| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||||
Geriatrics Section, Boston Medical Center and Boston University School of Medicine, Boston, Massachusetts.
 |
Abstract |
|---|
 Top Abstract MethodsResultsDiscussionReferences |
|---|
Methods. The health histories of centenarian offspring (
) and controls (
) were assessed from 1997–2000 using a cross-sectional study design. The offspring of 192 centenarian subjects enrolled in the nationwide New England Centenarian Study were recruited and enrolled. Controls consisted of offspring whose parents were born in the same years as the centenarians but at least 1 of whom died at an average life expectancy. Prevalence of age-related diseases including heart disease, hypertension, diabetes, cancer, osteoporosis, stroke, dementia, cataracts, glaucoma, macular degeneration, depression, Parkinson's disease, thyroid disease, and chronic obstructive pulmonary disease were compared between the two groups.
Results. Centenarian offspring had a 56% reduced relative prevalence of heart disease (odds ratio [OR] 0.44, 95% confidence interval [CI] 0.24, 0.80), a 66% reduced relative prevalence of hypertension (OR 0.34, 95% CI 0.21, 0.55), and 59% reduced relative prevalence of diabetes (OR 0.41, 95% CI 0.15, 1.12) after multivariate adjusted analyses.
Conclusions. The offspring of centenarians demonstrate a markedly reduced prevalence of diseases associated with aging, in particular for cardiovascular disease and cardiovascular risk factors. Along with their parents, the centenarian offspring, who are in their 70s and 80s, may prove to be a valuable cohort to study genetic and environmental factors conducive to the ability to live to very old age in good health.
CENTENARIANS delay, if not escape, age-associated diseases that normally cause mortality at earlier ages (1,2). In the New England Centenarian Study, 88% of centenarians were functioning independently at the age of 92 years (3). Most subjects experienced a decline in their cognitive function only in the last 3 to 5 years of their very long lives (4,5). Thus, we suspect that in order to achieve their age, centenarians survive better with, delay, or escape diseases normally associated with aging (6,7).
A significant component of the ability to survive to exceptional old age appears to be familial (8). Siblings of centenarians have themselves a marked increased probability of surviving to 100 years. Compared with the general survival experience of their birth cohort (born in 1896), the male and female siblings of centenarians have, respectively, a 17 and 8 times greater probability of living to 100 (9). In addition, the mean age of the parents of centenarians is higher than the average life expectancy for their birth cohort. Perls and colleagues have previously noted that 22% of mothers of centenarians and 13% of fathers of centenarians in their sample survived to at least age 90 or older (7). Among 444 centenarian pedigrees surveyed, approximately one half have at least 1 first degree relative or grandparent who lived beyond the age of 90 years (10).
Circumstantial evidence suggests a significant genetic component to this familial predisposition for exceptional longevity (9,11). Recently, the New England Centenarian Study reported a statistically significant linkage between a genetic locus on chromosome 4 and exceptional longevity among 137 sibships with at least 2 centenarian siblings (12). Patterns of genetic variations also have been noted among centenarians, most notably a decreased frequency of the apolipoprotein E 
-4 allele (13).
Given the evidence thus far for the familiality of exceptional longevity, and anecdotal reports of many centenarian offspring remaining in good health, we suspected that, like their parents, the offspring of centenarians delay or escape age-related diseases. With the exception of a study of the offspring of Ashkenazi Jewish centenarians by Barzilai and colleagues (14), little research has been done on the offspring of centenarians. Therefore, our objective was to determine the prevalence of age-related diseases in the offspring of centenarians and to generate hypotheses for future studies of the familiality of exceptional longevity.
|
METHODS |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
The study sample consisted of offspring of 192 centenarian subjects enrolled in the New England Centenarian Study, a nationwide study of centenarians. Of the 362 potential centenarian offspring who were alive and for whom contact information was available, 198 (55%) enrolled in the study. Of these, 177 (89%) successfully completed the study. Of the 192 centenarians, 60% were represented by at least 1 child that enrolled in this study. The offspring sibships ranged from 1 to 7, with a mean sibship of 1.8.
The centenarian offspring were initially contacted by telephone. The Committee for Clinical Investigations based at Beth Israel Deaconess Medical Center approved the study protocol.
The controls () consisted of offspring whose parents were born in the same years as the centenarians but at least 1 of whom died at age 73, the average life expectancy for that birth cohort. A list of approximately 4500 individuals who died in Massachusetts in the years 1970–1972 at the age of 73 years was obtained from the Massachusetts Department of Vital Records. Obituaries in the 1970–1972 Boston Globe were then searched for these individuals in order to find next-of-kin information. Next-of-kin names and addresses were located using internet-based resources. Of the 1552 obituaries found, 1097 listed offspring as next-of-kin, resulting in 3430 potential controls. The Social Security Death Index (SSDI) was used to determine if any of the next-of-kin had died. Of the 992 potential controls who were alive and for whom contact information was available, 203 (20%) enrolled in the study. A total of 166 (82%) successfully completed the study. Offspring of 96 deceased 73-year-olds were enrolled. The control sibships ranged from 1 to 4, with a mean sibship of 1.5.
Data Collection
Study participants were sent questionnaire packets that included the following: (a) a demographic and health questionnaire, (b) a functional status questionnaire, and (c) a physical examination form. The demographic and health questionnaire included questions about demographics, medical history, medications, alcohol and tobacco use, and exercise. Functional status was assessed using the Instrumental Activities of Daily Living (IADL) questionnaire (15). The physical examination form, filled out by the subject's physician during a regularly scheduled visit, requested height, weight, temperature, 3 blood pressures (sitting), and 3 pulse measurements. In addition, pedigree information was gathered by telephone from 1 or more offspring of each family.
Variables
Dependent variables included a determination of a history or the presence of the following age-related diseases: coronary artery disease, myocardial infarction, congestive heart failure, arrhythmia, hypertension, diabetes mellitus, cancer, stroke, dementia, osteoporosis, cataracts, glaucoma, macular degeneration, depression, Parkinson's disease, thyroid condition, and chronic obstructive pulmonary disease (COPD). A history of heart disease was noted if any of the following conditions existed: coronary artery disease, myocardial infarction, congestive heart failure, and/or arrhythmia. Covariates included age, gender, years of education, annual income, IADL score (instrumental activities of daily living), ethnicity, marital status, exercise, smoking, and alcohol use.
Statistical Analyses
Prevalence was determined for each of the diseases noted above. Preliminary examination of the differences between centenarian offspring and controls were performed using chi square analyses and analysis of variance (ANOVA). Logistic regression modeling was used to examine the relationship between specific diseases and offspring status. Clustered multivariate generalized estimating equations were used to control for potential confounders and colinearity between family members in the analyses. Recursive partitioning was used to examine the association of covariates with parent status.
Recursive partitioning is a nonparametric statistical technique that results in a tree-structured association of covariates with an outcome (16). In this case, the outcome was parent group, centenarian, or septuagenarian. Recursive partitioning is analogous to a maximally selected chi-square at each decision point. The technique uses information where it is available and surrogate information where data are missing. In the event that data are missing for the chosen variable at a decision node, observations are then classified using an empirically selected surrogate variable. Surrogates are variables correlated with decision node variables that cause a case to proceed in the same direction as the original variable.
|
RESULTS |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
]).
|
|
) (Figure 2).
|
|
), a 64% reduced relative prevalence of coronary artery disease (
), a 63% reduced relative prevalence of a prior myocardial infarction (
), and a 36% reduced relative prevalence of arrhythmias (
). Centenarian offspring also had a 66% reduced relative prevalence of hypertension () and a 59% reduced relative prevalence of diabetes (
) (Table 4). There were no differences between the 2 groups with respect to the prevalence of osteoporosis and cancer.
|
|
|
|
DISCUSSION |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
In addition, the centenarian offspring used fewer medications than controls did. These observations persisted after adjustment for age, gender, years of education, annual income, IADLs, ethnicity, marital status, exercise, smoking, and alcohol use. In a mortality analysis of 2092 siblings of 444 centenarians, we noted that siblings of centenarians experienced approximately half the mortality risk of their birth cohort, a risk reduction that was sustained over the course of the life span including old age (9). The marked disease-specific survival advantage observed among the offspring of centenarians is consistent with the overall survival advantage noted among the siblings of centenarians.
Interestingly, there were no significant differences in the prevalence of a number of other age-related diseases including cancer, stroke, dementia, osteoporosis, cataracts, glaucoma, macular degeneration, depression, Parkinson's disease, thyroid disease, and COPD. The lack of differences for these diseases may be a function of the sample size, the choice of the controls, or it may be that families with exceptional longevity do not have differential susceptibility to these diseases. Post hoc power calculations accounting for a moderate correlation due to familial clustering indicated that the sample size needed to detect differences of the magnitudes we observed would require samples in excess of 1500 per group.
Evidence for a Familial Component to Longevity
The decreased risk for age-related diseases in the offspring of centenarians is consistent with other research suggesting that long-lived parents have long-lived children (10,17–20). Prior studies suggest that both parents contribute to the heritability of life expectancy (17). Here, we compared the offspring who had 1 parent survive to exceptional old age to offspring who had 1 parent survive to the average life expectancy. This does not take into account the contribution of the other parent for either group. However, the effect of the other parent on each group should be similar, since the mean age of the other parent was statistically equivalent.
In this study, the centenarian offspring and the controls had similar frequencies of smoking, alcohol use, exercise, and similar socioeconomic status. While there were differences between the 2 groups in the number of years of education, this was controlled for in multivariate analyses. Other environmental influences, such as diet, were not included in the analyses and might explain an important component of the differences in disease prevalence and risk that were observed.
Role of Lower BMI in Longevity
One of our intriguing findings is a significantly lower BMI in the centenarian offspring compared with the controls. Recursive partitioning demonstrated that a low BMI was predictive of having a centenarian parent, although a substantial amount of missing data on this variable must temper the interpretation of these findings. Interestingly, the empirically derived BMI cut-point of 26.13 is remarkably similar to the World Health Organization's criteria for being overweight (
) (21). Decreased body fat, particularly abdominal fat, has been shown to extend life span and slow age-associated changes in several animal species including rodents and nonhuman primates (22). Increased BMI and abdominal fat have been associated with increased risk for diabetes (23–25), heart disease (26–28), and multiple metabolic syndrome. Multiple metabolic syndrome, a constellation of metabolic alterations associated with obesity including dyslipidemias, hypertension, and insulin resistance (28), could be less prevalent among the offspring of centenarians. Future studies examining the genetic and environmental factors associated with the syndrome among the offspring of centenarians and other cohorts are warranted.
Limitations
The relatively low control participation rate raises a concern regarding the role of selection bias. Offspring who were less healthy might have been less likely to participate. To explore this possibility, data about nonparticipants were collected and segregated by health and nonhealth related reasons. Self-reported health-related reasons accounted for 2% and 4% of nonparticipation for the centenarian offspring and controls, respectively. This suggests that participating controls were healthier than centenarian offspring, which would, if anything, bias findings towards the null. Nonhealth-related reasons, primarily a lack of interest or time to participate in a scientific study, accounted for nonparticipation among 54% of the centenarian offspring and 21% of the controls. Not surprisingly, more of the centenarian offspring opted for participation, since they were members of families that had a prior relationship with the New England Centenarian Study.
Ten percent of the centenarian offspring and 51% of the controls could not be located. One reason for the large discrepancy between the 2 groups was the difficulty in finding the controls due to 30-year-old obituary information. It is unclear how this might have influenced the control sample, as it may have resulted in subjects who were lost to death and disease or, alternatively, subjects who were healthier and moved to another location for retirement. In order to account for individuals who were not found or were not successfully contacted, pedigree information was collected from families. This information helped us determine whether individuals were alive but not interested in participating, were too ill to participate, or had died. In addition, the Social Security Death Index was used to determine whether individuals who were not found had died.
There is the potential for bias due to selective survival in both groups. The septuagenarian and octogenarian participants in both groups have already survived into relatively old age and are therefore potentially healthier than others in the general population. If anything, such selective survival suggests that cardiovascular and age-related disease advantages demonstrated in the centenarian offspring may be even stronger than those noted in this study.
The health questionnaire relies on self-report of medical conditions and habits. Other studies assessing the validity of self-administered questionnaires have demonstrated that subjects reliably and accurately report well-defined, serious, and chronic conditions such as heart disease, myocardial infarction, congestive heart failure, diabetes mellitus, and hypertension, all outcomes of interest for this study (29–31). Some studies have indicated lower reliability of self-report for older individuals (30), while others studies have not shown any differences. (32) Unfortunately, our attempts to collect information from the participants' physicians resulted in incomplete or missing data for a number of study participants.
Conclusions
The offspring of centenarians have a significantly reduced risk for age-related diseases including hypertension, coronary heart disease, and diabetes. Our results are consistent with previous studies, which show a familial component to longevity. Future genetic studies will help to untangle the role of genes versus the environment in longevity and the development of age-related diseases. Future studies including clinical examinations of cardiovascular function, serum measures of lipid profiles and markers of inflammation, and metabolic studies are warranted to gain a better understanding of the potential cardiovascular advantages in the offspring of centenarians. In addition, future studies using different comparison populations may help further our understanding of other age-related diseases such as cancer and cerebrovascular disease.
|
Acknowledgments |
|---|
Current location of authors: Dellara F. Terry, MD, MPH, Maegan A. McCormick, BS, and Thomas T. Perls, MD, MPH, Geriatrics Section, Department of Medicine, Boston Medical Center, 88 East Newton Street, F4, Boston, MA 02118; Marsha Wilcox, EdD, ScD, Boston University School of Medicine, Genetics Program, 715 Albany Street, L-320, Boston, MA 02118; Elizabeth Lawler, MPH, Boston University Gerontology Center/Maveric/Boston University School of Public Health, 52 Bay State Road, Boston, MA 02215.
Address correspondence to Dellara F. Terry, MD, MPH, The New England Centenarian Study, Geriatrics Section, Boston Medical Center, 88 East Newton Street, F-4, Boston, MA 02118. E-mail: laterry{at}bu.edu
Received September 18, 2002
Accepted November 26, 2002
|
References |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  C. B. Schechter Longevity and Cardiovascular Disease: Context and Overview Arch Intern Med, March 12, 2007; 167(5): 428 - 429. [Full Text] [PDF]
|
|
|
|
|
|
D. F. Terry, J. C. Evans, M. J. Pencina, J. M. Murabito, R. S. Vasan, P. A. Wolf, M. Kelly-Hayes, D. Levy, R. B. D'Agostino Sr, and E. J. Benjamin Characteristics of Framingham Offspring Participants With Long-lived Parents Arch Intern Med, March 12, 2007; 167(5): 438 - 444. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Melzer, A. J. Hurst, and T. Frayling Genetic Variation and Human Aging: Progress and Prospects J. Gerontol. A Biol. Sci. Med. Sci., March 1, 2007; 62(3): 301 - 307. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. T Perls The different paths to 100 Am. J. Clinical Nutrition, February 1, 2006; 83(2): 484S - 487S. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Zureik, P. Galan, S. Bertrais, D. Courbon, S. Czernichow, J. Blacher, P. Ducimetiere, M. E. Safar, and S. Hercberg Parental Longevity and 7-Year Changes in Blood Pressures in Adult Offspring Hypertension, August 1, 2005; 46(2): 287 - 294. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Karasik, S. Demissie, L. A. Cupples, and D. P. Kiel Disentangling the Genetic Determinants of Human Aging: Biological Age as an Alternative to the Use of Survival Measures J. Gerontol. A Biol. Sci. Med. Sci., May 1, 2005; 60(5): 574 - 587. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D Lio, G Candore, A Crivello, L Scola, G Colonna-Romano, L Cavallone, E Hoffmann, M Caruso, F Licastro, C M Caldarera, et al. Opposite effects of interleukin 10 common gene polymorphisms in cardiovascular diseases and in successful ageing: genetic background of male centenarians is protective against coronary heart disease J. Med. Genet., October 1, 2004; 41(10): 790 - 794. [Full Text] [PDF]
|
|
|
|
 |
|
 B. J. Geesaman, E. Benson, S. J. Brewster, L. M. Kunkel, H. Blanche, G. Thomas, T. T. Perls, M. J. Daly, and A. A. Puca Haplotype-based identification of a microsomal transfer protein marker associated with the human lifespan PNAS, November 25, 2003; 100(24): 14115 - 14120. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Chen Chasing 100 Sci. Aging Knowl. Environ., August 13, 2003; 2003(32): nf15 - 15. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
