International Journal of Obesity (1999) 23, 1066±1073
ß 1999 Stockton Press All rights reserved 0307±0565/99 $15.00
http://www.stockton-press.co.uk/ijo
Serum leptin concentration in moderate and
severe obesity: relationship with clinical,
anthropometric and metabolic factors
A Liuzzi1*, G Savia1, M Tagliaferri1, R Lucantoni1, ME Berselli1, ML Petroni2, C De Medici3 and GC Viberti4
1
Division of Endocrinology and Metabolic Diseases, Istituto Auxologico Italiano, Ospedale San Giuseppe IRCCS, Verbania Italy;
Clinical Nutrition Laboratory, Istituto Auxologico Italiano, Ospedale San Giuseppe IRCCS, Verbania Italy; 3Chemical Pathology
Laboratory, Istituto Auxologico Italiano, Ospedale San Giuseppe IRCCS, Verbania Italy; 4Unit for Metabolic Medicine, GKT School of
Medicine, Guy's Hospital, London, UK
2
OBJECTIVE: To study clinical, anthropometric and metabolic determinants of serum leptin concentrations in a series
of patients with a wide range of obesity.
SUBJECTS: 400 patients, 116 males and 284 females, aged 44 12.3 years with body mass index (BMI) ranging from 31
to 82 kg=m2 (mean 41.4 7.1).
MEASUREMENTS: Energy intake by 7-day recall, resting energy expenditure (REE) by indirect calorimetry, body
composition determined by bioelectrical impedance; C index, an anthropometric index of abdominal fat distribution,
and waist ± hip ratio (WHR), blood glucose serum leptin concentrations, total cholesterol, high-density lipoprotein
(HDL) cholesterol, triglycerides, uric acid, and insulin concentrations HOMA IRI (homeostastis model assessment of
insulin resistance index).
RESULTS: Leptin concentrations were higher in obese than in normal subjects and in females than in males without
differences between diabetic and non-diabetic patients; leptin concentrations were not related to age and showed a
strong negative association with energy intake only in the group of women with BMI less than 40. Leptin
concentrations showed a direct correlation with BMI and body fat values (expressed either as percentage of total
body mass or absolute fat mass) independent of age and sex. After adjustment for fat mass, leptin values higher than
predicted were found in women whereas concentrations lower than predicted were found predominantly in men.
Leptin showed an inverse correlation with WHR and C-index, the latter persisting also after correction for gender and
fat mass. REE, but not REE=kg fat-free mass (FFM) was inversely related to leptin also after correction for sex and
absolute fat mass. Leptin concentrations were directly associated with HOMA IRI, insulin and HDL cholesterol and
inversely associated with triglycerides and uric acid. The relationship of leptin with HOMA IRI was still evident after
adjusting for sex but was lost when absolute fat mass was added to the model; HDL cholesterol and triglycerides
appeared to be variables independent of leptin concentrations even when both sex and fat mass were added to the
model.
CONCLUSIONS: In a large group of obese patients (half of whom had severe obesity, gender, BMI and fat mass
accounted for the largest proportion of serum leptin concentrations variability. We found that in obese subjects there
is an effect of fat distribution on leptin concentrations and that, after excluding variability due to absolute fat mass,
patients with a greater amount of abdominal fat have relatively low leptin concentrations which in turn relates to a
metabolic pro®le compatible with an increased cardiovascular risk. Women with milder obesity may retain some
degree of control of food intake by leptin.
Keywords: leptin; obesity; fat distribution; cardiovascular risk factors
Introduction
After the discovery of leptin in 19941 efforts have
been made to elucidate the pathophysiological role
of this peptide in experimental animals and in
humans;2 ± 4 the availability of radio immunoassay
(RIA) methods for this peptide hormone produced
by adipose tissue have led to an impressive accumulation of clinical studies on the relationships between
*Correspondence: A Liuzzi, Division of Endocrinology and
Metabolic Diseases, Istituto Auxologico Italiano, Ospedale San
Giuseppe IRCCS, P.O. Box 1, 28044 Verbania, Italy.
Received 20 October 1998; revised 23 February 1999; accepted
26 April 1999
leptin and anthropometric, metabolic and hormonal
parameters in physiological and pathological conditions. Serum concentrations are elevated in human
obesity suggesting a condition of resistance to this
peptide, which would normally be associated with
reduced food intake.5 There is general agreement
that serum leptin concentrations correlate with body
fat mass; however, this correlation accounts only
partially for the variability in serum leptin concentrations.6 ± 8 Attempts have been made to gain insight into
the determinants of leptin concentrations,9±12 and
thus, into the pathophysiological meaning of leptin
in human obesity. Results have been partially con¯icting with some of these inconsistencies possibly
due to different ranges of patients' body mass index
(BMI) and gender ratio in the different series. The aim
Leptin in obesity
A Liuzzi et al
of the present study was to re-evaluate, in a very large
series of patients with severe obesity, several anthropometric and metabolic factors related with serum
leptin concentrations and to test the hypothesis that
leptin concentrations may be associated with factors
other than the absolute amount of adipose tissue and
gender, such as adipose tissue distribution (central vs
peripheral), energy expenditure and possibly with
metabolic risk factors for cardiovascular disease.
Patients and methods
Patients
We studied 400 obese patients (Table 1), 116 males
and 284 females, aged 44 12.3 y (mean s.d.; range
14 ± 84 y), with BMI ranging from 31±82 kg=m2
(mean 41.4 7.1) out of a larger series of patients
admitted to the Division of Endocrinology and Metabolic Diseases of S. Giuseppe Hospital, Verbania,
Italy, for diagnostic or therapeutic problems related
to obesity and its complications. Patients with concomitant severe renal, hepatic or cardiac disease were
excluded from the study; patients with ¯uid overload
according to vectorial analysis13 were also excluded to
minimise errors of bioelectrical impedance analysis
(BIA) in estimating fat and fat free-mass in severe
obesity. The study was approved by the Hospital
Ethics Committee; the aim and the design of the
study were explained to the patients who gave their
written informed consent. Eighty-six patients had noninsulin-dependent diabetes mellitus NIDDM, which
was well controlled (glycated haemoglobin
6.4 1.6%) by diet alone or combined with oral
hypoglycaemic agents; 10 of these patients were on
lipid lowering therapy with either ®brates (n ˆ 6) or
statins (n ˆ 4); none was taking beta blockers whereas
eight patients were on hydrochlorothiazide 12.5 mg=d.
None, except the diabetic patients, was taking drugs
interfering with glucose or lipid metabolism. All the
patients had stable weight during the month prior to
the study. To estimate their habitual caloric intake, all
patients underwent a careful 7-day dietary recall
carried out by trained dieticians according to a standardised interview technique (including the use of a
food-frequency questionnaire and of an atlas for
assessment of food-portion size and of snacks). As a
proxy for evaluating dietary under-reporting, we compared the estimated caloric intake with the measured
resting energy expenditure increased by a factor
allowing for sedentary physical activity. We assumed
that a negative difference between reported energy
intake minus resting energy expenditure multiplied by
1.2 (as a cautious estimate of total energy expenditure) would express under-reporting.
The patients underwent a diagnostic protocol which
included measurement of height, weight, abdominal
and hip girth for determination of body mass index
(BMI), waist ± hip ratio (WHR) and C-index14 as well
as evaluation of resting energy expenditure (REE) and
body composition. Blood chemistry included total
cholesterol, high-density lipoprotein (HDL) cholesterol, triglycerides, uric acid, fasting blood glucose
and insulin. Leptin concentrations were determined
on a blood sample taken at 8.00 a.m. after an overnight
fast.
Methods
Resting Energy expenditure (REE) was assessed by a
computerised, open-circuit, indirect calorimetry
system that measured resting oxygen uptake and
resting carbon dioxide production using a ventilated
Table 1 Values (mean s.d.) of the demographic, anthropometric, hormonal and metabolic
parameters in 400 obese patients reported as whole group and divided into men and women
Variable
n
Age
Body weight (kg)
BMI (kg=m2)
WHR
C-index
FAT mass (%)
FAT mass (kg)
Leptin (ng=mL)
Fasting glucose (mmol=L)
Fasting Insulin (mU=mL)
HOMA IRI
Uric Acid (mmol=L)
Total Cholesterol (mmol=L)
HDL Cholesterol (mmol=L)
Trygliceride (mmol=L)
Caloric intake (kcal=die)
REE (kcal=24 h)
REE=kgFFM (kcal=24 h)
All
400
Men
116
Women
284
46.7 13.8
109.8 21.2
41.4 7.1
0.9 0.1
1.5 0.08
43.7 5.5
48.3 13.4
43.3 24.5
5.6 1.8
13.4 7.7
3.44 2.86
384.8 131.6
5.5 1.1
1.2 0.4
1.8 0.9
2821.0 1335
1841.0 338
30.4 3.9
46.3 11.9
122.7 20.9
40.3 5.9
0.99 0.1
1.63 0.06
41.8 9.2
52.2 17.8
24.7 13.3
5.7 1.8
14.2 9.9
3.8 4.0
439.7 108.8
5.4 1.2
1.0 0.2
5.12 2.6
3613.0 1553
2218.0 318
32.4 5.4
46.9 14.5
104.6 19.1**
41.9 7.5*
0.86 0.06
1.51 0.07***
44.5 2.8
46.9 11***
51.0 23.7***
5.5 1.7
13.1 6.6
3.28 2.1
360.9 134.1***
5.5 1.2
1.3 0.5***
3.8 1.8***
2605.0 1190***
1726.0 250***
29.9 3.1***
*P < 0.05; **P < 0.01; ***P < 0.001; men vs women.
1067
Leptin in obesity
A Liuzzi et al
1068
canopy (Sensormedics, Milano, Italy). REE was measured at 08.00 h after an overnight fast, in a comfortable and thermoregulated (22 ± 24 C) room where
only the investigator and the patient were present.
After a 10-min period of steady-state, values were
recorded each minute for 30 min; the mean value was
then expressed as kcal=24 h.
Body composition, expressed as percentage of total
body mass and as absolute (kilograms) fat free mass
(FFM) and Fat Mass (FAT), was determined by BIA
(101=S, Akern, Firenze, Italy), in the morning, after
an overnight fast and after voiding. The two vector
components resistance (R) and reactance (Xc) were
recorded from single measurements: before each testing session, the external calibration of the instrument
was checked with a calibration circuit of known
impedance value. The mean coef®cient of variation
was 1% for within-day and 3% for weekly intraindividual measurements in the steady state condition
in either site and 2% for inter-operator variability.
Total body water and FFM of subjects were derived
from the equations by Lukaski et al.15,16 Body fat
distribution was measured by WHR and by C-index,
this latter being an anthropometric parameter believed
to be a reliable index for the assessment of central fat
distribution.14 The asserted advantages of this index vs
WHR are the following: (i) it has a theoretical
(expected) range, (ii) it includes a built-in adjustment
of waist circumference for height and weight, allowing direct comparisons of abdominal adiposity
between individuals or even between populations,
(iii) it does not require the hip circumference to
assess fat distribution. It is calculated according
to
p the formula C-index ˆ abdominal girth=0.109
weight=height.
The waist circumference was taken as the largest
standing horizontal circumference between the ribs
and the iliac crest; the hip circumference was taken as
the largest standing horizontal circumference of the
buttocks. Glucose, total cholesterol, HDL cholesterol,
triglycerides and uric acid were measured by enzymatic methods; insulin was determined by immunoenzymatic method (Tosoh, Kyobashi Chuo-Ku, Tokio,
Japan).
Serum leptin concentrations were measured by
radioimmunoassay using reagents supplied by Linco
Research Inc (St Louis, MO USA). In this assay,
detection limit is 0.15 ng=ml; the intra assay precision
(% CV) is 2.2% (6 ng=ml), 2.7% (25 ng=ml), and
5.9% (62.8 ng=ml); inter assay precision from 10
different runs of 3 patients serum samples was
4.3%, 4%, 6.9% at the concentration of 5.1, 21 and
56.2 ng=ml respectively; in 32 lean subjects (BMI
18 ± 25 kg=m2) reference limits (2.5 ± 97.5 percentiles) were 0.98 ± 5.17 ng=ml in men and 2.6 ±
17.4 ng=ml in women.
Index of insulin resistance was calculated by the
HOMA (Homeostasis Model Assessment) using fasting plasma insulin and blood glucose according to
Matthews et al. 19
Statistical analysis
Since leptin concentrations were not normally distributed, data were log transformed. The relationships
between leptin and the other parameters were assessed
by univariate regression analysis. To determine the
independent effects of several variables, data were
also analysed in stepwise multivariate analysis models
with leptin as dependent variable. SPSS statistical
software release 8.0 was used for statistical analysis
whereas the regression lines were ®tted used the
graphic Prism package release 2.01. Differences
were accepted as signi®cant at P < 0.05.
For simplicity data are presented as mean s.d. also
for not normally distributed variables.
Results
In the whole group (Table 1), mean basal leptin
concentration was 43.3 24.5 ng=ml (range 6.2 ±
169.4 ng=ml). Women had signi®cantly higher
(P < 0.001) mean plasma leptin concentrations
(51.1 23.7 ng=ml) than men (24.7 13.3 ng=ml)
even after adjustment for kg of FAT; in either sex
there were no age-related changes of leptin concentrations; in particular there were no differences in
leptin concentrations between women of fertile (50.8
24.1 ng=ml) or postmenopausal (51.1 23.5 ng=ml)
age. Leptin values were no different in diabetic
patients (39.13 23.18 ng=ml) vs non-diabetic
patients (43.6 22.5 ng=ml). Mean energy intake
was 2821 1335 kcal=d in the whole group of
patients; it was not related to leptin concentrations
in the whole group of patients, but a negative
( 7 0.188; P < 0.002) relation was evident when the
data were analysed in the subgroup of patients with
BMI below 40. This association persisted after adjusting for fat mass and fat distribution but not for gender;
indeed only women showed the negative correlation
between leptin and energy intake. There was no
signi®cant difference in proportion of dietary underreporting between women and men (33% vs 27%) or
between women with BMI < 40 kg=m2 and those with
BMI 40 kg=m2 (36% vs 31%).
Table 2 reports the univariate coef®cients of correlation before and after adjusting for gender and FAT
between serum leptin and the other parameters
considered. The correlation between leptin and BMI,
although very strong, also when patients were subgrouped according to gender, showed a very wide
intersubject variability, similar leptin concentrations
corresponding to markedly different BMI values
(Figure 1); for instance, patients with leptin values
of 100 ng=ml or higher could have a BMI ranging
between 35 and 82.
Leptin showed a strong association with FAT,
expressed either as percentage of total body composition or as absolute FAT. This relationship persisted
Leptin in obesity
A Liuzzi et al
Table 2 Coef®cients of correlation ( s.e.) between leptin and anthropometric, metabolic and
laboratory parameters in 400 obese subjects.
Age (yr)
BMI (kg=m2)
WHR
C-Index
Fasting glucose (mmol=l)
Fasting insulin (mU=l)
Homa IRI
Cholesterol
Cholesterol HDL (mg=dL)
Triglycerides (mg=dL)
Acid Uric (mg=dL)
Caloric Intake (Kcal=24 h)
REE (Kcal=24 h)
REE=kgFFM
Fat (%)
Fat (Kg)
A
Coefficients
P
7 0.08 0.08
1.6 0.16
7 032 0.11
7 6.2 0.41
7 0.18 0.10
0.24 0.05
7 0.13 0.05
7 0.11 0.13
0.54 0.10
7 0.35 0.06
7 0.26 0.09
7 0.14 0.07
7 0.57 0.19
7 0.61 0.27
2.30 0.19
0.82 0.10
ns
0.0001
0.001
0.00001
ns
0.0001
0.001
ns
0.00001
0.00001
0.005
ns
0.003
0.01
0.00001
0.00001
B
Coefficients
P
7 0.02 0.07
0.55 0.22
7 0.06 0.07
7 1.7 0.49
7 0.11 0.07
0.12 0.04
0.05 0.03
7 0.07 0.09
0.22 0.08
7 0.11 0.04
0.002 0.07
7 0.07 0.05
7 0.69 0.21
7 0.17 0.19
ns
0.01
ns
0.0001
ns
0.001
0.001
ns
0.05
0.01
ns
ns
0.001
ns
Column A: Univariate correlation coef®cients; column B: coef®cients from the multiple
regression adjusted for gender and fat mass (kg).
also after adjusting for age indicating that FAT is an
independent determinant of leptin concentrations; the
relationship between leptin and FAT was best
described by an exponential function (Figure 2).
When leptin values were adjusted for absolute fat
mass and the actual and the corrected values were
compared, leptin concentrations higher than expected
(that is, two standard deviations of the mean of the
differences) were found only in females, while lower
than expected leptin values were found predominantly
in males (Table 3).
Collectively, gender, BMI and percentage
explained 55% of variability in leptin concentration.
The relation of leptin to fat distribution assessed by
WHR or C-index (Figure 3) showed an inverse association with both indices (r2 ˆ 7 0.02; P < 0.004 and
r2 ˆ 7 0.33 < ; P < 0.001 respectively); however, the
relation with WHR, but not with C-index
(r2 ˆ 7 0.55; P < 0.001), disappeared when the
values were adjusted for gender and body fat mass;
thus only C-index seems to be independently correlated with leptin concentrations.
REE, even when expressed for kg of FFM, correlated inversely with leptin values. REE, but not
REE=kg FFM, maintained its independent association
with leptin concentrations also after adjusting for sex
and kg of FAT.
Leptin concentrations were also directly associated
with homeostasis model of insulin resistance index
(HOMA IRI), fasting serum insulin, and HDL cholesterol and inversely associated with triglycerides and
uric acid. The relationship of leptin with HOMA IRI
was still apparent after correction for sex but was lost
when kg of FAT was added to the model; HDL
Figure 1 Simple regression between unadjusted serum leptin
and body mass index (BMI) in 400 obese patients.
Figure 2 Simple regression between unadjusted serum leptin
and percentage fat mass measured by bioelectrical impedance.
1069
Leptin in obesity
A Liuzzi et al
1070
Figure 3 Simple regression between unadjusted serum leptin and C-index in 400 obese patients.
cholesterol and triglycerides appeared to be positive
and inverse independent correlates of leptin concentrations respectively even when both sex and FAT
were added to the model.
The results of multivariate analyses did not change
when the presence or not of diabetes was introduced
into the model and when patients on lipid lowering
drugs or hydrochlorothiazide were excluded from the
analysis.
Discussion
The results of our study, obtained in a large series of
patients including only markedly obese subjects, half
of whom had BMI > 40 kg=m2, add valuable information to the issue of determinants of circulating
leptin in relation to clinical, anthropometric and
metabolic parameters.
Table 3 Sex distribution between relatively low, normal and
high leptin patients in relation with sex. The patients are
allocated to the three groups on the basis of the difference
between their actual leptin values and those expected after
correction for fat mass
Males
Females
Relatively high
Relatively normal
Relatively low
0 (0%)
74 (26%)
33 (28.5%)
199 (70%)
83 (71.5%)
11 (4%)
Our data are in agreement with previous studies,
performed on more restricted series including
patients with only minor degrees of adiposity, concerning the ®nding of higher circulating concentrations of leptin in obese compared to normal subjects
and in females6,8,11,12 than in males; it is of interest
that the gender difference remains true also for very
high levels of BMI, that is, between 50 and 80.
Moreover, in both sexes of this group of exclusively
obese patients with a wide age range, we could not
con®rm age-related changes in leptin concentrations,
as described in miscellaneous conditions including
both overweight and normal weight subjects.6,18 In
particular, women of fertile or postmenopausal age
had similar leptin values. Thus, excess adiposity may
prevent the decline in plasma leptin due to aging and
estrogen de®ciency reported in non-obese subjects.
Leptin concentrations showed a high correlation
with BMI and with FAT either expressed as percentage or as absolute FAT con®rming that the degree of
adiposity is a major determinant of leptin concentration.6 ± 8 However, sex, BMI and FAT together
explained only 55% of serum leptin variability,
con®rming that factors other than the absolute
degree of adiposity are involved in controlling leptin
production.
The observation that the relationship between leptin
and percentage of body fat is best described by an
exponential function, in accordance with the data by
Ostlund et al, 6 indicates that above a certain threshold
Leptin in obesity
A Liuzzi et al
of adiposity, small increments of fatness may be
accompanied by signi®cant increments of serum
leptin. These ®ndings suggest that in situation of
severe adiposity, fat cells secrete disproportionately
amounts of leptin, which (as yet unidenti®ed) feedback mechanisms fail to respond to. A reduced clearance of leptin seems unlikely as a cause of the
elevated plasma concentrations since the metabolic
clearance rate of leptin has been found to be normal in
patients with obesity.19
More insight into this problem was gained when
leptin concentrations were corrected for absolute FAT
and actual and adjusted values were compared. Most
of the patients showed leptin concentrations in accordance with their theoretical values, but about 30% had
values lower or higher than those predicted by their
adiposity (Table 3). In particular, leptin concentrations lower than expected were more often found in
male subjects. A possible explanation for this observation may be the heterogeneity in leptin synthesis by
adipose tissue as reported by Cinti et al. 20 Levels of
leptin mRNA are lower in omental than in subcutaneous fat tissue21,22 and, more recently, van Harmelen
et al 23 have provided evidence that subcutaneous
tissue is two to three times more effective in synthesising leptin than omental tissue. Thus, patients with a
relatively low plasma leptin concentration should be
characterised by a higher visceral to subcutaneous fat
ratio than those with relatively normal or high leptin
values. The gender-related difference in plasma leptin,
which cannot be fully accounted for by differences in
the hormonal milieu, might indeed be explained by
the presence of more abundant subcutaneous fat in
women and of visceral fat in men. The relationship
between circulating concentrations of leptin and fat
distribution remains controversial:24 ± 27 our data
showed an inverse correlation between abdominal
fat distribution, measured as waist=hip ratio or Conicity-index and leptin, the latter association being
independent of gender and a body fat. Conicityindex appears to have several theoretical advantages
as a body fat distribution indicator compared to
WHR;16 thus, the prevalent abdominal fat distribution
is a negative determinant of leptin concentrations
independent of sex, a ®nding that suggests that a
greater amount of visceral fat is associated with a
relatively lower leptin value.
Exogenous leptin is known to increase thermogenesis28 in rat brown adipose tissue. Con¯icting
results have however been reported on the relationship
between serum leptin concentrations and energy
expenditure in humans, probably due to the heterogeneity of the patients studied; some studies reported
a positive correlation between serum leptin and resting energy expenditure or 24-h energy expenditure29
while others failed to show such a relation:30,31
Roberts et al 32 concluded that in normal weight
adult humans, leptin does not in¯uence energy regulation through adaptive variations in energy expenditure. Rosenbaum et al 33 found, in a mixed series of
31 normal weight and 19 obese subjects, a trend
towards a positive association between resting
energy expenditure and leptin concentrations which
disappeared when FAT and FFM were taken into
account. Niskanen et al 34 found in a series of 45
obese subjects an inverse association Ð after correction for fat mass, age and sex Ð between leptin concentrations and resting energy expenditure, respiratory
quotient and carbohydrate oxidation rate. In our large
series of markedly obese subjects, leptin concentrations were independently related with resting energy
expenditure in an inverse fashion, in accordance with
the view that obesity is a condition of resistance to
leptin. This correlation however was lost when the
values of REE were expressed for kilogram of FFM
suggesting that the resting thermogenesis of lean
tissue is independent of leptin at least in obese
subjects.
The ®nding of a negative correlation between leptin
and energy intake, similar to that reported in healthy
postmenopausal women,35 only in females with BMI
lower than 40, is unlikely to be accounted for by a
lower reported caloric intake36 ± 38 in our series of
obese patients; there was no difference in the proportion of estimated dietary under-reporting in women
with different degrees of obesity. A more convincing
explanation could be the persistence of some feedback regulation of energy intake by circulating leptin
in this group of less severe obese patients, possibly
re¯ecting different degrees of leptin effect at the CNS
level.
Con¯icting data also exist concerning the relationship between leptin, insulin and insulin sensitivity; our
data showing that insulin is an independent correlate
of leptin support the view that insulin has a stimulatory effect on leptin secretion,39 ± 42 and is compatible
with the demonstration that leptin has a facilitatory
effect on insulin release.43,44
On the other hand, HOMA IRI, an index of insulin
resistance, was no longer related to serum leptin after
adjusting for FAT which in¯uences both leptin concentrations and insulin resistance; this seems to rule
out a direct effect of leptin on plasma insulin; in other
words the amount of FAT is the link between the
elevated concentrations of leptin and the resistance to
insulin in obesity.
Whether high leptin concentration is a facet of the
metabolic syndrome and whether its measurement
may be of value in identifying patients at risk of
developing cardiovascular disease should also be
considered. The data of our series show that plasma
leptin is associated with serum HDL cholesterol, and
inversely related to triglycerides, and uric acid. However, in a multivariate analysis, the only independent
contributors to serum leptin were triglycerides and
HDL, whereas uric acid was sex related and HOMA
IRI was dependent on fat. These data suggest that,
after accounting for absolute FAT mass, the lower the
leptin concentration the higher the markers of cardiovascular risk; similar conclusions were reached by
1071
Leptin in obesity
A Liuzzi et al
1072
Lonnqvist et al: 9 they stated that although leptin
appeared not to be associated with the classical metabolic atherogenic complications of obesity, it might
protect subjects with a gynoid fat distribution from
metabolic complications.
In conclusion our study shows that in a large group
of obese patients, half of whom had severe obesity,
gender, BMI and FAT account for the largest proportion of variability of serum leptin concentrations. Our
observation supports the view that in obese subjects
there is an effect of fat distribution, as measured by
Conicity index, on leptin concentrations and that, after
excluding variability due to absolute FAT, patients
with a greater amount of abdominal fat, as indicated
by a higher Conicity index, have relatively low leptin
values which in turn relate to a metabolic pro®le
compatible with an increased cardiovascular risk.
Lastly, the ®nding of an inverse relationship between
serum leptin and energy intake only in women with
milder obesity, even after correction for adiposity, is
suggestive of the persistence of some degree of control of food intake by leptin in moderate human
obesity.
References
1 Zhang Y, Proenca R, Maffei M, Barone M, Leopoid L,
Friedman JM. Positional cloning of the mouse obese gene
and its human homologue. Nature 1994; 372: 425 ± 432.
2 Bray GA, York DA. Leptin and clinical medicine Ð a new
piece in the puzzle of obesity. J Clin Endocrinol Metab 1997;
82: 2771 ± 2776.
3 Rosenbaum M, Leibel R, Hirsch J. Obesity. N Engl J Med
1997; 337: 396 ± 407.
4 Auwerx J, Staels B. Leptin. Lancet 1998; 351: 737 ± 742.
5 Camp®eld LA, Smith JF, Burn P. OB protein: a hormonal
controller of central neural network mediating behavioral,
metabolic and neuroendocrine responses. Endocrinol Metab
1997; 4: 81 ± 102.
6 Ostlund RE, Yang JW, Klein S, Gingerich R. Relation
between plasma leptin concentration and body fat, gender,
diet, age, and metabolic covariates. J Clin Endocrinol Metab
1996; 81: 3909 ± 3913.
7 Havel PJ, Kasim-Karakas S, Mueller W, Johnson PR, Gingerich RL, Stern JS. Relationship of plasma leptin to plasma
insulin and adiposity in normal weight and overweight
women: effect of dietary fat content and sustained weight
loss. J Clin Endocrinol Metab 1996; 81: 4406 ± 4413.
8 Sumner AE, Falkner B, Kushner H, Considine RV. Relationship of leptin concentrations to gender, menopause, age,
diabetes, and fat mass in African Americans. Obes Res
1998; 6: 128 ± 133.
9 Lonnqvist F, Wennlund A, Arner P. Relationship between
circulating leptin and peripheral fat distribution in obese
subjects. Int J Obes 1997; 21: 255 ± 260.
10 Chapman IM, Wittert GA, Norman RJ. Circulating leptin
concentrations in polycystic ovary syndrome-relation to
anthropometric and metabolic parameters. Clin Endocrinol
1997; 46: 175 ± 181.
11 Saad MF, Damani S, Gingerich R, Riad-Gabriel MG, Khan A,
Boyadjian R, Jinagouda SD, el-Tawil K, Rude RK, Kamdar V.
Sexual dimorphism in plasma leptin concentration. J Clin
Endocrinol Metab 1997; 82: 579 ± 584.
12 Niskanen LK, Haffner SM, Karhunen LJ, Turpeinen AK,
Miettinen H, Uusitupa MIJ. Serum leptin in obesity is related
to gender and body fat topography but does not predict
successful weight loss. Eur J Endocrinol 1997; 137: 61 ± 67.
13 Piccoli A, Brunani A, Savia G, Pillon L, Favaro E, Berselli
ME, Cavagnini F. Discriminating between body fat and ¯uid
changes in the obese adult using bioimpedance vector analysis.
Int J Obes 1998; 22: 97 ± 104.
14 Lukaski HC, Bolonchuck WW. Estimation of body ¯uid
volumes using tetrapolar bioelectrical impedance measurements. Aviat Space Environ Med 1988; 59: 1163 ± 1169.
15 Lukaski HC, Bolonchuck WW, Hall CB, Siders WA. Validation of tetrapolar bioelectrical impedance method to assess
human body composition. J Appl Physiol 1986; 60: 1327 ±
1332.
16 Valdez R, Seidell J, Ahn Yi, Weiss KM. A new index
of abdominal adiposity as indicator of risk for cardiovascular disease. A cross population study. Int J Obes 1993;
17: 77 ± 82.
17 Matthews D, Hosher J, Rudenski A, Naylor BA, Treacher DF,
Turner RC. Homeostasis model assessment. Insulin resistance
and beta cell function from fasting plasma and insulin concentration in man. Diabetologia 1985; 28: 412 ± 419.
18 Rosenbaum M, Nicolson M, Hirsch J, Heyns®eld SB, Gallagher D, Chu F, Leibel RL. Effects of gender, body composition, and menopause on plasma concentrations of leptin. J Clin
Endocrinol Metab 1996; 81: 324 ± 327.
19 Klein S, Coppack SW, Mohamed-Ali V, Landt M. Adipose
tissue leptin production and plasma leptin kinetics in humans.
Diabetes 1996; 45: 984 ± 987.
20 Cinti S, Frederich RC, Zingaretti MC, De Matteis R, Flier JS,
Lowell BB. Immunohistochemical localization of leptin and
uncoupling protein in white and brown adipose tissue. Endocrinology 1997; 138: 797 ± 804.
21 Montague CT, Prins JB, Sanders L, Digby JE, O'Rahilly S.
Depot- and sex-speci®c differences in human leptin mRNA
expression: implications for the control of regional fat distribution. Diabetes 1997; 46: 342 ± 347.
22 Lefebvre AM, Laville M, Vega, N Riou JP, van Gaal L,
Auvwerx J, Vidal H. Depot-speci®c differences in adipose
tissue gene expression in lean and obese subjects. Diabetes
1998; 47: 98 ± 103.
23 Van Harmelen V, Reynisdottir S, Eriksson P, Thorne A,
Hoffstedt E, Lonnqvist F, Amer P. Leptin secretion from
subcutaneous and visceral adipose tissue in women. Diabetes
1998; 47: 913 ± 917.
24 Takahashi M, Funahashi T, Shimomura I, Miyaoka K, Matsuzawa Y. Plasma leptin concentrations and body fat distribution. Horm Metab Res 1996; 28: 751 ± 752.
25 Haffner SM, Gingerich R, Miettinen H, Stern MP. Leptin
concentration in relation to adiposity and regional adiposity.
Int J Obes 1996; 20: 904 ± 908.
26 Bennett FI, McFarlane-Anderson N, Wilks R, Luke A, Cooper
RS, Forrester TE. Leptin concentration in women is in¯uenced
by regional distribution of adipose tissue. Am J Clin Nutr
1997; 66: 1340 ± 1344.
27 Taylor RW, Goulding A. Plasma leptin in relation to regional
body fat in older New Zealand women. Au N Z J Med 1998;
28: 316 ± 321.
28 Scarpace PJ, Matheny M, Pollock BH, Tumer N. Leptin
increases uncoupling protein expression and energy expenditure. Am J Physiol 1997; 36: E226 ± E230.
29 Toth MJ, Gottlieb S, Fisher ML, Ryan AS, Nicklas BJ,
Poehlman ET. Plasma leptin concentrations and energy expenditure in heart failure patients. Metabolism 1997; 46: 450 ±
453.
30 Farooqui IS, Jebb S, Cook G, Cheetham CH, Lawrence E,
Prentice A, Hughes IA, McCamish M, O'Rahilly S. Treatment
of congenital leptin de®ciency in man. Presentation at the 8th
International Congress on Obesity. Paris, 1998.
31 Montague CT, Farooqi IS, Whitehead JP, Soos MA, Rau H,
Wareham NJ, Sewter CP, Digby JE, Mohammed SN, Hurst
JA, Cheetham CH, Earley AR, Barnett AH, Prins JB,
O'Rahilly S. Congenital leptin de®ciency is associated
with severe early-onset obesity in humans. Nature 1997;
387: 903 ± 908.
Leptin in obesity
A Liuzzi et al
32 Roberts SB, Nichoison M, Staten M, Dallal GE, Sawaya AL,
Heyman MB, Fuss P, Greenberg AS. Relationship between
circulating leptin and energy expenditure in adult men and
women aged 18 years to 81 years. Obes Res 1997; 5:
459 ± 463.
33 Rosenbaum M, Nicolson M, Hirsch J, Murphy E, Chu F,
Leibel RL. Effects of weight change on plasma leptin concentrations and energy expenditure. J Clin Endocrinol Metab
1997; 82: 3647 ± 3654.
34 Niskanen L, Haffner S, Karhunen LJ, Turpeinen AK, Miettinen H, Uusitupa MIJ. Serum leptin in relation to resting energy
expenditure and fuel metabolism in obese subjects. Int J Obes
1997: 21: 309 ± 313.
35 Larsson H, Elmstahl S, Berglund G, Ahren B. Evidence for
leptin regulation of food intake in humans. J Clin Endocrinol
Metab 1998; 83: 4382 ± 4385.
36 Black AE, Goldberg GR, Jebb SA, Livingstone MB, Cole TJ,
Prentice AM. Critical evaluation of energy intake data using
fundamental principles of energy physiology: 2. Evaluating
the results of published surveys. Eur J Clin Nutr 1991; 45:
583 ± 589.
37 Lafay L, Basdevant A, Charles MA, Vray M, Balkau B, Borys
JM, Eschwege E, Romon M. Determinants and nature of
dietary unserreporting in a free living population: the Fleurbaix Laventie Villesante (FLVS) study. Int J Obes 1997; 21:
567 ± 573.
38 Breifel RR, Sempos CT, McDowell MA, Chien S, Alaimo K.
Dietary methods research in the third National Health and
Nutrition Examination Survey: underreporting of energy
intake. Am J Clin Nutr 1997; 65: 1203S ± 1209S.
39 Kim-Motoyama H, Yamaguchi T, Katakura T, Myura M,
Ohashi Y, Yazaki I, Kadawaki T. Serum leptin levels are
associated with hyperinsulinemia independent of body mass
index but not with visceral obesity. Biochem Biophys Res
Commun 1997; 239: 340 ± 344.
40 Popovic V, Micic D, Danjanovic S, Zoric S, Djurovic M,
Obradovic S, Petakov M, Dieguez C, Casanueva F. Serum
leptin and insulin concentrations in patients with insulinoma
before and after surgery. Eur J Endocrinol 1997; 138: 86 ± 88.
41 Havel PJ, Uriu-Hare JY, Liu T, Stanhope KL, Stern JS, Keen
CL, Ahren B. Marked and rapid decrease of circulating leptin
in streptozotocin diabetic rats: reversal by insulin. Am J
Physiol 1998; 274: R1482 ± R1491.
42 Patel BK, Koenig JK, Kaplan LM, Hooi SC. Increase in
plasma leptin and Lep mRNA concentrations by food intake
is dependent on insulin. Metabolism 1998; 47: 603 ± 607.
43 Sivitz WI, Walsh SA, Morgan DA, Thomas MJ, Haynes WG.
Effects of leptin on insulin sensitivity in normal rats. Endocrinology 1997; 138: 3395 ± 3401.
44 Barzilai N, Wang J, Massilon D, Vuguin P, Hawkins M,
Rossetti L. Leptin selectively decreases visceral adiposity and
enhances insulin action. J Clin Invest 1997; 100: 3105 ± 3110.
1073