An important function of the large arteries is to cushion the pulsatile blood flow from the left ventricle, thereby reducing the afterload imposed on the heart. Ratios of change in blood pressure and change in arterial diameter with the cardiac cycle (e.g., arterial distensibility and the elastic modulus) quantify this cushioning function. A decrease in arterial distensibility is recognized as a potential marker of subclinical cardiovascular disease (CVD), is associated with numerous well-established CVD risk factors (male gender, age, dyslipidemia, insulin resistance, and type 2 diabetes), and is thought to accompany the initiation and/or progression of hypertension and atherosclerosis (3).
Habitual physical activity may retard the atherosclerotic process by reducing established CVD risk factors, including blood lipids, hypertension, hyperinsulinemia, and obesity, as well as slowing the aging of the arterial wall (3,8,9,16,28,32). In addition, vigorous exercise training is associated with beneficial structural and functional changes to the myocardium (12). Therefore, habitual physical activity, especially if performed at high intensity, may improve arterial distensibility through improvements in other CVD risk factors or through a direct impact on the large arteries. The few studies that assessed the possible association of habitual physical activity or &OV0312O2max with arterial distensibility or stiffness were conducted using animals and/or small samples of volunteers (7,17,29,30,33,34). To our knowledge, there are no studies of these associations in a population-based epidemiologic cohort. We tested the hypothesis that arterial distensibility is directly associated with self-reported habitual work, leisure, or sport physical activity in a cross-sectional analysis of the participants from the Atherosclerosis Risk in Communities (ARIC) study. Of specific interest was whether any observed association was independent of the known inverse association of habitual physical activity with arterial blood pressure (2,22). We also evaluated whether the associations differed by race, sex, or smoking status, as well as intensity level (nonvigorous vs vigorous) or pattern of physical activity.
The ARIC study is a prospective study of the etiology of atherosclerosis in a biracial, population-based cohort of men and women aged 45–64 yr in 1987–1989. The four clinical centers include Jackson, Mississippi (all African-Americans); Forsyth County, North Carolina; northwest suburbs of Minneapolis, Minnesota; and Washington County, Maryland. Details of study design and recruitment are published elsewhere (31). Carotid arterial pulsatile diameters were measured in 11,479 ARIC participants (9% at the first clinic visit, 91% at the second clinic visit). We excluded participants missing physical activity assessment (N = 53) and those with prevalent CVD at the time of the ultrasound examination of the carotid artery (N = 749), defined as a history of stroke or angina pectoris; self-reported history of myocardial infarction (MI); electrocardiographic evidence of MI; or physician documentation of MI, coronary angioplasty, carotid endarterectomy, or coronary bypass surgery. The target population for the inferences from this study is therefore limited to those without prevalent CVD. The final sample size for this analysis was 10,644 participants free of CVD.
Carotid artery ultrasound assessments.
Details of data collection and evaluation methods for common carotid arterial diameters have been published elsewhere (20,26). In brief, centrally trained and certified technicians used noninvasive B-mode ultrasound to measure mean intima-media wall thickness (IMT) of the extracranial left and right carotid arteries. In addition, continuous measurements of the left common carotid arterial diameters were made at the end of the B-mode ultrasound examination with an ultrasonic echo-tracking radiofrequency device (AUTREC 4881-AWT, Winston-Salem, NC). This wall tracking device electronically measured the pulse-echo velocity from the near to far wall media-adventitia borders, from diastole to systole, 1 cm proximal to the origin of the bifurcation in the left common carotid artery. At the time of the diameter measurements, participants had been resting in a supine position for at least 20 min. A mechanical holder steadied the transducer while the sonographer changed the transducer angle to maximize media-adventitia echoes. Electronic gates were positioned to track the near to far wall interfaces for the precise on-line measurement of the time interval between the arrival of the near and far wall echoes. The arterial diameter was displayed on an oscilloscope screen and printed on a strip chart recorder that the sonographer reviewed during the assessment. The digitized diameter data and the strip chart were sent to the ARIC Ultrasound Reading Center, where systolic and diastolic arterial diameters were calculated and averaged over as many cardiac cycles as possible with a maximum of 10 consecutive cardiac cycles (average, 5.5). ARIC study ultrasound readers were centrally trained and certified, periodically retrained and recertified, and subject to regular quality control monitoring (20,26). Arterial diameter change over the cardiac cycle was calculated as systolic arterial diameter minus diastolic arterial diameter. This change in diameter represents the intrinsic ability of the artery to respond to stress, where stress is defined as force per unit area or ((transmural pressure × vessel radius) ÷ vessel wall thickness). Arterial diameter change can be adjusted for blood pressure (a primary determinant of stress) in an ANCOVA model to allow for assessment of the potential association of arterial distensibility and habitual physical activity independent of blood pressure. Supine brachial arterial blood pressure was assessed immediately before and after the diameter measurements using a Dinamap automated blood pressure cuff (1846SX, Dinamap, Tampa, FL). Pulse pressure was defined as the difference between systolic and diastolic blood pressure.
Physical activity was assessed at the first ARIC visit using a slightly modified version of the Baecke physical activity questionnaire (4,10). The three ordinal indices that resulted from the Baecke survey ranged from low (1) to high (5) for physical activity from work, leisure, and sports. The work index included eight questions about self-rating of work’s vigor (compared to others of the same age), as well as the frequency of sitting, standing, walking, lifting, carrying heavy loads, sweating, and fatigue experienced after work. The work index was counted as missing for nonworkers. The leisure index included four questions that focused on frequency of television viewing and light-intensity activities: walking, bicycling, walking, or bicycling to or from work or shopping. The sport index was a sum of the yearly frequency, weekly duration, and coded intensity (low, medium, or high) of up to four self-reported sport activities, as well as self-rated amount of leisure time activity compared with others of the same age, frequency of sweating, and general frequency of sport play. For all quartile analyses, the physical activity indices were divided into quartiles on the basis of the entire ARIC sample distribution. In addition, participation in “regular” physical activity was defined as reporting a given sport activity for at least 1 h·wk−1 for 10 or more months of the past year. Vigorous activity was defined as those with an intensity of greater than 5 METs or greater according to intensity codes assigned to individual sports activities using standard published tables, as previously described (10). The cut-point of 5 METs for vigorous activity has been used in previous ARIC publications (10,22). The associations we report herein were not altered when we repeated analyses with the cut-point shifted to 6 METs.
The Baecke survey has been reported to have short- to long-term test-retest correlations of r = 0.65 to r = 0.93 (1,15,24). Correlation of the Baecke survey indices with maximal aerobic capacity has been reported to be r = 0.23 for the work index, r = 0.52 for the sport index, and r = 0.26 for the leisure index (15). These values compare favorably with reliability and validity of other physical activity measurement instruments (15).
Body mass index (BMI) was computed as weight in kilograms divided by height in meters squared. Smoking, education, medication use, dietary fat intake, race, and age were assessed by standardized questionnaires administered by trained interviewers (20). Fasting serum total cholesterol, triglycerides, and high-density lipoprotein cholesterol (HDL-C) were measured using standardized procedures (20). Low-density lipoprotein cholesterol (LDL-C) was calculated in participants with triglycerides under 4.53 mmol·L−1 (400 mg·dL−1) by the Friedewald formula (11). Diagnosis of diabetes mellitus was made using a fasting (8 h) serum glucose cut-point of > 6.93 mmol·L−1 (126 mg·dL−1), nonfasting serum glucose of > 11 mmol·L−1 (200 mg·dL−1), a physician’s diagnosis of diabetes mellitus, and/or a history of diabetes medication.
Differences between the highest and lowest physical activity quartiles with respect to descriptive characteristics were tested using Student’s t-tests for age; chi-square tests for race, sex, and education; and regression models adjusted for age, race, sex, clinical center, and education for nondemographic variables. ANCOVA was used to estimate adjusted mean levels of arterial diameter change by differing levels of physical activity. Covariates included were diastolic blood pressure, diastolic arterial diameter, pulse pressure, pulse pressure squared, sex, race, age, height, education, current smoking status, dietary fat intake, and clinical center. Covariates were chosen on the basis of associations with the physical activity and/or arterial diameter change. Pulse pressure squared was included because of the nonlinear relationship of pulse pressure and arterial diameter change. All models also included all three physical activity indices. Linearity of the relationship between physical activity and distensibility measures was assessed by treating quartiles of physical activity as continuous and testing for nonzero slope coefficients, using the “ESTIMATE” statement with four levels (corresponding to the four quartiles) in the general linear modeling procedure available in SAS version 6.12 (SAS Institute, Inc., Cary, NC). The adjusted arterial diameter change ANCOVA approach ostensibly allows for evaluation of the associations between arterial distensibility and physical activity independent of blood pressure.
To assess potential differences in any observed associations by race or gender, we added variables to the regression models that multiplied the predictor of interest (work, sport, or leisure activity, or pattern of physical activity) by race or gender and assessed the statistical significance of these variables using an F test, with statistical significance for these tests of effect modification set at 0.05.
We repeated the analysis, replacing arterial diameter change as the outcome variable with more traditional indices for arterial stiffness (the inverse of arterial distensibility) in multiple linear regression models, including Peterson’s elastic modulus, Young’s elastic modulus, and the beta stiffness index (13,14,23,27,36). The covariates for the models that used these arterial stiffness indices as outcome variables were the same as described above for the ANCOVA models minus diastolic arterial diameter, diastolic blood pressure, pulse pressure, and pulse pressure squared. The results from these additional models with the traditional stiffness indices did not differ substantively from one another. For comparison with other published articles, we present here results from models using adjusted arterial diameter change as well as from models using one of the traditional arterial stiffness indices (beta stiffness index (13,14)).
Table 1 presents a description of the participants free of CVD at the time of the carotid artery pulsatile diameter measurements (visit 2 for approximately 91% of the cohort, visit 1 for the remainder), within the lowest (Q1) and highest (Q4) quartiles of work, sport, and leisure physical activity. For the demographic variables included in Table 1, the association with physical activity was reversed for work versus sport and leisure physical activity. For example, there were more black participants in the highest than lowest quartile of work physical activity (P = 0.001), but there was a higher percentage of black participants in the lowest quartiles of sport and leisure physical activity (P = 0.001 for both). Arterial diameter change and the beta stiffness index were not associated with the physical activity indices.
Figure 1 presents the association of work, sport, and leisure physical activity quartiles with arterial diameter change from a model that included all three physical activity indices and was adjusted for diastolic arterial diameter, diastolic blood pressure, pulse pressure, pulse pressure squared, sex, race, age, height, education, dietary fat intake, smoking, and clinical center. Work quartiles were weakly associated with arterial diameter change. This association was not altered when potential mediators for this association were individually added to the model (IMT, BMI, HDL-C, LDL-C, and diabetes). No race or gender differences in the association of work physical activity and arterial diameter change were observed. Sport and leisure physical activity quartiles were not associated with arterial diameter change. Work, sport, and leisure physical activity quartiles were not associated with beta stiffness index (Fig. 2). We repeated these analyses with a single physical activity score that was the sum of all three indices and found no substantive differences in the findings from those presented in Figures 1 and 2.
We examined patterns of participation in any vigorous sport activity, in addition to regular participation in any activity, nonvigorous intensity activity, and vigorous intensity activity. The models included these physical activity patterns and work, leisure, and sport physical activity quartiles as covariates to test the hypothesis that specific patterns of physical activity participation were associated with arterial diameter change and/or beta stiffness index after accounting for overall physical activity level. Results from these models are presented in Table 2. The 12.7% of participants reporting any vigorous physical activity had significantly higher arterial diameter change and lower beta stiffness index compared with those not reporting any vigorous physical activity (P = 0.02 for arterial diameter change and P = 0.05 for beta stiffness index). Note that because beta stiffness index is an indicator of arterial stiffness (the inverse of distensibility, which we represent with adjusted arterial diameter change), a lower beta stiffness index in those reporting any vigorous activity is in agreement with a higher arterial diameter change. A difference in the association with vigorous physical activity was noted by gender for adjusted arterial diameter change (type 3, F = 4.81, P = 0.03), but not for beta stiffness index. The association of adjusted arterial diameter change and vigorous physical activity was observed in men but not in women. Adding potential covariates on the causal pathway (IMT, BMI, HDL-C, LDL-C, and diabetes) to the model predicting arterial diameter change from vigorous physical activity participation did not attenuate the association when arterial diameter change was the outcome variable (results not shown). When beta stiffness index was the outcome variable, the addition of IMT, HDL-C, or LDL-C to the model (but not BMI or diabetes) attenuated the association with any vigorous activity such that the association was no longer significant. The percentage change in the between-group beta stiffness index difference was −34.6%, −11.5%, and −11.5% for IMT, HDL-C, and LDL-C, respectively. Regular physical activity was not associated with arterial diameter change or beta stiffness index, regardless of intensity. The proportion of participants participating in regular vigorous physical activity was 4.8%.
The current findings do not support a strong association between physical activity and arterial distensibility in the ARIC cohort. We observed a weak statistically significant association between self-report of any vigorous physical activity for both outcome measures reported and a weak association of work physical activity with adjusted arterial diameter change (but not beta stiffness index). Participants who took antihypertensive medications were included in these analyses, in part because they represented 30% of those who completed the examination of the carotid arterial diameters. When we further excluded these participants, the association of vigorous activity remained, but the association of adjusted arterial diameter changes with the work physical activity index was no longer present. This may suggest that the observed association of adjusted arterial diameter changes and work physical activity is primarily because of residual confounding of the association by blood pressure. Alternatively, the change in results after excluding participants on antihypertensive medications may reflect differential effects on arterial distensibility of specific medications.
Prior smaller studies have reported strong associations between &OV0312O2max and some measure of arterial distensibility, compliance, or stiffness when comparing highly active versus sedentary individuals (29,30,33,34). A recent report observed a difference in arterial compliance comparing men who were sedentary with those who were endurance trained, but no difference between the sedentary and “recreationally active” men (30). Physical activity measured by questionnaire is only modestly associated with &OV0312O2max(15). The current findings are not inconsistent with the possibility that a high level of fitness, resulting from a genetic propensity for high fitness level, consistent vigorous exercise training, or some combination of the two, could be positively associated with arterial distensibility. However, the current findings do not support any association of arterial distensibility with a more moderate activity level that might be more commonly performed by the general population.
To our knowledge, the current study is the only observational investigation of the association of arterial distensibility (represented in this analysis by blood pressure adjusted carotid artery pulsatile diameter changes) and physical activity in a large population-based sample. No other studies of these associations have the gender and ethnic diversity of the ARIC cohort. The Baltimore Longitudinal Study of Aging (BLSA) (33) examined the association cross-sectionally in men. The BLSA cohort is almost entirely white, and the comparison of arterial stiffness across exercise training level was made on a subsample of 96 sedentary men who were part of the BLSA cohort and 14 age-matched older male athlete volunteers. They found a higher maximal aerobic capacity from a treadmill test attenuated an age-associated increase in arterial stiffness (as assessed by pulse wave velocity) in the BLSA men. Tanaka et al. (29) used similar methodology in a small cross-sectional study that found similar associations in 53 women.
Kupari et al. (17) conducted a small study of healthy volunteers and reported lower arterial distensibility with higher self-reported physical activity. This finding differs not just from the current ARIC results but from all other studies on the association of physical activity or fitness and distensibility, compliance, or stiffness (7,29,30,33,34). This difference in findings may be because of differences in the method of assessing arterial distensibility (or stiffness). In other observational studies (29,30,33,34), pulse-echo or applanation tonometry was used to assess arterial distensibility (or stiffness). Kupari et al. (17) measured aortic stiffness using magnetic resonance imaging (MRI).
Three exercise trials have examined changes in arterial distensibility in response to training. After an initial cross-sectional comparison showed higher femoral arterial diameters and greater brachial systolic to diastolic arterial diameter changes in trained versus untrained men (34), Wijnen et al. were unable to detect changes in carotid, femoral, or brachial artery diameters or systolic to diastolic diameter changes after 6 wk of vigorous intensity exercise training in healthy young men (35). Cameron and Dart (7) showed increases in compliance after a 4-wk exercise training intervention that were larger than could be explained by blood pressure changes, consistent with a primary change in arterial distensibility produced by exercise training independent of blood pressure. Tanaka et al. (30) recently reported improvements in blood pressure adjusted dynamic arterial compliance after 12 wk of endurance training four to six times weekly at 70 to 75% of maximal aerobic capacity. These three experimental studies were relatively short interventions in tightly controlled research settings. It has not been determined whether these favorable changes in arterial compliance or diameter changes would occur in the context of a more moderate exercise training program that might be more feasible for the general population to maintain long term. The proposed mechanisms by which exercise training may improve arterial distensibility include indirect effects through established CVD risk factors and/or direct effects on intrinsic structural or functional components of the arterial walls (3,8,9,12,16,28,32). We observed inconsistent evidence that the effect of vigorous exercise on arterial distensibility would be through established CVD risk factors. Furthermore, the direct structural and functional changes to the arterial walls may require training at an intensity level higher than what is required to achieve clinically significant improvements in blood lipids, blood pressure, insulin sensitivity, or obesity (3,8,9,12,16,28,32).
Matsuda et al. (19) observed an increase in the calcium content of the elastin fibers in the aortas of exercise-trained rats that resulted in increased arterial distensibility. In the current study, the association of vigorous activity and arterial diameter change was not attenuated by adjustment for CVD risk factors. This, combined with the 30% attenuation of the association of the beta stiffness index by IMT (a noninvasive estimate relating to arterial structure), is consistent with the hypothesis that the effect of vigorous exercise on the artery walls may result from structural changes in the arterial wall and is not mediated through other CVD risk factors. Further research is needed to determine if vigorous exercise training directly affects the structure or elastic components of the artery walls, as occurs in the myocardium after vigorous exercise training.
Assumptions and limitations.
It is possible that measurement error for physical activity and/or for carotid artery pulsatile diameters in this large epidemiologic study limited detection of an association of physical activity and arterial distensibility. It is also possible that the few statistically significant results reported here are the result of type I errors and a large sample size.
Measurement of arterial diameter change was made at the left common carotid artery, whereas blood pressures were measured at the brachial artery. Although these measures tend to differ in younger populations (6), in a homogenous group of older participants, the use of peripheral blood pressure may be an adequate surrogate of carotid blood pressure because these two measures may be very similar in this population (6,21). The arterial wall-tracking device has shown reasonably good reliability in repeated measures in a subsample of the ARIC cohort (26). However, time constraints did not allow for replication of the measurements at the right carotid artery. Furthermore, data for measurements of the common carotid artery were missing for 27% of the ARIC cohort either because participants did complete clinic visit 2 or because patient anatomy and/or physiology did not allow for adequate data quality. Therefore, a significant subset of the cohort was excluded from these analyses. Despite these limitations, adjusted arterial diameter changes are associated with retinal arteriolar narrowing measurements in the ARIC cohort (18).
The physical activity assessment was made by self-report, which may be subject to recall bias and misclassification. Furthermore, physical activity was assessed at the first ARIC clinic visit, whereas 91% of the arterial distensibility measurements were made at the second ARIC clinic visit, several years later. It was assumed that the ranking of physical activity levels within the ARIC cohort participants at clinic visit 2 would not have differed from the rankings at clinic visit 1. If this assumption is invalid, this could explain the null findings. Finally, because this is a cross-sectional analysis, no direction of association or causality can be inferred.
It is not possible, given the results of this and other studies conducted thus far, to draw conclusions about the public health significance of any association between physical activity and arterial distensibility. A recently developed measurement approach allows for measurement of arterial wall mechanics at the brachial artery over a wide range of transmural pressure (arterial pressure minus pressure from a water cuff around the brachial artery), without changing systemic pressure or reaction of the neurohormonal reflexes (5). Since the intrinsic elastic properties of the artery walls are affected by blood pressure, the ability to assess arterial elastic properties at transmural pressures below physiologic values improves the chance of detecting an association between aerobic exercise training and intrinsic arterial elastic properties.
The authors thank the staff and participants in the ARIC study for their important contributions.
Kathryn H. Schmitz was funded by training grant T32-HL07036 to Dr. Aaron Folsom from the National Institutes of Health.
The ARIC study was funded by contracts N01-HC-55016, N01-HC-55018, N01-HC-55019, N01-HC-55020, N01-HC-55021, and N01-HC-55022 from the U.S. National Heart, Lung, and Blood Institute.
Address for correspondence: Kathryn H. Schmitz, Division of Epidemiology, University of Minnesota, 1300 South 2nd St., Suite 300, Minneapolis, MN 55454; E-mail: email@example.com.
1. Ainsworth, B. E., D. R. Jacobs, Jr., A. S. Leon, M. T. Richardson, and H. J. Montoye. Assessment of the accuracy of physical activity questionnaire occupational data. J. Occup. Med. 35: 1017–1027, 1993.
2. American College of Sports Medicine. Position stand: physical activity, physical fitness, and hypertension. Med. Sci. Sports Exerc. 25: i–x, 1993.
3. Arnett, D. K., G. W. Evans, and W. A. Riley. Arterial stiffness: a new cardiovascular disease risk factor? Am. J. Epidemiol. 140: 669–682, 1994.
4. Baecke, J. A. H., J. Burema, and J. E. R. Fretters. A short questionnaire for the measurement of habitual physical ac-tivity in epidemiologic studies. Am. J. Clin. Nutr. 36: 932–942, 1982.
5. Bank, A. J. Physiologic aspects of drug therapy and large artery elastic properties. Vasc. Med. 2: 44–50, 1997.
6. Boutouyrie, P., C. Bussy, P. Lacolley, X. Girerd, B. Laloux, and S. Laurent. Association between local pulse pressure, mean blood pressure, and large-artery remodeling. Circulation 100: 1387–1393, 1999.
7. Cameron, J. D., and A. M. Dart. Exercise training increases total systemic arterial compliance in humans. Am. J. Physiol. 266: H693–H701, 1994.
8. Despres, J. P. Visceral obesity, insulin resistance, and dyslipidemia: contribution of endurance exercise training to the treatment of the plurimetabolic syndrome. Exerc. Sport Sci. Rev. 25: 271–300, 1997.
9. Durstine, J. L., and W. L. Haskell. Effects of exercise training on plasma lipids and lipoproteins. Exerc. Sport Sci. Rev. 22: 477–521, 1994.
10. Folsom, A. R., D. K. Arnett, R. G. Hutchinson, F. L. Liao, L. X. Clegg, and L. S. Cooper. Physical activity and incidence of coronary heart disease (CHD) in middle aged women and men. Med. Sci. Sports Exerc. 29: 901–909, 1997.
11. Friedewald, W. T., R. I. Levy, and D. S. Frederickson. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative centrifuge. Clin. Chem. 18: 499–502, 1972.
12. George, K. P., L. A. Wolfe, and G. W. Burggraf. The ‘athletic heart syndrome’: a critical review. Sports Med. 11: 300–331, 1991.
13. Hayashi, K., H. Handa, S. Nagasawa, A. Okumura, and K. Moritake. Stiffness and elastic behavior of human intracranial and extracranial arteries. J. Biomech. 13: 175–184, 1980.
14. Hirai, T., S. Sasayama, T. Kawasaki, and S. Yagi. Stiffness of systemic arteries in patients with myocardial infarction: a noninvasive method to predict severity of coronary atherosclerosis. Circulation 80: 78–86, 1989.
15. Jacobs, D. R., Jr., B. E. Ainsworth, R. J. Hartman, and A. S. Leon. A simultaneous evaluation of 10 commonly used physical activity questionnaires. Med. Sci. Sports Exerc. 25: 81–91, 1993.
16. Kriska, A. M., S. N. Blair, and M. A. Pereira. The potential role of physical activity in the prevention of non-insulin-dependent diabetes mellitus: the epidemiologic evidence. Exerc. Sport Sci. Rev. 22: 121–144, 1994.
17. Kupari, M., P. Hekali, P. Keto, V.-P. Poutanen, M. J. Tikkanen, and C.-G. Standertskjold-Nordenstam. Relation of aortic stiffness to factors modifying the risk of atherosclerosis in healthy people. Arterioscler. Thromb. 14: 386–394, 1994.
18. Liao, D., R. D in , J. M o , et al. Carotid arterial elasticity and retinal arteriolar narrowing. Circulation 99: 1109, 1999.
19. Matsuda, M., T. Nosaka, M. Sato, and N. Ohshima. Effects of physical exercise on the elasticity and elastic components of the rat aorta. Eur. J. Appl. Physiol. 66: 122–126, 1993.
20. National Heart, Lung, and Blood Institute. The ARIC Manuals of Operation. Chapel Hill, NC:ARIC Coordinating Center, School of Public Health, University of North Carolina,
21. Pauca, A. L., S. L. Wallenhaupt, N. D. Kon, and W. Y. Tucker. Does radial artery pressure accurately reflect aortic pressure? Chest 102: 1193–1198, 1992.
22. Pereira, M. A., A. R. Folsom, P. G. McGovern, et al., for the Atherosclerosis Risk in Communities Study Investigators. Physical activity and incident hypertension in black and white adults: the atherosclerosis risk in communities study. Prev. Med. 28: 304–312, 1999.
23. Peterson, L. H., R. E. Jensen, and J. Parnell. Mechanical properties of arteries in vivo. Circ. Res. 8: 622–639, 1960.
24. Pols, M. A., P. H. Peeter, H. B. Bueno-De-Mesquita, et al. Validity and repeatability of a modified Baecke questionnaire on physical activity. Int. J. Epidemiol. 24: 381–388, 1995.
25. Deleted in proof.
26. Riley, W. A., R. W. Barnes, G. W. Evans, O. S. Smith, and G. Heiss. Measurement of arterial distensibility in the atherosclerosis risk in communities (ARIC) cohort. In: Follow-Up and Prevention of Atherosclerotic Plaque, E. Borgatti (Ed.). Torino, Italy: Centro Scientifico, 1992, pp. 45–62.
27. Riley, W. A., R. W. Barnes, and H. M. Schey. An approach to the noninvasive periodic assessment of arterial elasticity in the young. Prev. Med. 13: 169–184, 1984.
28. Schmitz, K. H., D. R. Jacobs, Jr., A. S. Leon, P. J. Schreiner, and B. Sternfeld. Physical activity and body weight: associations over 10 years in the CARDIA study. Int. J. Obes. 24: 1475–1487, 2000.
29. Tanaka, H., C. A. Desouza, and D. R. Seals. Absence of age-related increase in central arterial stiffness in physically active women. Arterioscler. Thromb. Vasc. Biol. 18: 127–132, 1998.
30. Tanaka, H., F. A. Dinenno, K. D. Monahan, C. M. Clevenger, C. A. Desouza, and D. R. Seals. Aging, habitual exercise, and dynamic arterial compliance. Circulation 102: 1270–1275, 2000.
31. The ARIC Investigators. The Atherosclerosis Risk in Communities (ARIC) study: design and objectives. Am. J. Epidemiol. 129: 687–702, 1989.
32. U.S. Department of Health and Human Services. Physical Activity and Health: A Report of the Surgeon General. Washington, DC:U.S. Government Printing Office,
1996. Publication No. 737–212/40001.
33. Vaitkevicius, P. V., J. L. Fleg, J. H. Engel, et al. Effect of age and aerobic capacity on arterial stiffness in healthy adults. Circulation 88 (Pt. 1):1456–1462, 1993.
34. Wignen, J. A. G., H. Kuipers, M. J. F. Kool, et al. Vessel wall properties of large arteries in trained and sedentary subjects. Basic Res. Cardiol. 86: 25–29, 1991.
35. Wijnen, J. A., M. J. Kook, J. P. Kooman, et al. Vessel wall properties of large arteries and endurance training. J. Hypertens. 11 (Suppl. 5): S88–S89, 1993.
36. Young, T. On the functions of the heart and arteries: the Croonian Lecture. Philos. Trans. R. Soc. Lond. [Biol.] 99: 1–31, 1809.
Keywords:© 2001 Lippincott Williams & Wilkins, Inc.
ARTERIAL DISTENSIBILITY; EXERCISE; CARDIOVASCULAR DISEASE