Seminars in Pediatric Surgery (2009) 18, 126-135

Changes in physiology with increasing fat mass

Dara P. Schuster, MD, FACE

From the Divisions of Endocrinology and Metabolism, Departments of Pediatrics and Internal Medicine, The Ohio State
University Hospitals and Nationwide Children’s Hospital, Columbus, Ohio.

KEYWORDS Obesity has reached epidemic proportions in the USA with a nearly fourfold rise in the prevalence of
Obesity; childhood obesity. There are many possible etiologies of obesity as the adipose tissue plays a
Adipose tissue; significant, complex role in the physiology of fuel metabolism and hormone regulation. The develop-
Hormones ment of obesity represents a pathophysiologic increase in fat mass in which multiple metabolic
pathways are deranged. The consequences of these metabolic derangements, including insulin resis-
tance and inflammation, are reflected in obesity-related comorbidities and can be seen in the setting of
pediatric obesity. Obese adolescents demonstrate increased rates of early maturation, orthopedic growth
abnormalities, diabetes mellitus, obstructive sleep apnea, hypertension, steatosis, and polycystic ovarian
syndrome, placing this group of children at risk for long-term health problems and reduced quality of
life. Given the negative short- and long-term impact of obesity on children, careful attention should be
paid to the unique health issues of this “at-risk” population with both prevention and early intervention
strategies.
© 2009 Elsevier Inc. All rights reserved.

Obesity has reached epidemic proportions in the USA, begins is a matter of considerable debate, but with the rising
accounting for approximately 300,000 deaths annually.1 rates of obesity in younger individuals, the question of
These deaths are attributed mainly to the comorbidities whether there is a relationship between years of obesity and
associated with obesity, including hypertension (HTN), car- severity of associated medical conditions, because of the
diovascular disease (CVD), and diabetes mellitus (DM). cumulative effect of chronic obesity (ie, “pound years”),
Recent data have demonstrated a nearly fourfold rise in the will need to be answered. Studies have demonstrated higher
prevalence of childhood obesity, making the pediatric age rates of morbidity and mortality from CVD, stroke, and
group the fastest growing subpopulation of obese individu- colorectal cancer in the adults with a history of adoles-
als in this country.2 In addition, corresponding data exam- cent-onset obesity versus those who were not obese as
ining global obesity trends demonstrate similar patterns adolescents.4 Understanding the role of the adipose tissue
worldwide with rates of overweight— defined as having a in hormone secretion, appetite stimulation, and the patho-
body mass index (BMI) greater that 25 kg/m2— occurring in physiology of obesity will be the key to developing
40% of men and 30% of women and rates of obesity (BMI treatment and prevention strategies to reduce the rising
ⱖ 30 kg/m2) at approximately 25%.3 This increase in obe- incidence of obesity.
sity rates is a prelude to multiple obesity-associated health
problems. The age at which obesity-related comorbidities
Physiological role of adipose tissue in the
Address reprint requests and correspondence: Dara P. Schuster, body
MD, FACE, 491 McCampbell Hall, 1581 Dodd Drive, Columbus, OH
43210. Adipose tissue serves as the major storage site for fuel, in
E-mail address: dara.schuster@osumc.edu. the form of triglycerides, and is primarily made up of

1055-8586/$ -see front matter © 2009 Elsevier Inc. All rights reserved.
doi:10.1053/j.sempedsurg.2009.04.002
Schuster Physiology and Obesity 127

lipid-filled adipocytes held together by a framework of The adipose tissue is an active immunomodulator and en-
collagen fibers. Other cells found within the fat include docrine organ secreting leptin, adiponectin, adipsin, com-
stromal–vascular cells, leukocytes, macrophages, and pre- plement factor 3, free fatty acids, interleukin-6 (IL-6),
adipocytes.6-9 Mammalian adipose tissue can be divided TNF-␣, angiotensinogen, and plasminogen activation inhib-
into brown adipose tissue and white adipose tissue, with itor-1. Disruption and/or dysregulation of these hormonal
most fat in humans occurring as white adipose tissue. pathways has been shown to result in deleterious effects at
Brown adipose tissue is important for regulating heat and both the local and systemic levels via direct adipokine-
nonshivering thermogenesis and is seen predominately in related cytotoxicity and indirect upregulation of systemic
young and/or hibernating mammals. White adipose tissue, inflammatory cytokine levels (ie, secreted from liver, lung,
characterized by limited vascularity with blood flow varying etc.), respectively. The consequences of such pathophysio-
based on body weight and nutritional status, serves several logic events are believed to result in the development and
roles, including heat insulation, mechanical cushioning, and progression of multiple obesity-related comorbidities, in-
fuel storage.5 Adipose tissue also acts as a buffer for energy cluding CVD, obstructive sleep apnea, stroke, hypertension,
imbalances, especially for energy excess and storage as and lipid disorders to name a few.11
demonstrated by increased blood flow during states of nu-
tritional fasting.
Recent data have demonstrated that adipose tissue mass
is a function of both the absolute numbers of adipocytes and
Pathophysiologic causes of increasing fat
relative individual adipocyte size. Although developmental mass and obesity
mechanisms related to adipocyte maturation are poorly un-
Whereas “positive balance” refers to a normal state of
derstood, it appears that adipocyte number is controlled by
energy storage with the subcutaneous and visceral fat de-
a relative balance between mechanisms resulting in apopto-
posits, obesity refers to the development of excess body fat
sis and/or necrosis. This balance is believed to result from
resulting from an imbalance between energy intake and
differential signaling and subsequent development of adi-
energy expenditure. Though widely referred to as a simple
pocyte progenitor cells (ie, pre-adipocytes).7 An increase in
relationship between energy intake and expenditure, this
adipocyte mass can occur with differentiation of pre-
relationship involves interaction in multiple variables, such
adipocytes as well as with increased lipid deposition
within the adipocyte.8,9 In humans, there are two periods as growth hormone secretion, reproductive hormone secre-
of hyperplastic growth of adipose tissue, including the tion, age and genetics, as well as psychological issues that
third trimester of pregnancy and early in puberty. In may predict eating habits.
addition, hyperplastic growth can occur in adulthood. Once obese, reduction of excess body weight requires
Although poorly defined, it is believed that such growth great diligence due to the body’s feedback control system.
is related to a feedback mechanism whereby increasing Maintenance of reduced body weight is associated with
adipocyte size results in stimulation and differentiation of declines in energy expenditure that are greater than can be
precursor cells, culminating in an overall increase in accounted for by decline in metabolic mass. In the setting of
adipocyte number. Once new adipocytes are formed, weight reduction, there are decreased circulating triiodothy-
which can occur in the setting of long-term overfeeding, ronine and thyroxine concentrations, increased parasympa-
the resulting rise in adipocyte number is permanent. The thetic tone, and decreased sympathetic tone, and these
mechanism(s) responsible for related signal pathways are changes last for months to years. The importance of the
not well defined and may be, in part, genetically deter- feedback system was examined in a study of 18 obese and
mined. Interestingly, it appears that a differential rate of 23 never-obese subjects who were monitored during caloric
associated apoptosis, necrosis, and progenitor cell stim- restriction and caloric excess.12 Weight loss of 10-20% was
ulation exists among varying fat deposits (ie, peripheral associated with a decrease in total and resting energy ex-
and central depots). In addition, cellular response to penditure, a change that retarded further weight loss. Addi-
various stimuli, such as the effect of tumor necrosis tionally, with weight loss there is an increase in appetite and
factor-␣ (TNF-␣) on apoptosis and peroxisome prolifera- metabolic changes that promote weight gain. Similarly,
tor-activated receptor-gamma on proliferation, may vary weight gain is associated with an increase in energy expen-
depending on the nature of the specific fat tissue.10 diture, which retarded further weight gain. One conse-
In addition to serving as an energy storage reservoir, quence of these changes in energy expenditure is that a
adipose tissue plays an active role in various homeostatic formerly obese subject would require 15% fewer calories to
processes, including energy expenditure, appetite regula- maintain a “normal” body weight than someone with the
tion, and glucose regulation. In addition, fat tissue is critical same body composition who was never obese.12,13 The
for thyroid function, immune response, bone health main- reductions in energy expenditure persist, even in adults who
tenance, reproduction, and blood clotting through the secre- have maintained a reduced body weight for 3 to 5 years,13
tion of various hormones. These processes are regulated by indicating the necessity for lifelong commitment to weight
a variety of complex hormonal pathways and interactions. maintenance.
128 Seminars in Pediatric Surgery, Vol 18, No 3, August 2009

Genetic obesity ents.28 In a recent article, Wardle and coworkers29 exam-

ined 5000 pairs of twins age 8-11, revealing that genes were
Obesity has many potential etiologies,14 each of which responsible for 77% of the difference in BMI and waist
appears to have a genetic component.15 Obesity caused by circumference, and environmental influences accounted for
single-gene mutations14 has been well documented in mice 23% of the differences.
and other rodents but is relatively rare and not well defined Genetic influence of obesity is related to two primary
in humans. These include the agouti, leptin, and leptin processes: (1) susceptibility to overeating despite normal
receptor gene mutations. The agouti gene causes obesity by energy requirements, and (2) presence of a normal drive to
competing with melanocyte-stimulating hormone (MSH) eat despite low energy requirements. Appetite and the drive
for a melanocortin-4 receptor in the hypothalamus that mod- to eat may be impacted by several genetically programmed
ulates food intake.16 In humans, serum concentrations of the metabolic pathways,30 and this is demonstrated in the spe-
agouti protein are higher in obese than nonobese men, and cific but rare syndromes of Prader–Willi and leptin defi-
the values correlate with other measures of obesity, such as ciency.
BMI, indicating a possible role of this hypothalamic path- Obesity is a feature of at least 24 genetic disorders.15 The
way in obesity development.17 The leptin gene codes for a Bardet–Biedl and Prader–Willi syndromes are the best-
protein called leptin.18 Leptin is produced in fat cells, gut, known examples of these traits. The Prader–Willi syndrome
and the placenta and signals the brain about the quantity of is a neurodegenerative disorder that is characterized by poor
stored fat.19 In the setting of leptin-deficiency, mice (ob feeding and low tone as an infant and insatiable appetite,
mice) have hyperphagia, insulin resistance, hyperinsulin- food-seeking behavior, and obesity as an adult. The Bardet–
emia, and infertility. Administration of leptin to these mice Biedl syndrome is an autosomal recessive disorder charac-
reverses all the features of this syndrome. It is believed that, terized by obesity.31 Common obesity also appears to be a
with increasing adiposity, leptin acts as a negative feedback heritable trait, but the genes that contribute to the more
“adipostatic” signal to brain centers to reduce energy in- common forms of obesity have been difficult to identify.
take.20 With respect to food intake, leptin may act in the However, whole-genome scans in large populations have
arcuate nucleus to reduce the production of neuropeptide identified polymorphisms in genes that may be associated
Y,21,22 one of the most potent stimulators of food in- with obesity risk.
take.21,23 Obesity due to leptin deficiency has been reported
in two consanguineous families.24 Treatment of these indi-
viduals with physiological doses of leptin resulted in a
marked decrease in food intake and substantial loss of Determinants of energy expenditure
weight and fat mass.24 Leptin receptor gene defects appear
phenotypically similar to the leptin-deficient expression, Genes may help create the energy imbalance that leads to
and human obesity resulting from leptin receptor deficiency obesity. In addition to the heritability of weight, metabolic
has been described.24,25 Despite the rarity of these single rate, thermic response to food, and spontaneous physical
gene defects, recognition of the defects aid in focusing activity are to some extent heritable.15,32 These relation-
research to specific metabolic pathways that may play a role ships were illustrated in a study of overfeeding in identical
in human obesity. Transgenic mice are often used to study twins.33 With respect to body weight, percentage of fat, fat
gene products thought to play a role in obesity. In this mass, and estimated subcutaneous fat, there was approxi-
manner, candidate genes are isolated in the form of knock- mately three times more variance among pairs than within
out mice or developed to overexpress certain gene products, pairs, indicating that both current weight status and the
and the impact on weight is studied. For example, the MSH metabolic processes underlying weight gain have a strong
pathway has been studied using melanocortin-4 receptor inherited component.
deletion. Mice lacking this receptor become hyperphagic Low energy expenditure can promote weight gain.34
and markedly obese. These observations suggest that func- With regard to the issue of low energy requirements, evi-
tional receptors for MSH inhibit both food intake and the dence suggests that differences in resting energy expendi-
accumulation of body fat.26 In a similar manner, other ture and fuel storage have only a small impact on long-term
hypothalamic receptors are being studied, including seroto- weight gain.35 Approximately 70% of energy expenditure is
nin, neuropeptide Y, and gastric inhibitory polypeptide. used for basal or resting metabolic processes, and there is a
For most, the genetic susceptibility to obesity is more com- strong relationship between total daily energy expenditure or
plicated and involves interaction of both genetics and environ- resting energy expenditure and lean body mass. Differences in
ment. Studies of twins, adoptees, and families all suggest the lean body mass account for approximately 80% of the variance
existence of genetic factors in human obesity.27,28 The herita- in energy expenditure among subjects.34 Another 10% of en-
bility of obesity estimated from twin studies is high, ranging ergy expenditure is dissipated through thermic responses to
between 0.6 and 0.9, with only slightly lower values in food, and insulin resistance is associated with a low thermic
twins raised apart compared with those raised together.28 effect but does not cause it. Activity thermogenesis, which
Similarly, in adoptees, the BMI correlates with that of their includes both exercise and nonexercise activity thermogenesis,
biological parents rather than that of their adoptive par- is also a component of energy expenditure. A sedentary life-
Schuster Physiology and Obesity 129

style— one that promotes little thermogenesis—is an important Leptin, also produced by the adipocytes, circulates at levels
factor in the development of obesity. Finally, the propensity to proportionate to levels of body fat. It binds to receptors in the
physical inactivity appears to be determined in part by genet- hypothalamus, activating signals to inhibit food intake and
ics.36 Individuals may be prone to less physical activity due to increase energy expenditure. In addition, leptin affects other
physical findings, such as low rates of energy expenditure34 or hormones, both anorectic and orexigenic. As discussed previ-
less aerobic type 2 muscle fibers that reduce oxidative capacity ously, congenital leptin deficiency is rare, and it appears that
and make exercise more difficult.27 For such individuals, re- the typical state of obesity in humans is more likely to be
duced environmental exercise demands could predispose to related to leptin resistance rather than deficiency.22,42
weight gain.
The central nervous system provides the feedback con- Hormone signaling from the gastrointestinal tract
trol system for integration of energy expenditure and for
digestion, absorption, transport, and storage of nutrients and Previous research has implicated gut hormones or incretins
mobilization and use of fuels. Signals regarding alterations in appetite regulation, gastrointestinal motility, satiety, and
in fuel usage are tightly regulated and come primarily from changes in glucose, and in this regard may possibly play a
adipocytes and from the gastrointestinal tract. role in the occurrence of obesity.43 Gut hormones modulate
food processing, but the precise mechanism of hormone
regulation, communication, and interaction and their ability
to stimulate or modulate weight gain is unknown. In the
Hormone signaling from the adipocytes foregut, ghrelin, a 28-amino-acid orexigenic hormone pro-
duced in the stomach and duodenum, increases before meals
Adipose tissue is now known to play a significant role in and decreases after meals.44 Ghrelin is elevated in the set-
metabolic and immune function, possibly by the production of ting of obesity and increased after diet-induced weight loss,
pro-inflammatory cytokines and other hormones from adipo- suggesting a role in the compensatory changes in appetite
cytes.37,38 In the past decade, pro-inflammatory adipocyto- and energy expenditure that make maintenance of diet-
kines, including TNF-␣, IL-6, and adiponectin, leptin, resistin, induced weight loss difficult.45 A similar increase in serum
and many others, have received tremendous attention for their ghrelin occurs after exercise-induced weight loss (with no
potential role in fuel metabolism, glucose homeostasis, insulin change in caloric intake).46 In contrast to such observations,
resistance, and perhaps atherosclerosis.37,38 a recent study investigating changes in serum ghrelin con-
TNF-␣ has been demonstrated to reduce insulin sensi- centrations following surgical weight loss (ie, gastric by-
tivity and promote the development of diabetes mellitus, pass) found a decrease in detectable ghrelin postoperatively.
and has been associated with the development of athero- These results raise the possibility that a reduction in circu-
sclerosis.39-41 It has been suggested that obesity itself rep- lating ghrelin may be of fundamental importance in the
resents a chronic inflammatory state, whereby increased establishment of prolonged weight loss that often occurs
levels of IL-1, IL-6, and TNF-␣ stimulate liver production with gastric bypass.45 In the hind-gut, where the suppression
of C-reactive protein (CRP), which in turn stimulates addi- of gastrointestinal motility is modulated by hormones, such
tional proinflammatory cytokine release. The increased as peptide YY (PYY), neurotensin, glucagon-like peptide
proinflammatory cytokines are felt to be related to acceler- ⫺1 (GLP-1), and oxyntomodulin, perturbations in hormone
ated atherogenesis and an increase in the comorbid condi- release have been demonstrated with weight loss secondary
tions seen in obesity as discussed previously. These hor- to Roux-en-Y gastric bypass. Peptide YY is a hormone
mones have been demonstrated to play an active role in released after meals whose function is to reduce appetite.47
insulin resistance, inflammation, and thrombosis, with vis- In the setting of morbid obesity, PYY levels are reduced
ceral fat being the greatest contributor to this process. with a blunted postmeal response.48 Korner and cowork-
In contrast to the proinflammatory hormones, adiponectin, a ers49 found that postprandial levels of PYY are markedly
244-amino-acid peptide solely produced and secreted by adi- elevated after bariatric surgery. In addition, both bariatric
pose tissue, has been reported to improve insulin sensitivity and nonbariatric studies involving bypass of the foregut
and indeed could prevent diabetes mellitus and atherosclero- have shown increased levels of PYY, GLP-1, neurotensin,
sis.41 Unlike other adipose tissue-derived peptides that are and enteroglucagon in the fasting and postprandial
elevated in obesity, adiponectin levels are decreased in obese states.50-57 These changes in gut hormones appear to be
individuals with insulin resistance. Higher levels of adiponec- sustained for 15-20 years after the procedure. GLP-1 has
tin are associated with lower risk of diabetes in older men and been found to be reduced in patients with obesity and
women.40,41 Thus, the lower adiponectin levels in obese indi- DM,55,56 although there are inconsistent results regarding
viduals with larger adiposity are paradoxical, suggesting that, reduced levels and normalization after weight loss. With
in obese individuals, adiponectin gene may be downregulated. regard to other potential gut hormones, they appear to have
It is also well known that interventions that reduce percent various affects on appetite, satiety, and gastrointestinal mo-
adiposity mass, weight loss, and aerobic exercise indepen- tility. Kellum and Pappas and coworkers50,57 found no
dently increase adiponectin levels with a concomitant improve- change in vasoactive intestinal peptide, serotonin, or chole-
ment in insulin sensitivity. cystokinin after weight loss associated with gastric bypass.
130 Seminars in Pediatric Surgery, Vol 18, No 3, August 2009

Changes in gut hormone responses directly and indi- Consequences of increasing fat mass in
rectly affect fuel metabolism and play a significant role in children
regulation of insulin and glucagon secretion.49,58,59 In this
regard, changes in insulin sensitivity and resolution or im-
provement in hyperglycemia after weight loss are com-
Insulin resistance
monly seen.58,59 Hormones that appear to be particularly
Obesity has a negative impact on multiple associated alter-
important in this process include ghrelin, GLP-1, and PYY.
ations in the glucose/insulin axis and on lipid metabolism.
Reductions in ghrelin after Roux-en-Y gastric bypass may
In particular, obese adults demonstrate reduced glucose
have an antidiabetic effect by relatively decreasing levels of
disposal, primarily at the level of skeletal muscle (peripheral
growth hormone, cortisol, and epinephrine60 and improving
insulin resistance)64 as well as impairment in insulin action
peripheral glucose usage.61 The increase in GLP-1 levels often on nonesterified fatty acid oxidation,65 leading to insulin
seen after weight loss has a positive effect on insulin secretion, resistance and abnormal lipolysis. Yet it is unclear when
beta cell function, and possibly insulin sensitivity.62 these abnormalities occur in the setting of acute versus
Obestatin, a newly discovered hormone recently isolated chronic obesity. Polonsky and coworkers66 found that insulin
from rat stomach, is also encoded by the ghrelin gene, and secretion rates were significantly higher in the obese group
has been shown to oppose the effects of ghrelin on food when compared with the normal weight group. Furthermore,
intake, body weight, and gastric emptying. Despite these there was no difference in insulin clearance or hepatic insulin
initial observations, delineation of mechanistic features re- extraction between the groups (Figure 1). There was a dimin-
garding precise actions of obestatin, along with those of ished hepatic insulin extraction noted in a subset of the obese
other newly identified incretins, remains the topic of ongo- group that demonstrated a greater degree of hyperinsulin-
ing research and debate.63 Better understanding of the in- emia. To characterize this further, Monti and coworkers67
teractions of the gut hormones on appetite and fuel metab- examined obese children with normo-insulinemia to char-
olism will hopefully provide important information on the acterize early metabolic derangement and found peripheral
development of future obesity management strategies. insulin resistance in the obese children when compared with

Figure 1 Concentrations of glucose and insulin and c-peptide and insulin secretion in normal weight and obese subjects during
hyperglycemic clamp. (Reprinted with permission.66)
Schuster Physiology and Obesity 131

normal weight children but no significant differences in sulin resistance and HTN, interest in defining the mecha-
nonesterified fatty acids response to insulin infusion. Le nistic relationship between impaired glucose–insulin metab-
Stunff and coworkers68 found that the earliest abnormality olism and obesity has been recently highlighted by several
in glucose metabolism in obese children (those with short studies examining the prevalence of these comorbid condi-
duration of obesity) was an abnormal insulin response to a tions and morbid obesity.73-75 The link between these co-
meal stimuli and that maximal glucose uptake decreased morbid conditions becomes even more evident when con-
with age and obesity duration. So, it appears that one of the sidering their respective roles in the pathogenesis of
earliest negative effects of obesity is the development of “metabolic syndrome,” a condition generally accepted to
insulin resistance. include hyperinsulinemia, glucose intolerance, dyslipide-
mia, central adiposity, and HTN.73-75 Although evidence
exists that insulin resistance improves with the loss of ex-
Changes in body fat distribution
cess body weight, the mediators of the relationship between
glucose dysregulation and weight loss remain unclear.75
Total and abdominal adiposity increase the risk for CVD
and mortality. Increases in total body fat, visceral fat, and
fat deposition in the muscle occur with age and are associ-
ated with insulin resistance, dyslipidemia, and risk for DM Development of obesity-related
and CVD. Upper body fat distribution pattern has been developmental comorbidities in children
shown to be associated with alterations in glucose ho-
meostasis and cardiovascular risk factors in patients with Early maturation
DM and nondiabetic individuals. It has long been hypoth-
esized that abdominal obesity leads to more serious health Children that are obese tend to be taller, mature earlier, have
risks when compared with subcutaneous fat deposition. Ab- advanced bone age, and height acceleration occurs shortly after
dominal or central obesity includes fat surrounding the heart the excessive increase in weight. Early maturation, determined
and other organs, intestinal mesentery, and retroperitoneum by age of menarche and bone age, has been associated with
and is believed to be responsible for CVD possibly related increased adiposity in adulthood and greater truncal obesity in
to drainage into the portal system culminating in an upregu- women.76,77 Early maturation has also been associated with an
lation of normal hepatic metabolic responses. Recent obser- increased adiposity at the same chronologic age, indicating that
vations that DM appears to resolve quickly after Roux-en-Y those individuals were already fatter and at risk for obesity at
gastric bypass may be explained by changes in body fat the time the maturation process started.
distribution. Several recent studies examining changes in
body mass composition because of bariatric surgery have
shown that the first 6 months postoperative appear to be
Orthopedic issues
crucial in terms of changes on body composition.69 Using
total body magnetic resonance imaging to assess visceral fat Orthopedic issues occur related to stress and strain on bone
loss because of bariatric surgery, Busetto and coworkers70 and cartilage that was not designed to carry excess weight.
concluded that, during the initial rapid weight loss period The more common orthopedic problems include bowing of
(ie, 6 months), a preferential mobilization of visceral fat, as the tibia and femurs that result in overgrowth of the medical
compared with total and subcutaneous adipose tissue, can aspect of the proximal tibial metaphysis or Blount syndrome
occur and this may explain in part the metabolic improve- and slipped capital femoral epiphysis due to increased
ments that can occur early after bariatric surgery. As with weight on the growth plate of the hip.78
total body fat, the distribution of body fat between visceral
and subcutaneous compartments also has important genetic
determinants.27 As a component of central adiposity, altered Development of obesity-related metabolic
muscle composition, suggestive of increased fat accumula- comorbidities in children
tion, is an important independent marker of insulin resis-
tance in obesity.71
Type 2 diabetes

Increased inflammatory markers The prevalence of pediatric DM has increased significantly

over the last decade with significant variability based on
Although there are several hypotheses linking obesity with race and ethnicity. Obesity and DM have been tightly linked
various comorbid disease states,72 considerable debate re- in both animal models and adult humans. Recently, data on
mains regarding the identification of common pathophysi- the increasing incidence and prevalence of DM in adoles-
ologic pathways resulting in the development and/or pro- cents demonstrate a similar relationship of obesity and DM
gression of individual comorbid conditions. In conjunction with evidence that insulin resistance and increases in both
with the emerging concept of obesity being a state of total body fat and visceral fat play a role in DM develop-
chronic inflammation, potentiating the development of in- ment in adolescents.79,80
132 Seminars in Pediatric Surgery, Vol 18, No 3, August 2009

Obstructive sleep apnea ovaries. Up to 30% of women with polycystic ovarian

disease are overweight/obese. Obesity can worsen the pic-
Although there is limited data on obstructive sleep apnea ture of polycystic ovary syndrome (PCOS) by increasing
(OSA) in children, the published estimate of prevalence is insulin resistance, diabetes, and metabolic syndrome, but
7% in obese children. Obese children are six times more the increased degree of obesity itself has only a modest
likely than lean, age-matched children to have OSA. Neu- impact on disease development. The prevalence of impaired
rocognitive defects have been demonstrated in children with glucose tolerance in obese young women with PCOS has
OSA.81 Sleep disturbance is linked to higher inflammatory been estimated to be as high as 30-40%, with an additional
biomarkers, such as CRP and IL-6.82 Other hormone 5-10% having frank diabetes.90 Yet, there remains debate in
changes seen include a drop in leptin, an increase in ghrelin, the literature about the relationship between obesity and
increased insulin levels, and a decrease in insulin sensitiv- PCOS.
ity. The etiology for these changes is unclear but felt to be
related to the intermittent hypoxemia that may potentiate the
inflammatory cascade which triggers systemic inflamma- Lipid abnormalities
tion.83-85
Increased blood lipids occur among obese children and
adolescents. Lipid abnormalities are seen in 12-17% of
Hypertension overweight and obese children. The pattern is characterized
by elevated serum low-density lipoprotein– cholesterol and
HTN occurs with relatively low frequency in the pediatric triglycerides and reduced high-density lipoproteins. Central
population. However, when examining the relationship of
fat distribution, possibly through its impact on insulin lev-
obesity to HTN in children, elevated blood pressure oc-
els, may be the mediating variable between obesity and
curred more than nine times more frequently in the obese
lipids. Similar to that process seen in adults, increased free
group.86 Hyperinsulinemia appears to play a role in the
fatty acids produced by lipolysis by the visceral adipocytes
development of HTN. Studies in obese adolescents have
demonstrated increased sodium retention related to excess and hyperinsulinemia promotes hepatic triglyceride and
postprandial insulin secretion and/or hyperinsulinemia as a low-density lipoprotein– cholesterol synthesis.91-93 Weight
possible etiology for this relationship.87,88 reduction improves this profile.

Hepatic steatosis Cardiovascular disease

The incidence of hepatic steatosis is 38% in obese children. Obesity correlates strongly with traditional risk factors of
The underlying cause is unknown, but the condition is CVD, including DM, HTN, hyperlipidemia, and OSA. Obe-
associated with DM, obesity, rapid weight loss, and hyper- sity (central obesity in particular) is also associated with
lipidemia—all of which are characterized by impaired fat increased atherothrombotic events and increased inflamma-
metabolism. One study demonstrated that non-alcoholic tory markers as previously discussed.94,95 Little is known
fatty liver disease (NAFLD) is associated with insulin re- about the occurrence of CVD in young adulthood, but given
sistance even in patients with normal glucose tolerance. In the numbers of risk factors for CVD seen in the adolescent
the pediatric population, results demonstrate increased adi- obese population, close monitoring and surveillance is in-
pose tissue lipolytic activity in obese adolescents with dicated for this “at-risk” population, starting at a young age.
NAFLD with resulting increased rates of fatty acid release There is a group of obese individuals that do not dem-
into plasma throughout the day. This continual excess in onstrate any of the above discussed metabolic abnormali-
fatty acid flux supports the hypothesis that adipose insulin ties. These “healthy” obese appear to have less clustering of
resistance is implicated in the pathogenesis of steatosis and cardiovascular risk factors. In a study by Stefan and co-
contributes to the metabolic complications associated with workers,96 subjects were divided into three groups: normal
NAFLD.89 weight, obese–insulin sensitive, and obese–insulin resistant;
and cardiovascular risk factors were examined. The study
Polycystic ovarian disease found that the obese–insulin sensitive group had insulin
sensitivity, intimal medial thickness similar to that seen in
Polycystic ovarian disease is one of the most common the normal weight group, and when compared with the
endocrine abnormalities and typically begins in adoles- insulin-resistant group had less skeletal muscle fat, less
cence. It is often accompanied by hyperandrogenism and hepatic fat deposits, and lower intimal–media thickness of
hyperinsulinemia. Although the exact mechanism for this the common carotid artery, giving them a more favorable
relationship is unclear, it is postulated that high circulating risk profile despite similar weights. The genetic and envi-
levels of insulin play a role in ovarian cyst development due ronmental factors that place this particular group of obese
to the anabolic effect of insulin at the IGF receptors on the individuals at less risk is unknown.
Schuster Physiology and Obesity 133

Conclusion 9. Ailhaud G. Adipose cell differentiation in culture. Mol Cell Biochem

1982;49:17-31.
The adipose tissue plays a significant, complex role in the 10. Adams M, Montague CT, Prins JB, et al. Activators of peroxisome
proliferator-activated receptor gamma have depot-specific effects on
physiology of hormone regulation and fuel metabolism. The
human preadipocyte differentiation. J Clin Invest 1997;100:3149-53.
development of obesity is a pathophysiologic increase in fat 11. Kissebah AH, Vydelingum N, Murray M, et al. Relation of body fat
mass in which multiple metabolic pathways are deranged. distribution to metabolic complications of obesity. J Clin Endocrinol
There are many possible etiologies for the development of 1982;54:254-60.
obesity with common obesity resulting from both genetic 12. Leibel RL, Rosenbaum M, Hirsch J. Changes in energy expenditure
predisposition and environmental factors. With the environ- resulting from altered body weight. N Engl J Med 1995;332:621.
mental changes of a concomitant decline in physical activity 13. Leibel RL, Hirsch J. Diminished energy requirements in reduced-
obese patients. Metabolism 1984;33:164.
and increased availability of food, genetic predisposition 14. Bray GA, Fisler J, York DA. Neuroendocrine control of the develop-
has been more strongly manifested in a greater prevalence ment of obesity: Understanding gained from studies of experimental
of obesity in adults and children. animal models. Front Neuroendocrinol 1990;11:128.
Furthermore, with the increased prevalence of obesity, 15. Rankinen T, Perusse L, Weisnagel SJ, et al. The human obesity gene
there is an increased risk of obesity-related comorbidities, map: The 2001 update. Obes Res 2002;10:196.
such as DM, OSA, HTN, and CVD, and these deleterious 16. Shutter JR, Graham M, Kinsey AC, et al. Hypothalamic expression of
ART, a novel gene related to agouti, is up-regulated in obese and
side effects can be seen in the setting of pediatric obesity.
diabetic mutant mice. Genes Dev 1997;11:593.
Obese adolescents have increased rates of early maturation, 17. Katsuki A, Sumida Y, Gabazza EC, et al. Plasma levels of agouti-
orthopedic growth abnormalities, DM, OSA, HTN, steato- related protein are increased in obese men. J Clin Endocrinol Metab
sis, and PCOS, placing this group of children at risk for 2001;86:1921.
long-term health problems and reduced quality of life. In 18. Zhang YY, Proenca R, Maffei M, et al. Positional cloning of the mouse
addition to the known health consequences of obesity, there obese gene and its human homolog. Nature 1994;372:425.
19. Campfield LA, Smith FJ, Burn P. The OB protein (leptin) pathway: A
are also significant social barriers for the obese individual.
link between adipose tissue mass and central neural networks. Horm
Studies have demonstrated social and economic conse- Metab Res 1996;28:619.
quences of obesity in adolescents; women completed fewer 20. Chen H, Charlat O, Tartaglia LA, et al. Evidence that the diabetes gene
years of school and were less likely to be married, had lower encodes the leptin receptor: identification of a mutation in the leptin
household incomes and higher rates of poverty; men were receptor gene in db/db mice. Cell 1996;84:491.
less likely to be married.97 21. Schwartz MW, Seeley RJ, Campfield LA, et al. Identification of targets
of leptin action in rat hypothalamus. J Clin Invest 1996;98:1101.
Finally, despite the lack of full understanding of the
22. Bray GA, York DA. Leptin and clinical medicine: a new piece in the
pathophysiology of obesity, it is clear that there are consis- puzzle of obesity. J Clin Endocrinol Metab 1997;82:2771.
tent findings of insulin resistance and inflammation seen in 23. Gerald C, Walker MW, Criscione L, et al. A receptor subtype involved
both obese adults and children, and these metabolic abnor- in neuropeptide-Y-induced food intake. Nature 1996;382:168.
malities play a role in the development of obesity-related 24. Farooqi IS, Jebb SA, Langmack G, et al. Effects of recombinant leptin
comorbidities that are manifested early in the disease pro- therapy in a child with congenital leptin deficiency. N Engl J Med
cess. Given the negative short- and long-term impact of 1999;341:879.
25. Clement K, Vaisse C, Lahlou N, et al. A mutation in the human leptin
obesity on children, careful attention should be paid to the
receptor gene causes obesity and pituitary dysfunction. Nature 1998;
unique health issues of this “at-risk” population with both 392:398.
prevention and aggressive early intervention strategies. 26. Saad MF, Riad-Gabriel MG, Khan A, et al. Diurnal and ultradian
rhythmicity of plasma leptin: Effects of gender and adiposity. J Clin
Endocrinol Metab 1998;83:453.
References 27. Bouchard C, Despres JP, Mauriege P. Genetic and nongenetic deter-
minants of regional fat distribution. Endocr Rev 1993;14:72.
28. Stunkard AJ, Sorensen TI, Hanis C, et al. An adoption study of human
1. Flegal KM, Carroll MD, Ogden CL, et al. Prevalence and trends in
obesity. N Engl J Med 1986;314:193.
obesity among US adults, 1999-2000. J Am Med Assoc 2002;288:
29. Wardle J, Carnell S, Haworth CMA, et al. Evidence for a strong
1723-7.
genetic influence on childhood adiposity despite the force of the
2. Kimm YS, Obarzanek E. Childhood obesity: A new pandemic for the
obesogenic environment. Am J Clin Nutr 2008;87:398-404.
new millennium. Pediatrics 2002;110:1003-7.
30. Woods SC, Seeley RJ, Porte D Jr. Signals that regulate food intake and
3. Kirn TF. Worldwide survey finds adult obesity rate of around 25%.
Circulation 2007;epub doi 10.1161/circulationaha.106.676379. energy homeostasis. Science 1998;280:1378-83.
4. Must A, Jacques PF, Dallal GI, et al. Long-term morbidity and mor- 31. Parfrey PS, Davidson WS, Green JS. Clinical and genetic epidemiol-
tality of overweight adolescents. N Engl J Med 1992;327:1350-5. ogy of inherited renal disease in Newfoundland. Kidney Int 2002;61:
5. Rodeheffer MS, Birsoy K, Friedman JM. Identification of white adi- 1925.
pocyte progenitor cells in vivo. Cell 2008;135:240-9. 32. Fontaine E, Savard R, Tremblay A, et al. Resting metabolic rate in
6. Leibel R, Berry E, Hirsch J. Biochemistry and development of adipose monozygotic and dizygotic twins. Acta Genet Med Gemellol (Roma)
tissue in man. In: Conn HL Jr, DeFelice EA, Kuo P, eds. Health and 1985;34:41.
Obesity. New York, NY: Raven Press, 1983. 33. Bouchard C, Tremblay A, Despres JP, et al. The response to long-term
7. Tang W, Zeve D, Suh J. White fat progenitor cells reside in the adipose overfeeding in identical twins. N Engl J Med 1990;322:1477.
vasculature. Science 2008;322:583-6. 34. Ravussin E, Lillioja S, Knowler WC, et al. Reduced rate of energy
8. Knittle J, Timmers K, Ginsberg-Fellner F, et al. The growth of adipose expenditure as a risk factor for body-weight gain. N Engl J Med
tissue in children and adolescents. J Clin Invest 1979;63:239-46. 1988;318:467.
134 Seminars in Pediatric Surgery, Vol 18, No 3, August 2009

35. Weisnsier RL, Hunter GR, Heini AF, et al. The etiology of obesity: 58. Schauer PR, Burguera B, Ikramuddin S, et al. Effect of laparoscopic
relative contribution of metabolic factors, diet, and physical activity. Roux-en-Y gastric bypass on type 2 diabetes mellitus. Ann Surg
Am J Med 1998;105:145-50. 2003;238:467-85.
36. Perusse L, Tremblay A, Lablanc C, et al. Genetic and environmental 59. Sugerman HJ, Wolfe LG, Sica DA, et al. Diabetes and hypertension in
influences on level of habitual physical activity and exercise partici- severe obesity and effects of gastric bypass-induced weight loss. Ann
pation. Am J Epidemiol 1989;129:1012-22. Surg 2003;237:751-8.
37. Engeli S, Feldpaussch M, Gorzelniak K, et al. Association between 60. Wittgrove AC, Clark GW. Laparoscopic gastric bypass, Roux-en-Y
adiponectin and mediators of inflammation in obese women. Diabetes 500 patients: Technique and results, with 3-60-month follow-up. Obes
2003;52:942-7. Surg 2000;10:233-9.
38. Straczkowski MA, Kowalska I, Stepien A, et al. Increased plasma- 61. Schauer PR, Ikramuddin S, Gourash W, et al. Outcomes after laparo-
soluble tumor necrosis factor alpha-receptors level in lean, nondiabetic scopic Roux-en-Y gastric bypass for morbid obesity. Ann Surg 2000;
offspring of type 2 diabetic subjects. Diabetes Care 2002;25:1824-8. 232:515-29.
39. Krakoff J, Funahaser T, Stenhouwer CDA, et al. Inflammatory mark- 62. Zander M, Madsbad S, Madsen JL, et al. Effect of 6-week course of
ers, adiponectin and the risk of type 2 diabetes in the Pima Indians. glucagon-like peptide 1 on glycaemic control, insulin sensitivity, and
Diabetes Care 2003;26:1745-51. beta-cell function in type 2 diabetes: a parallel-group study. Lancet
40. Sprangler J, Kroke A, Mohlig M, et al. Adiponectin and protection 2002;359:824-30.
against type 2 diabetes. Lancet 2003;361:226-8. 63. Zhang JV, Ren PG, Avsian-Kretchmer O, et al. Obestatin, a peptide
41. Foster GD, Wadden TA, Mullen JL, et al. Resting energy expenditure, encoded by the ghrelin gene, opposes ghrelin’s effects on food intake.
body composition and excess weight in the obese. Metabolism 1988; Science 2005;310:996-9.
37:467-72. 64. Del Prato S, Enzi G, Vigili de Kreutzenberg S, et al. Insulin regulation
42. Considine RV, Sinha MK, Heiman ML, et al. Serum immunoreactive- of glucose and lipid metabolism in massive obesity. Diabetologia
leptin concentrations in normal-weight and obese humans. N Engl 1990;33:228-36.
J Med 1996;344:292. 65. Lillioja S, Foley J, Bogardus C, et al. Free fatty acid metabolism and
43. Griffen WO, Bivins BA, Bell RM, et al. Gastric bypass for morbid obesity in Man: in vivo and in vitro comparisons. Metabolism 1986;
obesity. World J Surg 1981;5:817-22. 35:505-14.
44. Hickey MS, Pories WJ, MacDonald KG, et al. A new paradigm for 66. Polonsky KS, Given BD, Lirsch L, et al. Quantitative study of insulin
type 2 diabetes mellitus: Could it be a disease of the foregut? Ann Surg secretion and clearance in normal and obese subjects. J Clin Invest
1998;227:637-44. 1988;81:435-41.
45. Cummings DE, Weigle DS, Frayo RS, et al. Plasma ghrelin levels after 67. Monti LD, Brambilla P, Stefani I, et al. Insulin regulation of glucose
diet-induced weight loss or gastric bypass surgery. N Engl J Med turnover and lipid levels in obese children with fasting normoinsuli-
2002;346:1623. naemia. Diabetologia 1995;38:739-47.
46. Foster-Schubert KE, McTiernan A, Frayo RS, et al. Human plasma 68. LeStunff C, Bourgeres P. Early changes in postprandial insulin secre-
ghrelin levels increase during a one-year exercise program. J Clin tion, not in insulin sensitivity, characterize juvenile obesity. Diabetes
Endocrinol Metab 2005;90:820. 1994;43:696-702.
47. Taylor IL. Role of peptide YY in the endocrine control of digestion. J 69. Giusti V, Suter M, Heraief E, et al. Effects of laparoscopic gastric
Dairy Sci 1993;76:2094-101. banding on body composition, metabolic profile and nutritional status
48. Batterham RL, Cohen MA, Ellis SM, et al. Inhibition of food intake in of obese women: 12-month follow-up. Obes Surg 2004;14:239-45.
obese subjects by peptide YY3-36. N Engl J Med 2003;349:941-8. 70. Busetto L, Tregnaghi A, Bussolotto M, et al. Visceral fat loss evalu-
49. Korner J, Bessler M, Cirilo LJ, et al. Effects of Roux-en-Y gastric ated by total body magnetic resonance imaging in obese women
bypass surgery on fasting and postprandial concentrations of plasma operated with laparoscopic adjustable silicone gastric banding. Int J
ghrelin, peptide YY, and insulin. J Clin Endocrinol Metab 2005;90: Obes 2000;24:60-9.
359-65. 71. Goodpaster BH. Subcutaneous abdominal fat and thigh muscle com-
50. Kellum JM, Kuemmerle JF, O’Dorisio TM, et al. Gastrointestinal position predict insulin sensitivity independently of visceral fat. Dia-
hormone responses to meals before and after gastric bypass and ver- betes 1997;46:1579-85.
tical banded gastroplasty. Ann Surg 1990;211:763-70. 72. Olshansky SJ, Passaro DJ, Hershow RC, et al. Potential decline in life
51. Naslund E, Backman L, Holst JJ, et al. Importance of small bowel expectancy in the United States in the 21st century. N Engl J Med
peptides for the improved glucose metabolism 20 years after jejuno- 2005;352:1138-45.
ileal bypass for obesity. Obes Surg 1998;8:253-60. 73. Gastrointestinal surgery for severe obesity. Proceedings of a National
52. Sarson DL, Scopinaro N, Bloom SR. Gut hormone changes after Institutes of Health Consensus Development Conference. March 25-
jejunoileal (JIB) or biliopancreatic (BPB) bypass surgery for morbid 27, 1991, Bethesda, MD. Am J Clin Nutr 1992;55:487S-619S (suppl).
obesity. Int J Obes 1981;5:471-80. 74. Long SD, O’Brien K, MacDonald KG Jr, et al. Weight loss in severely
53. Druce MR, Small CJ, Bloom SR. Minireview: Gut peptides regulating obese subjects prevents the progression of impaired glucose tolerance
satiety. Endocrinology 2004;145:2660-5. to type II diabetes. A longitudinal interventional study. Diabetes Care
54. Verdich C, Toubro S, Buemann B, et al. The role of postprandial 1994;17(5):372-5.
releases of insulin and incretin hormones in meal-induced satiety: 75. Dixon JB, Dixon AF, O‘Brien PE. Improvements in insulin sensitivity
Effect of obesity and weight reduction. Int J Obes Relat Metab Disord and beta-cell function (HOMA) with weight loss in the severely obese.
2001;25:1206-14. Homeostatic model assessment. Diabet Med 2003;20(2):127-34.
55. Velasquez-Mieyer PA, Cowan PA, Umpierrez GE, et al. Racial dif- 76. Garn SM, LaVelle M, Rosenberg KR, et al. Maturational timing as a
ferences in glucagon-like peptide-1 (GLP-1) concentrations and insu- factor in female fatness and obesity. Am J Nutr 1986;43:879-83.
lin dynamics during oral glucose tolerance test in obese subjects. Int J 77. Van Lenthe FJ, Kemper HCG, van Mecehelen W, et al. Biological
Obes Relat Metab Disord 2003;27:1359-64. maturation and the distribution of subcutaneous fat from adolescence
56. Feinle C, Chapman IM, Wishart J, et al. Plasma glucagon-like pep- into adulthood: the Amsterdam Growth and Health study. Int J Obes
tide-1 (GLP-1) responses to duodenal fat and glucose infusions in lean 1996;20:121-9.
and obese men. Peptides 2002;23:1491-5. 78. Kelsey JL, Keggi KJ, Southwick WO. The incidence and distribution
57. Pappas TN. Physiological satiety implications of gastrointestinal anti- of slipped femoral epiphysis in Connecticut and southwestern United
obesity surgery. Am J Clin Nutr 2002;55:571S-2S. States. J Bone Joint Surg 1970;52A:1203-16.
Schuster Physiology and Obesity 135

79. Pinhas-Hamiel O, Dolan LM, Daniels SR, et al. Increased incidence of 89. Fabbrini E, deHaseth D, Deivanayagam D, et al. Alterations in fatty
non-insulin-dependent diabetes mellitus among adolescents. J Pediatr acid kinetics in obese adolescents with increased intrahepatic triglyc-
1996;128:608-15. eride content. Obesity 2008;17:25-9.
80. Richards GE, Cavallo A, Meyer WJ III, et al. Obesity, acanthosis 90. Ehrmann DA, Barnes RB, Rosenfield RL, et al. Prevalence of impaired
nigricans, insulin resistance, and hyperandrogenemia: pediatric per- glucose tolerance and diabetes in women with polycystic ovary syn-
spective and natural history. J Pediatr 1985;107:893-7. drome. Diabetes Care 1999;22:141-6.
81. Rhodes SK, Shimoda KC, Waid LR, et al. Neurocognitive deficits in 91. Freedman DS, Srinivasan SR, Harsha DW, et al. Relation of body fat
morbidly obese children with obstructive sleep apnea. J Pediatr 1995; patterning to lipid and lipoprotein concentrations in children and ad-
127:741-4. olescents: The Bogalusa Heart Study. Am J Clin Nutr 1989;50:930-9.
82. Suarez EC. Self-reported symptoms of sleep disturbance and inflam- 92. Steinberger J, Moorehead C, Katch V, et al. Relationship between
mation, coagulation, insulin resistance and psychosocial distress: evi- insulin resistance and abnormal lipid profile in obese adolescents.
dence for gender disparity. Brain Behav Immun 2008;22:960-8. J Pediatr 1995;126:690-5.
83. Bonnet MH, Berry RB, Arand DL. Metabolism during normal, frag- 93. Caprio S, Hyman LD, Limb C, et al. Central adiposity and its meta-
mented, and recovery. J Appl Physiol 1991;71:1112-8. bolic correlates in obese adolescent girls. Am J Physiol 1995;269:
84. Somers VK, Dyken ME, Clary MP, et al. Sympathetic neural mech- E118-26.
anisms in obstructive sleep apnea. J Clin Invest 1995;96:1897-904. 94. Shimomura I, Funahashi T, Takahashi M, et al. Enhanced expression
85. Meslier N, Gagnadoux P, Giraud C, et al. Impaired glucose-insulin of PAI-1 in visceral fat: possible contributor to vascular disease in
metabolism in males with obstructive sleep apnea syndrome. Eur obesity. Nat Med 1996;2:800-3.
Respir J 2003;22:156-60. 95. Eriksson P, Reynisdottir S, Lonnqvist F, et al. Adipose tissue secretion
86. Lauer RM, Connor WE, Leaverton PE, et al. Coronary heart disease of plasminogen activator inhibitor-1 in non-obese and obese individ-
risk factors in school children: The Muscatine study. J Pediatr 1975; uals. Diabetologia 1998;41:65-71.
86:697-706. 96. Stefan N, Kantartzis K, Machann J, et al. Identification and character-
87. Finta KM, Rocchini AP, Moorehead C, et al. Urine sodium excretion ization of metabolically benign obesity in humans. Arch Intern Med
in response to an oral glucose tolerance test in obese and nonobese 2008;168:1609-16.
adolescents. Pediatrics 1992;90:442-6. 97. Gortmaker SL, Must A, Perrin JM, et al. Social and economic conse-
88. Rocchini AP, Katch V, Anderson J, et al. Blood pressure in obese quences of overweight in adolescence and young adulthood. N Engl
adolescents: Effect of weight loss. Pediatrics 1988;82:16-23. J Med 1993;329:1008-12.