CHAPTER 1
Introduction
This volume is the sequel to Normal Aging: Reports from the Duke Longitudinal Study, 1955-1969. Its purpose is to bring together and summarize the main findings which build upon and extend those presented in the first volume. It contains all the reports from the Duke Longitudinal Studies published since 1969 as well as an almost equal number of papers presented at professional meetings or written especially for this volume. These reports present new findings on aging among the older longitudinal panel as well as the first findings on the middle-aged panel.
Both of these volumes deal with "normal aging" in two senses: healthy aging and typical aging. The aged and middle-aged persons studied were relatively healthy in that they were noninstitutionalized, ambulatory, community residents who were willing and able to come to the Duke Medical Center for one or two days of tests and examinations. Second, the more common or typical patterns and problems of aging are focused upon rather than the unusual abnormalities. This volume deals with the typical physical changes that accompany aging, typical health care patterns, typical patterns of mental aging, some common forms of mental illness among the aged, the normal social roles, self concepts, satisfactions, leisure and sexual behavior, and the overall factors related to longevity.
As pointed out in the first volume, investigations of normal aging are of crucial importance in advancing the science of gerontology and in helping aged persons develop and enjoy a richer and longer life. When we can distinguish normal and inevitable processes of aging from those which may accompany aging simply because of accident, stress, maladjustment, or disuse, we can better focus our attention and efforts on those factors which can be changed and corrected.
Both volumes also emphasize the theoretical and methodological advantages of the longitudinal and interdisciplinary methods used in these studies. Since aging is by definition a process of change over time, it would seem that the best way to study aging is longitudinally, by repeated observations over time. This is not meant to deny the value of cross-sectional studies, nor to deny the technical and methodological problems connected with longitudinal study, but it is meant to reassert the unique advantages of longitudinal studies such as the following: each panel member can be used as his own control, consistent trends can be distinguished from temporary fluctuations, errors due to retrospective distortion are minimized, early warning signs of disease or death can be studied, cohort differences can be distinguished from age changes, and the effects of one kind of change on another kind of change at a later time period can be studied.
The interdisciplinary nature of the study is useful because aging affects many interrelated types of behavior and functioning. When specialists from different disciplines work together, the mutual stimulation, correction, and combination of perspectives can result in more accurate, thorough, and comprehensive understanding of the aging process.
However, this sequel differs from the first volume in several respects. While most of the reports in the first volume were previously published elsewhere, this sequel contains fifteen original papers written especially for this volume or presented to professional meetings but not published elsewhere. Second, while the first volume was limited to reports from the first longitudinal study of aging (persons over 60 years of age), this sequel is about one-third devoted to reports from the Adaptation Study of middle-aged persons (46-70 years of age). Third, this sequel shows that the tempo of reports produced has increased, because it contains thirty-five reports produced in the past four years (about nine per year) compared to the fifty basic reports in the first volume which were produced during the preceeding fourteen years (about three and one-half per year).
The reader who is interested in detailed description and discussion of the first longitudinal study of aging is referred to chapter 1 of the first Normal Aging volume. Suffice it to say here that the first longitudinal study of aging began in 1955 with 271 persons 60-90 years of age. The subjects were not a strictly representative sample, but were selected from a pool of volunteers who lived in the community so that the overall sample reflected the age, sex, ethnic, and socioeconomic characteristics of the older population in Durham, N.C. Each panelist was brought in to the Medical Center for a two-day long series of medical, psychiatric, psychological, and social examinations. These examinations were repeated every three or four years until 1965, every two years until 1972, and since then the examinations are being repeated on an annual basis. In 1973, the eighth round of examinations on the survivors of this study was completed.
The second longitudinal study, or Adaptation Study, was begun in 1968 with 502 persons 46-70 years of age. These panelists were a random sample of the members of the local health insurance association stratified by age and sex. The sample and study was designed so that at the end of five years there would remain approximately 40 persons in each of ten five-year age-sex cohorts. This design makes possible various kinds of cross-sequential types of analyses in order to separate the effects of aging from cohort differences and from changes in the environment over time. Each subject was brought in for a one-day long series of physical, mental, and social examinations. The panelists are reexamined at two-year intervals and in 1973 the third round of examinations was completed. For more details and discussion of the adaptation study design, the reader is referred to Appendix A in this volume.
The articles which are reprinted by permission from various journals are presented with no deletions except for duplicate passages such as those that repeat the description of the panelists. Deletions of a sentence or more in these articles are indicated by ellipses, and minor changes have been made for uniformity. Since most of the reports are interdisciplinary, the placement of some reports in one chapter rather than another was somewhat arbitrary, but an attempt was made to group reports by the type of central variable under investigation. Each chapter first presents the reports from the first longitudinal study and then reports from the Adaptation Study. Brief summaries of the reports introduce the chapters, and the book closes with an overall summary of the main findings and themes. The reader may wish to start with these summaries in order to decide which reports to read in more detail.
We appreciate the cooperation of the many authors and journals or publishers for their permission to reprint their articles. Special thanks are due to Mrs. Elke Gordon and to Mrs. Celenzy Chavis for their help in preparing the manuscript. Most of this research was supported in part by the U.S. Public Health Service, National Institute of Child Health and Human Development, Grant HD-00668.
CHAPTER 2
Physical Aging
We begin with physical aging because it is basic to all the other aspects of aging. And most basic among the physical aspects are possible changes in cell function with age. The first report presents the age differences in one essential product of the body cells, immunoglobulins, which produce our various immunities. The major rinding that the older persons (drawn from the older longitudinal study) had significantly lower average immunoglobulin concentration compared to younger persons suggests that lower immunity may partly account for the greater frequency of illness among the aged.
The next three reports present some of the first longitudinal data on dermatological, ophthalmological, and auditory changes among the normal, aged. The report on dermatological changes also examines the relationship between certain skin problems and longevity. The report on ophthalmological changes also examines the relationship of certain vision defects with longevity and with indicators of activity and satisfaction. The report on auditory changes shows that while hearing acuity generally diminishes with increasing age, there are substantial differences between sex and racial groups.
From The Journal of Immunology 105:4:964-972. Copyright 1970, The Williams & Wilkins Co. Reproduced by permission. This work was supported by grants from the United Public Health Service (AI-07499, 5-K3-AI-14797, HD-00668) and the American Medical Association (AMA-ERF).
Serum Immunoglobulin Concentrations C. E. Buckley III and F. C. Dorsey
The relationship between changes in serum immunoglobulin concentrations and immunologic diseases in adults is often obscure. Current difficulties in identifying alterations relevant to disease may result from several factors: the degree of departure from normal confidence intervals is usually small in comparison to dysimmunoglobulinemias occurring early in life; other sources of biologic variation, such as race (Rowe et al., 1968) and sex (Butterworth et al., 1967; Stoop et al., 1969) tend to obscure disease-related changes. Aging is another source of biologic variation only indirectly related to disease. Significant age-related changes occur prior to adult life (Buckley et al., 1968). Relatively little is known of the immunoglobulin content of the serum of humans late in life, a period associated with an increased frequency of disease. As a consequence observations of serum immunoglobulin concentrations in older patients are difficult to interpret. A systematic study of serum immunoglobulin concentrations throughout the life-span of man is a necessary requisite for the identification of disease-related changes.
The concentrations of IgG, IgA, and IgM were measured in the serum of 811 apparently healthy humans ranging in age from birth to 92 years. This work represents a part of continuing studies of the immunologic aspects of aging in man. Biometric analysis of these data suggests that valid contrasts of serum immunoglobulin concentrations between healthy and diseased adults are not possible without adequate control of biologic variation related to age.
Materials and Methods
Subjects
Sera were obtained from apparently healthy individuals who were free of complaints related to acute or chronic disease and exhibited no obvious abnormalities. The subjects included professional and paramedical hospital personnel, healthy family members and friends of patients who provided blood donations, and older individuals who are members of longitudinal study populations under periodic survey by the Duke University Center for the Study of Aging.
Preparation of Antigens and Antisera
Blood was collected by phlebotomy and serum separated within 1 hour. Sera were frozen at —20° C until analyzed and retained at 4° C during the period of analysis.
Purified human immunoglobulins were used as antigens and primary reference standards. Each immunoglobulin was prepared from serum by a combination of gel filtration and ion exchange chromatography. Human IgG was prepared from Cohn fraction II human γ globulins and from sera or patients with heterogenous hyperglobulinemia and myeloma. Human IgA was prepared from the sera of patients with IgA multiple myeloma and from one patient with increased heterogenous IgA. Human IgM was prepared from the sera of two patients with monoclonal macroglobulinemia, a patient with cryomacroglobulin (Turkington et al., 1966) and a patient with heterogenous macroglobulinemia (Whitehouse et al., 1967).
Gel exclusion chromatography with G-200 Sephadex or P-300 Bio-Gel was performed on a 15 cm × 150 cm column as an initial enrichment step. Immunoglobulins in eluate peaks were initially identified by microimmuno-electrophoresis (Scheidegger, 1955) with a polyvalent antiserum to human serum proteins. More recently single radial diffusion (Mancini et al., 1965) against monospecific antisera has also been used in order to identify eluate fractions enriched with the desired human immunoglobulin. Chromatography was repeated on the same column after concentration of the desired eluate fraction by lyophilization. Solutions of each partially purified human immunoglobulin were chromatographed on a 5 × 100 cm DEAE Sephadex A-5 column. The resin bed and protein solution were equilibrated with a 0.01 M ionic strength, pH 8, Tris-propionate buffer, and elution was carried out with a continuous gradient developed by pH 7, 0.05 M ionic strength Tris-propionate buffer adjusted to a final ionic strength of 0.45 M with sodium chloride. Eluate fractions were monitored by measurement of the optical density, evaluated immunochemically and concentrated by lyophilization. Purified antigens were further characterized by analytical ultracentrifugation and further immunochemical analysis. Preparations having immunochemical or physicochemical evidence of impurities were rechromatographed under the same conditions, or alternately were chromatographed on CM-Sephadex A-50 after equilibration with a pH 5.0, 0.01 M ionic strength acetate buffer. Elution was accomplished with a continuous gradient developed by a pH 6, 0.5 M ionic strength acetate buffer.
The apparently immunochemically pure globulins obtained were adjusted to a pH of 5.5 and dialyzed into 0.01 M sodium chloride. Each solution of IgG and IgA was adjusted to a concentration range between 1 percent and 4 percent and each IgM solution to between 0.3 percent and 0.5 percent. Each preparation was sterilized by ultrafiltration prior to measurement of concentration by drying duplicate amounts to minimal weight at 107° ± 0.2° C in a forced draft oven. All samples contained more than 20 mg protein and were weighed on an analytical balance having a precision of ±0.0002 g. Weights obtained were corrected for the salt content of the initial solution and the concentration of each primary reference standard solution expressed in milligrams/milliliter of solution. The concentrations of some primary reference standard solutions were also evaluated in the analytical ultracentrifuge using Rayleigh interference optics η = 0.000465 for 2.5 mg/ml, λ = 546 nm). Estimates of concentration by these two methods agree within ±2 percent.
All antisera were prepared in goats. The quantity of primary reference standard used as antigen solution for each injection varied between 0.3 and 1.0 mg in a volume of 1 ml. Each antigen was homogenized with an equal volume of mixed complete Freund's adjuvant and injected intramuscularly in divided doses into each goat at weekly intervals for a period of six to eight weeks. Booster doses of antigen prepared as an alum toxoid (Kabat, 1961) were used twice monthly thereafter. Approximately 200-ml bleedings were obtained every two weeks after the development of precipitating antibodies, usually after the fourth week of immunization.
Each harvest of antiserum was evaluated for antibody to other human serum proteins by microimmunoelectrophoresis. Apparently monospecific antisera were evaluated against a panel of 12 or more κ and λ purified human L chains (Bence Jones proteins). Antibody activity to L chains or to other human serum proteins was removed by absorption with an appropriate antigen containing serum or partially purified antigen. After absorption each pool of goat antiserum was precipitated three times with ammonium sulfate (24 g/100 ml of serum or solution), dialyzed into physiologic saline, sterilized by ultrafiltration, and portions were sealed in 10-ml vials closed with a rubber bung and aluminum seal. Each antiserum globulin preparation was again evaluated for potency and specificity, dated and labeled, and retained at —70° C until used. These goat antibody globulin preparations contained 90 percent or better 7S protein by ultracentrifugation and antibody activity was two- to fourfold greater than in neat serum.
