Amylin and its relation to insulin and lipids in obese children before and after weight loss.
ABSTRACT There are limited data concerning the relationships between amylin, weight status, lipids, insulin, and insulin resistance in obese humans. Therefore, the aim was to study these relationships in cross-sectional and longitudinal analyses.
Fasting amylin, insulin, glucose, triglycerides, low-density lipoprotein (LDL)- and high-density lipoprotein (HDL)-cholesterol, and percentage body fat based on skinfold measurements were determined in 37 obese children (median age, 11.5 years) and compared with 16 lean children of the same age and gender. Furthermore, we analyzed the changes of these variables in the obese children after participating in a one-year weight loss intervention program.
Obese children had significantly (p < 0.01) higher amylin, triglycerides, LDL-cholesterol, and insulin levels as compared with the lean children. In multiple linear regression analysis, amylin was significantly (p < 0.05) correlated to insulin and triglycerides, but not to age, gender, pubertal stage, or BMI. Changes of amylin correlated significantly (p < 0.001) to changes of insulin (r = 0.54) and triglycerides (r = 0.49), but not to changes of BMI or percentage body fat. Substantial weight loss in 17 children led to a significant (p < 0.05) decrease of amylin, triglycerides, and insulin, in contrast to the 20 children without substantial weight loss.
Amylin levels were related to insulin concentrations in both cross-sectional and longitudinal analyses, suggesting a relationship between amylin and insulin secretion. Amylin levels were reversibly increased in obesity and related to triglyceride concentrations.
- [Show abstract] [Hide abstract]
ABSTRACT: Background Gut hormones change with weight loss in adults but are not well studied in obese youth.Objective The primary aim was to evaluate how gut hormones and subjective appetite measure change with dietary weight loss in obese adolescents.Methods Participants were a subset of those taking part in the ‘Eat Smart Study’. They were aged 10–17 years with body mass index (BMI) > 90th centile and were randomized to one of three groups: wait-listed control, structured reduced carbohydrate or structured low-fat dietary intervention for 12 weeks. Outcomes were fasting glucose, insulin, leptin, adiponectin, total amylin, acylated ghrelin, active glucagon-like peptide-1, glucose-dependent insulinotropic polypeptide (GIP), pancreatic polypeptide (PP) and total peptide tyrosine-tyrosine. Pre- and postprandial subjective sensations of appetite were assessed using visual analogue scales.ResultsOf 87 ‘Eat Smart’ participants, 74 participated in this sub-study. The mean (standard deviation) BMI z-score was 2.1 (0.4) in the intervention groups at week 12 compared with 2.2 (0.4) in the control group. Fasting insulin (P = 0.05) and leptin (P = 0.03) levels decreased, while adiponectin levels increased (P = 0.05) in the intervention groups compared with control. The intervention groups were not significantly different from each other. A decrease in BMI z-score at week 12 was associated with decreased fasting insulin (P < 0.001), homeostatic model of assessment-insulin resistance (P < 0.001), leptin (P < 0.001), total amylin (P = 0.03), GIP (P = 0.01), PP (P = 0.02) and increased adiponectin (P < 0.001). There was no significant difference in appetite sensations.Conclusions Modest weight loss in obese adolescents leads to changes in some adipokines and gut hormones that may favour weight regain.Pediatric Obesity 01/2015; · 2.42 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: Background Insulin resistance is found in both psoriasis and Behçet's disease. No study has yet explored whether preptin and amylin, two hormones associated with insulin resistance, are involved in the insulin resistance observed in patients with psoriasis and Behçet's disease.Objectives We aimed to explore how the amounts of preptin and amylin change in psoriasis and Behçet's disease and whether they are involved in the etiopathology of these two diseases, by comparing hormone levels in patients and healthy controls.Methods The study registered 30 patients with psoriasis, 30 patients with Behçet's disease, and 30 healthy volunteers (as a control group). Fasting blood sugar, triglyceride, LDL, VLDL, HDL, total cholesterol, HbA1c, C-peptide, insulin, and serum preptin and amylin levels were measured in all subjects.ResultsSerum preptin and amylin levels were significantly lower in the patients with psoriasis and Behçet's disease than in the control group (P < 0.001, P = 0.004, and P = 0.008, respectively). A comparison of the serum preptin and amylin levels between the patients with psoriasis and Behçet's disease did not reveal a statistically significant difference. Serum insulin level and The homeostasis model assessment of insulin resistance (HOMA-IR) index were significantly lower in the psoriasis patient group relative to the control group (P = 0.02 and P = 0.03, respectively), while the values for the Behçet's disease group did not differ significantly from those for the control groupConclusions Serum levels of preptin and amylin were significantly lower in patients with psoriasis and Behçet's disease, indicating that these hormones may be a factor for development of metabolic syndrome in these two diseases.Journal of Clinical Laboratory Analysis 12/2014; · 1.14 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: The gastrointestinal (GI) tract comprises a large endocrine organ that regulates not only nutrient sensing and metabolising but also satiety and energy homeostasis. More than 20 hormones secreted from the stomach, intestine, and pancreas as well as signaling mediators of the gut microbiome are involved in this process. A better understanding of how related pathways affect body weight and food intake will help us to find new strategies and drugs to treat obesity. For example, weight loss secondary to lifestyle intervention is often accompanied by unfavorable changes in multiple GI hormones, which may cause difficulties in maintaining a lower body weight status. Conversely, bariatric surgery favorably changes the hormone profile to support improved satiety and metabolic function. This partially explains stronger sustained body weight reduction resulting in better long-term results of improved metabolic functions. This review focuses on GI hormones and signaling mediators of the microbiome involved in satiety regulation and energy homeostasis and summarizes their changes following weight loss. Furthermore, the potential role of GI hormones as anti-obesity drugs is discussed.Endocrine. 12/2014;
Amylin and Its Relation to Insulin and Lipids
in Obese Children Before and After
Thomas Reinehr,* Gideon de Sousa,* Petra Niklowitz,* and Christian L. Roth†
REINEHR, THOMAS, GIDEON DE SOUSA, PETRA
NIKLOWITZ, AND CHRISTIAN L. ROTH. Amylin and
its relation to insulin and lipids in obese children before and
after weight loss. Obesity. 2007;15:2006–2011.
Objective: There are limited data concerning the relation-
ships between amylin, weight status, lipids, insulin, and
insulin resistance in obese humans. Therefore, the aim was
to study these relationships in cross-sectional and longitu-
Research Methods and Procedures: Fasting amylin, insu-
lin, glucose, triglycerides, low-density lipoprotein (LDL)-
and high-density lipoprotein (HDL)-cholesterol, and per-
centage body fat based on skinfold measurements were
determined in 37 obese children (median age, 11.5 years)
and compared with 16 lean children of the same age and
gender. Furthermore, we analyzed the changes of these
variables in the obese children after participating in a one-
year weight loss intervention program.
Results: Obese children had significantly (p ? 0.01) higher
amylin, triglycerides, LDL-cholesterol, and insulin levels as
compared with the lean children. In multiple linear regres-
sion analysis, amylin was significantly (p ? 0.05) correlated
to insulin and triglycerides, but not to age, gender, pubertal
stage, or BMI. Changes of amylin correlated significantly
(p ? 0.001) to changes of insulin (r ? 0.54) and triglycer-
ides (r ? 0.49), but not to changes of BMI or percentage
body fat. Substantial weight loss in 17 children led to a
significant (p ? 0.05) decrease of amylin, triglycerides, and
insulin, in contrast to the 20 children without substantial
Conclusion: Amylin levels were related to insulin concen-
trations in both cross-sectional and longitudinal analyses,
suggesting a relationship between amylin and insulin secre-
tion. Amylin levels were reversibly increased in obesity and
related to triglyceride concentrations.
Key words: GI hormones, gastroenterology
Amylin, a 37-amino acid protein mainly secreted by the
pancreatic ? cells in response to nutrient stimuli, has mul-
tiple actions, including the inhibiting of gastric emptying
and the secretion of glucagon and insulin, lipase, and amy-
lase (1–4). Furthermore, an involvement of amylin in short-
and long-term effects of the regulation of food intake and
body weight is being discussed based on studies in rodents
(2,5). The central and peripheral administration of amylin in
rodents reduced their food intake in a dose-dependent man-
ner (6,7). When rats were given an amylin antagonist, a
significant increase of energy intake was observed (6).
Although the exact mechanisms remain unclear, amylin
binding sites have been identified in the nucleus of the
solitary tract, nucleus accumbens, and hypothalamus (8).
All these findings pave the way for amylin analogues as
new anti-obesity drugs (1).
In humans, there are only a few studies concerning the
relationship between amylin and overweight. These studies
demonstrated that amylin concentrations correlate with the
degree of overweight and that the basal amylin concentra-
tions are higher in obese than in lean human subjects
(1,9,10). Studies of amylin concentrations in obese humans
losing weight have not been performed as yet. Furthermore,
it is unclear whether the increase of amylin in obesity is a
consequence of obesity per se or a consequence of other
factors associated with obesity.
Because amylin is co-secreted with insulin in the ? cells
of the pancreas, the increase of amylin in obesity could also
Received for review October 4, 2006.
Accepted in final form January 2, 2007.
The costs of publication of this article were defrayed, in part, by the payment of page
charges. This article must, therefore, be hereby marked “advertisement” in accordance with
18 U.S.C. Section 1734 solely to indicate this fact.
*Vestische Hospital for Children and Adolescents Datteln, University of Witten/Herdecke,
Datteln, Germany; and †Department of Pediatrics, University of Bonn, Bonn, Germany.
Address correspondence to Thomas Reinehr, Vestische Hospital for Children and Adoles-
cents Datteln, University of Witten/Herdecke, Dr. F. Steiner Str. 5, 45711 Datteln, Germany.
Copyright © 2007 NAASO
2006OBESITY Vol. 15 No. 8 August 2007
be a consequence of increased insulin levels, which have
been demonstrated in obese children (11). Additionally, a
role for amylin has been suggested in the pathogenesis of
dyslipidemia in impaired glucose metabolism. Hyperamy-
linemia, which occurs in type 2 diabetes, may contribute to
a number of metabolic abnormalities present in the diabetic
state (12). Amylin caused insulin resistance in skeletal mus-
cles, stimulated lipolysis, and inhibited insulin secretion in
animal models (12–15). Moreover, amylin can reduce chy-
lomicron uptake, leading to an increased concentration of
triglyceride-rich remnant lipoproteins (12).
Because there are only limited data concerning the rela-
tionships between amylin, weight status, lipids, insulin, and
insulin resistance in obese humans, we studied amylin lev-
els and their changes in obese children as well as their
relations to insulin, lipids, and insulin resistance in the
course of one year.
Research Methods and Procedures
We examined anthropometric markers, fasting serum
amylin, glucose, insulin, triglycerides, and high-density li-
poprotein (HDL)1- and low-density lipoprotein (LDL)-cho-
lesterol concentrations in 37 obese white children and in 16
lean healthy white children of similar age, gender, and
pubertal stage. Furthermore, the obese children were studied
before and after participating in the one-year obesity inter-
vention program “Obeldicks,” which has been described in
detail elsewhere (16,17). Briefly, this outpatient interven-
tion program for obese children was based on physical
exercise, nutrition education, and behavior therapy includ-
ing individual psychological care of the child and his or her
family. The nutritional course was based on a fat and sugar
reduced diet as compared with the everyday nutrition of
German children (17): The diet contained 30% fat, 15%
proteins, and 55% carbohydrates, including 5% sugar.
Children with endocrine disorders, premature adrenarche,
or syndromal obesity were excluded from the study. Obesity
was defined according to the definition of the International
Task Force of Obesity using population-specific data (18).
Height was measured to the nearest centimeter using a
rigid stadiometer. Weight was measured unclothed to the
nearest 0.1 kg using a calibrated balance scale. Because
BMI is not normally distributed in childhood, we used the
lambda-mu-sigma (LMS) method to calculate standard de-
viation score (SDS)-BMI as a measurement for the degree
of overweight. The LMS method summarizes the data in
terms of three smooth age specific curves called L (?), M
(?), and S (?) based on German population-specific data
(19,20). The M and S curves correspond to the median and
coefficient of variation body mass index for German chil-
dren at each age and gender, whereas the L curve allows for
the substantial age-dependent skewness in the distribution
of BMI (19,20). The assumption underlying the LMS
method is that, after Box-Cox power transformation, the
data at each age are normally distributed (20).
The triceps and subscapularis skinfold thickness was
measured in duplicate by using a caliper and averaged to
calculate the percentage of body fat using a skinfold thick-
ness equation with the following formulas (21): boys, body
fat % ? 0.783x (subscapularis skinfold thickness ? triceps
skinfold thickness in mm) ? 1.6; girls, body fat % ? 0.546x
(subscapularis skinfold thickness ? triceps skinfold thick-
ness in mm) ? 9.7.
The pubertal developmental stage was determined ac-
cording to Marshall and Tanner and categorized into two
groups (prepubertal, boys with pubic hair and gonadal Stage
I, girls with pubic hair stage and breast Stage I; pubertal,
boys with pubic hair or gonadal Stage ?II and girls with
pubic hair stage or breast Stage ?II).
Blood sampling was performed in the fasting state at 8
AM. Serum amylin concentrations were measured by a high-
specific enzyme-linked immunosorbent assay (Human
Amylin ELISA Kit; Linco Research, St. Charles, MO). The
antibody does not cross-react with insulin, calcitonin, cal-
citonin gene-related peptide, glucagon-like peptide, pancre-
atic polypeptide, or other known gastrointestinal hormones.
The sensitivity was 1 pM/L. Insulin concentrations were
measured by microparticle enhanced immunometric assay
(MEIA; Abbott, Wiesbaden, Germany). Glucose levels
were determined by colorimetric test using a Vitros analyzer
(Ortho Clinical Diagnostics, Neckargmuend, Germany).
HDL- and LDL-cholesterol concentrations were measured
by an enzymatic test (HDL-C- Plus and LDL-C- Plus;
Roche Diagnostics, Mannheim, Germany), and triglyceride
concentrations by a colorimetric assay using a Vitros ana-
lyzer (Ortho Clinical Diagnostics). Intra- and inter-assay
coefficients of variation were ?5% in all methods. Ho-
meostasis model assessment (HOMA) was used to detect
the degree of insulin resistance (22); the resistance can be
assessed from the fasting glucose and insulin concentrations
by the formula: resistance (HOMA) ? [insulin (mU/L) ?
Substantial weight loss in the course of the one-year
intervention was defined by a reduction of SDS-BMI ?0.5
because, with a reduction of ?0.5 SDS-BMI, no improve-
ment of insulin resistance and cardiovascular risk factors
could be measured in obese children (23,24).
Statistical analysis was performed using the Winstat soft-
ware package. Apart from amylin, all variables were nor-
mally distributed tested by Kolmogorov-Smirnov test.
Therefore, amylin was log-transformed. Student’s t test for
paired and unpaired observations, and Mann Whitney U test
and Wilcoxon test were used as appropriate. Correlations
1Nonstandard abbreviations: HDL, high-density lipoprotein; LDL, low-density lipoprotein;
LMS, lambda-mu-sigma; SDS, standard deviation score; HOMA, homeostasis model as-
sessment; IQR, interquartile range.
Amylin in Obese Children, Reinehr et al.
OBESITY Vol. 15 No. 8 August 20072007
between log-transformed amylin, lipids, insulin, and insulin
resistance index (HOMA) at baseline, as well as correlations
between changes of weight status, amylin, lipids, and insu-
lin in the course of one year were calculated by Pearson’s
correlation. Changes were expressed as ? variable calcu-
lated by variable at baseline and variable one year later. A
backward multivariate linear regression analysis was con-
ducted for the dependent variable log-transformed amylin
including age, gender, pubertal stage, weight status (BMI),
lipids, and insulin as independent variables in the summa-
rized collective of 37 obese and 16 lean children. To express
the relationships, log-transformed amylin concentrations
were retransformed. A backward multivariate linear regres-
sion analysis was conducted for the dependent variable ?
amylin, including age, gender, pubertal stage, ? SDS-BMI,
? lipids, and ? insulin as independent variables in the 37
obese children. Gender and pubertal stage were used as
classified variables in these models. A p value ?0.05 was
considered significant. Data were presented as median and
interquartile range (IQR). Written informed consent was
obtained from all children and their parents. The study was
approved by the local ethics committee of the University of
The age, stage of puberty, gender, degree of overweight
(BMI, SDS-BMI), amylin levels, lipids, and insulin concen-
trations of the 37 obese and 16 lean children are shown in
Table 1. Obese children demonstrated a significantly higher
degree of insulin resistance index (HOMA), as well as
higher concentrations of amylin, triglycerides, LDL-choles-
terol, and insulin as compared with the lean children. The
lean and obese children did not differ significantly in terms
of age, gender, or pubertal status.
In the 53 obese and normal weight children, log-trans-
formed amylin correlated significantly to insulin (r ? 0.59,
p ? 0.001), triglycerides (r ? 0.55, p ? 0.001), LDL-
cholesterol (r ? 0.41, p ? 0.001), HDL-cholesterol (r ?
?0.30, p ? 0.012), insulin resistance index (HOMA) (r ?
0.42, p ? 0.001), SDS-BMI (r ? 0.46, p ? 0.001), and
percentage body fat (r ? 0.36, p ? 0.004) at baseline. We
found no significant correlation between amylin levels and
age (r ? ?0.11, p ? 0.204). In backward multivariate linear
regression analysis (r2? 0.42, n ? 53), log-transformed
amylin was significantly correlated to insulin (coefficient,
1.06; 95% confidence interval, 0.02 to 2.10, p ? 0.003) and
triglycerides (coefficient, 1.01; 95% confidence interval,
0.03 to 1.99, p ? 0.018), whereas log-transformed amylin
demonstrated no significant correlation to age, gender, stage
of puberty, or BMI. Multiple regression analyses with SDS-
BMI or percentage body fat instead of BMI demonstrated
no significant correlation between SDS-BMI or percentage
body fat and amylin.
resistance index (HOMA), insulin, glucose, lipids, and amylin concentrations in obese and lean children
Age, stage of puberty, gender, degree of overweight (BMI, SDS-BMI), percentage body fat, insulin
Triceps skinfold thickness (mm)
Subscapularis skinfold thickness (mm)
Percentage body fat
11.5 (9.0 to 12.0)
26.3 (23.6 to 28.7)
2.7 (2.3 to 3.0)
29 (25 to 33)
29 (26 to 32)
44 (37 to 50)
92 (71 to 145)
112 (95 to 131)
46 (42 to 53)
87 (84 to 90)
15 (12 to 20)
3.2 (2.4 to 4.0)
9.3 (6.3 to 13.9)
11.5 (9.9 to 12.3)
19.5 (18.1 to 20.6)
0.7 (0.6 to 0.9)
17 (15 to 19)
17 (15 to 19)
28 (25 to 31)
65 (54 to 83)
90 (87 to 96)
50 (48 to 57)
87 (84 to 90)
5 (4 to 7)
2.2 (1.6 to 2.8)
2.6 (1.1 to 5.2)
SDS, standard deviation score; HOMA, homeostasis model assessment; LDL, low-density lipoprotein; HDL, high-density lipoprotein. Data
are median (interquartile range).
Amylin in Obese Children, Reinehr et al.
2008OBESITY Vol. 15 No. 8 August 2007
The changes of insulin resistance index (HOMA), insulin,
glucose, lipids, and amylin concentrations in the course of
one year in the 17 obese children with substantial weight
loss and the 20 obese children without substantial weight
loss are shown in Table 2. Substantial weight loss led to a
significant decrease of amylin, triglycerides, LDL-choles-
terol, and insulin concentrations as well as to a decrease of
insulin resistance index (HOMA). In the obese children
without substantial weight loss, there were no significant
changes in insulin resistance index (HOMA), insulin, amy-
lin, and lipid levels apart from a decrease of LDL-choles-
At baseline, we found no significant differences in age
(p ? 0.070), gender (p ? 0.634), pubertal stage (p ? 0.902),
percentage body fat (p ? 0.862), and SDS-BMI (p ? 0.126)
between the obese children with and without substantial
weight loss. Furthermore, glucose (p ? 0.374), insulin (p ?
0.424), HOMA (p ? 0.301), triglycerides (p ? 0.284),
LDL-cholesterol (p ? 0.340), HDL-cholesterol (p ? 0.836),
and amylin concentrations (p ? 0.843) did not differ sig-
nificantly at baseline between the children with and without
substantial weight loss. One year later, percentage body fat
(p ? 0.031), triglycerides (p ? 0.031), insulin (p ? 0.022),
insulin resistance index (HOMA) (p ? 0.021), and amylin
concentrations (p ? 0.024) were significantly lower in the
children with substantial weight loss as compared with the
children without substantial weight loss.
In the 37 obese children, the changes of amylin concen-
trations in the course of one year correlated significantly to
those of insulin (r ? 0.54, p ? 0.001, Figure 1) and
triglycerides (r ? 0.49, p ? 0.001, Figure 2), but not to
changes of SDS-BMI (r ? 0.23, p ? 0.082), percentage
body fat (r ? 0.17, p ? 0.167), LDL-cholesterol (r ?
?0.12, p ? 0.238), or HDL-cholesterol (r ? ?0.12, p ?
0.241). In backward multivariate linear regression analysis
(r2? 0.37, n ? 37), changes of amylin were significantly
correlated to changes of insulin (coefficient, 0.22; 95%
confidence interval, 0.05 to 0.38, p ? 0.011) and triglycer-
ides (coefficient, 0.03; 95% confidence interval, 0.01 to
0.05, p ? 0.048), whereas changes of amylin demonstrated
no significant correlation to changes of SDS-BMI.
At baseline, the amylin levels of the 27 girls (median, 7.9;
IQR, 3.2 to 12.3 pM/L) did not differ significantly (p ?
0.917) from those of the 26 boys (median, 7.4; IQR, 4.7 to
lipids, and amylin concentrations in 17 obese children with substantial weight loss and 20 children with stable
weight status over a 1-year period
Changes of weight status, percentage body fat, insulin resistance index (HOMA), insulin, glucose,
Substantial weight lossNo change of weight status
Change of SDS-BMI
11.2 (8.9 to 11.9)
?0.6 (?0.7 to ?0.5)
26.8 (25.8 to 30.2)
2.4 (2.1 to 2.8)
30 (26 to 32)
30 (25 to 32)
43 (38 to 50)
82 (71 to 118)
109 (95 to 128)
46 (42 to 52)
16 (13 to 17)
85 (82 to 90)
3.1 (2.6 to 3.5)
11.6 (5.9 to 14.5)
11.9 (10.1 to 12.4)
0.0 (?0.1 to 0.3)
27.1 (25.5 to 30.1)
2.1 (1.9 to 2.5)
30 (26 to 33)
30 (25 to 34)
44 (37 to 52)
97 (68 to 120)
114 (95 to 141)
46 (41 to 56)
15 (9 to 23)
87 (85 to 89)
3.2 (1.8 to 5.1)
9.2 (6.5 to 12.8)
1 year later
23.1 (22.1 to 25.9)*
1.9 (1.4 to 2.1)*
22 (19 to 26)*
22 (19 to 25)*
33 (31 to 43)*
67 (51 to 101)*
88 (76 to 114)*
52 (45 to 58)
9 (6 to 14)*
85 (83 to 89)
1.7 (1.2 to 2.8)*
5.5 (4.2 to 9.2)*
1 year later
26.8 (24.7 to 29.5)
2.1 (1.9 to 2.5)
29 (26 to 34)
29 (26 to 34)
43 (33 to 50)
113 (57 to 144)
93 (78 to 111)*
51 (38 to 58)
16 (9 to 24)
88 (85 to 91)
3.4 (1.9 to 5.0)
10.3 (6.2 to 12.6)
Subscapularis ST (mm)
Triceps ST (mm)
Percentage body fat
HOMA, homeostasis model assessment; SDS, standard deviation score; ST, skinfold thickness; LDL, low-density lipoprotein; HDL,
high-density lipoprotein. Data are median (interquartile range).
* p ? 0.05 baseline vs. 1 year later.
Amylin in Obese Children, Reinehr et al.
OBESITY Vol. 15 No. 8 August 20072009
12.7 pM/L). Boys and girls did not differ with respect to age
(p ? 0.464), pubertal stage (p ? 0.502), and SDS-BMI (p ?
0.660). We found no significant difference (p ? 0.754)
between the amylin levels of the 29 prepubertal children
(median, 6.4; IQR, 2.8 to 11.0 pM/L) and the 24 pubertal
children (median, 8.7; IQR, 5.1 to 13.0 pM/L). Furthermore,
prepubertal and pubertal children did not differ with respect
to gender (p ? 0.502) and SDS-BMI (p ? 0.377).
To our knowledge, this is the first study analyzing the
cross-sectional and longitudinal relationships among amy-
lin, lipids, insulin, and weight status in childhood. We were
able to demonstrate that obese children had significantly
higher amylin levels as compared with lean children. The
amylin levels of the normal weight children were similar to
those of normal weight adults, while the amylin concentra-
tions in the obese children were higher than those of obese
adults (mean amylin, 4.7 pM/L) (1,9). Amylin levels de-
creased significantly in obese children who reduced their
overweight substantially, in contrast to obese children with-
out substantial weight loss in the course of one year. There-
fore, the increase of amylin in obesity seems to be a con-
sequence of obesity.
It is unclear how obesity leads to increased fasting amylin
levels in humans. Because percentage body fat and its
changes were not significantly related to changes of amylin,
a direct link between adipose tissue and amylin levels seems
unlikely. The release of amylin is proportional to the release
of insulin (2,3). In concordance, there was a significant
relationship between insulin and amylin levels at baseline
and their respective changes in the course of one year in our
study. Because insulin resistance is a common feature even
in childhood obesity (11), it could be speculated that the
elevated insulin levels due to insulin resistance are associ-
ated with an increase of amylin concentrations in childhood
obesity. Conversely, clamp studies with infusion of specific
amylin receptor antagonists demonstrated that endogenous
amylin modulates and/or restrains insulin secretion (14).
Hyperamylinemia, which occurs in obesity and impaired
glucose metabolism (12,25,26), may be central to a number
of metabolic abnormalities present in the prediabetic and
diabetic state. Amylin can reduce chylomicron uptake, most
probably by regulating lipoprotein receptors either directly
or via modulation of insulin activity (12). Increased levels
of amylin may contribute to the raised concentration of
triglyceride-rich remnant lipoproteins present in this disease
(12). In concordance, amylin correlated to triglycerides in-
dependently of insulin levels both in cross-sectional and
longitudinal analyses in our study. Because hyperamyline-
mia can lead to hypertriglyceridemia, the value of amylin
analogues as an anti-obesity drug is probably reduced.
This study has a few potential limitations. First, BMI
percentiles and skinfold measurements were used to classify
overweight. Although BMI and skinfold measurements are
a good measure for overweight, one needs to be aware of its
limitations as an indirect measure of fat mass. Second, the
HOMA model is only an assessment of insulin resistance
and clamp studies are the gold standard to analyze insulin
resistance. Because the HOMA model correlated to clamp
studies, it is a suitable method to assess insulin resistance in
field studies (27). Third, we measured amylin levels only in
the fasting state and not postprandially during weight loss.
Because impaired amylin release in response to a meal has
been reported in obese subjects (3), postprandial amylin
levels should be studied in subjects losing weight.
In summary, fasting amylin levels were independent of
age, gender, and pubertal stage. They were related to tri-
glycerides and were increased in obese children. This in-
crease tended to normalize after weight loss. Because amy-
lin concentrations were not significantly related to
Figure 1: Relationship between changes of amylin and insulin in
37 children in the course of one year (r ? 0.54, p ? 0.001) (?,
variable at baseline ? variable one year later).
Figure 2: Relationship between changes of amylin and triglycer-
ides in 37 children in the course of one year (r ? 0.49, p ? 0.001)
(?, variable at baseline ? variable one year later).
Amylin in Obese Children, Reinehr et al.
2010OBESITY Vol. 15 No. 8 August 2007