Does bone loss begin after weight loss ends? Results 2 years after weight loss or regain in postmenopausal women

Menopause: The Journal of The North American Menopause SocietyVol. 21, No. 5, pp. 000/000DOI: 10.1097/GME.0b013e3182a76fd5* 2013 by The North American Menopause Society

Does bone loss begin after weight loss ends? Results 2 years afterweight loss or regain in postmenopausal women

Nancy L. Von Thun, MS,1 Deeptha Sukumar, PhD,1 Steven B. Heymsfield, MD,2 and Sue A. Shapses, PhD1

AbstractObjective: Short-term weight loss is accompanied by bone loss in postmenopausal women. The longer-term

impact of weight loss on bone in reduced overweight/obese women compared with women who regained theirweight was examined in this study using a case-control design.

Methods: Postmenopausal women (N = 42; mean [SD] body mass index, 28.3 [2.8] kg/m2; mean [SD] age, 60.7[5.5] y) were recruited 2 years after the start of a 6-month weight loss trial; those who maintained their weight(weight loss maintainer [WL-M] group) were matched to a cohort of women who regained their weight (weight lossregainer [WL-R] group). Serum hormones and bone markers were measured in a subset. Bone mineral density(BMD) at the femoral neck, trochanter, spine, radius, and total body, and soft-tissue composition were taken atbaseline, 0.5 years, and 2 years.

Results: During weight loss, both groups lost 9.3% (3.4%) of body weight, with no significant difference be-tween the groups. After weight loss, weight change was j0.1% (2.7%) and 6.0% (3.3%) in the WL-M (n = 22)and WL-R (n = 20) groups, respectively. After 2 years, both groups lost BMD at the femoral neck and trochanter(P e 0.01), whereas only the WL-M group reduced BMD at the 1/3 radius (P G 0.001). There was greater BMD lossat the trochanter (j6.8% [5.7%]) and 1/3 radius (j4.5% [3.3%]) in the WL-M group compared with the WL-Rgroup after 2 years. Multiple linear regression showed that change in leg fat mass (but not trunk fat) contributed totrochanter BMD loss (P G 0.05).

Conclusions: After 2 years, there is no BMD recovery of weight reductionYinduced bone loss, irrespective ofweight regain. These data suggest that the period after weight loss may be an important point in time to prevent boneloss for those who maintain weight and those who regain weight.

Key Words: Bone Y Body composition Y Caloric restriction Y Postmenopausal Y Weight loss Y Weight cycle.

Obese and overweight people are strongly encouragedto lose 5% to 15% of their weight to reduce the risk ofcomorbid conditions. During the menopausal transi-

tion, women tend to gain weight and visceral adiposity; therefore,many of these women who were never overweight previously arenow encouraged to reduce weight to improve health. However,studies show that weight loss in even obese and overweightwomen results in bone mineral density (BMD) loss of approxi-mately 1% to 2% with 10% weight reduction and can partiallybe attenuated with certain interventions.1<4 Loss of bone due toweight reduction is more consistently shown in older individualscompared with younger individuals.4<7 In large retrospectivestudies, weight reduction whether weight loss was voluntary or

involuntaryVhas been associated with higher hip bone loss andfracture.8<10

Overweight and obese individuals who lose weight eithermaintain a reduced-obese state or regain lost weight primarilyas fat mass.11 Women who are restrained eaters, who go onlong-term diet, or who Bweight cycle[ (weight loss followed byregain) may be at greater risk for low bone mass or osteopo-rosis.10,12,13 A limited number of previous trials have examinedresponse to weight regain after weight reduction in a singlegroup of women to show that it leads to partial recoveryof bone at some anatomical sites.14<17 In this trial, we use acase-control design to examine whether the rate of bone lossdiffers in reduced-obese postmenopausal women who maintaintheir lost weight for 2 years compared with those who regaintheir lost weight.

METHODS

ParticipantsPostmenopausal women who successfully completed a 6-month

weight loss protocol in our laboratories were eligible for re-cruitment in this study. Participants were contacted approxi-mately 2 years after their initial inclusion in a 6-month weightreduction program reported previously1,2,18 or that was partof small unpublished pilot studies from 2002 to 2008. To be

Received May 4, 2013; revised and accepted July 18, 2013.

From the 1Department of Nutritional Sciences, Rutgers University, NewBrunswick, NJ; and 2Pennington Biomedical Research Center, BatonRouge, LA.

Funding/support: This study was supported by National Institutes ofHealth grant AG12161 and NJAES 0153866.

Financial disclosure/conflicts of interest: None reported.

Address correspondence to: Sue A. Shapses, PhD, Department of Nutri-tional Sciences, Rutgers University, 96 Lipman Drive, New Brunswick, NJ08901-8525. E-mail: [email protected]

Menopause, Vol. 21, No. 5, 2014 1

Copyright © 2013 The North American Menopause Society. Unauthorized reproduction of this article is prohibited.

shapses

Sticky Note

Menopause. 2014 May;21(5):501-8.

eligible, postmenopausal women had to be healthyVwithoutevidence of osteoporosis, metabolic bone disease, thyroiddisorders, immune disease, myocardial infarction, or stroke inthe past 6 months, or without kidney stones, diabetes, activecancers, or cancer therapy within the past 12 months. Partic-ipants were excluded if they changed their usual daily intakeof supplemental calcium or multivitamin/mineral, started anew exercise program, or were taking medications known toinfluence bone metabolism, including hormone therapy.These studies were approved by the Rutgers University In-stitutional Review Board. All participants signed an informedconsent form.

ProtocolParticipants were measured at three time points: baseline

(time 0), 6 months of weight reduction (0.5 y), and final (2 y).During the weight loss period (0-0.5 y), participantsunderwent 6 months of weight loss interventions in our lab-oratories. In this protocol, participants were counseledVonceweekly for the first 2 months and then twice monthlythereafterVby a registered dietitian to reduce their usual in-take by 500 to 600 kcal/day while maintaining usual physicalactivity levels, as described previously.1,2,18 During the 6-monthintervention, volunteers were given a multivitamin containing400 IU of vitamin D3, and total calcium intake was at least1,000 mg/day in all women. Upon completion of the inter-vention, all participants were counseled to consume approxi-mately 1.2 g of Ca and 400 IU of vitamin D daily through dietand supplementation. After weight loss, there was a no-intervention period (0.5-2.0 y), and participants were catego-rized according to weight change for this final measurement.Women who maintained their weight were recruited and age-matched to a cohort of women who did not meet these criteriaand regained their body weight (weight loss regainer [WL-R])in a case-control design. For eligibility in the weight lossmaintainer (WL-M) group, weight regain needed to be lessthan 25%. The WL-R group was composed of those whoregained more than this amount.

Bone and body composition measurements andserum markers

Weight and height were measured to the nearest 0.25 kgand 0.25 cm, respectively, at baseline, after weight loss(0.5 y), and at the 2-year final measurement with a balancebeam scale and a stadiometer, respectively (Detecto, WebbCity, MO). BMD was measured at the femoral neck, tro-chanter, spine, total body, 1/3 radius, and ultradistal (UD)radius by dual-energy x-ray absorptiometry (DXA; GE Lunar,Madison, WI; coefficient of variation G1% for all sites). Scanswere performed by using enCORE 2004 software (version8.10.027; GE Lunar). Bone mineral content at each site wasalso measured. Fat-free soft tissue (FFST), total fat mass,trunk fat, and leg fat were measured by DXA using total bodyscans and the manufacturer’s standard cut lines for leg andtrunk regions. Calcium intake was estimated using 3-day foodrecords and analyzed using the US Department of Agriculture

database (Food Works Software 10.1; Food Works, LongValley, NJ).

Fasting morning blood samples were collected from theentire population at baseline and from a subset (n = 22)after 2 years. The bone formation marker osteocalcin (BTI,Stoughton, MA; coefficient of variation G9%) was measuredby radioimmunoassay (RIA). Serum N-telopeptide of type Icollagen was measured by enzyme-linked immunosorbent as-say (Osteomark, Princeton, NJ; coefficient of variation G4.6%).Intact parathyroid hormone (PTH), 25-hydroxyvitamin D(25(OH)D), and estradiol were analyzed by RIA. The coeffi-cient of variation was less than 6.8% for PTH (DSL, Webster,TX; Scantibodies, Santee, CA), less than 12.5% for 25(OH)D(DiaSorin, Stillwater, MN), and less than 12.2% for estradiol(DSL, Webster, TX). Our laboratory participates in a vitamin Dexternal quality assessment scheme (DEQAS) to monitor theperformance of the RIA used for 25(OH)D assessment.

Statistical analysisChanges in body composition, hormones, and bone markers

at the three time points (baseline, 0.5 y, and 2 y) between thegroups (WL-R and WL-M) were analyzed by a two-factorrepeated-measures analysis of variance (ANOVA). If theF test was significant, post hoc analysis was performed usingTukey’s pairwise multiple comparison. Changes in bone markersbetween the groups from baseline to final measurement wereanalyzed using one-way ANOVA. Annual BMD loss was de-termined at each site by dividing the percent change in BMDby the number of months during the entire study period foreach individual. Multiple regression analysis was used toassess how the change in independent variables (age, leg fat,trunk fat, and FFST) across time influenced the change in BMDat each site. To determine whether weight loss and weight re-gain will have a similar effect on BMD, we performed a poweranalysis with > set to 0.05 and with A set to 0.90 using tro-chanter BMD change during weight loss or maintenance.5 Thisanalysis indicated that 15 participants per group would benecessary to avoid a type II error, and we included at least fiveadditional participants per group to account for two possiblebaseline covariates. Values are expressed as mean (SD), ex-cept in figures, which include SEM to improve visual clarity.Analysis was performed with SAS statistical software version9.2 (SAS Institute Inc, Cary, NC).

RESULTS

Ninety women who previously completed weight loss in-terventions in our laboratory were contacted for this study.Sixty-one responded to these inquiries, and 19 were excluded(owing to initiation of osteoporosis medications, initiation of arigorous exercise program, cancer diagnosis, initiation of di-abetes medications, or refusal to participate; Fig. 1). Forty-twopostmenopausal women agreed to participate and met criteriafor inclusion in this follow-up study. Participants were white(n = 40) and African American (n = 2). Measurements weretaken an average of 22 (6) months after the initiation of weightloss. At baseline, there were no significant differences in age,

2 Menopause, Vol. 21, No. 5, 2014 * 2013 The North American Menopause Society

VON THUN ET AL


weight, body mass index (BMI), time since menopause, bone,or soft-tissue data between the groups (Table 1). In both groupsof women, 12% to 15% had surgical or drug-induced meno-pause. For women who refused to participate or could not becontacted, we analyzed a subset (n = 31); their age, BMI, andweight loss did not differ from those included in this study.

Dietary intakeThe baseline total calcium intake (diet and supplement) was

948 (352) mg/day, which did not differ significantly between theWL-M group and theWL-R group. In the parent study (0-6 mo),there were equal numbers (n = 13) of women assigned to normalCa intake in each weight loss group, and the remaining womenwere assigned to higher Ca intake in the WL-M (n = 9) andWL-R (n = 7) groups. Total calcium intake during the weightloss intervention was 1,275 (406) mg/day in the WL-M group(n = 22) and 1,194 (389) mg/day in the WL-R group (n = 20),which did not differ significantly between the two groups. Inaddition, total calcium intake after 2 years (1,211 [512] mg/d)also did not differ between the groups. Participants consumed400 IU/day of vitamin D in their multivitamin throughout the2-year period. They consumed a small amount in their dietduring active weight loss (64 [43] IU/d) and during the no-intervention period (88 [84] IU/d), which did not vary signif-icantly between the groups.

Body weight and soft tissueAfter the 6-month weight loss protocol, there were no sig-

nificant differences in weight loss (9.3% [3.5%]) in the WL-Rand WL-M groups. However, weight loss differed signifi-cantly between the groups for two years (Table 1). During thepostYweight loss period (6 mo to 2 y), the WL-M group lostan average of 0.1 (2.0) kg, whereas the WL-R group gained4.1 (2.3) kg (P e 0.0001; Fig. 2). There were also significant

differences between the groups in total, trunk, and leg fat, withgreater gain in the WL-R group compared with the WL-Mgroup (P G 0.01; Table 1). FFST decreased after 2 years in allwomen (j1.2 [2.2] kg or j2.9% [5.6%]) and did not differsignificantly between the groups. Weight and fat mass (total,trunk, and leg) were significantly different between the groupsat 2 years (P G 0.001), with higher values in the WL-R groupcompared with the WL-M group.

BMD and content and serum markersAfter 6 months of weight loss, BMD at the trochanter

(j3.3% [4.6%]) and spine (j3.1% [5.8%]) decreased in bothgroups (P e 0.01). In the WL-M group, BMD also decreasedat the 1/3 radius and total body (Table 1). Bone mineral contentdecreased only at the 1/3 radius for both groups during the6 months of weight loss. Although there was a trend for greaterBMD loss at the UD radius during the weight loss period in theWL-M group compared with the WL-R group, no other sitesshowed differences between the groups (Table 1).

During the postintervention period (6 mo to 2 y), there was afurther decrease in BMD at the femoral neck in both groups(Fig. 3). The trochanter and 1/3 radius BMD continued to de-crease during the postintervention period in the WL-M group.Repeated-measures ANOVA showed significantly greaterBMD loss across time (2 y) between the groups at the tro-chanter (P G 0.02) and a trend at the 1/3 radius (Table 1, Fig. 3).Across 2 years, BMD decreased significantly at most sites andin both groups (Fig. 3). Bone mineral content decreased moreat the 1/3 radius and trochanter in the WL-M group than inthe WL-R group (P G 0.05) after 2 years (Table 1).

Bone turnover markers and hormone values did not dif-fer between the groups at baseline in the entire populationor subset (see Table 1, Supplemental Digital Content 1,http://links.lww.com/MENO/A66). Across the 2-year period,the bone resorption marker serum N-telopeptide of type Icollagen increased more in the WL-M group than in the WL-Rgroup (P G 0.05). In addition, serum PTH tended to increasemore in the WL-M group than in the WL-R group (P G 0.09).Changes in serum 25(OH)D, estradiol, and osteocalcin didnot differ between the WL-M group and the WL-R group.

Predictors of change in BMD for 2 yearsTo determine how age and soft tissue influence BMD, we

examined whether leg and trunk fat or FFST was important inpredicting the relationship using multiple regression analyses(Table 2). Age and body composition (leg or trunk fat or FFST)served as explanatory variables for each of the dependent var-iables. As expected, age did not have a significant independentassociation with bone loss because there was a relatively nar-row age range (53-72 y; Table 2). Changes in trochanter BMDwere largely explained by changes in leg fat (P G 0.05) andtended to be explained by FFST (P G 0.07). Similarly, leg fattended to explain changes in total BMD (P G 0.08).

DISCUSSION

It is well established that weight reduction leads to boneloss.19 This study was designed to determine whether bone

FIG. 1. Recruitment flow chart. WL-M, weight loss maintainer; WL-R,weight loss regainer.


BONE LOSS AFTER WEIGHT REDUCTION OR REGAIN


http://links.lww.com/MENO/A66

loss continues differently between women who maintain lostweight and women who regain lost weight. We show thatoverweight and obese postmenopausal women who experienceapproximately 10% weight reduction and regain approximately70% of the weight across 18 months have attenuated tro-chanter and 1/3 radius BMD loss compared with those whomaintain a reduced body weight. In women who regainweight, the bone lost during weight reduction is permanent;however, at 2 years, it does not exceed the 0.5% to 1% rateof annual BMD loss expected (depending on the site) becauseof normal aging.20<25 These findings suggest that only thosewomen who undergo short-term weight reduction andmaintain the lost weight for 2 years continue to lose moreBMD at several sites than those who regain the weight.

A few previous studies have addressed the influenceof weight loss and weight regain on site-specific BMD.14<17

The results of these studies have been contradictory, possiblyowing tomixed sexes, younger ages of the participants, or shorterstudy designs.14<16 In one study, premenopausal women wereexamined during a 3-month very-low-energy diet (13.2-kg loss),followed by 9 months of randomized controlled walking,and then followed for 2 years after a 62% weight regain.15

Fogelholm et al15 reported reduced lumbar spine and femoralneck BMD across 2 years for the entire group, but neithersite correlated with weight change. Avenell et al14 studiedBMD in 16 postmenopausal women for 6 months of weight

loss followed by 6 months of complete weight regain, andthey found greater loss of BMD at the lumbar spine (butnot at the femoral neck) than weight-stable women.14 In an-other study, 16 frail obese older adults were followed upat 30 months after a 1-year weight loss trial. Their weightremained below baseline and their hip BMD decreased, butthere no change in lumbar spine and whole-body BMD.17 Inthis same study, physical performance and metabolic profileremained improved after 30 months.17 In one other study, 23postmenopausal women were examined at 1 year after 6 monthsof 5% weight loss.16 The decrease in lumbar spine and hip BMDattributed to weight loss showed no further decrease during1 year of weight regain (93% fat mass gain). It was concludedthat bone did not recover with weight regain. However, theabsence of Bbone recovery[ is less surprising because it wouldnot be expected with fat mass gain16 and, in fact, BMD losswould be expected because of aging. Our findings show thatnot only is there no bone recovery with weight regain but alsothat BMD loss continues or begins at some bone sites. Forexample, there was no femoral neck BMD loss after 6 months,but there was a significant 2% loss after 2 years. This de-layed response to weight loss may be attributed to the boneremodeling transient that may take up to 2 years to complete.26

Annual bone loss in weight-stable postmenopausal womenreaches up to 1% per year,20<25 similar to women in the cur-rent trial who regained their weight (loss at the femoral

TABLE 1. Baseline and percent change in women who lost weight and either maintained or regained the weight

WL-M group (n = 22) WL-R group (n = 20)

Baselinea

% Change

Baseline

% Change

6 mo 2 y 6 mo 2 y

Age, y 61.3 (5.6) 60.1 (5.2)TSM, y 10.5 (7.7) 11.4 (7.3)BMI, kg/m2 27.7 (2.6) 28.9 (3.0)Body weight, kg 73.1 (7.1) j10.2 (3.2)b j10.3 (4.2)b 75.3 (7.0) j8.2 (3.3)b j2.8 (3.3)b,c

FFST, kg 36.0 (3.3) j3.3 (4.3)b j3.8 (5.5)b 36.4 (3.9) j3.4 (5.5)b j1.9 (6.4)Fat mass, kg 30.7 (4.8) j18.2 (8.7)b j16.6 (10.5)b 33.0 (5.1) j13.6 (6.7)b j5.4 (6.6)b,c

Trunk fat, kg 13.1 (2.4) j18.2 (14.0)b j19.1 (17.4)b 14.5 (2.5) j14.1 (6.7)b j4.9 (11.5)c

Leg fat, kg 10.8 (1.9) j13.9 (11.1)b j16.9 (13.0)b 11.2 (2.4) j12.0 (7.6)b j2.5 (9.9)c

BMD, g/cm2

Femoral neck 0.86 (0.11) j0.5 (3.1) j2.9 (4.3)b 0.87 (0.09) j0.1 (3.2) j2.2 (3.3)b

Trochanter 0.73 (0.11) j4.1 (4.4)b j6.8 (5.7)b 0.75 (0.11) j2.6 (4.7)b j2.5 (6.1)c,d

UD radius 0.31 (0.04) j1.5 (3.9)d j1.0 (4.6) 0.32 (0.05) 0.7 (3.4)e j2.0 (3.9)1/3 Radius 0.63 (0.09) j1.8 (2.3)b j4.5 (3.3)b 0.64 (0.07) j0.9 (2.9) j1.6 (3.8)c,d

Spine 0.99 (0.09) j3.3 (5.3)b j1.3 (6.4) 0.99 (0.13) j2.9 (4.7)b j2.5 (6.3)d

Total body 1.11 (0.08) j1.1 (1.8)b j1.5 (1.7)b 1.12 (0.08) j0.4 (1.7) j0.5 (1.9)e

BMC, gFemoral neck 4.40 (0.83) j1.3 (6.2) j1.4 (5.2) 4.34 (0.43) 1.4 (5.4) j1.5 (6.2)Trochanter 8.48 (1.82) j2.3 (8.8) j6.0 (9.0)b 8.97 (2.24) j3.5 (8.8) j1.3 (10.7)c

UD radius 1.11 (0.16) j1.3 (9.1) j4.7 (12.4) 1.13 (0.17) 1.7 (12.1) j4.1 (15.7)1/3 Radius 1.57 (0.21) j2.8 (2.6)b j4.1 (3.2)b 1.53 (0.19) j1.8 (4.3)b j1.3 (5.6)c

Total body 2,276 (271) j1.6 (5.3) j2.7 (5.1)b 2,309 (257) j0.6 (3.9) j1.7 (4.3)

Data are presented as mean (SD).Weight loss period refers to 0 to 0.5 y. Baseline to final period refers to 0 to 2 y.WL-M, weight loss maintainer; WL-R, weight loss regainer; TSM, time since menopause; BMI, body mass index; FFST, fat-free soft tissue; BMD, bone mineraldensity; UD, ultradistal; BMC, bone mineral content.aNo significant differences among groups were observed for any variable at baseline.bDiffers from baseline (P G 0.05).cDiffers from the WL-M group (P G 0.05).dP e 0.09.eP e 0.09 (same period).


VON THUN ET AL


neck,j1.1% per year; trochanter,j1.3% per year; UD radius,j1.0% per year). In addition, women who successfully main-tained a lower body weight for 2 years (WL-M group) showedan even higher annual BMD loss at the trochanter (È3.3% peryear) and 1/3 radius (j2.5% per year). There is also elevatedbone resorption at 2 years for women who maintained weightloss compared with those who regained body weight; this isconsistent with findings of higher bone resorption at 9 monthsafter a short-term (3mo)weight loss.27 These findings clarify thatbone loss is permanent and continues at a faster rate for at least2 years for those who do not regain weight. In addition, althoughthe current study observed only one weight loss cycle, bone losswould be expected to be greater with subsequent weight cycling,as shown in rodents.12 Furthermore, retrospective studies showthat multiple episodes of weight loss and regain (weight cycle)

increase fracture risk at some sites,10,15,28,29 suggesting that, evenwith a normal annual rate of bone loss observed in the Bregain[group for 2 years, bone quality may be compromised. It is pos-sible that bone loss during the postYweight loss period is de-layed because of the bone remodeling transient or because ofsystemic factors that were altered by lowering body weight,19,26

yet some anatomical sites showed faster, rather than slower,bone loss during this postYweight loss period.

The current study demonstrates that loss of FFST and legfat is a predictor of bone loss, whereas there was no relation-ship with trunk fat, which represents a site with greater vis-ceral fat. One recent 3-month weight loss study in young menand women30 examined how soft-tissue compartments explainBMD loss with and without 6 months of weight regain. Theauthors concluded that changes in soft-tissue composition hadlittle contribution to changes in BMD with weight loss.30

However, there was no bone loss with weight loss in theseparticipants, possibly owing to the short-term intervention and/orthe young age of the participants19; thus, how bone loss isregulated by soft-tissue changes could not be determined. Ourresults are encouraging because they suggest that fat loss in apredominantly visceral region is not associated with bone loss.Nevertheless, this hypothesis using DXA technology should beconfirmed using magnetic resonance imaging and/or quantita-tive CT to distinguish between different fat depots. Further-more, it is possible that loss of FFST is exacerbated with eachweight loss period,31 especially in older individuals32; this atleast partially explains the lower BMD found in weight cyclers.

This study is limited in its interpretation because it is acase-control design rather than a randomized controlled trial,such that women in the two groups were self-selected. In ad-dition, BMD measurement errors are a concern in the obesepopulation owing to excess and homogeneity of fat tissue sur-rounding bones33<36 and/or changes in soft tissue with weightreduction. However, we are less concerned about this poten-tial error for a few reasons. First, the results show bonechanges at both central and peripheral bone sites that havemore and less adiposity, respectively. In addition, adiposetissue change in this moderate weight loss/regain study isless than in studies showing bone measurement errors.33,34

For example, it has been found that 6 kg or more of fat layer-ing on the region being measured (spine or hip) will artifi-cially increase BMD, but the error was not found using smalleramounts of fat.37 In the current study, total body fat gain inthe weight regain group was only 2.8 kg and, therefore, waswell below the threshold where error would be expected; theBMD loss during weight stability (0.5-2 y) in the WL-M groupcannot be attributed to changes in fat tissue. In addition, theabsence of a control weight maintenance group is also a limi-tation; however, age-related bone loss has been extensivelystudied in the literature20<25 and thus is a smaller concern.

One strength of this study is that participants were measuredby the same certified radiology technician using identical in-strumentation across 2 years for all three measurements. Also,although participants had different calcium supplementationassignments during the 5-month weight loss, intake was at

FIG. 2. Weight (kg), fat, and fat-free soft tissue at baseline, at the end ofthe weight loss intervention (0.5 y), and after 2 years in the weight lossmaintainer (WL-M; n = 22) and weight loss regainer (WL-R; n = 20)groups. Data are presented as mean (SEM). *Repeated-measures analysisof variance differs between the groups, P G 0.05. The only time point thatdiffered between the groups was at 2 years for weight and fat (P G 0.001).Values with different letters are significantly different, P G 0.05. Upper-case and lower-case letters are used to denote WL-R and WL-M, re-spectively.




or above the recommended level in all individuals, there was arelatively even distribution of both levels of calcium betweenthe groups, and both groups had similar calcium intake duringmonths 6 to 24. It is also possible that this data set is biasedbecause it excludes certain women who were not eligible forfollow-up (ie, five women were excluded owing to initiationof osteoporosis medications, and nine women were excludedbecause they were not interested). However, it was encouragingto note that nearly all women contacted agreed to participate(if they met the inclusion criteria) and that those who we werenot contacted or did not participate had age, BMI, and weightloss similar to those included in the study. One further limitationis that we did not monitor physical activity during months 6 to24 and depended on participants to remember whether therewere any extreme changes in physical activity (the reasonfor the exclusion of one woman during screening). It would beexpected that those who maintained their body weight (vsregainers) would have been more physically active to maintaintheir lower weight and that this would have a beneficial effecton bone, yet BMD loss was greater or similar in the group thatsuccessfully maintained weight loss.

CONCLUSIONS

This study shows that weight reductionYinduced bone lossthat is apparent immediately after weight loss either continuesor first begins at the trochanter, femoral neck, and radius inpostmenopausal women. Weight regain does not result in re-covery of bone, but it prevents greater loss at the trochanter and1/3 radius compared with reduced-obese women. Even afterweight loss, about 50% of women in theWL-M group were stilloverweight. Thus, weight loss would be recommended for theseindividuals, who typically would not be considered at risk forosteoporosis based on their overweight status. There is concernthat repeated dieting would enhance bone loss and osteoporosisrisk in this population compared with obese women who havebeen weight-stable, possibly owing to poor bone quality,38,39

but specific studies have not been conducted. Hence, questionsabout weight history should be considered when evaluatingrisk of osteoporosis. Current recommendations appropriatelyencourage weight loss in overweight individuals to reduce therisk of comorbidities. Until future prospective studies haveaddressed how to prevent bone loss after a successful weight

FIG. 3. Bone mineral density (BMD) at baseline, at the end of the weight loss intervention (0.5 y), and on follow-up (2 y) in the weight loss maintainer(WL-M; n = 22) and weight loss regainer (WL-R; n = 20) groups. Data are presented as mean (SEM). *Repeated-measures analysis of variance differsbetween the groups (P G 0.02). †P G 0.08. There are no significant between-group differences at any time point. Values with different letters aresignificantly different. Upper-case and lower-case letters are used to denote WL-R and WL-M, respectively. UD, ultradistal.


VON THUN ET AL


reduction, therapies that have been shown to ameliorate boneloss during weight reductionVsuch as adequate calcium andvitamin D, higher protein intake, and increased bone-loadingexerciseVshould also be encouraged after weight stabilizes.Whether BMD changes in cortical or trabecular bone affectgeometry after weight loss remains an important question inthe prevention of fracture risk.

Acknowledgments: We thank Drs. Riedt and Cifuentes for recruitingvolunteers and for conducting the weight loss protocol. We also thankthe volunteers for their commitment to these studies.

REFERENCES

1. Riedt CS, Cifuentes M, Stahl T, Chowdhury HA, Schlussel Y, ShapsesSA. Overweight postmenopausal women lose bone with moderateweight reduction and 1 g/day calcium intake. J Bone Miner Res 2005;20:455-463.

2. Sukumar D, Ambia-Sobhan H, Zurfluh R, et al. Areal and volumetricbone mineral density and geometry at two levels of protein intake duringcaloric restriction: a randomized, controlled trial. J Bone Miner Res2011;26:1339-1348.

3. Villareal DT, Shah K, Banks MR, Sinacore DR, Klein S. Effect of weightloss and exercise therapy on bone metabolism and mass in obese olderadults: a one-year randomized controlled trial. J Clin Endocrinol Metab2008;93:2181-2187.

4. Gozansky WS, Van Pelt RE, Jankowski CM, Schwartz RS, Kohrt WM.Protection of bone mass by estrogens and raloxifene during exercise-induced weight Loss. J Clin Endocrinol Metab 2005;90:52-59.

5. Riedt CS, Schlussel Y, von Thun N, et al. Premenopausal overweightwomen do not lose bone during moderate weight loss with adequate orhigher calcium intake. Am J Clin Nutr 2007;85:972-980.

6. Shapses SA, Von Thun NL, Heymsfield SB, et al. Bone turnover anddensity in obese premenopausal women during moderate weight loss andcalcium supplementation. J Bone Miner Res 2001;16:1329-1336.

7. Redman LM, Rood J, Anton SD, Champagne C, Smith SR, Ravussin E.Calorie restriction and bone health in young, overweight individuals.Arch Intern Med 2008;168:1859-1866.

8. Ensrud KE, Fullman RL, Barrett-Connor E, et al. Voluntary weight re-duction in older men increases hip bone loss: the Osteoporotic Fracturesin Men Study. J Clin Endocrinol Metab 2005;90:1998-2004.

9. Ensrud KE, Ewing SK, Stone KL, Cauley JA, Bowman PJ, CummingsSR. Intentional and unintentional weight loss increase bone loss and hipfracture risk in older women. J Am Geriatr Soc 2003;51:1740-1747.

10. Meyer HE, Tverdal A, Selmer R. Weight variability, weight change andthe incidence of hip fracture: a prospective study of 39,000 middle-agedNorwegians. Osteoporos Int 1998;8:373-378.

11. Beavers KM, Lyles MF, Davis CC, Wang X, Beavers DP, Nicklas BJ. Islost lean mass from intentional weight loss recovered during weightregain in postmenopausal women? Am J Clin Nutr 2011;94:767-774.

12. Bogden JD, Kemp FW, Huang AE, et al. Bone mineral density andcontent during weight cycling in female rats: effects of dietary amylase-resistant starch. Nutr Metab (Lond) 2008;5:34.

13. Bacon L, Stern JS, Keim NL, Van Loan MD. Low bone mass inpremenopausal chronic dieting obese women. Eur J Clin Nutr 2004;58:966-971.

14. Avenell A, Richmond PR, Lean ME, Reid DM. Bone loss associated witha high fibre weight reduction diet in postmenopausal women. Eur J ClinNutr 1994;48:561-566.

15. Fogelholm GM, Sievanen HT, Kukkonen-Harjula TK, Pasanen ME. Bonemineral density during reduction, maintenance and regain of body weightin premenopausal, obese women. Osteoporos Int 2001;12:199-206.

16. Villalon KL, Gozansky WS, Van Pelt RE, et al. A losing battle: weightregain does not restore weight lossYinduced bone loss in postmenopausalwomen. Obesity (Silver Spring) 2011;19:2345-2350.

17. Waters DL, Vawter R, Qualls C, Chode S, Armamento-Villareal R,Villareal DT. Long-term maintenance of weight loss after lifestyle in-tervention in frail, obese older adults. J Nutr Health Aging 2013;17:3-7.

18. Ricci TA, Heymsfield SB, Pierson RN Jr, Stahl T, Chowdhury HA,Shapses SA. Moderate energy restriction increases bone resorption inobese postmenopausal women. Am J Clin Nutr 2001;73:347-352.

19. Shapses SA, Sukumar D. Bone metabolism in obesity and weight loss.Annu Rev Nutr 2012;32:287-309.

20. Makovey J, Naganathan V, Seibel M, Sambrook P. Gender differences inplasma ghrelin and its relations to body composition and boneVanopposite-sex twin study. Clin Endocrinol (Oxf) 2007;66:530-537.

21. Sirola J, Kroger H, Honkanen R, et al. Factors affecting bone loss aroundmenopause in women without HRT: a prospective study. Maturitas2003;45:159-167.

22. Nguyen TV, Sambrook PN, Eisman JA. Bone loss, physical activity, andweight change in elderly women: the Dubbo Osteoporosis EpidemiologyStudy. J Bone Miner Res 1998;13:1458-1467.

23. Young R, May H, Murphy S, Grey C, Compston JE. Rates of bone loss inperi- and postmenopausal women: a 4 year, prospective, population-based study. Clin Sci (Lond) 1996;91:307-312.

24. Sornay-Rendu E, Munoz F, Duboeuf F, Delmas PD. Rate of forearmbone loss is associated with an increased risk of fracture independently ofbone mass in postmenopausal women: the OFELY study. J Bone MinerRes 2005;20:1929-1935.

25. Daly RM, Ahlborg HG, Ringsberg K, Gardsell P, Sernbo I, KarlssonMK. Association between changes in habitual physical activity andchanges in bone density, muscle strength, and functional performance inelderly men and women. J Am Geriatr Soc 2008;56:2252-2260.

26. Heaney RP. The bone-remodeling transient: implications for the inter-pretation of clinical studies of bone mass change. J Bone Miner Res1994;9:1515-1523.

27. Hinton PS, LeCheminant JD, Smith BK, Rector RS, Donnelly JE. WeightlossYinduced alterations in serum markers of bone turnover persist duringweight maintenance in obese men and women. J Am Coll Nutr 2009;28:565-573.

28. Sogaard AJ, Meyer HE, Tonstad S, Haheim LL, Holme I. Weight cyclingand risk of forearm fractures: a 28-year follow-up of men in the OsloStudy. Am J Epidemiol 2008;167:1005-1013.

29. Gallagher KI, Jakicic JM, Kiel DP, Page ML, Ferguson ES, Marcus BH.Impact of weight-cycling history on bone density in obese women. ObesRes 2002;10:896-902.

TABLE 2.Multiple regression model of the relative contribution of age and change in body composition to change in BMD for all women for 2years (N = 42)

Variablea Explanatory variable A coefficient P Model R2 (%)

Trochanter BMD, g/cm2 Fat-free soft tissue 0.3250 0.066 19.5Leg fat 0.3370 0.043Age j0.0511 0.749

Trunk fat 0.0072 0.998Total BMD, g/cm2 Fat-free soft tissue j0.206 0.558 14.9

Leg fat j0.302 0.073Age 0.063 0.808

Trunk fat 0.058 0.836

BMD, bone mineral density.aVariables without significant findings or trends are not shown (femoral neck, total spine, ultradistal radius, and 1/3 radius).




30. Bosy-Westphal A, Later W, Schautz B, et al. Impact of intra- and extra-osseous soft tissue composition on changes in bone mineral density withweight loss and regain. Obesity (Silver Spring) 2011;19:1503-1510.

31. Lee JS, Visser M, Tylavsky FA, et al. Weight loss and regain and effectson body composition: the Health, Aging, and Body Composition Study.J Gerontol A Biol Sci Med Sci 2010;65:78-83.

32. Taaffe DR, Cauley JA, Danielson M, et al. Race and sex effects on theassociation between muscle strength, soft tissue, and bone mineral den-sity in healthy elders: the Health, Aging, and Body Composition Study.J Bone Miner Res 2001;16:1343-1352.

33. Tothill P. Dual-energy x-ray absorptiometry measurements of total-bodybone mineral during weight change. J Clin Densitom 2005;8:31-38.

34. Bolotin HH. Analytic and quantitative exposition of patient-specificsystematic inaccuracies inherent in planar DXA-derived in vivo BMDmeasurements. Med Phys 1998;25:139-151.

35. Blake GM, Fogelman I. How important are BMD accuracy errors forthe clinical interpretation of DXA scans? J Bone Miner Res 2008;23:457-462.

36. Svendsen OL, Hassager C, Skodt V, Christiansen C. Impact of softtissue on in vivo accuracy of bone mineral measurements in the spine,hip, and forearm: a human cadaver study. J Bone Miner Res 1995;10:868-873.

37. Yu EW, Thomas BJ, Brown JK, Finkelstein JS. Simulated increases inbody fat and errors in bone mineral density measurements by DXA andQCT. J Bone Miner Res 2012;27:119-124.

38. Premaor MO, Pilbrow L, Tonkin C, Parker RA, Compston J. Obesity andfractures in postmenopausal women. J Bone Miner Res 2010;25:292-297.

39. Sukumar D, Schlussel Y, Riedt CS, Gordon C, Stahl T, Shapses SA.Obesity alters cortical and trabecular bone density and geometry inwomen. Osteoporos Int 2011;22:635-645.


VON THUN ET AL


Does bone loss begin after weight loss ends? Results 2 years after weight loss or regain in postmenopausal women

Documents