International Journal of Obesity (1997) 21, 476±483
ß 1997 Stockton Press All rights reserved 0307±0565/97 $12.00
Impact of waist-hip-ratio and body-mass-index
on hormonal and metabolic parameters in
young, obese women
M Hollmann, B Runnebaum and I Gerhard
Department of Endocrinology and Reproduction, University Hospital of Obstetrics and Gynaecology, Voûstr. 9, Heidelberg, Germany
BACKGROUND: To investigate the impact of predominantly upper body fat localisation on the hormonal and
metabolic pro®le in obese, infertile women.
DESIGN: Prospective observational study of premenopausal women with obesity, infertility and menstrual dysfunction.
SETTING: Department of Endocrinology and Reproduction of the University Hospital of Obstetrics and Gynaecology of
Heidelberg.
SUBJECTS: Eighteen women with android type obesity (waist-hip-ratio ˆ WHR > 0.85, group 1) and 22 women with
gynoid type obesity (WHR 0.85, group 2) in a group of 58 premenopausal obese women (median age 28 y) with
infertility. Twenty-nine women took part in a weight reducing program lasting 32 14 (mean s.d.) weeks.
MEASUREMENTS: BMI, WHR and blood pressure. Plasma lipids and liver enzymes. Blood glucose, insulin, C-peptide
and different steroid and pituitary hormones during oral glucose loading.
RESULTS: In the total group of 58 obese women, WHR was directly correlated to plasma triglycerides, AST, ALT and
cholesterol/HDL-cholesterol-ratio. WHR correlated inversely with HDL-cholesterol. Insulin resistance was greater with
increasing WHR. Systolic blood pressure, glucose, insulin, C-peptide, triglycerides, cholesterol/HDL-cholesterol-ratio,
aspartate (AST) and alanine aminotransferase (ALT) were signi®cantly greater in group 1. Group 2 had greater HDLcholesterol levels. One subject in group 1, ®ve women in group 2 conceived spontaneously after weight reduction.
CONCLUSIONS: Determination of the WHR is a simple measurement to identify obese patients who are at a greater
risk of developing the metabolic syndrome. WHR is important in preventive medicine, as typical metabolic pro®les are
already present in young women before clinical manifestation. Women with android obesity seem to be more prone
to develop menstrual irregularity and infertility. The hyperinsulinaemia may be the pathway.
Keywords: insulin resistance; menstrual irregularity; obesity; waist-hip-ratio
Introduction
Obesity is a common health problem in western
populations and associated with an increase in mortality and morbidity.1±6 There is, for example, an
increased prevalence of diabetes, cardiovascular disease, hyperlipidemia and certain tumours, for example
rectal carcinoma.1,4±11 At least 30% of the population
of former West Germany is overweight.2,3 Body
weight and body mass index (BMI) are commonly
used to diagnose obesity in clinical practice.1 The
body fat distribution, simply measured by the waist to
hip girth, is supposedly a more important factor in
predicting the risk of related diseases than the grade of
obesity as determined by the BMI.1,3,6,12±17
Obesity is also considered a risk factor in the ®eld
of obstetrics and gynaecology.18±22 Obese women are
more likely to suffer from infertility and menstrual
Correspondence: Prof. Dr I Gerhard, UniversitaÈts-Frauenklinik,
Voûstr. 9, 69115 Heidelberg, Germany.
Received 6 June 1996; revised 27 November 1996; accepted 26
February 1997
dysfunction and respond less well to the usual hormonal treatment.23±26 If pregnancy occurs, the rate of
complications is higher during its course as well as
during delivery.27 There is also an increased prevalence of certain gynaecological tumours, for example
endometrial or breast cancer, in obese women.4,8,9,28
We investigated two subgroups of premenopausal,
obese women to evaluate whether the site of body fat
predominance does predict a speci®c metabolic and
hormonal pro®le already in young women. Furthermore, the effects of weight reduction on menstrual
irregularity and infertility in these subgroups were
evaluated. We previously found a pregnancy rate of
29% and an improvement of menstrual dysfunction in
80% in obese, infertile women after weight reduction.29
Materials and methods
Subjects
Fifty-eight obese, premenopausal women (BMI range
26.9±49.9 kg/m2) with menstrual irregularity, aged
WHR and BMI in young obese women
M Hollmann et al
17±41 y, were selected for this study between November 1988 and February 1990. All women were referred
to our department because of infertility of 6.5 5.0 y
(mean s.d.) duration. Inclusion criteria were: obesity, menstrual irregularity, normal thyroid function
and normal prolactin levels (< 20 ng/ml). Obesity was
de®ned as body mass index (BMI) greater than 25
kg/m2.3 BMI was calculated using the formula
BMI ˆ body weight (kg)/height2 (m2).1,3 Abnormal
length of menstrual cycles, chronic anovulation or
luteal insuf®ciency (progesterone < 10 ng/ml) was
considered as menstrual irregularity.30 Thirty-six
ovarian ultrasounds proved multicystic ovaries in six
subjects. Only one woman showed all criteria of the
polycystic ovary syndrome (PCO-syndrome: hyperandrogenism, polycystic ovaries, abnormal LH-FSHratio > 2.0).31 Body weight and height were determined in all 58 subjects. WHR was calculated by
dividing the waist circumference at the narrowest
place above the umbilicus by the hip circumference
at the widest place.1 As determinations of WHR were
only started in February 1989, 40 women were measured: 18 women showed upper body segment obesity
(android type, WHR > 0.85, group 1), 22 women
lower body segment obesity (gynoid type,
WHR 0.85, group 2). Systolic and diastolic blood
pressure (Riva-Rocci) were measured in lying position: normal values: readings up to 140/90; borderline
hypertension: 140±160/90±95; hypertensive readings:
> 160/95. Eleven women with android and 18 women
with gynoid obesity took part in a weight reduction
program with reduced caloric intake of 20930±
41860 kJ per week (5000±10000 kcal per week).
Oral glucose tolerance test
After following a 250±300 g carbohydrate diet for 3 d,
an oral glucose tolerance test (OGTT, 100 g, DextrooGT-glucose drink, Boehringer, Mannheim, Germany) was performed after an overnight fast at
0800 h during days 2±5 of the menstrual cycle. In
patients with amenorrhoea, menstrual bleeding was
induced by progestin challenge. Antecubital cannulation was performed 15 min prior to ®rst blood sampling. One fasting and four venous blood samples
after glucose loading were taken at hourly intervals (0,
60, 120, 180 and 240 min). Insulin, connecting peptide
(C-peptide), oestradiol, testosterone, dihydrotestosterone (DHT), dehydroepiandrosterone (DHEA), dehydroepiandrosterone sulphate (DHEAS), androstene
dione, cortisol, adrenocorticotrophic hormone
(ACTH) and growth hormone (GH) were assayed in
each sample. Single measurements (0 min) of prolactin, luteinizing hormone (LH), follicle-stimulating
hormone (FSH) and a biochemical pro®le including
triglycerides, cholesterol, high density lipoprotein
(HDL)-cholesterol, AST, ALT and gGT were performed. Hourly capillary blood sugar (Glucostix,
Bayer, Leverkusen, Germany) was measured photometrically with a re¯ectometer (Ames). Normal
values were: < 100 mg/dl at 0, 180, 240 min;
< 160 mg/dl at 60 min; < 120 mg/dl at 120 min.
Pathological values were: > 130 mg/dl at 0, 180,
240 min; > 220 mg/dl at 60 min; > 150 mg/dl at
120 min. All values between these areas were considered as borderline. Glucose tolerance was normal with
all values in the normal range or the peak value in the
borderline area. It was considered as pathological with
all values in the pathological range and the fasting
value in the borderline or pathological area. Any other
possibility was regarded as impaired glucose tolerance. Insulin resistance was calculated by dividing
fasting blood sugar by fasting insulin.1 Values less
than six were considered as pathological. Hyperinsulinaemia was de®ned as fasting insulin levels greater
than 24 mU/ml.
Biochemical determinations
LH, FSH and prolactin were assayed using immuno¯uorescence kits (LKB-Pharmazia, Helsinki, Finland).
The following hormone determinations were performed by radioimmunoassay (RIA): 17b-oestradiol
(TRAVENOL,
Dudingen,
Switzerland),
GH
(SERONO Diagnostika GmBH, Freiburg, Germany),
insulin and C-peptide (Diagnostic Products Corporation, Bad Nauheim, Germany) and ACTH (ACTHAllegro kit, Nichols-Institute, Bad Neuheim,
Germany). Testosterone and DHT were extracted with
help of dichlormethan and chromatography
in a benzol-heptan-methanol-water-system (300:500/
560:240). Separation was performed with testosterone-antibodies and radioactive, tritium-labelled 5DHT. DHEAS was assayed after 1:600 solution with
direct RIA. DHEA was determined with DHEA-antibodies after extraction with dichlormethan and chromatography in a benzol-heptan-methanl-water-system
(250:150/220:120). Androstenedione determination
was performed after extraction with isooctane. Cortisol was measured after extraction with ethanol. Determinations of AST, ALT, gGT, triglycerides,
cholesterol and HDL-cholesterol were performed in
the routine laboratory of the Medical University
Hospital of Heidelberg (Ludolf-Krehl-Klinik, Prof.
Dr. Fiehn) with standardized methods. Cholesterol/
HDL-cholesterol-ratio was calculated by dividing
cholesterol by HDL-cholesterol. Ratios greater than
four are associated with an increased cardiovascular
risk.32 The intraassay coef®cient of variation was less
than 10% for all methods applied. The interassay
coef®cient of variation was as follows: 11.1% for
LH, 10.8% for FSH, 10.4% for prolactin, 15.6% for
testosterone and DHT and 12±15% for ACTH, cortisol, androstenedione, DHEA and DHEAS. It was less
than 10% for all other methods applied.
Statistical analysis
Statistical analysis was performed using the Statistical
Analysis Program Package (SAS Institute, 1985).
Non-parametric methods were applied as the variables
477
WHR and BMI in young obese women
M Hollmann et al
478
were not normally distributed. Comparisons between
groups were made using the Mann-Whitney-Wilcoxon
U-test. Values are expressed as median (25th±75th
percentile) unless otherwise stated. w2 test and Fisher's exact test were used for ordinal variables, the
Spearman correlation coef®cient for correlation analysis.33 Differences were considered signi®cant at
P < 0.05.
monal differences between groups are shown in Fig
ures 1 and 2. Blood sugar, insulin and C-peptide levels
were signi®cantly higher in women with android
obesity versus group 2. Women with android obesity
were hyperinsulinaemic in 72%, subjects with gynoid
obesity in 38% (P < 0.05). No signi®cant differences
were found in the percentage of impaired or pathological glucose tolerance between groups (group 1: 61%
impaired, 6% pathological versus group 2: 45%
impaired). Despite a tendency to higher ACTH and
lower DHEA and cortisol concentrations in women
with android obesity, the level of signi®cance was
only reached at one time point following oral glucose
load for each hormone. No signi®cant differences
were found in LH, FSH, GH, androstenedione, testosterone, DHT, DHEAS, and oestradiol concentrations
Results
Women with android obesity had signi®cantly greater
systolic blood pressure readings (Table 1). The hor-
Table 1 Age, anthropometric data and blood pressure in the total group of obese patients (total), in the group of women with android
obesity and the group of women with gynoid obesity
Total
Android
Gynoid
Parameter unit
na
Median
25th^75thb
n
Median
25th^75th
n
Median
Age (y)
Body-Mass-Index kg/m2
Excess weight %
Syst. blood pressure mmHg
Diast. blood pressure mmHg
Waist-Hip-Ratio
58
58
58
44
44
40
28
34.6
46.7
123
80
0.84
25±32
37.3±39.8
31.0±46.5
110±140
73±90
0.8±0.88
18
18
18
18
18
18
28
36.1
53.8
133
80
0.88
24±29
33.9±39.3
41±67.9
120±140
70±90
0.86±0.93
22
22
22
22
22
22
28
33.7
39
120
80
0.80
a
b
25th^
75th\phi
Pc
23±30
30.4±35.4
30.8±50
110±130
80±90
0.77±0.82
ns
ns
ns
0.05
ns
0.001 >
n ˆ number of patients.
25th±75th ˆ 25th±75th percentile.
Figure 1 Blood glucose in the group of women with android obesity and the group of women with gynoid obesity. Time 0 fasting
values, 60±240 min after oral glucose load. u ˆ QD android: quartile difference 75th±25th percentile in the group of women with
android obesity (group 1). jj ˆ QD gynoid: quartile difference 75th±25th percentile in the group of women with gynoid obesity. ±r±
ˆ median android: median in the group of women with android obesity (group 1). ±j± ˆ Median gynoid: median in the group of women
with gynoid obesity (group 2). P ˆ signi®cance.
WHR and BMI in young obese women
M Hollmann et al
479
Figure 2 Insulin and C-peptide in the group of women with android obesity and the group of women with gynoid obesity. Time 0
fasting values, 60±240 min after oral glucose load. u ˆ QD android: quartile difference 75th±25th percentile in the group of women with
android obesity (group 1). jj ˆ QD gynoid: quartile difference 75th±25th percentile in the group of women with gynoid obesity. ±r±
ˆ median android: median in the group of women with android obesity (group 1). ±j± ˆ Median gynoid: median in the group of women
with gynoid obesity (group 2). P ˆ signi®cance.
WHR and BMI in young obese women
M Hollmann et al
480
between groups. In the total group of 58 obese
women, WHR correlated inversely with the ratio of
fasting glucose to insulin (Spearman correlation coef®cient r ˆ 70.35, P < 0.05). Thus, the more insulin
resistance a subject was, the higher was the WHR.
AST, gGT, tryglycerides, and the cholesterol/HDLcholesterol-ratio were signi®cantly higher in women
with android obesity (Table 2). Hypertriglyceridaemia
was found in 53% of group 1-patients and 9% of
group 2-patients (P < 0.01). Cardioprotective HDLcholesterol was signi®cantly greater in group 2.
All patients in group 1 vs 57% in group 2 had a
cholesterol/HDL-cholesterol-ratio greater than four
(P < 0.01).
In the total group of 58 obese women, WHR
correlated directly with triglycerides, cholesterol/
HDL-cholesterol-ratio, AST, gGT, and inversely
with HDL-cholesterol (Table 3). Triglycerides and
ALT were associated closer with the BMI and %
excess weight than with the WHR. Whereas no
correlation was found between the BMI or excess
weight and HDL-cholesterol or cholesterol/HDLcholesterol-ratio.
The achieved weight loss in group 1 was 7.5
(4±11) kg vs 10 (6.1±18.7) kg in group 2 after
32 14 weeks (mean s.d.) (n.s.). One patient in
group 1 vs 5 patients in group 2 conceived spontaneously after weight reduction without additional
therapy (n.s.). On dividing the group of 58 obese
women according to the BMI, 7 women had grade
1-(BMI 25.0±29.9 kg/m2), 43 women grade 2(BMI 30.0±40.0 kg/m2) and 8 women grade 3-obesity
(BMI > 40.0 kg/m2); 3). Fasting blood glucose was
signi®cantly greater in the group of women with
grade 3-obesity (85, 79.5±99.5 mg/dl) vs subjects
with grade 2-obesity (71, 62±82 mg/dl; P < 0.05).
The difference was also signi®cant at 120 min after
glucose load with 149 (138±173) mg/dl compared
with to the groups with grade 2-obesity (110, 96±
133 mg/dl) or grade 1-obesity (85, 73±108 mg/dl;
P < 0.05). Serum testosterone levels at 120 min
(41, 22±61 mg/dl) and 180 min (40, 19.6±54 mg/dl)
were signi®cantly higher in the group with grade 2obesity compared to grade 1 subjects (21, 14±25 mg/dl
and 19.3, 1743 mg/dl respectively; P < 0.05). However, no signi®cant difference was found compared
to grade 3 subjects. BMI correlated directly with
triglycerides and liver enzymes (Table 3).
Discussion
When analyzing the results of our study, one has to
consider that we investigated a highly selected population of obese, infertile women, Nevertheless, in this
group of young obese women with a mean age of
28.8 5 y, android obesity was associated with signi®cantly higher levels of systolic blood pressure,
blood glucose, AST and gGT. The risk of cardiovascular disease was increased as indicated by a
signi®cantly greater percentage of patients with
hypertriglyceridaemia,
higher
cholesterol/HDLcholesterol-ratios and suppressed HDL-cholesterol.
In agreement with our results, the associations
between android obesity and the metabolic syndrome
has been shown in previous studies.1,3,5,6,13,17 Assessing the WHR is a simple procedure to identify obese
risk patients.1,3 Interestingly, the differences in the
metabolic pro®le, which depend on the body fat
distribution, can already be found in young patients
Table 2 Liver enzymes and serum lipids in the group of women with android obesity and the group of women with gynoid obesity
Android
a
Parameter unit
n
Median
AST U/l
ALT U/l
gGT U/l
Triglycerides mg/dl
HDL-cholesterol mg/dl
Cholesterol/HDL-cholesterol-ratio
17
17
17
17
17
17
10
14
16
157
35
5.7
a
b
c
Gynoid
b
25th^75th
n
Median
9±13
12±21
11±18
86±199
32±38
4.8±6.8
22
22
22
22
21
21
8.5
11
9
97
41
4.7
25th^75th
Pc
7±10
9±14
5±13
69±131
37±47
3.8±5.2
0.05
ns
0.001
0.05
0.02
0.002
n ˆ number of patients.
25th±75th ˆ 25th±75th percentile.
P ˆ signi®cance (Mann-Whitney-Wilcoxon U-test).
Table 3 Correlations (Spearman coef®cient) between serum lipids or liver enzymes and the body weight (kg), body mass index (BMI,
kg/m2), excess body weight (%) and the waist-hip-ratio (WHR) in the total group of 58 obese, infertile women
Parameter
Body weight
Triglycerides
HDL-cholesterol
Cholesterol-HDL-ratio
AST
ALT
gGT
‡0.31*
70.06
‡0.13
‡0.24
‡0.32*
‡0.37**
Signi®cance: * P < 0.05; ** P < 0.01; *** P < 0.001.
BMI
‡0.41**
70.12
‡0.20
‡0.32*
‡0.42**
‡0.41**
Excess weight
‡0.42**
70.13
‡0.21
‡0.32*
‡0.44***
‡0.39**
WHR
‡0.34*
70.46**
‡0.53**
‡0.46**
‡0.34
‡0.49**
WHR and BMI in young obese women
M Hollmann et al
prior to clinical manifestation of obesity-related disorders. Thus, measuring the WHR facilitates preventive medicine through identi®cation and rigorous
treatment of these risk patients. On dividing the
group of obese women according to the BMI, we
could not associate speci®c metabolic pro®les with
different grades of obesity. As previously demonstrated, the WHR is of greater importance than the
BMI in predicting the risk of metabolic complications.1,3,6,16
Obesity, especially android type obesity, is commonly associated with the insulin resistant state and
secondary hyperinsulinaemia.1,3,6,12,18,20,34,35 We also
found signi®cantly higher insulin levels in women
with android obesity. The insulin resistance was
greater with increasing WHR. The insulin resistance
is discussed as the common pathway leading to the
metabolic complications in obesity.1,10,12,20,36,37 Insulin has been found to cause arteriosclerotic lesions and
a proliferation of vascular smooth muscle cells and to
promote renal tubular sodium and water absorption
which may contribute to the development of cardiovascular disease.1,10,37,38 Insulin resistance results in
an increased hepatic glucose release, and a decreased
catabolism of triglycerides contributing to an impaired
glucose tolerance and to reduced HDL-cholesterol
levels.1,10 The metabolic complications of android
obesity may be mediated by the more pronounced
hyperinsulinaemia. However, we did not ®nd any
correlations between insulin levels and metabolic
parameters.
In the ®eld of gynaecology, obesity is associated
with menstrual irregularity, chronic anovulation, and
the PCO-syndrome.23,24,34±36,39±41 Especially women
with android obesity seem to be prone to develop
menstrual disorders.12,15,18,20,22,34 Even in lean
women, an increasing WHR was associated with a
reduced chance of conception.24 The possible role of
insulin as the mediating factor has aroused a great deal
of interest in recent years. Receptors for insulin have
been demonstrated in human ovaries.12,42 In vivo and
in vitro studies showed a direct stimulatory effect of
insulin on ovarian androgen secretion acting in synergism with LH.12,19,42,43 In those obese women susceptible to the development of menstrual irregularity,
an additional, genetically determined defect of androgen-secreting tissues, resulting in a greater insulin
sensitivity than in normal ovaries, was postulated.12,20
The increased androgen concentrations serve as substrates for active aromatization into oestrogens in the
body fat compartment which is an important source of
oestrogens in obese women.9,12,20 The resulting
hyperoestrogenaemia is thought to modulate pituitary
gonadotrophin release by increasing LH and decreasing FSH secretion.12,15,23,34,39 Increased LH levels
result in further stimulation of the androgen production. Ovarian follicular maturation deteriorates due to
the lack of FSH and the increased androgen concentrations in the ovary. After a variable period of time,
these hormonal changes lead to the polycystic
morphology of the ovaries due to chronic
anovulation.12,20,39
Some studies observed a high prevalence of hyperandrogenaemia in women with android obesity.5,12,13,44 By contrast, this could not be con®rmed
by other authors in agreement with our results.13,45
However, they found increased free androgen and
decreased SHBG levels, thus increased free androgenic activity in subjects with android obesity.13,14,16,18,45 Other studies noted an inverse
association between insulin and SHBG. Insulin itself
could cause increased androgenic activity by decreasing SHBG levels.13,14,20,34,46,47 Despite increased
insulin levels in women with android obesity, we
could not ®nd any differences in androgens, oestrogens or gonadotrophins in comparison to women with
gynoid obesity. We did not observe any correlation
between WHR, insulin or androgens. Both groups of
obese women complained of infertility, however,
which may imply hormonal differences in comparison
to women with simple obesity. The prevalence of 45%
of women with android obesity in a group of women
with menstrual irregularity exceeded the normally
observed rate of 10±20%.5 Women with upper body
segment obesity seem to be more prone to develop
menstrual dysfunction. We could not identify a certain
mechanism, for example the hyperinsulinaemia, to be
responsible for this observation. However, the hyperinsulinaemia may act through mechanisms we did not
investigate in our study, for example free hormone
concentrations, SHBG levels or insulin-like growth
factors.
Weight loss and physical activity are recommended
as therapy of both types of obesity.1,29,36,48 We previously found a pregnancy rate of 29% and an
improvement of menstrual dysfunction in 80% in
obese, infertile women after simple weight reduction.29 Further controlled studies are necessary to
investigate whether weight loss, pregnancy rate and
changes of menstrual dysfunction generally differ
depending on the body fat distribution.
Conclusions
In summary, the WHR is a valuable parameter in
identifying those obese patients who are at a greater
risk of developing obesity-related disorders. As we
found higher levels of systolic blood pressure, triglcyerides, blood sugar, liver enzymes and lower
cardioprotective HDL-cholesterol concentrations
already in a group of young women with android
obesity compared to subjects with lower body fat
localisation, the WHR could play a role in preventive
medicine. One can identify risk patients prior to
manifestation of diseases and treat them more rigorously. Women with upper body segment obesity seem
to be more prone to develop menstrual disorders. The
more pronounced hyperinsulinaemia in women with
481
WHR and BMI in young obese women
M Hollmann et al
482
upper body fat localisation may be the link to the
abnormal ovarian function.
Acknowledgements
We thank Dr. Klinga and the laboratory staff of the
University Hospital of Obstetrics and Gynaecology
for their work and support. We further thank the
laboratories of the Medical Hospital (Ludolf-KrehlKlinik), of the Surgical Hospital and of the Institute of
Pharmacology of the University of Heidelberg for
their work.
References
1 Caro JF. Insulin resistance in obese and nonobese man. J Clin
Endocrinol Metab 1991; 73: 691±696.
2 Bergmann KE, Menzel R, Bergmann E, Tietze K, Stolzenberg
È bergewicht in der
H, Hoffmeister H. Verbreitung von U
erwachsenen BevoÈlkerung der Bundesrepublik Deutschland.
Akt ErnaÈhrungsmed 1989; 14: 205±208.
3 MuÈller MJ, Lautz HU, von zur MuÈhlen A, HoÈllwarth I,
Canzler H, Schmidt FW. Pathogenese und Therapie der
Adipositas. Dtsch Med Wochenschr 1990; 115: 789±794.
4 Galvao-Teles A, Baptista F, DoCarmo I. Obesity and cancer.
Mediographica 1991; 13: 16±18.
5 Hauner H, Ditschuneit HH, Pal SB, Pfeiffer EF. Fettgewebsverteilung und Adipositaskomplikationen bei uÈbergewichtigen
Frauen mit und ohne Hirsutismus. Dtsch Med Wochenschr
1987; 112: 709±713.
6 Kissebah AH, Vydelingum N, Murray R, Evans DJ, Hartz AJ,
Kalkhoff RK, Adams PW. Relation of body fat distribution to
metabolic complications of obesity. J Clin Endocrinol Metab
1982; 54: 254±260.
7 Alexander JK. Cardiac structure and function in obesity.
Mediographica 1991; 13: 9±12.
8 Glass AR. Endocrine aspects of obesity. Med Clin North Am
1989; 73: 139±160.
9 Hershcopf RJ, Bradlow HL. Obesity, diet, endogenous estrogens, and the risk of hormone-sensitive cancer. Am J Clin Nutr
1987; 45: 283±289.
10 Simonson DC. Hyperinsulinemia and its sequelae. Horm
Metab Res Suppl 1990; 22: 17±25.
11 Forney JP, Milewich L, Chen GT, Garlock JL, Schwarz BE,
Edman CD, MacDonald PC. Aromatization of androstendione
to estrone by human adipose tissue in vitro. Correlation of
adipose tissue mass, age, and endometrial neoplasia. J Clin
Endocrinol Metab 1981; 53: 192±199.
12 Barbieri RL, Smith S, Ryan KJ. The role of hyperinsulinemia
in the pathogenesis of ovarian hyperandrogenism. Fertil Steril
1988; 50: 197±212.
13 Evans DJ, Hoffmann RG, Kalkhoff RK, Kissebah AH. Relationship of androgenic activity to body fat topography, fat cell
morphology, and metabolic aberrations in premenopausal
women. J Clin Endocrinol Metab 1983; 57: 304±310.
14 Haffner SM, Valdez RA, Stern MP, Katz MS. Obesity, body
fat distribution and sex hormones in men. Int J Obes Relat
Metab Disord 1993; 17: 643±649.
15 Harlass FE, Plymate SR, Fariss BL, Belts RP. Weight loss is
associated with correction of gonadotropin and sex steroid
abnormalities in the obese anovulatory female. Fertil Steril
1984; 42: 649±652.
16 Marin P, Kvist H, Lindstedt G, Sjostrom L, Bjorntorp P. Low
concentrations of insulin like growth factor-1 in abdominal
obesity. Int J Obes Relat Metab Disord 1993; 17: 83±89.
17 Sonnichsen AC, Ritter MM, Mohrle W, Richter WO, Schwandt
P. The waist-to-hip ratio corrected for body mass index is related
to serum triglycerides and high-density lipoprotein cholesterol
but not to parameters of glucose metabolism in healthy premenopausal women. Clin Investig 1993; 71: 913±917.
18 Grenman S, RoÈnnemaa T, Irjala K, Kaihola HL, GroÈnroos M.
Sex steroid, gonadotropin, cortisol, and prolactin levels in
healthy, massively obese women: correlation with abdominal
fat cell size and effect of weight reduction. J Clin Endocrinol
Metab 1986; 63: 1257±1261.
19 Insler V, Shoham Z, Barash A, Koistinen R, Seppala M, Hen
M, Lunenfeld B, Zadik Z. Polycystic ovaries in non-obese and
obese patients: possible pathophysiological mechanism based
on new interpretation of facts and ®ndings. Hum Reprod 1993;
8: 379±384.
20 Nestler JE, Clore JN, Blackard WG. The central role of
obesity (hyperinsulinema) in the pathogenesis of the polycystic ovary syndrome. Am J Obstet Gynecol 1989; 161: 1095±
1097.
21 Tropeano G, Lucisano A, Liberale I, Barini A, Vuolo IP,
Martino G, Menini E, Dell'Acqua S. Insulin, C-peptide,
androgens, and beta-endorphin response to oral glucose in
patients with polycystic ovary syndrome. J Clin Endocrinol Metab 1994; 78: 305±309.
22 Zhang YW, Stern B, Rebar RW. Endocrine comparison of
obese menstruating and amenorrheic women. J Clin Endocrinol Metab 1984; 58: 1077±1083.
23 Rich-Edwards JW, Goldman MB, Willett WC, Hunter DJ,
Stampfer MJ, Colditz GA, Manson JA. Adolescent body mass
index and infertility caused by ovulatory disorder. Am J Obstet
Gynecol 1994; 171: 171±177.
24 Hartz AJ, Barboriak PN, Wong A, Katayama KP, Rimm
AA. The association of obesity with infertility and related
menstrual abnormalities in women. Int J Obes 1979; 3:
57±73.
25 Zaadstra BM, Seidell JC, Van Noord PAH, te Velde ER,
Habbema JDF, Vrieswijk B, Karbaat J. Fat and female
fecundity: prospective study of effect of body fat distribution
on conception rates. BMJ 1993; 306: 484±487.
26 Crosignani PG, Ragni G, Parazzini F, Wyssling H, Lombroso
G, Perotti L. Anthropometric indicators and response to
gonadotrophin for ovulation induction. Hum Reprod 1994; 9:
420±423.
27 Galtrier-Dereure F, Montypeyroux F, Boulot P, Bringer J,
Jaf®ol C. Weight excess before pregnancy: complications and
cost. Int J Obes 1995; 19: 443±448.
28 Yen SSC. The polycystic ovary syndrome. Clin Endocrinol
1980; 12: 177±208.
29 Hollmann M, Runnebaum B, Gerhard I. Effects of weight loss
on the hormonal pro®le in obese, infertile women. Hum
Reprod 1996; 11: 1884±1891.
30 Gerhard I, Reischmann T, Eggert-Kruse W, Runnebaum B.
Untersuchung der Faktoren, die den Schwangerschaftseintritt
und -ausgang durch medikamentoÈse Therapie bei sterilen
Frauen beein¯ussen. I. Anamnestische und klinische Parameter. FertilitaÈt 1990; 6: 123±135.
31 Balen AH, Conway GS, Kaltsas G, Techatraisak K, Manning
PJ, West C, Jacobs HS. Polycystic ovary syndrome: the
spectrum of the disorder in 1741 patients. Hum Reprod
1995; 10: 2107±2111.
32 Herold G. Innere medizin. In: Herold G (ed). Innere Medizin.
KoÈln, 1991, pp. 505±510
33 Immich H. Medizinische Statistik. Schattauer: Stuttgart, New
York, 1974.
34 Dunaif A, Mandeli J, Fluhr H, Dobrjansky A. The impact
of obesity and chronic hyperinsulinemia on gonadotropin
release and gonadal steroid secretion in the polycystic
ovary syndrome. J Clin Endocrinol Metab 1988; 66: 131±
139.
35 Peiris AN, Struve MF, Kissebah AH. Relationship of body fat
distribution to the metabolic clearance of insulin in premenopausal women. Int J Obes 1987; 11: 581±589.
36 Stout RW. Overview of the association between insulin and
atherosclerosis. Metabolism 1985; 34: 7±12.
37 Foster DW. Insulin resistanceÐa secret killer. N Engl J Med
1989; 320: 733±734.
WHR and BMI in young obese women
M Hollmann et al
38 Every NR, Boyko EJ, Keane EM, Marshall JA, Rewers M,
Hamman RF. Blood pressure, insulin, and C-peptide levels in
San Luis Valley, Colorado. Diabetes Care 1993; 16: 1543±
1550.
39 Pasquali R, Casimirri F. The impact of obesity on hyperandrogenism and polycystic ovary syndrome in premenopausal
women. Clin Endocrinol Oxf 1993; 39: 1±16.
40 Gerhard I, Postneek F. MoÈglichkeiten der Therapie durch
Ohrakupunktur bei weiblicher SterilitaÈt. Geburtsh u Frauenheilk 1988; 48: 165±171.
41 Gerhard I, Matthes J, Runnebaum B. The induction of ovulation with pulsatile gonadotrophin-releasing hormone (GnRH)
administration in hyperandrogenic women after down-regulation with buserelin or suppression with an oral contraceptive.
Hum Reprod 1993; 8: 2033±2038.
42 Poretsky L, Kalin MF. The gonadotropic function of insulin.
Endocr Rev 1987; 8: 132±141.
43 Conway GS, Clark PM, Wong D. Hyperinsulinemia in the
polycystic ovary syndrome con®rmed with a speci®c immunoradiometric assay for insulin. Clin Endocrinol Oxf 1993; 38:
219±222.
44 Pasquali R, Casimirri F, Cantobelli S, Labate AM, Venturoli
S, Paradisi R, Zannarini L. Insulin and androgen relationships
with abdominal body fat distribution in women with and
without hyperandrogenism. Horm Res 1993; 39: 179±187.
45 Peiris AN, Mueller RA, Struve MF, Smith GA, Kissebah AH.
Relationship of androgenic activity to splanchnic insulin
metabolism and peripheral glucose utilization in premenopausal women. J Clin Endocrinol Metab 1987; 64: 162±169.
46 Hamilton-Fairley D, Kiddy D, Anyaoku V, Koistinen R,
Seppala M, Franks S. Response of sex hormone binding
globulin and insulin-like growth factor binding protein-1 to
an oral glucose tolerance test in obese women with polycystic
ovary syndrome before and after caloric restriction. Clin
Endocrinol Oxf 1993; 39: 363±367.
47 Rajkhova M, Bicknell J, Jones M, Clayton RN. Insulin
sensitivity in women with polycystic ovary syndrome: relationship to hyperandrogenemia. Fertil Steril 1994; 61: 605±
612.
48 James WPT. Treatment of obesity and the quality of life.
Mediographica 1991; 13: 19±22.
483