˘ ˇ ˆuu.diva-portal.org/smash/get/diva2:314588/fulltext01.pdfphenotype without hyperandrogenism....

��

��

��

� � !��"#$%��&!

��'�()�(*)+�) �,'�-�.*-(*��/*�.+)*(��0��0��0��0��*(+1+/.

�� !��"#��$%"%��%�&%%�'��(��)��'��'�*(��(��+,��'� ��-.�/(��0��10��(.

��

�� .�$%"%.�� 2��3��'��(��*��4��1��.�5�� .�� 6%.�67��.� ��.�819:��6;<�"<��#<6;$�<".

*�� +*241-� �� (��=�� (��)�� (�� )��.� 8'��3�� 0�� 0��(� *241� �� 0��(� ��'�� )��0��(��'��0��.�/(��>�'��)��$��)��+8?/-��(��<��'��(�)(��*241�0��(��'�)��)��(��(��(��.@��0��(�*241� �� (�� >� '�� '�� (�)(� �� 0(��(��(��(��(��0��.�A��'�0��)��(��)<��3��0��0��(��'��*241�(��.�8��(��)<��'��0<��'�0��0��(�*241��;#�0��0��(��)��'�*241��0��"�;6��"��)��(��'��0<��B�7��)�<��(��<��)��'��0��.�� (�� '�� <��'��0��.�5��)��0��0��(�*241�(��)��(�� (�� '� �� )�� 0�� (�)(.� /(�� '��)�� 0��*241��(��'��)��)��(��0��(��'��<�C��(��))�� (�� (�� 0��'��)�.�*241�0��)��'��(��'��(��0��)��(��'��>�'��'��(��9 8�� )�� .� 8?/� �� $� �� '�� *241�0��.�,��)��<��'��0(��(�)��.�4��)��)��)��0��)(�<)��(��(��'��0<��.�9��<��'��0��*241�0�� 0��(� �� D�� .� 8�� '*241��(��(��'�*241��(��'��0<��)��0��0��(�*241�(��0��'��0��(��D��.�2��&�/(��)<��'�*241��0��0��(��.�5��(�)(��)��=��0��0��(�*241��<��'��(��(��>�'�8?/��$��.

�� )<��'��0<��< C��(��)��(��'��(��<��

�� !��"�� "�� #$��$��%�$�� &'()*+)��

E� ��$%"%

811:�"��"<�$%�819:��6;<�"<��#<6;$�<"��&�&��&��&��<"$7$#;�+(��&��.>�.��F��G��&�&��&��&��<"$7$#;-

To my family

List of Papers

This thesis is based on the following papers, which are referred to in the text by their Roman numerals.

I Hudecova M, Holte J, Olovsson M, Sundström-Poromaa I.

Long-term follow-up of patients with polycystic ovary syndrome – reproductive outcome and ovarian reserve. Hum Reprod. 2009; 24:1176-83.

II Hudecova M, Holte J, Olovsson M, Lind L, Sundström-

Poromaa I. Endothelial function in patients with polycystic ovary syndrome – a long-term follow-up study, Fertility and Sterility, in press.

III Hudecova M, Holte J, Moby L, Olovsson M, Stridsberg M,

Larsson A, Berglund L, Berne C, Sundström-Poromaa I. Androgen levels, insulin sensitivity and early insulin response in women with polycystic ovary syndrome – a long-term follow-up study, submitted.

IV Hudecova M, Holte J, Moby L, Olovsson M, Larsson A,

Berne C, Sundström-Poromaa I. Prevalence of diabetes, impaired glucose tolerance and insulin sensitivity in women with polycystic ovary syndrome – a long-term follow-up study, manuscript.

Reprints were made with permission from the respective publishers.

Contents

Introduction ............................................................................................. 11�Definition of PCOS ........................................................................ 11�Clinical features of PCOS .............................................................. 12�Pathophysiology of PCOS.............................................................. 12�Reproductive consequences of PCOS ............................................ 15�Ovarian reserve .............................................................................. 16�Metabolic consequences of PCOS ................................................. 17�Cardiovascular consequences of PCOS ......................................... 20�Prior long-term studies in women with PCOS ............................... 21�

Material and methods .............................................................................. 23�Patients ................................................................................................ 23�

PCOS diagnosis .............................................................................. 23�Design ................................................................................................. 25�

Paper I ............................................................................................ 25�Paper II ........................................................................................... 25�Paper III .......................................................................................... 25�Paper IV ......................................................................................... 26�

Methods .............................................................................................. 26�Paper I ............................................................................................ 26�Paper II ........................................................................................... 27�Paper III .......................................................................................... 29�Paper IV ......................................................................................... 30�

Results ................................................................................................ 31�Paper I ............................................................................................ 31�Paper II ........................................................................................... 34�Paper III .......................................................................................... 38�Paper IV ......................................................................................... 42�

Discussion ............................................................................................... 49�Methodological considerations ...................................................... 49�Reproductive outcome ................................................................... 51�Ovarian reserve .............................................................................. 52�Endothelial dysfunction ................................................................. 53�

Metabolic syndrome ....................................................................... 54�Hyperandrogenism ......................................................................... 54�Insulin resistance, glucose tolerance, early insulin response and diabetes mellitus ............................................................................. 56�

General conclusions ................................................................................ 58�

Acknowledgements ................................................................................. 60�

References ............................................................................................... 62�

Abbreviations

ACTH Adrenocorticotropic hormone AES The Androgen Excess Society AI Augmentation index AHS American Heart Association AMH Anti-müllerian hormone ApoA-I Apolipoprotein A-I ApoB Apolipoprotein B ASRM American Society for Reproductive

Medicine AUC Area under the curve BMI Body mass index CHD Coronary heart disease hs-CRP High-sensitivity C-reactive protein CV Coefficient of variation CVD Cardiovascular disease DHEAS Dehydroepiandrosterone EDV Endothelium-dependent vasodilation ESHRE European Society for Human

Reproduction FAI Free androgen index FFA Free fatty acids FSH Follicle-stimulating hormone GnRH Gonadotropin-releasing hormone HDL High density lipoprotein cholesterol HOMA-IR Homeostatic model assessment of

insulin resistance ICSI Intracytoplasmatic sperm injection IGFBP Insulin-like growth factor-binding

protein IGT Impaired glucose tolerance IR Insulin resistance IVF In vitro fertilization IVGTT Intravenous glucose tolerance test

LDL Low-density lipoprotein cholesterol LH Luteinizing hormone MetS Metabolic syndrome M/I Insulin sensitivity index NCEP/ATPIII The US National Cholesterol

Education Program Adult TreatmentPanel III

NIH National Institutes of Health OGTT Oral glucose tolerance test PCO Polycystic ovary PCOS Polycystic ovary syndrome QUICKI Quantitative insulin sensitivity check

index SHBG Sex-hormone binding globulin TSH Thyroid-stimulating hormone

11

Introduction

Definition of PCOS

Polycystic ovary syndrome (PCOS) is a complex and heterogeneous clinical condition characterized by hyperandrogenism and chronic oligo/anovulation. It is estimated that approximately 5 - 10 % of fertile women have PCOS, although no precise determinations have been possible to obtain [1-5].

The National Institutes of Health (NIH) defined PCOS in 1990 as chronic anovulation or oligomenorrhea, and clinical or biochemical hyperandrogenism [6]. In 2003 revised diagnostic criteria were proposed by the European Society for Human Reproduction and Embryology and the American Society for Reproductive Medicine (ASRM/ESHRE) which also included the typical ultrasonographic appearance of the ovaries [7]. In concordance with the revised diagnostic criteria two out of the three following criteria should be fulfilled for establishing the PCOS diagnosis:

1. Oligo- and/or anovulation 2. Clinical and/or biochemical signs of hyperandrogenism 3. Polycystic ovaries

and exclusion of other etiologies (congenital adrenal hyperplasia, androgen-secreting tumors, Cushing’s syndrome). ASRM/ESHRE consensus defines polycystic ovaries if at least one of the ovaries has at least one of the following: either 12 or more follicles measuring 2-9 mm in diameter, or increased ovarian volume (> 10 cm3). If there is a follicle > 10 mm in diameter, the scan should be repeated at a time of ovarian quiescence in order to calculate volume and area. Thus, in comparison with the NIH criteria, PCOS patients could present with at least one more possible phenotype i.e. with oligo/anovulation and polycystic ovaries (PCO) only. This led to the initiative of The Androgen Excess Society (AES) and in 2006, the, at the moment last version of PCOS definition was proposed. The AES definition is based on a presence of hyperandrogenism and ovarial dysfunction (expressed as oligomenorrhea/amenorrhea or polycystic ovaries), and exclusion of other androgen excess or related disorders. All possible phenotypes are based on the presence or absence of oligoanovulation, hyperandrogenemia, hirsutism and polycystic ovaries at ultrasound [8], but this definition excludes the previously mentioned

12

phenotype without hyperandrogenism. In this thesis PCOS is defined according to the Rotterdam criteria, as these criteria were the ones in use when our patients were diagnosed. Because of the ongoing debate on how to best define PCOS, results in this thesis are often also given for subjects fulfilling NIH criteria.

Clinical features of PCOS Oligomenorrhea and amenorrhea, as signs of anovulation, and hirsutism, as a sign of hyperandrogenism, are the most common complaints of patients with PCOS.

Menstrual cycles longer than 35 days or less then eight menstruation periods per year, are usually applied as definitions of oligo-/amenorrhea [9].

The Ferriman-Gallwey scoring system has been designed to assess the severity of hirsutism. The masculine pattern of body hair growth is described in four degrees on 11 different body places; the upper lip, chin, chest, upper back, lower back, upper abdomen, lower abdomen, arm, forearm, thigh, and lower leg [10]. Although often used in scientific reports on PCOS, the Ferriman-Gallwey score is hampered by the large inter-rater variability and also by the fact that many women have treated themselves prior to clinical evaluation.

The calculation of the free androgen index (FAI) from testosterone and sex steroid hormone binding globulin (SHBG) is widely used for assessment of biochemical hyperandrogenism. As about 80 % of testosterone is bound to SHBG and 19 % to albumin, and only 1 % of testosterone is free and biologically available, total testosterone is not sufficient for evaluation of androgens in women [11]. Pathophysiology of PCOS The pathogenesis of PCOS is multifactorial. Insulin resistance, defined as a diminished effect of a given dose of insulin on glucose homeostasis, is a highly prevalent feature of women with PCOS. Insulin resistance in PCOS is closely associated with an increase in truncal-abdominal fat mass, elevated free fatty acid (FFA) levels, increased androgens, particularly free testosterone through reduced SHBG levels, and anovulation. The causes of insulin resistance in PCOS are still unknown [12].

13

Obesity

Insulin resistance

Hyperinsulinemia

Direct effects

on the ovary

Decreased hepatic

SHBG synthesis

Hyperandrogenism

Anovulation

Polycystic ovaries

Genetic

predisposition

Figure 1. A simplified theoretical model of the pathophysiology of the syndrome. Adapted from Sundström Poromaa I: Patogenes av PCOS, ARG�rapport nr. 58, 2008.

There is evidence for a genetic component in PCOS based on familial aggregation of cases [13]. Data suggest that PCOS develops as the consequence of an interaction of key genes of ovarian androgen production, with environmental factors, particularly nutritional, or other factors causing hyperinsulinemia [13, 14]. Genes involved in the biosynthesis and metabolism of androgens and those involved in the secretion and action of insulin are, logically, in the centre of interest [15].

Chronic androgen excess of ovarian and/or adrenal origin results in

abdominal adiposity in affected women [16]. Long-term administration of testosterone in female-to-male transsexuals induces abdominal adiposity and insulin resistance [17], and androgen excess during fetal life and infancy is associated with increased risk of developing abdominal adiposity later in life [18-20]. Testosterone applied in supraphysiological doses in women is also followed by insulin resistance [21].

Conversely, abdominal fat and free fatty acids are strongly associated

with insulin resistance (IR). The increased flux of FFA from highly lipolytic abdominal fat to the liver and muscles may represent an important link between abdominal obesity and insulin resistance, as elevations of plasma FFA levels inhibit insulin-stimulated glucose utilization [22-24]. Abnormalities in insulin secretion and action might play a role in development and maintenance of adiposity. An overactive early insulin

14

secretion could stimulate carbohydrate intake and constitute a basis for weight gain [23]. Indeed, overweight (body mass index (BMI) > 25 kg/m2) and obesity (BMI > 30 kg/m2) are very common in this group of women. Between 38 and 88 % of women with PCOS are overweight [25], and 10-38 % are obese [1, 3, 26]. Insulin resistance induces hyperinsulinemia and subsequently stimulates the ovarian and adrenal hormonal production, inhibits SHBG production, and increases testosterone activity. Gambineri and colleagues have summarized links between hyperinsulinemia and hyperandrogenism in PCOS women as follows:

� Insulin causes a direct stimulation of ovarian androgen secretion, possibly through stimulatory effects on the 17�-hydroxylase/17-20 lyase and on 3�-hydroxysteroid dehydrogenase enzymes

� Insulin decreases levels of SHBG, with concomitant elevation of free androgen tissue availability

� Insulin decreases insulin-like growth factor-binding protein (IGFBP) 1 production, both in the liver and in the ovary

� Insulin upregulates ovarian insulin-like growth factor 1 (IGF-1) receptor type I with amplification of IGF-1, IGF-2 and insulin actions in the ovary

� Insulin increases the luteinizing hormone (LH) ovary receptor number and sensitizes LH secreting pituitary cells to gonadotropin-releasing hormone (GnRH) stimulation

� Insulin promotes ovarian growth and cyst formations � Insulin can cause a direct stimulation of adrenal androgen

secretion, possibly through stimulatory effects on the 17�-hydroxylase/17-20 lyase enzyme

� Insulin may increase the 17�-hydroxylase/17-20 lyase adrenal enzyme response to adrenocorticotropic hormone (ACTH) [27].

Anovulation and polycystic ovaries are the result of disturbances of folliculogenesis. The total number of follicles is determined early in life. Normal follicle development comprises initial recruitment, by which primordial follicles start to mature, and cyclic recruitment, which leads to the growth of a cohort of small antral follicles from which the dominant follicle, destined to ovulate, is subsequently selected [28]. Anti-müllerian hormone (AMH) is expressed in granulosa cells of primary follicles already in the 36th week of pregnancy [29]. AMH continues to be expressed in the growing follicles in the ovary until they have reached the size and differentiation state at which they are to be selected for dominance by the action of pituitary follicle-stimulating hormone (FSH) [30]. The selective

15

rise in FSH levels that occurs during the luteal to follicular transition appears as a potent stimulus for follicle selection. AMH may prevent multiple selections in the follicle cohort [31]. Granulosa cells start to produce inhibins and estradiol that cause a small but significant and progressive decline in the circulating FSH concentration, due to their inhibitory effects on pituitary secretion [32, 33]. The dominant follicle alone continues to grow. In the absence of pregnancy, luteolysis occurs, the negative feedback of inhibin A and estradiol decreases and the inter-cycle elevation of FSH starts again [31].

The number of primary growing follicles in woman with PCOS and

anovulation is six-fold higher than in normal ovulatory women [34]. Serum AMH levels are two to three-fold higher and correlate well with number of antral follicles, assessed by ultrasound [35-37]. This may contribute to the failed dominant follicle selection in women with PCOS [38]. Obesity, insulin resistance, elevated androgens and AMH levels are probably related to the development of the large pool of antral follicles and ovarian volume in women with PCOS. Obesity and insulin resistance may enhance the follicular excess through the dysregulation of AMH or through the pathway of hyperandrogenemia [39].

Reproductive consequences of PCOS Decreased rates of ovulation and metabolic alterations in women with PCOS are associated with subfertility [40]. Miscarriage rates among women with PCOS are believed to be increased compared with normal fertile women, although supporting evidence is limited. Endometrial abnormalities, poor quality of oocytes, obesity and other intrinsic factors have been suggested to affect implantation and increase the risk for miscarriage [41-43]. However, most studies on miscarriage rates in women with PCOS have been conducted in populations who have undergone assisted reproductive treatment, and the miscarriage rate in unselected PCOS populations is largely unknown.

Insulin resistance, obesity and hyperandrogenism, common in PCOS women, are associated with increased risk of preeclampsia, gestational diabetes, preterm birth and stillbirth [42-44].

Obese women with PCOS, conceiving after in vitro fertilization (IVF) and intracytoplasmatic sperm injection (ICSI), have higher gonadotropin requirements [45], fewer oocytes, a higher miscarriage rate and lower live-birth rate than their non-obese counterparts [46]. Furthermore, neonates coming from these pregnancies have a significantly higher risk of admission to a neonatal intensive care unit and a higher perinatal mortality [47].

16

Ovarian reserve The ovarian reserve is the capacity of the ovary to produce fertilizable eggs. Measurements of factors thought to reflect ovarian reserve are used as an important means to predict outcome of assisted reproduction techniques. Variables used to estimate the ovarian reserve include basal or stimulated levels of FSH, AMH, inhibin B and the number of antral follicles and ovarian volume, assessed by means of transvaginal ultrasound [48]. Recently, the menstrual cycle length was also suggested to be independently associated with success rates in assisted reproduction. The chance of delivery after IVF/ICSI was almost doubled for women with a menstrual cycle length of more than 34 days compared with women with a menstrual cycle length less than 26 days [49].

A number of previous reports in young women with PCOS lend support for an increased ovarian reserve in these women compared with age-matched controls [50]. Recently, it was proposed that serum FSH assessments are inferior to measuring AMH in identifying women with reduced ovarian reserve, largely because AMH is menstrual cycle-independent [51]. The ovarian reserve of premenopausal women with PCOS has previously not been established.

AMH AMH is a dimeric glycoprotein, produced by granulosa cells from pre-antral

and antral follicles, which plays an important role in both ovarian primordial follicle recruitment and dominant follicle selection [52]. AMH levels essentially reflect the ovarian follicular pool, as AMH levels gradually decrease throughout reproductive life, are undetectable after the menopause [53, 54], and are unchanged throughout the menstrual cycle [55]. Several studies have reported higher levels of AMH in women with PCOS than in controls [35, 56, 57].

FSH Results for basal levels of FSH in PCOS have been less conclusive [58], which presumably is the result of the high variability of this hormone across the menstrual cycle and the necessity to restrict the comparisons to strictly cycle-matched samples [59]. Roberts and colleagues suggested that a history of increased basal levels of FSH and normal FSH levels in the beginning of IVF treatment, in women younger than 40 years, predicted lower oocyte yield, but did not significantly compromise the embryo quality. However, in women older than 40 years with prior elevation of basal FSH, both the ovarian response and embryo quality were compromised. Thus, high age in combination with elevated FSH may predict particularly poor IVF outcome

17

[60]. Interestingly though, Dahlgren and colleagues found lower levels of FSH in 50-year old women with a previous PCOS diagnosis compared with controls, in spite of the fact that the women with PCOS had been wedge-resected [61].

Inhibin B Inhibin B is a dimeric non-steroidal glycoprotein produced by granulosa cells, which inhibits production and/or release of FSH in the pituitary gland [62]. In normal ovulatory cycles the serum concentration of inhibin-B is inversely correlated with FSH concentrations and increases up to the mid-point of the follicular phase, when it reaches a maximum peak together with the mass of granulosa cells. Thereafter, the concentration decreases progressively and remains low in the luteal phase, except for a brief new elevation after the LH surge [63]. Findings on inhibin B and ovarian reserve are more discrepant [64, 65] and its use is limited, presumably because of strong menstrual cycle-dependency.

Number of antral follicles and ovarian volume Polycystic ovaries are larger and contain more antral follicles, as assessed by transvaginal ultrasound [66, 67]. An important biopsy study indicated a much increased density of follicles at primary stages in polycystic ovaries compared with normal ovaries, suggesting that women with polycystic ovaries may actually be endowed with a larger ovarian reserve at birth [34].

Metabolic consequences of PCOS PCOS is associated with insulin resistance and hyperinsulinism, type 2 diabetes, dyslipidemia and possibly cardiovascular disease [68].

Insulin resistance Studies over the past two decades have indicated that insulin resistance is a pathogenic characteristic feature of the polycystic ovary syndrome, particularly in obese individuals [69].

Insulin resistance, i.e. impaired stimulation of glycogen formation in all major target tissues, such as skeletal muscle, liver, kidney, and adipose tissue, and dysfunction of beta-cells, i.e. inability to compensate for insulin resistance, characterize type 2 diabetes [70]. Approximately 40 % of newly diagnosed type 2 diabetes subjects already suffer from clinical macroangiopathy at the time of diagnosis [71]. Thus, health care providers and groups at risk should be aware of the fact that PCOS, which is a pre-diabetic state, is not harmless. Beck-Nielsen and Groop presented the following model for the development of type 2 diabetes: Insulin resistance is the trigger which increases the insulin demands thereby unmasking the

18

defect in the beta-cell. A normal beta-cell may compensate for the insulin resistance by increasing the insulin secretion while the dysfunctional beta-cell in the long term can not, thus resulting in hyperglycemia [71]. Insulin secretion, especially first phase, tends to be increased rather than decreased in the pre-diabetic phase and is appropriate for (i.e. correlated with) the level of insulin resistance [72].

Insulin resistance in women with PCOS is closely associated with abdominal obesity and hyperandrogenism [73-76] at a lower BMI than healthy controls [77, 78]. The heterogeneous nature of the PCOS population is, however, reflected by more conflicting data on insulin sensitivity in lean women with PCOS, where some studies indicate that they have reduced insulin sensitivity [79-83], whereas others fail to find differences in insulin sensitivity between lean women with PCOS and weight-matched controls [84-87]. Some studies indicate that patients without hyperandrogenic features do not develop insulin resistance [88, 89].

The diagnosis of insulin resistance is based on results of tolerance test such as the oral glucose tolerance test (OGTT), the euglycemic-hyperinsulinemic glucose clamp, or calculations based on peripheral basal fasting insulin and glucose such as the homeostatic model assessment of IR (HOMA-IR) [90], the quantitative insulin sensitivity check index (QUICKI) [91], or the Matsuda insulin sensitivity index, calculated from OGTT values [92]. The Matsuda index represents both hepatic and peripheral tissue sensitivity to insulin and is highly correlated with insulin sensitivity parameters of the euglycemic-hyperinsulinemic glucose clamp [92].

Several serum markers have been tested in order to simplify the diagnosis

of insulin resistance. For instance, hyperinsulinemia results in increased circulating levels of free IGF-1 and reduction in IGFBP-3 [93]. Both insulin and IGF-1 receptor genes are expressed in human thecal cells and can stimulate steroid production there [94]. Free IGF-1 is a potent regulator of cell growth [95]. The actions of IGF-I is regulated by six circulating bindings proteins [96]. IGFBP-3 has been demonstrated in follicular fluid of normally cycling women and increasing FSH levels during maturation of a dominant follicle inhibits basal secretion of IGFBP-3 [97].

Early insulin response Early insulin response is an expression of pancreatic beta-cell function. Beta-cell function can be estimated by the early phase insulin secretion in oGTT (expressed as the insulinogenic index) or by the intravenous glucose tolerance test (IVGTT) (expressed as the insulin peak, insulin increment, areas under the curves (AUC) for insulin and glucose) [98].

The Botnia clamp is an intravenous glucose tolerance test followed by a euglycemic-hyperinsulinemic clamp and provides reliable and independent measures of insulin sensitivity and beta-cell function during the same test

19

[99]. Studies in Pima Indians have revealed lowered insulin secretion in those with impaired glucose tolerance (IGT) or type 2 diabetes, but exaggerated early insulin secretion in Indians with normal glucose tolerance [100]. Similar results were obtained in PCOS women [85]. In the long-term follow-up of the off-springs of parents with type 2 diabetes, subjects who developed type 2 diabetes within a 10-20-year period presented with higher insulin responses to glucose, especially first phase of insulin secretion, compared with subjects who continued to be normoglycemic [72]. Thus, an enhanced early insulin response, which is not fully explained by insulin resistance, may be a feature of subjects prone to insulin resistance and type 2 diabetes [72, 85, 100].

Diabetes in women with PCOS Prevalence rates for glucose intolerance are reported to vary between 38 - 45 % in younger women with PCOS but these studies have included subjects who predominantly are overweight or obese [101, 102]. In lean women with PCOS, approximately 12 % are reported to display glucose intolerance (IGT or type 2 diabetes) [102]. In PCOS women from the Mediterranean region, 15.7 and 2.5% displayed impaired glucose tolerance and type 2 diabetes, respectively [103].

Metabolic disturbances in PCOS women The metabolic syndrome (MetS) is a cluster of metabolic disturbances that increases the risk of cardiovascular disease (CVD) and diabetes. The metabolic syndrome has several definitions and should not be regarded as a syndrome per se. According to a scientific statement of the American Heart Association (AHS) and the US National Cholesterol Education Program/Adult Treatment Panel III (NCEP/ATPIII), the metabolic syndrome is present if three of following five criteria are fulfilled: elevated waist circumference (� 88 cm), elevated triglycerides (� 1.7 mmol/L or on drug treatment for elevated triglycerides), reduced high density lipoprotein cholesterol (HDL) (< 1.1 mmol/L or on drug treatment for reduced HDL), elevated blood pressure (� 130 mm Hg systolic blood pressure or � 85 mm Hg diastolic blood pressure or on antihypertensive drug treatment in a patient with a history of hypertension), and elevated fasting glucose (� 5.6 mmol/L or on drug treatment for elevated glucose) [104]. Prevalence of the metabolic syndrome in PCOS women vary between 11 to 47.3 % in various studies [68, 105-111]. A considerable portion of women with PCOS fulfill the criteria for the metabolic syndrome in view of a higher reported incidence of hypertension, dyslipidemia, visceral obesity, insulin resistance and hyperinsulinemia in this population [106].

20

Ambulatory day-time systolic and mean arterial blood pressures has been described to be higher in women with PCOS compared with women with normal ovaries and regular menstruation, even when adjusted for BMI [112, 113]. Abnormal lipid profile was observed in 70 % of PCOS women [114], including elevated levels of cholesterol, triglycerides, and low-density lipoprotein cholesterol (LDL) and reduced levels of HDL-cholesterol and apolipoprotein A-1 [113, 115, 116]. Obesity has an important influence on the lipid profile [22, 117], but also lean PCOS women have been reported to display abnormal lipid levels [118]. Waist circumference is the best anthropometric correlate of the amount of visceral adipose tissue and is advised to be used in the assessment of cardiovascular risks [119] and insulin resistance [120].

Cardiovascular consequences of PCOS Cardiovascular disease is the leading cause of death in developed countries[121]. Most cardiovascular events in women are caused by coronary heart disease (CHD) that is most commonly presented by angina pectoris [122]. The major identified risk factors for CHD in women are tobacco use, hypertension, diabetes mellitus, dyslipidemia with elevated levels of fasting total cholesterol, and LDL, as well as low levels of HDL, obesity, particularly of abdominal type, sedentary lifestyle, and atherogenic diet [123].

Multiple studies have indicated that women with polycystic ovary syndrome have several risk factors for cardiovascular disease, including centripetal obesity, hypertension, type 2 diabetes mellitus and dyslipidemia [124-128]. However, it remains unclear whether women with PCOS actually experience an increased number of cardiovascular events compared to an appropriate reference population [129].

Endothelial function in women with PCOS Endothelial dysfunction is an early event in the development of atherosclerosis, preceding plaque formation and clinical disease. Endothelial function has been evaluated in PCOS patients, mostly by measurement of post-ischemic flow mediated dilatation of the brachial artery with high-resolution ultrasonography but also by measurements of arterial stiffness using pulse wave velocity [130-136]. However, these studies have mainly been conducted in young patients with PCOS and no long-term follow up studies on endothelial function in older PCOS patients have been conducted.

21

Prior long-term studies in women with PCOS

Reproductive outcome Very few studies assessing the long-term reproductive outcome and ovarian reserve in older women with previously confirmed PCOS have been conducted. The typical ultrasound features of PCOS appear to diminish with increasing age. In a Finnish population-based cross-sectional study, polycystic ovaries were prevalent in 6 % of healthy women 34 years of age and in 10 % of women of similar age who fulfilled criteria for the metabolic syndrome. In women 44 years of age, the prevalence of polycystic ovaries was 2 % and 5 % in healthy controls and patients with the metabolic syndrome, respectively [137].

Although studies on reproductive performance in older PCOS patients are few, some studies have indicated that menstrual cycles are normalized with increasing age, at least in some of the patients [61, 138]. This also holds true for subjects who have not undergone ovarian wedge resection [61].

Cardiovascular disease Two relatively small long-term follow-up studies of PCOS patients with prior history of ovarian wedge resection have indicated that these patients suffer an increased risk of myocardial infarction during the menopause [139, 140] and that they have an increased prevalence of coronary artery disease already in the perimenopausal period [140]. Register-based long-term follow-up studies have indicated an increased prevalence of cerebrovascular disease [141], however, morbidity and mortality from coronary heart disease among women with PCOS was not as high as previously predicted [141, 142]. More recently, postmenopausal women with a retrospective diagnosis of PCOS were reported to have more angiographic coronary artery disease and a lower cumulative 5-year cardiovascular event-free survival compared to women without clinical features of PCOS [143] . Furthermore, postmenopausal mothers of PCOS patients were demonstrated to have a higher prevalence of cardiovascular events than controls [144].

Insulin resistance and diabetes The, above cited, long-term follow-up studies of women with PCOS also indicated an increased risk of type 2 diabetes in the menopause [61, 140]. Longitudinal studies of women with PCOS with focus on age-related changes in plasma androgens, insulin sensitivity and insulin secretion are lacking. Also, no studies on the long-term risks for developing type 2 diabetes and impaired glucose tolerance among PCOS patients diagnosed according to Rotterdam criteria have been performed.

22

AIMS

The aims of this thesis were: to describe the reproductive outcome in middle-aged women with a

previous diagnosis of PCOS, both in terms of live birth rate, pregnancies and of miscarriages (paper I);

to assess the remaining ovarian reserve in middle-aged women with a

previous diagnosis of PCOS compared with age-matched controls (paper I); to investigate pulse wave reflection which depends on stiffness in the

aorta and resistance in lower extremities, and endothelial function in resistance arteries in women previously diagnosed with PCOS in comparison to age-matched healthy controls (paper II);

to examine changes over time in plasma androgen concentrations, early

insulin response and insulin sensitivity of women with a previous diagnosis of PCOS (paper III);

to examine glucose tolerance and insulin sensitivity in middle-aged

women previously diagnosed with PCOS in comparison to age-matched healthy controls (paper IV).

23

Material and methods

Patients PCOS diagnosis Eligible patients were identified by the out-patient register at Uppsala University Hospital. Inclusion criteria for the study were diagnosis of polycystic ovary syndrome between 1987 and 1995 and age at the follow-up > 35 years. As the emphasis in this study was to follow patients with a PCOS diagnosis made by transvaginal ultrasound the medical records at the time of diagnosis were screened to confirm the diagnosis and to verify the use of transvaginal ultrasound for the diagnosis. Only subjects with a diagnosis of PCOS according to the Rotterdam criteria [7] were included and one of the features had to be polycystic ovaries on ultrasound examination [59, 66]. In addition to the ultrasound criterion, one of the following two features had to be present for the PCOS diagnosis 1) oligomenorrhoea, with eight or fewer menstruations in the previous 12 months, or amenorrhea 2) clinical and/or biochemical signs of hyperandrogenism such as testosterone > 2.7 nmol/l, elevated dehydroepiandrosterone (DHEAS), free androgen index (FAI) > 5.0, or hirsutism (> 7 on the Ferriman and Gallway scale). The PCOS diagnosis also implied that no evidence of thyroid disease (normal thyroid-stimulating hormone (TSH)), adrenocortical dysfunction (normal 17-hydroxyprogesterone), or hyperprolactinemia (prolactin < 30 �g/l) was present at diagnosis. Subjects with a PCOS diagnosis based on clinical history and laboratory findings but without transvaginal ultrasound examination were not included.

Healthy controls were selected from population registrars. For each PCOS patient, three healthy controls residing in Uppsala County and born during the same month as the index patients were selected. The control women were invited by letter, and in case the first control did not respond to the invitation, the second (and third) control subject was subsequently invited. Healthy control status was assured by absence of polycystic ovaries on transvaginal ultrasound. Furthermore all control subjects denied a prior history of oligomenorrhoea or amenorrhea (lasting more than three months).

24

The patients and controls gave a written informed consent and the Independent Ethical Review Board at Uppsala University, Sweden approved the study.

In all, 174 PCOS patients had been previously identified with a PCOS

diagnosis. Of these, three patients (1.7%) were dead, 12 patients had emigrated (6.9 %), and 14 patients (8.0 %) were living too far from Uppsala to be invited. Hence 145 patients (83.3 % of the identified patients) were invited to participate in the study.

Among the 145 invited PCOS patients, 84 women (57.9 %) attended the clinical examination and seven women (4.8 %) filled out the questionnaires but were unable to visit the clinic. Among PCOS patients still living in the Uppsala area 61 women (66.3 %) attended the clinical examination and three patients (3.2 %) filled out the questionnaires, whereas the corresponding figures for PCOS patients living outside the Uppsala area were 23 (43.4%) and four (7.5 %), respectively. At the time of the original diagnosis, 58 (40.0 %) of invited PCOS patients had PCO and oligomenorrhoea only, 8 (5.5 %) had PCO and hirsutism only, and 79 (54.5 %) had PCO, oligomenorrhoea and hirsutism. Eighty-seven age-matched control women participated in the study.

The number of patients included in the four studies is described in figure 2. In the first paper all women, including those who only filled out the questionnaire were included. In paper II, only 67 PCOS patients participated in the pulse wave analyses due to logistic reasons. According to the protocol, hyperinsulinemic euglycemic clamp or oral glucose tolerance test should be performed in addition to the pulse wave analyses but these measurements had to be performed on different days. Both analyses required that patients should be fasting, and terbutaline administration results in increased blood glucose levels. For patients who needed to travel great distances it was thus not possible to perform all measurements. In essence, the 67 women who participated in this sub-study were the ones who could come in to the clinic twice, i.e. women still residing in Uppsala County. In paper III, only women who previously had been investigated with hyperinsulinemic euglycemic clamp and intravenous glucose tolerance test were included. Finally, in paper IV all women who came to the health care investigation were included.

There were no differences in parity between participating and non-participating PCOS patients, for details see paper I. Also, no differences in metabolic variables were found between patients who participated and those who did not participate in paper III.

25

Figure 2. Patients participating in the study I-IV.

Design Paper I

Long-term follow-up study describing the reproductive outcome in middle-aged women with a previous diagnosis of PCOS, both in terms of live birth rate, pregnancies, and miscarriages. A secondary aim was to assess the remaining ovarian reserve in women with PCOS compared with age-matched controls. Paper II Long-term follow-up study investigating pulse wave reflection (baseline aortic augmentation index (AI)) which depends on stiffness in aorta and resistance in lower extremities, and endothelial function in resistance arteries (expressed as change in radial AI after terbutaline provocation) in women previously diagnosed with PCOS in comparison to age-matched healthy controls, randomly selected from the general population. Paper III Long-term follow-up study examining changes over time in plasma androgen concentrations, early insulin response and insulin sensitivity in women with a previous diagnosis of PCOS.

PCOS women 174 identified

145 invited 84 examined + 7 questionnaires

Paper I 91 PCOS women

7 questionnaire-only 87 controls

Paper II 67 PCOS women

66 controls

Paper III 34 PCOS women

4 of them metabolic and anthropometric

measures only

Paper IV 84 PCOS women

87 controls

26

Paper IV Long-term follow-up study examining glucose tolerance and insulin sensitivity in middle-aged women previously diagnosed with PCOS in comparison to age-matched healthy controls, randomly selected from the general population.

Methods All subjects consenting to participate attended a health examination at the department of Women’s and Children’s Health, Uppsala University Hospital. As the reproductive status was variable among subjects (pre-menopausal with or without hormonal contraception, postmenopausal with or without hormone therapy (HT), the visit was not scheduled according to menstrual cycle phase. However, records were kept on last menstrual period and transvaginal ultrasound examination together with assays of estradiol, FSH and LH confirmed the cycle phase in subjects without hormonal contraception.

All blood samples for the endocrine measurements were obtained in a standardized manner between 08.00 h – 09.00 h during rest in the supine position and after an overnight fast.

Subjects were weighed on an electronic scale. The subjects wore light

clothes and no shoes and weight was measured to the nearest 0.2 kg. Height without shoes was measured to the nearest centimeter. Body mass index was calculated as weight (kg) divided by height (m2). Waist circumference was measured midway between the lower rib margin and iliac crest. The hip was measured at the maximum circumference over the buttocks and to the nearest 0.0 or 0.5 cm. Blood pressure was measured twice in every person after a five-minute rest in a sitting position with digital manometer. The mean value of the two measurements was used in the study. The free androgen index was calculated from testosterone/ SHBG x 100. Paper I The ultrasound examination The ultrasound examination was performed with a 7 MHz transvaginal probe by either of two physicians (Miriam Hudecova and Inger Sundström Poromaa). The ultrasound measurements were obtained in real-time

according to a standardized protocol, with examination of the ovaries made under the highest possible magnification. After the longest medial axis of the

ovary had been determined, the second dimension was measured, and then

27

the vaginal probe was rotated 90 degrees to obtain the third dimension. Ovarian volume was calculated in the largest ovary according to a simplified formula for an ellipsoid (0.523 x length x width x thickness [66]. The ovary was scanned in both longitudinal and transverse cross-section from the inner to the outer margins to count the total number of follicles. All follicles, antral and growing, were counted.

Questionnaire The participants filled out a questionnaire on reproductive health, including obstetric history (previous pregnancies, previous live births), infertility history (duration of infertility, infertility treatments and infertility treatment in relation to every live birth), menstrual cycle history (menstrual frequency during the last 12 months), treatment, and ongoing hormonal treatment (any treatment with ovarian steroids such as oral contraceptives, progestagens, hormonal IUD, postmenopausal hormone therapy).

Assays Serum concentrations of testosterone, SHBG, FSH, LH were analysed by

solid-phase chemiluminescent immunometric assays and estradiol was measured by competitive immunoassay, using commercial kits obtained from Siemens Medical Solutions, Germany. Detection limits for the estradiol essays was 73.0 pmol/l and for testosterone 0.5 nmol/l. Total coefficients of variation varied between 5.7 and 10.6% for these analyses. The serum concentrations of AMH were determined using enzymelinked immunoassay kits, from Immunotech Beckman Coulter Company, France. The detection limit for AMH was 0.07 pmol/l and levels below this limit were considered undetectable. Total coefficient of variation was 12.3% for the AMH analyses.

Paper II Pulse wave analyses Endothelium-dependent vasodilation (EDV) in resistance arteries was evaluated by aplanation tonometry of the radial artery, where the peripheral pulse pressure waveform in response to adrenergic �-2 receptor agonist stimulation may be captured and used as a valid measure of EDV [145, 146]. This method has been thoroughly evaluated and a reduced response to �-2

agonists has been demonstrated in patients with insulin resistance, hypercholesterolemia, coronary heart disease, diabetes mellitus, and the metabolic syndrome [146-149].

A micromanometer tipped probe (Sphygmocor, AtCor Medical Pty Ltd,West Ryde, Sydney, Australia) was applied to the skin surface overlying the radial artery and the peripheral radial pulse wave was continuously recorded. The

28

mean values of 10 pulse waves were used for analyses. Recordings were regarded as satisfactory if the variations in the systolic peak and the diastolic peak were 5% or below. The maximal systolic peak and the reflected waves were identified by the calculations of the first and second derivative of the pulse curve. After a baseline recording, terbutaline was subcutaneously administered (0.25 mg in the upper part of the arm), and a re-evaluation of the pulse wave was performed after 15 and 20 minutes. The maximal change occurring at either 15 or 20 minutes was used for calculations [145]. Each test was performed and evaluated by two examiners (Miriam Hudecova and Inger Sundström Poromaa) throughout the study. Intra- and interobserver coefficient of variation were 3.4% and 2.3 %, respectively. For this study we used two measures of augmentation index (AI): 1) AI-radial which is the augmentation index derived from the first reflected wave using the radial artery recording [147], and 2) AI-aorta which corresponds to the augmentation index obtained after transformation of the radial pulse wave to the corresponding central pulse wave by use of a validated transfer function [148]. These variables are reported as baseline values and AI-radial also as relative change from baseline after terbutaline administration. Thus, a large reduction of the pulse wave in response to terbutaline indicates a good response.

Figure 3. Radial and derived aortic pulse waveform obtained from pulse wave analysis. AI-radial = b/a, AI-aortic = b-a/a.

Assays Lipid variables, fasting plasma glucose and fasting insulin were measured

by standard laboratory techniques. Serum concentrations of testosterone and SHBG were analyzed by solid-phase chemiluminescent immunometric assays and. Homeostatic Model Assessment of Insulin Resistance Index

29

(HOMA-IR) was defined as: fasting blood glucose × serum insulin/22.5 [150]. Paper III Botnia clamp At the index assessment, an IVGTT and a hyperinsulinemic euglycemic clamp were performed on two different days. For the present study, the Botnia clamp technique was used. The Botnia clamp is an intravenous glucose tolerance test followed by a euglycemic-hyperinsulinemic clamp and provides reliable and independent measures of insulin sensitivity and beta-cell function during the same test [99]. Furthermore, the Botnia clamp technique provides measures of the early insulin response and insulin sensitivity comparable to measurements made on separate days [99].

After an injection of 300 mg/kg body weight of D-glucose, samples were drawn at -10, 2, 4, 6, 8, 10, 20, 40, 60 and 90 min for analyses of plasma insulin and glucose (time from initiation of the injection of glucose). The insulin peak was calculated as the mean value of the plasma insulin concentrations at 4, 6 and 8 min, and the insulin increment was calculated by subtracting the mean fasting plasma insulin concentrations from the peak value. The AUC for insulin and glucose were calculated as the deviations from the basal value integrated over the sampling time. The IVGTT procedures and calculations were similar to the index assessment [85].

After 90 min, a priming dose of insulin was given followed by an infusion (56 mU/m2) of short-acting human insulin (Actrapid; Novo Nordisk, Bagsvaerd, Denmark) for 120 min. Euglycemia (5.1 mmol/l plasma glucose) was maintained by measuring the plasma glucose every 5 min and adjusting the rate of infusion of a 20% glucose solution accordingly. Blood samples for insulin analyses were drawn at 60, 80 and 120 min after the start of the clamp. The coefficient of variation for the steady-state plasma glucose concentration for a single clamp was 5.5 %. The insulin sensitivity index (M/I) was calculated by dividing the amount of glucose infused during the last 60 min of the clamp with the mean steady state insulin concentration during the same time (milligram per kg body weight-¹ min-¹ x mU/L x 100). The insulin infusion rate and target plasma glucose during the clamp were similar to the index assessment [85].

Assays Apolipoprotein A-1, apolipoprotein B, triglycerides, total cholesterol,

high density lipoprotein cholesterol (HDL-cholesterol), low density lipoprotein cholesterol (LDL-cholesterol), high sensitive C-reactive protein (hs-CRP), and glucose, were analyzed on an Architect ci8200 (Abbott Laboratories, Abbott Park, IL, USA). The total CV were 0.9% at 2.25 g/l for

30

ApoA-I, 1.2% at 1.73 g/l for Apo B, 1.1% at 0.9 mmol/L, 0.5% at 5.7 mmol/L for total cholesterol, 2.3% at 1.0 mmol/L for HDL-cholesterol, 1.0% at 2.0 mmol/L for LDL-cholesterol, for triglycerides, 0.8% at 8 mg/L for hs-CRP, and 1.0% at 4.4 mmol/L for glucose.

Insulin, C-peptide and sex hormone binding globulin (SHBG) were analyzed on Modular E170 (Roche Diagnostics, Mannheim, Germany). The total imprecision of the methods, expressed as coefficient of variation (CV), were 1.6% at 7.0 mU/L for insulin, 1.5% at 43 nmol/L for SHBG, and 1.5% at 0.45 nmol/L for C-peptide. The analytical method for SHBG was changed in the time-interval between assessments, resulting in 13 % higher values at the follow-up investigation. For statistical comparisons, the SHBG values at the index assessment were transformed using a correction factor of 1.13. The SHBG values from the index assessment which are presented in text and tables were also transformed.

Testosterone was analysed by the same radioimmunoassay at both occasions (Diagnostic Products Corp., Los Angeles, CA). Within- and between-assay coefficients of variation for testosterone were 5.5 % and 7.0 %.

HOMA-IR was defined as: fasting blood glucose × serum insulin/22.5 [150]. Paper IV Oral glucose tolerance test With the exception of women with previously known diabetes and women who were subjected to euglycemic hyperinsulinemic clamp, all women performed a 75-g OGTT after over-night fasting. Blood samples for measurement of serum insulin and plasma glucose were obtained at 0, 30, and 120 minutes.

Assays Apolipoprotein A-I (ApoA-I), Apolipoprotein B (ApoB), triglycerides,

high-sensitivity C-reactive protein (hs-CRP) and glucose, were analyzed on an Architect ci8200 (Abbott Laboratories, Abbott Park, IL, USA). The total imprecision (coefficient of variation, CV) of the instrument for the analytes were 0.9% at 2.25 g/l for ApoA-I, 1.2% at 1.73 g/l for ApoB, 1.1% at 0.9 mmol/L for triglycerides, 0.8% at 8 mg/L for hs-CRP, and 1.0% at 4.4 mmol/L for glucose.

Insulin, C-peptide and SHBG were analyzed on Modular E170 (Roche Diagnostics, Mannheim, Germany), IGFBP-3 was analyzed on an Immulite 2500 (Siemens Healthcare Diagnostics, Deerfield, IL, USA), and proinsulin was analyzed with a sandwich ELISA (Mercodia, Uppsala, Sweden). The total imprecision of the methods were 1.6% at 7.0 mU/L for insulin, 1.5% at

31

43 nmol/L for SHBG, 1.5% at 0.45 nmol/L for C-peptide, 3.6% at 0.85 mg/L for IGFBP-3, and 4.1% at 21.7 pmol/L for proinsulin.

The Matsuda insulin sensitivity index, which is an index of whole-body insulin sensitivity was calculated according to this formula: (10,000/square root of [fasting glucose x fasting insulin] x [mean glucose x mean insulin during OGTT]) [92].

Beta-cell function was estimated by the early phase insulin secretion (the insulinogenic index) and calculated as [(30 min insulin - fasting insulin) / 30 min glucose] [98].

Impaired glucose tolerance and diabetes were defined using World Health Organization criteria.

Results Paper I

Reproductive outcome Among PCOS patients, 83 had attempted a pregnancy, 79 had become pregnant, and 72 had given birth at least once. There was no difference between PCOS patients and controls in number of women who had given birth. Among women who had attempted a pregnancy, 72 (86.7 %) of PCOS patients and 76 (91.6 %) of control subjects had given birth to at least one child. The overall rate of subjectively defined spontaneous conception was also high among PCOS patients. Among the 83 PCOS patients who had attempted a pregnancy, 56 (67.5 %) became spontaneously pregnant at least once. Among the PCOS patients who had given birth to a child, 53 (73.6 %) did so at least once following a spontaneous conception. Among PCOS patients who had attempted a pregnancy 63 (75.9%) had sought care for infertility and 56 (67.5%) patients had, at some point, been treated for infertility.

The rate of miscarriages was not increased in PCOS patients (PCOS

patients 0.6 ± 1.1 miscarriages vs. 0.6 ± 0.9 miscarriages in control subjects). Among women who conceived at least once, 36 (45.6 %) of PCOS patients and 41 (49.4 %) of control women had at least one miscarriage.

The majority of subjects were still premenopausal (87.9 % of PCOS

women and 80.5 % of controls). Pre-menopausal PCOS patients without hormonal contraception reported 8.8 ± 3.9 menstruations during the last year while control women reported 11.4 ± 2.2 menstrual periods during the last year, p < 0.01. The number of premenopausal PCOS patients without

32

hormonal contraception who had had eight or more menstrual bleedings during the last year was 38 (73.1%).

Ovarian reserve Mean ovarian volume and number of antral follicles in pre-menopausal PCOS patients without hormonal treatment were significantly higher than in corresponding control subjects, 9.5 ± 6.6 ml compared to 6.6 ± 4.1 ml (p < 0.001) and 11.7 ± 12.4 compared to 5.0 ± 2.4 (p < 0.001), respectively. Mean ovarian volume and number of antral follicles were also higher in pre-menopausal women with PCOS compared with controls, when the comparisons were restricted to women evaluated in the follicular phase, table I.

33

Table I. Mean ± SEM serum concentrations of ovarian volume, number of antral follicles, ovarian steroids and gonadotropins in 52 premenopausal PCOS patients and 56 premenopausal controls without hormone treatment.

Premenopausal women without hormone treatment

Premenopausal women in the follicular phase

PCOS

(n = 52)

Controls

(n = 56)

PCOS

(n = 16)

Controls

(n = 30)

Age, years 42.4 ± 4.5 41.5 ± 4.6 41.9 ± 3.8 42.7 ± 4.2

Ovarian volume, ml 9.5 ± 0.9** 6.6 ± 0.5 8.9 ± 1.3 6.5 ± 0.7

Antral follicles, n 11.7 ± 1.7*** 5.0 ± 0.3 10.1 ± 1.2*** 5.2 ± 0.4

Estradiol, pmol/l 514 ± 79* 710 ± 90 522 ± 156 734 ± 134

Testosterone, nmol/l 1.5 ± 0.1** 1.0 ± 0.1 1.6 ± 0.2* 1.0 ± 0.1

SHBG, nmol/L 50.4 ± 4.5* 64.0 ± 4.8 52.9 ± 6.9 63.0 ± 5.3

Free androgen index 5.1 ± 0.8** 2.0 ± 0.2 4.6 ± 1.2 2.0 ± 0.3

FSH, IU/L 6.2 ± 1.1* 9.6 ± 1.3 6.2 ± 0.8 9.7 ± 1.4

LH, IU/L 7.1 ± 0.7 10.4 ± 1.7 7.5 ± 1.1 12.1 ± 2.8

AMH, pmol/l 39.9 ± 6.1*** 15.7 ± 2.1 37.2 ± 7.9*** 9.9 ± 1.9

*p < 0.025 in comparison with respective control group, Mann-Whitney U test.

**p < 0.01 in comparison with respective control group, Mann-Whitney U test.

***p < 0.001 in comparison with respective control group, Mann-Whitney U test. Free androgen index was increased in PCOS patients compared to their

age-matched controls. AMH serum concentrations were significantly elevated in women with PCOS, also in the subgroup of women who were evaluated in the follicular phase. Premenopausal PCOS patients had lower levels of FSH and estradiol than control subjects, table I.

Ovarian volume and number of antral follicles were significantly

correlated with serum concentrations of AMH, but not to serum concentrations of FSH or estradiol, table II.

34

Table II. Correlation between ovarian volume and follicle count and reproductive hormones in 52 premenopausal PCOS patients and 56 premenopausal control subjects without hormone treatment.

Follicles count

E2 T SHBG FAI FSH AMH

Ovarian volume, ml 0.62** 0.09 0.11 -0.10 0.17 0.03 0.35**

Antral follicles, n -0.10 0.09 -0.18 0.19 -0.15 0.54**

Estradiol 0.04 0.04 -0.08 0.06 -0.12

Testosterone -0.22* 0.80*** -0.17 0.25**

SHBG -0.53*** -0.03 -0.20

FAI -0.17 0.30**

FSH -0.22*

LH -0.04

T = testosterone, E2 = estradiol, *p < 0.025 Pearson’s Correlation Coefficient, **p < 0.01 Pearson’s Correlation Coefficient, ***p < 0.001 Pearson’s Correlation Coefficient

The relationship between ovarian reserve, PCOS status and age is

depicted in figure 2 in paper I. Multiple regression analyses indicated that ovarian volume, number of antral follicles and AMH serum concentrations declined with age in both groups (p < 0.001, respectively) and that PCOS patients had higher values for the above measures across ages compared to controls (p < 0.001 for each ovarian reserve variable). Covariance analyses indicated no group by age interactions for the ovarian reserve measures, indicating that the age-related decline was similar between groups (ovarian volume F(1,121) = 0.76; number of antral follicles F(1,121) = 1.13: AMH F(1,123) = 1.74; p = 0.055)

Paper II In all, 67 PCOS patients and 66 control subjects were included in the

study. In two PCOS patients and two control subjects, satisfactory pulse wave recordings could not be obtained. In the remaining subjects, the

35

software did not properly recognize either AI-radial or AI-aorta in three control subjects.

At the index assessment 32 (47.8 %) of patients presented with PCO and oligomenorrhoea, 4 (6.0 %) had PCO and hyperandrogenism, and 31 (46.3 %) had PCO, oligomenorrhoea, and hyperandrogenism. At the follow-up investigation 22 (32.8 %) patients still fulfilled and 27 patients (40.3 %) no longer fulfilled diagnostic criteria for PCOS. In 18 patients (26.9 %) PCOS criteria could not be assessed because of postmenopausal status, prior surgery or ongoing use of hormonal contraception.

Analysis of basic characteristics and major cardiovascular risk factors at the follow-up investigation revealed that PCOS patients had significantly increased weight, BMI, waist circumference and waist:hip ratio in comparison to control subjects. Likewise, systolic and diastolic blood pressures, heart rate, HOMA-IR and triglycerides were higher in PCOS patients than in control subjects.

None of the PCOS patients or control subjects reported prior cardiovascular events such as nonfatal myocardial infarction, coronary revascularization, nonfatal stroke, or coronary disease. There were no significant differences in frequency of family history of myocardial infarction or stroke, but prevalence of diabetes was twice as high in first-degree relatives of patients with PCOS as in those of control subjects.

Pulse wave analyses There was no difference in baseline AI-aorta between PCOS patients and control subjects, table III. Baseline AI-radial was increased and percent decrease in AI-radial following terbutaline administration was less pronounced in PCOS patients in comparison with control subjects, table III. The difference in the change in AI-radial following terbutaline administration remained when adjusted for use of combined oral contraceptives/HRT as well as postmenopausal status (� = 0.197, p < 0.05), and smoking (� = 0.191, p < 0.05). However, after adjustment for BMI, none of these differences between PCOS patients and healthy controls remained, table IV. The strongest independent explanatory factor for the variability in endothelial function was HOMA-IR, table IV.

36

Tabl

e II

I. M

ean

± SD

for P

ulse

Wav

e A

naly

sis V

aria

bles

in P

CO

S Pa

tient

s, Su

bgro

ups o

f PC

OS

Patie

nts,

and

Con

trol S

ubje

cts

A

ll PC

OS

patie

nts

(n

= 6

5)

PCO

S pa

tient

s w

ith p

ersi

stin

g sy

mpt

oms

at fo

llow

-up

inve

stig

atio

n (n

= 2

1) a

PCO

S pa

tient

s w

ith re

solv

ed

sym

ptom

s at

follo

w-u

p in

vest

igat

ion

(n =

26)

a

PCO

and

ol

igom

enor

rhoe

a at

inde

x as

sess

men

t

(n =

32)

PCO

, ol

igom

enor

rhoe

a an

d hi

rsut

ism

at

inde

x as

sess

men

t

(n =

29)

a

Con

trol s

ubje

cts

(n

= 6

4)

BM

I 27

.4 ±

5.5

b

27.3

± 5

.9 b

25.9

± 4

.8

26.3

± 3

.3

28.5

± 5

.4b

25.5

± 3

.6

Bas

elin

e A

I-ra

dial

, %

84

± 1

4c 81

± 1

4 84

± 1

0 86

± 1

2 80

± 1

6 81

± 1

5

Bas

elin

e A

I-ao

rta, %

27

± 9

25

± 9

26

± 8

29

± 8

26

± 9

25

± 1

0

% c

hang

e A

I-ra

dial

-2

2 ±

14 c

-21

± 10

b 25

± 1

8 -2

3 ±

13

-21

± 16

-2

8 ±

17

a Dis

crep

anci

es in

num

bers

, in

com

paris

on to

the

text

, is b

ecau

se sa

tisfa

ctor

y pu

lse

wav

e re

cord

ings

cou

ld n

ot b

e ob

tain

ed in

two

PCO

S pa

tient

s b S

igni

fican

tly d

iffer

ent f

rom

con

trol s

ubje

cts,

one-

way

AN

OV

A w

ith T

ukey

Hon

estly

Sig

nific

ance

Diff

eren

ce p

ost h

oc te

st, p

< 0

.05.

c S

igni

fican

tly d

iffer

ent f

rom

con

trol s

ubje

cts i

n un

adju

sted

ana

lyse

s, p

< 0.

05. F

ollo

win

g ad

just

men

t for

BM

I, no

ne o

f the

se d

iffer

ence

s re

mai

ned.

37

Table IV. Results of a stepwise multivariate linear regression analysis of variables associated with percent change in AI-radial following terbutaline.

Variable Model 1

��

Model 2

�

Model 3

�

Model 4

�

R2 .04 .13 .18 .23

PCOS diagnosis 0.19* 0.14 0.14 0.12

Free androgen index 0.04 -0.03 -0.08 -0.06

Age 0.01 0.10 0.07

BMI 0.29** 0.13 0.07

HOMA-IR 0.29* 0.31*

Smoking 0.15

Current treatment for diabetes, hypertension or hyperlipidemia 0.14

Free androgen index, age, BMI and HOMA-IR entered as continuous variables.

��standardized regression coefficient. �

Pulse wave analyses in relation to PCOS phenotype at index assessment There were no significant differences in the pulse wave variables between PCOS patients who had PCO and oligomenorrhoea and those who had PCO, oligomenorrhoea, and hyperandrogenism at the index assessment, nor between any of these subgroups and control subjects, table III.

Pulse wave analyses in relation to PCOS phenotype at follow-up investigation There was no difference in pulse wave variables between PCOS patients who still fulfilled PCOS criteria at the time of the follow-up and those who no longer did, table 3. However, percent reduction in AI-radial following

38

terbutaline administration was significantly lower in PCOS patients with persisting symptoms in comparison to control subjects, table 3. Again, this finding did not remain when adjusted for BMI, table III.

Paper III In all, 73 women with PCOS had participated in the first study using IVGTT and hyperinsulinemic, euglycemic clamp between 1987 and 1995. Of these, 13 subjects were lost to follow up because personal data from the initial study was insufficient for tracing or because they had emigrated or lived too far away from the Uppsala region. Hence, 60 subjects were invited for the follow-up study. Of these, 30 (50.0 %) consented to participate in the Botnia clamp investigation, whereas four subjects only consented to blood sampling. Women who consented to participate in the follow-up investigation and non-participants did not differ on any of the metabolic measures at the index assessment, see paper III for details.

One PCOS patient had developed diet-treated type 2 diabetes since the

index assessment and one had an elevated fasting glucose. As results were not affected by the exclusion/inclusion of these two subjects, they were kept in the study. Three women were treated with combined oral contraceptives and one postmenopausal woman was on hormone replacement therapy. These subjects were excluded from analyses of plasma testosterone and SHBG. At the index assessment the distribution of lean, overweight and obese subjects were 15 (44.1 %), 10 (29.4 %) and 9 (26.5%), whereas corresponding figures at the follow-up investigation were 12 (34.3%), 6 (17.1%) and 17 (48.6%), respectively.

At the follow-up investigation 10 (29.4 %) women still fulfilled and 11 women (32.4 %) no longer fulfilled diagnostic criteria for PCOS. In 13 women (38.2 %) PCOS criteria could not be assessed because of postmenopausal status, prior surgery or ongoing use of hormonal contraception.

Metabolic and anthropometric variables over time Over time, women with PCOS displayed increased BMI, increased plasma SHBG concentration and increased diastolic blood pressure. In contrast, serum testosterone concentrations and free androgen index decreased from the time of index assessment to the follow-up investigation. There were no differences in systolic blood pressure or waist-hip ratio between the two time-points of assessments.

39

Insulin sensitivity and acute insulin response over time Insulin sensitivity expressed as M/I remained unaltered over time, table V. Similarly, insulin increment was unaltered between the two assessments. Other variables reflecting the early insulin response and glucose-stimulated

Table V. Difference in Botnia clamp parameters in PCOS patients between initial and follow-up evaluation

PCOS patients (n = 30)

At diagnosis

mean ± SD

Follow-up

mean ± SD

M/I60-120 6.7 ± 3.4 6.3 ± 2.6

Baseline insulin, mU/lc 11.0 ± 8.1 9.5 ± 7.9

Baseline glucose, mmol/lc 4.7 ± 0.6 4.8 ± 0.6

HOMA-IR 2.4 ± 2.0 2.1 ± 1.9

Insulin peak, mU/l 78.1 ± 55.7 60.2 ± 62.0a

Insulin increment, mU/l 67.7 ± 52.0 51.9 ± 59.4

AUC insulin 2425 ±1711 1823 ± 1711a

AUC glucose 281 ± 62 370 ± 118a

a Difference across time, p < 0.05 – 0.01

c Data available in 34 PCOS patients

insulin secretion, such as the insulin peak and AUC insulin decreased over time, table V. These results remained when the two subjects with type 2 diabetes were excluded from the analysis (data not shown). Also, there was no difference in M/I or insulin increment between women with resolved or unresolved PCOS (M/I; resolved PCOS 6.7 ± 2.3 vs. unresolved PCOS 7.6 ± 2.8, insulin increment; resolved PCOS 34.1 ± 20.0 mU/l vs. unresolved PCOS 48.4 ± 19.4 mU/l).

40

Insulin sensitivity at the follow-up investigation in relation to characteristics at the index assessment Comparisons of endocrine, metabolic and anthropometric variables at the index assessment, depending on current insulin sensitivity, are presented in table VI. PCOS patients who had M/I-levels in the lowest quartile at the follow-up investigation had already at the index assessment increased BMI, increased waist, higher plasma LDL-cholesterol and higher fasting plasma

Table VI. Differences in metabolic variables at baseline or across time between PCOS patients with insulin resistance or intermediate/normal insulin sensitivity at the follow-up.

Lowest M/I quartile at t2

(2.0 – 4.13)

n = 7

Remaining M/I quartiles at t2

(4.14 – 12.65)

n = 23

p-value

Age at t1 30.0 ± 7.0 25.7 ± 6.3 ns

BMI at t1, kg/m2 29.2 ± 2.8 25.4 ± 5.2 0.05

�BMI over time, kg/m2 4.8 ± 2.8 1.4 ± 3.9 0.05

Waist at t1, cm 92.7 ± 10.7 83.3 ± 12.8 0.05

FAI at t1 19.2 ± 13.4 16.3 ± 13.6 ns

Testosterone at t1, nmol/l 3.0 ± 1.5 2.8 ± 1.1 ns

SHBG at t1, nmol/l 17.7 ± 3.6 23.3 ± 13.6 ns

Total cholesterol at t1, mmol/l 5.2 ± 0.6 4.6 ± 0.7 ns

HDL-cholesterol at t1, mmol/l 1.0 ± 0.2 1.3 ± 0.3 ns

LDL-cholesterol at t1, mmol/l 3.7 ± 0.6 3.0 ± 0.7 0.05

Triglycerides at t1, mmol/l 1.50 ± 0.67 1.11 ± 0.80 ns

Insulin increment, IVGTT at t1 103 ± 68 56 ± 42 ns

Insulin at t1, mU/l 13.6 ± 6.3 8.9 ± 7.0 0.05

Glucose at t1, mmol/l 4.8 ± 0.7 4.6 ± 0.6 ns

41

insulin concentrations in comparison with subjects in the remaining M/I quartiles, table VI. Furthermore, PCOS patients who had M/I-levels in the lowest quartile at the follow-up investigation displayed a greater weight increase over time in comparison to patients with intermediate/normal insulin sensitivity at the follow-up. However, PCOS patients in the lowest M/I quartile had similar free androgen index, total testosterone and SHBG concentrations at the index assessment in comparison with patients with intermediate/normal insulin sensitive at the follow-up investigation.

Table VII. Variables associated with M/I at the follow-up

��

adjusted for age

Age -0.372*

WHR -0.502** -0.463*

FAI b -0.100 -0.154

Testosterone b 0.060 0.016

SHBG b 0.353a 0.308

CRP -0.565*** -0.498**

ApoB/ApoA1 -0.509** -0.460**

HDL 0.421* 0.456*

LDL -0.280 -0.232

Total cholesterol -0.162 -0.086

Triglycerides -0.541** -0.515**

C-peptide -0.425* -0.324*

a p = 0.064, b three patients with combined oral contraceptives and one postmenopausal woman on hormone replacement therapy were excluded from the analysis

Variables associated with insulin sensitivity at the follow-up investigation Table VII displays factors associated with M/I at the time of the follow-up investigation. Waist/hip ratio, hs-CRP, apoB/apoA1 ratio, HDL-cholesterol,

42

triglycerides, and C-peptide were all associated with insulin sensitivity even after adjustment for age. There was a trend towards a significant association with SHBG in the bivariate analysis (p = 0.064), which was lost when age was added to the model. Neither free androgen index nor total testosterone was associated with insulin sensitivity at the time of the follow-up investigation.

Paper IV 84 women with PCOS and 87 control subjects participated in paper IV. Of these, 49 performed OGTT, 32 were subjected to euglycemic hyperinsulinemic clamp/IVGTT (paper III), five subjects with known diabetes were not eligible for any of these interventions and four patients dropped out of the study prior to the clamp investigation. Six subjects participated in OGTT as well as clamp investigations. Among control subjects, 68 performed OGTT, 13 euglycemic hyperinsulinemic clamp/IVGTT (not reported in this thesis), and remaining six subjects dropped out. None of the controls participated in more than one tolerance test. Data from euglycemic hyperinsulinemic clamp/IVGTT are only used in this study for assessment of diabetes and impaired fasting glucose.

At the index assessment 36 (42.9 %) of women presented with PCO and oligomenorrhoea, 4 (4.8 %) had PCO and hyperandrogenism, and 44 (52.4 %) had PCO, oligomenorrhoea, and hyperandrogenism. At the follow-up investigation 31 (36.9 %) women still fulfilled and 30 women (35.7 %) no longer fulfilled diagnostic criteria for PCOS. In 23 women (27.4 %) PCOS criteria could not be assessed because of postmenopausal status, prior surgery or ongoing use of hormonal contraception.

Prevalence of the metabolic syndrome According to scientific statement of AHA and NCEP/ATPIII, the prevalence of MetS in PCOS women was 21.4 % and in healthy controls 6.9 %.

Prevalence of impaired glucose tolerance and diabetes Eighteen (21.4 %) PCOS patients had impaired glucose tolerance or diabetes at the follow-up which was significantly more common than in control subjects, p < 0.05. Two women with PCOS had developed type 1 diabetes (2.4 %), seven women had developed type 2 diabetes (8.3 %), eight women had IGT (9.5 %) and one had elevated fasting glucose (1.2 %). Four women with diabetes were currently treated with insulin. Corresponding figures for control subjects were one woman with type 2 diabetes (1.1 % diagnosed at euglycemic hyperinsulinemic clamp), two women with IGT (2.3 %), and one with elevated fasting glucose (1.1 %).

43

Metabolic and anthropometric parameters at the follow-up investigation Anthropometric measures in non-diabetic women with PCOS and their healthy counterparts are displayed in table VIII, metabolic parameters in table IX, and parameters from the OGTT in table X.

44

Tabl

e V

III.

Ant

hrop

omet

ric m

easu

res i

n al

l non

-dia

betic

PC

OS

patie

nts,

reso

lved

and

unr

esol

ved

PCO

S pa

tient

s, PC

OS

phen

otyp

ic su

bgro

ups

and

heal

thy

cont

rols

at t

he fo

llow

-up

inve

stig

atio

n.

A

ll PC

OS

patie

ntse

n

= 75

PCO

S pa

tient

s with

pe

rsis

ting

sym

ptom

s at

follo

w-u

pe

n =

27

PCO

S pa

tient

s with

re

solv

ed sy

mpt

oms

at fo

llow

-upe

n

= 27

PCO

S pa

tient

s with

hi

rsut

ism

, ol

igom

enor

rhea

and

PC

O a

t ind

ex

asse

ssm

ente

n =

40

PCO

S pa

tient

s with

ol

igom

enor

rhea

and

PC

O a

t ind

ex

asse

ssm

ente

n

= 32

Con

trol

subj

ects

n

= 86

B

MI,

kg/m

2

27.5

6 ±

5.59

a

28

.9 ±

6.0

b,c

25

.5 ±

4.0

29.2

± 5

.7 b

,d

26

.2 ±

5.2

25.6

± 4

.2

Wai

st, c

m

88.2

± 1

4.4 a

92

± 1

5b,

c 81

± 1

1 91

± 1

5 b,d

84 ±

13

82 ±

10

Wai

st:h

ip ra

tio

0.85

± 0

.09 a

0.

86 ±

0.0

9b

0.82

± 0

.08

0.85

± 0

.09 b

0.

84 ±

0.0

8 0.

81 ±

0.0

6 Fa

t per

cent

, %

36.6

± 9

.4 a

39.8

± 8

.9

31.3

± 8

.4

37.8

± 9

.4 b

34.6

± 9

.5

33.3

± 7

.4

Syst

olic

blo

od p

ress

ure,

m

m H

g 13

2 ±

20

129

± 19

12

9 ±

19

131

± 20

13

3 ±

22b

125

± 16

Dia

stol

ic b

lood

pre

ssur

e,

mm

HG

82

± 1

2 81

± 1

1 80

± 1

2 82

± 1

3 83

± 1

2 79

± 1

0

a Si

gnifi

cant

ly d

iffer

ent f

rom

con

trol s

ubje

cts,

Stud

ent’s

T-te

st (u

nadj

uste

d).

b Sig

nific

antly

diff

eren

t fro

m c

ontro

l sub

ject

s, A

NO

VA

pos

t-hoc

Tuk

ey H

SD, p

< 0

.05

– 0.

001

(una

djus

ted)

c S

igni

fican

tly d

iffer

ent f

rom

reso

lved

PC

OS

patie

nts,

AN

OV

A p

ost-h

oc T

ukey

HSD

, p <

0.0

5 (u

nadj

uste

d)

d Si

gnifi

cant

ly d

iffer

ent f

rom

PC

OS

patie

nts w

ith o

ligom

enor

rhoe

a an

d PC

O a

t ind

ex a

sses

smen

t, A

NO

VA

pos

t-hoc

Tuk

ey H

SD, p

< 0

.05

– 0.

001

(una

djus

ted)

e W

omen

with

dia

bete

s not

incl

uded

45

Tabl

e IX

. Met

abol

ic v

aria

bles

in a

ll no

n-di

abet

ic P

CO

S pa

tient

s, re

solv

ed a

nd u

nres

olve

d PC

OS

patie

nts,

PCO

S ph

enot

ypic

subg

roup

s and

he

alth

y co

ntro

ls a

t the

follo

w-u

p in

vest

igat

ion.

All

PCO

S pa

tient

se

n =

75

PCO

S pa

tient

s with

pe

rsis

ting

sym

ptom

s at

follo

w-u

pe

n

= 27

PCO

S pa

tient

s with

re

solv

ed sy

mpt

oms

at fo

llow

-upe

n =

27

PCO

S pa

tient

s with

hi

rsut

ism

, ol

igom

enor

rhea

and

PC

O a

t ind

ex

asse

ssm

ente

n =

40

PCO

S pa

tient

s with

ol

igom

enor

rhea

and

PC

O a

t ind

ex

asse

ssm

ente

n

= 32

Con

trol s

ubje

cts

n

= 86

Fa

stin

g in

sulin

, mU

/l 9.

78 ±

7.6

0 11

.6 ±

10.

2 8.

0 ±

4.8

10.5

± 8

.2

10.9

± 8

.9a

7.5

± 5.

8 Fa

stin

g gl

ucos

e, m

mol

/l 4.

71 ±

0.6

1 4.

8 ±

0.6

4.7

± 0.

6 4.

8 ±

0.6

4.7

± 0.

6 4.

7 ±

0.5

SHB

Gf

45.3

± 2

7.1 b

36

.0 ±

17.

7b,d

55.7

± 2

7.3

38.0

± 1

9.6b

51.7

± 2

8.0

60.3

± 3

0.8

C-p

eptid

e, n

mol

/l 0.

72 ±

0.2

8 0.

75 ±

0.2

8 0.

63 ±

0.2

0 0.

77 ±

0.3

2 0.

66 ±

0.2

2 0.

62 ±

0.2

8 IG

FBP3

, mg/

l 4.

20 ±

0.8

6 4.

23 ±

0.9

1 4.

25 ±

0.8

4 4.

09 ±

0.8

6 4.

25 ±

0.8

5 4.

12 ±

0.6

3 Pr

oins

ulin

pm

ol/l

9.14

± 8

.44 b

8.

99 ±

6.9

8 6.

61 ±

4.3

2 9.

87 ±

10.

23

8.56

± 6

.10a

6.71

± 6

.02

Hs-

CR

P, m

g/l

1.91

± 3

.0

2.45

± 4

.16

0.92

± 0

.74

1.98

± 2

.28

1.95

± 3

.80

2.06

± 3

.36

Trig

lyce

rides

, mm

ol/l

1.29

± 0

.81

1.28

± 0

.62

1.31

± 1

.02

1.32

± 0

.80

1.27

± 0

.86

1.01

± 0

.48

Apo

lipop

rote

in A

1, g

/l 1.

47 ±

0.2

8 1.

43 ±

0.3

0 1.

50 ±

0.2

4 1.

48 ±

0.2

5 1.

47 ±

0.3

1 1.

57 ±

0.2

8 A

polip

opro

tein

B, g

/l 0.

91 ±

0.2

5 0.

90 ±

0.2

4 0.

89 ±

0.2

4 0.

92 ±

0.2

7 0.

90 ±

0.2

3 0.

89 ±

0.2

6 A

poB

/Apo

A1

ratio

0.

64 ±

0.2

2 0.

66 ±

0.2

1 0.

61 ±

0.1

9 0.

64 ±

0.2

4 0.

64 ±

0.2

1 0.

59 ±

0.2

1 a S

igni

fican

tly d

iffer

ent f

rom

con

trol s

ubje

cts,

AN

OV

A p

ost-h

oc T

ukey

HSD

, p <

0.0

5 –

0.00

1 b S

igni

fican

tly d

iffer

ent f

rom

con

trol s

ubje

cts,

Line

ar re

gres

sion

with

adj

ustm

ent f

or B

MI,

p 0

.05

– 0.

01

c Bor

derli

ne d

iffer

ence

from

con

trol s

ubje

cts,

Line

ar re

gres

sion

with

adj

ustm

ent f

or B

MI,

p =

0.05

4 d S

igni

fican

tly d

iffer

ent f

rom

reso

lved

PC

OS

patie

nts,

Line

ar re

gres

sion

with

adj

ustm

ent f

or B

MI,

p <

0.05

e W

omen

with

dia

bete

s not

incl

uded

f O

nly

PCO

S pa

tient

s and

con

trols

with

out c

urre

nt u

se o

f com

bine

d or

al c

ontra

cept

ives

or h

orm

one

repl

acem

ent t

hera

py in

clud

ed in

the

anal

ysis

46

Tabl

e X

. OG

TT v

aria

bles

in a

ll no

n-di

abet

ic P

CO

S pa

tient

s, re

solv

ed a

nd u

nres

olve

d PC

OS

patie

nts,

PCO

S ph

enot

ypic

subg

roup

s and

hea

lthy

cont

rols

at t

he fo

llow

-up

inve

stig

atio

n.

A

ll PC

OS

patie

ntse

n

= 46

PCO

S pa

tient

s with

pe

rsis

ting

sym

ptom

s at

follo

w-u

pe

n

= 19

PCO

S pa

tient

s with

re

solv

ed sy

mpt

oms

at fo

llow

-upe

n =

18

PCO

S pa

tient

s with

hi

rsut

ism

, ol

igom

enor

rhea

and

PC

O a

t ind

ex

asse

ssm

ente

n =

20

PCO

S pa

tient

s with

ol

igom

enor

rhea

and

PC

O a

t ind

ex

asse

ssm

ente

n

= 24

Con

trol s

ubje

cts

n

= 68

30

-min

ute

gluc

ose,

O

GTT

, mm

ol/l

7.

6 ±

2.5 b

7.8

± 3.

0 b

7.

5 ±

2.0

8.

2 ±

2.6 b

7.3

± 2.

3

6.4

± 1.

3

120-

min

ute

gluc

ose,

O

GTT

, mm

ol/l

5.3

± 1.

9 5.

6 ±

2.1

4.9

± 1.

8 5.

7 ±

1.8

5.0

± 2.

0 4.

7 ±

1.3

30-m

inut

e in

sulin

, OG

TT,

mU

/l 81

.7 ±

63.

8 b

87.3

± 6

6.0

b 78

.6 ±

67.

2a

97.0

± 6

7.4 b

72

.2 ±

61.

4 42

.3 ±

20.

9

120-

min

ute

insu

lin,

OG

TT, m

U/l

72.7

± 8

7.8 b

72

.4 ±

95.

2 57

.8 ±

50.

7a

88.5

± 1

06.6

b

61.5

± 7

2.7

a 27

.4 ±

25.

1

Mat

suda

Inde

x 2.

01 ±

3.6

3b 2.

23 ±

4.4

5c 2.

16 ±

3.5

3a 1.

64 ±

3.9

9b 2.

36 ±

3.5

1a 3.

43 ±

3.9

9

Insu

linog

enic

Iind

ex

8.93

± 6

.35 b

9.

72 ±

7.5

1c

8.64

± 6

.04

a 10

.21

± 7.

28 b

8.10

± 5

.67

a 5.

50 ±

2.9

7

a Sig

nific

antly

diff

eren

t fro

m c

ontro

l sub

ject

s, A

NO

VA

pos

t-hoc

Tuk

ey H

SD, p

< 0

.05

– 0.

001

b Sig

nific

antly

diff

eren

t fro

m c

ontro

l sub

ject

s, Li

near

regr

essi

on w

ith a

djus

tmen

t for

BM

I, p

< 0

.05

- 0.0

01

c Bor

derli

ne d

iffer

ence

s fro

m c

ontro

l sub

ject

s, Li

near

regr

essi

on w

ith a

djus

tmen

t for

BM

I, p

= 0.

054

and

p =

0.07

8 d S

igni

fican

tly d

iffer

ent f

rom

reso

lved

PC

OS

patie

nts,

Line

ar re

gres

sion

with

adj

ustm

ent f

or B

MI,

p <

0.05

e W

omen

with

dia

bete

s not

incl

uded

47

PCOS patients had higher BMI, increased waist, increased waist/hip ratio, and increased fat percent in comparison with control subjects why subsequent statistical analyses have been adjusted for BMI, table VIII. Following the adjustment for BMI, insulin sensitivity measured by Matsuda insulin sensitivity index was significantly lower in women with PCOS, table X. The insulinogenic index as a measure of beta-cell function was elevated in PCOS patients, table X. In addition, women with PCOS had elevated levels of proinsulin, even after adjustment for BMI, table IX.

Metabolic and anthropometric parameters depending on PCOS phenotype at the initial assessment PCOS patients with clinical signs of hyperandrogenism in addition to menstrual disorders and PCO at index assessment had increased BMI, waist circumference, waist-hip ratio, and fat percent than controls, table VIII. BMI and waist circumference were even higher than in PCOS patients with menstrual irregularities and PCO only at index assessment. Otherwise there were no differences between women with different phenotypic expressions at the index assessment in lipid profile, inflammatory markers or parameters of insulin metabolism at follow-up investigation.

Women without clinical signs of hyperandrogenism at the index assessment displayed higher fasting insulin and proinsulin plasma concentrations than controls. In addition, they had lower Matsuda insulin sensitivity index and higher insulinogenic index than controls, table IX and X.

When adjusted for BMI, there was also a trend towards significantly

lower insulin sensitivity and increased insulinogenic index among women with hyperandrogenism at the index assessment in comparison with control subjects. Women with hyperandrogenic PCOS phenotype at the index assessment had lower serum concentrations of SHBG than controls at the follow-up investigation.

Metabolic and anthropometric parameters depending on persisting PCOS symptoms Women with persisting PCOS symptoms had increased waist circumference, higher WHR and BMI than controls, whereas women with resolved PCOS did not differ in any of these aspects from control subjects, table VIII. BMI and waist circumference in women with persisting PCOS symptoms were also increased in comparison to women with resolved PCOS symptoms at the follow-up investigation, table VIII and IX. With the exception of lower SHBG levels among women with unresolved PCOS, there were no differences between women with resolved and unresolved PCOS in lipid profile, inflammatory markers or parameters of insulin metabolism.

48

Table XI. Results of a multivariate linear regression analysis of variables associated with insulin sensitivity expressed by Matsuda Index

B

95 % CI ��

p

Group -0.183 -0.354 - -0.011 -0.144 0.05 Age -0.019 -0.034 - - 0.004 -0.159 0.05 BMI -0.055 -0.076 - -0.034 -0.401 0.001 Triglycerides -0.242 -0.368 - -0.115 -0.272 0.001 SHBG 0.003 0.001 – 0.006 0.175 0.05 hs-CRP -0.023 -0.54 – 0.008 -0.102 ns Variables entered into the equation are based on significant findings in bivariate analyses and colinearity analyses. R2 for final model was 0.62.

Insulin sensitivity measured as Matsuda index differed from controls in both women with persisting and resolved PCOS at the follow-up investigation. Likewise, beta cell function measured as insulinogenic index was significantly elevated in both women with resolved and unresolved PCOS in comparison with control subjects, table X.

Variables associated with insulin sensitivity at the follow-up investigation Table XI displays factors associated with Matsuda insulin sensitivity index at the time of the follow-up investigation. Age, BMI, triglycerides and low SHBG levels, together with PCOS diagnosis (resolved as well as unresolved) were all significantly and independently associated with insulin sensitivity. Free androgen index was not associated with insulin sensitivity in the bivariate analysis, and was hence not brought into the final model.

49

Discussion

Methodological considerations The emphasis in this study has been to provide the twenty-year old, newly diagnosed PCOS patient with relevant information as regards her future reproductive health and her future risk of developing cardiovascular and metabolic diseases. From her perspective, there is a fair chance that her PCOS will resolve with advancing age. In order to provide useful information to clinicians who meet young, internet-surfing women with PCOS we needed to include all our patients in the long-term analyses, not only the ones who continued to meet PCOS criteria in their forties. If only women with persisting PCOS had been included, we would have investigated a selected subgroup with worse outcome and the young patients would be presented with exaggerated risk figures.

Like any follow-up study, free living subjects tend to develop medical

disorders and utilize medications which may have confounded our results. Besides the use of antihypertensive drugs which may affect the pulse wave analysis, our patients were also using various types of contraception. In addition we also had a variable endocrine status among our PCOS patients with some menopausal subjects and with premenopausal subjects evaluated in the follicular or luteal phases of the menstrual cycle. If we had limited our study to patients with similar endocrine status we would have ended up with a much more limited sample for the follow-up investigation. Furthermore, many of our participating patients had to travel to Uppsala for 2 – 5 hours and a few also stayed over night. Given the efforts patients went through to participate in the study, we felt that it would be too much to ask to schedule their appointments according to menstrual cycle phase. Also, it was not feasible to request from patients that they should discontinue their use of oral contraceptives/HRT prior to participation.

As we knew beforehand that variable endocrine status was to be expected

we tried to use outcome measures that were not affected by menstrual cycle phase. For instance, endothelial function was evaluated by pulse wave analysis instead of flow-mediated dilatation of the brachial artery where changes across the menstrual cycle have been reported. However, for many

50

of the outcome measures in paper I this was not feasible, why data are reported in subgroups with more uniform endocrine status.

The choice of the control group was a major concern when this study was

planned. In the end, and with the young PCOS patient in perspective, we decided upon using a population-based control group. The use of a weight-matched control group was thoroughly discussed but discarded for several reasons. First, with the hypothesis (based on our clinical experience) that a proportion of PCOS patients improve with increasing age, and as it can be assumed that a weight-matched control group also would present with diabetes and hypertension at this age, we would risk underestimating the burden of PCOS with the use of a weight-matched control group. Secondly, by using a population-based age-matched control group, we would obtain a more reliable overall picture of PCOS patients’ risk profile in comparison with women of similar age. Third, we also had the possibility to adjust all statistical analyses for BMI. Finally, finding a weight-matched population without using selected hospital-based populations was not considered possible.

Another problem with the control group was that women with PCOS were

identified in the past, whereas control subjects were identified in the present. Given the resolution of PCOS in many of the cases, it is thus possible that there is a contamination in the control population. When the study was planned, we did discuss the possibility of using patient groups with infertility causes other than PCOS who had presented at the IVF clinic between 1987 and 1995 as controls. Again, this suggestion was rejected for ethical reasons and because life-time reproductive outcome also was one of our aims.

The precautions undertaken to ensure that no PCOS cases were hidden

among controls were 1) transvaginal ultrasound screening (although it could be assumed that some controls would have had PCO if investigated at an earlier age) and 2) all controls were screened for history of oligomenorrhea or amenorrhea. In fact, the question we asked was “Have you ever, on more than one occasion, had more than six weeks between menses without being pregnant?” A similar question was asked for amenorrhea. One control subject reported a history of amenorrhea (3-4 months) during her college-year in the US, but as that was more likely due to stress she was kept in the study. None of our controls had polycystic ovaries on ultrasound.

In addition, the invitation letter to control subjects specifically described

that they would act as a control group to women with PCOS. We also described the symptoms and problems of PCOS in the letter so that potential control subjects could judge for themselves whether or not they were

51

suitable for the study. Upon initial contact two potential control subjects were excluded because they identified themselves as PCOS patients.

Reproductive outcome Reproductive outcome did not differ between this unselected population of women with a previous diagnosis of PCOS and healthy control subjects, and the ovarian reserve seemed better preserved in the women with PCOS. The live birth rate and the rate of miscarriages were similar in PCOS patients and control women.

In our study almost all PCOS patients who attempted a pregnancy were successful. The cumulative delivery rate in our subjects was almost 90 %, which is in line with a population-based study of PCOS patients [151] and a previous long-term follow-up study of infertile women with ovulatory disorders, where a cumulative delivery rate of 82 % was reported [152]. More than two thirds of the PCOS patients in the present study reported at least one spontaneous pregnancy.

The menstrual frequency among PCOS patients who were premenopausal and without hormonal contraception was lower than in control subjects, but PCOS patients reported on average almost 9 menstrual cycles per year. Some clinical observations suggest improved fertility in ageing women with PCOS. A positive impact of ageing on cycle regularization in PCOS has recently been claimed, but the fertility outcome was not evaluated [138]. The authors concluded that the follicle loss through the process of ovarian ageing could explain the occurrence of regular cycles in older patients with PCOS [138]. In women overall, ageing is associated with a subtle shortening of menstrual cycles of a mean of 2-3 days from the age 20 to the early forties [49, 153]. Most likely, the mechanism is associated with a reduction in antral follicle numbers, and consequently a more rapid follicular growth and selection, thus shortening of the follicular phase. Careful monitoring has suggested that basal FSH levels are lower in young women with PCOS than in the early follicular phase of women with normal ovaries [59]. The mechanism behind these low levels, which may partly explain lack of follicular growth, is probably increased production of inhibin B from the increased number of antral follicles in polycystic ovaries [154]. Ovarian ageing results in diminution of the follicular cohort in both normal women and PCOS patients, and is associated with decreased inhibin B and AMH levels [48, 50]. Ageing may thus be followed by lower inhibin B levels, which will permit FSH enhancement and lead to full follicle maturation, more regular menstrual cycles and the appearance of ovulatory cycles in polycystic ovaries. A similar mechanism may be responsible for the increased number of ovulatory cycles after ovarian wedge resection or ovarian drilling, when the cohort of resting follicles is dramatically reduced [155]. Thus, in women with PCOS, such an age-dependent reduction in the

52

pool of resting follicles may eventually lead to more cycles becoming ovulatory and a catch-up of fecundity.

The rate of miscarriages was not increased in PCOS patients and the

number of subjects who had had at least one miscarriage throughout their fertile life did not differ in comparison with the control subjects. These findings are at odds with previous reports of increased rates of miscarriages in PCOS patients [41-43]. The discrepancy is most likely due to the fact that previous studies have merely included infertile PCOS patients undergoing assisted reproduction, whereas the present study included an unselected population, i.e. PCOS patients seeking help for various causes such as oligomenorrhea, hirsutism or infertility. However, more than half of our PCOS patients had at some point been treated for infertility. Our findings suggest that other factors than the PCOS diagnosis might contribute to the increased risk of miscarriage reported in prior studies, such as obesity, fertility treatment [41] or smoking [156]. Indeed, in women with polycystic ovaries, but not a full syndrome, miscarriage rates have been reported to be normal [151], as has prevalence of recurrent miscarriages [44].

Ovarian reserve Mean ovarian volume and number of antral follicles, measures of ovarian reserve, were significantly higher in PCOS patients. Likewise, PCOS patients had markedly higher serum concentrations of AMH. A number of previous reports for young women with PCOS lend support for an increased ovarian reserve in these women compared with age-matched controls [50]. Thus, polycystic ovaries are larger and contain more antral follicles as assessed by ovarian ultrasound [66, 67], and density of follicles at primary stages is increased [34]. Several studies have also reported higher levels of AMH in women with PCOS than in controls [35, 56, 57]. Findings on inhibin B are more discrepant [64, 65]. Our findings in older women with PCOS are in accordance with these results, suggesting a retained greater ovarian reserve in these women in their forties. Results published for basal levels of FSH in PCOS have been less conclusive [58]. However, Dahlgren and colleagues found lower levels of FSH in 50-year old women with a previous PCOS diagnosis compared with controls, in spite of the fact that the women with PCOS had been wedge-resected, which is in line with our findings. Recently, it was proposed that serum FSH assessments are inferior to measurements of AMH in identifying women with reduced ovarian reserve, largely because AMH is cycle-independent [51]. This assumption is substantiated in the present study, where AMH levels correlated strongly both with ovarian volume and follicle counts. Recent results indicate that AMH is a strong predictor for success rates of assisted reproduction, i.e. ovarian reserve [56, 157, 158], underlining the importance of the vast

53

differences in AMH levels we found between women with PCOS and controls.

Also, the menstrual cycle length has been reported to be an age-

independent marker of female fertility. Recently, it was shown that the mean menstrual cycle length correlated linearly with pregnancy and delivery rates after IVF/ICSI, even after age adjustment [49]. Although women with anovulatory PCOS were not included in that study, the authors found an association between mean menstrual cycle length and antral follicle counts, many of the patients with cycles > 28 days being women with multifollicular or ovulating polycystic ovaries. In line with those findings, others [159, 160] reported on higher pregnancy rates after IVF in women with PCOS than in women with normal ovaries.

Endothelial dysfunction PCOS patients, who at the time of follow-up had reached a mean age of 43 years, displayed signs of deteriorated endothelium-dependent vasodilatation in comparison to age-matched healthy controls and this finding was mainly driven by patients who still fulfilled diagnostic criteria for PCOS. However, this difference was primarily explained by the increased prevalence of cardiovascular risk factors among the PCOS patients, such as increased BMI and co-morbid diabetes, hypertension, and hyperlipidemia. Indeed, when the statistical analyses on endothelial function were adjusted for one of these risk factors, BMI, none of the significant differences between groups remained. This finding is in line with previous studies from our group, suggesting that PCOS per se is not associated with any metabolic disturbances that can be distinguished from co-morbid obesity [23, 85].

Endothelial function is influenced by a number of factors. Previous

studies have linked obesity to endothelial dysfunction [161-166], and recently Lind and colleagues demonstrated that EDV in resistance arteries was reduced in subjects with increased intra-abdominal adipose tissue mass [167]. EDV is also associated with insulin resistance and reduced in subjects with the metabolic syndrome [149]. In our middle-aged PCOS patients insulin resistance was a stronger independent explanatory variable of EDV than BMI. This finding is supported by studies of younger PCOS patients [132, 135] and by the fact that treatments aiming at decreasing insulin resistance in PCOS patients, such as metformin, have been shown to improve endothelial function [168-171].

A higher prevalence of cardiovascular risk factors in patients with PCOS

has been described in multiple studies [124-128]. In the line with these studies we have confirmed that PCOS patients between 40 – 50 years of age

54

present with increased BMI, waist/hip ratio, systolic and diastolic blood pressures and triglycerides. Presumably because of the relatively low age of subjects, we failed to demonstrate a higher prevalence of cardiovascular events in patients with PCOS.

Metabolic syndrome According to the scientific statement of AHA and NCEP/ATPIII, the prevalence of MetS in PCOS women was 21.4 % and in healthy controls 6.9 % in our study. This was nearly 2-fold lower than in previously published American and Asian studies [68, 107, 108, 172, 173], presumably because of lower BMI and differences in racial/ethnic composition of our probands in comparison to these studies.

In addition, frequency of isolated features of metabolic syndrome was more common in women with PCOS. This is in line with studies from central Europe and China who reported relatively low prevalence of overt metabolic syndrome, but frequent occurrence of metabolic and anthropometric disturbances [105, 109]. PCOS women between 40 – 50 years of age presented in our study with increased waist circumference and waist/hip ratio, BMI and triglycerides. These metabolic and anthropometric parameters and even HDL were also associated with insulin sensitivity. Both systolic and diastolic blood pressures were higher in PCOS women. This possibly pre-hypertensive state with increased day-time blood pressures in women with PCOS, persisting even after adjusting for BMI, body fat distribution and insulin resistance, was previously described in this group of women already at the index assessment [112]. Surprisingly, the regular intake of antihypertensive drugs at the follow-up was not higher in PCOS women than in controls. Further analyses are mandatory to reveal whether BMI and age are stronger contributing factors than hyperandrogenism and insulin resistance [109]. This hypothesis is supported by our results in Paper III, where insulin sensitivity ratio M/I was lower in women with high BMI at the index assessment and in those who progressively gained weight until the follow-up.

Hyperandrogenism PCOS patients displayed decreasing testosterone levels over time but they still had increased free androgen index compared to controls at the time of the follow-up. Clearly, although their menstrual frequency might be normalized across time, the androgen levels remain increased in comparison to controls. Persistently elevated androgen levels have previously been demonstrated in older PCOS patients in comparison with age-matched controls [174].

55

The decreasing testosterone levels over time is in line with a cross-sectional study in PCOS patients suggesting 40 – 60 % lower levels of total testosterone in PCOS patients 42 years or older as compared to younger PCOS patients [174]. In contrast to PCOS patients, a longitudinal study in healthy women recently showed that testosterone levels display a gradual increase from the age of 40 and onwards [175].

It has been argued that the combination of hyperandrogenism and

menstrual irregularity is associated with the most profound insulin resistance [176] and some studies even indicate that PCOS patients without clinical features of hyperandrogenism do not develop insulin resistance [88, 89]. According to our findings the phenotypic presentation at the index assessment had no influence on insulin sensitivity at the time of the follow-up. Both women with oligomenorrhea and PCOS as well as women with hyperandrogenism, oligomenorrhea and PCO displayed decreased insulin sensitivity, expressed as Matsuda Index, in comparison with control subjects and there was no difference in insulin sensitivity between PCOS phenotypic expressions. However, hyperandrogenism is less relevant for insulin sensitivity when women with PCOS have reached premenopausal ages, which is further corroborated by the fact that free androgen index is no longer associated with Matsuda index in the bivariate linear regression. Insulin sensitivity among our premenopausal women was instead independently associated with age, BMI, triglycerides, SHBG and PCOS diagnosis (resolved or unresolved). In the long run, however, women who presented with hyperandrogenism at the index presentation may presumably suffer an increased risk of subsequent development of type 2 diabetes as they had increased BMI, increased waist circumference and reduced SHBG levels in comparison with control subjects, all of these factors individually contributing to their life-time risk of diabetes [101-103, 177-180].

Previous studies have firmly concluded that SHBG is associated with insulin sensitivity in both women with PCOS and other women [178, 179]. Among our PCOS women in paper III we found a trend towards a significant association between SHBG and insulin sensitivity, but only in the bivariate analysis. In paper IV, on the contrary, SHBG was one of the factors independently associated with insulin sensitivity (see also chapter Insulin resistance, glucose tolerance, early insulin response and diabetes mellitus). The absence of a stronger relationship in paper III is presumably due to the relatively small sample size or the relatively high proportion of lean subjects.

SHBG increased over time which is at odds with findings in healthy

women. Longitudinal studies in non-PCOS women indicate that increased BMI over time is closely related to lowering of SHBG levels [181]. However, decreasing SHBG levels over time has also been reported to occur

56

in women irrespective of BMI changes, thus increasing the biologically available testosterone in pre- and perimenopausal women [175]. The changes in testosterone and SHBG levels over time in our cohort of women with PCOS suggest that hyperandrogenism may partly resolve before menopause in this group of women. This finding is in line with previous findings of an improved menstrual cycle pattern with increasing age in PCOS patients [61, 182] and the fact that in at least one third of our women no longer fulfilled PCOS criteria.

Insulin resistance, glucose tolerance, early insulin response and diabetes mellitus Prevalence of disturbances of glucose metabolism was significantly higher in PCOS women in comparison with healthy controls. Only 4.5 % of controls had impaired glucose tolerance or type 2 diabetes while in women with PCOS the corresponding figure was 16.7 % (if also cases with type 1 diabetes are included the corresponding figure is 21.4 %). This figure is, however, substantially lower than previously presented [101, 102]. Prevalence rates for glucose intolerance are reported to vary between 38 - 45 % in younger women with PCOS but these studies have included subjects who predominantly are overweight or obese. However, our data fits better with previous findings in lean women with PCOS where approximately 12 % are reported to display glucose intolerance (IGT or type 2 diabetes), in particular taking into consideration that our subjects had reached a more advanced age than subjects of previous studies. The overall relatively low rate of IGT and diabetes in our population are most likely explained by lower BMI and differences in racial/ethnic composition of our probands in comparison to previous studies. Possibly, also the inclusion of milder PCOS phenotypes may have contributed to the overall lower prevalence rates for IGT and type 2 diabetes, even if hyperandrogenism seems to be less relevant for insulin sensitivity in women with PCOS in their premenopausal age (see previous chapter Hyperandrogenism).

Insulin resistance and decreased insulin secretion, and the interaction between low acute insulin response and high insulin resistance are important predisposing factors in development of type 2 diabetes [183-185]. In our study, PCOS patients over time displayed decreased acute insulin response in the IVGTT and unchanged insulin sensitivity. The former finding might reflect impaired beta-cell function with increasing age, also suggested by increased AUC glucose during the IVGTT. Furthermore, this finding might confirm the theory of Cho and colleagues that the progression from PCOS to the development of type 2 diabetes is rather due to a progressive

57

deterioration of beta-cell function than further increase in insulin resistance [186]. Whether the deterioration of beta-cell function is genetically determined or reflects their exhaustion after compensation of insulin resistance over long time or even both, remain unclear [71, 187, 188]. Possible reasons why insulin sensitivity remained unchanged in our cohort may be due to a substantial fraction of women with resolved PCOS by unchanged WHR across time, or by decreasing androgen levels across time.

Another important finding of our study is that insulin sensitivity was

decreased compared with controls independent of whether women had persisting or resolved PCOS symptoms at the time of the follow-up. This finding indicates that although the clinical presentation of PCOS may be less obvious when women reach the age of 40 [61, 182], their risk of developing type 2 diabetes may not have diminished. This finding also emphasizes the need to monitor women with PCOS with respect to glucose tolerance, regardless of whether they are symptomatic or symptom free.

The insulinogenic index during OGTT, as a measure of beta cell function,

was increased in all PCOS patients, independent of whether PCOS symptoms were resolved or unresolved and also independent of the phenotypic presentation at the index assessment. The insulinogenic index has previously been shown to strongly correlate with the first phase insulin response following IVGTT [189] and prior findings of our group have indicated that PCOS women, independent of body weight, display an enhanced early insulin response during IVGTT [23, 85, 190]. Notably, whereas insulin sensitivity is restored following weight reduction, early insulin response is not reversed after weight reduction [23]. Whereas prior studies in young PCOS patients indicate that the increased early insulin response was not accounted for by the insulin resistance [85], the increased beta cell function of our premenopausal women with PCOS appears to be a normal beta cell adaptation to the decreased insulin sensitivity.

Insulin sensitivity also remained unchanged between the two time-points

of assessment. However, PCOS patients who already at young age presented with obesity and continued to gain weight were more prone to be insulin resistant at follow-up. Function of pancreatic beta cells decreases over time in PCOS patients, presumably both due to aging and exhaustion of already genetically altered pancreatic cells compensating insulin resistance in the long term.

58

General conclusions

Most women with PCOS had given birth and the rate of spontaneous pregnancies was relatively high. The rate of miscarriages was not increased in PCOS patients and the number of subjects who had had at least one miscarriage throughout their fertile life did not differ in comparison with the control subjects.

The ultrasound findings together with increased levels of AMH imply that

PCOS patients have an ovarian reserve possibly superior to women with normal ovaries.

PCOS women display signs of endothelial dysfunction at follow-up, but

this is largely due to the increased prevalence of independent risk factors for CVD found in this group. This finding might provide support to the hypothesis that PCOS patients have a cluster of factors that increase risk of CVD but that the syndrome per se is not the predictor of cardiovascular events.

Free androgen levels decrease over time in PCOS patients but are still

elevated compared to age-matched control subjects. Insulin sensitivity remained unchanged between the two time-points of

assessment but premenopausal women with PCOS had lower insulin sensitivity than age-matched controls.

PCOS patients who already at young age presented with obesity and

continued to gain weight were more prone to be insulin resistant at follow-up.

Function of pancreatic beta cells decreases over time in PCOS patients,

but premenopausal women with PCOS displayed increased beta cell function in comparison with age-matched control subjects.

IGT and type 2 diabetes occurred more often in PCOS patients than in

healthy controls.

59

Independent of PCOS phenotype at the index assessment and persistence of PCOS symptoms at the follow-up investigation, premenopausal women with PCOS had lower insulin sensitivity and increased beta cell function in comparison with control subjects.

60

Acknowledgements

This work was carried out at the Department of Women’s and Children’s Health, Uppsala University. I would like to express my sincere gratitude to all those who helped, supported and encouraged me during the work on this thesis:

I would especially like to thank: Inger Sundström Poromaa, professor at the Department of Women’s and

Children’s Health, Uppsala University, Sweden, my head supervisor, for sharing your great knowledge with me, for your unbelievable work-drive, for your patience when explaining complicated statistic rules, for your endless hospitality and tasteful dinners at your place … for everything.

Jan Holte, M.D. and Ph.D. at the Department of Women’s and Children’s

Health, Uppsala University, my co-supervisor, for trusting me, for sharing your world-appreciated knowledge and precious data on “your” PCOS women with me, for fruitful discussions over the lunch at the Sven Dufva Restaurant.

Matts Olovsson, associate professor at the Department of Women’s and

Children’s Health, Uppsala University, my second co-supervisor, for your precious notes and advices on manuscripts.

Ove Axelsson, professor and head of the Department of Women’s and

Children’s Health, Uppsala University, for support and encouragement. Elisabeth Darj and Gunilla Hallberg, chiefs of the gynaecologic and

obstetric departments, Uppsala University Hospital, for giving me the opportunity to carry on my research work at the out-patient department.

Lars Lind, professor at the Department of Internal Medicine, Uppsala

University, for introduction to the world of heart diseases and diagnostics. Christian Berne, professor at the Department of Internal Medicine,

Uppsala University, for everything you have taught me about diabetes, for vital and interesting lunch-discussions about pancreas function and sugar metabolism.

61

Lars Berglund for your precious help with statistics, Anders Larsson for

precise serum analysis, and Mats Stridberg for important hormonal analysis. Lena Moby, our research nurse, for your learning capacity, for all those

hours with clamp-examinations, for all help with our patients. Jan Hall, research nurse at the Department of Internal Medicine, Uppsala

University, for your very comprehensive and yet funny introduction to clamp- and endothelium- examinations.

Kristiina Kask, Alkistis Skalkidou Åsa Lindholm and Ragnheidur

Arnadottir for all those nice, funny and enriching moments at the laboratory, at your places and on conferences and seminaries.

All the staff at the Department of Women’s and Children’s Health,

Uppsala University, for being helpful and positive during my research work at the department.

All my colleagues, midwives and all the staff at the Nordland Hospital,

Norway, for support during all the years of my research work. My dear family in Slovakia, my mom and brother Branislav and his

family, for supporting me and believing in me no matter what. My beloved Paolo for taking care of me and cooking of fabulous dishes,

while I was doing my research work. And my dear Hugo, for relaxing walks that helped me to think clearly.

I would like to thank all the women participating in our studies for your

willingness to take part and understanding of the importance of research. These studies were supported financially by grants from the Family

Planning Foundation, Uppsala, and the Selander foundation.

62

References

1. Asuncion M, Calvo RM, San Millan JL, Sancho J, Avila S, Escobar-

Morreale HF. A prospective study of the prevalence of the polycystic ovary syndrome in unselected Caucasian women from Spain. J Clin Endocrinol Metab 2000;85:2434-8.

2. Azziz R, Woods KS, Reyna R, Key TJ, Knochenhauer ES, Yildiz BO. The prevalence and features of the polycystic ovary syndrome in an unselected population. J Clin Endocrinol Metab 2004;89:2745-9.

3. Diamanti-Kandarakis E, Kouli CR, Bergiele AT, Filandra FA, Tsianateli TC, Spina GG, Zapanti ED, Bartzis MI. A survey of the polycystic ovary syndrome in the Greek island of Lesbos: hormonal and metabolic profile. J Clin Endocrinol Metab 1999;84:4006-11.

4. Lindholm A, Andersson L, Eliasson M, Bixo M, Sundstrom-Poromaa I. Prevalence of symptoms associated with polycystic ovary syndrome. Int J Gynaecol Obstet 2008;102:39-43.

5. Taponen S, Martikainen H, Jarvelin MR, Laitinen J, Pouta A, Hartikainen AL, Sovio U, McCarthy MI, Franks S, Ruokonen A. Hormonal profile of women with self-reported symptoms of oligomenorrhea and/or hirsutism: Northern Finland birth cohort 1966 study. J Clin Endocrinol Metab 2003;88:141-7.

6. Zawadski, J., Diagnostic criteria for polycystic ovary syndrome: towards a rational approach. In: Dunaif A, Givens JR, Haseltine FP, Merriam GR, eds. Polycystic ovary syndrome., in Boston: Blackwell Scientific Publications. 1992. p. 377-384.

7. Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome. Fertil Steril 2004;81:19-25.

8. Azziz R, Carmina E, Dewailly D, Diamanti-Kandarakis E, Escobar-Morreale HF, Futterweit W, Janssen OE, Legro RS, Norman RJ, Taylor AE, Witchel SF. Positions statement: criteria for defining polycystic ovary syndrome as a predominantly hyperandrogenic syndrome: an Androgen Excess Society guideline. J Clin Endocrinol Metab 2006;91:4237-45.

9. Fehring RJ, Schneider M, Raviele K. Variability in the phases of the menstrual cycle. J Obstet Gynecol Neonatal Nurs 2006;35:376-84.

10. Ferriman D, Gallwey JD. Clinical assessment of body hair growth in women. J Clin Endocrinol Metab 1961;21:1440-7.

11. Manni A, Pardridge WM, Cefalu W, Nisula BC, Bardin CW, Santner SJ, Santen RJ. Bioavailability of albumin-bound testosterone. J Clin Endocrinol Metab 1985;61:705-10.

63

12. Holte J. Disturbances in insulin secretion and sensitivity in women with the polycystic ovary syndrome. Baillieres Clin Endocrinol Metab 1996;10:221-47.

13. Amato P, Simpson JL. The genetics of polycystic ovary syndrome. Best Pract Res Clin Obstet Gynaecol 2004;18:707-18.

14. Franks S, Gharani N, Waterworth D, Batty S, White D, Williamson R, McCarthy M. The genetic basis of polycystic ovary syndrome. Hum Reprod 1997;12:2641-8.

15. Franks S, Gharani N, McCarthy M. Candidate genes in polycystic ovary syndrome. Hum Reprod Update 2001;7:405-10.

16. Escobar-Morreale HF, San Millan JL. Abdominal adiposity and the polycystic ovary syndrome. Trends Endocrinol Metab 2007;18:266-72.

17. Elbers JM, Giltay EJ, Teerlink T, Scheffer PG, Asscheman H, Seidell JC, Gooren LJ. Effects of sex steroids on components of the insulin resistance syndrome in transsexual subjects. Clin Endocrinol (Oxf) 2003;58:562-71.

18. Alexanderson C, Eriksson E, Stener-Victorin E, Lystig T, Gabrielsson B, Lonn M, Holmang A. Postnatal testosterone exposure results in insulin resistance, enlarged mesenteric adipocytes, and an atherogenic lipid profile in adult female rats: comparisons with estradiol and dihydrotestosterone. Endocrinology 2007;148:5369-76.

19. Franks S. Adult polycystic ovary syndrome begins in childhood. Best Pract Res Clin Endocrinol Metab 2002;16:263-72.

20. Xita N, Tsatsoulis A. Review: fetal programming of polycystic ovary syndrome by androgen excess: evidence from experimental, clinical, and genetic association studies. J Clin Endocrinol Metab 2006;91:1660-6.

21. Holmang A, Bjorntorp P. [Hyperandrogenism or insulin resistance. What is the cause and effect in diabetes among women?]. Lakartidningen 1993;90:4313-5.

22. Holte J, Bergh T, Berne C, Lithell H. Serum lipoprotein lipid profile in women with the polycystic ovary syndrome: relation to anthropometric, endocrine and metabolic variables. Clin Endocrinol (Oxf) 1994;41:463-71.

23. Holte J, Bergh T, Berne C, Wide L, Lithell H. Restored insulin sensitivity but persistently increased early insulin secretion after weight loss in obese women with polycystic ovary syndrome. J Clin Endocrinol Metab 1995;80:2586-93.

24. Ferrannini E, Barrett EJ, Bevilacqua S, DeFronzo RA. Effect of fatty acids on glucose production and utilization in man. J Clin Invest 1983;72:1737-47.

25. Barber TM, McCarthy MI, Wass JA, Franks S. Obesity and polycystic ovary syndrome. Clin Endocrinol (Oxf) 2006;65:137-45.

26. Knochenhauer ES, Key TJ, Kahsar-Miller M, Waggoner W, Boots LR, Azziz R. Prevalence of the polycystic ovary syndrome in unselected black and white women of the southeastern United States: a prospective study. J Clin Endocrinol Metab 1998;83:3078-82.

27. Gambineri A, Pelusi C, Vicennati V, Pagotto U, Pasquali R. Obesity and the polycystic ovary syndrome. Int J Obes Relat Metab Disord 2002;26:883-96.

28. McGee EA, Hsueh AJ. Initial and cyclic recruitment of ovarian follicles. Endocr Rev 2000;21:200-14.

64

29. Rajpert-De Meyts E, Jorgensen N, Graem N, Muller J, Cate RL, Skakkebaek NE. Expression of anti-Mullerian hormone during normal and pathological gonadal development: association with differentiation of Sertoli and granulosa cells. J Clin Endocrinol Metab 1999;84:3836-44.

30. Visser JA, Themmen AP. Anti-Mullerian hormone and folliculogenesis. Mol Cell Endocrinol 2005;234:81-6.

31. Jonard S, Dewailly D. The follicular excess in polycystic ovaries, due to intra-ovarian hyperandrogenism, may be the main culprit for the follicular arrest. Hum Reprod Update 2004;10:107-17.

32. Zeleznik AJ. Premature elevation of systemic estradiol reduces serum levels of follicle-stimulating hormone and lengthens the follicular phase of the menstrual cycle in rhesus monkeys. Endocrinology 1981;109:352-5.

33. Roseff SJ, Bangah ML, Kettel LM, Vale W, Rivier J, Burger HG, Yen SS. Dynamic changes in circulating inhibin levels during the luteal-follicular transition of the human menstrual cycle. J Clin Endocrinol Metab 1989;69:1033-9.

34. Webber LJ, Stubbs S, Stark J, Trew GH, Margara R, Hardy K, Franks S. Formation and early development of follicles in the polycystic ovary. Lancet 2003;362:1017-21.

35. Pigny P, Merlen E, Robert Y, Cortet-Rudelli C, Decanter C, Jonard S, Dewailly D. Elevated serum level of anti-mullerian hormone in patients with polycystic ovary syndrome: relationship to the ovarian follicle excess and to the follicular arrest. J Clin Endocrinol Metab 2003;88:5957-62.

36. Laven JS, Mulders AG, Visser JA, Themmen AP, De Jong FH, Fauser BC. Anti-Mullerian hormone serum concentrations in normoovulatory and anovulatory women of reproductive age. J Clin Endocrinol Metab 2004;89:318-23.

37. Piltonen T, Morin-Papunen L, Koivunen R, Perheentupa A, Ruokonen A, Tapanainen JS. Serum anti-Mullerian hormone levels remain high until late reproductive age and decrease during metformin therapy in women with polycystic ovary syndrome. Hum Reprod 2005;20:1820-6.

38. Pellatt L, Hanna L, Brincat M, Galea R, Brain H, Whitehead S, Mason H. Granulosa cell production of anti-Mullerian hormone is increased in polycystic ovaries. J Clin Endocrinol Metab 2007;92:240-5.

39. Chen MJ, Yang WS, Chen CL, Wu MY, Yang YS, Ho HN. The relationship between anti-Mullerian hormone, androgen and insulin resistance on the number of antral follicles in women with polycystic ovary syndrome. Hum Reprod 2008;23:952-7.

40. Diejomaoh M, Jirous J, Al-Azemi M, Baig S, Gupta M, Tallat A. The relationship of recurrent spontaneous miscarriage with reproductive failure. Med Princ Pract 2003;12:107-11.

41. Wang JX, Davies MJ, Norman RJ. Polycystic ovarian syndrome and the risk of spontaneous abortion following assisted reproductive technology treatment. Hum Reprod 2001;16:2606-9.

42. Legro RS. Pregnancy considerations in women with polycystic ovary syndrome. Clin Obstet Gynecol 2007;50:295-304.

43. Boomsma CM, Fauser BC, Macklon NS. Pregnancy complications in women with polycystic ovary syndrome. Semin Reprod Med 2008;26:72-84.

65

44. Rai R, Backos M, Rushworth F, Regan L. Polycystic ovaries and recurrent miscarriage--a reappraisal. Hum Reprod 2000;15:612-5.

45. Fedorcsak P, Dale PO, Storeng R, Tanbo T, Abyholm T. The impact of obesity and insulin resistance on the outcome of IVF or ICSI in women with polycystic ovarian syndrome. Hum Reprod 2001;16:1086-91.

46. Fedorcsak P, Storeng R, Dale PO, Tanbo T, Abyholm T. Obesity is a risk factor for early pregnancy loss after IVF or ICSI. Acta Obstet Gynecol Scand 2000;79:43-8.

47. Boomsma CM, Eijkemans MJ, Hughes EG, Visser GH, Fauser BC, Macklon NS. A meta-analysis of pregnancy outcomes in women with polycystic ovary syndrome. Hum Reprod Update 2006;12:673-83.

48. Macklon NS, Fauser BC. Ovarian reserve. Semin Reprod Med 2005;23:248-56.

49. Brodin T, Bergh T, Berglund L, Hadziosmanovic N, Holte J. Menstrual cycle length is an age-independent marker of female fertility: Results from 6271 treatment cycles of in vitro fertilization. Fertil Steril 2007;

50. Nikolaou D, Gilling-Smith C. Early ovarian ageing: are women with polycystic ovaries protected? Hum Reprod 2004;19:2175-9.

51. Tremellen KP, Kolo M, Gilmore A, Lekamge DN. Anti-mullerian hormone as a marker of ovarian reserve. Aust N Z J Obstet Gynaecol 2005;45:20-4.

52. Weenen C, Laven JS, Von Bergh AR, Cranfield M, Groome NP, Visser JA, Kramer P, Fauser BC, Themmen AP. Anti-Mullerian hormone expression pattern in the human ovary: potential implications for initial and cyclic follicle recruitment. Mol Hum Reprod 2004;10:77-83.

53. La Marca A, De Leo V, Giulini S, Orvieto R, Malmusi S, Giannella L, Volpe A. Anti-Mullerian hormone in premenopausal women and after spontaneous or surgically induced menopause. J Soc Gynecol Investig 2005;12:545-8.

54. van Rooij IA, Broekmans FJ, Scheffer GJ, Looman CW, Habbema JD, de Jong FH, Fauser BJ, Themmen AP, te Velde ER. Serum antimullerian hormone levels best reflect the reproductive decline with age in normal women with proven fertility: a longitudinal study. Fertil Steril 2005;83:979-87.

55. La Marca A, Stabile G, Artenisio AC, Volpe A. Serum anti-Mullerian hormone throughout the human menstrual cycle. Hum Reprod 2006;21:3103-7.

56. La Marca A, Volpe A. Anti-Mullerian hormone (AMH) in female reproduction: is measurement of circulating AMH a useful tool? Clin Endocrinol (Oxf) 2006;64:603-10.

57. Catteau-Jonard S, Jamin SP, Leclerc A, Gonzales J, Dewailly D, di Clemente N. Anti-Mullerian Hormone, Its Receptor, Fsh Receptor, and Androgen Receptor Genes Are Overexpressed by Granulosa Cells from Stimulated Follicles in Women with Polycystic Ovary Syndrome. J Clin Endocrinol Metab 2008;

58. Franks S, Stark J, Hardy K. Follicle dynamics and anovulation in polycystic ovary syndrome. Hum Reprod Update 2008;14:367-78.

59. Holte J, Bergh T, Gennarelli G, Wide L. The independent effects of polycystic ovary syndrome and obesity on serum concentrations of

66

gonadotrophins and sex steroids in premenopausal women. Clin Endocrinol (Oxf) 1994;41:473-81.

60. Roberts JE, Spandorfer S, Fasouliotis SJ, Kashyap S, Rosenwaks Z. Taking a basal follicle-stimulating hormone history is essential before initiating in vitro fertilization. Fertil Steril 2005;83:37-41.

61. Dahlgren E, Johansson S, Lindstedt G, Knutsson F, Oden A, Janson PO, Mattson LA, Crona N, Lundberg PA. Women with polycystic ovary syndrome wedge resected in 1956 to 1965: a long-term follow-up focusing on natural history and circulating hormones. Fertil Steril 1992;57:505-13.

62. Robertson DM, Cahir N, Findlay JK, Burger HG, Groome N. The biological and immunological characterization of inhibin A and B forms in human follicular fluid and plasma. J Clin Endocrinol Metab 1997;82:889-96.

63. Groome NP, Illingworth PJ, O'Brien M, Pai R, Rodger FE, Mather JP, McNeilly AS. Measurement of dimeric inhibin B throughout the human menstrual cycle. J Clin Endocrinol Metab 1996;81:1401-5.

64. Chu MC, Carmina E, Wang J, Lobo RA. Mullerian-inhibiting substance reflects ovarian findings in women with polycystic ovary syndrome better than does inhibin B. Fertil Steril 2005;84:1685-8.

65. Tsigkou A, Luisi S, De Leo V, Patton L, Gambineri A, Reis FM, Pasquali R, Petraglia F. High serum concentration of total inhibin in polycystic ovary syndrome. Fertil Steril 2008;

66. Adams J, Polson DW, Franks S. Prevalence of polycystic ovaries in women with anovulation and idiopathic hirsutism. Br Med J (Clin Res Ed) 1986;293:355-9.

67. Balen A, Michelmore K. What is polycystic ovary syndrome? Are national views important? Hum Reprod 2002;17:2219-27.

68. Dokras A, Bochner M, Hollinrake E, Markham S, Vanvoorhis B, Jagasia DH. Screening women with polycystic ovary syndrome for metabolic syndrome. Obstet Gynecol 2005;106:131-7.

69. Dunaif A. Drug insight: insulin-sensitizing drugs in the treatment of polycystic ovary syndrome--a reappraisal. Nat Clin Pract Endocrinol Metab 2008;4:272-83.

70. Matthaei S, Stumvoll M, Kellerer M, Haring HU. Pathophysiology and pharmacological treatment of insulin resistance. Endocr Rev 2000;21:585-618.

71. Beck-Nielsen H, Groop LC. Metabolic and genetic characterization of prediabetic states. Sequence of events leading to non-insulin-dependent diabetes mellitus. J Clin Invest 1994;94:1714-21.

72. Martin BC, Warram JH, Krolewski AS, Bergman RN, Soeldner JS, Kahn CR. Role of glucose and insulin resistance in development of type 2 diabetes mellitus: results of a 25-year follow-up study. Lancet 1992;340:925-9.

73. Burghen GA, Givens JR, Kitabchi AE. Correlation of hyperandrogenism with hyperinsulinism in polycystic ovarian disease. J Clin Endocrinol Metab 1980;50:113-6.

74. Svendsen PF, Nilas L, Norgaard K, Jensen JE, Madsbad S. Obesity, body composition and metabolic disturbances in polycystic ovary syndrome. Hum Reprod 2008;23:2113-21.

67

75. Toprak S, Yonem A, Cakir B, Guler S, Azal O, Ozata M, Corakci A. Insulin resistance in nonobese patients with polycystic ovary syndrome. Horm Res 2001;55:65-70.

76. Yucel A, Noyan V, Sagsoz N. The association of serum androgens and insulin resistance with fat distribution in polycystic ovary syndrome. Eur J Obstet Gynecol Reprod Biol 2006;126:81-6.

77. Dunaif A. Insulin resistance and the polycystic ovary syndrome: mechanism and implications for pathogenesis. Endocr Rev 1997;18:774-800.

78. Gennarelli G, Holte J, Berglund L, Berne C, Massobrio M, Lithell H. Prediction models for insulin resistance in the polycystic ovary syndrome. Hum Reprod 2000;15:2098-102.

79. Dunaif A, Segal KR, Futterweit W, Dobrjansky A. Profound peripheral insulin resistance, independent of obesity, in polycystic ovary syndrome. Diabetes 1989;38:1165-74.

80. Diamanti-Kandarakis E, Mitrakou A, Hennes MM, Platanissiotis D, Kaklas N, Spina J, Georgiadou E, Hoffmann RG, Kissebah AH, Raptis S. Insulin sensitivity and antiandrogenic therapy in women with polycystic ovary syndrome. Metabolism 1995;44:525-31.

81. Kowalska I, Straczkowski M, Nikolajuk A, Adamska A, Karczewska-Kupczewska M, Otziomek E, Wolczynski S, Gorska M. Serum visfatin in relation to insulin resistance and markers of hyperandrogenism in lean and obese women with polycystic ovary syndrome. Hum Reprod 2007;22:1824-9.

82. Svendsen PF, Madsbad S, Nilas L, Paulsen SK, Pedersen SB. Expression of 11beta-hydroxysteroid dehydrogenase 1 and 2 in subcutaneous adipose tissue of lean and obese women with and without polycystic ovary syndrome. Int J Obes (Lond) 2009;33:1249-56.

83. Tosi F, Dorizzi R, Castello R, Maffeis C, Spiazzi G, Zoppini G, Muggeo M, Moghetti P. Body fat and insulin resistance independently predict increased serum C-reactive protein in hyperandrogenic women with polycystic ovary syndrome. Eur J Endocrinol 2009;161:737-45.

84. Baillargeon JP, Carpentier A. Role of insulin in the hyperandrogenemia of lean women with polycystic ovary syndrome and normal insulin sensitivity. Fertil Steril 2007;88:886-93.

85. Holte J, Bergh T, Berne C, Berglund L, Lithell H. Enhanced early insulin response to glucose in relation to insulin resistance in women with polycystic ovary syndrome and normal glucose tolerance. J Clin Endocrinol Metab 1994;78:1052-8.

86. Morin-Papunen LC, Vauhkonen I, Koivunen RM, Ruokonen A, Tapanainen JS. Insulin sensitivity, insulin secretion, and metabolic and hormonal parameters in healthy women and women with polycystic ovarian syndrome. Hum Reprod 2000;15:1266-74.

87. Vrbikova J, Cibula D, Dvorakova K, Stanicka S, Sindelka G, Hill M, Fanta M, Vondra K, Skrha J. Insulin sensitivity in women with polycystic ovary syndrome. J Clin Endocrinol Metab 2004;89:2942-5.

88. Belosi C, Selvaggi L, Apa R, Guido M, Romualdi D, Fulghesu AM, Lanzone A. Is the PCOS diagnosis solved by ESHRE/ASRM 2003

68

consensus or could it include ultrasound examination of the ovarian stroma? Hum Reprod 2006;21:3108-15.

89. Barber TM, Wass JA, McCarthy MI, Franks S. Metabolic characteristics of women with polycystic ovaries and oligo-amenorrhoea but normal androgen levels: implications for the management of polycystic ovary syndrome. Clin Endocrinol (Oxf) 2007;66:513-7.

90. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985;28:412-9.

91. Katz A, Nambi SS, Mather K, Baron AD, Follmann DA, Sullivan G, Quon MJ. Quantitative insulin sensitivity check index: a simple, accurate method for assessing insulin sensitivity in humans. J Clin Endocrinol Metab 2000;85:2402-10.

92. Matsuda M, DeFronzo RA. Insulin sensitivity indices obtained from oral glucose tolerance testing: comparison with the euglycemic insulin clamp. Diabetes Care 1999;22:1462-70.

93. Attia N, Tamborlane WV, Heptulla R, Maggs D, Grozman A, Sherwin RS, Caprio S. The metabolic syndrome and insulin-like growth factor I regulation in adolescent obesity. J Clin Endocrinol Metab 1998;83:1467-71.

94. Bergh C, Carlsson B, Olsson JH, Selleskog U, Hillensjo T. Regulation of androgen production in cultured human thecal cells by insulin-like growth factor I and insulin. Fertil Steril 1993;59:323-31.

95. Ferry RJ, Jr., Cerri RW, Cohen P. Insulin-like growth factor binding proteins: new proteins, new functions. Horm Res 1999;51:53-67.

96. Shimasaki S, Ling N. Identification and molecular characterization of insulin-like growth factor binding proteins (IGFBP-1, -2, -3, -4, -5 and -6). Prog Growth Factor Res 1991;3:243-66.

97. San Roman GA, Magoffin DA. Insulin-like growth factor binding proteins in ovarian follicles from women with polycystic ovarian disease: cellular source and levels in follicular fluid. J Clin Endocrinol Metab 1992;75:1010-6.

98. Wareham NJ, Phillips DI, Byrne CD, Hales CN. The 30 minute insulin incremental response in an oral glucose tolerance test as a measure of insulin secretion. Diabet Med 1995;12:931.

99. Tripathy D, Wessman Y, Gullstrom M, Tuomi T, Groop L. Importance of obtaining independent measures of insulin secretion and insulin sensitivity during the same test: results with the Botnia clamp. Diabetes Care 2003;26:1395-401.

100. Lillioja S, Nyomba BL, Saad MF, Ferraro R, Castillo C, Bennett PH, Bogardus C. Exaggerated early insulin release and insulin resistance in a diabetes-prone population: a metabolic comparison of Pima Indians and Caucasians. J Clin Endocrinol Metab 1991;73:866-76.

101. Ehrmann DA, Barnes RB, Rosenfield RL, Cavaghan MK, Imperial J. Prevalence of impaired glucose tolerance and diabetes in women with polycystic ovary syndrome. Diabetes Care 1999;22:141-6.

102. Legro RS, Kunselman AR, Dodson WC, Dunaif A. Prevalence and predictors of risk for type 2 diabetes mellitus and impaired glucose

69

tolerance in polycystic ovary syndrome: a prospective, controlled study in 254 affected women. J Clin Endocrinol Metab 1999;84:165-9.

103. Gambineri A, Pelusi C, Manicardi E, Vicennati V, Cacciari M, Morselli-Labate AM, Pagotto U, Pasquali R. Glucose intolerance in a large cohort of mediterranean women with polycystic ovary syndrome: phenotype and associated factors. Diabetes 2004;53:2353-8.

104. Grundy SM, Cleeman JI, Daniels SR, Donato KA, Eckel RH, Franklin BA, Gordon DJ, Krauss RM, Savage PJ, Smith SC, Jr., Spertus JA, Costa F. Diagnosis and management of the metabolic syndrome: an American Heart Association/National Heart, Lung, and Blood Institute scientific statement. Curr Opin Cardiol 2006;21:1-6.

105. Vrbikova J, Vondra K, Cibula D, Dvorakova K, Stanicka S, Sramkova D, Sindelka G, Hill M, Bendlova B, Skrha J. Metabolic syndrome in young Czech women with polycystic ovary syndrome. Hum Reprod 2005;20:3328-32.

106. Glueck CJ, Papanna R, Wang P, Goldenberg N, Sieve-Smith L. Incidence and treatment of metabolic syndrome in newly referred women with confirmed polycystic ovarian syndrome. Metabolism 2003;52:908-15.

107. Ehrmann DA, Liljenquist DR, Kasza K, Azziz R, Legro RS, Ghazzi MN. Prevalence and predictors of the metabolic syndrome in women with polycystic ovary syndrome. J Clin Endocrinol Metab 2006;91:48-53.

108. Apridonidze T, Essah PA, Iuorno MJ, Nestler JE. Prevalence and characteristics of the metabolic syndrome in women with polycystic ovary syndrome. J Clin Endocrinol Metab 2005;90:1929-35.

109. Ni RM, Mo Y, Chen X, Zhong J, Liu W, Yang D. Low prevalence of the metabolic syndrome but high occurrence of various metabolic disorders in Chinese women with polycystic ovary syndrome. Eur J Endocrinol 2009;161:411-8.

110. Roa Barrios M, Arata-Bellabarba G, Valeri L, Velazquez-Maldonado E. [Relationship between the triglyceride/high-density lipoprotein-cholesterol ratio, insulin resistance index and cardiometabolic risk factors in women with polycystic ovary syndrome]. Endocrinol Nutr 2009;56:59-65.

111. Dewailly D, Contestin M, Gallo C, Catteau-Jonard S. Metabolic syndrome in young women with the polycystic ovary syndrome: revisiting the threshold for an abnormally decreased high-density lipoprotein cholesterol serum level. BJOG 2010;117:175-80.

112. Holte J, Gennarelli G, Berne C, Bergh T, Lithell H. Elevated ambulatory day-time blood pressure in women with polycystic ovary syndrome: a sign of a pre-hypertensive state? Hum Reprod 1996;11:23-8.

113. Akram T, Hasan S, Imran M, Karim A, Arslan M. Association of polycystic ovary syndrome with cardiovascular risk factors. Gynecol 2001;26:47-53.

114. Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) final report. Circulation 2002;106:3143-421.

115. Legro RS, Blanche P, Krauss RM, Lobo RA. Alterations in low-density lipoprotein and high-density lipoprotein subclasses among Hispanic women

70

with polycystic ovary syndrome: influence of insulin and genetic factors. Fertil Steril 1999;72:990-5.

116. Wild RA, Alaupovic P, Parker IJ. Lipid and apolipoprotein abnormalities in hirsute women. I. The association with insulin resistance. Am J Obstet Gynecol 1992;166:1191-6; discussion 1196-7.

117. Rajkhowa M, Neary RH, Kumpatla P, Game FL, Jones PW, Obhrai MS, Clayton RN. Altered composition of high density lipoproteins in women with the polycystic ovary syndrome. J Clin Endocrinol Metab 1997;82:3389-94.

118. Conway GS, Agrawal R, Betteridge DJ, Jacobs HS. Risk factors for coronary artery disease in lean and obese women with the polycystic ovary syndrome. Clin Endocrinol (Oxf) 1992;37:119-25.

119. Pouliot MC, Despres JP, Lemieux S, Moorjani S, Bouchard C, Tremblay A, Nadeau A, Lupien PJ. Waist circumference and abdominal sagittal diameter: best simple anthropometric indexes of abdominal visceral adipose tissue accumulation and related cardiovascular risk in men and women. Am J Cardiol 1994;73:460-8.

120. Lord J, Thomas R, Fox B, Acharya U, Wilkin T. The central issue? Visceral fat mass is a good marker of insulin resistance and metabolic disturbance in women with polycystic ovary syndrome. BJOG 2006;113:1203-9.

121. Lloyd-Jones D, Adams RJ, Brown TM, Carnethon M, Dai S, De Simone G, Ferguson TB, Ford E, Furie K, Gillespie C, Go A, Greenlund K, Haase N, Hailpern S, Ho PM, Howard V, Kissela B, Kittner S, Lackland D, Lisabeth L, Marelli A, McDermott MM, Meigs J, Mozaffarian D, Mussolino M, Nichol G, Roger VL, Rosamond W, Sacco R, Sorlie P, Stafford R, Thom T, Wasserthiel-Smoller S, Wong ND, Wylie-Rosett J. Executive summary: heart disease and stroke statistics--2010 update: a report from the American Heart Association. Circulation 2010;121:948-54.

122. Bello N, Mosca L. Epidemiology of coronary heart disease in women. Prog Cardiovasc Dis 2004;46:287-95.

123. Mosca L, Manson JE, Sutherland SE, Langer RD, Manolio T, Barrett-Connor E. Cardiovascular disease in women: a statement for healthcare professionals from the American Heart Association. Writing Group. Circulation 1997;96:2468-82.

124. Carmina E, Chu MC, Longo RA, Rini GB, Lobo RA. Phenotypic variation in hyperandrogenic women influences the findings of abnormal metabolic and cardiovascular risk parameters. J Clin Endocrinol Metab 2005;90:2545-9.

125. Carmina E, Orio F, Palomba S, Longo RA, Cascella T, Colao A, Lombardi G, Rini GB, Lobo RA. Endothelial dysfunction in PCOS: role of obesity and adipose hormones. Am J Med 2006;119:356 e1-6.

126. Giallauria F, Orio F, Palomba S, Lombardi G, Colao A, Vigorito C. Cardiovascular risk in women with polycystic ovary syndrome. J Cardiovasc Med (Hagerstown) 2008;9:987-92.

127. Lo JC, Feigenbaum SL, Yang J, Pressman AR, Selby JV, Go AS. Epidemiology and adverse cardiovascular risk profile of diagnosed polycystic ovary syndrome. J Clin Endocrinol Metab 2006;91:1357-63.

71

128. Talbott EO, Zborowskii JV, Boudraux MY. Do women with polycystic ovary syndrome have an increased risk of cardiovascular disease? Review of the evidence. Minerva Ginecol 2004;56:27-39.

129. Legro RS. Polycystic ovary syndrome and cardiovascular disease: a premature association? Endocr Rev 2003;24:302-12.

130. Brinkworth GD, Noakes M, Moran LJ, Norman R, Clifton PM. Flow-mediated dilatation in overweight and obese women with polycystic ovary syndrome. BJOG 2006;113:1308-14.

131. Mather KJ, Verma S, Corenblum B, Anderson TJ. Normal endothelial function despite insulin resistance in healthy women with the polycystic ovary syndrome. J Clin Endocrinol Metab 2000;85:1851-6.

132. Kravariti M, Naka KK, Kalantaridou SN, Kazakos N, Katsouras CS, Makrigiannakis A, Paraskevaidis EA, Chrousos GP, Tsatsoulis A, Michalis LK. Predictors of endothelial dysfunction in young women with polycystic ovary syndrome. J Clin Endocrinol Metab 2005;90:5088-95.

133. Meyer C, McGrath BP, Teede HJ. Overweight women with polycystic ovary syndrome have evidence of subclinical cardiovascular disease. J Clin Endocrinol Metab 2005;90:5711-6.

134. Orio F, Jr., Palomba S, Cascella T, De Simone B, Di Biase S, Russo T, Labella D, Zullo F, Lombardi G, Colao A. Early impairment of endothelial structure and function in young normal-weight women with polycystic ovary syndrome. J Clin Endocrinol Metab 2004;89:4588-93.

135. Tarkun I, Arslan BC, Canturk Z, Turemen E, Sahin T, Duman C. Endothelial dysfunction in young women with polycystic ovary syndrome: relationship with insulin resistance and low-grade chronic inflammation. J Clin Endocrinol Metab 2004;89:5592-6.

136. Moran LJ, Hutchison SK, Meyer C, Zoungas S, Teede HJ. A comprehensive assessment of endothelial function in overweight women with and without polycystic ovary syndrome. Clin Sci (Lond) 2008;

137. Korhonen S, Hippelainen M, Vanhala M, Heinonen S, Niskanen L. The androgenic sex hormone profile is an essential feature of metabolic syndrome in premenopausal women: a controlled community-based study. Fertil Steril 2003;79:1327-34.

138. Elting MW, Korsen TJ, Rekers-Mombarg LT, Schoemaker J. Women with polycystic ovary syndrome gain regular menstrual cycles when ageing. Hum Reprod 2000;15:24-8.

139. Dahlgren E, Janson PO, Johansson S, Lapidus L, Oden A. Polycystic ovary syndrome and risk for myocardial infarction. Evaluated from a risk factor model based on a prospective population study of women. Acta Obstet Gynecol Scand 1992;71:599-604.

140. Cibula D, Cifkova R, Fanta M, Poledne R, Zivny J, Skibova J. Increased risk of non-insulin dependent diabetes mellitus, arterial hypertension and coronary artery disease in perimenopausal women with a history of the polycystic ovary syndrome. Hum Reprod 2000;15:785-9.

141. Wild S, Pierpoint T, McKeigue P, Jacobs H. Cardiovascular disease in women with polycystic ovary syndrome at long-term follow-up: a retrospective cohort study. Clin Endocrinol (Oxf) 2000;52:595-600.

72

142. Pierpoint T, McKeigue PM, Isaacs AJ, Wild SH, Jacobs HS. Mortality of women with polycystic ovary syndrome at long-term follow-up. J Clin Epidemiol 1998;51:581-6.

143. Shaw LJ, Bairey Merz CN, Azziz R, Stanczyk FZ, Sopko G, Braunstein GD, Kelsey SF, Kip KE, Cooper-Dehoff RM, Johnson BD, Vaccarino V, Reis SE, Bittner V, Hodgson TK, Rogers W, Pepine CJ. Postmenopausal women with a history of irregular menses and elevated androgen measurements at high risk for worsening cardiovascular event-free survival: results from the National Institutes of Health--National Heart, Lung, and Blood Institute sponsored Women's Ischemia Syndrome Evaluation. J Clin Endocrinol Metab 2008;93:1276-84.

144. Cheang KI, Nestler JE, Futterweit W. Risk of cardiovascular events in mothers of women with polycystic ovary syndrome. Endocr Pract 2008;14:1084-94.

145. Lind L, Pettersson K, Johansson K. Analysis of endothelium-dependent vasodilation by use of the radial artery pulse wave obtained by applanation tonometry. Clin Physiol Funct Imaging 2003;23:50-7.

146. Lind L, Fors N, Hall J, Marttala K, Stenborg A. A comparison of three different methods to evaluate endothelium-dependent vasodilation in the elderly: the Prospective Investigation of the Vasculature in Uppsala Seniors (PIVUS) study. Arterioscler Thromb Vasc Biol 2005;25:2368-75.

147. Hayward CS, Kraidly M, Webb CM, Collins P. Assessment of endothelial function using peripheral waveform analysis: a clinical application. J Am Coll Cardiol 2002;40:521-8.

148. Wilkinson IB, Hall IR, MacCallum H, Mackenzie IS, McEniery CM, van der Arend BJ, Shu YE, MacKay LS, Webb DJ, Cockcroft JR. Pulse-wave analysis: clinical evaluation of a noninvasive, widely applicable method for assessing endothelial function. Arterioscler Thromb Vasc Biol 2002;22:147-52.

149. Lind L. Endothelium-dependent vasodilation, insulin resistance and the metabolic syndrome in an elderly cohort: the Prospective Investigation of the Vasculature in Uppsala Seniors (PIVUS) study. Atherosclerosis 2008;196:795-802.

150. Albareda M, Rodriguez-Espinosa J, Murugo M, de Leiva A, Corcoy R. Assessment of insulin sensitivity and beta-cell function from measurements in the fasting state and during an oral glucose tolerance test. Diabetologia 2000;43:1507-11.

151. Koivunen R, Pouta A, Franks S, Martikainen H, Sovio U, Hartikainen AL, McCarthy MI, Ruokonen A, Bloigu A, Jarvelin MR, Morin-Papunen L. Fecundability and spontaneous abortions in women with self-reported oligo-amenorrhea and/or hirsutism: Northern Finland Birth Cohort 1966 Study. Hum Reprod 2008;23:2134-9.

152. Sundstrom I, Ildgruben A, Hogberg U. Treatment-related and treatment-independent deliveries among infertile couples, a long-term follow-up. Acta Obstet Gynecol Scand 1997;76:238-43.

153. Small CM, Manatunga AK, Klein M, Feigelson HS, Dominguez CE, McChesney R, Marcus M. Menstrual cycle characteristics: associations with fertility and spontaneous abortion. Epidemiology 2006;17:52-60.

73

154. Lockwood GM, Muttukrishna S, Groome NP, Matthews DR, Ledger WL. Mid-follicular phase pulses of inhibin B are absent in polycystic ovarian syndrome and are initiated by successful laparoscopic ovarian diathermy: a possible mechanism regulating emergence of the dominant follicle. J Clin Endocrinol Metab 1998;83:1730-5.

155. Farquhar C, Lilford RJ, Marjoribanks J, Vandekerckhove P. Laparoscopic 'drilling' by diathermy or laser for ovulation induction in anovulatory polycystic ovary syndrome. Cochrane Database Syst Rev 2007;CD001122.

156. George L, Granath F, Johansson AL, Anneren G, Cnattingius S. Environmental tobacco smoke and risk of spontaneous abortion. Epidemiology 2006;17:500-5.

157. Gnoth C, Schuring AN, Friol K, Tigges J, Mallmann P, Godehardt E. Relevance of anti-Mullerian hormone measurement in a routine IVF program. Hum Reprod 2008;23:1359-65.

158. Singer T, Barad DH, Weghofer A, Gleicher N. Correlation of antimullerian hormone and baseline follicle-stimulating hormone levels. Fertil Steril 2008;

159. Child TJ, Abdul-Jalil AK, Gulekli B, Tan SL. In vitro maturation and fertilization of oocytes from unstimulated normal ovaries, polycystic ovaries, and women with polycystic ovary syndrome. Fertil Steril 2001;76:936-42.

160. Tan SL, Child TJ. In-vitro maturation of oocytes from unstimulated polycystic ovaries. Reprod Biomed Online 2002;4 Suppl 1:18-23.

161. Steinberg HO, Chaker H, Leaming R, Johnson A, Brechtel G, Baron AD. Obesity/insulin resistance is associated with endothelial dysfunction. Implications for the syndrome of insulin resistance. J Clin Invest 1996;97:2601-10.

162. Suh HS, Park YW, Kang JH, Lee SH, Lee HS, Shim KW. Vascular endothelial dysfunction tested by blunted response to endothelium-dependent vasodilation by salbutamol and its related factors in uncomplicated pre-menopausal obese women. Int J Obes (Lond) 2005;29:217-22.

163. Perticone F, Ceravolo R, Candigliota M, Ventura G, Iacopino S, Sinopoli F, Mattioli PL. Obesity and body fat distribution induce endothelial dysfunction by oxidative stress: protective effect of vitamin C. Diabetes 2001;50:159-65.

164. Arcaro G, Zamboni M, Rossi L, Turcato E, Covi G, Armellini F, Bosello O, Lechi A. Body fat distribution predicts the degree of endothelial dysfunction in uncomplicated obesity. Int J Obes Relat Metab Disord 1999;23:936-42.

165. Brook RD, Bard RL, Rubenfire M, Ridker PM, Rajagopalan S. Usefulness of visceral obesity (waist/hip ratio) in predicting vascular endothelial function in healthy overweight adults. Am J Cardiol 2001;88:1264-9.

166. Park SH, Shim KW. Reduction in visceral adiposity is highly related to improvement in vascular endothelial dysfunction among obese women: an assessment of endothelial function by radial artery pulse wave analysis. Yonsei Med J 2005;46:511-8.

167. Lind L, Johansson L, Hulthe J, von Below C, Ahlstrom H. Vasodilation and visceral fat in elderly subjects: the Prospective Investigation of the

74

Vasculature in Uppsala Seniors (PIVUS) study. Atherosclerosis 2007;194:e64-71.

168. Diamanti-Kandarakis E, Alexandraki K, Protogerou A, Piperi C, Papamichael C, Aessopos A, Lekakis J, Mavrikakis M. Metformin administration improves endothelial function in women with polycystic ovary syndrome. Eur J Endocrinol 2005;152:749-56.

169. Paradisi G, Steinberg HO, Shepard MK, Hook G, Baron AD. Troglitazone therapy improves endothelial function to near normal levels in women with polycystic ovary syndrome. J Clin Endocrinol Metab 2003;88:576-80.

170. Jensterle M, Sebestjen M, Janez A, Prezelj J, Kocjan T, Keber I, Pfeifer M. Improvement of endothelial function with metformin and rosiglitazone treatment in women with polycystic ovary syndrome. Eur J Endocrinol 2008;159:399-406.

171. Romualdi D, Costantini B, Selvaggi L, Giuliani M, Cristello F, Macri F, Bompiani A, Lanzone A, Guido M. Metformin improves endothelial function in normoinsulinemic PCOS patients: a new prospective. Hum Reprod 2008;23:2127-33.

172. Hu WH, Qiao J, Wang LN, Tong J. [Clinical features of the metabolic syndrome in patients with polycystic ovary syndrome.]. Beijing 2010;42:159-63.

173. Weerakiet S, Bunnag P, Phakdeekitcharoen B, Wansumrith S, Chanprasertyothin S, Jultanmas R, Thakkinstian A. Prevalence of the metabolic syndrome in Asian women with polycystic ovary syndrome: using the International Diabetes Federation criteria. Gynecol Endocrinol 2007;23:153-60.

174. Winters SJ, Talbott E, Guzick DS, Zborowski J, McHugh KP. Serum testosterone levels decrease in middle age in women with the polycystic ovary syndrome. Fertil Steril 2000;73:724-9.

175. Sowers MF, Zheng H, McConnell D, Nan B, Karvonen-Gutierrez CA, Randolph JF, Jr. Testosterone, sex hormone-binding globulin and free androgen index among adult women: chronological and ovarian aging. Hum Reprod 2009;24:2276-85.

176. Robinson S, Kiddy D, Gelding SV, Willis D, Niththyananthan R, Bush A, Johnston DG, Franks S. The relationship of insulin insensitivity to menstrual pattern in women with hyperandrogenism and polycystic ovaries. Clin Endocrinol (Oxf) 1993;39:351-5.

177. Majumdar A, Singh TA. Comparison of clinical features and health manifestations in lean vs. obese Indian women with polycystic ovarian syndrome. J Hum Reprod Sci 2009;2:12-7.

178. Cibula D, Skrha J, Hill M, Fanta M, Haakova L, VrbIkova J, Zivny J. Prediction of insulin sensitivity in nonobese women with polycystic ovary syndrome. J Clin Endocrinol Metab 2002;87:5821-5.

179. Ding EL, Song Y, Malik VS, Liu S. Sex differences of endogenous sex hormones and risk of type 2 diabetes: a systematic review and meta-analysis. Jama 2006;295:1288-99.

180. Bhattacharya SM. Insulin resistance and overweight-obese women with polycystic ovary syndrome. Gynecol Endocrinol 2009;

181. Sternfeld B, Liu K, Quesenberry CP, Jr., Wang H, Jiang SF, Daviglus M, Fornage M, Lewis CE, Mahan J, Schreiner PJ, Schwartz SM, Sidney S,

75

Williams OD, Siscovick DS. Changes over 14 years in androgenicity and body mass index in a biracial cohort of reproductive-age women. J Clin Endocrinol Metab 2008;93:2158-65.

182. Hudecova M, Holte J, Olovsson M, Sundstrom Poromaa I. Long-term follow-up of patients with polycystic ovary syndrome: reproductive outcome and ovarian reserve. Hum Reprod 2009;24:1176-83.

183. Zethelius B, Berglund L, Hanni A, Berne C. The interaction between impaired acute insulin response and insulin resistance predict type 2 diabetes and impairment of fasting glucose: report from a 20-year follow-up in the Uppsala Longitudinal Study of Adult Men - ULSAM. Ups J Med Sci 2008;113:117-30.

184. Haffner SM, Miettinen H, Gaskill SP, Stern MP. Decreased insulin secretion and increased insulin resistance are independently related to the 7-year risk of NIDDM in Mexican-Americans. Diabetes 1995;44:1386-91.

185. Kahn BB. Type 2 diabetes: when insulin secretion fails to compensate for insulin resistance. Cell 1998;92:593-6.

186. Cho LW, Jayagopal V, Kilpatrick ES, Atkin SL. The mean and the biological variation of insulin resistance does not differ between polycystic ovary syndrome and type 2 diabetes. Ann Clin Biochem 2009;46:218-21.

187. Gerich JE. Is insulin resistance the principal cause of type 2 diabetes? Diabetes Obes Metab 1999;1:257-63.

188. Mauvais-Jarvis F, Kahn CR. Understanding the pathogenesis and treatment of insulin resistance and type 2 diabetes mellitus: what can we learn from transgenic and knockout mice? Diabetes Metab 2000;26:433-48.

189. Kosaka K, Kuzuya T, Hagura R, Yoshinaga H. Insulin response to oral glucose load is consistently decreased in established non-insulin-dependent diabetes mellitus: the usefulness of decreased early insulin response as a predictor of non-insulin-dependent diabetes mellitus. Diabet Med 1996;13:S109-19.

190. Holte J, Gennarelli G, Wide L, Lithell H, Berne C. High prevalence of polycystic ovaries and associated clinical, endocrine, and metabolic features in women with previous gestational diabetes mellitus. J Clin Endocrinol Metab 1998;83:1143-50.

!� ��#�� #��

%� �0�2��$�� 3��

!�� 3�� 4�#��#�� 4��5�!��6�� 7�� 8��6��4�6�� 9�� $�9� ��#��$�� 3�� 5�:�� ;��4�+��4� ��6�� <��#��$�� 3�� =5>

$�� 0�� 5��5��0��0��0��0��*(+1+/.

��

˘ ˇ ˆuu.diva-portal.org/smash/get/diva2:314588/fulltext01.pdfphenotype without hyperandrogenism....

Documents