| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Integrative Physiology, University of Colorado (P.E.G., C.L.G., D.R.S., D.D.C.), Boulder, Colorado 80309; and Department of Medicine (Cardiology and Geriatric Medicine), University of Colorado Health Sciences Center (D.R.S.), Denver, Colorado 80262
Address all correspondence and requests for reprints to: Phillip E. Gates, Ph.D., Human Cardiovascular Research Laboratory, Department of Integrative Physiology, 354 UCB, University of Colorado, Boulder, Colorado 80309. E-mail: phillip.gates{at}colorado.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
One such risk factor that worsens with age is adiposity. With advancing age, total and visceral adiposity increases, and the distribution of fat is characterized by more abdominal (central) relative to peripheral fat storage (3, 4). Increased total and visceral adiposity are associated with altered LV structure and function (5, 6, 7, 8) and elevated CVD risk (9, 10, 11). As such, we reasoned that increases in adiposity and/or shifting fat distribution with age may contribute to age-associated changes in LV structure and diastolic function.
The primary purpose of this study, therefore, was to examine the influence of adiposity in the association between age and LV structure and diastolic function in a cohort of healthy men over a broad age range. Our hypothesis was that the association of age with LV structure and diastolic function would be partially explained by adiposity. A secondary purpose was to identify which phenotypic expression of adiposity, i.e. percentage total, percentage abdominal, or fat distribution (abdominal/thigh fat ratio), contributed the most to the association of age with LV structure and diastolic function.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
A total of 113 men, aged 20–79 yr, were studied. Subjects were either sedentary or recreationally active. All subjects were normotensive at rest (seated systolic/diastolic, <140/90 mm Hg) and free of overt coronary artery disease, as assessed by medical history, physical examination, resting electrocardiogram, blood chemistry, and hematological evaluation (e.g. plasma glucose concentration <7.8 mmol/liter and total cholesterol <6.2 mmol/liter). Men older than 40 yr were further evaluated using electrocardiogram and blood pressure responses to incremental treadmill exercise performed to exhaustion. Subjects were not smoking or taking any medication. All procedures were approved by the human research committee of the University of Colorado (Boulder, CO). The nature, benefits, and risks of the study were explained to the volunteers, and their written informed consent was obtained before the study.
Study procedures
All measurements were performed while abstaining from caffeine and after fasting for at least 4 h (a 12-h overnight fast was required for determination of metabolic risk factors). Immediately before the testing session, each subject rested supine for at least 15 min in a quiet, temperature-controlled, semidarkened room.
Body composition
Body weight was measured to the nearest 0.1 kg with a physician’s balance scale. Whole body composition was determined using dual energy x-ray absorptiometry (DXA; software version 4.1, DPX-IQ, Lunar Radiation Corp., Madison, WI) with subjects in the supine position (12). Although DXA is unable to distinguish between visceral and sc fat depots, recent studies indicate that there is a strong correlation between abdominal fat estimated from regional analysis of DXA scans and visceral fat directly measured by magnetic resonance imaging (r = 0.84–0.89) (13, 14) and computed tomography (r = 0.66–0.90) (15). Regional adiposity was determined by manually defining specific regions of interest for the abdominal, and thigh areas: abdomen, from the upper edge of the first lumbar vertebra to the anterior superior iliac spine; thigh, from (but not including) the ischial tuberosities to the midpoint between the greater trochanter and the knee joint line. From all regions, fat, lean, bone, and total mass (kilograms) were calculated. Total adiposity (percentage) was determined as the portion of total fat to total body mass. Abdominal adiposity (percentage) was determined as the portion of fat in the abdominal region to the total mass in that region. To determine the association of body fat distribution, i.e. central vs. peripheral, to LV structure and diastolic function, we examined the abdominal to thigh fat ratio [fat in abdomen (kilograms)/fat in thigh (kilograms)]. All DXA scans were analyzed by the same investigator (C.G.). The scans of 10 representative subjects were analyzed 3 times for total and abdominal fat content and body fat distribution to assess intraobserver reliability. For each adiposity measure we calculated the coefficient of variation and performed intraclass reliability analysis. This demonstrated that intraobserver reliability was very high; coefficients of variation ranged from 0.1–3%, and intraclass reliability ranged from 0.9905-1.
Left ventricular structure and diastolic function
Ultrasound echocardiography (Toshiba SSH-140 interfaced with a 3.5-MHz transducer; Toshiba, Tochigi, Japan) images of the LV were obtained from the parasternal and apical windows with the subject in the left lateral decubitus position. Examinations were recorded on super-VHS videotape, and measurements were made using digitization software integral to the ultrasound system (mean of five consecutive cardiac cycles). Measurements of systolic and diastolic chamber dimensions and wall thickness were made from M-mode images following established guidelines (16). The ratio of LV posterior wall thickness to chamber radius (h:R), an index of LV morphology and concentric remodeling, was calculated from M-mode diastolic dimensions. Left ventricular mass was calculated using a validated formula (17). Transmitral blood flow from the left atrium to the LV was measured using pulsed Doppler to determine early (E) and late (A) diastolic peak filling velocity. The ratio of E to A velocity (E:A) was calculated and used as an index of diastolic function.
Scaling cardiac data to body composition
Left ventricular dimensions and mass are subject to the influence of body mass and body composition. To compare groups with different body masses and compositions, it is important to partition out this influence by using an appropriate scaling method. Studies from a variety of populations, as well as epidemiological data, have shown that fat-free mass (FFM) is a more appropriate variable for scaling than height, body mass, or body surface area for normalizing left ventricular wall thickness and mass (18, 19, 20, 21). Because fat free mass tends to be lower in older men, we attempted to partition out this influence by normalizing measurements of LV structure to FFM determined from DXA (described in detail above). In the subsample analyses (details below), absolute data were used because FFM was not different between the age-matched groups of men differing in adiposity.
Data analysis
Statistical analyses were performed using the SPSS (version 11.0, Chicago, IL) statistical package. To determine the bivariate relation of age to LV structure and function, we used Pearson zero order correlation coefficients. To determine the unique contribution of age and adiposity to LV structure and function, we used multiple linear regression analyses. Separate models were used for each measure of LV structure (LV wall thickness), LV morphology (h:R), and diastolic function (E:A) and each measure of adiposity (total and abdominal fat and abdominal/thigh fat). Measurements of LV structure and diastolic function were used as the dependent variables, whereas age and measurements of adiposity were used as the independent variables. Part correlation coefficients derived from the regression analysis were used to determine the independent contributions of age and adiposity to LV structure and diastolic function. Residual analyses to test the validity of the regression model assumptions were performed for all regression models.
To test our hypothesis further, we retrospectively analyzed separate subsamples of 20 older and 20 younger men. Older men were above the mean age of the entire cohort, and younger men were below it. Within each cohort, subjects were matched for age and systolic, diastolic, and mean arterial blood pressures, but were divided into groups of lower and higher adiposity. The purpose of this complementary analysis was to isolate adiposity as an independent variable that differed between groups of similar age. Within each age subsample, subjects with lower adiposity were compared with subjects with higher adiposity using t tests for independent samples. Statistical significance was set at P < 0.05 for all analyses. All data are reported as the mean ± SE.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Selected subject characteristics are presented in Table 1
, and mean values of LV structure and diastolic function are presented in Table 2. Aging was associated with increased levels of adiposity (percent total body fat, r = 0.54; percentage abdominal fat, r = 0.56; P < 0.0001) and central fat distribution (abdominal/thigh fat ratio, r = 0.55; P < 0.0001).
|
|
LV wall thickness and h:R increased with age, indicating LV concentric remodeling. Bivariate correlations of age with these measurements were 0.55 and 0.37, respectively (Fig. 1, A and B, 
). The correlation of LV mass with age was 0.23. Diastolic function was reduced with increasing age; the E:A-age correlation was -0.72 (Fig. 1C, ). All correlations were significant (at least P < 0.05).
|
Having established that age was associated with adiposity and differences in both LV structure and function, we determined the bivariate relation between adiposity and LV structure and function (Fig. 2). Mean LV wall thickness was related to all three expressions of adiposity (Fig. 2A), as was the h:R (Fig. 2B) and E:A (Fig. 2C). LV mass did not correlate with any of the expressions of adiposity. After controlling for age, the relation of each expression of adiposity with LV structure and diastolic function was still significant but reduced (mean wall thickness by 51–64%; h:R by 36–46%; and E:A by 56–69%).
|
) were reduced after controlling for percentage total body fat (by 38%), percentage abdominal fat (by 42%), and the abdominal/thigh fat ratio (by 35%). The age-h:R correlation (Fig. 1B) was reduced after controlling for percentage total body fat (by 54%), percentage abdominal fat (by 62%), and the abdominal/thigh fat ratio (by 46%). The diastolic function-age correlation (Fig. 1C) was reduced after controlling for percentage total body fat (by 35%), percentage abdominal fat (by 39%), and the abdominal/thigh fat ratio (by 29%). All three measures of adiposity significantly contributed to the association of age with LV wall thickness, E:A, and h:R (Fig. 1). There were no apparent differences in the contribution of percentage total body fat, percentage abdominal fat, or abdominal/thigh fat to the association of age with LV structure/diastolic function (Fig. 1). The outcome of all statistical analyses was similar if absolute (kilograms) total body and abdominal fat mass were used in place of percentage total body and abdominal fat. Subsample analysis
Age, systolic, diastolic and mean arterial blood pressures, and FFM were similar in the higher vs. lower adiposity subgroups of the older (Table 3) and younger (Table 4) men, whereas the age-matched subgroups differed in percent body fat, percent abdominal fat, and abdominal/thigh fat ratio. Consistent with the results of the regression analysis on the entire subject sample, in older men with higher vs. lower adiposity, mean LV wall thickness and h:R were greater, left ventricular mass was similar, and E:A was lower (Fig. 3). In contrast, in the young controls there were no differences in LV structure or diastolic function between subgroups differing in adiposity (Fig. 3).
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Age, adiposity, and LV structure and diastolic function
Our results are consistent with earlier findings establishing age-associated changes in LV structure and diastolic function in healthy men (1). Our data also support an independent association between total body and abdominal adiposity and the LV in this cohort of healthy men, in agreement with previous findings of an independent association of adiposity and LV structure and function in obese subjects (6, 7, 8, 22, 23, 24, 25, 26, 27). The present results extend these earlier findings by establishing for the first time a significant role for adiposity in the LV remodeling and impaired diastolic function observed with primary human aging. Indeed, our data show that the relation between age and LV structure (wall thickness and concentric remodeling) was reduced by 35–62% after controlling for adiposity phenotypes. Similarly, the age-diastolic function relation was reduced by 29–39% after controlling for the three expressions of adiposity.
In agreement with the analysis of the entire cohort of men, our analysis of subsamples of age-matched men differing in adiposity showed that older men with greater adiposity had significantly greater LV wall thickness, concentric remodeling, and reduced diastolic function compared with older men with lower adiposity. In younger men, however, LV structure and diastolic function were similar between groups with higher and lower levels of adiposity. This suggests that adiposity may exert an influence on LV structure and diastolic function, but that aging is a prerequisite, because this process is time dependent. This is consistent with Lakatta’s (2) suggestion that changes to the LV are mediated by increased exposure time to cardiovascular risk factors.
A secondary purpose of our study was to identify which phenotypic expression of adiposity contributes the most to the association of age with LV structure and diastolic function. Our findings suggest that total and abdominal adiposity exert similar influences on age-associated changes in LV structure and diastolic function.
Underlying mechanisms
The mechanisms by which adiposity influences LV structure and diastolic function have not been established. However, the ability of adipose tissue to act as an endocrine organ may provide a clue in this regard. In particular, the bioactivity of the renin-angiotensin system (RAS), both systemically and within adipose tissue, is elevated in obesity (28, 29, 30) and is reduced after weight loss (31, 32). Furthermore, the gene expression of various components of the RAS is augmented in overweight and obese subjects (33, 34, 35). Importantly, angiotensin II, a key cardiovascular-acting peptide of the RAS, is a potent growth factor for cardiac myocytes, is proinflammatory, and is associated with cell proliferation, hypertrophy, apoptosis, and fibrosis (36, 37, 38, 39, 40). Angiotensin II also plays an important role in LV remodeling and LV hypertrophy (20, 41, 42). Greater RAS activity associated with age-associated increases in adiposity may therefore provide a pathophysiological link among adipose tissue, aging, and changes in the LV.
The present findings clearly indicate that adiposity is one factor that contributes to age-associated changes in the LV. However, our analysis also shows that age contributes to changes in LV structure and diastolic function independently of adiposity. Several other factors may also contribute to LV aging. These include greater arterial stiffness, increased arterial blood pressure, greater aortic input impedance, changes to the LV extracellular matrix, and changes to cellular calcium release, excitation-relaxation coupling, and calcium reuptake. It is likely that there is interplay between all of these factors and ventricular aging, but this has not been determined.
Limitations
Our study used a cross-sectional design to examine associations between outcome measures of interest across age. We recognize that a longitudinal design in which serial measurements are performed in the same subjects over time may have yielded different results. However, results from cross-sectional and longitudinal study designs of such associations, including those from our laboratory (43, 44, 45), generally agree. Second, we were not able to measure sc and visceral fat separately. However, several studies have shown that increased abdominal adipose tissue is closely associated with increased visceral fat (13, 14, 15). As such, in this initial demonstration of a link between adiposity and age-associated changes in the LV we do not believe that the absence of direct estimation of visceral fat fundamentally affects our main conclusions.
Clinical implications
We believe that our findings have important clinical relevance. For example, the present results suggest that adiposity could play a significant role in the increased CVD risk associated with aging by contributing to adverse changes in LV structure and diastolic function in healthy men. As such, controlling adiposity throughout the adult life span may attenuate LV remodeling and the decline in diastolic function that is a characteristic of primary human aging. Accordingly, our findings provide further justification for the idea of age-associated obesity as a therapeutic target in prevention of the increased prevalence of CVD that occurs with advancing age. In support of this concept, several studies have shown that LV structure and diastolic function are improved after weight loss in obese patients (46, 47, 48, 49).
Conclusion
We conclude that increases in adiposity contribute to the LV remodeling and reduced diastolic function that occur with primary aging in healthy men. Our findings provide additional justification for effective control of body weight in the maintenance of cardiovascular health and functional capacity with adult aging.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: CVD, Cardiovascular disease; DXA, dual energy x-ray absorptiometry; E:A, ratio of early (E) and late (A) diastolic peak filling velocity; FFM, fat-free mass; h:R, ratio of LV posterior wall thickness to chamber radius; LV, left ventricle; RAS, renin-angiotensin system.
Received April 23, 2003.
Accepted July 2, 2003.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
B in aorta of normal rats and in vascular smooth muscle cells of AT1 knockout mice. Nephrol Dial Transplant 16(Suppl 1):27–33
This article has been cited by other articles:
 |
|
 |
|
  L. Mazzaro, S. J. Almasi, R. Shandas, D. R. Seals, and P. E. Gates Aortic Input Impedance Increases With Age in Healthy Men and Women Hypertension, June 1, 2005; 45(6): 1101 - 1106. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. R. Seals and P. E. Gates Stiffening Our Resolve Against Adult Weight Gain Hypertension, February 1, 2005; 45(2): 175 - 177. [Full Text] [PDF]
|
|
|
|
 |
|
 D. D. Christou, P. Parker Jones, A. E. Pimentel, and D. R. Seals Increased abdominal-to-peripheral fat distribution contributes to altered autonomic-circulatory control with human aging Am J Physiol Heart Circ Physiol, October 1, 2004; 287(4): H1530 - H1537. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. E. Gates, H. Tanaka, W. R. Hiatt, and D. R. Seals Dietary Sodium Restriction Rapidly Improves Large Elastic Artery Compliance in Older Adults With Systolic Hypertension Hypertension, July 1, 2004; 44(1): 35 - 41. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. P. Davy and J. E. Hall Obesity and hypertension: two epidemics or one? Am J Physiol Regulatory Integrative Comp Physiol, May 1, 2004; 286(5): R803 - R813. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. A. Tritos, W. J. Manning, and P. G. Danias Adiposity Contributes to Differences in Left Ventricular Structure and Diastolic Function with Age in Healthy Men J. Clin. Endocrinol. Metab., March 1, 2004; 89(3): 1485 - 1485. [Full Text] [PDF]
|
|
|
|
|
|
P. E. Gates, C. L. Gentile, D. R. Seals, and D. D. Christou Authors' Response: Adiposity Contributes to Differences in Left Ventricular Structure and Diastolic Function with Age in Healthy Men J. Clin. Endocrinol. Metab., March 1, 2004; 89(3): 1485 - 1486. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
), abdominal percentage fat (
), and central fat distribution (A:T; 
). All correlation coefficients are significant (P < 0.05) except the relation of age with h:R ratio independent of abdominal percentage fat.
