Metabolic Damage: do Negative Metabolic Adaptations During Underfeeding Persist After Refeeding in Non-Obese Populations?

Abstract

Several researches have proposed that a prolonged period of caloric restriction (CR) may have a permanent, adverse effect on basal metabolism, fostering the development of obesity. This reported metabolic slowing has been associated with a reduction in resting metabolic rate (RMR) beyond what is predicted by the change in body composition, promoting the idea that metabolism can be permanently damaged. This systematic review investigates if prolonged CR exhibits a permanent, negative effect on basal metabolism.
Here we review the literature reporting weight loss and weight regain of individuals who were initially within a healthy weight range, such as the long-term Minnesota starvation experiment, in addition to research on chronically undernourished individuals, such as patients with anorexia nervosa, before and after recovery. Quantification of basal metabolism before and after prolonged CR revealed that body composition is the most critical factor in determining absolute RMR in neutral energy balance. Changes in energy balance induce a rapid yet reversible increase or decrease in RMR. Previous reports may have come to erroneous conclusions in favor of the metabolic damage hypothesis because they did not examine the full recovery period in the Minnesota experiment or neglected the influence of energy balance on RMR. Our findings indicate that the theory of permanent, diet-induced metabolic slowing in non-obese individuals is not supported by the current literature.

Full text

Metabolic Damage: Do Negative Metabolic Adaptations During
Underfeeding Persist After Refeeding in Non-Obese Populations?
Authors:
Anastasia Zinchenko¹*
Menno Henselmans¹
1. Bayesian Bodybuilding R&D
Department, Bayesian
Bodybuilding, Lingsesdijk 46B,
4207 AE Gorinchem,
The Netherlands.
* Correspondence:
E-mails:
a.zinchenko@live.de
MennoHenselmans@Bayesian
Bodybuilding.com
Key words: metabolic damage,
metabolic slowing, weight regain,
weight loss, weight cycling
Abstract
Several researches have proposed that a prolonged period
of caloric restriction (CR) may have a permanent, adverse
effect on basal metabolism, fostering the development of
obesity. This reported metabolic slowing has been
associated with a reduction in resting metabolic rate
(RMR) beyond what is predicted by the change in body
composition, promoting the idea that metabolism can be
permanently damaged. This systematic review investigates
if prolonged CR exhibits a permanent, negative effect on
basal metabolism.
This paper reviews literature reporting weight loss and
weight regain of individuals who were initially within a
healthy weight range, such as the long-term Minnesota
starvation experiment, in addition to research on
chronically undernourished individuals, such as patients
with anorexia nervosa, before and after recovery.
Quantification of basal metabolism before and after
prolonged CR revealed that body composition is the most
critical factor in determining absolute RMR in neutral
energy balance. Changes in energy balance induce a rapid
yet reversible increase or decrease in RMR. Previous
reports may have come to erroneous conclusions in favor
of the metabolic damage hypothesis because they did not
examine the full recovery period in the Minnesota
experiment or neglected the influence of energy balance
on RMR. Our findings indicate that the theory of
permanent, diet-induced metabolic slowing in non-obese
individuals is not supported by the current literature.

Medical Research Archives, Vol. 4, Issue 8, December 2016
Metabolic Damage: Do Negative Metabolic Adaptations During Underfeeding
Persist After Refeeding in Non-Obese Populations?
Copyright 2016 KEI Journals. All Rights Reserved Page │2
1. Introduction
Previous research has suggested that
a prolonged period of caloric restriction
(CR) leads to a permanent decrease in
resting metabolic rate (RMR) that goes
beyond the value expected from the
present body composition [1–4].
Metabolic slowing, originating from
adaptive thermogenesis (AT) in a calorie-
restricted state, was hypothesized to be
irreversible upon the return to neutral
energy balance [1] and lead to „metabolic
damage‟. Metabolic damage refers to a
weight loss induced decrease in RMR that
is beyond the value expected from the
present body composition and persists
after weight regain.
It was suggested that metabolic
damage is responsible for post-diet weight
regain of overweight and obese
individuals if weight-reduced individuals
do not maintain a high level of physical
activity or a calorie restricted state [1]. A
recent study examining RMR of
participants of “The Biggest Loser”
challenge, who lost on average 58 kg in
30 weeks and regained approximately 41
kg in the following 6 years, found that the
participants had a significantly lower
RMR relative to their pre-dieting level
even after nearly reverting to their initial
body composition [4].
Reports reanalyzing the data from
the Minnesota experiment [5] suggested
that permanent metabolic slowing also
occurs in lean individuals. In this study,
32 healthy young men underwent severe
caloric restriction for 24 weeks with a
targeted weight loss of approximately 1 kg
per week. The weight regain phase
consisted of two phases: a 12-week
controlled rehabilitation period and an 8-
week ad libitum period, in which the
subjects did not follow a prescribed diet;
however, their food intake was still
recorded. After reanalyzing the data from
this experiment, Dulloo and coworkers
observed a disproportionate rate of fat
recovery relative to fat-free mass (FFM)
recovery and proposed that it is the reason
for favorable fat gain after CR [6–8].
Furthermore, the same researchers

Medical Research Archives, Vol. 4, Issue 8, December 2016
Metabolic Damage: Do Negative Metabolic Adaptations During Underfeeding
Persist After Refeeding in Non-Obese Populations?
Copyright 2016 KEI Journals. All Rights Reserved Page │3
suggested that weight loss diets make “the
lean fatter,” because CR-induced feedback
signals suppress RMR normalization until
the initial fat mass (FM) is fully recovered
(“fat-stores memory”) [8, 7]. The post-diet
shift towards an unfavorable body
composition may result in repeated dieting
attempts and increase the risk of becoming
overweight [8]. Indeed, repeated weight
cycling has been associated with weight
gain over time [9] and is speculated to
increase the risk of all-cause and
cardiovascular mortality [10].
Considering that the majority of the
population in Western countries is
dissatisfied with their appearance [11] and
about one half of individuals who are
within the reference weight range reported
undergoing weight loss interventions in
order to lose weight [12], a permanent
RMR reduction after dieting may present
a health hazard to the general population.
For this reason, this paper investigates
whether diet-induced metabolic slowing is
responsible for the difficulty in
maintaining reduced body weight.
Particular attention was devoted to
research on RMR related to the present
body composition of subjects who
previously underwent a prolonged period
of CR but were no longer in an energy-
restricted state. This context aimed to
control for body composition and
eliminate transient effects of CR, such as
the CR induced reduction in spontaneous
physical activity [13], that influence
energy expenditure. If a sustained
suppression of RMR occurs after weight
loss dieting, then a return to the initial
body composition should result in a lower
RMR relative to the baseline and
compared to weight-matched control
subjects without a history of weight
cycling („metabolic damage‟).
2. Methods
2.1. Used terms
In this review, the terms resting
metabolic rate (RMR), resting energy
expenditure (REE), basal metabolic rate
(BMR), basal energy expenditure (BEE),
sleeping metabolic rate (SMR) and
sleeping energy expenditure (SEE) are

Medical Research Archives, Vol. 4, Issue 8, December 2016
Metabolic Damage: Do Negative Metabolic Adaptations During Underfeeding
Persist After Refeeding in Non-Obese Populations?
Copyright 2016 KEI Journals. All Rights Reserved Page │4
used interchangeably to report previous
findings that differed in terminology but
used a comparable experimental
measurement methodology, specifically
indirect calorimetry after an overnight
fast.
2.2. Literature search
An electronic literature search was
conducted in the PubMed database using
the keywords caloric restriction, caloric
deficit, weight loss, diet, starvation,
underfeeding, malnourished, anorexia,
anorectic, weight regain, weight gain,
refeed, overfeeding, weight change,
weight maintenance, basal metabolic rate,
basal metabolism, metabolic rate,
metabolism, energy expenditure, BMR,
BEE, RMR, REE, SMR and SEE in any
combination using the Boolean operators
AND and OR. Titles and abstracts of
human studies published until December
2015 were assessed for their relevance for
this review.
Articles were selected when they
contained the information required to
determine changes of FM, FFM and RMR
following weight loss and subsequent
weight regain or weight regain relative to
the weight-reduced state for malnourished
subjects.
Bibliographies of relevant original
research articles and reviews were
screened for additional relevant
references.
2.3. Data analysis
To test the metabolic slowing
hypothesis, three equations were used that
were previously used to calculate RMR in
the literature [14–16]. The fat mass
percentage (FM%) based equations
developed by Bosy-Westphal and
coworkers [14], with each equation
specific to a particular FM% range, are
particularly suitable for the RMR
prediction of individuals who experienced
significant weight loss with a subsequent
weight regain. These equations were
shown to give a comparable precision to
accurate organ mass based RMR
predictions, unlike many other formulas
that only incorporate weight or only lean
body mass and fat mass [17]. Potential

Medical Research Archives, Vol. 4, Issue 8, December 2016
Metabolic Damage: Do Negative Metabolic Adaptations During Underfeeding
Persist After Refeeding in Non-Obese Populations?
Copyright 2016 KEI Journals. All Rights Reserved Page │5
error in the prediction of absolute RMR
due to individual variability [18] or
inaccuracies resulting from differences in
body composition measurements (e.g.
DEXA vs. BIA) [17] can be eliminated by
using predicted RMR as a normalization
factor [17]. If the measured RMR differs
from the predicted value, then the
difference between the predicted and
measured value should stay constant
throughout the study. Thus, the RMRm/p-
ratio of measured to predicted RMR – as
used previously [16] – was used to relate
the metabolic state of subject groups
before and after weight loss across a
number of studies.
𝑅𝑀𝑅𝑚/𝑝− 𝑟𝑎𝑡𝑖𝑜 =𝑅𝑀𝑅𝑚𝑒𝑎𝑠𝑢𝑟𝑒𝑑
𝑅𝑀𝑅𝑝𝑟𝑒𝑑𝑖𝑐𝑡𝑒𝑑
A constant RMRm/p-ratio throughout
the weight change intervention suggests
that the metabolic rate changes in
proportion to the changing body
composition. A decrease in RMRm/p-ratio
indicates a decreased basal metabolism
beyond what is expected based on the
body composition change; an increase
indicates an increase in metabolic rate
beyond what is expected based on the
body composition change.
For the conversion of energy units,
the relationship 1 kcal = 4.184 kJ was
used. RMR values for Minnesota
experiment subjects were calculated using
the factor 20.5 kJ/L oxygen [6].
2.4. Statistical analysis
Statistical analysis was conducted
using IBM SPSS Statistics 20.0.0 with
additional manual calculations in Excel
from Microsoft Office Professional Plus
2013. Data are presented as means (M)
plus standard deviations (SD) unless
otherwise specified.

Medical Research Archives, Vol. 4, Issue 8, December 2016
Metabolic Damage: Do Negative Metabolic Adaptations During Underfeeding
Persist After Refeeding in Non-Obese Populations?
Copyright 2016 KEI Journals. All Rights Reserved Page │6
3. Results
3.1. Metabolic damage during
weight loss and regain interventions in
normal weight individuals
To examine the relationship
between RMR and body composition over
time, we calculated the RMRm/p-ratio for
the subjects of the Minnesota experiment
(Table 1). The data calculated with all
three equations [14–16] was consistent in
showing that the RMRm/p-ratio decreased
over the semi-starvation period compared
to baseline and subsequently increased in
correspondence with body weight regain
over the controlled 12-week reefed period.
Two months after the subjects transitioned
to the ad libitum diet, the RMRm/p-ratio
calculated with the equations reported by
Bosy-Westphal and coworkers [14]
returned exactly to the baseline value
without a significant difference between
baseline and the end of the recovery
period (p = .98). The use of the equation
adapted from Doucet and coworkers [15]
showed a trend towards a higher post-
dieting RMRm/p-ratio (t(10) = 2.09,
p = .06). The calculations based on the
third equation [16] gave a significantly
higher RMRm/p-ratio after semi-starvation
and subsequent weight regain compared to
baseline (t(10) = 2.61, p = .03).
Collectively, these data suggest that the
subjects maintained their relative
metabolic rate or even experienced an
increase in their body composition
adjusted basal metabolism compared to
baseline. In fact, despite incomplete FFM
recovery after 20 weeks of rehabilitation
(t(10) = 2.7, p = .02), the subjects‟
measured RMR was significantly higher
than at baseline (t(10) = 2.42, p = .04).

Medical Research Archives, Vol. 4, Issue 8, December 2016
Metabolic Damage: Do Negative Metabolic Adaptations During Underfeeding
Persist After Refeeding in Non-Obese Populations?
Copyright 2016 KEI Journals. All Rights Reserved Page │7
Table 1. Changes in body composition and metabolic parameters over the
course of the Minnesota experiment
Body weight (kg)
FFM
(kg)
RMR (MJ)
(RMR (kcal))
M
SD
M
SD
M
SD
Baseline, n = 32
69.4
5.8
59.5
4.2
6.70
(1601)
0.40
(95)
S - 24 wk, n = 32
52.3
4
49.5
3.5
4.09
(977)
0.41
(98)
R - 12 wk, n = 32
59.1
3.9
52.7
3.7
5.38
(1285)
0.46
(111)
Recovery %
Total average
energy intake for R12
836 MJ
30
27
34
33
29
16
947 MJ
30
8
21
10
47
13
1097 MJ
42
17
36
11
56
17
1192 MJ
59
16
32
15
65
10
R - 20 wk, n - 20
70.5
4.5
57
3.4
7.06
(1688)
0.50
(119)
Baseline20
67
5
58.3
3.4
6.68
(1596)
0.35
(83)
RMRm/p
Calculated with equation
based on
Bosy-Westphal et
al. [14]
Camps et al.
[16]
Doucet et al.
[15]
M
SD
M
SD
M
SD
Baseline, n = 32
1.09
0.07
0.94
0.05
1.01
0.05
S - 24 wk, n = 32
0.83
0.08
0.69
0.07
0.73
0.07
R - 12 wk, n = 32
0.98
0.07
0.85
0.08
0.90
0.08
R - 20 wk, n - 20
1.13
0.05
1.08
0.05
1.01
0.05
Baseline20
1.13
0.07
1.03
0.05
0.95
0.05
M – mean; SD – standard deviation; S - 24 end of the semi-starvation period of 24
weeks; R -12 after 12 weeks controlled refeed;
R -20 after 20 weeks recovery (12 weeks controlled and 8 weeks ad libitum diet), n =
subject number, recovery % - percentage of the regained weight during controlled
refeed relative to the weight lost in semi-starvation period (recovery % = 100*[R12-
S24]/[baseline - S24]), Baseline20 – baseline data for the 20 subjects examined at the
time point R – 20

Medical Research Archives, Vol. 4, Issue 8, December 2016
Metabolic Damage: Do Negative Metabolic Adaptations During Underfeeding
Persist After Refeeding in Non-Obese Populations?
Copyright 2016 KEI Journals. All Rights Reserved Page │8
The rate of body fat gain and RMR
recovery in the controlled rehabilitation
period was influenced by energy intake
(Table 1). The subjects were divided into
4 groups that received diets differing in
caloric content. Initially, in the
experimental design, it was planned that
the groups‟ energy intakes increased in 1.6
MJ (400 kcal) increments, from the lowest
energy group consuming 400 kcal more
than each subject‟s estimated energy
maintenance requirements at the end of
starvation to a group consuming a diet
with an energy content of 6.7 MJ (1600
kcal) above energy maintenance
requirements. However, the subsequent
analysis of the diets showed that the
groups received supplements of a lower
calorific value than planned and the two
lowest calorie groups recovered only to a
small degree in the first 6 weeks. For this
reason, the energy intake was increased in
the second half of the controlled
rehabilitation period. In the entire 12-
week rehabilitation phase, not only body
weight, but also the RMRm/p-ratio, which
inherently controls for body composition,
increased in relation to energy intake
(Figure 1). The group with the highest
energy intake had a significantly higher
RMRm/p-ratio compared to the group with
the lowest energy intake (t(14) = 3.23, p =
.006). The RMRm/p-ratio correlated
strongly and significantly with total
energy intake in the refeeding period
using the equation based on Doucet et al.
[15] (r = .97, p = .034). For the other two
equations [14, 16], a clear trend was
observed for the association between
energy intake and the RMRm/p-ratio
(r = .95, p = .054; r = .94, p = .06;
respectively).
Previous computational modeling of
energy metabolism during semi-starvation
and refeeding also indicated that RMR is
higher in positive energy balance
compared to a calorie-restricted state
despite identical body composition [19].
The RMR in week 12 of the semi-
starvation phase was significantly lower
than the measured RMR in recovery week
12 despite identical FFM. The validity of
this finding remained after controlling for
body composition and including FM in the
model. Analysis based on all of our
equations [14–16] showed that the

Medical Research Archives, Vol. 4, Issue 8, December 2016
Metabolic Damage: Do Negative Metabolic Adaptations During Underfeeding
Persist After Refeeding in Non-Obese Populations?
Copyright 2016 KEI Journals. All Rights Reserved Page │9
RMRm/p-ratio during semi-starvation was
significantly lower than after 12 weeks of
controlled refeeding (Table 2). Moreover,
in the lowest energy group, the RMRm/p-
ratio in the recovery period was
significantly higher than during semi-
starvation even though the subjects had
significantly lower average FFM and
identical FM. Collectively, these findings
suggest that energy intake has an acute
effect on RMR.
Figure 1. Changes in the RMRm/p-ratio related to energy intake. Different energy intake
levels during the 12 week recovery period (gray) led to different recovery rates of the
RMRm/p-ratio. Despite an identical starting point after semi-starvation (white) and controlling
for body composition, groups receiving diets of a higher caloric content showed faster
RMRm/p-ratio recovery. Different shapes represent the RMRm/p-ratio values calculated using
different equations; circles - Bosy-Westphal et al. [14], squares - Camps et al. [16] and
rectangles - Doucet et al. [15].

Medical Research Archives, Vol. 4, Issue 8, December 2016
Metabolic Damage: Do Negative Metabolic Adaptations During Underfeeding
Persist After Refeeding in Non-Obese Populations?
Copyright 2016 KEI Journals. All Rights Reserved Page │10
Table 2. Paired samples t-tests for the RMR-ratio, calculated with three RMR
prediction equations, and body composition for week 12 of the semi-starvation
period and week 12 of the recovery period
Starvation
week 12
Recovery
week 12
All subjects (n = 32)
M
SD
M
SD
t
df
p value
RMR-ratio calculated with
equation based on
Bosy-Westphal et al. [14]
0.86
0.07
0.98
0.07
8.56
31
< .001
Camps et al. [16]
0.72
0.07
0.85
0.08
9.40
31
< .001
Doucet et al. [15]
0.77
0.07
0.90
0.08
9.12
31
< .001
Low-calorie group (n = 8)
M
SD
M
SD
t
df
p value
FM (kg)
4.84
3.63
5.05
3.09
0.31
7
.76
FFM (kg)
52.1
4.24
50.4
4.32
2.49
7
.04
RMR (MJ/d)
RMR (kcal/d)
4464
1067
378
90.3
4841
1157
250
59.8
2.08
7
.08
RMR-ratio calculated with
equation based on
Bosy-Westphal et al. [14]
0.84
0.07
0.92
0.07
3.37
7
.01
Camps et al. [16]
0.71
0.08
0.80
0.07
3.28
7
.01
Doucet et al. [15]
0.76
0.08
0.84
0.07
3.04
7
.02
M – mean; SD – standard deviation; t – paired t-test statistics; df – degrees of freedom.
The results of other studies from the
literature search are in line with the above
data. Grande and coworkers subjected 25
young, normal weight men to a three week
long caloric restriction phase (4.18 MJ/d
(1000 kcal/d)) followed by a 20 day
period of overfeeding (≥ 22.2 MJ/d (5300
kcal/d)) [20]. The subjects experienced
significant weight loss and a decrease in
RMR. Eight days after transitioning to an
energy surplus diet, the subjects` RMR
returned to baseline, even though their
body weight was still 3.2 kg below
baseline.
Müller et al. [21] performed a
sequential weight change experiment

Medical Research Archives, Vol. 4, Issue 8, December 2016
Metabolic Damage: Do Negative Metabolic Adaptations During Underfeeding
Persist After Refeeding in Non-Obese Populations?
Copyright 2016 KEI Journals. All Rights Reserved Page │11
consisting of a seven day long overfeeding
period, a 21 day long energy restriction
period and a subsequent 14 day long
refeeding period. Bodyweight, RMR and
RMRm/p-ratio increased as a result of
overfeeding (50% above energy
requirements) and decreased when the
subjects underwent severe CR (50%
below energy requirements). At the end of
the final refeeding period, when the
subjects were still in an energy surplus,
the RMRm/p-ratio was above the baseline
value (1.22 vs. 1.20), despite identical
body composition compared to baseline.
Similar results were obtained from
five individuals who spent two years in a
biosphere with a restricted energy intake
[13]. One week after the biosphere exit,
total sleeping energy expenditure (SEE)
was significantly lower compared to
control subjects. However, after reverting
to a body composition similar to that of
the control subjects, the difference was no
longer present.
3.2. Metabolic damage in
malnourished individuals
An extreme case of long-term
energy restriction occurs in chronically
undernourished individuals, such as
patients with anorexia nervosa or
malnourished populations with a poor
socioeconomic status. The RMR of these
individuals, when expressed in absolute
terms, is lower compared to well-
nourished, healthy individuals [22–26].
However, accounting for body
composition offsets the relative RMR
difference (Table 3). The weighted
average of the RMRm/p-ratio, calculated
with both the equation from Bosy-
Westphal and coworkers [14] and Doucet
and coworkers [15], showed that the RMR
of malnourished individuals corresponds
to their body composition, whereas a
slightly lower RMRm/p-ratio was found
using the equation from Camps and
coworkers [16]. The total RMR increases
in some cases even exceeded the predicted
values when malnourished individuals
regained weight [27–29]. Additionally, the
absolute RMR value of weight recovered
patients with anorexia nervosa does not

Medical Research Archives, Vol. 4, Issue 8, December 2016
Metabolic Damage: Do Negative Metabolic Adaptations During Underfeeding
Persist After Refeeding in Non-Obese Populations?
Copyright 2016 KEI Journals. All Rights Reserved Page │12
differ from weight-matched individuals
with no history of eating disorders [24,
30]. As such, metabolic damage does not
appear to reliably occur in malnourished
populations.
Table 3. The ratio of measured to predicted RMR (RMRm/p-ratio) calculated with
three different RMR prediction equations
Bosy-Westphal
et al. [14]
Camps et al. [16]
Doucet et al. [15]
Subjects, sample size
1.01
1.04
1.09
Anorectic women, 6
[24]
1.00
0.91
0.95
Anorectic women, 87
[28]
0.70
0.73
0.76
Anorectic women, 10
[29]
0.95
0.99
1.04
Anorectic women, 12
[23]
0.79
0.83
0.87
Anorectic women, 25
[25]
1.30
1.13
1.14
Severely undernourished men, 30
[22]
1.03
1.12
1.16
Severely undernourished women, 25
[22]
Weighted average RMRm/p-ratio
1.00
0.96
1.00

Medical Research Archives, Vol. 4, Issue 8, December 2016
Metabolic Damage: Do Negative Metabolic Adaptations During Underfeeding
Persist After Refeeding in Non-Obese Populations?
Copyright 2016 KEI Journals. All Rights Reserved Page │13
3.3. Metabolic damage in
competitive athletes
More data contradicting diet-
induced metabolic suppression comes
from two case studies of natural
bodybuilders undergoing contest
preparation [31, 32]. These athletic, young
males consumed a calorie-restricted diet
for up to six months to achieve body fat
levels of 4.5% and 7.5%. In both cases,
total RMR decreased along with weight
loss. In the study conducted by Robinson
and coworkers [32], the actual RMR
exceeded the predicted values during the
course of the entire study with a constant
RMRm/p-ratio of 1.19 - 1.20, suggesting
that the athlete did not experience
negative metabolic adaptations in the state
of caloric restriction beyond what is
expected from his changes in body
composition.
In contrast, another athlete [31]
experienced an RMRm/p-ratio decrease in
the weight loss phase. However, post-
contest weight regain increased his
RMRm/p-ratio. In fact, the ratio was
identical 13 weeks before and after the
competition (0.87) with a value in
between baseline (1.28) and after 26
weeks caloric restriction (0.74). Figure 2
depicts the body composition and
metabolic changes over time of both
athletes.
The same scenario was observed in
12 competitive wrestlers in a seven-month
weight cycling period [33]. These athletes
lost on average 4.8 kg during the
competition season and regained 6.8 kg in
five to six weeks after the last
competition. While in an energy deficit,
their average RMRm/p-ratio decreased
from 1.29 to 1.12; however, it increased to
1.31 upon weight regain.

Medical Research Archives, Vol. 4, Issue 8, December 2016
Metabolic Damage: Do Negative Metabolic Adaptations During Underfeeding
Persist After Refeeding in Non-Obese Populations?
Copyright 2016 KEI Journals. All Rights Reserved Page │14
Figure 2. Changes in FFM (gray), FM (black) and RMRm/p-ratio (white) during weight loss
(athlete 1 and 2) and weight regain (athlete 2). Week 0 indicates the end of caloric restriction
for both athletes.
4. Discussion
Our findings contradict the
hypothesis that basal metabolism
permanently slows down to favor fat gain
or is permanently damaged by undergoing
a prolonged period of caloric restriction
[1, 7].

Medical Research Archives, Vol. 4, Issue 8, December 2016
Metabolic Damage: Do Negative Metabolic Adaptations During Underfeeding
Persist After Refeeding in Non-Obese Populations?
Copyright 2016 KEI Journals. All Rights Reserved Page │15
The discrepancy between our
findings and previous reports has several
potential reasons. Previous studies
revising the Minnesota experiment did not
analyze the complete recovery period and
disregarded the varying recovery
conditions, i.e. energy balance, among the
subjects. The 12-week post-starvation
recovery period analyzed by Dulloo and
coworkers [7, 34] was not sufficient for
full metabolic recovery: full restoration of
the RMRm/p-ratio occurred after 20 weeks
of refeeding. The lack of complete
recovery within 12 weeks was not
surprising, considering that in the first half
of the 12-week controlled recovery period
the subjects received diets containing
fewer calories than initially planned and
one half of the subjects rehabilitated only
to a small degree. Notably, Dulloo and
coworkers [7] did not take into account
that the subjects were divided into 4
groups that received diets of a different
caloric content during the 12 weeks of
controlled rehabilitation. The groups
receiving diets of a higher caloric content,
consuming up to 4.2 MJ/d (1014 kcal/d)
more than the low calorie group,
experienced faster body weight and RMR
recovery than the low calorie group. Not
taking into account this significant
difference in energy intake may be the
reason why the researchers [7, 34]
concluded that RMR recovers in relation
to the degree of FM recovery without
being influenced by the recovery of the
more metabolically active FFM. The
findings here provide evidence that faster
RMR recovery and higher FM
replenishment resulted from a higher
caloric intake (Figure 1), contradicting the
idea that faster RMR recovery is primarily
a function of fat store replenishment.
The results here challenge the
hypothesis of a “metabolic memory,”
which states that thermogenesis is
suppressed until fat stores revert to their
initial size (“fat-stores memory”) [7]. The
findings here suggest that the suppression
of basal metabolism, which is a
component of thermogenesis [35], is a
function of caloric intake and changes in
body composition. Even though the
subjects in the low-calorie group did not
experience full metabolic recovery due to
their low energy intake, their RMRm/p-

Medical Research Archives, Vol. 4, Issue 8, December 2016
Metabolic Damage: Do Negative Metabolic Adaptations During Underfeeding
Persist After Refeeding in Non-Obese Populations?
Copyright 2016 KEI Journals. All Rights Reserved Page │16
ratio was still significantly higher in the
recovery period than in the semi-
starvation period, despite a less favorable
body composition (Table 2). In contrast,
positive energy balance following CR
readily undoes short-term AT and restores
thyroid hormone levels and SNS activity
[21, 28]. Thus, RMR measurements taken
in a calorie-restricted state, without
considering the influence of energy
balance on RMR, may potentially yield
misleading results in favor of the
metabolic damage hypothesis.
Another reason for the finding of
purported metabolic damage in previous
studies is that RMR predictions in favor of
the „metabolic slowing‟ hypothesis that
showed a decrease in RMR originated
from weight loss studies on obese
individuals. These studies appear to have
overestimated the actual RMR [1] of post-
obese individuals due to the fact that their
predictions were based on FM and FFM
changes without accounting for changes in
FFM relative to organ mass, as described
earlier [17, 36].
It is important to note that AT was
frequently observed in well-nourished
individuals subjected to a period of acute
energy restriction [37], even though here
the RMR of chronically undernourished
individuals corresponded to their body
composition (Table 3), suggesting that AT
was absent during chronic CR [37].
A limitation of this review is that the
data on individual subjects is lacking for
the majority of research studies. As such,
data on body composition and metabolic
parameters were only suitable for
statistical analysis in the Minnesota
experiment. Additionally, pre-anorexic
data on RMR from anorexia nervosa
patients are absent, making before and
after RMR comparisons impossible.
Another limitation of the current research
is that some research studies were limited
either by their length, consisting of a total
three week diet- and refeeding period, or
contrasted by single-case studies done on
bodybuilders that are limited in their
sample size. Future research may further
our understanding of human metabolic
changes by investigating multiple weight
cycling phases in various populations with
sufficient sample sizes to achieve
satisfactory statistical power.

Medical Research Archives, Vol. 4, Issue 8, December 2016
Metabolic Damage: Do Negative Metabolic Adaptations During Underfeeding
Persist After Refeeding in Non-Obese Populations?
Copyright 2016 KEI Journals. All Rights Reserved Page │17
5. Conclusion
Body composition is the most
influential determinant of RMR in neutral
energy balance. Short-term changes in
energy balance induce metabolic
adaptations that change RMR
correspondingly. During long-term caloric
restriction, AT occurs and RMR often
decreases relative to FM and FFM loss
[37]. Positive energy balance undoes the
negative metabolic effects of AT and
causes RMR to return to a level that is
appropriate for the body composition at
that time. The findings here show that
human metabolism is highly plastic and
rapidly adapts to changes in energy
availability and body composition. This
stands in contrast to the hypothesis of an
inflexible metabolism that is susceptible
to metabolic damage during prolonged
caloric restriction. As such, the presence
of metabolic damage in non-obese
individuals is not supported by the current
literature.
Conflict of interest: none
Financial support: none

Medical Research Archives, Vol. 4, Issue 8, December 2016
Metabolic Damage: Do Negative Metabolic Adaptations During Underfeeding
Persist After Refeeding in Non-Obese Populations?
Copyright 2016 KEI Journals. All Rights Reserved Page │18
6. References
1. Johannsen DL, Knuth ND, Huizenga
R et al. Metabolic Slowing with
Massive Weight Loss despite
Preservation of Fat-Free Mass. J. Clin.
Endocrinol. Metab. 2012; 97(7):2489–
2496. [PubMed: 22535969]
2. Rosenbaum M, Hirsch J, Gallagher
DA, Leibel RL. Long-term persistence
of adaptive thermogenesis in subjects
who have maintained a reduced body
weight. Am. J. Clin. Nutr. 2008;
88(4):906–912. [PubMed: 18842775]
3. Astrup A, Gøtzsche PC, van de
Werken K et al. Meta-analysis of
resting metabolic rate in formerly
obese subjects. Am. J. Clin. Nutr.
1999; 69:1117–1122. [PubMed:
10357728]
4. Fothergill E, Guo J, Howard L et al.
Persistent metabolic adaptation 6 years
after “The Biggest Loser” competition.
Obesity 2016; 24(8):1612–1619.
[PubMed: 27136388]
5. Keys A, Brozek J, Henschel A et al.
The Biology of Human Starvation,
University of Minnesota Press,
Minneapolis, MN., 1950.
6. Dulloo AG, Jacquet J. Adaptive
reduction in basal metabolic rate in
response to food deprivation in
humans: A role for feedback signals
from fat stores. Am. J. Clin. Nutr.
1998; 68:599–606. [PubMed:
9734736]
7. Dulloo AG, Jacquet J, Girardier L.
Autoregulation of body composition
during weight recovery in human: the
Minnesota Experiment revisited. Int. J.
Obes. 1996; 20:393–405. [PubMed:
8696417]
8. Dulloo AG, Jacquet J, Montani J-P.
How dieting makes the lean fatter:
from a perspective of body
composition autoregulation through
adipostats and proteinstats awaiting
discovery. Obes. Rev. 2015; 16(Suppl
1):25–35. [PubMed: 25614201]
9. Saarni SE, Rissanen A, Sarna S et al.

Medical Research Archives, Vol. 4, Issue 8, December 2016
Metabolic Damage: Do Negative Metabolic Adaptations During Underfeeding
Persist After Refeeding in Non-Obese Populations?
Copyright 2016 KEI Journals. All Rights Reserved Page │19
Weight cycling of athletes and
subsequent weight gain in middleage.
Int. J. Obes. 2006; 30(11):1639–44.
[PubMed: 16568134]
10. Dulloo AG, Jacquet J, Seydoux J,
Montani J-P. The thrifty “catch-up fat”
phenotype: its impact on insulin
sensitivity during growth trajectories
to obesity and metabolic syndrome.
Int. J. Obes. 2006; 30:S23–S35.
[PubMed: 17133232]
11. Diedrichs PC, Lee C, Kelly M. Seeing
the beauty in everyday people: a
qualitative study of young Australians‟
opinions on body image, the mass
media and models. Body Image 2011;
8(3):259–66. [PubMed: 21561819]
12. Kenardy J, Brown WJ, Vogt E.
Dieting and Health in Young
Australian Women. Eur. Eat. Disord.
Rev. 2001; 4(9):242–254. Retrieved
from
http://onlinelibrary.wiley.com/doi/10.1
002/erv.388/epdf
13. Weyer C, Walford RL, Harper IT et al.
Energy metabolism after 2 y of energy
restriction: The Biosphere 2
experiment. Am. J. Clin. Nutr. 2000;
72:946–953. [PubMed: 11010936]
14. Bosy-Westphal A, Müller MJ,
Boschmann M et al. Grade of
adiposity affects the impact of fat
mass on resting energy expenditure in
women. Br. J. Nutr. 2009;
101(4):474–7. [PubMed: 19230079]
15. Doucet E, St-Pierre S, Almeras N et
al. Evidence for the existence of
adaptive thermogenesis during weight
loss. Br J Nutr 2001; 85(6):715–723.
[PubMed: 11430776]
16. Camps SGJA, Verhoef SPM,
Westerterp KR. Weight loss, weight
maintenance, and adaptive
thermogenesis. Am. J. Clin. Nutr.
2013; 97(5):990–4. [PubMed:
23535105]
17. Bosy-Westphal A, Braun W, Schautz
B, Müller MJ. Issues in characterizing
resting energy expenditure in obesity
and after weight loss. Front. Physiol.

Medical Research Archives, Vol. 4, Issue 8, December 2016
Metabolic Damage: Do Negative Metabolic Adaptations During Underfeeding
Persist After Refeeding in Non-Obese Populations?
Copyright 2016 KEI Journals. All Rights Reserved Page │20
2013; 4(March):47. [PubMed:
23532370]
18. Johnstone AM, Murison SD, Duncan
JS et al. Factors influencing variation
in basal metabolic rate include fat-free
mass, fat mass, age, and circulating
thyroxine but not sex, circulating
leptin, or triiodothyronine. Am. J.
Clin. Nutr. 2005; 82:941–948.
[PubMed: 16280423]
19. Hall KD. Computational model of in
vivo human energy metabolism during
semistarvation and refeeding. Am. J.
Physiol. Endocrinol. Metab. 2006;
291:E23–E37. [PubMed: 16449298]
20. Grande F, Anderson JT, Keys A.
Changes of Basal Metabolic Rate in
Man in Semistarvation und Refeeding.
J. Appl. Physiol. 1958; 12:230–238.
[PubMed: 13525268]
21. Müller MJ, Enderle J, Pourhassan M
et al. Metabolic adaptation to caloric
restrication and subsequent refeeding:
the Minnesota Starvation Experiment
revisited. Am. J. Clin. Nutr. 2015;
102:807–819. [PubMed: 26399868]
22. Ferro-Luzzi A, Petracchi C, Kuriyan
R, Kurpad A V. Basal metabolism of
weight-stable chronically
undernourished men and women: lack
of metabolic adaptation and ethnic
differences. Am. J. Clin. Nutr. 1997;
66(5):1086–93. [PubMed: 9356524]
23. Zipfel S, Mack I, Baur LA et al.
Impact of exercise on energy
metabolism in anorexia nervosa. J Eat
Disord 2013; 1(1):37. [PubMed:
24499685]
24. Platte P, Pirke KM, Trimborn P et al.
Resting metabolic rate and total
energy expenditure in acute and
weight recovered patients with
anorexia nervosa and in healthy young
women. Int. J. Eat. Disord. 1994;
16(1):45–52. [PubMed: 7920580]
25. Kosmiski L, Schmiege SJ, Mascolo M
et al. Chronic starvation secondary to
anorexia nervosa is associated with an
adaptive suppression of resting energy
expenditure. J. Clin. Endocrinol.

Medical Research Archives, Vol. 4, Issue 8, December 2016
Metabolic Damage: Do Negative Metabolic Adaptations During Underfeeding
Persist After Refeeding in Non-Obese Populations?
Copyright 2016 KEI Journals. All Rights Reserved Page │21
Metab. 2014; 99(3):908–914.
[PubMed: 24302748]
26. Haas V, Onur S, Paul T et al. Leptin
and body weight regulation in patients
with anorexia nervosa before and
during weight recovery. Am. J. Clin.
Nutr. 2005; 81(4):889–896. [PubMed:
15817868]
27. Russell J, Baur L, Beumont P et al.
Refeeding of anorexics : wasteful not
wilful. Lancet 1998; 352:1445–1446.
[PubMed: 9808001]
28. Wymelbeke V Van, Brondel L, Brun
JM, Rigaud D. Factors associated with
the increase in resting energy
expenditure during refeeding in
malnourished anorexia nervosa
patients. Am. J. Clin. Nutr. 2004;
80:1469–1477. [PubMed: 15585757]
29. Obarzanek E, Lesem MD, Jimerson
DC. Resting metabolic rate of
anorexia nervosa patients during
weight gain. Am.J.Clin.Nutr. 1994;
60(5):666–675. [PubMed: 7942571]
30. Dellava JE, Policastro P, Hoffman DJ.
Energy metabolism and body
composition in long-term recovery
from anorexia nervosa. Int. J. Eat.
Disord. 2009; 42(5):415–21.
[PubMed: 19107831]
31. Rossow LM, Fukuda DH, Fahs CA et
al. Natural bodybuilding competition
preparation and recovery: A 12-month
case study. Int. J. Sports Physiol.
Perform. 2013; 8(5):582–592.
[PubMed: 23412685]
32. Robinson SL, Lambeth-Mansell A,
Gillibrand G et al. A nutrition and
conditioning intervention for natural
bodybuilding contest preparation: case
study. J. Int. Soc. Sports Nutr. 2015;
12:1–11. [PubMed: 25949233]
33. Melby CL, Schmidt WD, Corrigan D.
Resting metabolic rate in weight-
cycling collegiate wrestlers compared
with physically active, noncycling
control subjects. Am. J. Clin. Nutr.
1990; 52(3):409–414. [PubMed:
2393002]
34. Dulloo AG, Jacquet J, Girardier L.

Medical Research Archives, Vol. 4, Issue 8, December 2016
Metabolic Damage: Do Negative Metabolic Adaptations During Underfeeding
Persist After Refeeding in Non-Obese Populations?
Copyright 2016 KEI Journals. All Rights Reserved Page │22
Poststarvation hyperphagia and body
fat overshooting in humans: A role for
feedback signals from lean and fat
tissues. Am. J. Clin. Nutr. 1997;
65(3):717–723. [PubMed: 9062520]
35. Müller MJ, Bosy-Westphal A.
Adaptive thermogenesis with weight
loss in humans. Obesity 2013;
21(2):218–228. [PubMed: 23404923]
36. Bosy-Westphal A, Kahlhöfer J,
Lagerpusch M et al. Deep body
composition phenotyping during
weight cycling: relevance to metabolic
efficiency and metabolic risk. Obes
Rev 2015; 16(February):36–44.
[PubMed: 25614202]
37. Shetty PS. Adaptation to low energy
intakes: the responses and limits to
low intakes in infants, children and
adults. Eur. J. Clin. Nutr. 1999;
53:S14–S33. [PubMed: 10365978]