MUG – Mauritius – April 2009
Weight fluctuations
a risk factor for obesity & metabolic syndrome
Abdul G. Dulloo
Department of Medicine / Physiology
University of Fribourg
Switzerland
Large
fluctuations
in body entities
weight of the
Obesity
predisposes
to disease
isinsulin
also an
independent
risk factor
for:
resistance
(metabolic)
syndrome
Obesity
Abdominal
obesity
Diabetes
type II
Insulin resistance
Hyperinsulinaemia
Cardiovascular
diseases
Arterial
hypertension
Evolution typique du poids de l’obèse
‘traité’
Contrainte
Bilan énergétique
Echappement
Rebond
Poids
spontané
Récupération
sensation de faim
métabolisme
Post-obèse
Maintien
Facteurs opposants
Metabolique et psychique
Adapted from Guy-Grand (1980)
Evaluation of energy expenditure
VO2 and VCO2 measurements – indirect calorimetry
Thermogenesis
Physical
activity
Basal metabolic rate
Postprandial thermogenesis
in response to a mixed meal (300 kcal)
(Dulloo & Miller Am J Clin Nutr 49: 44-50, 1989).
1.20
% increase
above Basal
Metabolic Rate
Never obese
(n = 8)
1.10
Post-obese
1.0
(n = 8)
0
30
60
90
120
Time after meal (mins)
150
Dépense énergétique (kcal/jour)
Hypométabolisme compensatoire
ENERGIE - ECONOMIE
Thermogenèse
2500
Obligatoire
Phys Act.
Comportement
Régulatoire
2000
Masse
Activité
Thermogenèse
1500
1000
BMR
500
0
- 20 Kg Poids
Dulloo AG: Nutrition 9: 366-372 (1993).
Is weight cycling detrimental to health?
A review of the literature in humans
Erik Muls et al. IJO, 19 (Suppl 3):S48-50, 1995
Points of controversy
WC makes subsequent weight loss more difficult
WC affects body composition (fat accumulation)
WC decreases resting energy expenditure
WC increases dietary preference for fat
WC promotes cardiovascular diseases
Montani, Viecelli, Prévot, Dulloo. Int J Obesity 30: S58-S66 (2006)
Weight fluctuations / weight cycling
Body
weight
obese
Lean
Growth
Age
Adult
Long-term Weight Cycling and Metabolic Syndrome
Cross-sectional study of 664 Japanese men
BW at age 20, 25, 30, 5y prior and current (40-49)
Simple linear regression model (slope, RMSE)
Classification by quartiles of weight fluctuation
Hypertension, HyperTG, Low HDL-chol, High glucose
Significant association when BMI < 25 kg/m2
but not when BMI > 25 kg/m2
Body
Weight
20
Zhang, Circ J, 2005
25
30
-5
40 - 49
Age (years)
Intentional Weight Cycling
in Young Lean Japanese Women
25
20
Change in SBP
Change in DBP
15
Significant increases in
• Systolic blood pressure
• Diastolic blood pressure
• Plasma Triglycerides
mmHg
10
5
0
-5
-10
-15
-20 52.1 kg
-25
0
53.3 kg
30
47.6 kg
Kajioka, Metabolism, 2001
44
Days
51.7 kg
74
48.7 kg
180
Large weight fluctuations in non-obese adults :
an independent risk factor for CV diseases & obesity
Body
weight
obese
lean
• weight cycling for a slim image
(yoyo dieting)
• Weight cycling in athletes
(power sports)
• Unintentional weight losses
(chronic diseases with remissions)
Growth
Age
Adult
Weight fluctuations in athletes & subsequent weight gain
in middle age
Saarni et al. Int J Obesity (2006) 30: 16391639-44
(Boxers, wrestlers, weight-lifters)
BMI
Age
Weight fluctuations during growth ?
Body
weight
obese
lean
• Weight loss program in adolescents
• Dieting for a slim image
• Slimming in athletes (competition)
• Common fluctuations (teething, fever, ...)
• Chronic diseases with remissions
• Food shortage (poverty, war, famine, ...)
Growth
Age
Adult
Weight fluctuations during early growth?
The strongest predictor of future
cardiovascular disease
(and its risk factors)
a combination of
thinness at birth and/or early infancy
followed by
catch-up growth during infancy/childhood
Early growth and later coronary heart disease in later life
Growth of 357 boys who later developed CHD in a cohort of 4630 boys born in Helsinki
Eriksson …Barker BMJ 322:949-953 (2001)
Barker… Eriksson NEJM 353:1802-1809 (2005)
Cohort
Standard
deviation
(Z-score)
BMI
Height
Weight
Age (years)
Higher risks for
coronary events
was associated with
insulin resistance
Coronary heart disease death rates
in men born in Helsinki during 1924-1933
Data from Forsen et. al. BMJ 319: 1403-1407 (1997)
150
Standardized
125
mortality ratio
100
P < 0.0001 for trend by log-linear
regression
75
50
25
Thin
babies
0
< 25
-27
-29
> 29
Ponderal index at birth (Kg / m3)
(a measure of thinness)
Risk for type 2 diabetes
in adult women born between 1921-1946
The Nurses’ Health study, USA
Rich-Edwards et al.
Ann Intern Med (1999) 130: 278-284
2.5
2
Relative Risk
(P < 0.001)
adjusted for age, 1.5
BMI, maternal
history of diabetes
1
Referent
0.5
0
<2.3 2.3-2.5
2.5-3.2
3.2-3.8
Birthweight
(Kg)
3.9-4.5
>4.5
Insulin-mediated glucose disposal in humans
born SGA ( ) and with normal birth weight ( )
PGU. plasma glucose uptake
NOXGDR: non-oxidative
glucose disposal rate
GOXR: glucose oxidation rate
Jaquet, D. et al. J Clin Endocrinol Metab 2001;86:3266-3271
From weight fluctuations
to obesity & chronic metabolic diseases
obese
Body
weight
weight cycling
BMI <30
Higher CV risks,
Higher obesity risks
Unintentional
weight loss/recovery
Catch-up growth
in infants & children
Increased risks for later
CV diseases, diabetes, obesity
after fetal/neonatal growth retardation
Growth
For review:
Age
Adult
Dulloo et al.: Int J Obesity 30 (suppl. 4): S23-S35 (2006)
Strongest predictor
Later
Catch-up growth
Thin-short
metabolic diseases
babies/infants
State of
hyperinsulinemia
catch-up of lean tissue
or catch-up of fat tissue
Preferential catch-up fat
after low birth weight / poor neonatal growth
Children born small for gestational
age (SGA) have more fat, and
less FFM (i.e less lean tissue)
Sas et al. J Clin Endocrinol 85: 3786-3792 (2000)
Martins et al. Br J Nutr 92: 819-825 (2004)
Jornayvaz et al. Metabolism 53: 847-851 (2004)
At any adult BMI,
those who were smaller at birth
have less FFM and more fat
(∆ 4 kg)
Eriksson et al. Horm Metab Res 34:72-76 (2002)
Low birth weight in Mauritius
Health Statistics Report 2007
20
18
16
14
12
10
8
6
4
2
0
Low birth weight
(as a % of live birth with known birth weight)
∆
∆
∆
∆
■
∆
■
■
■ ∆ ∆■ ■ ■ ∆■ ∆■ ■ ■
∆
■
∆
∆
■
■
∆
∆
■
8%
1
9
9
0
■
■
■ 16%
∆
∆
∆
Still-birth rate
(per thousand total live births and still births)
1
9
9
2
1
9
9
4
1
9
9
6
1
9
9
8
2
0
0
0
2
0
0
2
2
0
0
4
2
0
0
6
2
0
0
8
The 1st longitudinal study of body composition (by DEXA scanning)
in human infants born small for gestational age (SGA)
Fat mass
(kg)
4
● SGA (Small for gestational age, n=29)
3
○
●
2
Lean mass
(kg)
○ AGA (Appropriate for gestational age, n= 22))
○
●
□ AGA
■ SGA
12
11
□
■
10
□
■
9
Fat mass/lean mass
ratio
■ SGA
0.30
■
□
0.25
□
□
■
AGA
Figure built from a reanalysis of raw data derived from
Ibinez et al. JCEM (2006) and published in:
0.20
2
Infants born small show preferential
catch-up fat & lower insulin sensitivity
(HOMA index)
3
Age (yrs)
4
Dulloo et al. Int J Obesity 30 (suppl. 4): S23-S35 (2006)
Past reports of ‘‘Rapid’’ fat tissue recovery (catch-up fat)
with lean tissue recovery ‘‘lagging behind’’
Kornfeld & Schuller (1931)
Debray et al. (1946)
Keys et al. (1950)
• Emaciated patients in Vienna
• Prisoners from concentration camps
• Men after experimental starvation
Ashworth (1969)
McLean & Graham (1980)
Castilla-Serna et al. (1996)
• Infants / children recovering from
protein-energy malnutrition
Barac-Nieto et al. (1979)
Forbes et al. (1984)
Mitchell & Truswell (1987)
• Adults after substantial weight loss
(independently of protein level)
• Anorectics regaining weight
Van Eys (1985)
Streat et al. (1987)
Kotler et al. (1990)
• Cancer patients
• Septic intensive care patients
• AIDS patients - parenteral nutrition
Source of references: Dulloo et al. Int J Obesity (2002) 26: S46-S57
Preferential catch-up fat:
ubiquitous phenomenon of weight recovery throughout lifecycle
obese
Body
weight
Relapse of obesity
Catch-up fat
lean
Weight recovery in adults
Catch-up fat
Catch-up growth
Catch-up fat
Growth
Age
Adult
Fundamental questions ?
What are the control systems that regulate
during 'physiological' catch-up fat
fat storage
?
via increased food intake
(compensatory hyperphagia)
via nutrient partitioning
Shift from lean to fat tissue
via increased metabolic efficiency
thrifty metabolism - Suppressed thermogenesis
How does 'physiological' catch-up fat turn
pathophysiological ?
Working hypothesis
Dulloo, Seydoux, Girardier (1990)
Adipose-muscle crosslinks
underlie the thrifty metabolism that drives catch-up fat
Signal ?
Skeletal muscle
Adipose  Fat
tissue
Thermogenesis
Hyperinsulinemia
Catch-up fat
Spared Glucose
Insulin
resistance
Decreased Glucose
utilisation
‘Thrifty metabolism’ driving catch-up fat
in humans ??
Body
weight
Modern life
Evidence from experimental
starvation and refeeding
The Classic
Minnesota Experiment
(Keys et al 1950)
Growth
Age
Adult
Design of the Minnesota Experiment
(Keys et al. 1950)
n = 32 (normal weight healthy men)
Body
weight
Weeks
Weight loss:
25-29%
No hyperphagia Hyperphagia
Low fat intake High fat intake
20% by energy 35% by energy
Control Semistarvation
Refeeding Refeeding
period
restricted
ad libitum
Food intake
++++++++++++++++++++++++++++++++++++++++++++++++
BMR
+
+
+
+
+
+
+
Body composition
+
+
+
+
+
The men volunteers in the 'Minnesota Experiment':
after nearly 6 months of experimental semistarvation
Keys et al. 1950 : The Biology of Human Starvation
Minnesota Experiment (Keys et al. 1950)
Low dietary fat intake
~ 20% by energy
Recovery
of Fat and FFM
(% loss)
Fat mass
‘Preferential catch-up fat
& fat overshooting’
Fat free mass
Recovery of weight (% loss)
After 12 weeks of restricted refeeding (n=32)
‘Minnesota Experiment’ Revisited
Adapted from Dulloo, Girardier & Jacquet (1996, 1998)
 + 20%
+ 10%

Body weight
(kg)
-10% 
-10%
- 20%
 - 25%

20

10
Suppressed Thermogenesis
0
( adjusted BMR
-10
% initial BMR)
BMR
20
Energy intake
(kcal/d x 100 )
Semistarvation Refeeding
Minnesota Experiment Revisited
Dulloo et al. Am J Clin Nutr 1998;68:599-606
Feedback control system between fat stores & thermogenesis
Concept of dual-adaptive thermogenesis
Dulloo et al. Int J Obesity 25 (Suppl. 5):S22-S29. (2001)
Refeeding
Starvation
Energy intake
Non-specific
Thermogenesis
Thermogenesis
Sympatheticthyroid axis
(Down)
(Down)
adipose-specific
Thermogenesis
100%
50%
Suppressed thermogenesis
specific for 'catch-up fat'
0%
Body fat
(% initial value)
50%
100%
Catch-up fat resulting only from suppressed thermogenesis:
Adapted from Dulloo, Seydoux &
a rat model
Girardier (1990, 1991)
Body protein
(g)
60
40
45
Body fat
(g)
**
25
5
Energy intake
(kJ wk-1 )
2600
Energy
expenditure
(kJ wk-1)
2000
**
Refed
Control
catch-up fat
***
2200
1200
**
**
Semi
starvation
0
Refeeding
7
Suppressed thermogenesis
specific for 'catch-up fat'
14
Days of refeeding
21
Suppressed thermogenesis favouring catch-up fat:
a state of hyperinsulinemia / Insulin resistance
Body fat
(g)
Plasma glucose
45
(mg/100ml)
GTT
**
25
400
**
Control
WM
RefedRF
300
200
5
***
Clamp
Refed
Control
Tissue glucose utilization index
100
0
0
20
40
60
80 100 120
Energy
intake
(kJ wk-1 )
mins after glucose load
2600
Plasma Insulin
15
2200
(ng/ml)
**
Energy
expenditure
(kJ wk-1)
10
*
2000
1200
**
Semistarvation
**
*
**
Refed
*
5
Control
Refeeding
0
14
7
Days of refeeding
**
0
21
0
20
40
60
80 100 120
mins after glucose load
Crescenzo, ……, Dulloo .: Diabetes 52: 1090-1097. (2003)
Suppressed thermogenesis favouring catch-up fat:
Glucose redistribution from skeletal muscle to adipose tissue
Skeletal muscle
Enhanced de novo lipogenesis
in adipose tissue
25
20
100
15
10
5
FAS activity
U/g prot/min
Glucose utilization index
(ng/min/mg tissue)
30
Adipose tissue
Control
Refed
**
**
80
Glucose utilization index
(ng/min/mg tissue)
*
**
60
***
***
**
40
***
20
0
0
b
Ti
lis
ia
nt
A
Q
io
er
dr
a
u
r
ps
e
ic
w
te
i
h
Q
s
ep
ic
d
re
i
us
w
te
i
h
m
ne
em
c
n
ro
oc
t
r
s
a
st
a
G
G
dr
a
u
iu
s
d
re
us
le
o
S
al
al
al
n
e
m
i
on
gu
dy
t
i
n
i
I
id
er
p
Ep
ro
t
Re
Cettour-Rose, ….., Dulloo. Diabetes 54: 751-756. (2005)
Suppressed
thermogenesis
Adipose tissue
insulin hyperresponsiveness
Skeletal muscle
insulin resistance
Hyperinsulinemia
Catch-up fat
Excess adiposity
Skeletal muscle insulin resistance and hyperinsulinemia
precede the development of excess adiposity
Suppressed thermogenesis favouring catch-up fat
(on a low-fat diet): Glucose redistribution
Signal ?
Skeletal muscle
Adipose  Fat
tissue
Thermogenesis
Hyperinsulinemia
Insulin
Hyperresponsiveness
Increased Glucose
utilisation for lipogenesis
catch-up fat
Euglycemia
Insulin
resistance
Decreased Glucose
utilisation
Will an increase in dietary fat
compromise this homeostatic system ?
Role of suppressed thermogenesis in susceptibility to
insulin resistance & glucose intolerance
Adaptive
Catch-up fat
on low-fat diet
maladaptive
Catch-up fat
on high-fat diet
Thermogenesis
Insulin
Glucose tolerance
Muscle PI3K
Muscle AMPK
Summarized from: Summermatter, ….., Dulloo. FASEB J 22: 774-785 (2008)
Putative molecular-physiological mechanisms
underlying adipose-muscle crosstalks in thrifty catch-up fat phenotype
Brain
Insulin
leptin
SNS/thyroid
catecholamines
Insulin
+
+
AMPK
PI3K
adiponectin
?
l
a
n
Sig
Pancreas
Adipose  Fat
tissue
-
-
-
Substrate cycling
Skeletal muscle
Thermogenesis
glucose spared
catch-up fat
Dulloo et al.: Int J Obesity 30 (suppl. 4): S23-S35 (2006)
Take home message
A role for a ‘thrifty catch-up fat phenotype’
in pathways to insulin-resistance syndrome
Body
weight
weight cycling
Higher CV risks,
Higher obesity risks
Unintentional
weight loss/recovery
Increased risks for later
Catch-up growth
CV diseases, diabetes, obesity
in infants & children
after fetal/neonatal growth retardation
Growth
Age
Adult
For review: Dulloo et al.: Int J Obesity 30 (suppl. 4): S23-S35 (2006)
Take home message I
Genes/reprogramming/lifestyle
Interactions
Thin-short babies/infants
Adults (normal/overweight)
Catch-up growth
Weight recovery
Later insulin-related
chronic diseases
Obesity,
Obesity, diabetes,
diabetes, CV diseases
state of
catch-up fat
Regulation of fat storage via suppressed thermogenesis
(energy conservation for the purpose of catch-up fat)
constitutes a thrifty ‘catch-up fat’ phenotype
that predisposes individuals with large weight fluctuations
to insulin resistance (metabolic) syndrome
Pathophysiology of weight fluctuations across the life cycle:
the role of preferential catch-up fat
Body
weight
How body fat is acquired is at least as important
as excess fat per se in the pathogenesis of
insulin-related chronic diseases
Growth
Age
Adult
Acknowledgements
University of Fribourg
Jean-Pierre Montani
Giovanni Solinas
Davide Mainieri
Serge Summermatter
University of Geneva
Josiane Seydoux
Lucien Girardier
Jean Jacquet
Françoise Rohner-Jeanrenaud
Philippe Cettour-rose
Françoise Assimacopoulos-Jeannet
University of Frederico II
Naples
Raffaella Crescenzo
Susanna Iossa
Hippocrates 400 BC
‘‘ Do not allow the body to attain extreme thinness,
for that, too, is treacherous,
but bring it only to a condition
that will naturally continue unchanged,
whatever that may be ’’