Why would one care to know or measure a resting metabolic equivalent or MET? The MET is a practical means of quantifying the energy cost of a particular physical activity or workload and, especially in exercise stress testing, is a means of quantifying the work performed at various stages of exercise intensity.1,2 One MET of energy expenditure was originally defined as the resting level of oxygen uptake of a healthy man weighing 70 kg while sitting, or 3.5 mL O₂ [middle dot] kg^-1 [middle dot] min^-1.3 Therefore, any physical activity can be typically viewed as a multiple of this measure, with threshold requirements identified in individuals who perform unique activities. In this issue of Journal of Cardiopulmonary Rehabilitation and Prevention, however, Savage et al4 raise several issues regarding the determination of the MET, which adds insight into problems that we and others have raised in the past.

ESTIMATED VERSUS MEASURED METS

The evaluation of functional capacity is an important component of the exercise stress test, providing data that have both prognostic and diagnostic implications and are the foundation for developing an exercise prescription. Functional capacity is related to the maximal (or peak) ventilatory oxygen uptake (VO₂max), which represents the greatest amount of oxygen transported and used in cellular metabolism and is the best index of exercise capacity. VO₂max is seldom directly measured in most clinical stress testing laboratories and is generally estimated by linear regression equations, although these equations tend to be population specific and nonlinear and often do not apply to patients with chronic conditions.1-3 We have demonstrated, however, that these formulas grossly overestimate the true exercise capacity, more so in younger patients, and even more so after an exercise training program.3,5 For example, 2 individuals of similar weights and percent body fat may both exercise for 9 minutes on a Bruce protocol treadmill test, and both would have a peak estimated exercise capacity of 10 METs. However, on cardiopulmonary assessment, 1 patient may have a peak oxygen consumption of 17.5 mL O₂ [middle dot] kg^-1 [middle dot] min^-1 (or 5 METs), whereas the other patient has 24.5 mL O₂ [middle dot] kg^-1 [middle dot] min^-1 (or 7 METs). Although the regression equations overestimated the peak VO₂ in both instances, the overestimation was considerably greater for 1 patient than for the other.

FAT ADJUSTED PEAK VO₂

Although peak VO₂ provides a better assessment of exercise capacity than simply estimating METs by treadmill speed and incline, generally when reporting peak VO₂, this parameter is adjusted for total weight as opposed to lean weight, despite the fact that fat is, for practical purposes, aerobically inactive. As Buskirk wrote in a dissertation nearly 50 years ago, "It is concluded that when VO₂ is used to examine the performance of the respiratory-cardiovascular system, the value should be expressed as VO₂ per kg of fat-free weight."6 Nevertheless, these measures are still typically corrected for total as opposed to lean body mass. We have previously demonstrated that the differences noted between men and women and between obese and lean patients are mostly explained by the differences in percent body fat.7-11 When correcting values for lean body mass, the differences between sexes as well as between obese and lean are essentially eliminated in our studies and in the present report by Savage and colleagues.4 This adjustment is most applicable for the evaluation of patients with advanced heart failure because peak VO₂ has been used as a major prognostic discriminator of patients with heart failure and is a major factor in determining listing for heart transplantation. We have demonstrated that correcting peak VO₂ for lean as opposed to total weight, as well as correcting oxygen pulse (peak VO₂/peak heart rate) for body fat, provides better prognostic stratification for patients with advanced heart failure.9,10

THE RESTING MET

In the present evaluation by Savage and colleagues4 in this issue of the journal, the entire concept of the MET is seriously questioned. Although 3.5 mL O₂ [middle dot] kg^-1 [middle dot] min^-1 has been the accepted definition of a MET for many decades, these investigators demonstrate that the average MET in patients with coronary heart disease is considerably lower than this widely accepted value. Although variations in the "true" MET may be partly explained by differences in percent body fat, which raises the idea that the "true" MET should probably also be adjusted for lean as opposed to total body weight, still even this does not totally explain the discrepancy between the measured values in patients with coronary heart disease and the typical 3.5 mL O₂ [middle dot] kg^-1 [middle dot] min^-1 value, which has been dogma for decades. Recognizing that the original population used for defining METs was a young and healthy one free of disease, other factors accounting for these differences would likely include age, physical conditioning, and effects of chronic disease in determining resting metabolic activity.

Considering these points, how should we proceed? The concepts illustrated in the study by Savage and colleagues4 provide important insights for future investigators to consider in studies using exercise cardiopulmonary metabolism. However, at the present time, the current data probably raise more questions than provide answers. For example, should peak exercise capacity be described as a peak VO₂ in mL O₂ [middle dot] kg^-1 [middle dot] min^-1, or should this be corrected for the patients true basal resting VO₂ (which will not be known for most patients), or an average baseline basal resting VO₂? In other words, would peak VO₂ be more important, or would peak VO₂/basal resting VO₂ be superior? Obviously, the present data and studies in the near future will not answer these potentially important questions.

At the present time, when performing cardiopulmonary exercise assessments, we would advise reporting the data in mL O₂ [middle dot] kg^-1 [middle dot] min^-1, preferably correcting for lean as opposed to total weight, as opposed to simply reporting METs. A simple estimation of percent body fat by the sum of the skin-fold method is a fast, reliable, and inexpensive method readily available to determine lean body mass. When using routine treadmill testing without cardiopulmonary assessment, investigators and clinicians must recognize the limitations in this crude assessment and thus report these data as estimated METs (as opposed to VO₂ in mL O₂ [middle dot] kg^-1 [middle dot] min^-1). Nevertheless, numerous investigators have demonstrated the prognostic impact of estimated METs obtained from treadmill speed and incline,12,13 and at present, this "inflated" value should remain a major part of current cardiovascular medicine.

References