Effect of bolus fluid intake on energy expenditure values as determined by the doubly labeled water method D. DREWS AND T. P. STEIN Department of Surgery, Robert Wood Johnson Medical School and School of Osteopathic Medicine, University of Medicine and Dentistry of New Jersey, Camden, New Jersey 08103 DREWS, D., AND T. P. STEIN. Effect of bolus fluid intake on energy expenditure values as determined by the doubly labeled zuater method. J. Appl. Physiol. 72(l): 82-86, 1992.-The dou-
bly labeled water (DLW, 2H2180) method is a highly accurate method for measuring energy expenditure (EE). A possible source of error is bolus fluid intake before body water sampling. If there is bolus fluid intake immediately before body water sampling, the saliva may reflect the ingested water disproportionately, because the ingested water may not have had time to mix fully with the body water pool. To ascertain the magnitude of this problem, EE was measured over a S-day period by the DLW method. Six subjects were dosed with 2H,180. After the reference salivas for the two-point determination were obtained, subjects drank water (700-1,000 ml), and serial saliva samples were collected for the next 3 h. Expressing the postbolus saliva enrichments as a percentage of the prebolus value, we found I) a minimum in the saliva isotopic enrichments were reached at -30 min with the minimum for 2H (95.48 t 0.43%) being significantly lower than the minimum for 180 (97.55 & 0.44, P < 0.05) and 2) EE values calculated using the postbolus
because the ingested water may not have had time to mix fully with the body water pool. The result would be anomalously low isotopic enrichments in the saliva. This could lead to serious errors in the calculated energy expenditure (8, 15). The objectives of this series of experiments were to determine whether bolus fluid intake before body water sampling was a potential source of error in the DLW method using saliva as the sampling medium. METHODS
Informed consent for this study was obtained in accordance with the policies of the University of Medicine and Dentistry of New Jersey-School of Osteopathic Medicine Committee for the protection of human subjects. The experiment lasted 8 days. Energy expenditure was measured between days 3 and 8 for five subjects using the two-point method. The protocol is diagrammed in Table isotopic enrichments are appreciably higher (19.9 k 7.5%) than the prebolus reference values. In conclusion, it is not advisable 1. Beginning at t = 0 on day 1, subjects were not allowed to collect saliva samples for DLW measurements within -1 h to eat or drink (npo) for the next 6 h. After 2 h, a backof bolus fluid intake. ground saliva was collected. Subjects l-3 and 5 then drank ‘H,180 (batch 1, 165 ml, 3.88% 2H, ICONS, Sumbody water pool; saliva mit, NJ and 9.45% 180, Isotec, Miamisburg, OH). Subject 4 was given 150 ml of 2H2180 (batch 2, 9.34% 180 and 2.84% 2H). Salivas were collected 3 and 4 h post-isotope THEDOUBLYLABELEDWATER (DLW)methodformeaingestion for determination of the total body water. Subsuring human energy expenditure has been validated in ject 6 was given 192 ml of water from batch 2, with saliva the controlled laboratory environment by several inde- being collected for the total body water determination on pendent groups (2,13,14,18). It is generally agreed that day 1, the bolus given 17 h later on day 2, and the final the precision of the method is >5% (1, 3, 17). However, saliva collected 120 h after administration of the bolus. the controlled conditions of the research laboratory are For days 3-8, subjects l-5 recorded their dietary inoften difficult to reproduce for field studies. In field stud- take. Macronutrient intakes were calculated using the ies the reported precision may not be as good (5, 1518). Campbell Master Nutrient Data Base, which is based on For example, Westerterp et al. (18) measured energy ex- the Michigan State University Numerical Data Base and penditure during exercise in the laboratory and again includes additional material from the US Department of during the “Tour de France.” For the laboratory study, Agriculture and the Campbell Soup Co. (Campbell Soup, the agreement between intake balance and the DLW Camden, NJ). Sample collection was resumed on day 3 method was >5%. In contrast, for the Tour de France with the subjects being npo for 2 h before collecting a study, the DLW values were sometimes higher by as saliva sample. This saliva sample was defined as the “refmuch as 30% (18). It is therefore important to determine erence” saliva for day 3. The subjects then voided and what the potential error error are for field studies so that drank either 700 ml (subject 5) or 1,000 ml (subjects l-4 they can be avoided in future studies. and 6) of water over a 5-min period. Serial saliva samples We have previously suggested that bolus fluid intake were then collected 20,40,60,120, and 180 min after the in the period immediately before body water sampling is subjects drank the water. Five days later, day 8 (4 days one such potential source of error (15). If there is bolus for subject 6), a terminal reference saliva was collected fluid inta .ke immediately before body water sampling, sa- after the subject had been npo for 2 h. During the 3 h liva may refleet the ingested water disproportionately, after the bolus subjects were not allowed to eat or drink.
82
0161~7567/92
$2.00 Copyright
0 1992 the American
Physiological
Society
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 14, 2019.
EFFECT TABLE
1. Experimental
1
0 2.0
1 1
2.0 5.0 6.0 0.0 2.0 2.0 2.33 2.67 3.0 4.0 5.0 0 2.0
1
1 3 3 3 3 3 3 3 3 8 8
FLUID
83
INTAKE
rC0, = 0.481 N(l.O1 kO - 1.04 kH)
protocol
- 0.0258 N(kO - kH)
Elapsed Time, h
Day
OF BOLUS
Task
Start npo Collect background saliva Take 2H,180 Collect saliva for TBW Collect saliva for TBW Start npo Collect (reference) saliva Void, drink water (700~1,000 ml) Collect saliva Collect saliva Collect saliva Collect saliva Void, collect saliva Start npo Collect (reference) saliva
Day I salivas collected at 5 and 6 h were used for total body water (TBW) determinations (Table 2). Day 3 and 8 reference salivas were used for calculating energy expenditure rate for subjects 1-5. For subject 6, bolus was given on day 2 and final saliva taken on day 6.
(3)
where rC0, is the rate of CO, production in mol/day and N is the total body water in moles. The Weir equation (Eq. 4) was used to convert the rate of CO, production into energy expenditure values. The equation is based on the facts that the caloric equivalent of 1 liter of oxygen and the amount of CO, produced differs for the three major foodstuffs, protein, fat, and carbohydrate, and that the amount of protein oxidized can be determined from the urinary nitrogen excretion (16). VU, is the rate of 0, utilization, VCO, is the rate of CO, production in liters/day, EE is the energy expended in kcal/day, and U is the nitrogen excretion rate. For an estimate of the nitrogen excretion rate (in g nitrogen/ day) we used the nitrogen intake. The respiratory quotient (RQ) was estimated from the &day dietary records EE = 3.941b2 rC0,
They were allowed to walk around, but for the most part they were sedentary.
+ 1.106 i7c0, - 2.17U
= %0,/22.4
(4) (5)
RQ = Vco,/V0,
(6)
RESULTS ANALYSIS
The saliva samples were stored frozen in sealed containers until they were vacuum distilled as previously described (15). The resultant water from subjects l-5 was analyzed for I80 and 2H enrichment by Global Geochemistry (Canoga Park, CA). The saliva from subject 6 was not distilled and was analyzed for I80 and 2H enrichment by Metabolic Solutions (Acton, MA). CALCULATIONS
Isotope distribution spaces and the total body water. The 180 and 2H isotope distribution spaces (IDS, in grams) were calculated from Eqs. 1 and 2
IDS = (d APE,. 18.02. f)/(MW,
delta,) (1) where d is the dose given in grams, APE, is its enrichment in atoms percent excess relative to standard mean ocean water, MWd is the molecular weight of the dose water, Delta, is the enrichment of the saliva relative to the predose enrichment expressed in delta units, and f is the fractionation effect (1.047 for CO, and 1.0 for 2H). The total body water (TBW, in grams) was assumed to be equal to the “0 isotope distribution space divided by 1.01 and the 2H isotope distribution space divided by 1.04 (9). Energy expenditure. The rate of isotope loss from the body was calculated from Eq. 2 l
.
100
.0.00208
l
The results are summarized in Tables 2-4 and Figs. 1 and 2, and those data are means t SE. Statistics were by paired t tests (Fig. 1) or by a repeated measures design/ gnalysis of variance (Fig. 2). Table 2 gives the actual isotopic enrichments found. Subject characteristics and the body water values for the five subjects are given in Table 3. Table 4 gives the mean dietary intakes for days 3-8 and the energy expenditure values calculated by the twopoint method using the day 3 (reference) and day 8 points (days 2 and 6 for subject 6). For subjects 1, 2, and 5, reasonable agreement between intake and expenditure was found. Subjects 3 and 4 were in negative energy balance for the study period. Subject 3 was attempting to lose weight by restricting intake and exercising during
-*
102
T
‘80
mm0 2H
k, or k, = (In delta, - In delta&?
(2) where delta, and delta, are the differences in isotopic enrichments of the sample and the predosing background for 180 or 2H on days 3 and 8, respectively, k, and 12,arethefractionalH2180 and2H20turnoverrates,and t is time. The rate of CO, production was calculated from the equation of Schoeller et al. (Eq. 3 in Ref. 12) Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 14, 2019.
EFFECT
OF BOLUS
FLUID
INTAKE
are the average of the 20- and 40-min data (“30”- min data). Using the minima 180 and ‘H enrichments (Fig. 1) the energy expenditure values are 19.9 t 7.5% greater than those found using the two reference data points (P < 0.05). Using the mean of the 120- and 180-min values gives a value of 9.8 t 6.4%, which is significantly less than the 30-min point (P < 0.05) but not significantly different from the reference point (P = 0.195, Fig. 2). DISCUSSION
REFti%iNCE
30Yn TlUF .
I...-
t5U min IminI \*
I
..I
./
2. Effect of bolus fluid intake on calculated enerm expenditure. Data are expressed as percentage of reference value. Mean of ZOand 40-min values (valley) is significantly greater than prebolus (reference) values or mean of 12O- and 18O-min values (*P < 0.05). FIG.
the study period; the data show she was successful in reducing intake below energy expenditure. Figure 1 shows the percent change in the saliva enrichments with time after the subjects took the bolus. The postbolus saliva enrichments have been expressed as a percentage of the prebolus (reference) value. Both 180 and 2H show a significant dip with a minimum at -30 min after the subjects drank the water, but the ‘dip” (mean of the 20- and 40-min data points) with 2H is significantly greater than that for 180 (97.55 t 0.44% for I80 vs. 95.48 t 0.43% for 2H, P < 0.05). By ~1 h after dosing a new plateau has been approached for both isotopes. By using the mean of 120- and 180-min points to estimate the plateau, the values were 97.30 t 0.41% and 96.51 t 0.62% of th e predose level for 180 and 2H, respectively (NS) . Figure 2 shows the calculated energy expenditure values using 1) the two reference points, 2) the minima in the isotopic enrichments and the day 8 reference point and 3) the plateau and the day 8 reference point. Figure 1 shows that the minima in isotopic enrichment occurred at -30 min postbolus. Therefore the data used for Fig. 2 TABLE
Selection of experimental conditions. We selected saliva over urine or --blood- as the body fluid to sample for practical reasons. Blood is an invasive way of sampling body water, and serial blood sampling is not always acceptable to test subjects (6). Furthermore, the collection of saliva is often more socially acceptable than collection of urine for subjects outside a clinical setting. A disadvantage of saliva sampling is that there is a small fractionation effect for deuterium (4, 15). If the objective is to compare similar groups run under a common protocol rather than to obtain absolute values, a small consistent error is not a serious problem. Another disadvantage of saliva is the small sample obtained, and this can preclude repeat analyses. The DLW methodology appears deceptively simple. However it has not proven to be a simple matter to evaluate the relative importance of overlapping complicating factors (10, 11). These factors include fractionation effects, variations in enrichment of exogenous water intake during the study period, changes in body composition during the study period, and the differences between the measured 180 and 2H spaces (7,9). For the present study we wanted to avoid the problems associated with the 2H space and equilibration rates being different from the 180 space. Two days were therefore allowed for water equilibration. By allowing two days we could be sure that any residual exchange of 2H with body protein was minimal (6). And basing the EE calculation on the 180 space by using the 1986 equation of Schoeller et al. (12) avoided any potential problems with using the 2H space.
2. Summary of experimental data Subj 1
Subj 2
Subj 3
Subj 4
Subj 5
Subj 6
180
2H
i8u
2H
‘W
2H
180
2H
180
2H
180
2H
0.4 221.4 220.7
-16 1,190 1,185
-3.2 175.0 174.3
-35 664 685
-1.5 176.1 179.1
-32 912 921
-1.0 161.5 161.1
-18 820 854
-3.0 192.4 190.8
-33 864 862
225.7 225.1
1,103 1,136
173.9
1,001
145.7
799
707
130.0 130.8 129.8 131.0 128.0
753 719 711 738 714 719
168.1
127.7 127.7 126.3 128.7 128.8
719 703 683 716 710
135.3
144.6 144.5 143.8 143.4 144.3
602 580 551 578 586 566
132.3
169.6 170.9
163.3 166.4 165.1 165.4 163.5
766 772 767 773 775
213.1 207.2 207.2 207.6 206.7 207.5
1,071 1,036 1,031 1,041 1,043 1,055
86.1
399
64.6
418
82.8
536
109.8
581
142.7
802
Day 1
Pre 3h 4h Day 3 Oh 0.33 h 0.67 h 1.0 h 2.0 h 3.0 h
168.8 168.7 168.7
944 955 961 956 943
Day 8 120 h
102.7
666
Isotopic enrichments are given in delta units (103) relative to standard mean ocean water. Salivas were collected 3 and 4 h post-isotope ingestion for determination of total body water. Subject 4 was given 150 ml of 2Hz18O (batch 2,9.34% la0 and 2.84% 2H). Time schedule for subject 6 was a little different; bolus was given 17 h postdosing and final saliva collected 120 h postdosing. Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 14, 2019.
EFFECT
OF BOLUS
Figure 1 shows that the bolus fluid intake had a large and disproportionate effect on the saliva water 180 and 2H enrichments. At 20 and 40 min, the increase in 2H dilution is significantly greater than the dilution in the I80 dilution. Using the “valley” data overestimates the calculated EE by -20% (Fig. 2). We suggest that the valley in the isotopic enrichment of the saliva occurs because the ingested water first enters the plasma compartment from the gut and then is distributed more slowly into the various extravascular pools such as saliva. By - 1 h postbolus, the system approaches a new equilibrium, and the energy values decrease toward the reference values. The following explanation is proposed for the differential effect between ‘$0 and 2H. The saliva isotopic enrichments reflect the vascular pool because saliva is derived from the vascular pool. After a large bolus of water has been ingested, two processes occur: I) water from the gut enters the vascular compartment, and 2) there is exchange of H+, OH-, and H,O between the gut water and the vascular pool. The first process will lead to an increase in the vascular pool size if the rate at which water enters the vascular compartment is greater than that at which it leaves the vascular pool for urine or the extravascular pool. Because the bolus was not enriched, the plasma compartment enrichments will be less than that of the extravascular pool. The second process, differential exchange rates of H+ and OH- can account for the greater decrease in 2H enrichment relative to 180 in the plasma compartment. If the rate of H entry from the gut into the vascular pool is faster than that for 0, the transient dip in the isotopic enrichments in the plasma water pool, and hence in the saliva, will be greater for H than 0. Because H+ has a greater ionic mobility than OH-, it is not unreasonable to argue that (unlabeled) H+ will move into the vascular compartment more rapidly than (unlabeled) OH- (1). An alternate explanation would be that the effect is specific to saliva, being due to incomplete equilibration in the mouth after the bolus. Such an explanation is possible. Carefully collected urines were avaliable for subject 6. Energy expenditure values were 2,890,2,793, and 2,708 kcal/day for the reference, 30-min, and 150-min times, 3. Subject characteristics water values
and total body
TABLE
Subj No. I
49%yr Ht, cm w kg TBW (18U, 3 h) TBW (180, 4 h) TBW (‘H, 3 h) TBW (‘H, 4 h) TBW,
34 1’76 54
2
35 168 69.4
3
26 176 66.7
4
28 173 79.3
5
6
34 64 60.5
50 178 78
34.83
38.82
43.35
47.37
39.40
40.07
34.95
38.87
42.63
47.49
39.72
40.17
34.03
40.91
43.47
48.97
45.75
44.02
34.17
39.66
43.06
47.06
45.55
43.63
FLUID TABLE
85
INTAKE
4. Dietary
expenditure
intake data and energy
values Subj No.
Dietary intake Carbohydrate, % Fat, % Protein, % Intake, kcal/day EE-1 by DLW, kcal/day
I
2
3
4
5
49 35 16 2,270 2,235
52 32 16 2,260 2,475
56 32 12 1,875 4,196
47 39 14 2,804 3,807
55 31 14 2,133 2,255
Intake is mean daily energy intake calculated from 5 days of intake records. EE-1, energy expenditure value calculated by doubly labeled water (DLW) method using the &zy 3 and 8 reference saliva values. Energy intake was not determined for subject 6. Energy expenditure was 2,387 kcal/day.
respectively. The corresponding saliva values were 2,947, 3,195, and 2,807 kcal/day. However, urine data do not necessarily have to reflect in the time period immediately after the bolus. It depends on when the urine enters the bladder after the bolus. If the bolus was given at t = 0, increased urine production may not likely start instantaneously and proceed at a uniform rate until the bolus is voided over l-2 h. More likely, the rate of urine production will increase to a maximum and then return to baseline during that period. Only if the rate of urine production was uniform or most of the bolus voided in the first -40 min would an effect be seen. If most of the urine is produced toward the end of the 1st h no or minimal effect will be seen because the effect is largely over by then. If the mean saliva valley values for the 180 and 2H enrichments (mean of the 120- and 180-min points) is used in conjunction with the day 8 reference value, the resultant EE values are 21% higher than the reference EE values (Fig. 2). Using the mean of the 120- and MOmin (150,min) values as estimates of the plateau gives a lower value, -lo%, which is not significantly lower than the value obtained at 30 min. The amount of fluid taken in by our subjects may have been somewhat large. Had the dose been halved the error would still be substantial. From our data it can be estimated that a 300-ml bolus (a can of soda) can result in an overestimating EE by -8% if saliva sampling is at the minima. The point is that while there remain several uncertainties in such complex parameters with the method such as which fractionation constant to use and how to handle the difference between the 2H and ‘$0 spaces, neglect of a simple experimental precaution like avoiding bolus fluid intake during the hour before sampling can lead to significant and avoidable errors in the calculated energy expenditure values. This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant DK-35612. Address for reprint requests: T. P. Stein, Dept. of Surgery, School of Osteopathic Medicine, University of Medicine and Dentistry of New Jersey, 401 Haddon Ave., Camden, NJ 08103. Received 25 June 1990; accepted in final form 19 June 1991.
total body water.
REFERENCES P. W. Physical 1978, p. 819-848.
1. ATKINS,
Chemistry.
San Francisco, CA: Freeman,
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 14, 2019.
86
EFFECT
OF BOLUS
2. COWARD,W. A., ANDA. M. PRENTICE.Isotope method for the measurement of carbon dioxide production rate in man. Am. J. Clin. Nutr. 41: 659-661,1985. 3. DELANY,J. P., D. A. SCHUELLER,R. W. HOYT, E. W. ASKEW, AND M. A. SHARP.Field use of D,% to measure energy expenditure of soldiers at different energy intakes. J. AppC Physiol. 67: 1922-1929, 1989. 4. HALLIDAY, D., AND A. G. MILLER. Precise measurement of total body water using trace quantities of deuterium oxide. Biomed. Mass. Spec. 4: 82-87, 1977. 5. HOYT, R. W., T. P. STEIN, H. R. LIEBERMAN,T. E. MORGAN,E. J. IWONYK,E. W. ASKEW,AND A. CYMERMAN.Doubly labeled water (DLW) method accurately estimates energy expenditure during field operation. J. Appl. Physiol. 71: 16-22, 1991. 6. HUSS-ASHMORE,R., J. L. GOODMAN,T. E. SIBNA, AND T. P. STEIN. Energy expenditure of young Swazi women as measured by the doubly labeled water method. Eur. J. Clin. Nutr. 43: 739-748, 1989. 7. NAGY,K. A. CO, production in animals: analysis of potential errors in the doubly labeled water method. Am. J* Physiol. 238 (Regulatory Integrative Comp. Physiol. 7): R466-R473, 1980. 8. NOVICK, W. M., M. NUSBAUM,AND T. P. STEIN. The energy costs of surgery as measured by the doubly labeled water (‘H,“O) method. Surgery 103: 99-105,1987. 9. SCHOELLER,D. A. Energy expenditure from doubly labeled water: some fundamental considerations in humans. Am. J. Clin. Nutr. 38: 999-1005,1984. 10. SCHOELLER,D.A.,W.H. DIETZ, E. VAN~ANTEN,AND P.D. KLEIN. Validation of saliva sampling for total body water determination by Hs180 dilution. Am. J. Clin. Nutr. 35: 591-594, 1982.
FLUID 11.
12.
13. 14.
15 .
“* 17,
18.
INTAKE
SCHOELLER,D. A., C. A. LEITCH, AND C. BROWN. Doubly labeled water method: in vivo oxygen and hydrogen isotope fractionation. Am. J. Physiol. 250 (Regulatory Integrative Camp. Physiol. 19): R823-R830,1986. SCHOELLER,D. A.,E. RAVUSSIN,Y. SCHUTZ, K. J. ACHESON,P. BAERTSCHI,AND E. JEQUIER. Energy expenditure by doubly labeled water: validation in humans and proposed calculation. Am. J. Physiol. 250 (Regulatory Integrative Comp. Physiol. 19): R823R830,1986. SCHOELLER,D. A., AND E. VAN SANTEN. Measurement of energy expenditure in humans by doubly labeled water method. J. Appl. Physiol. 63: 955-959,1982. SEALE,J., W. V. RUMPLER, AND J. M. CONWAY.Comparison of energy estimation methods using double labeled water, intake balance and direct/indirect calorimetry in man. Am. J. Clin. Nutr. 52: 66-71,199O. STEIN, T. P., R. W. HOYT, M. O'TOOLE, R. G. SETTLE, AND W. D. B. HILLER. Determination of energy expenditure during heavy exercise, normal daily activity and sleep using the doubly labeled water method. Am. J. CZin. Nutr. 45: 534~539,1986. WEIR, J. B. DEV. New methods for calculating metabolic rate with special reference to protein metabolism. J. Physiol. Land. 109: l-9, 1949. WESTERTERP,K, R., F. BROUNS,W. H. M. SARIS, AND F. TEN HOOR. Comparison of doubly labeled water with respirometry at low-and high-activity levels. J. Appl. Physiol. 65: 53-56, 1988. WESTERTERP,K. R., W. H. M. SARIS, M. VAN Es, AND F. TEN HOOR.Use of the doubly labeled water technique in humans during heavy sustained exercise. J. Appl. Physiol. 61: 2162-2167, 1986.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 14, 2019.
bly labeled water (DLW, 2H2180) method is a highly accurate method for measuring energy expenditure (EE). A possible source of error is bolus fluid intake before body water sampling. If there is bolus fluid intake immediately before body water sampling, the saliva may reflect the ingested water disproportionately, because the ingested water may not have had time to mix fully with the body water pool. To ascertain the magnitude of this problem, EE was measured over a S-day period by the DLW method. Six subjects were dosed with 2H,180. After the reference salivas for the two-point determination were obtained, subjects drank water (700-1,000 ml), and serial saliva samples were collected for the next 3 h. Expressing the postbolus saliva enrichments as a percentage of the prebolus value, we found I) a minimum in the saliva isotopic enrichments were reached at -30 min with the minimum for 2H (95.48 t 0.43%) being significantly lower than the minimum for 180 (97.55 & 0.44, P < 0.05) and 2) EE values calculated using the postbolus
because the ingested water may not have had time to mix fully with the body water pool. The result would be anomalously low isotopic enrichments in the saliva. This could lead to serious errors in the calculated energy expenditure (8, 15). The objectives of this series of experiments were to determine whether bolus fluid intake before body water sampling was a potential source of error in the DLW method using saliva as the sampling medium. METHODS
Informed consent for this study was obtained in accordance with the policies of the University of Medicine and Dentistry of New Jersey-School of Osteopathic Medicine Committee for the protection of human subjects. The experiment lasted 8 days. Energy expenditure was measured between days 3 and 8 for five subjects using the two-point method. The protocol is diagrammed in Table isotopic enrichments are appreciably higher (19.9 k 7.5%) than the prebolus reference values. In conclusion, it is not advisable 1. Beginning at t = 0 on day 1, subjects were not allowed to collect saliva samples for DLW measurements within -1 h to eat or drink (npo) for the next 6 h. After 2 h, a backof bolus fluid intake. ground saliva was collected. Subjects l-3 and 5 then drank ‘H,180 (batch 1, 165 ml, 3.88% 2H, ICONS, Sumbody water pool; saliva mit, NJ and 9.45% 180, Isotec, Miamisburg, OH). Subject 4 was given 150 ml of 2H2180 (batch 2, 9.34% 180 and 2.84% 2H). Salivas were collected 3 and 4 h post-isotope THEDOUBLYLABELEDWATER (DLW)methodformeaingestion for determination of the total body water. Subsuring human energy expenditure has been validated in ject 6 was given 192 ml of water from batch 2, with saliva the controlled laboratory environment by several inde- being collected for the total body water determination on pendent groups (2,13,14,18). It is generally agreed that day 1, the bolus given 17 h later on day 2, and the final the precision of the method is >5% (1, 3, 17). However, saliva collected 120 h after administration of the bolus. the controlled conditions of the research laboratory are For days 3-8, subjects l-5 recorded their dietary inoften difficult to reproduce for field studies. In field stud- take. Macronutrient intakes were calculated using the ies the reported precision may not be as good (5, 1518). Campbell Master Nutrient Data Base, which is based on For example, Westerterp et al. (18) measured energy ex- the Michigan State University Numerical Data Base and penditure during exercise in the laboratory and again includes additional material from the US Department of during the “Tour de France.” For the laboratory study, Agriculture and the Campbell Soup Co. (Campbell Soup, the agreement between intake balance and the DLW Camden, NJ). Sample collection was resumed on day 3 method was >5%. In contrast, for the Tour de France with the subjects being npo for 2 h before collecting a study, the DLW values were sometimes higher by as saliva sample. This saliva sample was defined as the “refmuch as 30% (18). It is therefore important to determine erence” saliva for day 3. The subjects then voided and what the potential error error are for field studies so that drank either 700 ml (subject 5) or 1,000 ml (subjects l-4 they can be avoided in future studies. and 6) of water over a 5-min period. Serial saliva samples We have previously suggested that bolus fluid intake were then collected 20,40,60,120, and 180 min after the in the period immediately before body water sampling is subjects drank the water. Five days later, day 8 (4 days one such potential source of error (15). If there is bolus for subject 6), a terminal reference saliva was collected fluid inta .ke immediately before body water sampling, sa- after the subject had been npo for 2 h. During the 3 h liva may refleet the ingested water disproportionately, after the bolus subjects were not allowed to eat or drink.
82
0161~7567/92
$2.00 Copyright
0 1992 the American
Physiological
Society
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 14, 2019.
EFFECT TABLE
1. Experimental
1
0 2.0
1 1
2.0 5.0 6.0 0.0 2.0 2.0 2.33 2.67 3.0 4.0 5.0 0 2.0
1
1 3 3 3 3 3 3 3 3 8 8
FLUID
83
INTAKE
rC0, = 0.481 N(l.O1 kO - 1.04 kH)
protocol
- 0.0258 N(kO - kH)
Elapsed Time, h
Day
OF BOLUS
Task
Start npo Collect background saliva Take 2H,180 Collect saliva for TBW Collect saliva for TBW Start npo Collect (reference) saliva Void, drink water (700~1,000 ml) Collect saliva Collect saliva Collect saliva Collect saliva Void, collect saliva Start npo Collect (reference) saliva
Day I salivas collected at 5 and 6 h were used for total body water (TBW) determinations (Table 2). Day 3 and 8 reference salivas were used for calculating energy expenditure rate for subjects 1-5. For subject 6, bolus was given on day 2 and final saliva taken on day 6.
(3)
where rC0, is the rate of CO, production in mol/day and N is the total body water in moles. The Weir equation (Eq. 4) was used to convert the rate of CO, production into energy expenditure values. The equation is based on the facts that the caloric equivalent of 1 liter of oxygen and the amount of CO, produced differs for the three major foodstuffs, protein, fat, and carbohydrate, and that the amount of protein oxidized can be determined from the urinary nitrogen excretion (16). VU, is the rate of 0, utilization, VCO, is the rate of CO, production in liters/day, EE is the energy expended in kcal/day, and U is the nitrogen excretion rate. For an estimate of the nitrogen excretion rate (in g nitrogen/ day) we used the nitrogen intake. The respiratory quotient (RQ) was estimated from the &day dietary records EE = 3.941b2 rC0,
They were allowed to walk around, but for the most part they were sedentary.
+ 1.106 i7c0, - 2.17U
= %0,/22.4
(4) (5)
RQ = Vco,/V0,
(6)
RESULTS ANALYSIS
The saliva samples were stored frozen in sealed containers until they were vacuum distilled as previously described (15). The resultant water from subjects l-5 was analyzed for I80 and 2H enrichment by Global Geochemistry (Canoga Park, CA). The saliva from subject 6 was not distilled and was analyzed for I80 and 2H enrichment by Metabolic Solutions (Acton, MA). CALCULATIONS
Isotope distribution spaces and the total body water. The 180 and 2H isotope distribution spaces (IDS, in grams) were calculated from Eqs. 1 and 2
IDS = (d APE,. 18.02. f)/(MW,
delta,) (1) where d is the dose given in grams, APE, is its enrichment in atoms percent excess relative to standard mean ocean water, MWd is the molecular weight of the dose water, Delta, is the enrichment of the saliva relative to the predose enrichment expressed in delta units, and f is the fractionation effect (1.047 for CO, and 1.0 for 2H). The total body water (TBW, in grams) was assumed to be equal to the “0 isotope distribution space divided by 1.01 and the 2H isotope distribution space divided by 1.04 (9). Energy expenditure. The rate of isotope loss from the body was calculated from Eq. 2 l
.
100
.0.00208
l
The results are summarized in Tables 2-4 and Figs. 1 and 2, and those data are means t SE. Statistics were by paired t tests (Fig. 1) or by a repeated measures design/ gnalysis of variance (Fig. 2). Table 2 gives the actual isotopic enrichments found. Subject characteristics and the body water values for the five subjects are given in Table 3. Table 4 gives the mean dietary intakes for days 3-8 and the energy expenditure values calculated by the twopoint method using the day 3 (reference) and day 8 points (days 2 and 6 for subject 6). For subjects 1, 2, and 5, reasonable agreement between intake and expenditure was found. Subjects 3 and 4 were in negative energy balance for the study period. Subject 3 was attempting to lose weight by restricting intake and exercising during
-*
102
T
‘80
mm0 2H
k, or k, = (In delta, - In delta&?
(2) where delta, and delta, are the differences in isotopic enrichments of the sample and the predosing background for 180 or 2H on days 3 and 8, respectively, k, and 12,arethefractionalH2180 and2H20turnoverrates,and t is time. The rate of CO, production was calculated from the equation of Schoeller et al. (Eq. 3 in Ref. 12) Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 14, 2019.
EFFECT
OF BOLUS
FLUID
INTAKE
are the average of the 20- and 40-min data (“30”- min data). Using the minima 180 and ‘H enrichments (Fig. 1) the energy expenditure values are 19.9 t 7.5% greater than those found using the two reference data points (P < 0.05). Using the mean of the 120- and 180-min values gives a value of 9.8 t 6.4%, which is significantly less than the 30-min point (P < 0.05) but not significantly different from the reference point (P = 0.195, Fig. 2). DISCUSSION
REFti%iNCE
30Yn TlUF .
I...-
t5U min IminI \*
I
..I
./
2. Effect of bolus fluid intake on calculated enerm expenditure. Data are expressed as percentage of reference value. Mean of ZOand 40-min values (valley) is significantly greater than prebolus (reference) values or mean of 12O- and 18O-min values (*P < 0.05). FIG.
the study period; the data show she was successful in reducing intake below energy expenditure. Figure 1 shows the percent change in the saliva enrichments with time after the subjects took the bolus. The postbolus saliva enrichments have been expressed as a percentage of the prebolus (reference) value. Both 180 and 2H show a significant dip with a minimum at -30 min after the subjects drank the water, but the ‘dip” (mean of the 20- and 40-min data points) with 2H is significantly greater than that for 180 (97.55 t 0.44% for I80 vs. 95.48 t 0.43% for 2H, P < 0.05). By ~1 h after dosing a new plateau has been approached for both isotopes. By using the mean of 120- and 180-min points to estimate the plateau, the values were 97.30 t 0.41% and 96.51 t 0.62% of th e predose level for 180 and 2H, respectively (NS) . Figure 2 shows the calculated energy expenditure values using 1) the two reference points, 2) the minima in the isotopic enrichments and the day 8 reference point and 3) the plateau and the day 8 reference point. Figure 1 shows that the minima in isotopic enrichment occurred at -30 min postbolus. Therefore the data used for Fig. 2 TABLE
Selection of experimental conditions. We selected saliva over urine or --blood- as the body fluid to sample for practical reasons. Blood is an invasive way of sampling body water, and serial blood sampling is not always acceptable to test subjects (6). Furthermore, the collection of saliva is often more socially acceptable than collection of urine for subjects outside a clinical setting. A disadvantage of saliva sampling is that there is a small fractionation effect for deuterium (4, 15). If the objective is to compare similar groups run under a common protocol rather than to obtain absolute values, a small consistent error is not a serious problem. Another disadvantage of saliva is the small sample obtained, and this can preclude repeat analyses. The DLW methodology appears deceptively simple. However it has not proven to be a simple matter to evaluate the relative importance of overlapping complicating factors (10, 11). These factors include fractionation effects, variations in enrichment of exogenous water intake during the study period, changes in body composition during the study period, and the differences between the measured 180 and 2H spaces (7,9). For the present study we wanted to avoid the problems associated with the 2H space and equilibration rates being different from the 180 space. Two days were therefore allowed for water equilibration. By allowing two days we could be sure that any residual exchange of 2H with body protein was minimal (6). And basing the EE calculation on the 180 space by using the 1986 equation of Schoeller et al. (12) avoided any potential problems with using the 2H space.
2. Summary of experimental data Subj 1
Subj 2
Subj 3
Subj 4
Subj 5
Subj 6
180
2H
i8u
2H
‘W
2H
180
2H
180
2H
180
2H
0.4 221.4 220.7
-16 1,190 1,185
-3.2 175.0 174.3
-35 664 685
-1.5 176.1 179.1
-32 912 921
-1.0 161.5 161.1
-18 820 854
-3.0 192.4 190.8
-33 864 862
225.7 225.1
1,103 1,136
173.9
1,001
145.7
799
707
130.0 130.8 129.8 131.0 128.0
753 719 711 738 714 719
168.1
127.7 127.7 126.3 128.7 128.8
719 703 683 716 710
135.3
144.6 144.5 143.8 143.4 144.3
602 580 551 578 586 566
132.3
169.6 170.9
163.3 166.4 165.1 165.4 163.5
766 772 767 773 775
213.1 207.2 207.2 207.6 206.7 207.5
1,071 1,036 1,031 1,041 1,043 1,055
86.1
399
64.6
418
82.8
536
109.8
581
142.7
802
Day 1
Pre 3h 4h Day 3 Oh 0.33 h 0.67 h 1.0 h 2.0 h 3.0 h
168.8 168.7 168.7
944 955 961 956 943
Day 8 120 h
102.7
666
Isotopic enrichments are given in delta units (103) relative to standard mean ocean water. Salivas were collected 3 and 4 h post-isotope ingestion for determination of total body water. Subject 4 was given 150 ml of 2Hz18O (batch 2,9.34% la0 and 2.84% 2H). Time schedule for subject 6 was a little different; bolus was given 17 h postdosing and final saliva collected 120 h postdosing. Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 14, 2019.
EFFECT
OF BOLUS
Figure 1 shows that the bolus fluid intake had a large and disproportionate effect on the saliva water 180 and 2H enrichments. At 20 and 40 min, the increase in 2H dilution is significantly greater than the dilution in the I80 dilution. Using the “valley” data overestimates the calculated EE by -20% (Fig. 2). We suggest that the valley in the isotopic enrichment of the saliva occurs because the ingested water first enters the plasma compartment from the gut and then is distributed more slowly into the various extravascular pools such as saliva. By - 1 h postbolus, the system approaches a new equilibrium, and the energy values decrease toward the reference values. The following explanation is proposed for the differential effect between ‘$0 and 2H. The saliva isotopic enrichments reflect the vascular pool because saliva is derived from the vascular pool. After a large bolus of water has been ingested, two processes occur: I) water from the gut enters the vascular compartment, and 2) there is exchange of H+, OH-, and H,O between the gut water and the vascular pool. The first process will lead to an increase in the vascular pool size if the rate at which water enters the vascular compartment is greater than that at which it leaves the vascular pool for urine or the extravascular pool. Because the bolus was not enriched, the plasma compartment enrichments will be less than that of the extravascular pool. The second process, differential exchange rates of H+ and OH- can account for the greater decrease in 2H enrichment relative to 180 in the plasma compartment. If the rate of H entry from the gut into the vascular pool is faster than that for 0, the transient dip in the isotopic enrichments in the plasma water pool, and hence in the saliva, will be greater for H than 0. Because H+ has a greater ionic mobility than OH-, it is not unreasonable to argue that (unlabeled) H+ will move into the vascular compartment more rapidly than (unlabeled) OH- (1). An alternate explanation would be that the effect is specific to saliva, being due to incomplete equilibration in the mouth after the bolus. Such an explanation is possible. Carefully collected urines were avaliable for subject 6. Energy expenditure values were 2,890,2,793, and 2,708 kcal/day for the reference, 30-min, and 150-min times, 3. Subject characteristics water values
and total body
TABLE
Subj No. I
49%yr Ht, cm w kg TBW (18U, 3 h) TBW (180, 4 h) TBW (‘H, 3 h) TBW (‘H, 4 h) TBW,
34 1’76 54
2
35 168 69.4
3
26 176 66.7
4
28 173 79.3
5
6
34 64 60.5
50 178 78
34.83
38.82
43.35
47.37
39.40
40.07
34.95
38.87
42.63
47.49
39.72
40.17
34.03
40.91
43.47
48.97
45.75
44.02
34.17
39.66
43.06
47.06
45.55
43.63
FLUID TABLE
85
INTAKE
4. Dietary
expenditure
intake data and energy
values Subj No.
Dietary intake Carbohydrate, % Fat, % Protein, % Intake, kcal/day EE-1 by DLW, kcal/day
I
2
3
4
5
49 35 16 2,270 2,235
52 32 16 2,260 2,475
56 32 12 1,875 4,196
47 39 14 2,804 3,807
55 31 14 2,133 2,255
Intake is mean daily energy intake calculated from 5 days of intake records. EE-1, energy expenditure value calculated by doubly labeled water (DLW) method using the &zy 3 and 8 reference saliva values. Energy intake was not determined for subject 6. Energy expenditure was 2,387 kcal/day.
respectively. The corresponding saliva values were 2,947, 3,195, and 2,807 kcal/day. However, urine data do not necessarily have to reflect in the time period immediately after the bolus. It depends on when the urine enters the bladder after the bolus. If the bolus was given at t = 0, increased urine production may not likely start instantaneously and proceed at a uniform rate until the bolus is voided over l-2 h. More likely, the rate of urine production will increase to a maximum and then return to baseline during that period. Only if the rate of urine production was uniform or most of the bolus voided in the first -40 min would an effect be seen. If most of the urine is produced toward the end of the 1st h no or minimal effect will be seen because the effect is largely over by then. If the mean saliva valley values for the 180 and 2H enrichments (mean of the 120- and 180-min points) is used in conjunction with the day 8 reference value, the resultant EE values are 21% higher than the reference EE values (Fig. 2). Using the mean of the 120- and MOmin (150,min) values as estimates of the plateau gives a lower value, -lo%, which is not significantly lower than the value obtained at 30 min. The amount of fluid taken in by our subjects may have been somewhat large. Had the dose been halved the error would still be substantial. From our data it can be estimated that a 300-ml bolus (a can of soda) can result in an overestimating EE by -8% if saliva sampling is at the minima. The point is that while there remain several uncertainties in such complex parameters with the method such as which fractionation constant to use and how to handle the difference between the 2H and ‘$0 spaces, neglect of a simple experimental precaution like avoiding bolus fluid intake during the hour before sampling can lead to significant and avoidable errors in the calculated energy expenditure values. This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant DK-35612. Address for reprint requests: T. P. Stein, Dept. of Surgery, School of Osteopathic Medicine, University of Medicine and Dentistry of New Jersey, 401 Haddon Ave., Camden, NJ 08103. Received 25 June 1990; accepted in final form 19 June 1991.
total body water.
REFERENCES P. W. Physical 1978, p. 819-848.
1. ATKINS,
Chemistry.
San Francisco, CA: Freeman,
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 14, 2019.
86
EFFECT
OF BOLUS
2. COWARD,W. A., ANDA. M. PRENTICE.Isotope method for the measurement of carbon dioxide production rate in man. Am. J. Clin. Nutr. 41: 659-661,1985. 3. DELANY,J. P., D. A. SCHUELLER,R. W. HOYT, E. W. ASKEW, AND M. A. SHARP.Field use of D,% to measure energy expenditure of soldiers at different energy intakes. J. AppC Physiol. 67: 1922-1929, 1989. 4. HALLIDAY, D., AND A. G. MILLER. Precise measurement of total body water using trace quantities of deuterium oxide. Biomed. Mass. Spec. 4: 82-87, 1977. 5. HOYT, R. W., T. P. STEIN, H. R. LIEBERMAN,T. E. MORGAN,E. J. IWONYK,E. W. ASKEW,AND A. CYMERMAN.Doubly labeled water (DLW) method accurately estimates energy expenditure during field operation. J. Appl. Physiol. 71: 16-22, 1991. 6. HUSS-ASHMORE,R., J. L. GOODMAN,T. E. SIBNA, AND T. P. STEIN. Energy expenditure of young Swazi women as measured by the doubly labeled water method. Eur. J. Clin. Nutr. 43: 739-748, 1989. 7. NAGY,K. A. CO, production in animals: analysis of potential errors in the doubly labeled water method. Am. J* Physiol. 238 (Regulatory Integrative Comp. Physiol. 7): R466-R473, 1980. 8. NOVICK, W. M., M. NUSBAUM,AND T. P. STEIN. The energy costs of surgery as measured by the doubly labeled water (‘H,“O) method. Surgery 103: 99-105,1987. 9. SCHOELLER,D. A. Energy expenditure from doubly labeled water: some fundamental considerations in humans. Am. J. Clin. Nutr. 38: 999-1005,1984. 10. SCHOELLER,D.A.,W.H. DIETZ, E. VAN~ANTEN,AND P.D. KLEIN. Validation of saliva sampling for total body water determination by Hs180 dilution. Am. J. Clin. Nutr. 35: 591-594, 1982.
FLUID 11.
12.
13. 14.
15 .
“* 17,
18.
INTAKE
SCHOELLER,D. A., C. A. LEITCH, AND C. BROWN. Doubly labeled water method: in vivo oxygen and hydrogen isotope fractionation. Am. J. Physiol. 250 (Regulatory Integrative Camp. Physiol. 19): R823-R830,1986. SCHOELLER,D. A.,E. RAVUSSIN,Y. SCHUTZ, K. J. ACHESON,P. BAERTSCHI,AND E. JEQUIER. Energy expenditure by doubly labeled water: validation in humans and proposed calculation. Am. J. Physiol. 250 (Regulatory Integrative Comp. Physiol. 19): R823R830,1986. SCHOELLER,D. A., AND E. VAN SANTEN. Measurement of energy expenditure in humans by doubly labeled water method. J. Appl. Physiol. 63: 955-959,1982. SEALE,J., W. V. RUMPLER, AND J. M. CONWAY.Comparison of energy estimation methods using double labeled water, intake balance and direct/indirect calorimetry in man. Am. J. Clin. Nutr. 52: 66-71,199O. STEIN, T. P., R. W. HOYT, M. O'TOOLE, R. G. SETTLE, AND W. D. B. HILLER. Determination of energy expenditure during heavy exercise, normal daily activity and sleep using the doubly labeled water method. Am. J. CZin. Nutr. 45: 534~539,1986. WEIR, J. B. DEV. New methods for calculating metabolic rate with special reference to protein metabolism. J. Physiol. Land. 109: l-9, 1949. WESTERTERP,K, R., F. BROUNS,W. H. M. SARIS, AND F. TEN HOOR. Comparison of doubly labeled water with respirometry at low-and high-activity levels. J. Appl. Physiol. 65: 53-56, 1988. WESTERTERP,K. R., W. H. M. SARIS, M. VAN Es, AND F. TEN HOOR.Use of the doubly labeled water technique in humans during heavy sustained exercise. J. Appl. Physiol. 61: 2162-2167, 1986.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 14, 2019.
