PDA

View Full Version : Energy-speed relationship for running



rmiller
11-17-2011, 04:35 PM
Hi everyone,

I was talking with some colleagues lately about a topic that I thought might be of interest to the board, and would love to hear the thoughts / opinions of others.

The metabolic cost ("cost of transport", CoT, the metabolic energy consumed per unit distance traveled) for human running was classically reported by Margaria et al. (1963) as being invariant with respect to the speed of progression. That study had a grand total of two subjects, but it seems to me that there's a general belief that unlike walking, running has no "energetically optimal" speed. The CoT is clearly less sensitive to speed for running than for walking, but later and more recent studies have suggested that the CoT is probably not entirely independent of running speed (e.g. Mayhew, 1977; Steudel-Numbers & Wall-Scheffler, 2009). Sometimes the CoT even appears to decrease slightly with increasing speed (e.g. see recent data from Ferris & Sawicki, 2011). It seems that large amounts of data, perhaps collected over multiple repeated sessions on different days, are needed to detect nonlinearity in the CoT-speed relationship for running (Steudel-Numbers & Wall-Scheffler, 2009).

My question is whether the measurements we take for calculating the CoT are really capturing the full metabolic energy consumption for a movement like running. As speed increases, I would expect that anaerobic metabolism starts to come into play, which the pulmonary gas-exchange-based measurements that we typically take to calculate energy expenditure in locomotion don't account for. It seems like the true CoT-speed relationship for running could be more nonlinear than we give it credit for, if we aren't accurately measuring the energy cost at faster speeds. I'm not aware of any energy-speed studies for running that have addressed this issue.

Is it well known (or even vaguely known) what the proportional contributions of aerobic and anaerobic metabolism are to the total energy expenditure as speed increases? Are there ways of accurately measuring anaerobic energy expenditure that can be reconciled with the typical cart-based measurements for aerobic energy expenditure (i.e. producing the joules of metabolic energy consumed anaerobically)? I found some studies using the "accumulated oxygen deficit" method to quantify the anaerobic contribution. Is this a generally accepted and valid metric for anaerobic energy expenditure?

Ross

bsellers37
11-21-2011, 04:21 AM
The reality is that this sort of experiment is harder to do than it sounds. Respirometry isn't very accurate (see e.g. Webb, P., Annis, J. F. & Troutman, S. J., Jr. 1980 Energy balance in man measured by direct and indirect calorimetry. Am J Clin Nutr 33, 1287-1298.). This is for two reasons. Firstly we never know what the subject is actually respiring and even simultaneously measuring CO2 and water production as well as O2 consumption can't eliminate this uncertainty. Secondly there are always errors in the technique: sealed approaches always have some leaks and open systems always have some mixing problems. This explains why you per trial error is likely to be c. 25% although with sufficient repeats you get a mean error of a few percent. You also have considerable individual variation both in the amount of error from these sources and also probably in the actual cost of locomotion and optimal speed. Finally the difference you are looking for is probably quite small. If it was reasonably big like it is in walking (>x2) then you can pick it up with a moderate sized experiment, but if the speed dependent change in COT is 30% I predict you'd need a really big, well controlled experiment to show it. Someone could certainly do a power analysis to find out just how big! Room-based direct calorimetry would probably help but almost nowhere has facilities for this any more.

Anaerobic + aerobic respiration is actually a problem because all these techniques only really work once steady state has been reached. There will be some anaerobic respiration but it will have saturated in the minutes before you start taking the gas measurements. That's actually one of the difficulties - at high speeds you would need subjects who can maintain 8 m/s for 5 minutes or so! You can measure for shorter periods but then you have to make all sorts of assumptions which may or may not be correct. There are also even more indirect techniques for estimating metabolic energy cost (heart rate, thermography, doppler flow, or based on kinematics) but these make a lot of assumptions too.

Cheers
Bill

--
Dr. Bill Sellers Email: William.Sellers@manchester.ac.uk
Programme Director of Zoology Skype: wisellers
Faculty of Life Sciences Tel. 0161 2751719
The University of Manchester Fax: 0161 2755082
D1239 Michael Smith Building Mob: 0785 7655786
Oxford Road, Manchester, M13 9PT, UK http://www.animalsimulation.org