![[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}
|
|
|
|
|
||
USDA-Agricultural Research Service Children's Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine, Houston, TX 77030 and * Unidad de Investigacion en Nutricion, Instituto Mexicano del Seguro Social, CMN, SXXI, Mexico City
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGMENTS
LITERATURE CITED
ABSTRACT 
Energy-sparing mechanisms may be elicited to meet increased energy requirements imposed by lactation on women who reside in poor, rural communities in developing countries. The objectives of this study were to measure total energy expenditure and its components, basal and activity energy expenditure, and to investigate their relationships with lactation performance in a total of 40 rural Mesoamerindians stratified according to postpartum body mass index. Total energy expenditure and fat-free mass were measured by the doubly labeled water method, and basal metabolic rate was determined by indirect calorimetry at 3 and 6 mo postpartum. Physical activity level was taken as the ratio of total energy expenditure to basal metabolic rate. Milk energy output, which is not included in total energy expenditure, was computed from 24-h milk intake (test-weighing) and energy concentration of milk (bomb calorimetry). Anthropometric measurements revealed negligible mobilization of tissue stores. Mean (± SD) total energy expenditures were 8912 ± 1296 kJ/d and 9253 ± 1298 kJ/d for the lower and higher body mass index groups, respectively. Adjusted for fat-free mass, total energy expenditure was higher in the lower body mass index group (P = 0.05). Adjusted for fat-free mass, basal metabolic rate did not differ between groups. Physical activity level was significantly higher in the lower body mass index group (P = 0.03). Lactation performance did not differ between groups and was not associated with total energy expenditure or its components. Despite the maintenance of energy balance and heightened energy requirements of lactation, energy-sparing mechanisms were not evident in these lactating Mesoamerindians.
Key words: energy expenditure, basal metabolic rate, physical activity, breastfeeding, humans.INTRODUCTION
Lactation is the most energy-demanding phase of the human reproductive cycle. The incremental energy cost of lactation adds a burden amounting to approximately 25% of the nonlactating requirement. The FAO/WHO/UNU Expert Consultation (1985) recommends an additional energy allowance of 2.2 MJ/d for the first 6 mo of lactation, based on an average milk production of 796 mL/d with an energy content of 2.9 MJ/L, an 80% efficiency of conversion of food energy to milk, and mobilization of maternal fat stores equivalent to 835 kJ/d. If adequate fat is not deposited during pregnancy, the full energy allowance (2.9 MJ/d) is recommended.
The energy cost of lactation may be met by an increase in food intake, mobilization of body fat reserves, reduced energy expenditure or increased metabolic efficiency. In poor, undernourished populations, the extra energy cost of lactation is imposed upon a limited food supply, and lactating women may not have the option to increase their energy intake. Food intake studies from developing countries have consistently reported that energy intakes during lactation are considerably lower than recommendations (Adair and Pollitt 1982
, Prentice et al. 1981, Reynolds et al. 1988, Schutz et al. 1980). However, studies in which total energy expenditure and energy intake have been measured suggest a high likelihood of underreporting of food intake (Singh et al. 1989). Although some studies from developing countries document substantial weight loss during lactation, others report remarkable weight stability (Prentice and Prentice 1990).
These observations raise the alternative possibility of energy-sparing mechanisms. A decrease in basal metabolism or physical activity level, or increased work efficiency, would offset the energy cost of lactation. Evidence of energy-sparing mechanisms in lactating women from developing societies is conflicting, although adaptations indeed may vary according to women's nutritional status and work demands. The basal metabolic rate (BMR)4 of lactating women from rural Guatemala (Schutz et al. 1980), The Gambia (Frigerio et al. 1991, Singh et al. 1989) and India (Madhavapeddi and Rao 1992) was not any different from that of nonlactating women. In contrast, lower BMR were reported in Philippine lactating women compared with pregnant women at 13 wk gestation (Guillermo-Tuazon et al. 1992), and higher values were reported in Indian lactating women compared with nonpregnant-nonlactating women (Khan and Belavady 1973).
Physical activity patterns in rural lactating women have been studied by time-motion observations (Guillermo-Tuazon et al. 1992, Madhavapeddi and Rao 1992, Panter-Brick 1993, Roberts et al. 1982), heart rate monitoring (Schutz et al. 1980) and the doubly labeled water (DLW) method (Singh et al. 1989). Changes in physical activity have been documented during lactation, but adaptability depended on the stage of lactation and the women's work load. By and large, women were expected to resume their normal agricultural and household tasks after the first few months of lactation. The effect of total energy expenditure (TEE) or physical activity on lactation performance of women living in rural, developing areas, however, has not been investigated explicitly.
The specific aims of this study were to measure TEE and its components, basal and activity energy expenditure, and to investigate their relationships with lactation performance in rural Mesoamerindians stratified according to postpartum body mass index (BMI). Possible energy-sparing mechanisms were explored in these women faced with increased energy requirements of lactation. The lactation performance of women with lower BMI and high physical activity was expected to be most vulnerable, and it was in this group that energy-sparing mechanisms were expected. These data were used to re-evaluate the energy requirements of lactating women living under similar conditions.
MATERIALS AND METHODS
), subject selection criterion was defined for a lower BMI group (<23 kg/m2) and a higher BMI group (
23 kg/m2). Women also were required to be between the ages of 18 and 35 y, parity 
6, and free of any chronic diseases, medications and alcohol consumption. Women who met the subject selection criteria were recruited from the local prenatal clinic, which serves 70% of all pregnant women in the community. Total energy expenditure, BMR, physical activity and lactation performance were measured at 3 and 6 mo postpartum. The study was conducted in San Mateo Capulhuac, a rural community southwest of Mexico City, which is 2800 m above sea level and encompasses approximately 25 km2. Its 5500 inhabitants are primarily Otomi Indians who depend on subsistence farming. The study was reviewed and approved by the Instituto Mexicano del Seguro Social and Baylor Affiliates Review Boards for Human Subject Research. Written informed consent was obtained for all studies.
Subjects.
Twenty-one women with lower BMI (21.4 ± 0.9 kg/m2, range 18.5-22.8) and 19 with higher BMI (25.7 ± 1.9 kg/m2, range 23.1-29.1) were recruited immediately after giving birth to term infants with birth weights greater than 2500 g. Two women (one from each group) did not return for the 6 mo study interval due to family objections.
Maternal anthropometry.
Weight and height were measured using a beam balance and stadiometer (Holtain Limited, Crymych, U.K.), respectively, at 0.5, 3 and 6 mo postpartum.
Dietary intake.
A combination weighed-recall dietary method was used to measure food intake for three consecutive days. A research assistant was present in each woman's home from 0700 and 1900 h to weigh and record all food ingredients as they were prepared and the portion consumed by the woman. Raw ingredients of mixed dishes were weighed, as well as the final cooked dish. Between 1900 h that evening and 0700 h the next morning, all food items eaten were recorded by recall. Nutrient intakes were computed using Mexican food composition tables (Hernandez et al. 1980).
Total energy expenditure.
The doubly labeled water (DLW) method was used to measure total energy expenditure (TEE). After collection of baseline urine samples, a dose equivalent to 100 mg 2H2O/kg body wt and 125 mg 18O/kg body wt was administered to the mother orally (Isotec Inc., Miamisburg, OH). Post-dose samples of urine were collected for 14 d and frozen at 
20°C. Urine samples were analyzed for 2H abundance using a Finnigan Delta-E gas-isotope-ratio mass spectrometer (Finnigan MAT, San Jose, CA) and for 18O abundances using a VG-SIRA 12/ISOPREP-18 mass spectrometer system (Isotech Limited, Middlewich, Cheshire, U.K.) (Wong et al. 1987 and 1988). The 2H and 18O natural abundances of urine samples were measured with an average precision of 1.1
and 0.16, respectively. Enriched levels of 2H (570) and 18O (250) in urine were measured with an average precision of 3.1% and 0.84% and an average accuracy of 5.0 and 0.02, respectively. Total energy expenditure was estimated using the multipoint slope-intercept method of calculation (International Dietary Energy Consulting Group 1990). Carbon dioxide production rates (rCO2) were calculated from the isotope dilution spaces and fractional turnover rates of 2H and 18O. Total fractionated water loss was calculated as 2.3 rCO2. Precision of rCO2 was calculated according to the methods of Cole and Coward (1992) and Ritz et al. (1996). Oxygen consumption (rO2) was calculated from rCO2 using the food quotient 0.92, which was computed on the basis of the macronutrient content of the diet of the Otomi women (Black et al. 1986). The Weir equation was used to convert rCO2 and rO2 to TEE (de V. Weir 1949). Total energy expenditure was referenced to fat-free mass (FFM) estimated from deuterium dilution, using a FFM hydration constant of 0.73. Throughout the 14-d study, the ambient temperature and humidity were recorded continuously in the subject's home on a hygrothermograph (Cole Palmer, Chicago, IL). Morbidity records also were kept by the fieldworker who visited the subject daily to collect urine samples.
Basal metabolic rate.
The woman and her infant were brought to the clinic facility in San Mateo Capulhuac the evening before the basal metabolic rate (BMR) measurement. A 15-min sham measurement was performed to acquaint the woman with the hood apparatus and procedure. The woman slept at the clinic; food and energy-containing beverages were withheld for 12 h prior to the BMR measurement. In the morning, the woman was awakened and the infant was separated from the mother, who remained in bed clothed and covered with blankets. An electric heater was used to warm the room to 20-22°C. A 40-min BMR measurement was performed using the DeltaTrac hood apparatus (SensorMedics, Yorba Linda, CA). Prior to each measurement, the DeltaTrac was calibrated with SensorMedics standard gases and an ethanol burn test was performed; N2/CO2 gas infusion tests also were performed periodically to calibrate the system.
Activity energy expenditure (AEE).
Energy expended on physical activity was calculated as AEE = TEE BMR (0.1TEE). The 0.1 fraction of TEE was designated for the thermic effect of feeding.
Physical activity level (PAL).
The level of physical activity also was expressed as a ratio of TEE/BMR.
Activity pattern.
At 3 and 6 mo postpartum, a research assistant visited each woman at home to record physical activities over a 12-h period from 0700 to 1900 h. The woman was encouraged to engage in her normal activities, and the research assistant accompanied her if excursions were made outside the house. A description of each activity and the time spent in that activity were recorded on a minute grid. The daytime physical activities were collapsed into seven categories: household tasks, food preparation and cooking, sitting, washing clothes and dishes, standing, walking, and walking with load. The number of minutes spent in each category was calculated; for the 12 h of unobserved time it was assumed that the women rested for 4 h and slept for 8 h. Energy expenditure in the specified categories was calculated as multiples of the women's measured BMR as follows: 2.7× BMR for household tasks, 1.8× BMR for food preparation and cooking, 1.4× BMR for sitting, 3.0× BMR for washing clothes and dishes, 1.5× BMR for standing, 3.0× BMR for walking, 4.0× BMR for walking with load, 1.2× BMR for resting and 1.0× BMR for sleeping (FAO/WHO/UNU Expert Consultation 1985).
Milk energy output.
Milk energy output was computed from the infant's 24-h milk intake and energy concentration of a daily representative milk sample. Milk intake was measured by a community fieldworker using the 24-h test-weighing technique in the home (model 3862MP8, Sartorius, Gottingen, Germany). On a day separate from the day of test-weighing, milk composition was determined on a pooled milk sample expressed at 1000, 1400, and 1800 h (Stafford et al. 1994). The milk contents of one breast, which the infant was not allowed to suckle for 2 h before the procedure, were completely expressed using an electric breast pump (Egnell, Inc., Cary, IL). The energy content of the pooled milk sample was analyzed by bomb calorimetry (Parr Instruments, Moline, IL).
Statistical analysis.
Minitab Statistical Software program (release 10.5×, 1995, Minitab Inc., 1995, State College, PA) was used for data reduction and statistical analysis. Chi-square and unpaired t-tests were used to test for differences between groups. Relationships between physiological variables were examined using Pearson correlation coefficient and linear regression. Slopes of the regression lines of the lower and higher BMI groups were compared by examining the interaction between group and the independent variable. Results did not indicate that these slopes were different, and therefore correlations are presented combining across both groups. Analysis of variance and covariance with repeated measures (BMDP2V, Statistical Software, Inc., 1993, Los Angeles, CA) was used for the statistical comparisons between the lower and higher BMI groups and between the 3 and 6 mo postpartum periods. Analysis of covariance was used to normalize TEE, dietary intake, BMR and AEE for differences in body weight or FFM between the lower and higher BMI groups, thereby avoiding the assumption of a zero intercept. Group (lower or higher BMI) and time (3 or 6 mo postpartum) were treated as main effects, with a group × time interaction; weight or FFM was entered as a covariate. Statistical significance was set at P 0.05. All values are presented as means ± SD.
RESULTS
Table 1.
Subject characteristics1
|
Table 2. Dietary intake of the Otomi women at 3 and 6 mo postpartum1 |
0.05). The fractional turnover rates of 18O and 2H were higher in the lower BMI group. Carbon dioxide production was not different between groups or time periods. Precision of the rCO2 measurements was 18.9 L/d arising from analytical error and physiological day-to-day variation and 20.1 L/d arising from natural abundance variation; the overall precision was equal to 29.6 L/d or 7.4% of rCO2.
|
Table 3. Isotope dilution spaces, fractional turnover rates of 2H and 18O, fractionated water loss, and total energy expenditure of the Otomi women at 3 and 6 mo postpartum1 |
|
Table 4. Basal metabolic rates, activity energy expenditure and physical activity levels of the Otomi women at 3 and 6 mos postpartum1 |
0.30, 0.42; P = 0.06, 0.01). Activity energy expenditure (AEE) in absolute terms was not different in the two groups. However, adjusted for body weight, AEE was significantly higher among the women with lower BMI rather than higher BMI (adjusted means 2786 vs. 2022 kJ/d) (P = 0.05). Adjusted for FFM, AEE was higher in the lower BMI group (adjusted means 2906 vs. 1872 kJ/d) (P = 0.006).
DISCUSSION
, Goldberg et al. 1991, Lovelady et al. 1993, Singh et al. 1989), of which only one (Singh et al. 1989) was from the Third World. Potential sources of error unique to lactation are 1) the isotope exchange and sequestration during de novo synthesis of milk fat and lactose and 2) the increased water flux into milk. On the basis of theoretical calculations, the export of exchangeable hydrogen bound to solids in milk may result in a 1.0-1.3% underestimation of carbon dioxide production (International Dietary Energy Consulting Group 1990). 2H sequestration may increase this underestimation to 1.5 to 3.4%. The higher water turnover in these lactating women might have contributed to our higher overall precision (7.4%) relative to other DLW reports (4.9%) (Ritz et al. 1996).
) and low values (8.9 MJ/d or 1.54 × BMR) reported in British lactating women (Goldberg et al. 1991). Values reported for American women (10.1 MJ or 1.76 × BMR) (Lovelady et al. 1993) and Swedish women (10.7 MJ/d or 1.80 × BMR) (Forsum et al. 1992) were somewhat higher during lactation. The physical activity level of the Otomis would be classified as moderate in the lower BMI group and light in the higher BMI group according to FAO/WHO/UNU (1985). There was no effect of season or stage of lactation on TEE. A higher level of physical activity was expected in these rural women, but their observed patterns of activity were consistent with the light-to-moderate levels of expenditure. The time-motion data yielded mean estimates of TEE that were closely matched to the DLW results. The women for the most part stayed close to home, caring for children and performing their household tasks. In the community of San Mateo Capulhuac, strenuous physical activities such as farming and hauling firewood are performed exclusively by men. The majority of homes had piped water, eliminating the laborious task of hauling water. The women partake in the shucking and cleaning of corn, but nowadays most women use the central mill for grinding.
, Illingworth et al. 1986, Madhavapeddi and Rao 1992, Motil et al. 1990, Piers et al. 1995, Schutz et al. 1980, Singh et al. 1989), whereas others reported an increase (Forsum et al. 1992, Khan and Belavady 1973, Sadurskis et al. 1988, Spaaij et al. 1994) or a decrease (Guillermo-Tuazon et al. 1992, Lawrence et al. 1986). The high altitude environment may have contributed to the higher BMR because of thermic responses to cold or hypoxic conditions. Elevated BMR has been observed at high altitude in several studies (Gill and Pugh 1964, Grover 1963, Nair et al. 1971, Picon-Reategui 1961).
), the percentage of body fat was inversely related to PAL. The women with higher BMI may maintain their body energy stores by being relatively less active.
). In The Gambia, physical activity returns to prepregnancy levels after the first month of lactation (Roberts et al. 1982). High TEE (9660 kJ/d or PAL = 1.94) was observed in lactating Nepali women during the monsoon season when obligatory work demands could not be curtailed (Panter-Brick 1993). As in The Gambia (Lawrence and Whitehead 1988), the high TEE of the Nepali women resulted from many hours of performing tasks of relatively low energy cost. The role of decreased physical activity as an energy-sparing mechanism is now questioned under conditions prevailing in the Third World (Ferro-Luzzi 1990). Although women's involvement in agriculture in Latin America does not approach that of African and Asian women, time-motion studies indicate that women increase their participation as agricultural field workers as the economy changes from subsistence farming to cash cropping (McGuire 1979). The community of San Mateo Capulhuac is still based on a traditional subsistence economy.
1·d1) of the Otomi women were substantially below international recommendations for moderately active, lactating women (10.8-11.7 MJ/d or 219 kJ·kg1·d1) (FAO/WHO/UNU Expert Consultation 1985). Total energy requirements of these lactating Otomi women were estimated from the sum of TEE and milk energy output. Total energy requirements were 11.0 MJ/d for the lower BMI group and 11.4 MJ/d for the higher BMI group, of which 2.2 MJ/d was exported into milk. The assumptions underlying the FAO/WHO/UNU full energy allowance for lactation (2.9 MJ/d) were valid in these Otomi women, because fat mobilization was negligible. From energy balance data (Barbosa et al. 1996), we know that weight loss offsets only a fraction of the total energy requirement (0.20 and 0.06 MJ/d in the two respective BMI groups); therefore, food intake must have been underestimated on the order of 20%. The Otomi women must have been consuming sufficient energy on a daily basis to support their milk production.
FOOTNOTES
Manuscript received 8 May 1996. Initial reviews completed 31 July 1996. Revision accepted 10 October 1996.
ACKNOWLEDGMENTS
We gratefully acknowledge A. Romero, C. Fonseca, I. Amerlink, M. Perez, M. Juvenal and M. Vidal for data collection, N. Mehta, S. Zhang, D. Roose, K. Usuki and Z. Colon for technical assistance, L. Loddeke for editorial review and I. Tapper for manuscript preparation.
LITERATURE CITED
This article has been cited by other articles:
|
|
 |
|
  S. Villalpando, M. D. Prado, A. Lance, E. Alfonso, M. Rodriguez, H. Demmelmair, and B. Koletzko [ Am. J. Clinical Nutrition, December 1, 2001; 74(6): 827 - 832. [Abstract] [Full Text]
|
|
|
|
 |
|
 N. F. Butte, W. W. Wong, and J. M. Hopkinson Energy Requirements of Lactating Women Derived from Doubly Labeled Water and Milk Energy Output J. Nutr., January 1, 2001; 131(1): 53 - 58. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
