Total body bone mineral content and density during weight ... - Nature

European Journal of Clinical Nutrition (2010) 64, 392–399

& 2010 Macmillan Publishers Limited All rights reserved 0954-3007/10 $32.00 www.nature.com/ejcn

ORIGINAL ARTICLE

Total body bone mineral content and density during weight loss and maintenance on a low- or recommended-dairy weight-maintenance diet in obese men and women PS Hinton1, R Scott Rector2, JE Donnelly3, BK Smith3 and B Bailey4 1 Department of Nutritional Sciences, University of Missouri, Columbia, MO, USA; 2School of Medicine—Gastroenterology (RSR), University of Missouri, Columbia, MO, USA; 3Energy Balance Laboratory & The Center for Physical Activity, Nutrition & Weight Management, Schiefelbusch Life Span Institute, University of Kansas, Lawrence, KS, USA and 4Brigham Young University, Provo, UT, USA

Background: Diets high in either dairy or calcium during moderate weight reduction both prevent loss of bone mineral density (BMD) and suppress bone turnover. The purpose of this study was to determine whether recommended dairy and calcium intakes during weight maintenance favorably affect total body BMD (TBBMD) and bone mineral content (TBBMC) in obese adults. Methods: Obese men (n ¼ 49) and women (n ¼ 64), aged 40.8±0.6 years, underwent 12 weeks of moderate energy restriction (B1200 kcal/day) followed by 24 weeks on either a low or recommended dairy weight maintenance diet. The TBBMC and TBBMD values were measured using dual energy X-ray absorptiometry at baseline, 12, 24 and 36 weeks. Concentrations of calcium, intact parathyroid hormone (iPTH), 25OH and 1,25 (OH)2 vitamin D in plasma were also measured. Data were analyzed using a two-factor repeated measures analysis of variance. Results: After weight loss, women exhibited a small, but statistically significant, increase in TBBMC (1.17±0.57%), whereas TBBMD increased in the men (1.34±0.28%). The iPTH concentration decreased significantly in all subjects. Despite significantly greater intakes of calcium, vitamin D and protein compared with the recommended dairy diet, there were no treatment-related differences in outcome variables after 24 weeks of weight maintenance. The TBBMC remained unchanged in women during weight stabilization; both TBBMC and TBBMD decreased in men (1.59±0.51% and 0.70±0.25%, respectively). Conclusions: In summary, results of this study do not provide convincing evidence that moderate weight loss through energy restriction and mild exercise reduces TBBMC in obese men and women. Similarly, a weight-maintenance diet providing the recommended daily servings of dairy does not seem to affect changes in BMC after weight loss.

European Journal of Clinical Nutrition (2010) 64, 392–399; doi:10.1038/ejcn.2009.156; published online 13 January 2010 Keywords: weight loss; bone mineral density; parathyroid hormone; vitamin D

Introduction Weight loss improves metabolic fitness and reduces morbidity and mortality associated with overweight and obesity

Correspondence: Dr PS Hinton, Department of Nutritional Sciences, University of Missouri-Columbia, 106 McKee, Columbia, MO 65211, USA. E-mail: [email protected] Received 28 June 2009; revised 24 October 2009; accepted 6 November 2009; published online 13 January 2010

(Fernandez, 2007). However, even moderate weight reduction (B10%) in overweight or obese individuals both decreases bone mineral density (BMD) (Avenell et al., 1994; Pritchard et al., 1996; Ricci et al., 1998, 2001; Van Loan et al., 1998; Fogelholm et al., 2001; Jensen et al., 2001; Riedt et al., 2005; Villareal et al., 2006, 2008) and accelerates bone turnover, as assessed by serum biomarkers of bone formation and resorption (Ricci et al., 1998, 2001; Salamone et al., 1999; Villareal et al., 2006; Holecki et al., 2007). Although these relatively short-term weight loss intervention studies, which are typically 3–6-month long, have demonstrated significant

Bone mineral during weight loss and maintenance PS Hinton et al

393 reductions in total body or regional BMD, long-term consequences of weight loss on bone mass, bone turnover and subsequent fracture remain relatively unknown. However, results from the few studies that included a post-weight loss follow-up suggest that alterations in BMD and bone turnover may persist during weight maintenance (Avenell et al., 1994; Fogelholm et al., 2001; Jensen et al., 2001; Hinton et al., 2010; Villareal et al., 2008). Consistent with these findings, observational data indicate that both weight loss and repeated cycles of weight loss and regain increase fracture risk in both men and women (Langlois et al., 1998, 2001; Meyer et al., 1998; Bacon et al., 2004). Studies in overweight pre- and post-menopausal women have demonstrated that high calcium intakes (1.5–1.8 g/day) both prevent the reduction in bone mass associated with moderate weight loss, and suppress bone turnover and parathyroid hormone (PTH) secretion (Ricci et al., 1998; Jensen et al., 2001; Shapses et al., 2001). Weight-loss diets that are high in dairy and, therefore, in calcium, vitamin D and protein content also have been shown to preserve BMD and suppress bone turnover after weight reduction (Bowen et al., 2004; Thorpe et al., 2008). Although diets high in calcium content seem to protect bone during moderate weight loss, potential benefits of adequate dietary calcium intake after weight loss—normalization of bone turnover and bone mass accretion—have not been investigated. Thus, the purpose of this study was twofold: (1) to evaluate the effects of moderate weight loss followed by weight maintenance on total body BMD (TBBMD) and content (TBBMC) in obese middle-aged men and women using the repeated measures study design; and (2) to determine whether a weight-maintenance diet adequate in dairy and, therefore, also in calcium, content favorably affects TBBMD and TBBMC after weight loss compared with a low dairy weight maintenance diet. We hypothesized that moderate weight loss would result in significant reductions in TBBMD and TBBMC, which would persist in both women and men participants. We also hypothesized that high dietary dairy and calcium intake during 24 weeks of weight maintenance would favorably affect TBBMD and TBBMC compared with inadequate dairy intake in both men and women.

Materials and methods Study participants Sedentary, obese (body mass index: 30–40 kg/m2) men and women (non-pregnant and non-lactating), aged 19–50 years, were eligible to participate in this study. Individuals who smoked, were taking calcium supplements, were lactose intolerant, had lost 44.5 kg in the prior 3 months, were diabetic or taking medication that alters metabolism (for example, thyroid hormone and b-blockers) were excluded from the study. For practical reasons, participants were recruited in cohorts of 30–40 participants between May 2004 and Feb 2006, with a new cohort recruited approximately every 3 months.

A total of 183 participants agreed to participate in this study and 127 completed the entire 9-month study. The primary reasons for not completing the study were scheduling conflicts (n ¼ 21, 37.5%), attendance (n ¼ 13, 23.2%), participant withdrawal (n ¼ 13, 23.2%), less than 10% weight loss (n ¼ 3, 5.4%) and other (n ¼ 6; 10.7%). In addition, post-menopausal women (n ¼ 14) were excluded from this post hoc analysis that examined bone-related outcomes. The Human Subjects Committees at the University of Kansas and the University of Missouri-Columbia approved this study. All procedures were in accordance with the Declaration of Helsinki 1975, as revised in 1983. Before enrollment, participants provided written informed consent.

Experimental design This study was part of a larger randomized, clinical trial that evaluated the effects of moderate weight loss and subsequent weight stabilization on either a low- or recommended-dairy weight-maintenance diet on body weight, body composition and metabolic fitness (Zemel et al., 2008). All participants were prescribed a reduced energy diet designed to result in, at least, a 10% reduction in initial body weight during the 12-week weight-loss phase of the study. At baseline, participants were randomized to either a low-dairy (p1 serving per day) or recommended-dairy (X3 servings per day; Fulgoni et al., 2004) weight-maintenance diet for the 24-week weight-stabilization period; both participants and study staff were blinded to the assignment during the weight-loss phase of the study, that is, during weeks 0–12, to minimize bias. Each participant attended a 90-min weight-management clinic once a week for the duration of the 36-week study. These clinics provided instruction in three areas: (1) nutrition, (2) behavior/lifestyle and (3) physical activity. Each participant received a progressive, moderate-intensity, lifestyle physical activity program designed to provide 150 min of planned physical activity per week by week 13 (the start of the weightmaintenance phase of the study). Walking was the mode of exercise promoted during this weight-management study.

Intervention Weight loss. During the initial 12-week weight-loss phase, participants were prescribed a daily energy intake of 5.0 MJ/day (1200 kcal/day) and were provided prepackaged meals and weight-loss shakes. In addition to the diet prescription, participants began the exercise program described above. Participants who did not lose a minimum of 10% of their initial body weight by week 13 were dropped from the study. Three participants failed to lose 10% of their initial body weight and were excluded from the weight-maintenance phase of the study. Weight maintenance. After weight loss, energy intake was prescribed to maintain the reduced body weight. Participants consumed a low-fat weight-maintenance diet that emphasized adequate intake of fruits and vegetables and European Journal of Clinical Nutrition


394 included either one or fewer servings of dairy (low dairy) or three or more servings of dairy (recommended dairy). Participants on the low-dairy diet were instructed to limit their intake of high-calcium foods and to avoid calciumfortified foods and calcium supplements. Participants continued to exercise (weekly target of 150 min/week) and attend weekly weight-management clinics during the weightmaintenance phase of the study. Compliance. Compliance with the diet and exercise prescription was monitored through self-reported written records that were reviewed by study personnel at the weekly clinics. Measures of compliance with the weight-loss and weight-maintenance diets were recorded by the participants daily and data were collected from participants once a week. Measures of compliance with the diets included number of shakes, pre-packaged meals and servings of fruits and vegetables consumed per day. In addition to the dietary compliance data, participants completed a prospective, written 3-day food record at baseline, weight loss and weight maintenance. The 3-day food records were reviewed by a registered dietitian upon submission and any clarification needed was provided by the participants. The 3-day food records were analyzed for nutrient content and using the Nutrition Data System for Research (2006).

Outcome measures Body weight (BW) was measured weekly on a digital scale (Befour, Model # PS6600, Saukville, WI, USA) for the duration of the study and was used to calculated body mass index. The TBBMD and TBBMC values were measured at baseline, 12, 24 and 36 weeks using dual energy X-ray absorptiometry (DXA; Lunar DPX-IQ Software version 4.6b, GE Medical Systems, Madison, WI, USA). The coefficient of variations for TBBMD and TBBMC are o1%. The blood also was collected between 0700 and 1000 hours after a 12-h overnight fast at baseline, 12, 24 and 36 weeks. The concentration of glucose, insulin, intact PTH, 25OH vitamin D, 1,25(OH)2 vitamin D and calcium in plasma was measured at an independent clinical laboratory (LabOne, Lenexa, KS, USA). The homeostasis model assessment of insulin resistance was estimated using fasting glucose and insulin concentrations, as previously described (Matthews et al., 1985). As both insulin resistance and hyperglycemia negatively affect bone, we evaluated these potential confounders in our population at risk for the metabolic syndrome (Hofbauer et al., 2007).

Statistics A repeated measures three-factor analysis of variance (RM ANOVA) was used to test for significant main effects of time (baseline, 12 , 24 , 36 weeks), gender and diet ( low dairy vs recommended dairy) and for significant interactions. As there were significant TIME GENDER interactions for European Journal of Clinical Nutrition

several variables (body weight, percent body fat, fat mass, serum 25OH vitamin D and TBBMC), separate two-factor RM ANOVA values (TIME DIET) were performed for males (n ¼ 49) and females (n ¼ 64). When appropriate, that is, identification of a significant main effect of time or significant interaction between time and dietary treatment, post hoc pairwise comparisons were performed using the least significant difference test. As there is seasonal variability in 25OH vitamin D concentrations (Bolland et al., 2007), we attempted to control for this source of variability in the statistical analysis. First, a one-way ANOVA was used to compare 25OH vitamin D concentrations in participants enrolled during different seasons. Baseline 25OH vitamin D concentrations were lowest in subjects enrolled in the spring (51.3±5.4 ng/ml, n ¼ 14) and highest in participants enrolled in the fall (101.6±7.6 ng/ml, n ¼ 24), whereas subjects enrolled in winter (82.4±4.9 ng/ml, n ¼ 29) and summer (79.6±4.4 ng/ml, n ¼ 46) had intermediate values, as expected (Bolland et al., 2007). Therefore, in an attempt to control for seasonal variability in serum 25OH vitamin D concentrations, season was included as a covariate in the RM ANOVA for vitamin D; there was a significant SEASON TIME interaction, indicating that change in 25OH vitamin D level over time was dependent on season of enrollment. Data are presented as means±S.E.M. P-values o0.05 were considered statistically significant.

Results Results of the two-factor RM ANOVA revealed significant main effects for time, and the posthoc pairwise comparisons of mean values (±S.E.M.) among time points for the anthropometric variables and blood biochemistry are shown in Tables 1 and 2. There were no significant TIME DIET interactions for any of the anthropometric or blood biochemistry variables for either men or women; therefore, no post hoc tests were performed. Results for both women and men participants are shown by dietary treatment in Supplementary Tables 1a and b and 2a and b. However, because of differences in the nutrient content of weightmaintenance diets, there were significant TIME DIET interactions for intake of some nutrients. Results of the post hoc pairwise comparisons of nutrient intake (means±S.E.M.) for women and men are shown in Tables 3 and 4, respectively.

Weight loss A total of 64 obese premenopausal women, aged 40.6±0.8 years, and 49 obese men, aged 41.0±1.0 years, completed this 36-week weight-management study. At baseline, there were no significant differences between the low and recommended dairy groups in anthropometric or serum biochemistry variables in either women or men participants (Supplementary Tables 1a and b and 2a and b). The targeted


395 Table 1 Anthropometrics, total body BMC and BMD at baseline, and after 12 weeks of weight loss and 24 weeks of weight maintenance in obese women and men Women (n ¼ 64) Intervention Time point (week) BW (kg) BMI (kg/m2) FM (kg) LBM (kg) BF (%) BMD (g/cm2) BMC (g)

Weight loss 0

Men (n ¼ 49)

Weight maintenance 12

24

Weight loss

36

0

Weight maintenance 12

24

36

96.6±1.6a 85.6±1.5b 83.9±1.5c 84.5±1.6 b 108.8±1.7a 94.5±1.6b,c 93.7±1.8c 95.1±1.9b 34.8±0.4a 30.9±0.4b 30.2±0.4c 30.4±0.4 b 33.8±0.4a 29.4±0.4b,c 29.1±0.4c 29.5±0.5b 47.3±1.1a 38.3±1.1b 35.8±1.1d 36.5±1.2c 38.5±1.0a 27.2±1.1b 25.5±1.1c 26.7±1.2b 44.3±0.6a 43.1±0.6c 43.9±0.6b 43.7±0.6 b 62.3±0.8a 61.3±0.8b 62.4±0.8a 62.0±0.8a 51.3±0.6a 46.6±0.7b 44.5±0.7d 45.0±0.7c 38.0± 0.6a 30.4±0.9b 28.6±0.9c 29.6±0.9b 1.277±0.012b 1.286±0.010b 1.286±0.010b 1.293±0.011a 1.341±0.012c 1.358±0.013a 1.353±0.013a,b 1.349±0.013b 2724±44b 2764±45a 2751±43a,b 2740±44a,b 3402±49a 3396±57a 3374±57a 3343±57b

Abbreviations: BMC, bone mineral content; BMD, bone mineral density; BW, body weight; BMI, body mass index; FM, fat mass; LBM, lean body mass; BF (%), percent body fat. Data are means±S.E.M. There were no significant effects of the weight maintenance diet on outcome measures, that is, no significant TIME DIET interactions; thus, only the significant main effect of time is shown. Within each gender, means with different letter superscripts are significantly different by post hoc pairwise comparison (LSD), Po0.05.

Table 2 Serum biochemistry at baseline and after 12 weeks of weight loss and 24 weeks of weight maintenance in obese women and men Women (n ¼ 64) Intervention Time point (week) Calcium (mg/dl) 25OH D (ng/ml) 1,25OH D (pg/ml) iPTH (pg/ml) Glucose (mg/dl) Insulin (mU/ml) HOMA

Weight loss

Men (n ¼ 49)

Weight maintenance

Weight loss

Weight maintenance

0

12

24

36

0

12

24

36

9.13±0.03b,c 84.8±4.1a 41.9±1.6 40.1±2.4a 89.0±1.1a 8.53±0.64a 1.92±0.16a

9.22±0.04a 88.9±6.6a 44.9±2.1 33.4±1.7b 86.9±0.8b 5.54±0.35c 1.21±0.08b

9.07±0.04c 58.7±5.3b 45.8±2.0 34.7±2.1b 87.2±1.0a,b 6.22±0.44b 1.36±0.11b

9.15±0.04b 48.2±3.2c 42.6±1.7 37.5±2.4a,b 86.2±0.8b 5.63±0.39b,c 1.22±0.09b

9.38±0.04 80.2±4.4a 41.5±1.6b 41.0±2.4a 96.2±4.7a 10.2±1.0a 2.47±0.28a

9.37±0.05 84.2±6.7a 50.5±2.1a 30.4±1.9c 86.1±1.2b 5.7±0.8c 1.22±0.18c

9.34±0.05 77.1±6.8a 49.5±2.3a 37.0±2.4b 90.1±1.0a 7.3±0.9b 1.64±0.21b

9.39±0.04 65.6±4.3b 45.8±2.0a,b 38.1±2.2a,b 89.7±1.3a 6.6±0.8b,c 1.50±0.18b,c

Reference

8.5–10.2 X30 X45 10–70 90–130 5–20 o2.0

Abbreviations: ANOVA, analysis of variance; HOMA, homeostasis model assessment; iPTH, intact PTH; RM, repeated measures. Data are means±S.E.M. There were no significant effects of the weight maintenance diet on outcome measures, that is, no significant TIME DIET interactions; thus, only the significant main effect of time is presented. Within each gender, means with different letter superscripts are significantly different by pairwise post hoc comparison (LSD), Po0.05. 25OH vitamin D data were adjusted for season in the RM two-way ANOVA.

energy intake of 5.0 MJ/day during the weight-loss phase of the study was achieved in both female (Table 3) and male (Table 4) participants, who achieved a 10% reduction in initial body weight by week 13 and were allowed to continue the weight-maintenance phase of the study. On average, participants engaged in 137±4 minutes of physical activity per week during the weight-loss phase. Calcium intake during weight loss was less than the adequate intake for adults, aged 19–50 years, of 1000 mg/day (Food Nutrition Board IoM (1997); men: 867±17 mg/day; women: 900±13 mg/day), and was significantly greater than calcium intake at baseline (men: 755±58 mg/day; women: 717±42 mg/day; Po0.05). During the initial 12-week weight-loss phase of the study, women lost an average of 11.3±0.4% of initial body weight and men lost 13.0±0.6% (Table 1). The women exhibited a statistically significant increase in TBBMC (1.17±0.57%), but TBBMD remained unchanged (0.33±0.20%), after 12 weeks of weight reduction (Table 1). By contrast, in men, TBBMD

increased significantly (1.34±0.28%), but TBBMC did not (0.12±0.69%), as shown in Table 1. Measurement and reporting of bone area is useful in determining whether changes in BMC or BMD are artifacts of DXA. Unfortunately, as is the case in most weight-loss studies reported to date, bone area of each participant was not recorded in this study. Thus, the best approximation of bone area under the circumstances is to calculate area from the mean BMC and BMD at each time point. Taking this approach, using the data in Table 1, changes in bone area with weight loss were small in both women (2133 vs 2149 cm2) and men (2537 vs 2501 cm2). The intact PTH concentration decreased significantly in both genders during weight loss, whereas 25OH level did not change from baseline to post-weight loss (Table 2). As expected, serum insulin and glucose concentrations decreased after weight loss, as did homeostasis model assessment (Table 2). Inclusion of glucose, insulin or homeostasis model assessment, as a covariate, in RM ANOVA of TBBMD and TBBMC did not alter the results. European Journal of Clinical Nutrition


396 Table 3 Nutrient intakes at baseline and after 12 weeks of weight loss and 24 weeks of weight maintenance in obese women Low dairy (n ¼ 28) Intervention

Weight loss

Recommended dairy (n ¼ 36)

Weight maintenance

Weight loss

Weight maintenance

Time point (week)

0

12

36

0

12

36

Energy (MJ) Protein (g) Fat (g) Carbohydrate (g) Calcium (mg) Vitamin D (mg)

8.2±0.5a 76±4 83±7a 228±12a 679±64b 4.2±0.6b

5.2±0.1c 80±2 19±1c 198±6b 857±19a 6.6±0.3a

6.8±0.2b 75±3 50±3b 220±10b 587±45b 2.9±0.4 b

8.4±0.4a 80±4b 86±6a 228±12a,b 745±55c 3.6±0.5b

5.3±0.1b 81±2b 19±1c 205±6b 933±16b,* 6.2±0.3a

7.6±0.2a,* 92±2a,* 54±3b 250±9a,* 1330±39a,* 7.1±0.4a,*

Data are means±S.E.M. There was a significant TIME DIET interaction for protein, calcium and vitamin D; therefore, post hoc pairwise comparisons were performed. Within each dietary treatment, means with different letter superscripts are significantly different, Po0.05.*, significantly different than low dairy at respective time point.

Table 4 Nutrient intakes at baseline and after 12 weeks of weight loss and 24 weeks of weight maintenance in obese men Low dairy (n ¼ 28) Intervention

Weight loss

Recommended dairy (n ¼ 21) Weight maintenance

Weight loss

Weight maintenance

Time point (week)

0

12

36

0

12

36

Energy (MJ) Protein (g) Fat (g) Carbohydrate (g) Calcium (mg) Vitamin D (mg)

9.5±0.5a 98±4a 94±7a 250±18 771±78a,b 4.1±0.7b

6.1±.02c 78±3b 18±1c 209±9 847±22a 5.7±0.5a

7.3±0.3b 88±4a,b 54±4b 228±14 624±36b 3.8±0.5b

9.9±0.6a 92±5b 102±7a 262±20b 734±88b 4.9±0.8c

5.7±0.2b 82±3b 22±2c* 221±10b 913±25a,b 7.2±0.5b

9.3±0.4a,* 111±5a,* 66±4b,* 303±16a,* 1448±41a,* 9.7±0.6a,*

Data are means±S.E.M. Significant TIME DIET interaction for energy, protein, carbohydrate, calcium and vitamin D. Within each dietary treatment, means with different letter superscripts are significantly different, Po0.05.*, significantly different than low dairy at respective time point.

Weight maintenance During weight maintenance, energy intake increased significantly in both dietary treatments, primarily because of increased fat consumption in both women and men, as shown in Tables 3 and 4, respectively. As expected, protein, calcium and vitamin D intakes were significantly greater for participants on the recommended dairy diet compared with low dairy diet (Tables 3 and 4). Similarly, protein intake relative to body weight also was greater with the recommended dairy diet for both men (recommended dairy ¼ 1.19±0.06 g/kg BW; low dairy ¼ 0.93±0.04 g/kg BW) and women (recommended dairy ¼ 1.07±0.02 g/kg BW; low dairy ¼ 0.94±0.02 g/kg BW). Participants continued to exercise during weight maintenance and there were no differences between dietary treatments in either women (recommended dairy ¼ 129±4; low dairy ¼ 132±3 min/week) or men (recommended dairy ¼ 141±11; low dairy ¼ 151± 10 min/week). As mentioned above, there were no statistically significant effects of the weight-maintenance diet on outcome measures, that is, no significant TIME DIET interactions. During weight maintenance, average weight change did not differ between low and recommended dairy diets (data not shown) and was 1.4±0.7% for women and 0.4±0.7% for men. Thus, body weight at week 36 was similar to post-weight loss weight in European Journal of Clinical Nutrition

both women and men (Table 1). In women, TBBMC after weight maintenance was similar to TBBMC at baseline (Table 1). The TBBMD increased and was significantly greater than baseline; although statistically significant, the percent increase was small—0.52±0.20%. By contrast, in men, both TBBMD and TBBMC decreased significantly during weight maintenance, and TBBMC was less than the mean at baseline (Table 1). Nevertheless, the percent decrease in TBBMD during weight maintenance was only 0.70±0.25%, whereas TBBMC decreased by 1.59±0.51%. These small changes in TBBMD might have been due to the small changes in bone area observed from week 12 to week 36 in both women (2149 vs 2119 cm2) and men (2501 vs 2478 cm2). Inclusion of glucose, insulin or homeostasis model assessment as a covariate in RM ANOVA of TBBMD and TBBMC did not alter the results. Regardless of dietary treatment, serum 25OH vitamin D concentration decreased significantly during weight maintenance in both women and men, whereas the intact PTH concentration tended to increase (Table 2).

Discussion In this study, we observed a statistically significant increase in bone mineral after moderate weight loss in obese,


397 middle-aged adults—in women, the increase in TBBMC was statistically significant, whereas in men, TBBMD increased. In addition, a diet providing the recommended servings of dairy had no effect on bone mineral or the calcium–vitamin D–PTH axis during 24 weeks of weight maintenance compared with a diet inadequate in dairy. The effect of weight loss on bone mineral remains controversial. Although weight loss in postmenopausal women apparently results in loss of bone mass (Avenell et al., 1994; Ricci et al., 1998, 2001; Riedt et al., 2005; Villareal et al., 2006), findings in premenopausal women (Salamone et al., 1999; Fogelholm et al., 2001; Shapses et al., 2001; Park et al., 2007; Riedt et al., 2007) and in middle-aged men (Pritchard et al., 1996; Villareal et al., 2006) are inconsistent. The discrepant results may be explained, in part, by differences in skeletal site examined, baseline body weight, rate or magnitude of weight loss, calcium or protein intake, physical activity and duration of post-weight loss follow-up. However, an underappreciated and often unrecognized methodological limitation of these studies, which may contribute to the contradictory findings, is the effect of weight loss on the assessment of bone mineral using DXA. For example, several weight-loss studies have reported discordant changes in total body BMC and BMD, that is, BMC decreases and BMD increases (Van Loan et al., 1998; Riedt et al., 2005), or as observed in this study, BMC changes without a corresponding statistically significant change in BMD or vice versa (Fogelholm et al., 2001; Ricci et al., 2001; Riedt et al., 2005). Areal BMD is calculated from BMC and bone area. In adults, BMD and BMC should change in parallel, as bone area is not expected to change appreciably over the short duration typical of weight-loss intervention studies. Thus, discordant BMD and BMC results are biologically implausible, and are probably artifacts of DXA, which result from errors in determination of bone area. Tothill et al. (1999) and Tothill (2005) evaluated the effect of changes in fat mass or distribution on measurement of BMC, BMD and bone area by placement of known quantities of lard on human volunteers or a phantom with BMC, BMD and bone area approximately equivalent to that of an adult male (Tothill et al., 1999). Addition of B6 kg of lard resulted in erroneous increases in BMC and bone area as measured using DXA scanners of various manufacturers, including the Lunar DPX used in this study. Although the effect of the increased fat mass on BMD was smaller using the Lunar DXA compared with the Hologic and Norland instruments, others have reported significant reductions in BMD with no change in BMC as a result of moderate weight loss using the Lunar DPX (Ricci et al., 1998; Van Loan et al., 1998). Despite the fact that DXA has been used in numerous studies to examine the effects of weight reduction on bone mineral, changes in BMC or BMD, which are associated with change in body weight or composition must be interpreted with caution. Although the percent changes in TBBMC and TBBMD observed during 12 weeks of weight loss and 24

weeks of weight maintenance are within the range of values previously reported for premenopausal women (Ramsdale and Bassey, 1994; Van Loan et al., 1998) and men (Pritchard et al., 1996), the strength of the data is limited by both the small percent changes (that is, approximately equivalent to the coefficients of variation for TBBMD and TBBMC) and by the absence of consistent parallel changes in TBBMC and TBBMD, which might have been due to small variations in bone area. Thus, results of this study do not provide convincing evidence that moderate weight loss followed by weight maintenance results in loss of bone mineral in obese men and women. Clearly, the recommended dairy weight maintenance diet had no effect on total body bone mineral during the 24-week weight-stabilization period, despite significantly greater intakes of protein, calcium and vitamin D by participants on the recommended dairy diet compared with the low dairy diet (Tables 3 and 4). There are several possible reasons because of which we did not observe an effect of increased dairy consumption on bone mineral. First, to minimize radiation exposure, BMC and BMD of the hip, spine and wrist were not evaluated; therefore, it is possible that sitespecific alterations in BMC occurred, but were not detected. On the basis of the time required to detect changes in bone mass, that is, several remodeling cycles, it is possible that differences between the low- and recommended dairy weight-maintenance diets may have been detected after a longer follow-up interval. Alternatively, compensatory changes in calcium absorption and excretion in response to either altered calcium intake (Rafferty and Heaney, 2008) or weight loss per se (Cifuentes et al., 2004) might have reduced differences between the low and recommended dairy diets. Similar to the results of this study, Bowen et al. (2004) reported that a high dairy-protein diet (2400 mg Ca2 þ /day) had no effect on TBBMD during 12 weeks of energy restriction and 4 weeks of energy balance in obese men and women compared with a mixed protein diet (500 mg Ca2 þ /day). By contrast, Thorpe et al., (2008) reported that a high-protein diet (1000 mg Ca2 þ /day) positively affected TBBMD during weight stabilization compared with a high-carbohydrate diet (700 mg Ca2 þ /day) in overweight men and women. Clearly, more studies are needed to identify independent effects of dietary calcium and protein on bone during weight loss and maintenance. In addition to the lack of effect on bone mineral, there was no difference in plasma 25OH D concentrations between the recommended and low dairy weight maintenance diets. Supplementation of the diet with 400–480 IU vitamin D in food (skim milk) has been shown to increase serum 25OH D concentrations by 16 nmol/l in adult men and women (Cranney et al., 2007). That is, for each additional microgram of vitamin D consumed in the diet, the concentration of 25OH D increased by 0.583 ng/ml. Thus, on the basis of the differences in vitamin D intake between the low and recommended dairy diets, one would expect the magnitude of the diet effect on 25OH D concentrations to be B3 ng/ml, European Journal of Clinical Nutrition


398 which is much less than that observed in both dietary treatments. Similarly, the expected effect of changes in fat mass on 25OH D concentrations in men and women ( þ 0.0521 and þ 0.208 ng/ml per kg fat lost, respectively; (Bolland et al., 2007)) was markedly less than that observed. The most likely explanation is that seasonal variation has a much greater effect on plasma 25OH D concentrations than either weight loss (or gain) or consumption of a diet adequate in dairy. In a study on adult men and women living at a distance from the equator similar to that of the participants in this study (that is, 38.97 1N), the difference in peak 25OH D concentrations between seasons averaged 15.6 ng/ml for men and 8.4 ng/ml for women (Bolland et al., 2007). Thus, the potential magnitude of the seasonal effect is considerably greater than that of diet or fat mass. In fact, because the majority of our participants were enrolled when 25OH D concentration was elevated due to greater ultraviolet B exposure, 25OH D should have decreased to due seasonality. As a result, it was not possible to detect changes that were due to weight loss or altered vitamin D intake, and the 25OH D and intact PTH results are thus difficult to interpret. In summary, results of this study do not provide convincing evidence that moderate weight loss through energy restriction and mild exercise reduces total body bone mineral content in obese men and women. Similarly, a weightmaintenance diet providing the recommended daily servings of dairy does not seem to affect changes in bone mineral content after weight loss.

Conflict of interest Dr JE Donnelly received a grant from the National Dairy Council (USA). Other authors declare no conflict of interest.

Acknowledgements This study was supported by the Department of Nutritional Sciences, University of Missouri (PSH) and National Dairy Council (JED)

References Avenell A, Richmond PR, Lean ME, Reid DM (1994). Bone loss associated with a high fibre weight reduction diet in postmenopausal women. Eur J Clin Nutr 48, 561–566. Bacon L, Stern JS, Keim NL, Van Loan MD (2004). Low bone mass in premenopausal chronic dieting obese women. Eur J Clin Nutr 58, 966–971. Bolland MJ, Grey AB, Ames RW, Mason BH, Horne AM, Gamble GD et al. (2007). The effects of seasonal variation of 25-hydroxyvitamin D and fat mass on a diagnosis of vitamin D sufficiency. Am J Clin Nutr 86, 959–964.

European Journal of Clinical Nutrition

Bowen J, Noakes M, Clifton PM (2004). A high dairy protein, highcalcium diet minimizes bone turnover in overweight adults during weight loss. J Nutr 134, 568–573. Cifuentes M, Riedt CS, Brolin RE, Field MP, Sherrell RM, Shapses SA (2004). Weight loss and calcium intake influence calcium absorption in overweight postmenopausal women. Am J Clin Nutr 80, 123–130. Cranney A, Horsely T, O’Donnell S, Weiler HA, Puil L, Ooi DS et al. (2007). Effectiveness and safety of vitamin D in relation to bone health. Evidence report/technology assessment no. 158 (prepared by the University of Ottawa Evidence-based Practice Center (UO-EPC) under Contract No. 290-02-0021. Agency for Healthcare Research and Quality: Rockville, MD. Fernandez ML (2007). The metabolic syndrome. Nutr Rev 65, S30–S34. Fogelholm GM, Sievanen HT, Kukkonen-Harjula TK, Pasanen ME (2001). Bone mineral density during reduction, maintenance and regain of body weight in premenopausal, obese women. Osteoporos Int 12, 199–206. Food Nutrition Board IoM (1997). DRI Dietary Reference Intakes for calcium, magnesium, vitamin D, and fluoride. National Academy Press: Washington, DC. Fulgoni 3rd VL, Huth PJ, DiRienzo DB, Miller GD (2004). Determination of the optimal number of dairy servings to ensure a low prevalence of inadequate calcium intake in Americans. J Am Coll Nutr 23, 651–659. Hinton PS, LeCheminant JS, Smith BK, Rector RS, Donnelly JE (2010). Weight loss-induced alterations in serum markers of bone turnover persist during weight maintenance in obese men and women. J Am Coll Nutr (In press). Hofbauer LC, Brueck CC, Singh SK, Dobnig H (2007). Osteoporosis in patients with diabetes mellitus. J Bone Miner Res 22, 1317–1328. Holecki M, Zahorska-Markiewicz B, Janowska J, Nieszporek T, Wojaczynska-Stanek K, Zak-Golab A et al. (2007). The influence of weight loss on serum osteoprotegerin concentration in obese perimenopausal women. Obesity (Silver Spring) 15, 1925–1929. Jensen LB, Kollerup G, Quaade F, Sorensen OH (2001). Bone minerals changes in obese women during a moderate weight loss with and without calcium supplementation. J Bone Miner Res 16, 141–147. Langlois JA, Mussolino ME, Visser M, Looker AC, Harris T, Madans J (2001). Weight loss from maximum body weight among middleaged and older white women and the risk of hip fracture: the NHANES I epidemiologic follow-up study. Osteoporos Int 12, 763–768. Langlois JA, Visser M, Davidovic LS, Maggi S, Li G, Harris TB (1998). Hip fracture risk in older white men is associated with change in body weight from age 50 years to old age. Arch Intern Med 158, 990–996. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC (1985). Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 28, 412–419. Meyer HE, Tverdal A, Selmer R (1998). Weight variability, weight change and the incidence of hip fracture: a prospective study of 39 000 middle-aged Norwegians. Osteoporos Int 8, 373–378. Park HA, Lee JS, Kuller LH, Cauley JA (2007). Effects of weight control during the menopausal transition on bone mineral density. J Clin Endocrinol Metab 92, 3809–3815. Pritchard JE, Nowson CA, Wark JD (1996). Bone loss accompanying diet-induced or exercise-induced weight loss: a randomised controlled study. Int J Obes Relat Metab Disord 20, 513–520. Rafferty K, Heaney RP (2008). Nutrient effects on the calcium economy: emphasizing the potassium controversy. J Nutr 138, 166S–171S. Ramsdale SJ, Bassey EJ (1994). Changes in bone mineral density associated with dietary-induced loss of body mass in young women. Clin Sci (Lond) 87, 343–348. Ricci TA, Chowdhury HA, Heymsfield SB, Stahl T, Pierson Jr RN, Shapses SA (1998). Calcium supplementation suppresses bone turnover during weight reduction in postmenopausal women. J Bone Miner Res 13, 1045–1050.


399 Ricci TA, Heymsfield SB, Pierson Jr RN, Stahl T, Chowdhury HA, Shapses SA (2001). Moderate energy restriction increases bone resorption in obese postmenopausal women. Am J Clin Nutr 73, 347–352. Riedt CS, Cifuentes M, Stahl T, Chowdhury HA, Schlussel Y, Shapses SA (2005). Overweight postmenopausal women lose bone with moderate weight reduction and 1 g/day calcium intake. J Bone Miner Res 20, 455–463. Riedt CS, Schlussel Y, von Thun N, Ambia-Sobhan H, Stahl T, Field MP et al. (2007). Premenopausal overweight women do not lose bone during moderate weight loss with adequate or higher calcium intake. Am J Clin Nutr 85, 972–980. Salamone LM, Cauley JA, Black DM, Simkin-Silverman L, Lang W, Gregg E et al. (1999). Effect of a lifestyle intervention on bone mineral density in premenopausal women: a randomized trial. Am J Clin Nutr 70, 97–103. Shapses SA, Von Thun NL, Heymsfield SB, Ricci TA, Ospina M, Pierson Jr RN et al. (2001). Bone turnover and density in obese premenopausal women during moderate weight loss and calcium supplementation. J Bone Miner Res 16, 1329–1336. Thorpe MP, Jacobson EH, Layman DK, He X, Kris-Etherton PM, Evans EM (2008). A diet high in protein, dairy, and calcium attenuates bone loss over twelve months of weight loss and maintenance relative to a conventional high-carbohydrate diet in adults. J Nutr 138, 1096–1100.

Tothill P (2005). Dual-energy X-ray absorptiometry measurements of total-body bone mineral during weight change. J Clin Densitom 8, 31–38. Tothill P, Laskey MA, Orphanidou CI, van Wijk M (1999). Anomalies in dual energy X-ray absorptiometry measurements of total-body bone mineral during weight change using Lunar, Hologic and Norland instruments. Br J Radiol 72, 661–669. Van Loan MD, Johnson HL, Barbieri TF (1998). Effect of weight loss on bone mineral content and bone mineral density in obese women. Am J Clin Nutr 67, 734–738. Villareal DT, Banks M, Sinacore DR, Siener C, Klein S (2006). Effect of weight loss and exercise on frailty in obese older adults. Arch Intern Med 166, 860–866. Villareal DT, Fontana L, Weiss EP, Racette SB, Steger-May K, Schechtman KB et al. (2006). Bone mineral density response to caloric restriction-induced weight loss or exercise-induced weight loss: a randomized controlled trial. Arch Intern Med 166, 2502–2510. Villareal DT, Shah K, Banks MR, Sinacore DR, Klein S (2008). Effect of weight loss and exercise therapy on bone metabolism and mass in obese older adults: a one-year randomized controlled trial. J Clin Endocrinol Metab 93, 2181–2187. Zemel MB, Donnelly JE, Smith BK, Sullivan DK, Richards J, Morgan-Hanusa D et al. (2008). Effects of dairy intake on weight maintenance. Nutr Metab (Lond) 5, 28.

Supplementary Information accompanies the paper on European Journal of Clinical Nutrition website (http://www.nature.com/ejcn)

European Journal of Clinical Nutrition