| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Section of Applied Physiology, Division of Geriatrics and Nutritional Sciences, and 2Cardiovascular Division, Department of Internal Medicine, Washington University School of Medicine, St. Louis, Missouri
Submitted 4 November 2005 ; accepted in final form 15 February 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
O2 max) and the influence of gender on the decline in O2 max and its determinants in old and very old men and women. Sedentary, 60- to 92-yr-old women (n = 71) and men (n = 29), with no evidence of cardiovascular disease, underwent maximal treadmill exercise tests during which O2 max and maximal cardiac output (
max) were determined. O2 max and age were inversely related in both women (–23 ± 2 ml·min–1·yr–1; P < 0.0001) and men (–57 ± 5 ml·min–1·yr–1; P < 0.0001). The absolute slope of the O2 max vs. age relationship was twofold steeper in men than in women (P < 0.0001). max was also inversely related to age in a gender-specific manner (women = –87 ± 25 ml·min–1·yr–1, P = 0.0009; men = –215 ± 50 ml·min–1·yr–1, P = 0.0002; P = 0.01 women vs. men). Age-related changes in maximal exercise arteriovenous oxygen content difference (a-vDO2) were marginally different (P = 0.08) between women (–0.12 ± 0.03 ml·dl–1·yr–1, P = 0.0003) and men (–0.22 ± 0.04 ml·dl–1·yr–1, P < 0.0001). Age-associated decreases in max and a-vDO2 contributed equally to the declines in O2 max in both men and women. In the later stages of life, O2 max, max, and a-vDO2 decrease with age more rapidly in older men than they do in older women. As a result, the gender differences dissipate in the later decades of life. Declines in max and a-vDO2 contribute equally to the age-related decrease in O2 max in men and women. exercise; cardiac output; hemodynamics; maximal oxygen uptake; aging
O2 max) decreases progressively with age (3), with men showing a greater decline than women (3, 7, 12, 13, 26, 29). O2 max is determined by the capacity of the cardiovascular system to provide oxygenated blood to the working muscles, as reflected in maximal cardiac output (max), and the capacity of the working muscle to extract oxygen from the blood, as manifested by arteriovenous oxygen content difference (a-vDO2). Therefore, the age-related decline in O2 max can be a consequence of a reduction in max, a-vDO2, or both. Several studies have suggested that age-related reduction in max is at least partly responsible for the age-related decrease in O2 max (1, 10, 11, 14, 19). However, others challenged this notion and concluded that max may not change with age (18, 20). Furthermore, it is unclear whether the mechanisms underlying diminished aerobic capacity in older individuals are affected by gender. Therefore, we sought to study these issues in a large cohort of sedentary, older men and women. In an earlier study (19), our laboratory found that both determinants of O2 max (i.e., max and a-vDO2) were considerably lower in 60- to 70-yr-old men and women than in younger subjects. As an extension of these earlier findings, and because the number of people living to very old age is increasing rapidly, we focused on older, 60- to 92-yr-old men and women in the present study. We hypothesized that O2 max declines in older men and women as a result of age-related reductions in both max and a-vDO2. Furthermore, because there is a paucity of data comparing age-associated changes in maximal exercise capacity and cardiovascular function during maximal aerobic exercise in older men and women, we hypothesized that the age-related decline in O2 max is greater in men than in women and that reductions in max and a-vDO2 contribute to these declines in both men and women. |
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Graded exercise test and O2 max.
O2 max was determined by indirect calorimetry during a graded treadmill exercise test and as described previously (4, 15). Metabolic data were measured with a computer-interfaced system, including a dry gas meter (CD-4, Parkinson-Cowan), oxygen analyzer (S3-A, Applied Electrochemistry), CO2 analyzer (LB-2, Beckman), and 5-liter mixing chamber. Oxygen uptake (O2) data were generated by the computer every 30 s. The highest average for two consecutive 30-s data points was considered O2 max. Blood pressure, measured via auscultation, and 12-lead ECG were monitored throughout the test. The incremental test started at a speed determined, during a warm-up period, to elicit 
70% of age-predicted maximum heart rate (HRmax) and remained constant throughout the test while grade was increased by 1–2% every 1–2 min. The test continued until the subjects could no longer exercise due to exhaustion or until other conditions, such as ECG changes or development of symptoms, made it unsafe to continue. None of the subjects whose test was stopped because of cardiac symptoms or significant ECG changes was included in the present analyses. Peak respiratory exchange ratio (RER) was higher (P < 0.0001) in men (1.19 ± 0.01) than in women (1.13 ± 0.01). However, RER was not related to age in men (slope = 0.0011 ± 0.0018 RER units/yr, P = 0.55) or in women (slope = –0.0014 ± 0.0010 RER units/yr, P = 0.17), indicating that age did not affect the subjects' ability to give maximal effort during the test.
Cardiac output.
On a separate occasion, and at least 1 wk after the initial exercise test, another exercise test was performed to measure max, O2 max, HRmax, and maximal exercise blood pressures. max was determined noninvasively using closed-circuit acetylene rebreathing during treadmill exercise, as described in detail previously (25, 30). After the subjects performed a few minutes of warm-up exercise, the treadmill speed and grade were progressively increased to the lowest settings found to elicit O2 max during the initial O2 max test. In general, larger increments in speed and grade were used for subjects with greater fitness so that the duration of exercise before maximal exercise (4–6 min) was not a function of fitness. This protocol allowed us to avoid the potentially confounding effect of test duration on measured max and maximum stroke volume (SVmax) (17). O2 was monitored continuously (as described above), and, when the O2 was at or near the previously measured O2 max value, blood pressure was measured by auscultation, and the subject was switched from the open-circuit O2 measurement system to a closed-circuit rebreathing system containing a mixture of 0.5% C2H2, 10% He, 45% O2, and 44.5% N2 for measurement of cardiac output. End-tidal gas concentrations were monitored using a capillary sampling line attached to the mouthpiece and a Perkin-Elmer mass spectrometer (MGA 1100), which was interfaced to a computer. The calculation of cardiac output was based on the exponential decay of end-tidal C2H2 concentrations across 8–10 serial breaths. Tests in which HRmax was <85% of age-predicted HRmax (220 – age) were deemed submaximal (8) and were not included in the analyses for the present report. O2 max and HRmax values from the cardiac output test were slightly but not significantly lower than those measured in the initial O2 max test (O2 max: difference = 13 ± 10 ml/min, P = 0.20; HRmax: difference = 1.6 ± 0.9 beats/min, P = 0.10). O2 max, HRmax, and maximal blood pressure data obtained from the cardiac output test were used for outcome analyses, since they were collected in the same test as the cardiac output data, and this made it possible to elucidate the mechanisms underlying the age- and gender-related differences in O2 max. SVmax, a-vDO2, maximal exercise total peripheral resistance (TPRmax), and maximal exercise mean blood pressure (MBPmax) were calculated using standard equations, as described previously (25).
Statistics.
Mixed-model analysis of variance was used to assess whether the rates of age-associated decline in O2 max and its determinants were different between men and women (as indicated by an interactions between age and gender). Linear regression was used for analyses of relationships among quantitative variables. Error terms are presented as standard errors (SEs), unless noted otherwise. Analyses were performed at an alpha error rate of 0.05. SAS software (SAS version 8, SAS Institute, Cary, NC) was used for all analyses.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
O2 max values reflect the older, sedentary state of the subjects. Although the men were slightly older than the women, on average, this difference was small relative to the wide age range in both groups. As would be expected, men were heavier, taller, and had a greater average O2 max than women.
|
O2 max were evident in both men (–57 ± 5 ml·min–1·yr–1; P < 0.0001) and women (–23 ± 2 ml·min–1·yr–1; P < 0.0001), but the slope of this relationship was twofold steeper (P < 0.0001) in men than in women (Fig. 1). The gender difference in the relationship between age and aerobic capacity was also evident when O2 max was expressed relative to body weight (men: –0.51 ± 0.08 ml·kg–1·min–1·yr–1; P < 0.0001; women: –0.22 ± 0.04 ml·kg–1·min–1·yr–1; P < 0.0001; P = 0.0006 for comparison between men and women). We also analyzed O2 max data as percentages of the estimated values for 60-yr-old men and women (where the estimated values for 60 yr olds were calculated from the regression equations depicted in Fig. 1A). Despite this adjustment to account for the higher starting values seen in men, the results were similar in that the slope of the relationship between age and O2 max was still steeper (P = 0.03) in men (–2.1 ± 0.2%/yr; P < 0.0001) than in women (–1.5 ± 0.2%/yr; P < 0.0001) (Fig. 1).
|
max in these older subjects, with men showing a steeper (P = 0.01) slope than women (men: –215 ± 50 ml·min–1·yr–1, P = 0.0002; women: –87 ± 25 ml·min–1·yr–1, P = 0.0009) (Fig. 2). The gender difference in the age vs. max slopes was still evident (P = 0.03) after normalizing max for body surface area (men: –95 ± 26 ml·min–1·m–2·yr–1, P = 0.001; women: –35 ± 14 ml·min–1·m–2·yr–1, P = 0.02). When max was expressed in relative terms as a percentage of the estimated max for 60 yr olds (using the same approach as described above for O2 max; men: –1.25 ± 0.29%/yr, P = 0.0002; women: –0.77 ± 0.22%/yr, P = 0.0009), the gender difference in the slopes for the relationships between age and max was no longer significant (P = 0.23, Fig. 2).
|
0.33) (Fig. 3). Normalization of SVmax for differences in body surface area decreased the magnitude of difference between men and women. However, the difference was still significant (47.3 ± 1.6 vs. 39.9 ± 0.9 ml/m2 in men and women, respectively; P < 0.0001). An inverse relationship between age and HRmax was present for the group as a whole (–1.3 ± 0.2 beats·min–1·yr–1); however, this relationship was not significantly influenced by gender (P = 0.40) (Fig. 3).
|
O2 max was quantified by calculating O2 max values that should have resulted from differences in one hemodynamic factor while holding the value(s) for the other factor(s) constant (O2 max = max x a-vDO2). For example, to determine how much of the observed –57 ml·min–1·yr–1 slope in O2 max in the men was due to age-related differences in max, the relationship between age, and the product of the observed max values and a fixed value for a-vDO2, was assessed (where the fixed value for a-vDO2 was the estimated value for subjects in the midpoint of the age range, i.e., 76 yr). In men, the portion of the age vs. O2 max slope that was estimated to be exclusively due to reductions in max was –28 ± 7 ml·min–1·yr–1 (P = 0.0002), and the portion that was estimated to be due to reductions in a-vDO2 was –30 ± 5 ml·min–1·yr–1 (P < 0.0001). This suggests that decreases in max and a-vDO2 contribute equally to the age-related decline in O2 max. Likewise, in women, the portions of the slope for the age vs. O2 max relationship that were estimated to be exclusively due to max (–10 ± 3 ml·min–1·yr–1, P = 0.0009) and to a-vDO2 (–12 ± 3 ml·min–1·yr–1, P = 0.0003) were similar. None of the age vs. O2 max relationship was attributable to age-related variation in SVmax in either men (P = 0.33) or women (P = 0.93). Therefore, the portions of the slopes in the age vs. O2 max relationships for men and women that were due to decreases in max were entirely due to age-associated decrements in HRmax (men: –19 ± 4 ml·min–1·yr–1, P < 0.0001; women: –10 ± 1 ml·min–1·yr–1, P < 0.0001). TPRmax was related to age in both men (8.4 ± 3.0 dyn·s–1·cm–5·yr–1, P = 0.01) and women (16.5 ± 3.1 dyn·s–1·cm–5·yr–1, P < 0.0001), and the slopes for the relationships in men and women were not statistically different (P = 0.12). In contrast, men and women were different (P = 0.003) with respect to the relationship between age and MBPmax. In women, we found a positive relationship between age and MBPmax (0.6 ± 0.2 mmHg/yr, P = 0.0006), while in men, MBPmax was not associated with age (P = 0.18).
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
O2 max after the age of 60 yr in sedentary men and women, and to assess the relative contribution of cardiac output and a-vDO2 to the deterioration of O2 max in advancing age. Second, we sought to determine whether gender influences the age-associated decline in O2 max after the age of 60 yr and to explore the physiological basis for gender-related differences. Results from the present study provide evidence that, in 60- to 92-yr-old subjects, the age-related decrease in O2 max is attributable to decrements in both max and a-vDO2 and that HRmax is solely responsible for the decline in max. Since the age-associated decline in max was accompanied by an age-related increase in TPRmax in both men and women, MBPmax did not decline with age, and even increased with age in women. Another important finding of the present study is that, although the rate of age-associated decline in O2 max, max, and maximal a-vDO2 is greater in men than in women, the relative contribution of max and a-vDO2 to the decline in aerobic power appears to be similar for men and women.
Most previous studies have reported that the age-related decline in O2 max is attributable to reductions in both a-vDO2 and max. a-vDO2 has been shown, unequivocally, to decrease with advancing age in 18- to 75-yr-old subjects (1, 14, 18, 19), and our findings in subjects as old as 92 yr are consistent with these observations. A number of studies have also shown that max declines with age (1, 11, 13, 14, 19) in subjects ranging from 18 to 75 yr old. However, reports from a longitudinal study on 20- to 50-yr-old men (18) and from a cross-sectional study on 25- to 79-yr-old men and women (20) suggested that max is not influenced by age, because the age-related decrease in HRmax is completely offset by an age-associated increase in SVmax. Our data from older subjects support the findings of most other investigators (1, 11, 13, 14, 19) and show a significant decline in max with increasing age. Furthermore, we did not observe any evidence that SVmax increases with age to compensate for the decrease in HRmax and prevent a decrease in max.
The lack of an age-associated change in SVmax reported in the present study may appear contradictory to a previous report from our laboratory (19) in which we found that 60- to 72-yr-old sedentary men and women had a lower SVmax than 20 to 31 yr olds. One explanation is that the data from the earlier report reflect the changes in SVmax that occur during the early and middle decades of adulthood, while data from the present study reflect changes that occur in the later decades of life. It is plausible, therefore, that SVmax decreases with age until late adulthood, after which the decline in SVmax ceases.
A greater age-associated decline in O2 max in men, compared with women, has been reported by others (3, 7, 12, 13, 26, 29), and our findings confirm these previous reports. Furthermore, our data suggest that the decrease in O2 max is likely to be accelerated after the age of 60 yr. Although men have higher O2 max values than women through most of the adult lifespan, the greater age-associated decrements in O2 max in men should eventually result in the elimination of this gender-specific difference in advanced age. In fact, based on our data, the regression lines for men and women intersect at age 94 yr, only slightly beyond the age of the oldest individuals in the present study. Our data show that both the greater decline in max and the greater decline in a-vDO2 in men than in women account for the gender-specific difference in the rate at which O2 max declines with advancing age. Most previous studies have reported the annual age-associated decline in O2 max to be 24 to 35 ml·min–1·yr–1 in sedentary men (12, 21, 22) and 13 to 16 ml·min–1·yr–1 in sedentary women (12, 23, 28). These age-related declines are much slower than the respective –57 ± 5 and –23 ± 2 ml·min–1·yr–1 decrements seen in the present study. One explanation for this apparent discrepancy is that we did not study young individuals, and many of the subjects in our study were older than those in the previous studies. The upper age range for the previous studies was 75–84 yr, and, furthermore, there were only a few subjects in the upper end of the age spectrum (12, 21–23, 28). In contrast, 31 of our 100 subjects were 80–92 yr of age. It is plausible, therefore, that the greater decline in aerobic capacity may reflect a greater proportion of very old subjects who, by virtue of their old age, had mild frailty and physical inactivity. Thus very old men and women may exhibit more rapid deterioration in physiological function than their younger counterparts. The age-associated decline in O2 max is commonly described as a fixed percent per decade (11, 19, 22, 23, 28, 29). Mathematically, this suggests that the absolute reduction in O2 max per decade should progressively decrease as people age. Despite this premise that the absolute decline in O2 max should decrease with increasing age, the comparison of our data on very old subjects with that reported in the literature for 20- to 84-yr-old subjects (12, 21–23, 28) suggests otherwise, i.e., the rate of age-associated decline in O2 max increases with age rather than decreases. In support of this notion, a recent comparison of longitudinal changes in O2 max among subjects from a wide age range (8) indicated that the 8-yr decline in O2 max is much greater in men and women over 60 yr old (50 and 24 ml·min–1·yr–1, respectively) than it is in 30–39 yr olds (17 and 13 ml·min–1·yr–1, respectively).
The reason for the more rapid decline in O2 max, max, and a-vDO2 among men compared with women is not clear. One possibility is that the men decline at a greater rate, simply because they have greater absolute cardiovascular functional capacity to begin with. To address this possibility, we calculated O2 max, max, and a-vDO2 as percentages of the values that would be expected for the average 60-yr-old man and woman. Despite this normalization of men and women to a uniform starting value, the men still demonstrated a more rapid age-associated deterioration in O2 max than women. In contrast, the declines in max and a-vDO2, when expressed relative to age 60-yr values, were no longer significantly different between men and women. Although the relative declines in max and a-vDO2 are not statistically significantly different between men and women (P = 0.23 and P = 0.24, respectively), we may have lacked the statistical power to detect significance for these outcomes, since they are responsible for the age-related decline in O2 max. It seems that at least some of the more rapid decline in maximal cardiovascular function in men is due to initially greater function, and some is due to other factors. One factor could be a greater age-associated decline in inotropic sensitivity to 
-adrenergic agonist in men than in women, which is partly due to a greater -adrenergic-stimulated increase in left ventricular systolic function in younger men than in younger women (31).
Another potential explanation for gender differences in the decline in cardiovascular function is that older men decrease their physical activity levels more than women as they age. Although we do not have leisure time physical activity data to assess this possibility in the present study, this explanation seems unlikely, since population-based data from the Minnesota Heart Survey suggest that women decrease their leisure time physical activity levels more than men (9). It is also important to recognize that the main type of physical activity that has a tangible effect on O2 max is structured vigorous endurance exercise. Because the men and women in the present study were specifically selected as individuals who did not participate in endurance training, it is unlikely that differential changes in other, nonendurance-training activities between men and women could explain the gender differences in the rate of O2 max decline. It has been reported that variations in habitual physical activity only account for 
5% of the variation in O2 max among elderly men and women (27).
Although the absolute age-related declines in O2 max, a-vDO2, max, and HRmax were greater in men than in women, we found that the age-associated reductions in max and a-vDO2 contributed equally to the age-associated declines in O2 max for men and women. Furthermore, the entire age-associated decline in max was attributable to decreases in HRmax for both men and women. These data suggest that aging of the skeletal muscle and of the central cardiovascular system contribute equally to the decline in aerobic capacity with advancing age during the later stages of life and that this is true for both men and women.
The O2 max values of individuals in the upper end of the age range studied were extremely low. In fact, the average O2 max for the oldest two men and the oldest two women was only 971 and 820 ml/min (13.0 and 13.6 ml·kg–1·min–1), respectively. To put this into perspective, the energy expenditure required for these elderly subjects to stand statically (16) would require 50% of O2 max. This serious deficiency in cardiovascular function in very old men and women illustrates the importance of interventions known to increase O2 max, such as exercise training, to delay the age at which cardiorespiratory fitness becomes so limiting that an individual can no longer function independently in activities of daily living.
Several limitations should be considered when interpreting the results of the present study. First, young subjects and those with clinical evidence of heart disease were not included in our study. The findings, therefore, are only applicable to individuals in the later decades of life who are free from clinical heart disease. Second, some of the older men and women had mild to moderate frailty, which may have affected the results. However, because the prevalence of physical frailty increases with age, the data on our oldest subjects more closely represent the elderly population than it would have if we had excluded all frail subjects. Third, although we specifically recruited individuals who did not perform habitual structured exercise, it is quite conceivable that the older subjects were less active in their daily living than the younger subjects, and this could have contributed to the decline in aerobic capacity, independent of the effects of aging per se. Another limitation is that undetected coronary disease may have been present in some participants, despite the relatively thorough screening process used in the present study (6). Because the frequency of occult heart disease increases with age and is more common in older men than in older women (6), coronary disease might have been partly responsible for the more rapid decline in O2 max seen in our older subjects, compared with younger subjects, and it might have also been partly responsible for the greater rate of decline seen in men than in women. Lastly, we used a cross-sectional design, which has inherent limitations related to the inability to draw all of the subjects from the same population. While it cannot be known if this limitation affected our results, it is noteworthy that our cross-sectional estimates of the rate of decline in the present study are similar to those recently reported from a longitudinal study (8).
In summary, findings from the present study suggest that, after the age of 60 yr, O2 max decreases with age due to reductions in both max and a-vDO2 and that these reductions occur more rapidly in men than in women. Furthermore, the age-related declines in max and a-vDO2 contribute equally to the age-associated reductions in O2 max in men and women. The age-related decline in max is due to decreases in HRmax with no significant effect of age on maximal exercise stroke volume. Because the rates of decline in O2 max and its physiological determinants are considerably greater in older men than in older women, the gender-associated differences tend to dissipate in the later decades of life.
|
GRANTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
O2 max. J Appl Physiol 65: 1147–1151, 1988.O2 max to training in 60–71 yr olds. J Appl Physiol 71: 2004–2011, 1991.O2 max with aging in master athletes and sedentary men. J Appl Physiol 68: 2195–2199, 1990.O2 in men and women. J Appl Physiol 77: 647–652, 1994.This article has been cited by other articles:
|
|
 |
|
  R J Shephard Maximal oxygen intake and independence in old age Br. J. Sports Med., May 1, 2009; 43(5): 342 - 346. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. McGavock, J. L. Hastings, P. G. Snell, D. K. McGuire, E. L. Pacini, B. D. Levine, and J. H. Mitchell A Forty-Year Follow-Up of the Dallas Bed Rest and Training Study: The Effect of Age on the Cardiovascular Response to Exercise in Men J Gerontol A Biol Sci Med Sci, February 5, 2009; (2009) gln025v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Grewal, R. B. McCully, G. C. Kane, C. Lam, and P. A. Pellikka Left Ventricular Function and Exercise Capacity JAMA, January 21, 2009; 301(3): 286 - 294. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
P 
