the maximum life span of rodents (Finkel and Holbrook, 2000; Masoro, 2000).

One hypothesis of aging is that it is regulated by specific genes. Support for such a hypothesis has been gained mainly from yeast, nematodes, fruit flies, and models of in vitro aging. Several genes in yeast, nematodes, and fruit flies have been found to extend the species life span. They appear to reinforce the importance of metabolic capacity and stress responses in aging. By DNA microarrays, relatively few genes changed in human fibroblasts with aging, and downregulation of genes involved in control of mitosis was proposed as a possible general cause of aging (Ly et al., 2000). However, others have not shown the same gene

P.9

pattern changes in other aging tissues, suggesting that different changes underlie aging in different tissues.

In adulthood, cells can be placed into one of three categories based on their replicative capacity: continuously replicating, replicating in response to a challenge, and nonreplicating. Epidermal, gastrointestinal, and hematopoietic cells are continuously renewed; liver can regenerate in response to injury; while neurons and cardiac and skeletal muscle do not regenerate.

In vitro replication is closely related to in vivo proliferation. Neurons and cardiac myocytes from adults can be maintained in culture but do not divide, whereas hepatocytes, marrow cells, endothelial cells, and fibroblasts replicate in vitro. Because they are easily obtained from skin, fibroblasts have been the most extensively studied. Although some cells continuously replicate in vivo, they have a finite replicative life. For fibroblasts in vitro, this is about 50 doublings (Hayflick, 1979). Replicative life in vitro correlates with the age of the donor, such that the older the donor, the fewer the doublings in vitro. With time in culture, doubling time decreases and ultimately stops. Several lines of evidence suggest that replicative senescence evolved to protect higher organisms from developing cancer (Campisi, 2000).

With each cell division, a portion of the terminal end of chromosomes (the telomere) is not replicated and therefore shortens. It is proposed that telomere shortening is the clock that results in the shift to a senescent pattern of gene expression and ultimately cell senescence (Fossel, 1998). Telomerase is an enzyme that acts by adding DNA bases to telomeres. Transfection of the catalytic component of this enzyme into senescent cells extends their telomeres as well as the replicative life span of the cells and induces a pattern of gene expression typical of young cells. It is now possible to explore the role of replicative senescence in aging and associated chronic disease processes.

These experiments help define the finite life span of cells in vitro but do not themselves explain in vivo aging. However, factors associated with finite cell replication may more directly influence in vivo aging. Fibroblasts aged in vitro or obtained from older adult donors are less sensitive to a host of growth factors. Such changes occur at both the receptor and postreceptor levels. A decrease in such growth factors, a change in sensitivity to growth factors, and/or a slowing of the cell cycle may slow wound healing and thus place the older individual at greater risk for infection.

For tissues with nonreplicating cells, cell loss may lead to a permanent deficit. With aging, dopaminergic neurons are lost, thus influencing gait and balance and the susceptibility to drug side effects. With further decrements such as ischemia or viral infection, Parkinson's disease may develop. Similar cell loss and/or functional deficits may occur in other neurotransmitter systems and lead to autonomic dysfunction as well as alteration in mental function and neuroendocrine control.

The immune system demonstrates similar phenomena. Lymphocytes from older adults have a diminished proliferative response to a host of mitogens. This

P.10

appears to be a result of both a decrease in lymphokines and a decrease in response to extracellular signals. As the thymus involutes after puberty, levels of thymic hormones (thymosins) decrease.

Basal and stimulated interleukin-2 (IL-2) production and IL- 2 responsiveness also diminish with age. The latter appears to be due, at least in part, to a decreased expression of IL-2 receptors. Some immune functions can be restored by the addition of these hormones to lymphocytes in vitro, or, in vivo, by their administration to aged animals. The proliferative defect can also be reversed in vitro by calcium ionophores and activators of protein kinase C, suggesting that the T-cell defect may be in transduction of extracellular signals to intracellular function.

In vivo, molecular mechanisms such as those described above contribute to physiologic deficits and altered homeostatic mechanisms that predispose older individuals to dysfunction in the face of stress and disease.

The gene for Werner's syndrome, a progeric syndrome associated with early onset of age-related changes such as gray hair, balding, atherosclerosis, insulin resistance, and cataracts, but not Alzheimer's disease was recently cloned. The gene codes for a helicase involved in DNA replication. There is great interest in understanding how a defect in just this one gene leads to the multiple abnormalities of this syndrome.

Molecular geneticists have also cloned several genes related to early onset familial Alzheimer's disease and identified susceptibility genes for the late-onset form of the disease (Tanzi et al., 1996).

A small number of families have mutations in the amyloid precursor protein located on chromosome 21. The largest number of families with early onset familial Alzheimer's disease have a mutation in a gene on chromosome 14. This gene has been named presenilin 1. A similar gene has been identified on chromosome 1 and labeled presenilin 2. The role of the presenilins in Alzheimer's disease pathology is not yet known, but the identification of the three loci mentioned above has led to much excitement for the potential understanding of pathophysiologic mechanisms in this devastating disease. Similarly, the identification of apo E alleles as risk factors for late-onset Alzheimer's disease has raised interest in both the diagnosis and pathology of this disease.

Stem cells offer hope to greatly extend the numbers and range of patients who could benefit from cell replacement therapy to treat debilitating diseases such as diabetes, Parkinson's, Huntington's, and Alzheimer's. There is a long way to go in basic research before new therapies will be established, but clinical trials in some diseases are underway (Lovell-Badge, 2001).

CLINICAL IMPLICATIONS

As we try to understand aging, we appreciate the limitations of available information. As noted earlier, most of the data cited to document changes with

P.11

age come from cross-sectional studies in which individuals of different ages are compared in terms of group averages. Such an approach generally reveals a gradual decline in organ function with age, beginning in early middle life. A few studies have followed cohorts of people longitudinally as they age. Their conclusions are quite different. In several parameters, performance actually increases with age. For example, cognitive function can improve over time among older persons. Similarly, cardiac function in subjects free of heart disease does not show inevitable decline with age (Lakatta, 1999).

The physician must be able to take data derived from group studies and apply them to the individual. It is essential to keep in mind the principle of individual variation. The best predictor of a given patient's performance now is that person's earlier performance rather than an average age-related decline documented in cross-sectional studies. Thus, an 80-year-old runner may well have better cardiovascular function than a 50-year-old sedentary doctor.

Aging is not simply a series of biological changes. That is, when one looks in the mirror and confronts an old person, the noted changes are associated with a variety of alterations in life. Aging is a time of losses: loss of social role (usually through retirement), loss of income, loss of friends and relatives (through death and mobility). It can also be a time of fear: fear for personal safety, fear of financial insecurity, fear of dependency.

In the face of these enormous threats, we should pause to rethink our views about older adults. Rather than being victims, they are the survivors. Most elderly persons have developed mechanisms to cope with multiple limitations. Most nevertheless continue to function. The physician's role is to enhance this coping ability by identifying and treating remediable problems and facilitating changes in the environment to maximize function in the face of those problems that remain.

In some instances, the patient's coping skills may make the physician's task more difficult. The elderly patient has often adapted to problems by denying or ignoring them. In such cases, it will be difficult to obtain a good history. Other patients cope with their disabilities by employing adaptive techniques. For example, a person who is hard of hearing may talk a great deal to hide a hearing deficit. A particularly troublesome problem is the skillful compensation for cognitive losses.

At least once in every physician's career, the physician will encounter a patient who carries on a perfectly lucid conversation, only to discover on closer examination that the patient is completely disoriented to time, place, and person. Because it is easy to miss these cognitive deficits, we recommend that an evaluation of older persons include a formal screening for mental status. A simple method for doing this is described in Chap. 6.

One of the hallmarks of aging is the reduced response to stress, including the stress of disease. Thus, the symptom intensity may be dampened by the aged body's decreased responsiveness. The presentation of illness in the geriatric patient can be thought of as a combination of dampened primary sound in the presence of background noise.

P.12

In treating an older adult, it is useful to keep in mind that an individual's ability to function depends on a combination of his or her characteristics (e.g., innate capacity, motivation, pain tolerance) and the setting in which that person is expected to function. The same individual may be functional in one setting and dependent in another.

The physician's first responsibility is to treat the patient, to remedy the remediable by searching for and dealing with those conditions that are treatable. Having improved the patient's ability (physiologically and psychologically) as much as possible, the physician's next task is to structure an environment that will facilitate the patient's functioning with maximum autonomy. This latter mandate should not rest exclusively on the doctor's shoulders. A variety of health-related professionals are available in most situations to play major roles in locating and utilizing supportive environments. But the physician must not abrogate this task. To ignore the environment of a disabled individual is tantamount to prescribing drugs and ignoring the patient's compliance with the treatment regimen.

Conversely, the environment can produce dysfunction. At the simplest level, it may produce hazards that lead to falls (see Chap. 9). At a more subtle level, it may necessitate a level of effort that produces decompensation. For example, an elderly person with dyspnea on exertion may get along reasonably well in a ground-floor apartment but become unmanageable in an apartment on the second floor of a building with no elevator. Similarly, patients with compromised pulmonary or cardiac function will show increased morbidity and mortality as air pollution levels increase. Finally, the environment may create disability by fostering dependency.

At a somewhat more subtle level, physicians must be aware of the forces among caregivers that foster dependency. Patients may be immobile because of the care they get. One important factor is risk aversion. Nursing personnel may be reluctant to mobilize patients for fear that they will fall and sustain injuries. We must provide assurance to staff that they will not be penalized for activating patients appropriately.

Nor is risk aversion confined to professionals. Families may be equally protective, insisting on limiting an older relative's activities or moving him or her to a more closely supervised situation. Such fears are often infused with guilt and may manifest as anger. Families can be helped to see the dangers of such restricted activity.

CLASSIFYING GERIATRIC PROBLEMS

Because diagnoses often do not tell the whole story in geriatrics, it is more helpful to think in terms of presenting problems. One aid to recalling some of the common problems of geriatrics uses a series of I's:

Immobility
Instability

P.13

Incontinence
Intellectual impairment
Infection
Impairment of vision and hearing
Irritable colon
Isolation (depression)
Inanition (malnutrition)
Impecunity
Iatrogenesis
Insomnia
Immune deficiency
Impotence

The list is important for several reasons. Especially with older patients, the expression of the problem may not be a good clue to the etiology. Conversely, a problem may occur for a variety of reasons.

For example, an individual may be immobilized by a broken hip, by severe angina, or by arthritis. But the patient may also be immobilized by fear. The elderly patient with a successfully repaired hip fracture may be unwilling to walk again for fear of falling and sustaining another fracture. An elderly person living in a deteriorated neighborhood may be confined to the home not by physical limitations but because of a fear of being molested. Such an individual may decide to enter a long-term-care institution to seek a safer environment. In each instance, the physician and coworkers must obtain a sufficient history to understand the true etiology of the problem if they are to develop a successful approach to remedying the condition.

Another factor in generating dependency is cost. It is often much easier and cheaper to do things for people with functional limitations than to invest the effort needed to encourage them to do for themselves.

Unfortunately such savings are short-ranged; they will increase the level of dependency and ultimately the amount of care needed.

Among the list of I's is iatrogenesis. The least desirable outcome of medical care is a decrease in the patient's health as a result of contact with the care system. In some cases, there is a real risk that untoward consequences of treatment may worsen a patient's health. The risk:benefit calculation as a basis for urging intervention must be performed carefully for each elderly patient in the context of the patient's condition. Many risks are within the ordinary bounds of medicine.

We are concerned here with those events that result from indifferent or superficial care. The physician who casually adds another drug to the patient's polypharmacy portfolio is playing with a living chemistry set. The reduced rate of drug metabolism and excretion in many elderly persons exacerbates the problem of drug interactions (Chap. 5). Even more dangerous is the careless, hasty application of clinical labels. The patient who becomes confused and disoriented in the hospital may not be suffering from dementia. The individual who has an

P.14

occasional urinary accident is not necessarily incontinent. Labeling patients as demented or incontinent is too often the first step toward their placement in a nursing home, a setting that can make such labels self-fulfilling prophecies. We must exercise great caution in applying these potent labels. They should be reserved for patients who have been carefully evaluated, lest we unnecessarily condemn countless persons to lifetimes of institutionalization.

DIAGNOSIS VERSUS FUNCTIONAL STATUS

One of the persistent problems surrounding growing discussions about care of the elderly has arisen from the emphasis on functioning. This emphasis on the need to direct clinical attention to the patient's functional status as well as to specific medical conditions has occasionally been misinterpreted. The point is not that functional status is more important or more useful than diagnosis but that both are needed. One is incomplete without the other. Functioning is the result of the innate abilities of the patient and the environment that supports those abilities.

Clearly, the optimal management of an elderly patient involves identifying a correctable problem and correcting it. The first and principal task of the physician is to do precisely that. No amount of rehabilitation, compassionate care, or environmental manipulation will compensate for missing a remediable diagnosis. However, diagnoses alone are usually insufficient. The elderly are the repositories of chronic disease more often cared for than cured.

The process of geriatrics is thus twofold: (1) careful clinical assessment and management to identify remediable problems and (2) equally careful and competent functional assessment to ascertain how the patient's autonomy can be maximized by appropriate human and mechanical assistance and environmental manipulations.

Our goal in orienting primary care providers is to raise their consciousness about the need to consider the whole patient and the patient's environment, but never at the cost of neglecting the search for correctable causes for the patient's problems. In that search for causes, the a priori probabilities will often differ substantially from those of younger patients. For this reason, a problem-focused approach, like that of the I's, outlined above, may prove useful.

References

Bengston VL, Rice CJ, Johnson ML: Are theories of aging important? Models and explanations in gerontology at the turn of the century, in Bengston VC, Schaie KW (eds): Handbook of Theories of Aging. New York, Springer, 1999.

Birren JE, Schaie KW (eds): Handbook of the Psychology of Aging, 5th ed. San Diego, Academic Press, 2001.

Campisi J: Cancer, aging and cellular senescence. In Vivo 14:183 188, 2000.

P.15

Cockcroft DW, Gault MH: Prediction of creatinine clearance from serum creatinine. Nephron 16:31 41, 1976.

Finch CE, Tanzi RE: Genetics of aging. Science 278:407 411, 1997.

Finkel T, Holbrook NJ: Oxidants, oxidative stress and the biology of ageing. Nature 408:239 247, 2000.

Fossel M: Telomerase and the aging cell. JAMA 279:1732 1735, 1998.

Freedman VA, Martin LG, Schoeni RF: Recent trends in disability and functioning among older adults in the United States. JAMA 288:3137 3146, 2002.

Hayflick L: Cell biology of aging. Fed Proc 38:1847 1850, 1979.

Kirkwood TBL, Austad SN: Why do we age? Nature 408:233 238, 2000.

Ly DH, Lockhart DJ, Lerner RA, et al: Mitotic misregulation and human aging. Science 287:2486 2492, 2000.

Lakatta EG: Circulatory function in younger and older humans in health, in WR Hazzard, JP Blass, EL Bierman (eds): Principles of Geriatric Medicine and Gerontology 4th ed New York, McGraw-Hill, 1999.

Lovell-Badge R: The future for stem cell research. Nature 414:88 91, 2001.

Masoro EJ: Caloric restriction and aging: an update. Exp Gerontol 35:299 305, 2000.

Masoro EJ, Austad SN (eds): Handbook of the Biology of Aging, 5th ed. San Diego, Academic Press, 2001.

Rowe JW, Kahn RL: Human aging: usual and successful. Science 237:143 149, 1987.

Tanzi RE, Kovacs DM, Kim T-W, et al: The gene defects responsible for familial Alzheimer's disease. Neurobiol Dis 3(16):159 168, 1996.

Vijg J, Wei JY: Understanding the biology of aging: the key to prevention and therapy. J Am Geriatr Soc 43:426 434, 1995.