Lactation, weaning, and calcium supplementation: effects on body composition in postpartum women 1 3

Lactation, weaning, and calcium supplementation: effects on body composition in postpartum women 1 3 Karen S Wosje and Heidi J Kalkwarf ABSTRACT Background: Concern that long-term weight retention after pregnancy contributes to obesity underscores the need to identify factors that facilitate postpartum weight loss. Lactation is believed to facilitate postpartum weight loss and fat loss. intake also has been hypothesized to promote weight loss and fat loss. Objective: We addressed the following questions: 1) whether lactation enhances loss of fat mass, and 2) whether loss of fat mass during lactation and after weaning is greater in women receiving calcium supplementation than in women receiving placebo. Design: We used data from 87 lactating and 81 nonlactating women enrolled in a randomized, double-blind, calcium supplementation trial from 2 wk to 6 mo postpartum and data from 76 previously lactating and 82 nonlactating women enrolled in a parallel trial from 6 to 12 mo postpartum. Body fat and lean masses were measured by using dual-energy X-ray absorptiometry. Results: Nonlactating women lost whole-body, arm, and leg fat at a faster rate than did lactating women between 2 wk and 6 mo postpartum (lactation group time effect, P 0.01). Fat mass of the trunk, arms, and legs decreased between 6 and 12 mo postpartum regardless of previous lactation status (time effect, P 0.001). supplementation did not affect postpartum fat loss. Conclusions: Body-composition changes occur differently in nonlactating and lactating women during the first 6 mo postpartum and occur at some sites until 12 mo postpartum regardless of previous lactation status. Clinicians should use caution when advising lactating mothers about expected rates of postpartum fat loss. supplementation (1 g/d) does not promote postpartum weight loss or fat loss. Am J Clin Nutr 2004;80:423 9. KEY WORDS Lactation, weaning, postpartum women, body composition, calcium, fat mass, obesity, breastfeeding INTRODUCTION Concerns that long-term weight retention after pregnancy contributes to increasing rates of obesity (1) underscore the importance of understanding postpartum energy utilization. The dietary reference intake for energy for lactating women is 330 kcal/d (1383 kj/d) higher than that for nonlactating women. The dietary energy requirement for lactating women during the first 6 mo postpartum is based on estimates of energy expenditure, milk energy output (500 kcal/d, or 2095 kj/d), and energy mobilization from fat (170 kcal/d, or 712 kj/d) (2). The dietary reference intake assumes that 66% of the energy needed for milk synthesis is obtained from dietary intake and that the remaining 34% comes from mobilization of fat deposited during pregnancy (2). The extent to which lactation results in fat loss is unclear. Many studies indicate that weight lost during lactation consists primarily of fat mass (3 5). Some (6, 7), but not all (8, 9), studies found no difference in postpartum anthropometric or bodycomposition changes according to lactation practice when other factors, such as gestational weight gain, were taken into account. Butte and Hopkinson (9) described differences in postpartum body-composition changes between lactating and nonlactating women as subtle and short term. Because of the hypothesized fat mobilization during lactation, breastfeeding is often cited as a factor that facilitates postpartum weight loss. Dewey et al (8) showed that breastfeeding mothers lost more weight than formula-feeding mothers between 1 and 12 mo postpartum and that greater frequency of and time spent breastfeeding were associated with greater weight loss. However, another study indicated that lactation status or amount of breastfeeding did not significantly influence postpartum weight change (10). One reason for this lack of influence may be higher energy intakes in lactating women than in nonlactating women (11 13). The higher energy intake of lactating women may be a physiologic consequence of elevated concentrations of prolactin, an appetite stimulant (14). Theoretically, the reduction in prolactin concentrations after weaning may result in decreased appetite and energy intake, but little is known about the influence of weaning on postpartum changes in weight and body composition. Parikh and Yanovski (15) recently reviewed evidence on calcium and adiposity and concluded that increasing calcium or dairy intake may inhibit gains in weight or fat mass. Zemel (16) proposed that low dietary calcium intake increases circulating 1,25-dihydroxyvitamin D [1,25(OH)D 2 ] concentrations, which stimulates calcium influx into adipocytes and thereby results in suppression of lipolysis and increased lipogenesis. Influences of supplemental calcium on changes in weight and fat mass in 1 From the Cincinnati Children s Hospital Medical Center, Division of General and Community Pediatrics, Cincinnati. 2 Supported by grants R01 AR41366 and M01 RR08084 from the General Clinical Research Centers program, National Center for Research Resources, National Institutes of Health. 3 Address reprint requests to KS Wosje, Cincinnati Children s Hospital Medical Center, 3333 Burnet Avenue, MLC 7035, Cincinnati, OH 45229. E-mail: karen.wosje@cchmc.org. Received December 4, 2003. Accepted for publication February 25, 2004. Am J Clin Nutr 2004;80:423 9. Printed in USA. 2004 American Society for Clinical Nutrition 423

424 WOSJE AND KALKWARF postpartum women are unknown. We used data from a randomized, double-blind trial designed to determine the effects of calcium on bone mass during lactation and after weaning to address the following questions: 1) whether lactation enhances loss of fat mass during lactation and after weaning, and 2) whether loss of fat mass during lactation and after weaning is greater in women receiving calcium supplementation than in women receiving placebo. SUBJECTS AND METHODS Subjects Information on the subjects and methods in the original randomized, double-blind trial on the effects of calcium supplementation on bone density during lactation and after weaning are described in detail elsewhere (17). Briefly, 4 groups of women were recruited according to their time since delivery and their lactation status to participate in either the lactation study (2 wk to 6 mo postpartum) or the weaning study (6 12 mo postpartum). To maximize the ability to determine effects of calcium on bone, the original study design called for selection of women with a low habitual calcium intake ( 800 mg/d). None of the participants were taking medications or using hormonal contraceptives. All participants had singleton pregnancies of 37 wk of gestation and had taken vitamins during pregnancy. None of the women participated in both the lactation study and the weaning study. A total of 97 lactating and 99 nonlactating women were enrolled in the lactation study. The lactating women intended to breastfeed for 6 mo and to provide no more than one formula feeding per day. The nonlactating women fed their infants exclusively with formula. Measurements were obtained at enrollment [2.3 0.2 wk postpartum (x SD)] and at 3 and6mo postpartum. A total of 95 lactating and 92 nonlactating women were enrolled in the weaning study. In the weaning study, the women were enrolled at 23.3 3.4 wk postpartum (baseline), and measurements were obtained at baseline and 9 and 12 mo postpartum. The lactating women in the weaning study were fully breastfeeding at enrollment and weaned their infants during the 2 mo after enrollment. The nonlactating women from the weaning cohort either had not breastfed their infants at all or had breastfed for 2 wk. Of the 383 women enrolled, 8 lost interest (6 from the lactation study), 13 became ill or required medication or iron supplementation (7 from the lactation study), 5 started using hormonal contraceptives (all from the lactation study), 6 were lost to follow-up (3 from the lactation study), 11 became pregnant (3 from the lactation study), 4 weaned their infants early (all from the lactation study), 1 could not swallow pills (from the weaning study), and 9 did not wean their infants within 3 mo after enrollment (all from the weaning study). The original study was approved by the Institutional Review Board of Cincinnati Children s Hospital Medical Center. All participants provided written informed consent. Study procedures In both the lactation study and the weaning study, participants were randomly assigned to receive either 1 g CaCO 3 /d (Os-Cal; Marion Merrell Dow, Kansas City, MO) or placebo. Compliance with calcium or placebo supplementation was determined by pill count. Randomization was blocked according to age, race, and lactation status (breastfeeding or formula feeding). All the women received a daily multivitamin containing 400 IU vitamin D (Dayalets; Abbott Laboratories, North Chicago, IL). The participants kept records of the occurrence of menses and provided 3-d food-consumption records and 7-d physical activity logs during the week preceding both the second and third visits. Estimates of average daily intakes of energy and calcium were calculated by using NUTRITION DATA SYSTEM (Nutrition Coordinating Center, University of Minnesota, Minneapolis). Hours per week spent sleeping and participating in moderate (eg, house painting and lawn mowing), hard (eg, heavy gardening and walking fast), and very hard (eg, jogging and physical labor) exercise were estimated from the 7-d physical activity logs on the basis of the methods of Sallis et al (18). Body composition (fat and lean masses) was measured by using dual-energy X-ray absorptiometry (QDR 1000W or 2000; Hologic, Waltham, MA). All measurements from a given subject were obtained with the same machine. Trunk, arm, and leg values for fat mass and lean mass were obtained from the subregion analysis of the whole-body scan. The average of the left and right limbs was used for arm and leg measures. Whole-body percentage fat values were calculated as whole-body fat (in kg) divided by whole-body mass (in kg). The CVs ranged from 1.30% to 2.03% for whole-body lean mass and from 0.88% to 1.13% for whole-body fat mass. Statistical analyses Student s t test or Pearson chi-square was used to compare baseline and descriptive characteristics between lactation groups. Data for the studies of lactation and weaning were analyzed separately by using similar statistical methods. Repeatedmeasures analyses were carried out on the raw data from all 3 time points for each response variable. Whole-body, trunk, arm, and leg fat masses were considered as primary outcomes of interest. Whole-body percentage fat, lean mass, and body weight were also included as outcomes. Data were analyzed with a mixed model approach by using the MIXED procedure of SAS version 8.02 (SAS Institute Inc, Cary, NC), and the REPEATED statement was used to model within-subject (subject within lactation group by calcium group) variation with covariance structure autoregressive order one. A model containing the effects of lactation group, calcium group, and time (ie, visit), as well as all possible interactions, was first obtained. The three-way interaction term was not significant and was therefore not considered in any further model-selection procedures. In determining the statistical significance (P 0.05) of lactation group time (ie, the lactation effect) and calcium supplementation group time (ie, the calcium effect), the following covariates were included in all statistical models because of their potential influence on body fat and lean masses: height, dietary calcium intake (averaged from the second and third visits), and gestational weight gain. If either interaction term was significant, both interaction terms were retained in the final model. If neither interaction term was significant, a reduced model to test for the main effect of time was examined. After determining final models by using the above procedures, we simultaneously entered energy intake and hours spent in hard or very hard activity (both variables were averages from the second and third visits) into the models. We did not include energy intake or activity as covariates in the initial model-selection procedures because of the potential for overcontrolling for these factors, which we felt might be influenced

LACTATION, WEANING, CALCIUM, AND FAT MASS 425 TABLE 1 Descriptive characteristics of the study participants at baseline of the lactation study (2 wk postpartum) and of the weaning study (6 mo postpartum) according to lactation group and calcium supplementation group 1 Lactation study Weaning study (n 41) Nonlactating Lactating Nonlactating Lactating (n 40) (n 45) (n 42) (n 40) (n 42) (n 38) (n 38) Age (y) 30.2 3.7 29.7 3.8 30.1 3.4 30.0 3.5 30.8 3.1 31.2 3.8 30.5 3.2 30.8 3.1 Height (cm) 2 167 6 165 5 164 6 163 6 165 7 165 6 164 7 164 7 Weight (kg) 3 68.0 11.5 67.9 9.2 65.0 9.5 67.9 11.0 65.1 11.8 66.4 10.5 60.7 14.0 61.4 10.8 WB percentage fat (%) 3 30.2 6.0 30.6 5.1 28.7 4.8 31.0 5.6 30.2 6.8 30.1 5.1 27.2 6.3 28.6 7.2 Fat mass WB (kg) 3 21.0 7.0 21.1 6.3 18.9 5.4 21.4 7.1 20.2 7.9 20.4 6.2 17.1 8.2 18.1 7.8 Trunk (kg) 3 8.0 3.4 8.1 3.2 7.3 2.8 8.3 3.4 8.0 4.4 8.3 3.3 6.3 4.0 7.0 4.0 Arm (g) 3 1295 436 1269 356 1131 293 1315 449 1272 501 1290 359 1121 487 1125 423 Leg (g) 4768 1458 4875 1314 4335 1165 4870 1518 4447 1390 4368 1261 3967 1650 4070 1564 Lean mass WB (kg) 3 44.6 6.0 44.3 4.4 43.7 5.2 44.1 5.4 42.6 5.8 43.7 5.0 41.5 6.1 41.0 4.6 Trunk (kg) 3 24.0 3.1 23.7 2.0 23.9 2.9 23.9 2.9 22.6 3.0 23.2 2.7 21.7 3.1 21.9 2.6 Arm (g) 3 2036 353 2024 256 1951 304 1976 345 1943 309 2025 281 1886 365 1832 267 Leg (g) 6684 1097 6726 875 6388 875 6538 988 6537 1103 6666 867 6444 1165 6208 813 Reported gestational 34.8 12.5 34.7 11.7 32.7 8.9 31.0 9.2 32.5 12.5 34.6 11.9 29.7 10.8 32.7 10.4 weight gain (kg) intake (mg/d) 2,4 739 240 656 218 868 293 843 216 744 214 678 212 684 239 776 222 Energy intake (kcal/d) 2,4 1768 456 1773 367 2196 511 2000 385 1821 440 1800 440 1786 458 1930 405 Exercise (h/wk) 5 Moderate 28 14 25 11 24 11 27 12 28 12 27 13 29 12 27 13 Hard and very hard 2 7 5 10 9 7 7 5 5 6 5 7 6 8 7 8 11 1 All values are x SD. WB, whole body. 2 Significant difference between nonlactating and lactating women in the lactation study, P 0.05. 3 Significant difference between nonlactating and lactating women in the weaning study, P 0.05. 4 Average of two 3-d dietary records (3- and 6-mo visits for the lactation study and 9- and 12-mo visits for the weaning study). 5 Average of two 7-d physical activity records (3- and 6-mo visits for the lactation study and 9- and 12-mo visits for the weaning study). by lactation. The inclusion of energy intake and activity in the final models did not affect our conclusions; therefore, the data presented are from models that did not include these covariates. To illustrate the pertinent findings, data are presented as changes over time. Least-squares mean estimates were obtained from the SAS output for the repeated-measures analyses, which included in the model the lactation group calcium group time interaction and the covariates described above. For example, the least-squares mean for change in weight between visits 1 (2 wk postpartum) and 2 (3 mo postpartum) in nonlactating women receiving calcium was taken from the SAS output as the difference between the least-squares mean for visit 1 and the least-squares mean for visit 2 for that group of women. SAS also provided the SE for each estimate of the difference between least-squares means, and these SEs are reported here. RESULTS General characteristics of the study participants are described in Table 1. Ninety-two percent of the women took 80% of their pills, and 73% of the women took 90%. In both studies, the calcium and placebo groups did not differ significantly in any of the measures presented in Table 1. In the lactation and weaning studies, 97% and 95% of the women, respectively, were white. Parity was 2 1(x SD) across all groups of women. In the lactation study, the nonlactating women were slightly taller than the lactating women and had significantly lower energy and calcium intakes and reported spending significantly more time engaged in hard and very hard exercise during the study. All the nonlactating women reported resumption of menses by 6 mo postpartum, in comparison with only 22% and 29% of the lactating women who were receiving calcium and placebo, respectively. Body weight, whole-body percentage fat, and whole-body fat and lean masses were not significantly different between the nonlactating and the lactating women in the lactation study. Regional fat and lean masses also were not significantly (P 0.1 for all) different between lactation groups at enrollment in the lactation study. In the weaning study, the nonlactating women were significantly heavier at enrollment than were the lactating women and had significantly higher whole-body percentage fat and fat and lean masses in the whole body and in all subregions (P 0.05) except for the leg (fat mass, P 0.10; lean mass, P 0.08). In the weaning cohort, 98% of the nonlactating women reported resumption of menses by 6 mo postpartum, in comparison with 13% and 16% of the previously lactating women who were receiving calcium and placebo, respectively. Effects of lactation and calcium supplementation on body composition: the lactation study (2 wk to 6 mo postpartum) Changes in fat mass during the lactation study are shown in Figure 1. As shown, the nonlactating women lost fat mass from the whole body, arms, and legs at a faster rate than did the lactating women; calcium supplementation had no significant

426 WOSJE AND KALKWARF FIGURE 1. Least-squares mean ( SE) changes in fat mass during lactation and after weaning in nonlactating (NL) and lactating (L) mothers who received supplementation with calcium (Ca) or placebo (P). Each value is the difference between the least-squares mean for the indicated time point and the least-squares mean for the baseline (within lactation and calcium groups) obtained from the repeated-measures analyses, which included in the model the lactation group calcium group time interaction and the covariates of height, dietary calcium intake, and gestational weight gain. influence on fat loss (calcium group time, all sites, P 0.9). Trunk fat loss occurred between 2 wk and 6 mo postpartum in all the women regardless of lactation or calcium group. Changes in body weight, whole-body percentage fat, and lean mass between 2 wk and 6 mo postpartum are shown in Table 2. The rate of decrease in body weight was not significantly influenced by lactation group or calcium group. Whole-body percentage fat decreased in the nonlactating women but did not change significantly in the lactating women between 2 wk and 6 mo postpartum; calcium supplementation did not significantly affect

LACTATION, WEANING, CALCIUM, AND FAT MASS 427 TABLE 2 Change ( ) in weight, whole-body percentage fat, and lean mass between 2 wk and 6 mo postpartum in nonlactating and lactating women receiving calcium or placebo in the lactation study 1 2wkto3mo 3 6 mo Model effect (P) Nonlactating Lactating Nonlactating Lactating Lactation group time 2 group time 2 Time 3 Weight (kg) 0.15 0.14 0.001 2.6 0.4 4 1.4 0.4 0.7 0.4 0.6 0.4 1.6 0.4 1.4 0.4 1.2 0.4 1.1 0.4 Whole-body percentage fat (%) 0.01 0.5 NA 0.9 0.3 0.4 0.3 1.3 0.3 0.5 0.3 0.7 0.3 0.2 0.3 0.6 0.3 0.6 0.3 Lean mass 0.4 0.001 NA Whole body (kg) 1.4 0.2 1.3 0.2 0.4 0.2 0.2 0.2 0.6 0.2 0.9 0.2 0.5 0.2 0.4 0.2 Trunk (kg) 0.6 0.01 NA 1.3 0.2 1.2 0.1 0.2 0.2 0.1 0.1 0.8 0.1 0.9 0.1 0.3 0.1 0.3 0.1 Arm (g) 0.7 0.2 0.05 25 17 31 16 25 17 4 16 16 17 16 16 8 17 14 16 Leg (g) 0.3 0.14 0.4 11 40 24 37 45 40 20 37 72 39 19 37 36 39 58 37 1 n 41 for nonlacting women receiving calcium, 45 for lactating women receiving calcium, 40 for nonlactating women receiving placebo, and 42 for lactating women receiving placebo. NA, not applicable because time was part of a significant interaction with lactation group or calcium group. 2 Determined without the lactation group calcium group time interaction in the model because it was not significant. 3 Determined without interaction terms in the model. 4 Least-squares mean SE (all such values). Each value is the difference between the least-squares means for the time points indicated in the column headings (within lactation and calcium groups) obtained from the repeated-measures analyses, which included in the model the lactation group calcium group time interaction and the covariates of height, dietary calcium intake, and gestational weight gain. changes in whole-body percentage fat. Lactation had no significant influence on changes in lean mass at any site. The significant effect of calcium supplementation on changes in wholebody and trunk lean mass was due to higher rates of lean mass loss between 2 wk and 3 mo postpartum in the calcium groups than in the placebo groups; however, gains or smaller losses in lean mass between 3 and 6 mo postpartum in the calcium groups than in the placebo groups appeared to offset these initial changes, and no net effect of calcium was apparent [mean ( SE) changes from 2 wk to 6 mo: for the whole body, 1.0 0.3 kg (nonlactating calcium group), 1.5 0.3 kg (lactating calcium group), 1.1 0.3 kg (nonlactating placebo group), and 1.3 0.3 kg (lactating placebo group); for the trunk, 1.1 0.2 kg (nonlactating calcium group), 1.3 0.2 kg (lactating calcium group), 1.1 0.2 kg (nonlactating placebo group), and 1.2 0.2 kg (lactating placebo group)]. Effects of weaning and calcium supplementation on body composition: the weaning study (6 12 mo postpartum) As shown in Figure 1, there was no influence of lactation (lactation group time) or calcium supplementation (calcium group time) on fat mass losses in the women in the weaning study. Changes in body weight, whole-body percentage fat, and lean mass between 6 and 12 mo postpartum are shown in Table 3. The rates of decrease in body weight and whole-body percentage fat were not significantly influenced by lactation or calcium supplementation. Changes in whole-body and trunk lean mass were not significant and did not differ significantly by lactation group or calcium group. Differences in changes in arm lean mass between the nonlactating and the previously lactating women between 6 and 12 mo postpartum were not significant. However, differences in changes in leg lean mass between the nonlactating and the lactating women were significant and were particularly apparent between 6 and 9 mo postpartum (Table 3). supplementation had no significant influence on changes in arm and leg lean masses. DISCUSSION Our findings from the lactation study indicate that fat mass changes occur differently in nonlactating and lactating mothers between 2 wk and 6 mo postpartum. Although the average weight loss was not significantly different between the nonlactating and the lactating women, higher rates of fat loss from the whole body, arms (where the lactating women gained fat), and legs occurred in the nonlactating women. Our data agree with those from previous reports that trunk fat and thigh fat are the primary energy depots mobilized to support lactation (9). However, we found that fat mass loss from both central and peripheral sites occurs more rapidly in nonlactating women than in lactating women during the first 6 mo postpartum. The composition of the body weight lost in the nonlactating women was such that whole-body percentage fat decreased from 2 wk to 6 mo postpartum; a significant change in whole-body percentage fat was not apparent in the lactating women. We observed no beneficial influence of calcium supplementation on changes in weight or fat mass. Thus,

428 WOSJE AND KALKWARF TABLE 3 Change ( ) in weight, whole-body percentage fat, and lean mass between 6 and 12 mo postpartum in nonlactating and previously lactating women receiving calcium or placebo in the weaning study 1 6 9 mo 9 12 mo Model effect (P) Nonlactating Previously lactating Nonlactating Previously lactating Lactation group time 2 group time 2 Time 3 Weight (kg) 0.12 0.7 0.01 0.1 0.3 4 0.1 0.3 0.6 0.3 0.6 0.3 0.6 0.3 0.3 0.3 0.6 0.3 0.2 0.3 Whole-body percentage fat (%) 0.9 0.2 0.001 0.2 0.2 0.3 0.2 0.6 0.2 0.5 0.2 0.6 0.2 0.3 0.2 0.2 0.2 0.3 0.2 Lean mass 0.3 0.9 0.2 Whole body (kg) 0 0.2 0.1 0.2 0.1 0.2 0.1 0.2 0.1 0.2 0.3 0.2 0.3 0.2 0 0.2 Trunk (kg) 0.9 0.6 0.16 5 123 144 124 40 120 10 121 101 117 44 124 154 117 140 125 Arm (g) 0.16 0.7 0.001 15 35 28 14 20 14 8 14 20 13 36 14 21 13 10 14 Leg (g) 0.01 0.7 NA 2 37 111 37 26 36 30 36 15 35 129 37 29 35 32 38 1 n 40 for nonlactating women receiving calcium, 38 for previously lactating women receiving calcium, 42 for nonlactating women receiving placebo, and 38 for previously lactating women receiving placebo. NA, not applicable because time was part of a significant interaction with lactation group or calcium group. 2 Determined without the lactation group calcium group time interaction in the model because it was not significant. 3 Determined without interaction terms in the model. 4 Least-squares mean SE (all such values). Each value is the difference between the least-squares means for the time points indicated in the column headings (within lactation and calcium groups) obtained from the repeated-measures analyses, which included in the model the lactation group calcium group time interaction and the covariates of height, dietary calcium intake, and gestational weight gain. our findings indicate that neither breastfeeding nor calcium supplementation facilitates weight loss or fat loss during the first 6 mo postpartum. Pregnancy-induced body fat deposition occurs largely at central sites (ie, trunk and thighs) (9). Although data on alterations in body fat distribution during lactation and after weaning are sparse, these same fat depots also appear to be the primary energy sources mobilized to support lactation (9). We found that whereas the nonlactating women lost fat mass at all sites and achieved reductions in whole-body percentage fat, the lactating women appeared to support milk energy output by mobilizing trunk and leg fat (but gained arm fat) and did not experience significant changes in whole-body percentage fat. Our findings of higher rates of fat loss in the nonlactating women differ from those of Butte et al (19), who reported that changes in wholebody fat mass and fat-free mass, as measured by using a 4-compartment model, did not differ between nonlactating and lactating women during the first 6 mo postpartum. Furthermore, Butte and Hopkinson (9) found no indication that postpartum changes in arm and leg fat were different between nonlactating and lactating women over 1yoffollow-up. To our knowledge, there are no other reports on changes in regional fat and lean masses among postpartum women who used different infant feeding practices. Thus, further clarification as to whether postpartum changes in regional fat distribution differ by lactation status is necessary. Estrogen and androgens are endocrine regulators of energy partitioning and body composition. Hypoestrogenemia and an increased androgen-estrogen ratio are accompanied by an increase in the ratio of trunk fat to leg fat (ie, shift to abdominal fat distribution) in women as they transition through menopause. Lactating women are hypoestrogenemic, and there is evidence that circulating concentrations of the androgen dehydroepiandrosterone sulfate increase between late pregnancy and 3 mo postpartum in lactating women but not in nonlactating women (19). However, in the present study, the hypoestrogenemic state (or increased androgen-estrogen ratio) induced by lactation did not result in a shift to abdominal fat distribution compared with the nonlactating women who had resumed menses. Increased circulating prolactin concentrations stimulate appetite in lactating women (14). Thus, we presume that our failure to find that lactating women lose more fat and weight than do nonlactating women was due to the influence of prolactin on appetite stimulation. We found that the lactating women had higher energy intakes (ie, 327 kcal/d, or 1370 kj/d) than did the nonlactating women in the first 6 mo postpartum. One limitation of our study, however, was that we did not know whether the women were trying to lose weight. It is possible that the nonlactating women intentionally restricted their energy intake whereas the lactating women did not because they presumed that they would lose weight by lactating. It is also possible that lactation status influences postpartum physical activity levels; the nonlactating mothers in our

LACTATION, WEANING, CALCIUM, AND FAT MASS 429 study reported spending significantly more time in hard and very hard activity than did the lactating mothers during the first 6 mo postpartum. Our results suggest that compensations in energy intake and physical activity occur during the first 6 mo postpartum to meet the energy needs for lactation and help spare fat mass. On average, the women in the weaning study lost fat mass at all sites, which indicates that body-composition changes occur until 12 mo postpartum. The gain in leg lean mass between 6 and 9 mo postpartum in the previously lactating women compared with the nonlactating women supports the possibility that lean tissue energy stores from the thigh region were previously used to support lactation, with subsequent recovery of lean tissue after weaning. This speculation should be viewed with caution because our findings from the lactation study did not show that changes in leg lean mass occurred differently in the nonlactating and the lactating women during the first 6 mo postpartum. It remains possible that our crude measure of physical activity was inadequate for determining the potential effects of exercise on lean mass changes in the women in this study. The present study had enough power (80%) to detect a 1.6-kg difference in weight loss due to calcium supplementation during the first 6 mo postpartum. This corresponds to a weight loss of 3.2 kg/y, which is a clinically relevant weight change. On the basis of the speculation by Davies et al (20) that differences in calcium intakes might be associated with changes in body weight of 0.4 kg/y, an extremely large trial would be necessary to detect a significant calcium effect. Zemel (16) suggested that high calcium intakes might suppress serum 1,25(OH)D 2 concentrations and thereby result in reduced calcium influx into adipocytes. Because calcium influx into cells stimulates lipogenesis and inhibits lipolysis, suppressing 1,25(OH)D 2 might have antiobesity effects. Serum 1,25(OH)D 2 concentrations were previously shown to be lower in the calcium-supplemented women than in the placebo-supplemented women from the present study (21); thus, our calcium intervention was sufficient to elicit a response that has been proposed as a mechanism for weight loss, but that response had no significant weight-loss effect. We did find that in the women in the lactation study, higher rates of lean mass loss in the whole body and trunk occurred between 2 wk and 3 mo postpartum in those who received calcium than in those who received placebo. Because higher rates of lean mass loss occurred between 3 and 6 mo postpartum in the placebo group than in the calcium group, there was no significant difference between the groups in net lean mass loss (ie, the placebo group caught up to the calcium group). We do not have a biologically plausible explanation for these findings, and there was no significant effect of calcium supplementation on body-composition changes in the weaning study. In summary, our findings are in agreement with those of previous reports that lean mass is spared during lactation (5, 11), because we found no difference in postpartum lean mass changes between the nonlactating and the lactating women. Postpartum weight loss consists of both fat and lean masses in both lactating and nonlactating women. Our results also differ both from those of previous studies that suggest that breastfeeding facilitates postpartum weight loss (8, 10) and from anecdotal reports about increased rates of weight loss after weaning. supplementation of 1 g/d does not appear to influence fat mass loss during the postpartum period in women with low habitual calcium intakes. The combined findings from the lactation and weaning studies presented herein indicate that postpartum bodycomposition changes 1) occur at different rates between lactating and nonlactating women during the first 6 mo postpartum, 2) are not beneficially influenced by calcium supplementation, and 3) occur at some sites until 12 mo postpartum. Finally, we recommend that when promoting breastfeeding, clinicians use caution in advising mothers with respect to expected rates of weight and fat loss after pregnancy. KSW was responsible for data analysis, interpretation, and manuscript preparation. HJK was responsible for the original study design and for data analysis, interpretation, and manuscript preparation. Neither KSW nor HJK had any conflicts of interest. REFERENCES 1. Mokdad AH, Serdula MK, Dietz WH, Bowman BA, Marks JS, Koplan JP. The spread of the obesity epidemic in the United States, 1991 1998. JAMA 1999;27:1519 22. 2. Institute of Medicine. Dietary reference intakes for energy, carbohydrates, fiber, fat, protein, and amino acids (macronutrients). Washington, DC: National Academy Press, 1997. 3. Sadurkis A, Kabir N, Wager J, Forsum E. Energy metabolism, body composition, and milk production in healthy Swedish women during lactation. Am J Clin Nutr 1988;48:44 9. 4. Butte NF, Garza C, Stuff JE, O Brian Smith E, Nichols BL. Effect of maternal diet and body composition on lactational performance. Am J Clin Nutr 1984;39:296 306. 5. Sohlstrom A, Forsum E. Changes in adipose tissue volume and distribution during reproduction in Swedish women as assessed by magnetic resonance imaging. Am J Clin Nutr 1995;61:287 95. 6. Haiek LN, Kramer MS, Ciampi A, Tirado R. Postpartum weight loss and infant feeding. J Am Board Fam Pract 2001;14:85 94. 7. Kramer FM, Stunkard AJ, Marshall KA, McKinney S, Liebschutz J. Breast-feeding reduces maternal lower-body fat. J Am Diet Assoc 1993; 93:429 33. 8. Dewey KG, Heinig MJ, Nommsen LA. Maternal weight-loss patterns during prolonged lactation. Am J Clin Nutr 1993;58:162 6. 9. Butte NF, Hopkinson JM. Body composition changes during lactation are highly variable among women. J Nutr 1998;128(suppl):381S 5S. 10. Brewer MM, Bates MR, Vannoy LP. Postpartum changes in maternal weight and body fat depots in lactating vs nonlactating women. Am J Clin Nutr 1989;49:259 65. 11. Motil KJ, Sheng H, Kertz BL, Montandon CM, Ellis KJ. Lean body mass of well-nourished women is preserved during lactation. Am J Clin Nutr 1998;67:292 300. 12. Chou T, Chan GM, Moyer-Mileur L. Postpartum body composition changes in lactating and non-lactating primiparas. Nutrition 1999;15: 481 4. 13. Manning-Dalton C, Allen LH. The effects of lactation on energy and protein consumption, postpartum weight change and body composition of well nourished North American women. Nutr Res 1983;3:293 308. 14. Grattan DR. The actions of prolactin in the brain during pregnancy and lactation. Prog Brain Res 2001;133:153 71. 15. Parikh SJ, Yanovski JA. intake and adiposity. Am J Clin Nutr 2003;77:281 7. 16. Zemel MB. Regulation of adiposity and obesity risk by dietary calcium: mechanisms and implications. J Am Coll Nutr 2002;21:146S 51S. 17. Kalkwarf HJ, Specker BL, Bianchi DC, Ranz J, Ho M. The effect of calcium supplementation on bone density during lactation and after weaning. N Engl J Med 1997;337:523 8. 18. Sallis JF, Haskell WL, Wood PD. Physical activity assessment methodology in the Five-City Project. Am J Epidemiol 1985;121:91 106. 19. Butte NF, Hopkinson JM, Mehta N, Moon JK, O Brian Smith E. Adjustments in energy expenditure and substrate utilization during late pregnancy and lactation. Am J Clin Nutr 1999;69:299 307. 20. Davies KM, Heaney RP, Recker RR, et al. intake and body weight. J Clin Endocrinol Metab 2000;85:4635 8. 21. Kalkwarf HJ, Specker BL, Ho M. Effects of calcium supplementation on calcium homeostasis and bone turnover in lactating women. J Clin Endocrinol Metab 1999;84:464 70.