![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
*
Fred Hutchinson Cancer Research Center, Cancer Prevention Research Program and the
Department of Epidemiology, University of Washington, Seattle, WA 98109-1024.
2To whom correspondence should be addressed. E-mail: mneuhous{at}fhcrc.org
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
KEY WORDS: • indirect calorimetry • energy expenditure • postmenopausal women • dietary intake reporting error
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
,2)
. Random
error in self-reported dietary intake has been recognized for many
years. However, recent studies using objective measures indicate that
energy intake is generally underreported and that underreporting varies
by subject characteristics (3
4
5
6
7)
. These various sources
of error threaten inferences made about dietary exposure and disease
relationships in studies that rely upon self-reported dietary
intake. Therefore, large-scale epidemiologic studies should likely
include objective measures of energy intake and other dietary exposures
for characterization of the error specific to the dietary assessment
instrument being used and to the population under study. Such objective
measures, on appropriate subsamples, can then be used to calibrate the
self-report data and to adjust exposure-disease association
estimates.
Measuring total energy expenditure
(TEE)3
in weight-stable subjects has been accepted as an objective proxy
for energy intake (8)
. Prediction equations offer a quick,
low cost approximation of energy expenditure but require an estimate of
activity-related energy expenditure. In addition, many of these
equations were derived using a narrow age range of Caucasian subjects
and, therefore, may not adequately represent specific population groups
being studied. Doubly labeled water provides an integrated measure of
TEE that includes resting metabolic rate (RMR), all
activity-related energy expenditure and the thermic effect of food
over a 1- to 2-wk period without undue subject burden (9)
.
However, oxygen-18 and isotope ratio mass spectrometry analyses remain
extremely costly for large-scale studies. Additionally, a recent
shortage of oxygen-18 isotope has limited the incorporation of doubly
labeled water techniques in study protocols (10)
.
Another practical and objective measure of TEE can be achieved by
combining measures of RMR with measures of activity-related energy
expenditure and an estimate of the thermic effect of food
(11)
. RMR represents 
60–75% of TEE and is typically
measured using indirect calorimetry (IC), which derives expenditure
estimates from oxygen consumption (VO2) and
carbon dioxide production measures from expired gases
(12)
. The bulk of IC work published addresses energy needs
of metabolic unit or hospitalized patients. However, IC has been used
in small studies to measure RMR in healthy individuals
(13
14
15)
, in larger studies aimed at producing or
assessing prediction equations for RMR (16
17
18
19)
, and in
conjunction with doubly labeled water to estimate activity-related
energy expenditure (20
,21)
. In the above instances, the IC
protocols were quite demanding and constituted significant subject
burden.
The aim of this article is to develop IC protocols that are practical for use in large-scale studies and produce reliable estimates of RMR in free-living, postmenopausal women. Specifically, we address the following: 1) the length of the data collection period; 2) the value of duplicate measures; and 3) the pros and cons of conducting IC until achievement of specific steady-state criteria for minute ventilation (VE), VO2 and respiratory quotient (RQ). Streamlining the IC protocol for use in public health research may translate into substantial resource savings and reduction of subject burden in large-scale studies for which doubly labeled water is not an option.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This project was part of a larger study of biomarkers and dietary self-report among postmenopausal women. Subjects were 102 female volunteers 50–79 y old from the greater Seattle area. Postmenopausal was defined as greater than or equal to 55 y of age or at least 12 mo since last menses. Women were recruited through direct mailings using the Washington State Department of Licensure list, flyers and newspaper advertisements. Those interested in participating contacted us by phone. We provided callers with a description of the study and screened for history of conditions that might interfere with nutrient utilization or RMR measurements (bowel disease, diabetes or hypoglycemia, renal disease, chronic lung disease, liver disease, claustrophobia, weight change in excess of 4.5 kg in the 2 mo before enrollment and alcohol intake > 2 servings per day). The Institutional Review Board of the Fred Hutchinson Cancer Research Center approved all procedures.
Protocol.
Eligible women agreeing to participate were mailed questionnaires and urine collection materials. They were scheduled for two fasting visits 1 wk apart during which IC, height, weight, hip and waist measurements were taken. Each woman completed a 24-h urine collection at home before each visit. RMR was measured using a VMAX 2900 indirect calorimeter and standard manufacturer calibrations were performed (Sensormedics, Loma Linda, CA). One trained technician conducted all IC measurements.
Subjects were instructed to abstain from food and beverages, except water, for a minimum of 8 h, and to avoid strenuous activity for 48 h before each visit. Before IC, subjects rested quietly in a recliner for 30 min in a thermally neutral testing room and were provided an explanation of procedures. The mixing chamber pump was turned on and the plastic canopy was placed over the reclined subject’s head and neck with the vinyl skirt covering the torso. Two minutes of data were allowed to expire before initiating formal data collection to allow for acclimation to the apparatus. Data points were collected every 30 s and steady-state was defined as 10 min during which the volume of oxygen consumed, VE and RQ did not vary >10%. If 10 min of ‘steady state’ was achieved by 30 min of data collection, the test was concluded. If not, the test was continued until 10 min of ‘steady state’ was achieved or at 45 min of data collection, whichever occurred first.
Body composition was estimated using urinary creatinine from duplicate
24-h urine collections. Urinary creatinine concentrations were
determined by a kinetic modification of the Jaffe alkaline picarate
reaction on a Cobas Mira Plus Analyzer (Roche Diagnostics, Brandburg,
NJ). The interassay coefficients of variation for low, medium
and high urine quality control pool levels were 1.2%, 1.6% and 1.6%,
respectively. Daily creatinine excretions were excluded if <780 mg/d
or if urine collections were <23 h or >25 h (22)
. We
excluded values for seven women due to incomplete urine collections.
Fat-free mass (FFM) was calculated using Welle’s formula for older
adults (23)
. We derived fat mass (kg) and % body fat from
FFM (weight - FFM = kg fat; kg fat/weight x 100 = % body fat).
Statistical analyses.
Analytical goals included determining whether there was any advantage to conducting IC for a specified length of time or until achievement of defined stability criteria. RMR for segments had near normal distributions with slight left shifts. We used the natural logarithmic transformation of all RMR to improve normal distribution approximations, followed by paired t test comparisons. Pearson correlation coefficients comparing mean RMR and body composition indices were calculated using the 5- to 10-min and 5- to 30-min IC segments. We first compared various lengths of the IC testing period during any one clinic visit to determine whether longer data collection produced significantly different RMR. We then compared visit 1 time segments with those of visit 2 to ascertain whether significant differences could be detected. To determine whether more stringent stability definitions would produce different RMR estimates, we compared RMR conforming to each of the definitions by visit and as a two-visit mean. To assess bias, we compared mean differences between measured (time segments and segments meeting each stability definition) and predicted RMR from four published equations. Pearson correlation coefficients were calculated between predictive equation RMR and measured RMR for visit 1, visit 2 and a 2-d mean. We excluded RMR measurements from one subject because she was unable to achieve a resting state during the second visit. All analyses were performed using SAS, Version 6.12 (SAS Institute, Cary, NC). Statistical significance of difference was defined as P < 0.05. Values are means ± SD.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
23.1 to <27.3 kg/m2),
20% as overweight (27.3 to <32.2 kg/m2), and
15% as obese (>32.2 kg/m2) using the Surgeon
General’s Report on Nutrition and Health categories (24)
.
FFM in this group was 47.3 ± 4.7 kg, percent body fat was 32.4%
± 10%, and mean body surface area was 1.79 ± 0.18
m2. All subjects met or exceeded the 8-h fast
before IC (8.5–17.0 h; mean: 12.6 ± 1.6 h). RMR recorded
from 5 to 10 min was positively correlated with FFM (r
= 0.62, P < 0.0001), body mass index
(r = 0.67, P < 0.0001), body surface
area (r = 0.83, P < 0.0001) and
percent body fat (r = 0.36, P < 0.0002). Relationships between these measures and the mean RMR recorded
from 5 to 30 min were almost identical (data not shown.)
Table 1
gives percentages of subjects meeting varying degrees of
steady-state criteria.
Stringent steady-state criteria (10 min with VO2, VE and RQ varying <5%) were too difficult for most women to achieve, even after 45 min of data collection. Conversely, nearly all women achieved 5 min with <10% variability in VO2, VE and RQ. In general, more subjects achieved steady-state criteria more quickly and more often at visit 2 than at visit 1. For subjects not achieving 10 min with <10% variability in VO2, VE and RQ by 30 min, we observed a mean increase of 3–6 kcal (13–25 kJ) in RMR when we continued IC until either this criteria was achieved or 45 min elapsed (P = 0.005; data not shown).
|
). Within visit 1, the RMR for 0–5 min and 0–30 min were higher than
segments that excluded the first 5 min. Compared with visit 1, RMR were
lower at visit 2 for the 0–5 (P = 0.001), 5–15
(P = 0.04), and 0- to 30-min (P = 0.05)
segments. RMR calculated from the full length of IC (entire study) were
lower for visit 2 (P = 0.06). RMR correlations between
visits ranged from 0.86 to 0.92 by time segment (data not shown). Based
on these results, we chose the 5- to 10-min time segment of measurement
for additional RMR comparisons because it was the least burdensome and
produced RMR estimates that were not different from values generated
from longer data collection.
|
Table 2
gives mean differences between predicted RMR using equations of Harris
and Benedict (25)
, Mifflin et al. (16)
, World
Health Organization (26)
and Arciero et al.
(17)
and measured RMR. Predicted RMR were 100–200 kcal
(418–836 kJ) higher than measured values using the 5- to 10-min
segment of IC from each visit and a two-visit mean (P
= 0.0001). RMR measured from segments that met the three
achievable definitions of steady-state were 10–30 kcal (42–126
kJ) closer to predicted values. However, differences between measured
and predicted RMR remained (P < 0.02; data not shown).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, suggesting it may not reflect RMR and
should be discarded (14
,27)
.
At the second visit, we observed that more subjects were able to
achieve stability criteria by 30 min and there was a 10- to 20-kcal
(42–84 kJ) reduction in mean RMR compared with the first visit. This
is thought to reflect reduced subject anxiety related to familiarity
with the study protocol. Others working with healthy volunteers
(14)
have reported this adaptation or training effect.
These findings indicate that addressing subject orientation/anxiety is
important when designing IC protocols in free-living groups.
Nonetheless, the average difference in RMR between the two visit
measures was <20 kcal (84 kJ) and correlation coefficients between the
two visits were 0.90, which suggest that doing a second measure is
of limited value. We found that conducting IC to achieve predetermined
steady-state criteria offered few improvements in RMR accuracy,
increased subject burden and would reduce sample size because some
subjects were unable to meet the criteria.
This study offers recommendations for minimizing subject burden in IC protocols without sacrificing measurement quality for inclusion in TEE calculations. However, our study had several limitations. We compared time segments of continuous IC studies instead of conducting studies of various lengths on each subject, which may underestimate differences that occur when the shorter measures are taken. Also, we only assessed RMR measures using different IC study lengths, duplicate measures and three steady-state criteria definitions. We did not explore various lengths of relaxation before measurements. Both the 30-min relaxation period and the IC measurement were conducted while the participant was reclining in a recliner rather than lying flat. This may have increased variability. Additional measurement of subjects not able to achieve 10 min with <10% variability in VE, VO2 and RQ within 30 min resulted in an increase in RMR of only 3–6 kcal (13–25 kJ; P = 0.005), a clinically unimportant difference.
It is interesting to note that our measured RMR tended to be 100–200
kcal (418–836 kJ) less than predicted values from equations despite
attention to standardized protocols and environment. Equations were
selected for breadth of use and for including postmenopausal women in
derivation but were derived by fitting IC data and have inherent
limitations. The equation of Arciero et al. (17)
was based
on data most similar to our sample with 75 women, 50–81 y old, using a
canopy system. However, the women in the sample of Arciero et al.
(17)
were smaller with lower FFM (43.8 ± 4 kg)
compared with the crudely estimated FFM in our sample (47.3 ± 4.7
kg). Their studies were conducted under inpatient conditions, which
should theoretically produce RMR 8% lower than a free-living
protocol. The equation of Harris and Benedict (25)
includes few postmenopausal women (n = 16) and would be
expected to overestimate RMR for our sample. The World Health
Organization (26)
equation is known to overestimate RMR in
North Americans and data on those older than 60 y are limited. The
equation of Mifflin et al. (16)
was developed with a
reasonable number of postmenopausal women (n = 50)
representing a wide weight range (46–120 kg) and likely offers the
best comparison.
These results suggest that for public health protocols including IC to assess RMR as a component of TEE in postmenopausal women, a 10-min canopy study (excluding the first 5 min of data collection) produces reliable results with minimal subject burden. Most free-living postmenopausal women, after following the protocol outlined, were able to achieve 5 min of <10% criteria variation by 45 min, but use of these criteria offered no additional precision and clinically unimportant differences in mean RMR measures. More stringent steady-state criteria were not feasible for 40–60% of the sample. Although our suggestions result in a considerably shorter IC protocol, additional time savings may be feasible by streamlining other aspects of the preparation protocol.
|
FOOTNOTES |
|---|
3 Abbreviations used: FFM, fat-free mass; IC, indirect calorimetry; RMR, resting metabolic rate; RQ, respiratory quotient; TEE, total energy expenditure; VE, minute ventilation; VO2, oxygen consumption.
Manuscript received February 20, 2001. Revision accepted May 28, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1. Livingstone M. B. Assessment of food intakes: are we measuring what people eat?. Br. J. Biomed. Sci. 1995;52:58-67[Medline]
2. Prentice, R. L., Sugar, E., Wang, C. Y., Neuhouser, M. & Patterson, R. (2001) Research strategies and the use of nutrient biomarkers in studies of diet and chronic disease. Public Health Nutr. (in press).
3. Schoeller D. A., Bandini L. G., Dietz W. H. Inaccuracies in self-reported intake identified by comparison with the doubly labeled water method. Can. J. Physiol. Pharmacol. 1990;68:941-949[Medline]
4. Black A. Measurements of total energy expenditure provide insights into the validity of dietary measurements of energy intake. J. Am. Diet. Assoc. 1993;93:572[Medline]
5.
Heitmann B. L., Lissner L. Dietary underreporting by obese individuals—is it specific or nonspecific?. Br. Med. J. 1995;311:986-989
6.
Prentice R. L. Measurement error and results from analytic epidemiology: dietary fat and breast cancer. J. Natl. Cancer Inst. 1996;88:1738-1747
7.
Sawaya A., Tucker K., Tsay R., Willett W., Saltzman E., Dallal G. E., Roberts S. B. Evaluation of four methods for determining energy intake in young and older women: comparison with doubly labeled water measurements of total energy expenditure. Am. J. Clin. Nutr. 1996;63:491-499
8.
Seale J. Energy expenditure measurements in relation to energy requirements. Am. J. Clin. Nutr. 1995;62:1042S-1046S
9. Speakman J. R. The history and theory of the doubly labeled water technique. Am. J. Clin. Nutr. 1998;68:S932-S938[Abstract]
10. North American Association for the Study of Obesity (2000) Report on the supply and demand of 18O-enriched water. Available at: http://www.naaso.org/newsflash/oxygen.htm, 2000.
11. Poehlman E. T., Melby C. L., Badylak S. F. Relation of age and physical exercise status on metabolic rate in younger and older healthy men. J. Gerontol. 1991;46:B54-B58[Medline]
12. Kriketos A. D., Peters J. C., Hill J. O. Cellular and whole-animal energetics. Stipanuk M. H. eds. Biochemical & Physiological Aspects of Human Nutrition 2000:411-424 WB Saunders and Company Philadelphia, PA.
13. McAnena O. J., Harvey L. P., Katzeff H. L., Daly J. M. Indirect calorimetry: comparison of hood and mask systems for measuring resting energy expenditure in healthy volunteers. J. Parenter. Enteral. Nutr. 1986;10:555-557[Abstract]
14. Leff M. L., Hill J. O., Yates A. A., Cotsonis G. A., Heymsfield S. B. Resting metabolic rate: measurement reliability. J. Parenter. Enteral. Nutr. 1987;11:354-359[Abstract]
15.
Warwick P. M., Edmundson H. M., Thomson E. S. Prediction of energy expenditure: simplified FAO/WHO/UNU factorial method vs. continuous respirometry and habitual energy intake. Am. J. Clin. Nutr. 1988;48:1188-1196
16.
Mifflin M. D., St Jeor S. T., Hill L. A., Scott B. J., Daugherty S. A., Koh Y. O. A new predictive equation for resting energy expenditure in healthy individuals. Am. J. Clin. Nutr. 1990;51:241-247
17. Arciero P. J., Goran M. I., Gardner A. M., Ades P. A., Tyzbir R. S., Poehlman E. T. A practical equation to predict resting metabolic rate in older females. J. Am. Geriatr. Soc. 1993;41:389-395[Medline]
18. Heshka S., Feld K., Yang M. U., Allison D. B., Heymsfield S. B. Resting energy expenditure in the obese: a cross-validation and comparison of prediction equations. J. Am. Diet. Assoc. 1993;93:1031-1036[Medline]
19. Garrel D. R., Jobin N., de Jonge L. H. Should we still use the Harris and Benedict equations?. Nutr. Clin. Pract. 1996;11:99-103[Abstract]
20.
Starling R. D., Toth M. J., Carpenter W. H., Matthews D. E., Poehlman E. T. Energy requirements and physical activity in free-living older women and men—a doubly labeled water study. J. Appl. Physiol. 1998;85:1063-1069
21. Westerterp K. R. Alterations in energy balance with exercise. Am. J. Clin. Nutr. 1998;68:970S-974S[Abstract]
22. Forbes G. B. Human Body Composition 1987 Springer-Verlag New York, NY.
23.
Welle S., Thornton C., Totterman S., Forbes G. Utility of creatinine excretion in body-composition studies of healthy men and women older than 60 y. Am. J. Clin. Nutr. 1996;63:151-156
24. U. S. Surgeon General (1988) The Surgeon General’s Report on Nutrition and Health. DHHS (PHS) publication 88-50210. U. S. Department of Health and Human Services, Washington, DC.
25. Harris, J. A. & Benedict, F. G. (1919) A biometric study of basal metabolism in man. Carnegie Institute of Washington publication 279. Carnegie Institution of Washington, Washington, DC.
26. World Health Organization, Food and Agriculture Organization & United Nations University (1985) Energy and Protein Requirements, Series 724. World Health Organization, Geneva, Switzerland.
27. Isbell T. R., Klesges R. C., Meyers A. W., Klesges L. M. Measurement reliability and reactivity using repeated measurements of resting energy expenditure with a face mask, mouthpiece, and ventilated canopy. J. Parenter. Enteral. Nutr. 1991;15:165-168[Abstract]
This article has been cited by other articles:
|
|
 |
|
  D. M. Roffey, N. M. Byrne, and A. P. Hills Day-to-Day Variance in Measurement of Resting Metabolic Rate Using Ventilated-Hood and Mouthpiece & Nose-Clip Indirect Calorimetry Systems JPEN J Parenter Enteral Nutr, September 1, 2006; 30(5): 426 - 432. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. M. Reeves, P. S. W. Davies, J. Bauer, and D. Battistutta Reducing the time period of steady state does not affect the accuracy of energy expenditure measurements by indirect calorimetry J Appl Physiol, July 1, 2004; 97(1): 130 - 134. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. K Horner, R. E Patterson, M. L Neuhouser, J. W Lampe, S. A Beresford, and R. L Prentice Participant characteristics associated with errors in self-reported energy intake from the Women's Health Initiative food-frequency questionnaire Am. J. Clinical Nutrition, October 1, 2002; 76(4): 766 - 773. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
