How to Accurately Measure Energy Expenditure

Measurement accuracy from metabolic phenotyping systems cannot be assumed. This video shows how the accuracy of energy expenditure measurements is dependent on the flow rate of air through mouse metabolic cages. Only higher air flow rates enable you to make accurate resting and active energy expenditure measurements.

VIDEO NOT LOADING? Viewing videos requires acceptance of all cookies. To change consent settings, open the MANAGE CONSENT tab at the bottom right of this window and click ACCEPT.

In this short video, my goal is to show you the importance of airflow rates when making accurate metabolic phenotyping measurements. In any investigation of energetics, it is valuable not only to measure total energy expenditure, but to quantify how different activities contribute to overall energy budgets.  This story begins with a single mouse who divides her days between eating, wheel running, and generally lounging around her cage.

On the first day of the study, we used a flow rate of 500 milliliters per minute through the cage. We chose this flow rate because other metabolic phenotyping systems currently use this or even lower flow rates for their measurements. In this graph, the orange trace shows the wheel running activity of the mouse. You can see that she sits around for a few hours and then has bouts of running that last between 10 to 15 minutes throughout this 14-hour period. The dark blue trace illustrates her energy expenditure over the same period. You see she has an average metabolic rate of about 0.5 kcals per hour, which is what you might expect for a 40-gram mouse.

Knowing average energy expenditure over a long period is a nice starting point, but most researchers – especially those who plan to publish their results – will need to accurately quantify the energy expenditure associated with different types of activities. From this graph, it’s reasonable to conclude that this mouse’s resting energy expenditure is around 0.4 kcals per hour and that her active energy expenditure is around 0.65 kcals per hour. Unfortunately, these conclusions are far from accurate and the reason for this has to do with the low flow rates within the cage.

Here is the same mouse the next day, but this time we increase the flow rate to 2,000 milliliters per minute. This is four times the previous value. This is the standard flow rate for mice in Promethion phenotyping systems, and just as before, her long lounges are separated by wheel running activity. Again, the dark blue function illustrates her energy expenditure, and just like the day before, her average energy expenditure was 0.5 kcals per hour. But because we used a higher flow rate, this graph provides us with much more information.

During the periods of wheel running activity, we see that the active energy expenditure values were actually around 0.85 kcals per hour. The low flow measurement underestimated this value by over 30 percent. These high flow rates also allow us to better identify when the mouse is rested and allows us to see that the resting energy expenditure is actually 0.3 kcals per hour. The previous graph gave us a 25 percent overestimate. You might ask “why do our competitors continue to use these low flow rates?” Well the short answer is that they need to because they use low resolution gas analyzers that cannot detect the small changes in oxygen and CO2 concentrations that occur at higher flow rates.

Animal ethics committees worldwide recommend that mouse cages receive 15 air exchanges per hour. A flow rate of 500 milliliters per minute in an average 8-liter cage gives the animal less than half of the fresh air that they need, putting the animals under stress and at risk of toxic hypercapnia. Equally important for accurate data is the fact that low flow rates chronically overestimate resting energy expenditure, and they underestimate active energy expenditure.

In the end, I ask you to consider the flow rate you’re using for your metabolic measurements and whether those measurements are showing you what you need to best understand your model system.