Doubly labelled water assessment of energy expenditure: principle, practice, and promise
- 3.6k Downloads
The doubly labelled water method for the assessment of energy expenditure was first published in 1955, application in humans started in 1982, and it has become the gold standard for human energy requirement under daily living conditions. The method involves enriching the body water of a subject with heavy hydrogen (2H) and heavy oxygen (18O), and then determining the difference in washout kinetics between both isotopes, being a function of carbon dioxide production. In practice, subjects get a measured amount of doubly labelled water (2H 2 18 O) to increase background enrichment of body water for 18O of 2000 ppm with at least 180 ppm and background enrichment of body water for 2H of 150 ppm with 120 ppm. Subsequently, the difference between the apparent turnover rates of the hydrogen and oxygen of body water is assessed from blood-, saliva-, or urine samples, collected at the start and end of the observation interval of 1–3 weeks. Samples are analyzed for 18O and 2H with isotope ratio mass spectrometry. The doubly labelled water method is the indicated method to measure energy expenditure in any environment, especially with regard to activity energy expenditure, without interference with the behavior of the subjects. Applications include the assessment of energy requirement from total energy expenditure, validation of dietary assessment methods and validation of physical activity assessment methods with doubly labelled water measured energy expenditure as reference, and studies on body mass regulation with energy expenditure as a determinant of energy balance.
KeywordsDoubly labelled water Carbon dioxide production Energy expenditure Physical activity Body composition Energy requirement Energy intake Motion sensors Body mass regulation
Food and Agriculture Organization of the united nations/World Health Organization/United Nations University
Elimination rate of heavy hydrogen
Elimination rate of heavy oxygen
Number of body water molecules
Parts per million
Measurement of whole-body metabolic rate is performed with direct calorimetry, based on the measurement of heat loss, or with indirect calorimetry, based on the measurement of oxygen consumption, carbon dioxide production, and urine-nitrogen loss for energy production from carbohydrate, protein, and fat. Measurement of energy expenditure with doubly labelled water is an innovative variant on indirect calorimetry, where energy expenditure is derived from the measurement of carbon dioxide production.
Key articles in development of the doubly labelled water method for the measurement of energy expenditure
Discovery that the oxygen in respiratory carbon dioxide is in isotopic equilibrium with the oxygen in body water (Lifson et al. 1949)
Description of the measurement of carbon dioxide production from isotopic analysis of body water (Lifson and McClintock 1955)
Doubly labelled water assessment of the energy cost of flying in pigeons (LeFebvre 1964)
Review paper on a model for doubly labelled water assessment of energy expenditure in free-ranging animals (Lifson and McClintock 1966)
Validation of the doubly labelled water method for the measurement of energy expenditure in humans (Schoeller and Van Santen 1982)
Workshop on standardization of the doubly labelled water method for the measurement of energy expenditure in humans (Prentice 1990)
Symposium on advances in the doubly labelled water technique (Speakman and Roberts 1995)
Labelled water, water labelled with 2H and/or 18O, was initially applied to measure body composition. In a two-compartment model for body composition with the component fat-free mass and fat mass, water is an important constituent of fat-free mass (Pace and Rathbun 1945). Fat is free of water and the hydration of fat-free mass among healthy subjects is relatively constant. The hydration of fat-free mass decreases from 80% at birth to 73% at adult age Fomon et al. 1982). Thus, fat-free mass can be calculated from a total body water measurement with 2H and/or 18O dilution after drinking labelled water (Schoeller et al. 1980). Assessment of body composition by measuring total body water with isotope dilution is one of the two available methods, with the measurement of body volume with densitometry, for in vivo body composition measurement (Westerterp 2011). The assumptions on the hydration and the density of body components are derived from carcass analysis. Alternative body composition techniques, including skinfold thickness, body impedance, total body electrical conductivity, dual energy X-ray absorptiometry, and magnetic resonance imaging, are double indirect. The techniques are validated against the two indirect methods including isotope dilution. A recent example is the validation of the measurement of changes in body composition with bio impedance spectroscopy (Ellegård et al. 2016).
The first applications of the doubly labelled water method for the measurement of energy expenditure were restricted to small animals because of the high cost of 18O-enriched water. An early example is the measurement of the energy cost of flying in pigeons (LeFebvre 1964). My first application of the method was measurement of the energy budget of breeding House Martins, a 20-g bird spending most of the day on the wing (Bryant and Westerterp 1980). Advances in analytical devices and reductions in the costs of 18O-enriched water led to the suggestion that the method was useful for application to the field of human energy metabolism (Lifson et al. 1975). Subsequently, the first validation for the measurement of energy expenditure in humans was in 1982 (Schoeller and Van Santen 1982).
Schoeller and Van Santen (1982) administered approximately 10 g of 18O en 5 g 2H to four young adults, one female and three males, after collecting a baseline urine sample for measurement of the background enrichment of body water for the two isotopes. Subsequent urine samples were collected at 6 h after the dose, directly after equilibration of the isotope dose with total body water, and 14 days later at the end of the observation interval. During this interval, energy expenditure was calculated from energy balance by taking the sum of dietary intake and the change in body stores as reference. The energy expenditure from the doubly labelled water method differed from dietary intake plus change in body composition by 2 ± 6%. Thus, it was the first study showing the validity of the doubly labelled water method for the measurement of energy expenditure in unrestricted humans. Later, validation studies, comparing doubly labelled water-assessed energy expenditure with simultaneously measured energy expenditure in a respiration chamber, showed that the method is accurate and has a precision of 2–8% (Schoeller 1988).
In 1988, the method was already applied to humans in eight research centres and all users came together to discuss standardization of the method (Prentice 1990). Aspects included were isotope mass spectrometric analysis, calculation of isotopic pool sizes and flux rates, isotope fractionation corrections, changes in isotopic background, and converting carbon dioxide production to energy expenditure. A similar group presented further advances in the doubly labelled water technique in a workshop at the 78th annual Experimental Biology meeting (Speakman and Roberts 1995). Here, the protocol and analytical requirements for application of the doubly labelled water method were further defined and results were presented on a between laboratory comparison.
Since the mid 1980s, the use of the method has expanded enormously (Speakman 1998). The number of human studies as published in peer-reviewed journals rapidly increased to a steady level around 40 per year from 1995 onwards. It has become the gold standard for human energy requirement under daily living conditions (FAO/WHO/UNU 1985, 2004). Further applications include energy cost of clinical conditions (Van der Kuip et al. 2007), activity costs in relation to body mass (Prentice et al. 1996) and age (Speakman and Westerterp 2010), and under extreme conditions like endurance exercise (Westerterp et al. 1986; Cooper et al. 2011) and at high altitude (Westerterp et al. 1992).
Validation studies were typically performed with simultaneously measurement of carbon dioxide production in a respiration chamber. Thus, the accuracy was shown to be within 2% with a coefficient of variation of 2–12% (Schoeller 1988; Speakman 1998). Validation studies included sedentary conditions as well as measurements at high-activity levels where subjects performed heavy bicycle ergometer work (Westerterp et al. 1988). The results demonstrated the utility of the doubly labelled water method for the determination of carbon dioxide production in the range of activity levels in daily life.
Critical methodological aspects of the doubly labelled water method for the measurement of energy expenditure in humans
1 to 3 weeks
>1.8 g water/kg body water with 10% 18O
>0.12 g water/kg body water with 99% 2H
Background sample before dose
Two independent samples directly after dose equilibration
Two independent samples at the end of the observation interval
Accuracy 0.5 ppm for both isotopes
With regard to the protocol, the optimal observation interval for the measurement of energy expenditure with doubly labelled water is one to three times the biological half-life of the isotopes (Lifson and McClintock 1966). The biological half-life of 18O ranges from 3 days in young children or extremely active adult subjects to about 10 days in very sedentary adult and old subjects (Westerterp 1999). In addition, measuring energy expenditure under daily living conditions with weekly cycles of physical activity routines, one often measures over multiples of a week. Thus, a typical observation interval is 1 week in children and endurance athletes, 2 weeks in adult subjects, and 3 weeks in the elderly.
With regard to the isotope dose, body water volume of the subject mainly determines the isotope dose. Larger subjects require more isotopes, as do men compared to similar mass women, because of their larger body water containing fat-free mass. The ultimate determinant of the dose is the final body water enrichment of the isotopes at the end of the observation interval, in combination with the variation in background enrichment and the analysis precision of body water samples. With the current sample analysis precision with isotope ratio mass spectrometry, subjects are dosed with at least 1.8 g water/kg body water of water with 10% 18O atoms and 0.12 g water/kg body water of water with 99% 2H atoms (International Atomic Energy Agency 2009). Thus, the initial background enrichment of body water for 18O of 2000 ppm is increased with 180 ppm and the initial background enrichment of body water for 2H of 150 ppm is increased with 120 ppm, or slightly higher values. The doubly labelled water method is considered safe for subjects of all ages, with no indications of health effects of the two isotopes at these concentrations (Leatherdale and Jones 1991).
With regard to sample collection, body water enrichment for 18O and 2H is measured in blood-, saliva-, or urine samples. Different body fluids are subject to different degrees of isotope fractionation and thus are not combined in the same protocol (Prentice 1990). Usually urine is chosen, where one should ensure that the urine has not been in the bladder for longer time, preventing accurate timing of the sample. A typical protocol starts with collection of a sample before isotope dose administration, to measure the background body water enrichment. The next sample is collected directly after dose administration and equilibration with the total body water. In healthy young adults, steady-state enrichment of the body water is reached around 2.5 h of ingestion of the isotope (Jankowski et al. 2004). Most studies allow a 4–6 h equilibration time. In larger subjects with a larger body water compartment, where equilibration takes longer, a protocol is adopted with overnight equilibration (Van Marken Lichtenbelt et al. 1994). In the simplest protocol, subsequent isotope elimination is determined by collecting one sample at the end of the observation interval, usually after 1–3 weeks, the two-point method.
Initially, it was preferred to calculate isotope elimination from serial samples, collected on a daily basis, the multipoint method (Cole and Coward 1992). Daily sampling improves the precision of the measurement because the results are calculated from more analyses, reducing the analytical variation by averaging (Schoeller et al. 1995). However, there might be substantial day-to-day differences in isotope elimination rate due to daily variation in energy expenditure. Nowadays, a protocol is indicated with at least two independent samples at the start and at the end of the observation interval. A protocol allowing for differences in equilibration time between subjects and for analytical variation is the ‘Maastricht protocol’ (Westerterp et al. 1995). The isotope dose is administered, after collecting a background sample, as a last consumption before the night. After overnight equilibration, two independent early morning samples are collected at the first day and with weekly intervals of the subsequent observation interval. Samples are preferably stored in glass vials with airtight caps to prevent isotope exchange through perfusion (Westerterp et al. 1995) and isotope fractionation through evaporation (Schoeller et al. 1986a).
Nutrients and resulting energy equivalents of oxygen and carbon dioxide when metabolized (Brouwer 1957)
Carbon dioxide (kJ/l)
The doubly labelled water method is the only method to measure energy expenditure in any environment, especially with regard to activity energy expenditure, where there is no interference with the behaviour of the subjects. Thus, the method is primarily applied for the measurement of the physical activity level of subjects. Additional applications include the assessment of energy requirement from total energy expenditure, validation of dietary assessment methods and validation of physical activity assessment methods with doubly labelled water measured energy expenditure as reference, and studies on body mass regulation with energy expenditure as a determinant of energy balance.
Measuring physical activity level
Validation of dietary- and physical activity assessment methods
The doubly labelled water method has become the gold standard for the assessment of energy requirement (FAO/WHO/UNU 1985, 2004). Assessment of energy expenditure with doubly labelled water has demonstrated that self-report measures of food intake and physical activity are not accurate (Dhurandhar et al. 2015). There is not yet a method for the accurate determination of habitual dietary intake, and thus, energy requirement is derived from measured energy expenditure (Schoeller et al. 1990; Trabulsi and Schoeller 2001). Most studies show a lower value for reported energy intake compared with measured total energy expenditure. Underreporting of habitual intake can be explained by under-recording and under-eating (Goris et al. 2000; Goris and Westerterp 1999). In a clinical setting, the increased awareness of the nursing staff resulted in the opposite, overreporting of intake. Carefully collected energy intake data were inaccurate to such a degree as to be useless (Stallings et al. 1996). Reported intake remains not a suitable marker for energy requirement as evaluated with doubly labelled water (Lopes et al. 2016; Lins et al. 2016). Current development for improvement includes the use of micro-cameras to take pre- and post-meal photographs (Ptomey et al. 2015; Delisle Nyström et al. 2016; Pettitt et al. 2016). So far, doubly labelled water remains the reference for energy requirement, in healthy subjects as well as in patients (Prelack et al. 2017) and in athletes (Morehen et al. 2016).
Doubly labelled water measured energy expenditure not only is the current reference for energy requirement and improvement of dietary assessment methods but also for physical activity assessment methods.
New technology has resulted in a growing number of physical activity monitors to evaluate the activity level of an individual. The validity of activity monitors, as derived from studies using doubly labelled water-assessed activity energy expenditure as a reference, is not always sufficient. In a review on the performance of 11 different monitors, only three could explain more than 50% of the variation in activity-induced energy expenditure or physical activity level between subjects (Westerterp 2014). A more recent review confirmed that there is a large heterogeneity across studies in the performance of accelerometers to estimate activity-induced energy expenditure (Jeran et al. 2016). Most wearable devices do not yet provide a valid estimate of total energy expenditure (Murakami et al. 2016).
Studies on body mass regulation
Doubly labelled water-assessed energy expenditure is a key parameter in studies on body mass regulation, where energy expenditure is a determinant of energy balance. The last decades, the prevalence of overweight and obesity has increased worldwide. Initially, it was suggested that modern inactive lifestyles were the predominant factor in the increasing prevalence of overweight and obesity (Prentice and Jebb 1995). The physical activity level and thus energy needs should have declined faster than energy intake as encouraged by the increasing availability of highly palatable foods. However, analysis of doubly labelled water-assessed physical activity level for trends over time showed that activity-induced energy expenditure has not declined over the same period that obesity rates have increased, and total energy expenditure of modern man is in line with energy expenditure in wild mammals (Westerterp and Speakman 2008). The relation between daily energy expenditure and body mass suggests that increase in energy intake has driven the increase in body mass (Swinburn et al. 2009). Physical activity level, as assessed with doubly labelled water, was shown to promote normal growth and accretion of fat-free mass (Butte et al. 2016). At adult age, an increase in physical activity level results only in modest mass loss or in some cases mass gain (Thomas et al. 2012).
Limitations of the doubly labelled water method for the measurement of energy expenditure are the length of the observation interval and the time-consuming sample analysis procedure. The observation interval ranges between 3 days in young children or extremely active adult subjects and 3 or 4 weeks in very sedentary and old subjects (Schoeller 1983; Westerterp 1999). Over shorter intervals, the change in isotope enrichment is too small to obtain a precise measure of the elimination rate. After longer intervals, the final enrichment cannot be measured precisely, getting too close to background values. Isotope analysis remains time-consuming, despite the development of on alternative for the traditional sample preparation and subsequent measurement of isotope enrichment with isotope ratio mass spectrometry. Immediate measurement of isotope enrichment in biological samples with laser absorption spectrometry can be performed in minutes (Thorsen et al. 2011; Berman et al. 2012). However, memory problems necessitate conditioning of the system by repeated sample injections, increasing the analysis time to 1 h for both isotopes. In addition, one cannot measure the initial samples with high enrichment and the final samples with low enrichment on the same day.
In conclusion, the doubly labelled water method for the measurement of total energy expenditure is a valuable addition to the spectrum of indirect calorimetry methods. It is the indicated method for the measurement of energy expenditure in unrestrained subjects in their normal surroundings. Then, the typical observation interval of one or more weeks can cover the regular cycle of activity routines or effects of exercise interventions on total energy expenditure.
Compliance with ethical standards
Conflict of interest
The author has no conflict of interest.
- Berman ES, Fortson SL, Snaith SP, Gupta M, Baer DS, Chery I, Blanc S, Melanson EL, Thomson PJ, Speakman JR (2012) Direct analysis of ∂2H and ∂18O in natural and enriched human urine laser-based, off-axis integrated cavity output spectroscopy. Anal Chem 84:9768–9773CrossRefPubMedPubMedCentralGoogle Scholar
- Black AE, Prentice AM, Coward WA (1986) Use of food quotients to predict respiratory quotients for the doubly-labelled water method of measuring energy expenditure. Hum Nutr: Clin Nutr 40C:381–391Google Scholar
- Brouwer E (1957) On simple formulae for calculating the heat expenditure and the quantities of carbohydrate and fat oxidized in metabolism of men and animals from gaseous exchange) oxygen intake and carbonic acid output) and urine-N. Acta Physiol Neerl 6:795–802Google Scholar
- Bryant DM, Westerterp KR (1980) The energy budget of the House Martin (Delichon urbica). Ardea 68:91–102Google Scholar
- Delisle Nyström C, Forsum E, Henriksson H, Trolle-Lagerros Y, Larsson C, Maddison R, Timpka T, Löf M (2016) A mobile phone based method to assess energy and food intake in young children: a validation study against the doubly labelled water method and 24 h dietary recalls. Nutrients 8(1):E50. doi: 10.3390/nu8010050 CrossRefPubMedGoogle Scholar
- FAO/WHO/UNU (1985) Energy and protein requirements. WHO Technical Report Series 724. Geneva, SwitzerlandGoogle Scholar
- FAO/WHO/UNU (2004) Human energy requirements. FAO food and nutrition report series 1. Rome, ItalyGoogle Scholar
- Fomon SJ, Haschke F, Ziegler EE, Nelson SE (1982) Body composition of reference children from birth to age 10 years. Am J Clin Nutr 35(Suppl 5):1169–1175Google Scholar
- International Atomic Energy Agency (2009) Assessment of body composition and total energy expenditure in humans using stable isotope techniques. IAEA human health series 3. Vienna, AustriaGoogle Scholar
- Lifson N, Gordon GB, Visscher MB, Nier AO (1949) The fate of utilized molecular oxygen and the source of the oxygen of respiratory carbon dioxide, studied with the aid of heavy oxygen. J Biol Chem A 180:803–811Google Scholar
- Lins IL, Bueno NB, Grotti Clementa AP, Pfrimer K, Sawaya AL, de Menezes Toledo Florencio TM (2016) Energy intake in socially vulnerable women living in Brazil: assessment of the accuracy of two methods of dietary intake recording using doubly labeled water. J Acad Nutr Diet 116:1560–1567CrossRefPubMedGoogle Scholar
- Morehen JC, Bradley WJ, Clarke J, Twist C, Hambly C, Speakman JR, Morton JP, Close GL (2016) The assessment of total energy expenditure during a 14-day ‘In season’ period of professional rugby league players using the doubly labelled water method. Int J Sport Nutr Exerc Metab 26:464–472CrossRefPubMedGoogle Scholar
- Pace N, Rathbun EN (1945) Studies on body composition, III: the body water and chemically combined nitrogen content in relation to fat content. J Biol Chem 158:685–691Google Scholar
- Prentice AM (1990) The doubly labelled water method for measuring energy expenditure, technical recommendations for use in humans. Nahres-4, International Atomic Energy Agency, ViennaGoogle Scholar
- Speakman JR (1997) Doubly labelled water. Springer, BerlinGoogle Scholar
- Thomas DM, Bouchard C, Church T, Slentz C, Kraus WE, Redman LM, Martin CK, Silva AM, Vossen M, Westerterp K, Heymsfield SB (2012) Why do individuals not lose more weight from an exercise intervention at a defined dose? An energy balance analysis. Obes Rev 13:835–847CrossRefPubMedPubMedCentralGoogle Scholar
- Trabulsi J, Schoeller DA (2001) Evaluation of dietary assessment instruments against doubly labeled water, a biomarker of habitual energy intake. Am J Physiol 281:E891–E899Google Scholar
- Westerterp KR (2011) Body composition and its measurement. In: Sobotka L, Allison SP, Forbes A, Ljungqvist O, Meier RF, Pertkiewicz M, Soeters PB (eds) Basics in clinical nutrition, 4th edn. Espen, Galén, Czech RepublicGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.