Association between Fat-Free Mass Loss after

Diet and Exercise Interventions and Weight

CLINICAL SCIENCES
Regain in Women with Overweight
Downloaded from http://journals.lww.com/acsm-msse by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

CATIA MARTINS, BARBARA A. GOWER, and GARY R. HUNTER

Department of Nutrition Sciences, University of Alabama at Birmingham, Birmingham, AL
CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdgGj2MwlZLeI= on 02/08/2024

ABSTRACT
MARTINS, C., B. A. GOWER, and G. R. HUNTER. Association between Fat-Free Mass Loss after Diet and Exercise Interventions and
Weight Regain in Women with Overweight. Med. Sci. Sports Exerc., Vol. 54, No. 12, pp. 2031-2036, 2022. Purpose: This study aimed
to determine if percent fat-free mass loss (% FFML) after diet alone, diet plus aerobic, or diet plus resistance exercise is a predictor of weight
regain in women with overweight. Methods: One hundred and forty-one premenopausal women with overweight (body mass index,
28 ± 1 kg·m−2; age, 35 ± 6 yr) enrolled in a weight loss program to achieve a body mass index <25 kg·m−2 (diet alone, diet plus resistance,
or diet plus aerobic exercise) and were followed for 1 yr. Body weight and composition (with dual-energy x-ray absorptiometry) were mea-
sured at baseline, after weight loss, and at 1 yr. Results: Participants lost 12.1 ± 2.6 kg of body weight, 11.3 ± 2.5 kg of fat mass, and
0.5 ± 1.6 kg of fat-free mass during the weight loss intervention, followed by weight regain at 1 yr (6.0 ± 4.4 kg, 51.3% ± 37.8%;
P < 0.001 for all). % FFML was −3.6 ± 12.4, and a greater % FFML was associated with more weight regain (r = −0.216, P = 0.01,
n = 141), even after adjusting for the intervention group (β = −0.07; 95% confidence interval, −0.13 to −0.01; P = 0.017). Conclusions: %
FFML is a significant predictor of weight regain in premenopausal women with overweight. These results support strategies for conserving
fat-free mass during weight loss, such as resistance training. Future research should try to identify the mechanisms, at the level of both appetite
and energy expenditure, responsible for this association. Key Words: LEAN TISSUE, DIET, AEROBIC, RESISTANCE, WEIGHT LOSS

W
eight regain remains the main challenge in obesity arising from the adipose tissue, in particular leptin, a growing
management. Even though clinically significant body of evidence suggests that FFM also plays a role in the
weight loss is achievable in the short term, drive to eat (7–9). It has been proposed that FFM can modulate
long-term results are disappointing (1,2), with some patients energy intake and body weight both directly, through feed-
relapsing to or above baseline weight. The exact reasons for back signaling between FFM and brain centers involved in ap-
weight recidivist remain unexplained (3,4). The search for new petite control (10,11), and indirectly, via its effects on resting
metabolic determinants of relapse in obesity management is energy expenditure (EE) (12,13) and free-living activity–
therefore needed to better understand weight regain and help de- related EE (14–16). Loss of FFM after lifestyle interventions
velop new strategies to improve long-term weight loss outcomes. may therefore contribute to weight regain, not only due to the
Recently, there has been increased interest in the functional body’s attempt to restore FFM by overeating but also due to
role of body composition, in particular fat-free mass (FFM), in the resulting lower EE and energy needs.
modulating eating behavior and body weight (5,6). Even In a reanalysis of the Minnesota Starvation Experiment by
though the traditional view has been that body weight homeo- Dulloo and colleagues (17), a greater initial loss of FFM was
static and appetite control are regulated through feedback signals associated with a stronger hyperphagic response during
refeeding, even after adjusting for fat mass (FM) loss. Surpris-
ingly, hyperphagia persisted despite complete recovery of
body weight and FM, until FFM was totally restored to base-
Address for correspondence: Gary Hunter, Ph.D., Department of Nutrition Sci- line values. This landmark article is, nevertheless, limited by
ences, University of Alabama at Birmingham, Webb Nutrition Sciences Building, the fact that it included lean men only, and the energy deficit
1675 University Boulevard, Birmingham, AL 35233; E-mail: [email protected]. was extreme (20%–26% loss of initial body weight due to
Submitted for publication April 2022.
Accepted for publication June 2022. semistarvation). It is unknown if a similar response is activated
Supplemental digital content is available for this article. Direct URL citations in individuals with overweight or obesity after less restricted
appear in the printed text and are provided in the HTML and PDF versions lifestyle interventions, and if FFM loss modulates weight re-
of this article on the journal’s Web site (www.acsm-msse.org). gain. A greater fractional loss of FFM (% FFML), after
0195-9131/22/5412-2031/0 diet-induced weight loss in 55 individuals with overweight
MEDICINE & SCIENCE IN SPORTS & EXERCISE® and obesity, was found to be predictive of weight regain at
Copyright © 2022 by the American College of Sports Medicine 9 months, even after adjusting for baseline percent FM (18).
DOI: 10.1249/MSS.0000000000002992 A recent meta-analysis of more than 2000 individuals with

2031

Copyright © 2022 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 overweight or obesity suggested that reductions in FM and group meetings (bimonthly dietary education classes aimed
FFM better predict weight regain compared with weight loss at weight maintenance for the first 6 months, followed by
alone (19). monthly meetings for months 6 to 12) and to continue with
CLINICAL SCIENCES

In a recent reanalysis of the DiOGenes project, % FFML their exercise program, if applicable. For detailed informa-
predicted weight regain at 6-month follow-up in men but not tion about the two studies, see Weinsier et al. (23) and Hunter
in women, and a larger % FFML was associated with a larger et al. (21).
increase in postprandial hunger and desire to eat in men (20). Testing was done, after a 4-wk weight stabilization period
All the available evidence regarding a potential role for % FFML (aiming to maintain body weight within a 2.5-kg range), at
Downloaded from http://journals.lww.com/acsm-msse by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

on weight regain arises from studies where weight loss was baseline, after weight loss, and at 1-yr follow-up. During the
induced by low-energy or very-low energy diets (5,18,19). 4-wk weight stabilization period, participants were weighed
It remains to be determined if % FFML is associated with three times a week the first 2 wk while eating own food and
CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdgGj2MwlZLeI= on 02/08/2024

weight regain when weight loss is induced by combined life- weighed five times a week with food provided by GCRC the
style interventions, including both diet and exercise, given that last 2 wk. Variation in body weight during the last 2 wk of
exercise, in particular resistance exercise, is known to prevent the stabilization period, after weight loss, was −1.0 ± 1.4 kg.
the decrease in FFM that usually accompanies diet-induced Testing was done 30 ± 2 d after the end of the weight loss
weight loss (15,21,22). phase. All testing was conducted in the follicular phase of
Therefore, the aims of this exploratory post hoc analysis the menstrual cycle during a 4-d GCRC inpatient stay (to en-
were to determine the association between % FFML after sure that physical activity and diet were standardized). Testing
diet alone, diet plus resistance exercise, or diet plus aerobic ex- was done in a fasted state in the morning after spending the
ercise, and weight regain at 1-yr follow-up in a group of night in the GCRC.
women with overweight. We hypothesized that a greater pro- Data collection. Body weight and composition (FM and
portional reduction in FFM (% FFML) would result in greater FFM) were assessed by dual-energy x-ray absorptiometry
weight regain at 1-yr follow-up. (DXA; DPX-L; Lunar Corp, Madison, WI) with the use of
software version 1.5g (Lunar Corp), at baseline, after weight
loss, and at 1-yr follow-up, after a 4-wk weight stabilization
METHODS period (at all time points).
Participants. Participants in this analysis were premeno- Statistical analysis. Statistical analysis was performed
pausal women with overweight (25 ≤ body mass index with SPSS version 22 (SPSS Inc., Chicago, IL), and data were
(BMI) ≤ 30 kg·m−2). They were 20–41 yr of age, sedentary presented as mean ± SD, unless otherwise stated. Statistical
(no more than one time per week regular exercise), had normal significance was set at P < 0.05. All variables were assessed
glucose tolerance (2–h glucose ≤140 mg·dL−1 after 75-g oral for normality by visual inspection of Q–Q plots and histo-
dose), had a family history of overweight/obesity in at least grams. Participants from all groups were combined and in-
one first-degree relative, and had no use of medications that af- cluded in the analysis if they had anthropometric data at base-
fect body composition or metabolism. All women were non- line, after weight loss, and at 1-yr follow-up (n = 141). This
smokers and reported a regular menstrual cycle. The two stud- was done to ensure enough variance in the response variable
ies included in this retrospective analysis were both approved (loss of FFM) and to allow for sufficient sample size required
by the Institutional Review Board for Human Use at the Uni- to have robust examination of multiple regression models.
versity of Alabama at Birmingham. All women provided in- Changes in body weight, FM, and FFM over time were
formed consent before participating in the study. assessed by a repeated-measures ANOVA, with Bonferroni
Study design. Participants included in this retrospective correction for post hoc pairwise comparisons.
analysis come from two different studies performed at the De- The proportion of weight lost as FFM was calculated as the
partment of Nutrition Sciences at the University of Alabama at change in FFM during weight loss divided by total weight loss
Birmingham, with the same sequence of events and same (i.e., % FFML = (ΔFFM/Δweight)  100). There was one ex-
methodology, and both aiming to identify metabolic predictors treme value for % FFML (40.5%), which was excluded from
of weight regain. In the first study, all participants achieved the analysis. The association between % FFML and weight re-
weight loss with diet alone (single-arm longitudinal study with gain was investigated with Pearson correlation. Univariate lin-
repeated measurements). In the second study, participants ear regressions were also conducted to investigate crude asso-
were randomly assigned to one of three groups: 1) diet alone ciations between potential predictors and the outcome variable
(same diet as in the first study), 2) diet plus aerobic exercise (weight regain). β coefficients were reported as unstandardized
training three times a week, and 3) diet plus resistance exercise estimates and 95% confidence intervals (CI), representing the
training three times a week. All participants were provided an estimate and confidence of a 1-unit change in the predictor
800-kcal diet until reaching a BMI <25 kg·m−2. Food was variable per 1-kg change in weight regain at 1 yr (1 yr minus
provided (20%–22% fat, 20%–22% protein, and 56%–58% postintervention). Next, multivariate linear regression models
carbohydrate) by the General Clinical Research Center were generated for all individuals. Adjustments were consid-
(GCRC) Kitchen. During follow-up (1 yr), participants were ered for the intervention group, the amount of weight lost (be-
encouraged, but not mandated, to attend regular support cause weight loss has been shown to be a strong predictor of

2032 Official Journal of the American College of Sports Medicine http://www.acsm-msse.org

Copyright © 2022 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 TABLE 1. Anthropometric variables over time (n = 141). TABLE 2. Univariate regression analysis predicting weight regain at 1 yr.
Baseline Postintervention 1 yr Predictor β Coefficient Adjusted R 2 P
−2 a,b a,c b,c
BMI (kg·m ) 28.3 ± 1.3 23.9 ± 1.0 26.1 ± 2.2 Age 0.067 (−0.047 to 0.181) 0.002 0.248
77.3 ± 7.1a,b 65.1 ± 6.3a,c 71.1 ± 8.1b,c

CLINICAL SCIENCES
Weight (kg) Race 0.793 (−0.654 to 2.241) 0.001 0.280
FM (kg) 33.3 ± 4.8a,b 21.7 ± 4.3a,c 27.6 ± 6.1b,c Group 0.821 (−0.090 to 1.732) 0.015 0.077
FFM (kg) 44.0 ± 4.0a,b 43.5 ± 4.1a 43.5 ± 4.4b Baseline weight (kg) 0.080 (−0.022 to 0.183) 0.01 0.124
% FFML −3.6 ± 12.4 Baseline FFM (kg) 0.220 (0.041 to 0.399) 0.034 0.016
Weight regain (kg) 6.0 ± 4.4 Baseline FM (kg) −0.021 (−0.132 to 0.173) 0.001 0.789
Weight regain (%) 51.3 ± 37.8 Baseline FM (%) −0.132 (−0.336 to 0.071) 0.005 0.201
FM regain (kg) 5.9 ± 3.9 Weight loss (kg) −0.050 (−0.329 to 0.230) 0.001 0.727
FFM regain (kg) −0.1 ± 1.5 FM loss (kg) 0.195 (−0.097 to 0.488) 0.005 0.188
Downloaded from http://journals.lww.com/acsm-msse by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

FFM loss (kg) −0.628 (−1.083 to −0.174) 0.044 0.007

Data presented as mean ± SD.
FFML (%) −0.076 (−0.134 to −0.019) 0.040 0.010
Means sharing the same superscript letters denote significant changes over time:
a,b,c
P < 0.001. Univariate linear regression analyses predicting weight regain at 1 yr (from week 9). Each
unstandardized β coefficient represents 1-kg weight regain at 1 yr per unit of the predictor.
CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdgGj2MwlZLeI= on 02/08/2024

weight regain) (19), and initial FM, given that baseline body
composition has been shown to modulate body composition the diet group (−9.7% ± 9.9%) compared with both diet + aerobic
changes with weight loss (24). Multicollinearity was tested (−2.2% ± 10.0%) and diet + resistance exercise (3.6% ± 12.8%;
by examining the variance inflation factors of the model vari- P = 0.007 and P < 0.001, respectively), and there was also a ten-
ables and was deemed acceptable. dency for % FFML to be larger in the diet + aerobic versus the
diet + resistance exercise group ( P = 0.068).
There was a significant association between % FFML and
RESULTS
weight regain at 1 yr (r = −0.216, P = 0.01, n = 141), with a
One hundred and forty-one women (68 White and 73 greater % FFML (or lower accretion) being associated with
Black) were included in this analysis. They had an average more weight regain (Fig. 1). Table 2 provides univariate re-
age of 35 ± 6 yr and a BMI of 28 ± 1 kg·m−2. Anthropometrics gression results predicting weight regain (from end of weight
at baseline, after weight loss, and at 1-yr follow-up are pre- loss to 1 yr). Baseline FFM (in kilograms), but not weight or
sented in Table 1. Significant reductions in BMI, body weight, FM (in kg or percent), significantly predicted weight regain
and FM (in kilograms) were seen both after weight loss and at (β = 0.220 kg; 95% CI, 0.041 to 0.399 kg; R2 = 0.0334,
1 yr, compared with baseline ( P < 0.001 for all). Participants P = 0.016). Moreover, FFM loss (in kilograms; β =
lost, on average, 4.5 ± 1.0 kg·m−2, 12.1 ± 2.6 kg of body −0.628 kg; 95% CI, −1.083 to −0.174 kg; R2 = 0.044,
weight, 11.6 ± 2.5 kg of FM, and 0.5 ± 1.6 kg of FFM (range, P = 0.007) and % FFML (β = −0.076 kg; 95% CI, −0.134 to
−5.5 to 3.1 kg) during the weight loss intervention. This was −0.019 kg; R2 = 0.040, P = 0.01) were also predictors of weight
followed by weight (6.0 ± 4.4 kg) and FM (5.9 ± 3.9 kg) regain regain. Age, race, and weight or FM loss were not significant
at 1 yr ( P < 0.001 for both), even though values were still below predictors of weight regain. There was a tendency for the inter-
baseline at 1-yr follow-up ( P < 0.001 for both). FFM (in kilo- vention group to predict weight regain (β = 0.821 kg; 95% CI,
grams) was maintained between end of intervention and 1 yr −0.090 to 1.732 kg; R2 = 0.015, P = 0.077).
(−0.1 ± 1.5 kg, P = 0.579). % FFML after intervention was Table 3 reports a multivariate linear regression model
−3.6% ± 12.4% (range, −32.7% to 28.2%). Average weight regain predicting weight regain. % FFML was a significant predictor
from end of weight loss to 1 yr was 6.0 ± 4.4 kg (51.3% ± 37.8%), of weight regain at 1 yr (β = −0.07; 95% CI, −0.13 to −0.01 kg;
with a very large interindividual variation (−3.6 to 20.3 kg), even P = 0.017), even after adjusting for the intervention group, with
though 95% of the participants regained weight (anything the model explaining 5% of the variation in weight regain.
above zero) over time. % FFML was significantly larger in Baseline FM (in kilograms) and FM loss (in kilograms) were
not included in the model, as they were not significant predic-
tors and reduced the R2 and statistical significance of the model
(Supplemental Table, Supplemental Digital Content, Multivar-
iate linear regression models predicting weight regain at 1 yr,
http://links.lww.com/MSS/C665).
Those with a negative % FFML (n = 88), denoting loss of
FFM during the intervention, regained significantly more weight

TABLE 3. Multivariate linear regression model predicting weight regain at 1 yr.

Predictor β Coefficient (95% CI) P Adjusted R 2
A. Multivariate model 0.012 0.048
Constant 4.77 (2.60 to 6.94) <0.001
Group 0.50 (−0.43 to 1.454) 0.292
% FFML −0.07 (−0.13 to −0.01) 0.017
FIGURE 1—Scatterplot for the association between the proportion of Multivariate linear regression analyses predicting weight regain at 1 yr. Each unstandardized
weight lost as % FFML with the weight loss intervention and weight re- β coefficient represents 1-kg weight regain at 1 yr per unit of the predictor. Variance inflation
gain at 1 yr. factors <1.1.

FAT-FREE MASS LOSS AND WEIGHT REGAIN Medicine & Science in Sports & Exercise® 2033

Copyright © 2022 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 compared with those with a positive % FFML (n = 53), denoting men, 3.0 kg (23%)), likely because of the shorter follow-up
an increase in FFM during weight loss (6.4 ± 4.9 vs 5.2 ± 3.1 kg, (6 months). Body composition in the present study was measured
P = 0.004). under conditions of weight stability, which is in line with the
CLINICAL SCIENCES

methodology followed by Vink et al. (18). This is opposite to

the study by Turicchi et al., where body composition was mea-
DISCUSSION
sured immediately after weight loss (in negative energy balance
The present secondary analysis aimed to determine the (EB)). Moreover, Turicchi et al. used an 800 kcal·d−1 diet. Neg-
association between % FFML during weight loss induced ative EB is accompanied by glycogen depletion and, with it,
Downloaded from http://journals.lww.com/acsm-msse by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

by diet alone, diet plus aerobic exercise, or diet plus resis- water loss, whereas refeeding is followed by glycogen replen-
tance exercise, on weight regain at 1-yr follow-up in women ishment and, with it, increased water content (27,28). Turicchi
with overweight. There was an inverse association between et al. (20) used DXA, a method that fails to account for changes
CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdgGj2MwlZLeI= on 02/08/2024

% FFML after weight loss and weight regain at 1 yr, with a in the hydration of lean tissue, and as such, losses of FFM
greater % FFML (or lower accretion) being associated with might be exaggerated when measurements are done under neg-
more weight regain. Moreover, those with a negative % FFML, ative EB, particularly when in glycogen-depleted conditions.
denoting loss of FFM during the weight loss intervention, The exact mechanisms through which % FFML modulates
regained significantly more weight at 1-yr follow-up (on aver- eating behavior and body weight homeostasis remain to be
age, >1.2 kg), compared with those with a positive % FFML fully elucidated. However, changes in the size and functional
(i.e., gained FFM during the intervention). Finally, % FFML was integrity of FFM may be involved (17,29). It has been sug-
a significant predictor of weight regain, even after adjusting gested that proteinstats, signals released from the muscle (both
for the intervention group. organs and skeletal muscle) during FFM loss, may act at the
Previous studies have found an association between % FFML level of the brain to modulate appetite control (30). In addition,
and weight regain in men (20) or in a sample composed FFM may also modulate the hedonic system. In a 12-wk aerobic
mainly (90%) of men (18), and in heavier samples: Vink et al. exercise intervention study, in young adults with overweight/
(average BMI, 31 kg·m−2) (18) and Turicchi et al. (weight, obesity, Flack and colleagues (31) showed a greater increase in
99.6 ± 16.3 kg) (20). Weight regain is likely to be weaker reward-driven feeding (using a behavioral choice task) in those
in fatter individuals, given that the proportion of weight lost who lost the greatest amount of FFM with the exercise interven-
as FFM decreases with increasing BMI and FM (25). More- tion (31). Another potential mechanism for increased weight re-
over, the association between % FFML and weight regain is gain after FFM loss may be because of reduced muscle function.
likely to be stronger in men, given that they have less FM It has been shown that ease of walking is related to participation
compared with women at baseline and are therefore more in free-living physical activity (14,15). It has also been estab-
likely to lose FFM after weight loss interventions (26). This lished that increased muscle and strength results in increased
study adds to the literature by showing an association be- ease of walking (16). It is thus possible that those individuals
tween % FFML and weight regain in premenopausal women who maintained muscle mass with weight loss were able to
with overweight. walk with less effort and were thus inclined to be more phys-
Previous studies on the association between % FFML and ically active. Indeed, we have previously shown that resistance
weight regain have used diet alone interventions to induce training during weight loss may decrease weight regain (32).
weight loss (18,20). The present analysis represents the first Regardless of the mechanisms, these results suggest that strat-
study where a combination of interventions was used, includ- egies to maintain, or increase, muscle mass during weight loss
ing diet alone, diet plus aerobic exercise, and diet plus resis- should be followed to limit weight regain in the long-term.
tance exercise. This helps to explain why average % FFML This study has both strengths and limitations. A strength of
values in the present study were only −3.6% ± 12.4%, −30% this study is that measurements of body composition were
in the study by Turicchi and colleagues (20), and −8.8% and done under conditions of weight stability at all time points.
−1.3% in the study by Vink et al. (18), in the very-low- This is important because it ensures that body composition re-
energy diet and low-energy diet groups, respectively. Exer- sults were not affected by negative EB and/or shifts in body
cise, in particular resistance exercise, has been shown to min- water compartments that accompany active weight loss. More-
imize the decrease in FFM that follows diet-induced weight over, all testing was conducted in the follicular phase of the
loss or even to increase FFM (15,21,22). The great variation menstrual cycle. Menstrual cycle is known to modulate body
in % FFML in the present study, from loss (−33%) to accretion composition, even when performed by DXA, with FM (in ki-
(28%), probably helped in finding an association between lograms and percent) being lower in the late follicular phase
% FFML and weight regain. compared with the midluteal phase in women in a natural men-
In the present study, weight regain at 1-yr follow-up was strual cycle (33). Finally, the present analysis displayed a large
6.0 ± 4.4 kg (51.3% ± 37.8%), ranging from −3.6 to 20.3 kg. range of % FFML from loss to accretion and a large range of
This is in line with the study by Vink et al. (18), 4.2 kg weight regain, important to establish an association between
(59%) in the low-energy diet group and 4.5 kg (55%) in the these two variables. However, this study also suffers from
very-low-energy diet group at 1-yr follow-up, but much larger some limitations. First, it includes a very homogenous sample
than in the study by Turicchi et al. (20) (1.6 kg (14%) overall; of premenopausal (20–41 yr) women with overweight. This

2034 Official Journal of the American College of Sports Medicine http://www.acsm-msse.org

Copyright © 2022 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 prevents the generalization of our results to men, other BMI suggesting that strategies such as resistance training during
groups, and older subjects. Second, this is a secondary analysis weight loss may be beneficial for future weight loss mainte-
of data, and as such, it is likely underpowered to examine the nance. Future research should try to identify the mecha-

CLINICAL SCIENCES
association between % FFML and weight regain. More stud- nisms, at the level of both appetite and EE, responsible for
ies, with larger sample sizes, a balanced sex distribution that this association.
allows for the investigation of potential important sex interac-
tions, and a large range of BMI and FM are needed. In addi-
tion, further research should use advanced methods of body Disclosure of funding from the National Institutes of Health: R01
Downloaded from http://journals.lww.com/acsm-msse by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

composition and multicompartment models to try to establish DK049779, P30 DK56336, P60 DK079626, and UL 1RR025777.
the underlying mechanisms involved. The authors declare no conflicts of interest. The results of this study
are presented clearly, honestly, and without fabrication, falsifica-
tion, or inappropriate data manipulation. The results of the present
CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdgGj2MwlZLeI= on 02/08/2024

CONCLUSIONS study do not constitute endorsement by the American College of

Sports Medicine.
In conclusion, % FFML is a significant predictor of weight ClinicalTrials.gov Identifier: NCT00067873.
regain at 1 yr in premenopausal women with overweight, URL: https://clinicaltrials.gov/ct2/show/NCT00067873.

REFERENCES
1. Anderson JW, Konz EC, Frederich RC, Wood CL. Long-term and unsuccessful at maintaining a normal body weight. Am J Clin
weight-loss maintenance: a meta-analysis of US studies. Am J Clin Nutr. 2002;75(3):499–504.
Nutr. 2001;74(5):579–84. 15. Hunter GR, McCarthy JP, Bamman MM. Effects of resistance train-
2. Kraschnewski JL, Boan J, Esposito J, et al. Long-term weight loss ing on older adults. Sports Med. 2004;34(5):329–48.
maintenance in the United States. Int J Obes (Lond). 2010;34(11): 16. Landers KA, Hunter GR, Wetzstein CJ, Bamman MM, Weinsier
1644–54. RL. The interrelationship among muscle mass, strength, and the
3. DeBenedictis JN, Nymo S, Ollestad KH, et al. Changes in the ho- ability to perform physical tasks of daily living in younger and
meostatic appetite system after weight loss reflect a normalization to- older women. J Gerontol A Biol Sci Med Sci. 2001;56(10):
ward a lower body weight. J Clin Endocrinol Metab. 2020;105(7): B443–8.
e2538–46. 17. Dulloo AG, Jacquet J, Girardier L. Poststarvation hyperphagia and
4. Martins C, Gower BA, Hunter GR. Baseline metabolic variables do body fat overshooting in humans: a role for feedback signals from
not predict weight regain in premenopausal women. Obesity (Silver lean and fat tissues. Am J Clin Nutr. 1997;65(3):717–23.
Spring). 2020;28(5):902–6. 18. Vink RG, Roumans NJ, Arkenbosch LA, Mariman EC, van Baak
5. Dulloo AG. Physiology of weight regain: lessons from the classic MA. The effect of rate of weight loss on long-term weight regain in
Minnesota Starvation Experiment on human body composition regu- adults with overweight and obesity. Obesity (Silver Spring). 2016;
lation. Obes Rev. 2021;22(2 Suppl):e13189. 24(2):321–7.
6. Dulloo AG, Jacquet J, Miles-Chan JL, Schutz Y. Passive and active 19. Turicchi J, O’Driscoll R, Finlayson G, Beaulieu K, Deighton K,
roles of fat-free mass in the control of energy intake and body compo- Stubbs RJ. Associations between the rate, amount, and composition
sition regulation. Eur J Clin Nutr. 2017;71(3):353–7. of weight loss as predictors of spontaneous weight regain in adults
7. Blundell JE, Caudwell P, Gibbons C, et al. Body composition and ap- achieving clinically significant weight loss: a systematic review and
petite: fat-free mass (but not fat mass or BMI) is positively associated meta-regression. Obes Rev. 2019;20(7):935–46.
with self-determined meal size and daily energy intake in humans. Br 20. Turicchi J, O’Driscoll R, Finlayson G, et al. Associations between the
J Nutr. 2012;107(3):445–9. proportion of fat-free mass loss during weight loss, changes in ap-
8. Cameron JD, Sigal RJ, Kenny GP, et al. Body composition and en- petite, and subsequent weight change: results from a randomized
ergy intake—skeletal muscle mass is the strongest predictor of food 2-stage dietary intervention trial. Am J Clin Nutr. 2020;111(3):
intake in obese adolescents: the HEARTY trial. Appl Physiol Nutr 536–44.
Metab. 2016;41(6):611–7. 21. Hunter GR, Byrne NM, Sirikul B, et al. Resistance training conserves
9. Weise CM, Hohenadel MG, Krakoff J, Votruba SB. Body composi- fat-free mass and resting energy expenditure following weight loss.
tion and energy expenditure predict ad-libitum food and macronutri- Obesity (Silver Spring). 2008;16(5):1045–51.
ent intake in humans. Int J Obes (Lond). 2014;38(2):243–51. 22. Campbell WW, Crim MC, Young VR, Evans WJ. Increased energy
10. Weise CM, Thiyyagura P, Reiman EM, Chen K, Krakoff J. A poten- requirements and changes in body composition with resistance train-
tial role for the midbrain in integrating fat-free mass determined en- ing in older adults. Am J Clin Nutr. 1994;60(2):167–75.
ergy needs: an H2(15)O PET study. Hum Brain Mapp. 2015;36(6): 23. Weinsier RL, Hunter GR, Zuckerman PA, et al. Energy expenditure
2406–15. and free-living physical activity in Black and White women: com-
11. Tekin S, Erden Y, Ozyalin F, Cigremis Y, Colak C, Sandal S. The ef- parison before and after weight loss. Am J Clin Nutr. 2000;71(5):
fects of intracerebroventricular infusion of irisin on feeding behaviour 1138–46.
in rats. Neurosci Lett. 2017;645:25–32. 24. Hall KD. Body fat and fat-free mass inter-relationships: Forbes’s the-
12. Caudwell P, Finlayson G, Gibbons C, et al. Resting metabolic rate is ory revisited. Br J Nutr. 2007;97(6):1059–63.
associated with hunger, self-determined meal size, and daily energy 25. Dulloo AG, Jacquet J. The control of partitioning between protein
intake and may represent a marker for appetite. Am J Clin Nutr. and fat during human starvation: its internal determinants and biolog-
2013;97(1):7–14. ical significance. Br J Nutr. 1999;82(5):339–56.
13. Piaggi P, Thearle MS, Krakoff J, Votruba SB. Higher daily energy 26. Millward DJ, Truby H, Fox KR, Livingstone MB, Macdonald IA,
expenditure and respiratory quotient, rather than fat-free mass, inde- Tothill P. Sex differences in the composition of weight gain and
pendently determine greater ad libitum overeating. J Clin Endocrinol loss in overweight and obese adults. Br J Nutr. 2014;111(5):
Metab. 2015;100(8):3011–20. 933–43.
14. Weinsier RL, Hunter GR, Desmond RA, Byrne NM, Zuckerman PA, 27. Olsson KE, Saltin B. Variation in total body water with muscle gly-
Darnell BE. Free-living activity energy expenditure in women successful cogen changes in man. Acta Physiol Scand. 1970;80(1):11–8.

FAT-FREE MASS LOSS AND WEIGHT REGAIN Medicine & Science in Sports & Exercise® 2035

Copyright © 2022 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 28. Kreitzman SN, Coxon AY, Szaz KF. Glycogen storage: illusions of 31. Flack KD, Hays HM, Moreland J. The consequences of exercise-induced
easy weight loss, excessive weight regain, and distortions in estimates weight loss on food reinforcement. A randomized controlled trial.
of body composition. Am J Clin Nutr. 1992;56(1 Suppl):292S–3S. PLoS One. 2020;15(6):e0234692.
CLINICAL SCIENCES

29. Stubbs RJ, Hopkins M, Finlayson GS, Duarte C, Gibbons C, 32. Hunter GR, Brock DW, Byrne NM, Chandler-Laney PC, Del
Blundell JE. Potential effects of fat mass and fat-free mass on energy Corral P, Gower BA. Exercise training prevents regain of visceral
intake in different states of energy balance. Eur J Clin Nutr. 2018; fat for 1 year following weight loss. Obesity (Silver Spring). 2010;
72(5):698–709. 18(4):690–5.
30. Dulloo AG, Jacquet J, Montani JP, Schutz Y. How dieting makes the 33. Thompson BM, Hillebrandt HL, Sculley DV, Barba-Moreno L, Janse
lean fatter: from a perspective of body composition autoregulation de Jonge XAK. The acute effect of the menstrual cycle and oral con-
through adipostats and proteinstats awaiting discovery. Obes Rev. traceptive cycle on measures of body composition. Eur J Appl Physiol.
Downloaded from http://journals.lww.com/acsm-msse by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

2015;16(1 Suppl):25–35. 2021;121(11):3051–9.

CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdgGj2MwlZLeI= on 02/08/2024

2036 Official Journal of the American College of Sports Medicine http://www.acsm-msse.org

Copyright © 2022 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.