Parental Longevity and 7-Year Changes in Blood Pressures in Adult Offspring
In this report, we examined the cross-sectional and the 7-year longitudinal changes in blood pressures in adult offspring according to parental longevity. A population of volunteers free of symptomatic cardiovascular diseases who participated to the Supplementation en Vitamines et en Minéraux Antioxydants (SUVIMAX) Vascular Study (mean age 52.3 years; 48.3% women) were examined at baseline and 7 years later. Paternal (n=994) and maternal (n=896) longevity were analyzed separately. The prevalence of hypertension at baseline in subjects whose father died at <65 years of age, in those whose fathers were alive by age 65 but died by 80 years of age, and in those whose fathers were alive by age 80 was respectively 34.9%, 28.5%, and 20.2% (P<0.001). The means of systolic blood pressure in the 3 groups of paternal longevity were respectively 128.4 (±16.0), 125.3 (±14.2), and 123.6 (±14.4) mm Hg (P<0.001). During the follow-up, the mean systolic blood pressure increases in the 3 groups of paternal longevity were respectively 5.3 (±17.0), 4.2 (±14.0), and 1.6 (±13.2) mm Hg (P<0.001). In subjects without hypertension at baseline, hypertension occurred during the follow-up in 26.6%, 17.7%, and 15.3% (P<0.009), respectively. Multivariate analyses adjusted for baseline or changes in cardiovascular risk factors did not modify these results. In contrast, there was no relationship between maternal longevity and blood pressure measurements in either cross-sectional or longitudinal analyses. This study suggests that paternal premature death was associated with accelerated progression of systolic blood pressure and higher occurrence of hypertension in offspring. These results indicate that there are dynamic and continuous processes linking paternal longevity to blood pressure in adults.
Longevity, which may involve diverse biological processes that protect from a number of age-related diseases, has been observed to cluster in families.1,2 This phenomenon could be explained by shared genetic or environmental risk factors. However, these factors are largely unknown.
Increased systolic blood pressure (BP) and hypertension have been shown to be part of the most important determinants of cardiovascular morbidity and mortality and all-cause mortality.3–7 In addition, a consistent association between hypertension and cancer mortality has been reported in numerous population-based studies,8 although hypertension is probably a marker rather than a risk factor, per se, for cancer mortality in adults. Whatever, all these results suggest that hypertension and increased systolic BP are powerful predictors of premature death and decreased longevity.
In the few cross-sectional studies that previously reported the relationships of BPs with parental longevity, hypertension tends to be less frequent in offspring of parents who lived longer.9–14 To our knowledge, no longitudinal studies on this focus are available. Longitudinal studies are necessary to confirm and extend the cross-sectional findings and may indicate the timing and the patterns of BP changes according to parental longevity. In fact, it is not known whether the differences of BP measurements in offspring associated with parental longevity are constant and static or whether they are dependent on dynamic processes leading to modify these differences over time.
In the present report based on the longitudinal data of the Supplementation en Vitamines et en Minéraux Antioxygants (SUVIMAX) Vascular Study, we assessed the associations of parental longevity with changes in BP measurements and occurrence of hypertension during the 7-year follow-up in a population of middle-aged subjects free of cardiovascular diseases at baseline.
The design and methods of the SUVIMAX Vascular Study as well as characteristics of the participants have been described in detail previously.15 In brief, the SUVIMAX Vascular Study is a substudy of the SUVIMAX Study, which is a randomized double-blind, placebo-controlled, cardiovascular and cancer primary prevention trial, with follow-up over an average of 7.2±0.3 years.16 Eligible participants (free of symptomatic chronic diseases and apparently healthy) were randomly allocated to receive either a combination of antioxidants (120 mg vitamin C, 30 mg vitamin E, 6 mg beta carotene, 100 μg selenium, and 20 mg zinc) or placebo in a single daily capsule.
Eligible subjects for this specific substudy were participants living in the Paris area, >50 years of age in 2002, whose end-trial visit should take place in one facility (Conservatoire National des Arts Métiers [CNAM] center). These subjects were blindly examined between January and July 2002. Detailed cardiovascular risk factors assessment, carotid ultrasound examination, and pulse wave velocity measurement of this substudy protocol were added to the standard measures performed during the end-trial visit. All subjects gave their informed written consent to the study, which was approved by the Comité Consultatif de Protection des Personnes dans la Recherche Biomédicale.
Baseline and End-Trial Risk Factors Assessment
At baseline (1994 to 1995) and end-trial visits (2002), participants were asked through a standardized questionnaire to provide information about demographic background, occupation, medical history, and personal habits. The body mass index was computed as weight (in kilograms) divided by height (in meters) squared. Subjects were classified as never smokers, former smokers, or current smokers. Fasting cholesterol total, triglycerides, and glycemia were measured using enzymatic method (Technicon DAX).
During the standard physical examinations, BP was measured at baseline and end-trial visits with a random-zero sphygmomanometer. Lying brachial systolic and diastolic BPs were measured using a standard mercury sphygmomanometer in subjects who were lying down for 10 minutes. If systolic pressure was >160 mm Hg or diastolic pressure >100 mm Hg, pressures were remeasured after a second rest period of 5 minutes, and the lowest value was kept. Mean arterial pressure (MAP) and pulse pressure were calculated using the formula MAP=2/3 diastolic BP+1/3 systolic BP; pulse pressure=systolic BP−diastolic BP. For each examination, subjects using antihypertensive drugs or subjects who had systolic BP ≥140 mm Hg or diastolic BP ≥90 mm Hg were considered hypertensive. At end-trial visit, 2 additional BP measurements with a digital electronic tensiometer (model CP750; Omron) were obtained before and after the ultrasound examinations. Substituting the BP values obtained from the sphygmomanometers by those from the digital electronic tensiometer yielded very similar results (data available from authors).
At the end-trial visit, subjects were administered a standardized questionnaire about parental history. Subjects were asked, in separate questions, whether their mothers and fathers were still alive and their ages. If a parent was dead, each subject was asked at what age death had occurred. A total of 83.9% of the fathers and 58.7% of the mothers were dead. The mean age at death for the fathers was 70.7 (±14.3) years, and those who where still alive were an average of 84.1 (±5.8) years of age. The corresponding figure for mothers were 74.1 (±14.7) years, and 83.0 (±6.1) years, respectively.
For the present analysis, we classified the subjects into 3 groups according to the ages of their parents (at death or current age). Father and mother longevity classes were analyzed separately. Premature death in parents was defined as a paternal (or maternal) death age <65 years. A paternal (or maternal) current age of or death at >80 years was considered to represent increased paternal (or maternal) longevity. The intermediate group included subjects whose fathers (or mothers) were alive by age 65 but dead by age 80. The subjects whose parents were still alive and between ages 65 and 80 years (53 fathers and 151 mothers) were excluded from the analyses because their parents may live >80 years and therefore cannot be classified in the intermediate group.
Standard procedures from the Statistical Analysis System were used for univariate and multivariate analyses. Longitudinal changes in BP measurements (systolic, diastolic, MAP, and pulse pressure) were computed as the differences between end-trial and baseline values. Occurrence of hypertension during the follow-up was assessed in subjects without hypertension at baseline. In cross-sectional (baseline values of BPs) and longitudinal analyses, the associations between BP measurements and parental longevity groups were assessed by ANOVA. Multivariate cross-sectional and longitudinal associations (performed by analysis of covariance) were simultaneously adjusted for sex, supplementation group, educational level, and baseline age, body mass index, diabetes, hypercholesterolemia, and smoking habits. For each longitudinal BP variable, adjustment for the corresponding baseline values was also performed. Further adjustment in the longitudinal analyses for changes over time in cardiovascular risk factors yielded very close results. Therefore, only associations adjusted for baseline values of these factors are presented in this report.
The relationships of baseline hypertension status and occurrence of hypertension during the follow-up with parental longevity groups were assessed by χ2 and multiple logistic regression models. Multiple logistic regression models were adjusted for sex, supplementation group, educational level, and baseline age, body mass index, diabetes, hypercholesterolemia, and smoking habits. All reported P values are 2-tailed, and P<0.05 was considered significant.
Of the 1162 subjects included in the SUVIMAX Vascular Study, complete data of parental longevity, cross-sectional and longitudinal BP measurements, and baseline cardiovascular risk factors were available for 1047 subjects. At baseline, there were no significant differences in cardiovascular risk factors between subjects with and without complete data. After the exclusion of subjects whose parents were still alive and between 65 and 80 years of age, statistical analyses were performed on 994 subjects for paternal longevity and 896 for maternal longevity.
The percentage of subjects whose fathers died at <65 years of age, of those whose fathers were alive by 65 years of age but died by age 80, and of those whose fathers were alive by age 80 was respectively 27.4% (n=272), 36.7% (n=365), and 35.9% (n=357). The corresponding percentages for mothers were 15.2% (n=136), 25.5% (n=228), and 59.4% (n=532). Ages at death of fathers and mothers were positively but weakly associated (Spearman correlation coefficient 0.11; P<0.01).
High educational levels of offspring were associated with increased paternal longevity. Mean values of body mass index and triglycerides were higher in subjects with paternal premature death and lower in those with increased paternal longevity. Systolic BP, diastolic BP, and MAP were negatively associated with paternal longevity (Table 1). In multivariate models adjusted for age, sex, supplementation group, diabetes, hypercholesterolemia, and smoking habits, BP measurements (systolic, diastolic, or MAP), triglycerides, and education level but not body mass index were independently associated with paternal longevity. The multivariate-adjusted means of systolic BP (±SEM) in the 3 groups of paternal longevity were respectively 130.7 (±1.5), 128.5 (±1.4), and 126.1 (±1.4) mm Hg (P<0.001). The corresponding multivariate-adjusted means of diastolic BP in the 3 groups of paternal longevity were respectively 83.9 (±1.0), 82.4 (±0.9), and 81.5 (±0.9) mm Hg (P<0.001).
The prevalence of hypertension in the 3 groups of paternal longevity was respectively 34.9%, 28.5%, and 20.2% (P<0.001; Table 1). The multivariate odds ratio for hypertension in subjects with paternal premature death compared with those with increased paternal longevity was 1.94 (95% confidence interval, 1.33 to 2.88; P<0.001).
For maternal longevity, age, educational level, and body mass index were significantly associated with maternal longevity (Table 2). Neither BP measurements nor hypertension were associated with maternal longevity (Table 2).
During the follow-up, significant increases in systolic BP (3.4±14.6 mm Hg; P<0.001) and in pulse pressure (7.6±10.0 mm Hg; P<0.001) were observed, along with a significant decrease in diastolic BP (−4.1±9.1 mm Hg; P<0.001). MAP increased slightly over time (0.6±9.1 mm Hg; P=0.05).
The highest increase over time in systolic BP was observed in subjects with paternal premature death, and the lowest increase was in those with increased paternal longevity (Table 3). Changes in diastolic BP were not related to paternal longevity. Multivariate analyses adjusted for corresponding baseline BP variable or for conventional baseline cardiovascular risk factors did not modify these results (Table 3). Changes in pulse pressure and, to a lesser extent, in MAP were negatively associated with paternal longevity (Table 3). The latter association appeared only after adjustment for baseline MAP values (Table 3).
All these results were not markedly modified when subjects who had cardiovascular diseases at the end-trial examination (16 myocardial infarction and 12 stroke) were excluded from the analyses. Analyses repeated separately in men and women also yielded similar results.
Of the 762 subjects without hypertension at baseline, hypertension occurred in 143 subjects during the follow-up. The percentages of hypertension occurrence in subjects whose fathers died at <65 years of age, in those whose fathers were alive by 65 but died by age 80, and in those whose fathers were alive by 80 years were respectively 26.6%, 17.6%, and 15.3% (P<0.009). This relationship remained significant after multivariate adjustment. The multivariate odds ratio for hypertension occurrence in subjects with paternal premature death compared with those with increased paternal longevity was 1.67 (95% confidence interval, 1.18 to 2.72; P<0.01).
There were no significant differences in hypertension occurrence or in any BP measurements (systolic, diastolic, pulse, and MAP) across maternal longevity groups (Table 4).
When all cross-sectional and longitudinal analyses were repeated after the exclusion of subjects with antihypertensive medication at baseline (for cross-sectional analyses) and after the exclusion of those with antihypertensive medication at baseline or during the follow-up (for longitudinal analyses), the relationships of BP measurements with paternal and maternal longevity were of similar patterns to those observed in the whole population (Tables 5 and 6⇓).
The main findings of this large-scale longitudinal study conducted in middle-aged subjects are that paternal premature death was associated with markedly increased systolic BP in offspring in cross-sectional as well as in longitudinal analysis. Prevalence and occurrence of hypertension were also more frequent in subjects whose fathers died at <65 years of age. In contrast, no relationships between maternal longevity and BP measurements were observed in either cross-sectional or longitudinal analyses. To our knowledge, the present study is the first to report changes over time in BP measurements according to parental longevity.
Few cross-sectional studies have previously reported the relationships of BP with age at death of parents and longevity. Their results were similar to ours. In 1971, Hammond et al reported that self-reported “history of high BP” was more frequent in subjects with the “shortest-lived” parents (a group defined by both parents having died at <70 years of age).9 In the New England Centenarian Study, the offspring of centenarians, compared with controls, had a reduced prevalence of aging-associated diseases, particularly hypertension.10 In the National Heart, Lung, and Blood Institute Twin Study, twins whose fathers died at a younger age had unfavorable midlife cardiovascular risk factor levels (higher total cholesterol, ratio of total cholesterol to high-density lipoprotein, and systolic BP).11 In the PRIME Study (étude Prospective de l’Infarctus du MyocardE), systolic and diastolic BPs were lower in subjects whose both parents were alive at 80 years.12 These results were observed in France and Northern Ireland cohorts.12
Several studies have suggested that longevity has a familial pattern. Inherited susceptibility or shared environmental exposures could be involved in this phenomenon. Twin studies have shown some higher similarities in age at death for identical (monozygotic) twins, indicating a moderate genetic component to longevity.17,18 Environmental modifiable risk factors such as level of education and social classes are also known to be associated with mortality. In our report, educational level of offspring was related to parental longevity. However, controlling for this factor in the multivariate analysis did not modify the relationships of BP measurements with parental longevity. On the other hand, hypertension also has a familial component,19 but it is not clear whether the clustering of longevity could be explained by clustering of higher levels of BP or by some unidentified familial factors.
Arterial BP might be divided into 2 components: a steady component, which is the virtual pressure that might be present in the aorta and its major arteries during a given cycle if the cardiac output was nonpulsatile, and a pulsatile component (pulse pressure), which is the oscillation around the mean arterial BP.20 A large pulse pressure might occur by means of 3 principal factors: an increase in the velocity of ventricular ejection, a reduction in the visco-elastic properties of the arterial wall, and a modification in the timing of reflected waves.21,22 Our cross-sectional results emphasize the predominant effects of MAP, whereas longitudinal results show the accelerated progression over time of systolic and pulse pressure and, to a lesser extent, MAP in the offspring of fathers with premature death. The Framingham Study23 has shown that whereas systolic BP and pulse pressure increase markedly and exponentially with age, MAP is poorly influenced by age, and diastolic BP tends to fall at >55 years of age. These findings are consistent with several hemodynamic findings indicating that large artery stiffness is highly sensitive to age but mainly in central arteries, whereas age affects poorly muscular arteries and resistant arterioles.22 Our present data agree with these findings and suggest that the age of death of fathers might predict in the offspring not only the degree of large artery stiffness but also the steepness of the increase of systolic BP and pulse pressure with age and, therefore, the long-term occurrence of cardiovascular complications. Such hemodynamic changes may be modulated by genetic or environmental factors or a combination of both. The role of genetic polymorphisms such as those of the renin-angiotensin system has been emphasized recently24–26 and could even be triggered through changes in telomere length.27 The present data suggest not only the role of genetic factors, but they also do not exclude that environmental or associated factors may be involved. Indeed, traditional cardiovascular risk factors such as diabetes, dyslipidemia, or smoking do not seem to play a substantial role in this sample.
The reasons of the differential associations of BP with paternal and maternal longevity are unclear. Women have a longer life span than men, and different ages at death might be used in statistical analyses to define maternal and paternal premature deaths. When the cutoff of 70 years was used instead of 65 for maternal premature death, the lack of association between maternal longevity and BP measurements persisted. Another explanation of the discrepancy between results of fathers and mothers is that longevity in women might be a more heterogeneous phenotype involving many and different genetic and environmental factors. This heterogeneity may lead to the dilution of its eventual association with BP measurements. Another plausible explanation is that paternal premature death from cardiovascular diseases is more frequent than maternal premature death from cardiovascular diseases. This may explain why some cardiovascular risk factors, especially BP measurements, are more strongly associated with paternal premature death. This hypothesis is partially supported by our results. In fact, subjects were also administered a standardized questionnaire that gave the cause of death (cardiovascular causes: sudden death, myocardial infarction and stroke, cancer, and other causes). A total of 84 paternal premature cardiovascular-related deaths and only 28 maternal premature cardiovascular-related deaths have been reported. For paternal history, the highest baseline and the highest changes in BP measurements were observed in subjects with paternal premature cardiovascular mortality, followed by those with paternal premature noncardiovascular mortality and then by the other subjects. For example, the means of baseline systolic BP in the 3 groups were respectively 129.6 (±16.2), 128.0 (±15.9), and 124.4 (±14.3) mm Hg (P<0.001). For maternal history, the subjects with maternal premature cardiovascular mortality had significantly higher BP measurements than those with maternal premature noncardiovascular mortality or than the other subjects. However, because of the low number of subjects with maternal premature cardiovascular-related deaths, these results should be interpreted with caution.
Although all middle-aged adults should be encouraged to be screened for the presence of major cardiovascular risk factors (hypertension, diabetes, obesity, dyslipidemia, etc) regardless of their parental longevity, paternal longevity may be considered, if our results are confirmed, as a useful additional marker for screening adult offspring who may be at higher risk of developing hypertension over time and may help determine the appropriate aggressiveness of risk factor modifications. Considering the high frequency of hypertension in the population and the major role of hypertension on morbidity and mortality from several conditions, this would be of particular interest in terms of prevention and intervention. Physicians should also be aware that the control of hypertension in such subjects should be more rigorous than usual.
In conclusion, this study suggests that increased paternal longevity was associated with a low frequency of hypertension and decreased systolic BP in cross-sectional as well as in longitudinal analysis. These results may indicate that there are dynamic and continuous processes linking paternal longevity to BP in adult offspring.
The SUVIMAX project was supported by a grant from the National Institute of Health and Medical Research. It received also supports from several public and private sectors.
- Received March 18, 2005.
- Revision received April 5, 2005.
- Accepted May 26, 2005.
Pearl R. Studies on human longevity. IV. The inheritance of longevity, preliminary report. Hum Biol. 1931; 3: 245–269.
Perls TT, Wilmoth J, Levenson R, Drinkwater M, Cohen M, Bogan H, Joyce E, Brewster S, Kunkel L, Puca A. Lifelong sustained mortality advantage of siblings of centenarians. Proc Natl Acad Sci U S A. 2002; 99: 84428447.
Psaty BM, Furberg CD, Kuller LH, Cushman M, Savage PJ, Levine D, O’Leary DH, Bryan RN, Anderson M, Lumley T. Association between blood pressure level and the risk of myocardial infarction, stroke, and total mortality: the cardiovascular health study. Arch Intern Med. 2001; 161: 1183–1192.
Hammond EC, Garfinkel L, Seidman H. Longevity of parents and grandparents in relation to coronary heart disease and associated variables. Circulation. 1971; 43: 31–44.
Terry DF, Wilcox M, McCormick MA, Lawler E, Perls TT. Cardiovascular advantages among the offspring of centenarians. J Gerontol A Biol Sci Med Sci. 2003; 58: M425–M431.
Yarnell J, Yu S, Patterson C, Cambien F, Arveiler D, Amouyel P, Ferrieres J, Luc G, Evans A, Ducimetiere P. Family history, longevity, and risk of coronary heart disease: the PRIME Study. Int J Epidemiol. 2003; 32: 71–77.
Zureik M, Galan P, Bertrais S, Mennen L, Czernichow S, Blacher J, Ducimetiere P, Hercberg S. Effects of long-term daily low-dose supplementation with antioxidant vitamins and minerals on structure and function of large arteries. Arterioscler Thromb Vasc Biol. 2004; 24: 1485–1491.
Nicholas WW, O’Rouke MF. Properties of the arterial wall. In: McDonald’s Blood Flow in Arteries: Theoretical, Experimental, and Clinical Principles. 3rd ed. London, UK: Edward Arnold; 1990: 77–114.
Franklin SS, Gustin W IV, Wong ND, Larson MG, Weber MA, Kannel WB, Levy D. Hemodynamic patterns of age-related changes in blood pressure. The Framingham Heart Study. Circulation. 1997; 96: 308–315.
Mourad JJ, Ducailar G, Rudnicki A, Lajemi M, Mimran A, Safar ME. Age-related increase of pulse pressure and gene polymorphisms in essential hypertension: a preliminary study. J Renin Angiotensin Aldosterone Syst. 2002; 3: 109–115.
Safar ME, Lajemi M, Rudnichi A, Asmar R, Benetos A. Angiotensin-converting enzyme D/I gene polymorphism and age-related changes in pulse pressure in subjects with hypertension. Arterioscler Thromb Vasc Biol. 2004; 24: 782–786.
Benetos A, Okuda K, Lajemi M, Kimura M, Thomas F, Skurnick J, Labat C, Bean K, Aviv A. Telomere length as an indicator of biological aging: the gender effect and relation with pulse pressure and pulse wave velocity. Hypertension. 2001; 37: 381–385.