Exercise, Physiological Function, and the Selection of Participants for Aging Research

Abstract

Background.

Regular and vigorous exercisers appear to be the logical choice for studying the inherent aging process as they are essentially free from the complications of disuse. Cross-sectional studies of aging tend to depict an essentially smooth and progressive decrement of physiological function with increasing chronological age. On closer examination of such data, it is seen that although the young have high functional values and the very old low, between these limits, values are widely scattered.

Methods.

We have reevaluated published data from a meta-analysis of 242 studies on men and from a similar study on women. From both data sets, where $\stackrel{·}{V}{\text{O}}_{2,\max }$ was plotted against chronological age, we stratified the $\stackrel{·}{V}{\text{O}}_{2,\max }$ values into bandwidth intervals of 5 ml/kg/minute and then allocated data points to their respective bandwidth irrespective of chronological age.

Results.

When replotted into bandwidths of functional equivalence, these data show that at the extremes of function, the young are separated from the old. Between these values, each functional bandwidth accommodates a wide age range. The decrement in function with chronological age is not smooth or well defined.

Conclusions.

We suggest that participants for research into healthy aging should be initially segregated into bands of functionally equivalent $\stackrel{·}{V}{\text{O}}_{2,\max }$ values irrespective of their chronological age. Subsequently, other physiological measurements should be made on every participant in the band in order to begin to define the physiological profile of the participants. By conducting longitudinal studies on every individual, it will be possible to chart the physiological history of each participant through various ages. Segregating participants into cohorts of functional equivalence with data handling blinded to chronological age may be of great utility in increasing our understanding of the inherent aging process.

Exercise, aging, V˙O_2max

STUDIES on the mechanisms underlying the physiology of healthy aging are often based on comparing and contrasting young and older participants in cross-sectional studies. Underlying this comparison is the assumption that the older the participant, the greater the decrement in physiological function. Furthermore, because physical activity is known to have a profound effect on physiological function, it is also likely to have a profound effect on the relationship between physiological function and aging. In this perspective, we have reevaluated published data on the relationship between age, physiological function, and exercise. We have used this reevaluation to formulate criteria for the selection of participants for the study of the inherent human aging process. The main body of data reevaluated are those that have related maximum oxygen uptake (⁠$\stackrel{·}{V}{\text{O}}_{2,\max }$⁠) to chronological age and are from two groups of participants: vigorous exercisers and nonexercisers. Other physiological functions not related to $\stackrel{·}{V}{\text{O}}_{2,\max }$ are also briefly discussed. The conclusions from these reevaluations have four components. (i) Vigorous exercisers are the participants of choice for the investigation of the inherent aging process. (ii) Equivalence rather than decrement of function can be demonstrated across a wide range of age. (iii) Participants for studies of the effects of aging on physiological function should be segregated into matching $\stackrel{·}{V}{\text{O}}_{2,\max }$ groups irrespective of age. (iv) Vigorous exercising individuals who show similar $\stackrel{·}{V}{\text{O}}_{2,\max }$ values will probably also show functional equivalence across other systems. We outline the rationale underlying these conclusions later.

REEVALUATION OF AGING AND$\stackrel{·}{V}{\text{O}}_{2,\mathrm{MAX}}$ DATA

$\stackrel{·}{V}{\text{O}}_{2,\max }$ is the maximum rate at which the body can utilize oxygen and is ultimately a reflection of the integrated performance of the cardiovascular, respiratory, and neuromuscular systems during maximal whole body exercise. It is viewed as the gold standard measure of health and physical fitness with low values being a key predictor of mortality (1). The data we have reevaluated are taken from two studies. The first is a meta-analysis of 242 studies on men (2) and the second from a very similar study (but not a meta-analysis) on women (3). For both data sets, we have taken the individual data points from figures in these articles, which plotted $\stackrel{·}{V}{\text{O}}_{2,\max }$ against chronological age and have simply stratified the $\stackrel{·}{V}{\text{O}}_{2,\max }$ values into bandwidth intervals of equivalent functionality (5 ml/kg/minute) irrespective of chronological age (Figure 1). This is the only manipulation of the data we have performed. Published figures were enlarged, and all discernable points were plotted. In the male exercisers (Figure 1), we have not plotted $\stackrel{·}{V}{\text{O}}_{2,\max }$ values for the ranges 65–80 ml/kg/minute because these ranges are the realm of young elite athletes, and the exclusion of this cohort from the figure does not alter in any significant way the particular conclusions drawn on this in the perspective.

Figure 1.

$\stackrel{·}{V}{\text{O}}_{2,\max }$ values taken from Wilson and Tanaka (2) (men, lower panel) and Tanaka and Seals (3) (women, upper panel) have been allocated to bandwidths of functional equivalence (5 ml/kg/minute). The data for men represent mean values from studies used in the meta-analysis and for women each data point represents the raw value for a given individual. The graphs separate vigorous exercisers and sedentary people and show that bandwidths representing different functional equivalence encompass a wide range of ages. Note: The data points plotted are only those that could be discerned from the original publications and a number of points are overlaid.

We note the following key observations in these data sets. First, in both genders, the data form two distinct groups, with $\stackrel{·}{V}{\text{O}}_{2,\max }$ values being consistently lower in the nonexercisers as compared with vigorous exercisers. Therefore, by comparison with the vigorous exercisers, the nonexercisers can be described as having suboptimal function. Although aging clearly declines to the same end point, namely death, these data suggest that the pathway to that end point for vigorous exercisers and nonexercisers may not be the same because of the well-described ameliorating effects of exercise on diseases associated with aging. Although there has been some attempt (4), there is no body of work that defines accurately the amount and intensity of exercise necessary to prevent the complications of disuse. Therefore, until this is defined, individuals who have exercised vigorously throughout their life span or for many years (like those in Figure 1) are the cohort of choice in which to study intrinsic human aging.

The second observation, which is clear in the male data where the data set is larger, is that in the exercisers, the highest functional group comprises only the youngest individuals. These young male athletes can clearly be distinguished from the oldest athletes who comprise the oldest functional group. Therefore, at the extremes, $\stackrel{·}{V}{\text{O}}_{2,\max }$ can differentiate clearly between both age and athleticism. However, in the middle functional ranges (45–60 ml/kg/minute), although $\stackrel{·}{V}{\text{O}}_{2,\max }$ segregates individuals into differing levels of athleticism, it does not definitively differentiate individuals on the basis of age. In other words, in the middle functional ranges, a wide age range is contained with each functional level and there is no clear-cut age-related decline.

We extend this concept of age-related functional equivalence to hypothesize the following. Exercisers who share a similar $\stackrel{·}{V}{\text{O}}_{2,\max }$ (eg, 55–60 ml/kg/minute) will also show functional equivalency across a wide range of other physiological, biochemical, hormonal, and immunological indices, even though there is wide age range encompassed within this group. If we then compare and contrast this functionally equivalent group with another group, for example, those in the lower 45–50 ml/kg/minute range, we further hypothesize that the diminution in $\stackrel{·}{V}{\text{O}}_{2,\max }$ will also be accompanied by diminution in other functionalities. This would imply that as inherent aging progresses (as we have outlined earlier this only occurs in vigorous exercisers) and physiological function diminishes, the synchrony and integration of physiology is maintained, at least up to a point. We are not aware of any published experimental data to support this particular interpretation, and published data that may be used to support our hypothesis are limited to examining those publications where individual data points have been plotted. However, examination of athletic performance data (eg, running, cycling, and swimming), which shows an essentially linear decline up to around the eighth decade followed by a marked increase in the rate of decline, suggests maintenance of synchrony and integration of physiological function existing, at least up to around the eighth decade (3,5). One study (6) that followed aerobic exercisers for more than a 33-year period may be interpreted as supporting the hypothesis given the relatively high $\stackrel{·}{V}{\text{O}}_{2,\max }$⁠, low measures of resting blood pressure, mean arterial pressure, heart rate reserve, and body fat. Our assumptions can easily be tested by carrying out an extensive analysis of other physiological functions in participants of differing ages sharing the same $\stackrel{·}{V}{\text{O}}_{2,\max }$ values. Grouping exercisers by functional equivalency rather than chronological age and then examining physiological function across a range of indices may allow us for the first time to map accurately those physiological indices (not pertinent to exercise) that are linked to $\stackrel{·}{V}{\text{O}}_{2,\max }$ and therefore to inherent aging.

This perspective is directed mainly at exercisers, those who show inherent aging, but if we analyze the behavior of the sedentary male individuals, we see that again those with high function (young) can be differentiated clearly from those with very low function (old). In the middle functional ranges, in this instance ranging from ∼30 to 50 ml/kg/minute, there is again no ability of $\stackrel{·}{V}{\text{O}}_{2,\max }$ to discriminate between ages. We emphasize the point once more that at every $\stackrel{·}{V}{\text{O}}_{2,\max }$ functional level, the sedentary show diminished function compared with exercisers of similar age.

FUNCTIONAL EQUIVALENCY IN OTHER PHYSIOLOGICAL SYSTEMS

We have also examined another physiological function that is unrelated to integrated cardiovascular respiratory function and lower limb explosive muscle power (7). When the power data are similarly stratified into groups of functional equivalence (Figure 2), it can be seen that at the extremes, power like $\stackrel{·}{V}{\text{O}}_{2,\max }$ can differentiate between athletes and nonathletes, but in contrast to $\stackrel{·}{V}{\text{O}}_{2,\max }$ in the middle functional groups, which are shared by the sedentary and the athletes, power cannot differentiate between either age or athleticism. This analysis shows the importance of measuring the appropriate functionality if the relationship between age and athleticism needs to be explored.

Figure 2.

Peak lower limb power output in trained male master weightlifters and sedentary individuals taken from Pearson and colleagues (7). Data have been allocated to bandwidths of functional equivalence (200 W). The data show functional equivalency across chronological ages but less clear segregation between groups than seen in Figure 1.

We have also considered data from top master performers in a task, 10 pin bowling, which requires an integration of neuromuscular systems, balance coordination, and skill and contrasts with the other two measures (8). Here, competition performance data were also plotted in bands of functional equivalence against age (Figure 3). In this cohort where athleticism of the individual is not defined, the measurement of function does not clearly differentiate the relationship between age and functionality.

Figure 3.

Competition scores for professional 10 pin bowlers taken from DeVan and Tanaka (8). Data were allocated in bandwidths of functional equivalence (10 points). These data show functional equivalence over a wide range of chronological age in these competitors.

CONCLUSION

Despite many decades of research into healthy aging, we are unaware of any studies that give unequivocal insights into the inherent aging process. The majority of human aging studies are of cross-sectional design, which is an inherent weakness. However, we suggest that interpretation of the aging process is weakened further where no information about the exercise status of participants is given (9). Because exercise has a powerful influence on physiological function, this information would seem to be mandatory in order to dissect the mechanisms underlying the intrinsic aging process. Vigorous exercisers, because of a relative lack of disuse-mediated health complications, show a decline in functional capacity that probably reflects the inherent aging process. The importance of knowing the exercise background of participants for studies has already been emphasized by other authors (10). We endorse and extend their ideas and propose that participants selected for the study of intrinsic aging should only be recruited from vigorous exercisers. We advocate the use of exercisers neither as models for sporting prowess nor as models for treating or ameliorating illness in the aged but because we are unaware of any procedure other than exercise that can ameliorate the effects of nonusage. That said, we acknowledge that this group is not entirely without limitations and may also show signs of disease (11).

Our analysis of the data has revealed that there are a large number of individuals who fall within the middle functional ranges. These individuals do not show a linear relation between age and functionality. We suggest that because $\stackrel{·}{V}{\text{O}}_{2,\max }$ is the best marker of integrated and synchronous whole body function, it is the function of choice with which to initially segregate participants for aging research. Every functional range of $\stackrel{·}{V}{\text{O}}_{2,\max }$ will no doubt include a wide age range. Subsequently, other physiological measurements, both exercise related and nonexercise related, should be carried out in order to begin to define the physiological profile of each of the participants in a particular functional range. We suggest that those exercisers who are grouped by $\stackrel{·}{V}{\text{O}}_{2,\max }$ equivalence will also show equivalency in other unrelated functions, irrespective of the ages of the individuals in each group. By measuring and comparing a range of physiological indices in each participant in a particular $\stackrel{·}{V}{\text{O}}_{2,\max }$ grouping and then doing a longitudinal follow-up of each participant, it should be possible to test the hypothesis that as master athletes age their physiology diminishes in a synchronous and coherent manner, as their continued high levels of performance suggest.

Author notes

Decision Editor: Darryl Wieland, MD, PhD