SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. METHODS
  4. DATA ANALYSIS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments:
  8. References

Little has been reported on the association of derived body composition data and cardiovascular mortality. The authors defined body composition profiles based on one- and two-variable measures from dual-energy x-ray absorptiometry (DXA) total body scans. Scan results are labeled “apple” if Z score for percent of total fat in trunk is >0 and “pear” if Z score for height-corrected limb fat is ≥0. The fat measures were combined to define four body composition profiles: “pickle,”“avocado,”“mango,” and “barrel.” A third axis, the Z score of height-corrected limb lean tissue, is an index of skeletal muscle mass and was used to label subjects as “hard” or “soft.” Subjects (n=324) who were in good health from Malmö, Sweden, underwent body composition analysis using DXA and were followed for 10 years. The distribution of body composition profiles was similar for both genders and across age groups. Among subjects aged 50–74 years at baseline (n=116), there were 21 deaths. Barrel had the highest mortality rate: 13/39 (33.3%) mortality for barrels, compared with 8/77 (10.4%) mortality for non-barrels; mortality odds ratio, 3.2; 95% confidence interval, 1.45–7.08. The increased mortality was principally attributable to cardiovascular cause-related deaths. Soft (sarcopenia) was also associated with increased mortality (25.9%; p=0.05), but not cardiovascular cause-related deaths, whereas the total mortality among apples was not significantly increased but cardiovascular cause-related deaths were predominant (75%; p=0.02). The authors propose that DXA-body composition profiles can identify increased mortality risk of magnitude similar to major cardiovascular risk factors and may prove useful in health assessment.

Obesity, as defined by body mass index (BMI) ≥30 (measured as height [meters] divided by weight [kilograms] squared), is a worldwide epidemic and a major predictor of morbidity and mortality.1–5 Abdominal obesity is associated with the metabolic syndrome and its sequels, including increased cardiovascular mortality.6,7 Abnormal body fat distribution is increasingly recognized as a risk factor, even at normal BMI, and waist circumference is recommended to supplement the BMI in risk assessment. Regional body fat, lean mass, and bone mass can be accurately quantified using dual-energy x-ray absorptiometry (DXA), computed tomography, or magnetic resonance imaging. These methods all provide measures of visceral/truncal fat that correlate well with each other.8–10 Lean tissue by DXA in the trunk includes the viscera, whereas in the limbs it closely approximates skeletal muscle mass.11,12 DXA has the advantage over the other techniques of providing central and extremity measurements from a single scan at a very low radiation exposure.13 However, due in part to factors of historical association of and reimbursement exclusively for skeletal measurements of bone mineral density, DXA body composition (DXA-BC) has had little clinical application.14 Furthermore, the current format of total body DXA scan reports provides normative data only for bone and total fat mass, and there are no official guidelines for body composition reporting.

We propose a software-automated classification for DXA-BC, based on Z scores computed relative to the mean of normal controls matched by gender and 10-year age groups from 20 to 80+ years. The three DXA variables were selected for their previously reported association with mortality or cardiovascular risk. We set the “fat” plane with Z_%trunk_fat as the horizontal axis and Z_limb_fat on the vertical axis. Figure 1 offers a pictoral description of the body composition profiles designated by quadrant (avocado, mango, barrel, pickle). We categorized subjects by DXA-determined percent trunk fat alone to be apples if Z_%trunk_fat >0; that is, inclusive of mangos and barrels (Figure 1). A subject was categorized as pear if Z_limb_fat ≥0, inclusive of the avocados and mangos (Figure 1). The third axis, Z_limb_lean, is a measure of skeletal muscle mass and modifies the fat plane body composition profiles as “soft” (sarcopenia) or “hard” (above-average skeletal muscle mass).

image

Figure 1. Measured fat distribution profile. Apple body type (mango and barrel) are those with scans where Z_%trunk_fat>0; pear body type (avocado and mango) are those with scans where Z_limb_fat≥=0.

Download figure to PowerPoint

We were particularly interested in the barrel profile because this body habitus is suggestive of the adipose tissue distribution seen in extreme form in partial lipodystrophy, a condition that is associated with multiple cardiovascular risk factors.15–19 Furthermore, in a study of insulin resistance and dyslipidemia, both higher DXA-determined trunk fat and lower-leg fat were independently associated with a heightened risk for cardiovascular disease.20 In patients receiving chronic hemodialysis, a low ratio of limb to total lean tissue was a significant determinant of 5-year mortality.21 In this study, we tested the hypothesis that in a population cohort, the baseline DXA-determined barrel build of combined high trunk/total fat and low limb fat mass would be associated with an increased 10-year cardiovascular mortality. A secondary hypothesis was that mortality associations with single DXA variables; that is, sarcopenia (Z_limb_lean)—apples and pears—would be less strong than for fat plane quadrants (barrel profile).

METHODS

  1. Top of page
  2. Abstract
  3. METHODS
  4. DATA ANALYSIS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments:
  8. References

Subjects were 324 (173 women, 151 men) residents of Malmö, Sweden, aged between 20 and 89 years who underwent total body DXA scanning with a Lunar DPX (Lunar Corp., Madison, WI) instrument.22,23 Vital status after 10 years of observation was obtained from the Swedish national registry, along with dates of death for those deceased in the interim. The registry provided an International Classification of Diseases (ICD) 9 or ICD 10 (depending on year of death) cause of death code for those subjects deceased before 2000.

The variables analyzed included the measured scale weight and height, age at scanning, and gender, with all subjects being of Swedish ethnic background. DXA variables were bone mineral content (BMC), soft tissue mass, and fat mass for the total body, arms, legs, and trunk.

DATA ANALYSIS

  1. Top of page
  2. Abstract
  3. METHODS
  4. DATA ANALYSIS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments:
  8. References

The Z scores for body composition variables adjusted for gender and age by decade were computed from the study cohort using standard formulas. Among the 324 subjects, we principally sought to test our hypotheses in the 116 subjects aged 50–74 years at baseline (10-year mortality 18.1%) on account of the low observed 10-year mortality in the 140 subjects younger than age 50 years (2.1%) and high 10-year mortality in the 68 persons aged 75 years and older (61.8%). Due to sample size limitations, ICD codes for cause of death were designated as either cardiovascular or noncardiovascular. The mortality rates were compared between groups by χ2 tests for the quadrant fat plane body composition profiles and by the Fisher exact test for the single variable profiles. Odds ratios for mortality and survival were calculated with 95% confidence intervals (CIs) from 2 × 2 tables. According to the a priori hypotheses, the barrel, soft, and apple profiles would be associated with a greater subsequent (cardiovascular) mortality. Significance was set at p<0.05. All analyses were performed using SSPS for Windows (version 7.5, SPSS, Inc., Chicago, IL).

RESULTS

  1. Top of page
  2. Abstract
  3. METHODS
  4. DATA ANALYSIS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments:
  8. References

At baseline, the 324 subjects were nearly equally represented among the four fat plane body composition profiles (avocado 26.5%, mango 19.8%, pickle 23.8%, and barrel 29.9%; χ2 test for equal frequencies, p=0.065) and three single variable profiles (apple, pear, and soft). There was no gender difference among the profiles (Mann-Whitney U test of the four fat plane body composition profiles, p=0.53). The avocados and mangos had a predominance of higher than average muscle mass (60.5% and 59.4% hard, respectively), whereas the pickles and barrels where more often soft (63.6% and 58.8%, respectively; χ2 test for difference among the four groups, p=0.002). A low but significant correlation was found between the measures of limb fat and limb lean (r=0.25; p<0.001) and between percent trunk fat and limb fat (r=−0.13; p=0.024) but not between percent trunk fat and limb lean (r=0.007; p=0.9). The relative prevalence of body composition profiles did not differ significantly between age groupings (<50 years, 50–75 years,≥75 years at baseline).

During the 10-year follow-up, there were 21 deaths among the 116 subjects aged 50–74 years. Tables I and  II compare the anthropomorphic, basic DXA data, and body composition profiles between the 21 subjects who died and the 95 survivors. Only height showed a significant difference (survivors were shorter). Mortality differed for the fat plane quadrant body composition profiles with a significant excess mortality among barrels. In Table III, the 116 subjects were divided according to profiles and were found not to differ significantly by age or gender. Total mortality (second to last row in Table III) was significantly increased for barrel and soft but not apple profiles. As displayed in Figures 2 and  3, CIs for the odds ratio (OR) for mortality were significant for barrel body composition profile (OR, 3.2; 95% CI, 1.4–7.1) and borderline significant for soft (OR, 2.3; 95% CI, 1.0–5.3). The complementary OR for survival was significantly decreased for barrel (OR, 0.74; 95% CI, 0.59–0.94).

Table I. Ten-Year Mortality in 116 Subjects Aged 50–74 Years at Baseline Based on Anthropomorphic and Dual-Energy X-Ray Absorptiometry Data
 Deceased (n=21)Alive (n=95)Probability*
Age (years)62.9±7.260.8±8.00.3
Height (cm)173.2±9.4168.5±8.60.03
Weight (kg)73.1±12.571.5±12.50.6
Body mass index**24.4±3.325.5±3.30.2
Total lean (kg)50.2±11.948.1±12.30.5
Total fat (kg)20.7±6.322.1±7.80.5
Total bone mineral content (kg)2.66±0.662.52±0.520.3
Trunk/total fat0.53±0.0700.50±0.0660.08
Limb fat (kg)8.8±3.69.9±4.00.2
Limb lean (kg)21.71±5.6221.44±5.730.2
*For two-tailed, unpaired t test; **measured as weight (kilograms) divided by height (meters) squared
Table III. Body Composition Profiles and 10-Year Mortality in 116 Subjects Aged 50–74 Years at Baseline
ProfileAvocadoMangoPickleBarrelp*Applep**Pearp**Softp**,†
Subjects (n [%])35 (30.2)19 (16.4)23 (19.8)39 (33.6) 58 (50) 54 (46.6) 54 (46.5) 
Age (years)60.8±7.763.0±9.560.1±7.161.3±7.80.761.8±8.40.461.6±8.30.661.4±7.80.8
Height (cm)170.7±7.3164.5±5.8170.8±9.9170.8±9.90.059168.7±9.20.4168.5±7.40.3170.3±8.80.3
Weight (kg)77.7±12.774.9±10.564.1±10.569.4±11.4<0.00171.2±11.30.676.8±11.9<0.001‡68.3±9.80.005
Body mass index26.6±3.027.6±2.722.3±2.024.0±2.4<0.00124.9±3.00.926.9±2.90.923.5±2.3<0.001
Total lean (kg)50.3±11.944.5±9.645.7±11.550.4±13.60.248.5±12.71.048.2±11.41.045.5±9.90.012
Total fat (kg)26.3±6.328.7±5.616.4±4.617.6±5.5<0.00121.3±7.60.427.1±6.10.420.8±6.10.2
Total bone mineral content (kg)2.71±0.452.47±0.512.40±0.522.52±0.640.22.50±0.600.42.62±0.480.42.5±0.60.1
Trunk/total fat0.47±0.0590.51±0.0380.47±0.0580.55±0.062<0.0010.54±0.057<0.00††0.48±0.057<0.001‡‡0.51±0.0630.4
Limb fat (kg)12.5±3.512.6±3.17.7±2.57.1±2.8<0.0018.9±3.80.02†††12.5±3.30.02‡‡‡9.1±3.20.08
Limb lean (kg)22.7±5.019.2±4.320.7±5.722.0±6.60.221.1±6.10.421.1±6.10.419.8±4.70.003
Mortality (n [%])5 (14.3)1 (5.3)2 (8.7)13 (33.3)0.01914 (24.1)0.156 (11.1)0.0914 (25.9)0.054
Cardiovascular death (n [%])0 (0.0)1 (100)1 (50)8 (72.7)0.0419 (75.0)0.021 (16.7)0.0577 (53.8)0.9
*For analysis of variance for heterogeneity between the four groups, Pearson χ2 for mortality comparisons; **for unpaired t test, apple (pear, soft) vs. non-apple (non-pear, non-soft), Fisher exact test for mortality comparisons; †all significant differences except for mortality are for lower values soft; ††vs. trunk/total fat non-apple=0.46±0.059; †††vs. limb fat non-apple=10.6±3.92 kg; ‡vs. weight non-pear=67.4±11.3 kg; ‡‡vs. trunk/total fat non-pear=0.52±0.072; ‡‡‡vs. limb fat non-pear=7.31±2.65 kg
image

Figure 2. Odds ratio and 95% confidence intervals for 10-year mortality compared to the average mortality among the 116 subjects aged 50–74 years at baseline

Download figure to PowerPoint

image

Figure 3. Odds ratio and 95% confidence intervals for 10-year survival compared to the observed overall survival among the 116 subjects aged 50–74 years at baseline

Download figure to PowerPoint

Cause of death was available for 19 of the 21 deceased subjects (10 cardiovascular, nine other causes) with significant excess of cardiovascular deaths among deceased with barrel and apple profiles, but not for soft or pear (see Table III).

Among the fat plane quadrant body composition profiles, BMI (displayed in Table III) was significantly lower for pickle and barrel as it was for soft vs. non-soft. None of the single variables was significantly correlated with age at death, except height (r=−0.26; p=0.006).

The three limb tissue compartments were significantly correlated with each other for the entire study population and for the age 50–74 years segment. A high positive correlation was found between limb lean tissue and BMC (r=0.90; n=324; p<0.001)

with weaker negative correlation between lean and fat mass (r=−0.28; n=324; p<0.001) and between fat mass and BMC (r=−0.18; n=324; p=0.002).

Tables IV and  V display summary body composition variables for each gender by decade along with number of subjects and observed mortality at 10-year follow-up. The genders differed significantly for each of the body composition variables (p<0.001). Limb lean tissue mass was higher in men and declined with age by decade. In women the decline in limb lean tissue was less consistent and of a lower magnitude. In contrast, limb fat in women was consistently higher than in men and increased with age. Correspondingly, men had higher percent trunk fat than women, especially in the middle age group, (overall 52.5%±4.9% for men vs. 44.0%±5.4% for women; p<0.001). Although not shown in Tables IV and  V, BMI did not differ significantly by age group.

Table IV. Study Men (n=151) Age Stratified by Decade and Observed Mortality*
Age GroupTotalDeceasedBarrelsTrunk/TotalLimb LeanLimb Fat
(Years)(n)(n)(%)Fat(kg)(kg)
20–2925032.00.49±0.03729.7±4.05.5±2.8
30–3917217.60.50±0.02728.7±3.66.9±2.7
40–4921038.10.52±0.05527.9±4.27.3±3.2
50–5925540.00.56±0.04928.4±3.26.7±3.5
60–6924733.30.55±0.05025.5±3.68.2±3.2
70–79211223.80.52±0.03123.7±3.79.3±3.3
80–89181227.80.53±0.04023.3±3.17.8±2.0
p**   <0.001<0.0010.003
*Total mortality for men=38/151 (25.2%); **one-way analysis of variance for between age groups
Table Tablet V. Study Women (n=173) Age Stratified by Decade and Observed Mortality*
Age GroupTotalDeceasedBarrelsTrunk/TotalLimb LeanLimb Fat
(Years)(n)(n)(%)Fat(kg)(kg)
20–2929024.10.41±0.04018.6±2.28.4±2.4
30–3922013.60.42±0.05218.0±2.48.4±2.3
40–4926134.60.43±0.05519.0±3.211.2±4.3
50–5927229.60.44±0.05617.7±31.611.2±3.3
60–6918233.30.47±0.04216.4±2.411.7±3.6
70–79351034.30.47±0.04816.5±2.011.6±4.3
80–89161331.30.44±0.05517.2±2.511.0±4.4
p**   <0.001<0.0010.001
*Total mortality for women=28/173 (16.2%); **one-way analysis of variance for between age groups

DISCUSSION

  1. Top of page
  2. Abstract
  3. METHODS
  4. DATA ANALYSIS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments:
  8. References

Obesity is epidemic worldwide and is associated with enormous mortality, morbidity, and cost.24 It appears that a truncal fat distribution is important for predicting the cardiovascular and metabolic consequences of obesity.25–27 Identifying persons whose BMI suggests low risk but who are nevertheless at high mortality risk due to body build could help focus efforts for lifestyle modification and for aggressive medical intervention. We defined a spectrum of DXA-derived body composition profiles and found evidence that the barrel quadrant of higher than average percent trunk fat in conjunction with lower than average limb fat mass is associated with increased mortality predominately from cardiovascular causes. As defined, about one third of our study subjects had the barrel body composition profile. Unlike what is observed with abdominal obesity per se, our normative definition showed a uniformity of the fraction of barrels between ages and genders, suggesting perhaps a constitutional rather than an acquired trait. One might speculate about garden variants of the genetic defects of familial partial lipodystrophy.16,17 On the other hand, genetic predisposition for the increase in abdominal adiposity with age in men (as in our subjects) has been demonstrated, for example, with the DD ACE genotype.28

We found increased 10-year mortality with the barrel body composition, evidence that this body composition profile is harmful. The significant OR (3.2) we observed compares closely to relative mortality risk determinations for the metabolic syndrome in middle-aged men (OR, 2.9)7 and for low fitness in men and women (OR, 2.0, 2.2).25

A recent report of the cardiovascular risk association of abdominal obesity as measured by waist-hip ratio (apples) found that in a large population sample only those women in the lower tail of BMI values had increased event rates.26 In our much smaller population, we also found statistically significant evidence for increased mortality among subjects with the barrel body composition profile despite largely normal BMI (24.0±2.4). We suggest that our findings, rather than decreasing the concern for obesity as defined by BMI, emphasize that the population at cardiovascular risk associated with abnormalities of body composition includes normal levels of BMI.

Our data are consistent with studies of other populations demonstrating large structural differences between genders and with aging.29–31 We found increased overall (but not cardiovascular) mortality associated with sarcopenia in our normal subjects as reported in chronic hemodialysis.21

Our data also support a previously reported association of height and mortality.32,33 In fact, the mean correlation between life span and height found for six different populations was remarkably identical to our data (r=−0.26) and to another collection of 10 populations (r=−0.25) in unpublished data (T. T. Samaras, personal communication, 2002).

We recognize several limitations in our study. Although our subjects were from a homogenous population, our sample size was small. We hope our study will lead others to examine larger populations. Our subjects were not originally recruited for a mortality study and specific data on traditional risk factors and comorbidities are lacking. We cannot assess the degree of overlap with well defined traditional risk factors and our data leave open to speculation what leads to the observed association of barrel body composition with cardiovascular mortality. Possibly, relatively increased skeletal muscle mass, peripheral fat, or both attenuate the adverse metabolic consequences of central adiposity to account for the association of DXA measurements with cardiovascular risk factors and mortality.20 Current research findings show that DXA fat measurements are inversely associated with serum adiponectin, low levels of which seem to be involved in the pathogenesis of insulin resistance.34

DXA measured adiposity also explains a substantial percentage of age related left ventricular wall thickening, concentric remodeling, and diastolic dysfunction in otherwise healthy men.35 Despite the limitations of our study, we would hope that the method of body composition profile assessment derived directly from DXA total body scan and normative data will facilitate the interpretation of lean and fat tissue mass distribution. We feel that the mesenchymal assessment afforded by DXA-BC has the potential to effectively identify persons at increased cardiovascular risk who might otherwise not be targeted adequately for risk factor modification focused on BMI.

Based on our findings, subjects with DXA-defined sarcopenia and barrel body composition profile are at increased mortality risk and may be particularly likely to benefit from early intervention.36,37

Acknowledgments:

  1. Top of page
  2. Abstract
  3. METHODS
  4. DATA ANALYSIS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments:
  8. References

The authors thank M. Balasubramaniam, Department of Biostatistics, William Beaumont Hospital, Royal Oak, MI, for help with data handling and statistical advice; M.A. Letwak, MSN, NP, RN, for help with the figures; and H.S. Barden, PhD, T.M. Rifai, MD, and N.Y. Krakauer; MS, for critical reading of the manuscript.

References

  1. Top of page
  2. Abstract
  3. METHODS
  4. DATA ANALYSIS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments:
  8. References
  • 1
    Fontaine KR, Redden DT, Wang C, et al. Years of life lost due to obesity. JAMA. 2003;289:187193.
  • 2
    Peeters A, Barendregt JJ, Willekens F, et al. Obesity in adulthood and its consequences for life expectancy: a life-table analysis. Ann Intern Med. 2003;138:2432.
  • 3
    Mokdad AH, Ford ES, Bowman BA, et al. Prevalence of obesity, diabetes and obesity-related health risk factors, 2001. JAMA. 2003;289:7679.
  • 4
    Wilson PWF, D'Agostino RB, Sullivan L, et al. Overweight and obesity as determinants of cardiovascular risk. Arch Intern Med. 2002;162:18671872.
  • 5
    Kurth T, Gaziano MJ, Rexrode KM, et al. Body mass index and the risk of stroke in men. Arch Intern Med. 2002;162:25572562.
  • 6
    Ford ES, Giles WH, Dietz WH. Prevalence of the metabolic syndrome among US adults. JAMA. 2002;287:356359.
  • 7
    Lakka HM, Laaksonen DE, Lakka TA, et al. The metabolic syndrome and total and cardiovascular disease mortality in middle-aged men. JAMA. 2002;288:27092716.
  • 8
    Heymsfield SB, Wang ZM, Baumgartner RN, et al. Human body composition: advances in models and methods. Annu Rev Nutr. 1997;17:527558.
  • 9
    Wajchenberg BL. Subcutaneous and visceral adipose tissue: their relation to the metabolic syndrome. Endocr Rev. 2000;21:697738.
  • 10
    Wang SW, Wang Z, Punyanita M, et al. Adipose tissue quantification by imaging methods: a proposed classification. Obes Res. 2003;11:516.
  • 11
    Fuller NJ, Hardingham CR, Graves M, et al. Assessment of limb muscle and adipose tissue by dual-energy x-ray absorptiometry using magnetic resonance imaging for comparison. Int J Obes Relat Metab Disord. 1999;23:12951302.
  • 12
    Levine JA, Abboud L, Barry M, et al. Measuring leg muscle and fat mass in humans: comparison of CT and dual-energy x-ray absorptiometry. J Appl Physiol. 2000;88:452456.
  • 13
    Mazess RB, Barden HS, Bisek JP, et al. Dual-energy x-ray absorptiometry for total-body and regional bonemineral and soft-tissue composition. Am J Clin Nutr. 1990;51:11061112.
  • 14
    Albanese CV, Diessel E, Genant HK. Clinical applications of body composition measurements using DXA. J Clin Densitom. 2003;6:7585.
  • 15
    Garg A, Misra A. Hepatic steatosis, insulin resistance and adipose tissue disorders. J Clin Endocrinol Metab. 2002;87:30193022.
  • 16
    Hagele RA, Cao H, Anderson CM, et al. Heterogeneity of nuclear lamin A mutations in Dunnigan-type familial partial lipodystrophy. J Clin Endocrinol Metab. 1992;75:779782.
  • 17
    Agarwal AK, Garg A. A novel heterozygous mutation in perosixome proliferator-activated receptor-γ gene in a patient with familial partial lipodystrophy. J Clin Endocrinol Metab. 2002;87:408411.
  • 18
    Simha V, Garg A. Body fat distribution and metabolic derangements in patients with familial partial lipodystrophy associated with mandibuloacral dysplasia. J Clin Endocrinol Metab. 2002;87:776785.
  • 19
    Graber AL. Syndrome of lipodystrophy, hyperlipidemia, insulin resistance, and diabetes in treated patients with human immunodeficiency virus infection. Endocr Pract. 2001;7:430437.
  • 20
    Van Pelt RE, Evans EM, Schechtman KB, et al. Contributions of total and regional fat mass to risk for cardiovascular disease in older women. Am J Physiol Endocrinol Metab. 2002;282:E1023E1028.
  • 21
    Kato A, Odamaki M, Yamamoto T, et al. Influence of body composition on 5 year mortality in patients on regular haemodialysis. Nephrol Dial Transplant. 2003;18:333340.
  • 22
    Karlsson MK, Gardsell P, Johnell O, et al. Bone mineral normative data in Malmo, Sweden. Acta Orthop Scand. 1993;64:168172.
  • 23
    Karlsson MK, Obrant KJ, Nilsson BE, et al. Changes in bone mineral, lean body mass and fat content as measured by dual energy x-ray absorptiometry: a longitudinal study. Calcif Tissue Int. 2000;66:9799.
  • 24
    Franklin BA. The downside of our technological revolution? An obesity-conducive environment. Am J Cardiol. 2001;87:10931095.
  • 25
    Lee CD, Blair SN, Jackson AS. Cardiorespiratory fitness, body composition, and all-cause and cardiovascular disease mortality in men. Am J Clin Nutr. 1999;69:373380.
  • 26
    Kanaya AM, Shlipak M, Vittinghoff E, et al. Association between overall and abdominal obesity and coronary heart disease: the Heart & Estrogen/Progestin Replacement Study. J Gen Intern Med. 2002;17(suppl 1):113.
  • 27
    Gan SK, Kriketos AD, Ellis BY, et al. Changes in aerobic capacity and visceral fat but not myocyte lipid levels predict increased insulin action after exercise in overweight and obese men. Diabetes Care. 2003;26:17061713.
  • 28
    Strazzullo P, Iacone R, Iacoviello L, et al. Genetic variation in the renin-angiotensin system and abdominal adiposity in men: the Olivetti prospective heart study. Ann Intern Med. 2003;138:1723.
  • 29
    Nuti R, Martini C, Gennari C. Age-related changes of whole skeleton and body composition in healthy men. Calcif Tissue Int. 1995;57:336339.
  • 30
    Baumgartner RN, Stauber PM, McHugh D, et al. Cross-sectional age differences in body composition in persons 60+ years of age. J Gerontol A Biol Sci Med Sci. 1995;50:M307M316.
  • 31
    Reid IR, Plank LD, Evans MC. Fat mass is an important determinant of whole body bone density in premenopausal women but not in men. J Clin Endocrinol Metab. 2002;87:408411.
  • 32
    Samaras TT, Elrick H. Height, body size and longevity. Acta Med Okayama. 1999;53:149169.
  • 33
    Samaras TT, Elrick H, Storms LH. Is height related to longevity? Life Sci. 2003;72:17811802.
  • 34
    Gavrila A, Chan JL, Yiannakouris N, et al. Serum adiponectin levels are inversely associated with overall and central fat distribution but are not directly regulated by acute fasting or leptin administration in human: cross-sectional and interventional studies. J Clin Endocrinol Metab. 2003;88:48234831.
  • 35
    Gates PE, Gentile CL, Seals et al. Adiposity contributes to differences in left ventricular structure and diastolic function with age in healthy men. J Clin Endocrinol Metab. 2003;88:48844890.
  • 36
    Gan SK, Kriketos AD, Ellis BA, et al. Changes in aerobic capacity and visceral fat but not myocyte lipid levels predict increased insulin action after exercise in overweight and obese men. Diabetes Care. 2003;26:17061713.
  • 37
    Franklin BA. New AHA guidelines for primary prevention: “moving” toward cardiovascular risk reduction. Am J Med Sports. 2002;4:343346.