Impaired walking performance is a key predictor of morbidity among older adults. A distinctive characteristic of impaired walking performance among older adults is a greater metabolic cost (worse economy) compared to young adults. However, older adults who consistently run have been shown to retain a similar running economy as young runners. Unfortunately, those running studies did not measure the metabolic cost of walking. Thus, it is unclear if running exercise can prevent the deterioration of walking economy.
To determine if and how regular walking vs. running exercise affects the economy of locomotion in older adults.
15 older adults (69±3 years) who walk ≥30 min, 3x/week for exercise, “walkers” and 15 older adults (69±5 years) who run ≥30 min, 3x/week, “runners” walked on a force-instrumented treadmill at three speeds (0.75, 1.25, and 1.75 m/s). We determined walking economy using expired gas analysis and walking mechanics via ground reaction forces during the last 2 minutes of each 5 minute trial. We compared walking economy between the two groups and to non-aerobically trained young and older adults from a prior study.
Older runners had a 7–10% better walking economy than older walkers over the range of speeds tested (p = .016) and had walking economy similar to young sedentary adults over a similar range of speeds (p = .237). We found no substantial biomechanical differences between older walkers and runners. In contrast to older runners, older walkers had similar walking economy as older sedentary adults (p = .461) and ∼26% worse walking economy than young adults (p<.0001).
Running mitigates the age-related deterioration of walking economy whereas walking for exercise appears to have minimal effect on the age-related deterioration in walking economy.
Citation: Ortega JD, Beck ON, Roby JM, Turney AL, Kram R (2014) Running for Exercise Mitigates Age-Related Deterioration of Walking Economy. PLoS ONE 9(11): e113471. doi:10.1371/journal.pone.0113471
Editor: Yuri P. Ivanenko, Scientific Institute Foundation Santa Lucia, Italy
Received: September 8, 2014; Accepted: October 23, 2014; Published: November 20, 2014
Copyright: © 2014 Ortega et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The authors confirm that all data underlying the findings are fully available without restriction. All relevant data are within the paper.
Funding: Support was provided by the California State University Program for Education and Research in Biotechnology New Investigator [Grant #: HM531] (http://www.calstate.edu/csuperb/grants/). This funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Support was also provided by the National Institutes of Health Clinical and Translational Science Award [Grant #: UL1 TR000154] (http://www.ncats.nih.gov/research/cts/ctsa/funding/funding.html). This funder provided the facility where the data was collected.
Competing interests: The authors have declared that no competing interests exist.
Walking performance typically deteriorates with advanced age , and impaired walking performance is a key predictor of morbidity among older adults . A distinctive characteristic of impaired walking performance among older adults is a 15–20% greater metabolic cost for walking (worse economy) compared to young adults –. Several factors are known to determine the metabolic cost of walking in humans across all ages. These major biomechanical factors include the costs associated with: supporting body weight, performing mechanical work, leg swing and balance –. Studies investigating age-related biomechanical determinants of walking cost have found that older adults have a similar cost of balance and perform a similar amount, or even less, external mechanical work during walking as young adults , , . Despite these similarities, other studies suggest that a decrease in muscular efficiency and an increase in antagonist leg muscle co-activation, contribute to the greater cost of walking in both healthy sedentary and active older adults , , , , . Yet, no study has found a sole mechanical determinant that accounts for the 15–20% greater metabolic cost of walking in older adults. Therefore, interventions for improving walking economy in older age have been elusive.
Recent studies by Thomas et al.  and Malatesta et al.  show that vigorous walking interval training effectively reduces the metabolic cost of walking in older adults by as much as 20%. Yet, the mechanisms for the decreases were not elucidated. Conversely, a generalized year-long training program that included resistance, aerobic and balance exercises had no effect on post-training walking economy in older adults . The different effects of these exercise interventions, high intensity aerobic versus generalized exercise with only a moderate aerobic component, suggest higher intensity aerobic activities may mitigate the typical age-related decrease in walking economy, and consequently, preserve mobility into older age.
In contrast, running economy does not exhibit the same age-related trend as walking economy. Two studies have reported that adults (45–61 years) who consistently participated in running exercise retain a similar metabolic economy of running as young runners (23–27 years) , . Although these results seem to support the hypothesis that vigorous aerobic exercise mitigates the decline in locomotion economy, i.e. metabolic cost of running and walking, it is also possible that a decline in running economy does not occur until late into the 6th decade of life, as observed with walking economy . Perhaps the subjects in these studies ,  were not “old” enough to exhibit declines in locomotion economy. Another possible explanation is that running economy, unlike walking economy, is simply not affected by age. However, since these running studies did not measure walking economy, it remains unclear if regular participation in running exercise mitigates the typical age-related deterioration of walking economy.
Our purpose was to determine if and how regular participation in walking or running exercise affects the metabolic cost and biomechanics of walking in older adults. We hypothesized that older runners would consume less metabolic energy for walking than older walkers. Further, we also investigated whether the two groups demonstrate different walking biomechanics. We measured metabolic rates, ground reaction forces and spatio-temporal stride variables of two groups, older walkers and older runners, while they walked on a dual-belt, force-sensing treadmill at three speeds.
Thirty healthy older adults (15 males and 15 female) who either walk (4 Male, 11 Female) or run (10 Male, 5 Female) regularly for exercise volunteered. Table 1 summarizes the anthropometric characteristics of the subjects. We recruited subjects with a minimum age of 65 years, which is in accordance with prior studies reporting age-related impairments of walking performance become most apparent at this age , –. All subjects were free of neurological, orthopedic and cardiovascular disorders. Walkers self-reported walking for exercise three or more times per week for at least 30 minutes per bout and for at least six months prior to the study. Runners self-reported running for exercise three or more times per week for at least 30 minutes per bout and for at least six months prior to the study. The experiment was performed in accordance with the ethical standards of the 1964 Declaration of Helsinki and was approved by the Humboldt State University and University of Colorado Institutional Review Boards. All subjects gave written informed consent prior to participation in the study.
Subjects completed three sessions. In the first session, subjects underwent a physician's examination to determine neurological, orthopedic and cardiovascular health, a body composition test (DXA) to determine percent body fat and lean tissue mass and a VO2 max treadmill test to determine maximal aerobic capacity. In the second session, at least five days following the first session, we measured standing metabolic rate and familiarized the subjects to treadmill walking. For the treadmill familiarization, subjects walked on a dual-belt, force-instrumented treadmill (FIT, Bertec Corporation, Columbus, OH, USA) at three speeds (0.75, 1.25 and 1.75 m/s) for at least 7 minutes at each speed. These speeds correspond to 1.67, 2.80, 3.91 MPH. Thus, subjects completed a minimum of 21 minutes total of walking familiarization. This familiarization period is over double the recommended minimum treadmill habituation time of 10 minutes , . In the third session, at least two days following familiarization, we measured each subject's metabolic rate during quiet standing and while walking on the treadmill at three speeds (0.75, 1.25 and 1.75 m/s) in random order. All trials were five minutes in duration with at least five minutes of rest between trials. Throughout each trial, we measured the rates of oxygen consumption (VO2) and carbon dioxide production (VCO2) in order to determine metabolic rate. We calculated the average VO2 and VCO2 for the last two minutes of each trial. We also measured ground reaction forces (GRFs) from the force-instrumented treadmill for 1 minute during the last 2.5 minutes of each trial to determine kinetics and spatio-temporal stride variables.
Metabolic Power Consumption
We measured VO2 and VCO2 using an open-circuit expired gas analysis system (TrueOne 2400, ParvoMedic, Sandy, UT, USA). We calculated average gross metabolic power per kilogram body mass (W/kg)  using the average VO2 (mlO2/min) and VCO2 (mlCO2/min) for the last two minutes of each trial, when VO2 and respiratory exchange ratio reached steady state ensuring that each subject was working sub-maximally and oxidative metabolism was the main metabolic pathway. We then divided gross metabolic power by speed to calculate gross metabolic cost of transport (CoT) (J/kg/m) for walking.
Ground Reaction Forces and Spatio-temporal Stride Variables
For each walking trial, we collected the ground reaction forces (vertical and horizontal components) of each leg from the force-sensing treadmill at 2000 Hz for a 1 minute period during the last 2.5 minutes of each trial. A custom MATLAB script (Math Work Inc., Natick, Mass) was then used to process all force data. The GRF data were filtered with a 4th order zero-lag low pass Butterworth filter with a cutoff frequency of 30 Hz. For each trial, we calculated vertical and horizontal peak GRFs across all 10 strides. Using the filtered GRF data, we determined gait cycle events and spatio-temporal stride variables (stride frequency, stance time, and duty factor as percent of the gait cycle) for 10 strides of each trial (10 steps per each leg).
We used a repeated-measures ANOVA (p<.05) to determine statistical differences due to exercise group (walkers vs. runners) and walking speed, as well as, the exercise group-walking speed interaction. When a significant main effect of exercise group was found, we performed independent-samples t-tests with Bonferroni correction to determine at which speed(s) the differences occurred. To determine if difference in metabolic cost and GRF was related to sex differences in our runner and walker groups, we examined differences in metabolic cost, ground reaction forces and spatio-temporal stride variables due to sex among each group and analyzed difference in metabolic cost, GRFs and spatio-temproal stride variables using sex as a covariate. We found no effect of sex on any dependent variable and differences between runners and walkers were not affected by sex. We performed all statistical analyses using SPSS 21.0 (SPSS, Inc.) software. In addition to our comparison between older walkers and runners, we used a mixed-model repeated-measures ANOVA (p<.05) to make further post-hoc comparisons of gross metabolic cost in walkers and runners collected in the present study to data for young and older sedentary adults previously collected in our lab at similar speeds . To make these comparison between exercise/age group (old walkers, old runners, old sedentary and young sedentary) using a linear mixed model, walking speed squared (m/s)2 was used as the repeated measure.
In support of our hypothesis, older runners consumed 7–10% less metabolic energy for walking than older walkers across the range of speeds tested (Fig. 1; p = .016). Gross metabolic power consumption increased significantly across the range of walking speeds tested in both older runners and walkers, (p<.0001). Compared to walking at the slowest speed of 0.75 m/s, gross metabolic power increased by 95% to walk at 1.75 m/s in older walkers but only 86% in older runners (speed X group interaction, p = .009). Mass-specific standing metabolic rates were similar between older runners and walkers (p = .250; Table 1).
Lines represent least square regression for older walkers (y = 2.709x2–3.539x+4.523, r2 = 0.86) and older runners (y = 2.382x2–3.189x+4.233, r2 = 0.89). Symbols shown on vertical axis represent standing metabolic rate of both groups. Asterisks (*) indicate significant differences between older runners and walkers (p<0.05).
Following from the metabolic rate data, the older runners had an average of 7–10% lower gross metabolic cost of transport compared to the older walkers. Older walkers and runners exhibited similar U-shape relations between gross CoT and walking speed (Fig. 2). Between the three speeds, gross CoT was significantly lower at the intermediate speed of 1.25 m/s as compared to the faster and slower walking speeds in both the older walkers (3.49±0.09 J/kg/m, p<.0001) and older runners (3.18±0.08 J/kg/m, p<.0001). Although there were a greater number of male runners in the study, our statistical analysis showed that the difference in metabolic cost between runners and walkers was not due to sex or any other anthropometric variable.
Asterisks (*) indicate significant differences between older walkers and runners (p<.05).
Despite the substantial differences in walking economy, older walkers and runners exhibited nearly identical spatio-temporal stride variables and kinetics across the range of speeds (Table 2). Among spatio-temporal gait characteristics, we found no significant differences between older walkers and older runners in regards to stride time, stride frequency (p = .879), single leg stance time (p = .126) or duty factor (p = .126). However, older runners walked with slightly (6%) shorter strides in relation to their leg length compared to older walkers (p = .033). This difference remained nearly constant across the range of speeds. With regards to ground reaction forces, older walkers and runners exhibited similar first (p = .838) and second (p = .282) peak vertical ground reaction force (Figure 3). Additionally, peak anterior-posterior braking (p = .182) and propulsive (p = .056) ground reaction forces were similar for both exercise groups.
We also compared gross metabolic cost of walking for older walkers and older runners to data from young and older sedentary adults collected in our lab from a prior study over a similar range of speeds . The speeds used in these two studies were slightly different. Thus, in order to statistically make this comparison using a linear mixed model repeated measures ANOVA, we determined gross metabolic power as a function of speed squared (Fig. 4). The results of this analysis showed that across the range of speeds, older walkers consume metabolic energy at a similar rate as sedentary older adults (p = .461) and 14–22% faster than young sedentary adults (p<.0001). In contrast, older runners consume metabolic energy at a slower rate compared to older sedentary adults (p = .016). However, our most striking finding was that older runners consumed metabolic energy at a similar rate as young sedentary adults across the range of walking speeds (p = .237).
Lines denote least square regression within each group (older sedentary: y = 1.46x+2.30, r2 = 0.91; older walkers: y = 1.31x+2.52, r2 = 0.86; older runners: y = 1.12x+2.42, r2 = 0.88; young sedentary: y = 1.01x+2.27, r2 = 0.87). Symbols on vertical axis represent standing metabolic rate of each group.
Discussion and Conclusions
In this study, we distinguished the effects of regular walking vs. running exercise on the metabolic cost and biomechanics of walking in older adults. In support of our hypothesis, older runners consumed less metabolic energy for walking than older walkers. Although the older runners consumed less metabolic energy for walking than the older walkers, the two groups had almost identical walking biomechanics.
Given that there were virtually no differences in walking biomechanics between the older walkers and runners, other factors must underlie the lower cost of walking observed for the older runners. One factor may be muscle co-activation. Older adults, both sedentary and active walkers, use 30–50% greater co-activation of antagonist leg muscles compared to young adults , , . It has been suggested that older adults may use greater co-activation to increase joint stiffness and the stabilization of the body, thus reducing the risk of walking related falls . Yet, increased co-activation has been associated with increased metabolic cost of walking in older adults , . It is possible that older runners are able to maintain a lower metabolic cost of walking compared to older walkers because they use less antagonist leg muscle activation. Some research shows that older adults who participated in a lower limb strength training program reduce leg muscle co-activation by 5–10% . Perhaps, by regularly running three or more times per week for 30 minutes per bout, older runners are able to maintain or even increase leg muscle strength and reduce co-activation. However, a decrease in co-activation associated with running that is similar in magnitude to the decrease observed after strength training is likely not sufficient to explain the 7–10% difference in metabolic cost of walking. It is also possible that other neuromuscular factors such as widening of EMG/motoneuronal bursts  may also help to explain the difference in metabolic cost between older runners and walkers.
Better muscular efficiency may also help explain why older runners have a lower metabolic cost of walking than older walkers. Aging has been associated with reduced muscular efficiency , . More specifically, mitochondrial dysfunction associated with the uncoupling of oxidative phosphorylation (reduced ATP synthesis per O2 uptake) effectively reduces muscular efficiency and increased the metabolic cost of muscle activation . Interestingly, recent evidence suggests that aerobic exercise training may ameliorate mitochondrial uncoupling and improve muscular efficiency in older adults .
Perhaps studies of cycling efficiency in older adults can provide insight. In contrast to the effects of running we have observed, the muscular efficiency of cycling declines with age despite regular cycling exercise . More recently, Brisswalter et al.  measured the cycling efficiency of active triathletes (who regularly swim, bike, and run for exercise) across age-groups and found a decline in cycling efficiency past the 5th decade. These data suggest that older cyclist and triathletes are unable to maintain muscular efficiency with age. However, Peiffer et al.  found no difference in cycling efficiency between their youngest age group (39±3 years) and their oldest (65±4 years). Intriguingly, their oldest training group cycled 58 km more per week (359 km per week) than the youngest group. Possibly the greater quantity of aerobic cycling exercise mitigated the decrease in muscular efficiency with age.
Alternatively, the intensity of exercise may hold the key to maintaining or improving muscular efficiency. Two prior studies have found that 6–7 weeks of vigorous aerobic exercise (fast walking) that elicits a heart rate close to the ventilatory threshold can improve walking economy by 8–20% , . More vigorous aerobic exercise such as walking uphill, fast walking or running may be required to elicit improvement in walking economy. Clinicians and others who work with older adults to improve their fitness may need to prescribe more vigorous, more prolonged and/or more frequent aerobic exercise to prevent the decline in walking performance. To test this hypothesis and help guide clinicians, a future study should investigate the effects of different intensity aerobic exercises on muscular efficiency and more specifically, the economy of walking.
One limitation of the current study is the cross-sectional design. It is possible that older runners may not be economical walkers because of the effect of running exercise but rather they run because they are more economical in their locomotion. To better address this issue, a future study might quantify the longitudinal effects of a running training program. One such study conducted by Trappe et al.  on the longitudinal effect of running exercise on running economy spanned 22 years. In that study, Trappe et al.  showed that running economy did not decline in older adults who maintained their health and fitness over the 22 year period, whereas runners who became unfit had worse running economy. Although these results suggest that running may help to prevent a decline in running economy, Trappe et al.  did not measure walking economy.
Another potential limitation of the current study is the different numbers of male and female participants in each group. Although the sex difference may have influenced the difference in anthropometrics between runners and walker, our results showed no main effect of sex on walking economy (p = .211) and no sex difference in walking economy among older runners (p = .131) or older walkers (p = .331). Based on post-hoc power analysis, it is clear that we did not have sufficient statistical power to detect sex differences that might exist but that would require ∼300 subjects. However, when treated as covariates, sex and anthropometrics did not statistically account for the difference in walking economy between runners and walkers. Thus, while it would have been preferable to have a larger sample size with more similar sex and anthropometric matched cohorts, it would not have changed our overall conclusion.
Based on the results of this study and others, future studies of the effect of age and exercise on walking economy are warranted. Although the average age of our runners and walkers was 69 years, a future study might look to see if running exercise continues to prevent or slow the decline in walking economy in even older runners (over the age of 80 years). It seems plausible that at some age that exercise may not be able to sufficiently offset the normal decline in muscular efficiency and walking economy associated with aging. It is also not known whether there is an intensity threshold of aerobic exercise that is needed to prevent the decline in walking economy. Thus, it would be beneficial for future studies to investigate the relative effect of exercises with different levels of aerobic intensity on walking economy.
In conclusion, older runners mitigate the age-related deterioration of walking economy. However, older walkers are unable to forestall the decline of walking economy as they require the same metabolic consumption as sedentary older adults. The difference in walking economy between older runners and older walkers remains unexplained due to no substantial differences found in either the kinetic or spatio-temporal data between the groups. Other factors such as decreased muscle co-activation and/or increased muscular efficiency may contribute to the superior walking economy exhibited by the older runners.
Conceived and designed the experiments: JO OB JR RK. Performed the experiments: JO OB JR AT RK. Analyzed the data: JO OB JR AT RK. Contributed reagents/materials/analysis tools: JO RK. Wrote the paper: JO OB JR AT RK.
- 1. Himann JE, Cunningham DA, Rechnitzer PA, Paterson DH (1988) Age-Related-Changes in Speed of Walking. Med Sci Sports Exerc 20: 161–166. doi: 10.1249/00005768-198804001-00010
- 2. Studenski S, Perera S, Patel K, Rosano C, Faulkner K, et al. (2011) Gait speed and survival in older adults. JAMA 305: 50–58. doi: 10.1001/jama.2010.1923
- 3. Martin PE, Rothstein DE, Larish DD (1992) Effects of age and physical activity status on the speed-aerobic demand relationship of walking. J Appl Physiol 73: 200–206.
- 4. Waters RL, Lunsford BR, Perry J, Byrd R (1988) Energy-speed relationship of walking: standard tables. J Orthop Res 6: 215–222. doi: 10.1002/jor.1100060208
- 5. Ortega JD, Farley CT (2007) Individual limb work does not explain the greater metabolic cost of walking in elderly adults. J Appl Physiol 102: 2266–2273. doi: 10.1152/japplphysiol.00583.2006
- 6. Ortega JD, Farley CT (2014) Effects of aging on mechanical efficiency and muscle activation during level and uphill walking. J Electromyogr Kinesiol 0.
- 7. Ortega JD, Fehlman LA, Farley CT (2008) Effects of aging and arm swing on the metabolic cost of stability in human walking. J Biomech 41: 3303–3308. Epub 2008 Sep 3323.
- 8. Gottschall JS, Kram R (2005) Energy cost and muscular activity required for leg swing during walking. J Appl Physiol 99: 23–30. doi: 10.1152/japplphysiol.01190.2004
- 9. Franz JR, Lyddon NE, Kram R (2012) Mechanical work performed by the individual legs during uphill and downhill walking. J Biomech 45: 257–262. doi: 10.1016/j.jbiomech.2011.10.034
- 10. Mian OS, Thom JM, Ardigo LP, Narici MV, Minetti AE (2006) Metabolic cost, mechanical work, and efficiency during walking in young and older men. Acta Physiol Scand 186: 127–139. doi: 10.1111/j.1748-1716.2006.01522.x
- 11. Malatesta D, Simar D, Dauvilliers Y, Candau R, Borrani F, et al. (2003) Energy cost of walking and gait instability in healthy 65- and 80-yr-olds. J Appl Physiol 95: 2248–2256.
- 12. Thomas EE, Vito GD, Macaluso A (2007) Speed training with body weight unloading improves walking energy cost and maximal speed in 75- to 85-year-old healthy women. J Appl Physiol 103: 1598–1603. doi: 10.1152/japplphysiol.00399.2007
- 13. Malatesta D, Simar D, Ben Saad H, Prefaut C, Caillaud C (2010) Effect of an overground walking training on gait performance in healthy 65- to 80-year-olds. Exp Gerontol 45: 427–434. doi: 10.1016/j.exger.2010.03.009
- 14. Mian OS, Thom JM, Ardigo LP, Morse CI, Narici MV, et al. (2007) Effect of a 12-month physical conditioning programme on the metabolic cost of walking in healthy older adults. Eur J Appl Physiol 100: 499–505. doi: 10.1007/s00421-006-0141-9
- 15. Quinn TJ, Manley MJ, Aziz J, Padham JL, MacKenzie AM (2011) Aging and factors related to running economy. Journal of strength and conditioning research/National Strength & Conditioning Association 25: 2971–2979. doi: 10.1519/jsc.0b013e318212dd0e
- 16. Trappe SW, Costill DL, Vukovich MD, Jones J, Melham T (1996) Aging among elite distance runners: a 22-yr longitudinal study. J Appl Physiol 80: 285–290.
- 17. Prince F, Corriveau H, Hebert R, Winter DA (1997) Gait in the elderly. Gait Posture 5: 128–135. doi: 10.1016/s0966-6362(97)01118-1
- 18. Studenski S, Perera S, Patel K, Rosano C, Faulkner K, et al. (2011) Gait speed and survival in older adults. JAMA: the journal of the American Medical Association 305: 50–58. doi: 10.1001/jama.2010.1923
- 19. Himann JE, Cunningham DA, Rechnitzer PA, Paterson DH (1988) Age-related changes in speed of walking. Medicine and science in sports and exercise 20: 161–166. doi: 10.1249/00005768-198804001-00010
- 20. Murray MP, Kory RC, Clarkson BH (1969) Walking patterns in healthy old men. J Gerontol 24: 169–178. doi: 10.1093/geronj/24.2.169
- 21. Wall JC, Charteris J (1981) A kinematic study of long-term habituation to treadmill walking. Ergonomics 24: 531–542. doi: 10.1080/00140138108924874
- 22. Van de Putte M, Hagemeister N, St-Onge N, Parent G, de Guise JA (2006) Habituation to treadmill walking. Biomed Mater Eng 16: 43–52.
- 23. Brockway JM (1987) Derivation of formulae used to calculate energy expenditure in man. Hum Nutr Clin Nutr 41: 463–471.
- 24. Franz JR, Kram R (2012) How does age affect leg muscle activity/coactivity during uphill and downhill walking? Gait Posture 37: 378–384. doi: 10.1016/j.gaitpost.2012.08.004
- 25. Finley JM, Dhaher YY, Perreault EJ (2012) Contributions of feed-forward and feedback strategies at the human ankle during control of unstable loads. Exp Brain Res 217: 53–66. doi: 10.1007/s00221-011-2972-9
- 26. Hakkinen K, Kallinen M, Izquierdo M, Jokelainen K, Lassila H, et al. (1998) Changes in agonist-antagonist EMG, muscle CSA, and force during strength training in middle-aged and older people. J Appl Physiol 84: 1341–1349.
- 27. Monaco V, Ghionzoli A, Micera S (2010) Age-related modifications of muscle synergies and spinal cord activity during locomotion. J Neurophysiol 104: 2092–2102. doi: 10.1152/jn.00525.2009
- 28. Amara CE, Shankland EG, Jubrias SA, Marcinek DJ, Kushmerick MJ, et al. (2007) Mild mitochondrial uncoupling impacts cellular aging in human muscles in vivo. Proc Natl Acad Sci U S A 104: 1057–1062. doi: 10.1073/pnas.0610131104
- 29. Conley KE, Jubrias SA, Amara CE, Marcinek DJ (2007) Mitochondrial dysfunction: impact on exercise performance and cellular aging. Exerc Sport Sci Rev 35: 43–49. doi: 10.1249/jes.0b013e31803e88e9
- 30. Sacchetti M, Lenti M, Di Palumbo AS, De Vito G (2010) Different effect of cadence on cycling efficiency between young and older cyclists. Med Sci Sports Exerc 42: 2128–2133. doi: 10.1249/mss.0b013e3181e05526
- 31. Brisswalter J, Wu SX, Sultana F, Bernard T, Abbiss C (2014) Age difference in efficiency of locomotion and maximal power output in well-trained triathletes. Eur J Appl Physiol: 1–8.
- 32. Peiffer JJ, Abbiss CR, Chapman D, Laursen PB, Parker DL (2008) Physiological characteristics of masters-level cyclists. J Strength Cond Res 22: 1434–1440. doi: 10.1519/jsc.0b013e318181a0d2