| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Division of Reproductive Medicine, Department of Obstetrics and Gynaecology (I.J.G.K., T.J.M.K., P.G.A.H., R.H., C.B.L.) and Department of Internal Medicine (E.H.S., Y.M.S., R.T.d.J.), de Boelelaan, 1117 Amsterdam, The Netherlands; and Department of Internal Medicine (C.D.A.S.), Academic Hospital Maastricht, P. Debyelaan 25, 6202 AZ Maastricht, The Netherlands
Address all correspondence and requests for reprints to: Iris J. G. Ketel, de Boelelaan 1118, Room 0Z106, 1081 HV Amsterdam, The Netherlands. E-mail: ijg.ketel{at}vumc.nl.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Objective: The objective of the study was to investigate whether insulin sensitivity and insulin’s effects on the microcirculation are impaired in normal-weight and obese women with PCOS.
Design and Population: Thirty-five women with PCOS (19 normal weight and 16 obese) and 27 age- and body mass index-matched controls (14 normal weight and 13 obese) were included. Metabolic Insulin sensitivity (isoglycemic-hyperinsulinemic clamp) and microvascular insulin sensitivity [endothelium dependent (acetylcholine [ACh])] and endothelium-independent [sodium nitroprusside (SNP)] vasodilation with laser Doppler flowmetry was assessed at baseline and during hyperinsulinemia.
Main Outcome Measures: Metabolic insulin sensitivity (M/I value) and the area under the response curves to ACh and SNP curves were measured to assess microcirculatory function at baseline and during insulin infusion (microvascular insulin sensitivity).
Results: Obese women were more insulin resistant than normal-weight women (P < 0.001), and obese PCOS women were more resistant than obese controls (P = 0.02). In contrast, normal-weight women with PCOS had similar insulin sensitivity, compared with normal-weight women without PCOS. Baseline responses to ACh showed no difference in the four groups. ACh responses during insulin infusion were significantly greater in normal-weight PCOS and controls than in obese PCOS and controls. PCOS per se had no significant influence on ACh responses during insulin infusion. During hyperinsulinemia, SNP-dependent vasodilatation did not significantly increase, compared with baseline in the four groups.
Conclusion: PCOS per se was not associated with impaired metabolic insulin sensitivity in normal-weight women but aggravates impairment of metabolic insulin sensitivity in obese women. In obese but not normal-weight women, microvascular and metabolic insulin sensitivity are decreased, independent of PCOS. Therefore, obese PCOS women in particular may be at increased risk of metabolic and cardiovascular diseases.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
However, a substantial number of women with PCOS are not obese. Therefore, the question arises whether such women are also at a greater risk of diabetes and cardiovascular disease, i.e. whether PCOS increases these risks independently of the presence of obesity. Similarly, in obese women with PCOS, the question remains whether the risks conferred by obesity and PCOS are additive.
To gain more insight into these issues, we studied insulin sensitivity and microvascular function (including the microvascular response to insulin) in normal-weight and obese women with and without PCOS. Microvascular dysfunction in particular may contribute to: 1) hypertension by raising peripheral vascular resistance and large artery stiffness (5, 6); 2) insulin resistance and diabetes by limiting the timely access of glucose, and 3) insulin to their target tissues (7, 8) and cardiovascular diseases that are in part caused by microangiopathy, such as heart failure and nephropathy (9, 10). Insulin resistance and microvascular dysfunction are thus thought to be important pathways linking obesity and PCOS to cardiovascular diseases (4, 11, 12). Normally, insulin increases nitric oxide synthesis at the site of the endothelium of the microcirculation and causes vasodilation in the skeletal muscle of healthy individuals (13). Insulin also enhances endothelium-dependent vasodilation induced by acetylcholine (14). Microvascular insulin resistance is characterized by impairments in these effects of insulin (4). As a consequence, nutritive capillary perfusion in skeletal muscle is diminished, which impairs glucose uptake and thus causes metabolic insulin resistance (4).
Impaired microvascular function is thus associated with less vasodilatory response to acetylcholine, which can be measured by skin flow changes after iontophoretic application. When measured under hyperinsulinemic conditions, the amount of acetylcholine-mediated vasodilation can be considered as the degree of microvascular insulin sensitivity. To address this issue of microvascular functioning, in particular the reaction to insulin, the aim of the study was to measure insulin sensitivity by means of the hyperinsulinemic, isoglycemic clamp and the microvascular response to acetylcholine under baseline and hyperinsulinemic conditions in normal-weight and obese women with and without PCOS.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Four groups of subjects were recruited: 19 normal-weight (body mass index <25 kg/m2) women with PCOS, 14 normal-weight controls, 16 obese (body mass index >30 kg/m2) women with PCOS, and 13 obese controls. All cases had the diagnosis PCOS according to the European Society of Human Reproduction and Embryology/American Society of Reproductive Medicine criteria (15). All women with PCOS had oligo- or amenorrhea and polycystic ovarian morphology on transvaginal ultrasound; 13 obese women and 16 normal-weight women had hyperandrogenism estimated by clinical and/or laboratory measurements, and 13 obese and 18 lean PCOS patients fulfilled all three Rotterdam criteria. The controls were as a group matched for age and weight with the PCOS groups. All were healthy as judged by medical history, normal fasting glucose levels, nondiabetic (16), and normotensive (<140/90 mm Hg) as determined by triplicate office blood pressure measurement. All controls had a normal ovarian morphology on ultrasound, a regular cycle, and no clinical or laboratory features of PCOS. All participants were Caucasian and nonsmokers and had not used any medication or oral contraceptives for the last three months. Exclusion criteria for all subjects included abnormal TSH, prolactin, 17
-hydroxyprogesterone (17-OHP) or dehydroepiandrosterone sulfate (DHEAS) serum levels, and pregnancy. All PCOS patients and some controls were recruited from the department of Reproductive Medicine of the Vrije Universiteit University Medical Center, Amsterdam. Other controls were recruited through advertisements.
All measurements were scheduled for PCOS patients on cycle d 13–16 and for controls on cycle d 5–9 because metabolic and vascular physiology in these periods is thought to be least influenced by ovulation and progesterone production (17, 18). The study was approved by the Medical Ethical committee of the Vrije Universiteit University Medical Center and all participants signed an informed consent.
Study design
All individuals underwent the study protocol as shown in Fig. 1
. All participants came to the clinic after a 10-h overnight fast. They were asked to refrain from caffeine and alcohol containing beverages for 10 h preceding the test. All experiments were conducted in a quiet, temperature-controlled (23.4 ± 0.4 C) room after 30 min of acclimatization.
|
Microvascular endothelium-(in)dependent vasodilatation was evaluated by iontophoresis of acetylcholine (ACh) and sodium nitroprusside (SNP) in combination with laser Doppler flowmetry (Periflux 5010; Perimed, Stockholm, Sweden) in the skin as described previously (6, 14, 19) both before and during the hyperinsulinemic isoglycemic clamp (Fig. 1
). The measurements during the clamp reflect the responses of microcirculatory function to hyperinsulinemia, which is an important determinant of the timely access of glucose and insulin to their targets tissues (i.e. microvascular insulin sensitivity) (4, 12).
Areas under the curve (AUC) were used for further analyses of the blood flow responses to ACh and SNP, respectively. The response of ACh- and SNP-induced vasodilatation to insulin was defined as the AUC value during minus the value before insulin infusion. The same operator performed the measurements for each patient to avoid interobserver variability. The intraobserver coefficients of variation determined in eight subjects on two occasions were 15.2 ± 10.8% for ACh-mediated vasodilation and 16.6 ± 10.7% for SNP-mediated vasodilation. ACh and SNP were administered in their respective vehicle solutions. Because responses to vehicle are known to be negligible compared with responses to ACh and SNP (4), we refrained from doing vehicle control studies. Finally, Fig. 1 makes clear that the response of ACh- and SNP-induced vasodilatation to insulin, as defined above, may be affected by nonspecific changes due to time and volume expansion. In a previous study, we have addressed this issue and observed that such nonspecific changes are relatively small in normal-weight and obese women without PCOS (4). To further address this, we did a time- and volume-control study in 10 normal-weight and 10 obese women with PCOS included in the present study and found small effects that were similar to those in women without PCOS (4). We therefore here report all data without adjustment for such affects, thus assuming that any such effects are similar in all four groups.
Insulin sensitivity
Insulin sensitivity with regard to glucose uptake (metabolic insulin sensitivity) was determined with the hyperinsulinemic, isoglycemic clamp technique as described previously (20), with glucose concentrations clamped at fasting level and an insulin infusion rate of 40mU·m–2·min–1. The M-value was defined as the glucose infusion rate during the second hour of the clamp expressed per kilogram of body weight. The metabolic insulin sensitivity (M/I value) is the M-value expressed per unit of plasma insulin concentration.
Blood pressure
Ambulatory 24-h blood pressure monitoring (Spacelabs 90207, Redmond, WA) was performed as described previously (6, 19). During study days, blood pressure measurements were determined before and during the hyperinsulinemic isoglycemic clamp (Colin Press-Mate BP-8800; Colin, Komaki City, Japan).
Laboratory data
Blood was collected in the fasting state and during hyperinsulinemia (Fig. 1). All serum samples were immediately centrifuged at 4 C and stored at –80 C. Insulin was measured by using an immunometric assay with luminescence (Bayer Diagnostics, Mijdrecht, The Netherlands). The interassay coefficient of variation was less than 5% and the intraassay coefficient of variation was less than 10%. Glucose was measured with the hexokinase method (Roche Diagnostics, Mannheim, Germany). The inter- and intraassay coefficients of variation were both less than 2%. An immunometric assay (Delfia; Wallac, Turku, Finland) was used to measure LH and FSH in serum. Estradiol levels were measured using a RIA with double antibody (Diasorin, Saluggia Italy). A competitive immunoassay with luminescence was used to measure progesterone (Abbott Laboratories Diagnostic Division, Abbott Park, IL). Testosterone and 17-OHP levels were measured with a competitive immunoassay (DRG Instruments, Marburg, Germany). The SHBG level was measured by using an immunometric assay (Immulite 2500; Diagnostic Products Corp., Los Angeles, CA). The free androgen index was calculated as the total testosterone divided by the SHBG level x 100. A RIA was used to measure levels of androstenedione (Diagnostic Systems Laboratories, Webster, TX) and DHEAS (Diagnostic Products). TSH, total cholesterol, high-density lipoprotein (HDL)-cholesterol, and triglycerides were measured with an enzymatic colorimetric assay (Roche Diagnostics). The low-density lipoprotein-cholesterol level was calculated according to the Friedewald formula. Free fatty acids (FFAs) were measured with an enzymatic colorimetric test (NEFA-C; Wako Chemicals, Neuss Germany). All laboratory assays were performed in the clinical chemistry/endocrine laboratory of the Vrije Universiteit University Medical Center.
Statistical analyses
All analyses were performed with SPSS 15.0 (Chicago, IL). de Jongh et al. (4) detected an increase of ACh mediated vasodilation of 537 ± 133% in normal-weight women, compared with 345 ± 159% in obese women. An of 0.05 and a β of 0.05 therefore require a minimal number of 13 subjects to detect the same differences in PCOS patients.
The distribution of variables was tested for normality. Data are expressed as mean ± SEM or as median (interquartile range), as appropriate. To examine the differences in anthropometric and laboratory differences between normal-weight PCOS, obese PCOS, and matched controls, we used one-way ANOVA with post hoc analyses (Bonferroni). Insulin sensitivity, expressed as M/I, was also tested with one-way ANOVA with post hoc analyses (Bonferroni). The general linear model (GLM) univariate procedure, with body mass index (BMI) group (i.e. normal weight and obese), PCOS group (i.e. PCOS and control), and BMI group vs. PCOS group interaction as factors in the model was used. GLM tests were done for differences of ACh and SNP responses before and during insulin infusion between PCOS vs. controls, normal-weight vs. obese, and the possible effect of BMI on PCOS vs. controls. A probability value of P < 0.05 was considered significant.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In Table 1 the anthropometrics are displayed; obese PCOS women and controls had a larger waist and a higher waist to hip ratio, compared with normal-weight PCOS women and controls. Obese controls were slightly more obese with a smaller waist circumference, compared with obese women with PCOS.
|
Two obese PCOS individuals, two normal-weight PCOS individuals, two obese controls, and one normal-weight control did not complete the blood pressure measurements because of intolerance to the continuous presence of the cuff around the arm during the night (of these individuals we report only daytime measurements). Hemodynamics were not different between normal-weight PCOS and normal-weight controls. Twenty-four-hour heart rate was higher in obese PCOS women, compared with normal-weight PCOS women. Next, 24-h systolic blood pressure was higher in obese controls, compared with the normal-weight controls. With regard to endocrine profiles, ovarian androgen levels and LH were elevated in both normal-weight and obese PCOS patients, and FSH was slightly lower in the normal-weight PCOS women, compared with normal-weight controls.
Metabolic insulin sensitivity
The obese groups had a significantly lower M/I value than the normal-weight groups (Table 1 and Fig. 2). Obese women with PCOS had a significantly lower M/I value, compared with obese controls. Insulin sensitivity was not different between normal-weight PCOS and controls.
|
Endothelium-dependent vasodilation (response to ACh)
The microvascular responses to ACh before insulin infusion showed no significant difference in the PCOS women, compared with controls without PCOS, or in the obese group, compared with the normal-weight group (Table 2 and Fig. 3).
|
|
). Microvascular function during insulin infusion (microvascular insulin sensitivity)
Endothelium-dependent vasodilation (response to insulin)
ACh responses during, and the ACh response to, insulin infusion were significantly diminished in obese compared with normal-weight women, and this difference was similar for women with or without PCOS (Table 2 and Fig. 3). PCOS per se had no significant influence on the ACh response during insulin infusion.
Endothelium-independent vasodilation (response to insulin)
Insulin infusion did not increase the response to SNP. The response to SNP during insulin infusion was not significantly different in women with vs. those without PCOS or in obese vs. normal-weight women (Table 2).
There was no evidence for a statistically significant interaction between PCOS and BMI with regard to microvascular function (Table 2).
Time- and volume-control study
A time- and volume-control study was performed in 10 normal-weight and 10 obese women with PCOS included in the present study. There was no difference of the baseline of the time- and volume-control study as compared with the baseline of the clamp day (565.1 ± 409.9 vs. 442.8 ± 329.4 for ACh and 1095.4 ± 927.7 vs. 1043.3 ± 1154.6 for SNP in obese women with PCOS; and 499.7 ± 279.8 vs. 534.0 ± 349.1 for ACh and 1127.7 ± 714.7 vs. 1139.5 ± 703.1 for SNP in normal weight women with PCOS). In addition, there was no microvascular response to saline infusion (442.8 ± 329.4 vs. 536.4 ± 439.7 for ACh and 1043.3 ± 1154 vs. 1154.6 ± 648.0 for SNP in obese women with PCOS; and 534.0 ± 349.1 vs. 582.1 ± 305.1 for ACh and 1139.5 ± 703.1 vs. 1174.0 ± 586.4 for SNP in normal weight women with PCOS).
Skin temperature and insulin-mediated changes in skin temperature during microvascular measurements did not differ among the four groups (data not shown).
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
We focused on insulin resistance and microvascular function because these are thought to be important pathways linking obesity and, possibly, PCOS to cardiovascular and metabolic diseases, notably hypertension, diabetes, heart failure, and nephropathy (4, 5, 6, 7, 8, 9, 10, 11, 12). We acknowledge that the literature addressing the association between PCOS and cardiovascular disease is contradictory (21, 22). Despite the increased prevalence of cardiovascular risk factors in middle-aged PCOS women, large cohort studies (22, 23) did not find a higher risk of cardiovascular death in women with PCOS. However, large cohort studies might suffer from mixing populations of obese and normal-weight women, which, when analyzed together, might average out differences between subgroups.
The data of the present study thus suggest that much of the cardiovascular and metabolic risk in women with PCOS is caused by obesity. In the presence of obesity, PCOS may further increase this risk because obese women with PCOS had greater metabolic insulin resistance than obese women without PCOS. In the absence of obesity, women with PCOS had normal microvascular function and normal metabolic insulin sensitivity. To our knowledge the present study is the first to demonstrate that microvascular insulin resistance is a feature of obesity but not of PCOS per se.
Various other studies have addressed the vasodilator effect of insulin in women with PCOS (24, 25, 26). First, Carmassi et al. (24) measured endothelium-dependent vasodilatory effects of intraarterial insulin infusion in PCOS women with and without insulin resistance and found an impaired effect of insulin only in metabolically insulin-resistant women with PCOS, compared with metabolically insulin-sensitive women with and without PCOS. Second, ex vivo studies showed absence of insulin reducing contractor response to norepinephrine in PCOS patients indicative for vascular stiffness (25). Finally, Paradisi et al. demonstrated impaired leg blood flow responses to hyperinsulinemia during a euglycemic hyperinsulinemic clamp in obese PCOS compared with obese controls (26).
An important contrast to these three reports is that the present study evaluated the effects of insulin on the microcirculation, which is important for insulin mediated glucose uptake (i.e. microvascular insulin resistance) rather than on nonnutritive resistance vessels and total blood flow.
Somewhat unexpectedly, cholinergic microvascular endothelium-dependent vasodilation was similar in all groups, i.e. appeared unaffected by the presence of obesity or PCOS. We and others have previously demonstrated impaired microvascular function in obesity (4, 27). Although the obese women we studied had severe obesity, they were selected for otherwise being very young and healthy, and this may explain why cholinergic microvascular vasodilation was comparable with that in normal-weight women, although microvascular function clearly was not entirely normal, as shown by the impaired response to insulin in the obese women. Studies on microvascular function in PCOS have shown inconsistent results. In accordance with our data, Bickerton et al. (28) found no differences in reactive hyperemic blood flow of the forearm microcirculation. On the other hand, other studies (29, 30) found impaired endothelial function of the forearm microcirculation of obese women with PCOS. An explanation for this discrepancy might be that in these studies (29, 30), women with PCOS were significantly more obese than their controls. Therefore, these data suggest that obesity and not PCOS might play a more important role in the development of microcirculatory dysfunction. Indeed, Dokras et al. (31), measuring forearm blood flow responses with occlusion plethysmography to intraarterial infusion of endothelium-dependent and endothelium independent dilators, found only an impaired microvascular function in obese, insulin-resistant women independent of PCOS, compared with normal-weight, insulin-sensitive women with and without PCOS.
PCOS women in the present study exhibited an increased baseline response of SNP-mediated vasodilation. SNP acts directly on vascular smooth muscle as nitric oxide donor. Thus, microcirculatory smooth muscle in PCOS patients may have greater than normal sensitivity to nitric oxide. On the other hand, the few other studies reporting on SNP effects in PCOS are contradicting with no or opposite effects (25, 30, 31), and further studies of this issue are needed.
Metabolic insulin resistance is thought to play a central role in the pathogenesis of PCOS. In the present study, we did not find resistance to insulin in normal-weight women with PCOS as measured with the gold standard technique, i.e. the isoglycemic hyperinsulinemic clamp. Because of the small sample size of our study, small differences in insulin sensitivity in normal-weight PCOS women, compared with normal-weight controls, could have been missed. Nevertheless, the one study that was able to asses insulin sensitivity, as measured by euglycemic hyperinsulinemic clamp, in a large group of normal-weight women with and without PCOS (32) also found that there was no difference in insulin sensitivity between normal-weight women with PCOS, compared with normal-weight controls. It must be stressed, however, that we studied Caucasian women, and our findings are not necessarily valid for other ethnicities.
This study has several limitations. First, it was cross-sectional, and therefore, any causal inference on the link between microcirculatory function and metabolic insulin resistance should be made with caution. Second, although muscle tissue is the main peripheral site of insulin-mediated glucose uptake and vascular resistance, we studied skin and not muscle microvascular function because in skin, iontophoresis can be applied noninvasively. However, comparable insulin-mediated metabolic and microvascular effects can be demonstrated in skin and muscle (14). In addition, skin microvascular function is associated with blood pressure (6), and hypertension is characterized by defects in both muscle and skin microvascular function (19, 33). Thus, the study of skin microvascular function seems a reasonable model of muscle microvascular function. Third, the small sample size of the present study may have led to type 2 errors, especially because microvascular measurements have a relatively large variability (4), so that small differences may have been missed. However, power calculations suggested that we studied a sufficient number of individuals to account for this. In addition, we selected a robust type of PCOS by including almost only PCOS women who fulfilled all three criteria for the diagnosis. Moreover, the measurements were done at a predetermined point in the PCOS cycle and matched with the hormonal status of controls. In addition, the women were divided in groups according to their PCOS and BMI status. Therefore, outcome measures are comparable without the (confounding) effects of hormones and/or amount of body fat.
In conclusion, we are the first to report that obesity and not PCOS per se is associated with microvascular insulin resistance, as measured with endothelium-dependent vasodilation during insulin infusion. Remarkably, we demonstrated that metabolic insulin resistance was not present in normal-weight women with PCOS, compared with normal-weight controls. In contrast, obese women with PCOS showed a significantly lower metabolic insulin resistance, compared with obese controls. These data may be relevant to understanding and managing metabolic and cardiovascular risk in women with PCOS, suggesting that the focus should be on weight control. In turn, normal-weight women with PCOS may be at relatively normal risk of metabolic and cardiovascular disease. These findings are relevant for a substantial number of women because PCOS is one of the most common endocrinopathies, affecting about 5% of premenopausal women (15, 34).
|
Footnotes |
|---|
Disclosure Statement: The authors have nothing to disclose.
First Published Online July 1, 2008
Abbreviations: ACh, Acetylcholine; AUC, area under the curve; BMI, body mass index; DHEAS, dehydroepiandrosterone sulfate; FFA, free fatty acid; GLM, general linear model; HDL, high-density lipoprotein; M/I, metabolic insulin sensitivity; 17-OHP, 17-hydroxyprogesterone; PCOS, polycystic ovary syndrome; SNP, sodium nitroprusside.
Received March 18, 2008.
Accepted June 25, 2008.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  M. Luque-Ramirez, C. Mendieta-Azcona, J. M del Rey Sanchez, M. Maties, and H. F Escobar-Morreale Effects of an antiandrogenic oral contraceptive pill compared with metformin on blood coagulation tests and endothelial function in women with the polycystic ovary syndrome: influence of obesity and smoking Eur. J. Endocrinol., March 1, 2009; 160(3): 469 - 480. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. M Nelson Obesity, insulin resistance and assisted conception The British Journal of Diabetes & Vascular Disease, January 1, 2009; 9(1): 23 - 26. [Abstract] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
