Endurance performance in masters athletes

Age-related declines in endurance performance have been observed in marathon and road running [8, 28, 51, 92, 99, 107], track endurance running [5, 35, 65, 90, 109], orienteering [7], indoor rowing [83], and long distance swimming [9, 27, 30, 45, 75, 93].

The age-related decrease in endurance performance of elite level masters endurance athletes appears curvilinear from age 35 years until approximately age 60–70 years and exponential thereafter. Specifically, endurance performance appears (Figs. 2 and 3) maintained until approximately 35 years of age, followed by modest decreases until 50 years of age, with a progressive decrease in performance thereafter with the greatest declines occurring after the age of 70 years in endurance running events [5, 94] and after age 70–80 years in 1,500 m [27, 30, 93] or 1-h swimming events [9].

A cross-sectional study of swimming performance from the U.S. Masters Swimming Championships observed a greater rate of decline in swim performance in females than males across all swim events from 50 to 1,500 m [93]. In endurance running events, the decrease in performance is greater in women compared to men, possibly due to either biological or sociological differences [94]. These authors suggested that these gender differences may partly be explained by selection bias. That is, there are a smaller number of female runners in the older age groups. This suggestion was supported by recent research examining the running times, age, and gender of 415,000 run performances in the New York Marathon between 1983 and 1999 [51]. They observed that female marathon participation showed a significantly greater percentage increase in all age groups compared to the males and that the finishing times for the top 50 male and female finishers over the past two decades showed significantly greater improvement in the masters’ age groups than the younger age groups, especially in the older female athletes [51].

The above age-related declines in endurance performance have been suggested to be due to decreased training volumes and intensities as a result of increased work and family commitments [87, 88, 103], behavioral factors such as reduced motivation to train [64, 94], few masters athletes having coaches [87], and masters athletes spending less overall time in training than international caliber younger athletes, even though the percentage time spent in the various types of training appears similar [87, 103]. Apart from these sociopsychological factors, previous research has shown that age-related reductions in VO_2max, and changes in lactate threshold and economy in masters athletes may also affect endurance performance.

Coyle [24] has identified that maximal aerobic power (VO_2max) is a major contributor to endurance performance. In both young [66] and older [28, 96] endurance athletes, VO_2max is a strong predictor of running performance with stronger correlations observed between endurance performance and VO_2max in populations heterogenous for VO_2max [52, 94].

VO_2max is estimated to decline approximately 10% per decade after the age of 25 years in healthy sedentary aging individuals [15]. An age-related reduction in VO_2max has also been well documented in both cross-sectional and longitudinal studies of male and female endurance athletes undertaking high levels of endurance training into older age [11, 32, 40, 42, 48, 53, 56, 57, 71, 72, 78, 106, 108], suggesting that some decrease in VO_2max appears inevitable with aging, despite physical training into older age. However, the rates of decline in VO_2max have been reported to be reduced [54, 55, 78], similar [108], or greater [47, 71] than age-matched sedentary individuals.

Earlier studies suggested the rate of decline in VO_2max of masters endurance athletes to be only half that observed in sedentary aging individuals [40, 48, 53]. In a classic longitudinal study, Kasch and others [53] measured VO_2max in 12 physically active men aged 44–79 years 10, 15, 20, and 25 years of follow-up after an initial investigation. The VO_2max of the group was 60% greater than those reported for sedentary men of similar age [15]. The decline in VO_2max was 13% (5% per decade) over the 25-year period. The rate of decline per annum of 0.24 mLkg⁻¹min⁻¹ was half of the 0.45 mLkg⁻¹min⁻¹ reported for sedentary aging men [15]. However, more recent studies have suggested accelerated decline in VO_2max in older endurance athletes as a result of decreased training volumes and intensities (32, 71) or the examined athletes possessing higher than normal initial VO_2max values at baseline (94).

Gender differences in the rate of age-related decline in VO_2max are commonly observed in masters endurance athletes [11, 104]. For example, Brown et al. [11] recently reported declines of 0.65 and 0.39 mLkg⁻¹min⁻¹year⁻¹ in male (17–64 years) and female (16–54 years) high performance cyclists, respectively. Fitzgerald et al. [32] used meta-analysis to suggest that VO_2max in aging females declines at different rates in sedentary (0.35 mLkg⁻¹min⁻¹year⁻¹), active (0.44 mLkg⁻¹min⁻¹year⁻¹), and endurance-trained (0.62 mLkg⁻¹min⁻¹year⁻¹) subjects. Wilson and Tanaka [106] also used meta-analysis to suggest age-related declines in VO_2max of 0.40, 0.39, and 0.46 mLkg⁻¹min⁻¹year⁻¹ in sedentary (n = 6,231), active (n = 5,261), and endurance-trained (n = 1,961) males. Hawkins et al. [47] reported a longitudinal study of 86 male (53.9 ± 1.1 year) and 49 female (49 ± 1.2 year) masters endurance runners tested 8.5 years after initial measures of VO_2max. Their findings suggested that VO_2max declines in male and female masters athletes at a rate similar to or greater than that expected in sedentary older adults. Soon after, Pimental et al. [71] observed that VO_2max declined by 0.54 mLkg⁻¹min⁻¹year⁻¹ in 89 endurance-trained men aged 21–74 years compared to a decline of 0.39 mL kg⁻¹.min^-1.year⁻¹ in 64 sedentary men aged 20–75 years. However, when stratifying for age, the endurance-trained subjects VO_2max declined by 0.2 mLkg⁻¹min^-1year⁻¹ between 20–50 years, increasing to 0.89 mLkg⁻¹min⁻¹year⁻¹ between ages 50 and 74 years.

These differences in rates of decline in VO_2max may be explained by the decreases in training volume, frequency, and intensity commonly observed in masters endurance athletes [42, 61, 71, 100, 106, 108]. These changes in training habits have been shown to influence the rate of decline in VO_2max in both male and female masters endurance athletes [56, 71, 72, 94, 101]. In their classic longitudinal study, Trappe et al. [101] initially examined 53 highly trained and competitive distance runners. Twenty-two years later, these men were classified as highly trained (n = 10), fitness trained (n = 18), untrained (n = 15), and fit older (n = 10), depending on their continued level of training and age. All groups experienced a significant decrease in VO_2max that was related to the amount of training between evaluations. The reductions were 6, 10, 15, and 15% lower per decade for the highly trained, fitness trained, untrained, and fit older, respectively. Similarly, Katzel et al. [56] compared longitudinal changes in VO_2max in 41 male endurance runners who, after initial testing 4–13 years prior, were divided into ‘high training” (n = 7), “moderate training” (n = 21), and “low training” (n = 13) groups, depending on their training regimes and exhibited age-related declines in VO_2max of 2.6, 4.6, and 4.7%, respectively. Taken together, these results suggest that high-intensity training and maintenance of training volume may mediate the age-related declines in VO_2max and endurance performance. This would suggest that high intensity and high volume training is important in maintaining or attenuating age-related decreases in VO_2max and endurance performance, a suggestion strongly supported by research focused on maximizing endurance performance in younger athletes [14].

In summary, in both male and female masters endurance athletes, there is an age-related decrease in VO_2max that contributes to decreased endurance performance. However, the decline in VO_2max is dependent on the degree to which the athletes maintain training volume and intensity, the age of the subjects, or the heterogeneity of the group(s) studied.

According to the modified Fick equation, VO_2max is the product of both maximal heart rate and maximal stroke volume (central factors), and maximal arteriovenous oxygen difference (peripheral factors [4]). Arteriovenous oxygen difference is further influenced by a variety of factors including muscle mass, the capacity of the blood to transport and relinquish oxygen (blood volume, hemoglobin), and the capacity of the working tissues to take up and utilize oxygen (capillarization, muscle fiber type, aerobic enzyme activity). To fully understand the age-related declines in VO_2max, an examination of each of these factors in masters endurance athletes is warranted.

Coyle [24] identified that lactate threshold is a major determining factor in endurance performance. Lactate threshold is defined as the exercise intensity (velocity or %VO_2max) at which blood lactate increases significantly above baseline level. Research has consistently shown that both VO₂ at lactate threshold [31] and velocity at lactate threshold are better predictors of endurance running performance than VO_2max [6, 68] in younger distance runners. In older runners, it appears that endurance running performance is correlated with both VO_2max and velocity at lactate threshold in male runners [50, 96] and highly trained older female runners [28, 108]. The latter study used stepwise regression analyses on subjects pooled from different age groups and determined that the majority (60%) of the variability in performance for runners aged 23–47 year was explained by the running velocity at which LT occurred, whereas VO_2max explained the majority (74%) of the variability for the runners aged 37–56 years. They concluded that both decreases in VO_2max and running velocity at LT were the two physiological phenomena most closely associated with age-related declines in 10-km run performance in highly trained female runners. However, the contributions of these two mechanisms to the declines in endurance performance appear not to be uniform with advancing age.

When expressed as a percentage of VO_2max, lactate threshold does not appear to decrease significantly with age in either male [2, 107] or female [28, 108] distance runners. Allen et al. [2] observed that male masters runners (56 ± 5 year) had a significantly (9%) lower VO_2max than young runners (25 ± 3 year), despite being matched for 10-km run performance. While running economy was not significantly different between the groups, the masters runners attained a 2.5-mmolL⁻¹ blood lactate level during steady-state exercise at a higher percentage of their VO_2max (92 vs 89%), although both groups attained this lactate level at the same running speed and VO₂. Similarly, the later longitudinal study by Wiswell et al. [107] observed that lactate threshold as %VO_2max did not differ between male and female endurance runners and increased significantly with age in both groups. In a heterogeneous population of aging endurance athletes, it is the exception to the rule to have older athletes performing and younger athletes in any given event. Thus, when older and younger endurance athletes are not matched for performance, endurance performance velocity decreases with age [5‐9, 27, 39, 35, 51, 92, 94, 109] suggesting that the physiological steady-state during endurance performance may be influenced by age.

While the research examining lactate threshold in masters athletes suggests aging may improve or have minimal effect on lactate threshold as a percentage of VO_2max [5, 6, 107], more recent research has examined the maximal lactate steady state in masters athletes [63]. Maximal lactate steady state is defined as the highest concentration of blood lactate that can be maintained (±1 mmolL⁻¹) during the last 20 min of a 30-min constant workload test. Mattern et al. [63] determined VO_2max and maximal lactate steady state in three age groups: young (25.9 ± 1.0 year), middle-aged (43.2 ± 1.0 year), and older (64.6 ± 2.7 year) of male competitive cyclists and triathletes matched for training intensity and duration. The researchers observed significant differences in VO_2max among all age groups: 67.7 ± 1.2 mLkg⁻¹min⁻¹, 56.0 ± 2.6 mLkg⁻¹min⁻¹, 47.0 ± 2.6 mLkg⁻¹min⁻¹ in the young, middle-aged and older athletes, respectively. When expressed as a percentage of VO_2max, there was also a significant age-related decrease in the relative maximal lactate steady-state exercise intensity: 80.8 ± 0.9%, 76.1 ± 1.4%, 69.9 ± 1.5% in the young, middle-aged, and older athletes, respectively. Thus, it would appear that when masters endurance athletes are not matched for performance, both the velocity at race pace and lactate steady state decline with age.

The above studies examining the effect of age on maximal lactate steady state or lactate threshold have used cross-sectional study designs. Marcell et al. [61] used both a cross-sectional and longitudinal study design to explore the relationship between lactate threshold and running performance in male (n = 51) and female (n = 23) masters endurance runners aged 39 to 77 years over a period of 5.8 ± 1.6 years. They observed a significant decrease in both VO_2max and training volume irrespective of age and gender and an increase in lactate threshold as a percent of VO_2max with age. Crucially, the changes in lactate threshold were not related to changes in fitness or performance and were very unstable over time. They concluded that the lactate threshold may be less precise than VO_2max or performance in the prescription of exercise intensities in older endurance athletes, supporting the earlier suggestion of Wiswell et al. [107] who also observed age-related increases in lactate threshold as a percent of VO_2max in male and female endurance runners.

In summary, the available data suggests that when expressed as a performance velocity, there is an age-related reduction in lactate threshold and maximal lactate steady state that contributes to a reduction in endurance performance in masters athletes. However, when expressed as a percentage of VO_2max, lactate threshold increases in masters endurance athletes, suggesting this decrease in lactate threshold exercise intensity with age may be secondary to age-related decreases in VO_2max.

In modeling endurance performance, Coyle [24] identified that economy is a major determining factor in endurance performance. Economy is defined as the oxygen cost to exercise at a given exercise intensity (velocity or %VO_2max) and has been shown to be a stronger predictor of endurance performance than VO_2max in a homogenous group of endurance athletes [2, 19].

A few studies have examined economy in masters endurance athletes [2, 28]. In a study of male 10-km runners matched for run performance, running economy was similar in both young (25 ± 3 year) and older (56 ± 5 year) runners [2]. Similarly, Evans et al. [28] tested the hypothesis that declines in 10-km run performance in females was associated with decreases in VO_2max, lactate threshold, and running economy. In 31 highly trained female runners aged 23–56 years, they observed that both 10-km performance and age were significantly correlated with VO_2max and running velocity and VO₂ at lactate threshold. However, both 10-km performance and age were not correlated with running economy in the highly trained and competitive female endurance runners [28].

In summary, exercise economy appears to not appear to change with age in masters endurance athletes suggesting that economy does not contribute significantly to age-related decreases in endurance performance commonly observed in masters athletes.

A decreased muscle mass has been suggested as a contributor to the age-related decline in VO_2max in sedentary aging individuals [15, 33]. Fleg and Lakatta [33] measured 24-h urinary creatinine excretion, (an index of muscle mass) in 184 healthy volunteers aged 22–87 years from the Baltimore Longitudinal Study of Aging. They observed a significant positive correlation between VO_2max and creatinine excretion in both men and women. VO_2max showed a strong negative linear relationship with age in both men and women. When VO_2max was normalized for creatinine excretion, the variance in the VO_2max decline attributable to age declined from 60 to 14% in men and from 50 to 8% in women, suggesting that muscle mass may influence the age-related decline in VO_2max observed with age in healthy adults.

Muscle atrophy may also play a role in the observed age-related decrease in VO_2max in masters endurance athletes [47, 73]. Early studies examining declines in endurance performance and VO_2max in masters endurance athletes measured lean body mass and observed age-related declines in VO_2max relative to both total body weight or lean body mass [33, 48, 70]. While these early studies observe age-related declines in VO_2max relative to lean body mass in masters endurance athletes, they did not statistically adjust their data for age or gender differences in body composition, a consistent age-related finding in masters endurance athletes [47, 72, 101].

In the first study to examine the effect of decreases in muscle mass on VO_2max in older athletes, Proctor and Joyner [73] observed that approximately 50% of the difference in whole body lean body mass could be explained by reduced appendicular muscle mass in older vs younger endurance athletes. Moreover, they observed that approximately 14% of the age difference in VO_2max relative to appendicular muscle mass could be explained by this age-related difference. Hawkins et al. [47] later examined longitudinal changes in VO_2max, HR_max, and training volume in 86 male (53.9 ± 1.1 year) and 49 female (49.1 ± 1.2 year) masters endurance runners over 8.5 years. VO_2max and HR_max declined significantly regardless of gender or age. Men with the greatest loss in VO_2max had the greatest loss in lean body mass, suggesting that decreased muscle mass may have contributed to the age-related reduction in VO_2max. However, studies such as the one by Hawkins et al. [47] that have suggested age-related decreases in VO_2max in masters endurance athletes are due to decreases in lean body mass are difficult to interpret due to the confounding effects of age-related changes in body fat and muscle oxidative capacity [15, 33].

One of the most commonly observed declines with aging is a decrease in muscle mass [13, 28]. While several cross-sectional studies of strength- or power-trained masters athletes observed greater muscle mass compared to sedentary age-matched controls [58, 86], it appears that chronic endurance training into older age does not maintain overall muscle mass [44, 58, 85]. These studies showed that masters endurance athletes had similar muscle mass as age-matched sedentary controls and reduced muscle mass compared to younger sedentary or young endurance-trained individuals.

Recent studies have been able to measure appendicular muscle mass in aging sedentary and endurance trained individuals using dual-energy X-ray absorptiometry [13, 91]. Sugawara et al. [91] studied 131 healthy men aged 20–79 years who were either sedentary or endurance-trained. Appendicular (arms and legs) muscle mass in the endurance-trained middle-aged and older men was significantly higher than their sedentary peers. However, the rate of decline in appendicular muscle mass with age (1.3 gkg⁻¹ BWyear⁻¹) was similar between the two groups. The same investigators observed a positive correlation (p < 0.01) between VO_2max and appendicular muscle mass relative to total body mass, suggesting that the age-related decrease in appendicular muscle mass contributes to the age-related decrease in VO_2max in masters endurance athletes.

While active muscle mass may decrease in size in masters endurance athletes, a number of studies have suggested better muscle structure and architecture in older endurance-trained individuals compared to age-matched sedentary controls [58, 85, 86]. In an investigation of the quadriceps size in male [85] and female [86] endurance athletes who had a lifelong training history in competitive sports, the athletes exhibited firmer fasciae and connective tissue septa but less fat and connective tissue infiltration than age-matched sedentary controls.

In summary, it would appear that endurance training into older age may not reduce the age-related loss of muscle mass observed in a sedentary aging population, but the quantity and quality of muscle may be enhanced through maintenance of contractile tissue. Moreover, it appears that the age-related decrease in lean body mass and appendicular muscle mass in particular, contributes to the age-related decrease in VO_2max and thus endurance performance in masters athletes.

The modified Fick equation suggests that VO_2max is the product of both central factors as well as maximal arteriovenous oxygen difference (peripheral factors [4]). A common observation in studies of masters endurance athletes is that there is an age-related decrease in arteriovenous oxygen difference [76, 94]. However, it appears that when young and older endurance athletes are matched for training volume, this age-related decrease is attenuated or disappears [34, 40], suggesting that the well-observed decrease in training volume and intensity with age in masters endurance athletes [54, 56, 100, 101] may explain the decline in arteriovenous oxygen difference.

To explain this age-related decline in arteriovenous oxygen difference, the peripheral factors identified by Coyle [24] as influencing endurance performance will be discussed. These include muscle fiber composition and size, muscle capillarization, and aerobic enzyme activity.