| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Endocrine Care |
Endocrine Section, First Department of Medicine, Laiko General Hospital, University of Athens, Athens, Greece
Address all correspondence and requests for reprints to: Dr. E. Diamanti-Kandarakis, 1A Zefyrou strasse, Ekali, 145 78, Athens, Greece.
Abstract
Women with polycystic ovary syndrome who present with hyperandrogenemia, hyperinsulinemia, and insulin resistance appear to be at high risk of cardiovascular disease. Elevated levels of endothelin-1, a marker of vasculopathy, have been reported in insulin-resistant subjects with endothelial dysfunction. Male gender also seems to be an aggravating factor for cardiovascular disease.
In this study we investigated endothelin-1 levels in women with polycystic ovary syndrome, and we evaluated the effect of an insulin sensitizer, metformin, on endothelin-1 levels. Plasma endothelin-1 levels were measured in 23 obese (mean age, 24.3 ± 4.6 yr; body mass index, 35 ± 5.6 kg/m2) and 20 nonobese women with polycystic ovary syndrome (24.1 ± 3.6 yr; body mass index, 21.8 ± 2.5 kg/m2) as well as in 7 obese and 10 nonobese healthy, normal cycling, age-matched women. Additionally, endothelin-1 levels were evaluated in a subgroup of women with polycystic ovary syndrome (10 obese and 10 nonobese) 6 months postmetformin administration (1700 mg daily).
Our results showed that obese and nonobese women with polycystic ovary syndrome had higher levels of endothelin-1 compared with the controls [obese, 2.52 ± 1.87 vs. 0.44 ± 0.23 pmol/liter (by analysis of covariance, P < 0.02); nonobese, 1.95 ± 1.6 vs. 0.43 ± 0.65 pmol/liter (P < 0.009)]. All of the participating women with polycystic ovary syndrome (n = 43) when compared with the total group of controls (n = 17) demonstrated hyperinsulinemia (polycystic ovary syndrome, 24.5 ± 19.6; controls, 11.2 ± 3.4 U/liter; P < 0.03), lower glucose utilization (M40) during the hyperinsulinemic euglycemic clamps (3.4 ± 2.4 vs. 5.6 ± 1.75 mg/kg·min; P < 0.045, by one-tailed test), and higher levels of endothelin-1 (polycystic ovary syndrome, 2.52 ± 1.87; controls, 0.44 ± 0.23 pmol/liter; P < 0.02, analysis of covariance covariate for body mass index). A positive correlation of endothelin-1 with free T levels was also shown (r = 0.4, P = 0.002) as well as a negative correlation of endothelin-1 with glucose utilization (r = -0.3; P = 0.033) in the total studied population.
Finally, after metformin therapy, endothelin-1 levels were significantly reduced in obese (endothelin-1 before, 3.25 ± 2.2; endothelin-1 after, 1.1 ± 0.9 pmol/liter; P < 0.003) and nonobese (endothelin-1 before, 2.7 ± 2; endothelin-1 after, 0.7 ± 0.4 pmol/liter; P < 0.01) women with polycystic ovary syndrome, with no change in body mass index. Moreover, after metformin therapy, hyperandrogenemia and hyperinsulinemia were normalized, and glucose utilization improved [obese before: total T, 0.9 ± 0.15 ng/ml; fasting insulin, 22.2 ± 12.1 U/liter; glucose utilization, 2.15 ± 0.5 mg/kg·min; obese after: total T, 0.5 ± 0.2 ng/ml; fasting insulin, 11.6 ± 6 U/liter; glucose utilization, 4.7 ± 1.4 mg/kg·min 9P < 0.003, P < 0.006, and P < 0.002, respectively); nonobese before: total T, 1 ± 0.5 ng/ml; fasting insulin, 15.5 ± 7.6 U/liter; glucose utilization, 3.4 ± 0.7 mg/kg·min; nonobese after: total T, 0.8 ± 0.5 ng/ml; fasting insulin, 9 ± 3.8 U/liter; glucose utilization, 6 ± 1.7 mg/kg·min (P < 0.04, P < 0.02, and P < 0.0008, respectively)].
In conclusion, our data clearly demonstrate that women with polycystic ovary syndrome, obese and nonobese, have elevated endothelin-1 levels compared with the age-matched control group. In addition, 6 months of metformin therapy reduces endothelin-1 levels and improves their hormonal and metabolic profile.
POLYCYSTIC OVARY SYNDROME (PCOS) is the most common endocrinopathy in women of reproductive age; recent studies from United States and Europe have confirmed a comparative prevalence of 4–6% (1, 2). This syndrome in addition to chronic anovulation and hyperandrogenemia (3) is also characterized by several metabolic aberrations, including a high incidence of impaired glucose tolerance (4), hyperinsulinemia, and insulin resistance (5, 6), which, in turn, have shown to increase the risk of cardiovascular disease (7, 8, 9, 10, 11).
Cardiovascular disease is the leading cause of death in women; particularly those with PCOS are at a 7-fold or greater risk for myocardial infarction (11, 12). One of the early signs of cardiovascular lesions is the endothelial injury (13). The mechanism via which the vascular bed is affected under the influence of various metabolic and hormonal abnormalities is not clear. Several hypotheses have been formulated, and a cluster of factors appear to have a synergistic role in this process. The insulin resistance appears to play a key role in the development of the endothelial damage (14). Birdsall et al. (10) demonstrated extensive coronary artery disease in women with polycystic ovary syndrome, and subsequent studies reported that the increased insulin levels in PCOS are associated with decreased cardiac flow (15).
In addition, hyperandrogenemia, a typical finding of the PCOS, may underlie the acceleration of the endothelial injury process (16). Previous studies have reported a higher prevalence of atherosclerosis among hirsute women (17). Interestingly, it has been demonstrated that long-term androgen administration in female to male transsexual adults is associated with impaired vascular reactivity (16). These findings strengthen the possibility that hyperandrogenic women may have early endothelial dysfunction compared with normal young women.
Although study of the pathogenetic mechanisms is a strong research challenge, the assessment of the early signs of endothelial dysfunction has obvious clinical significance. For this purpose different types of approaches (morphological, functional, and biological) have recently been used (18, 19).
Endothelin-1 is one of the several circulating molecules of endothelial injury products. It has endothelial mitogenic effects (20), seems to play a role in the early events of endothelial dysfunction, and has been used as a marker of abnormal vascular reactivity (21). Elevated endothelin-1 (ET-1) levels have been reported in some insulin-resistant states such as obesity (22, 23), diabetes mellitus (DM) (24, 25, 26, 27), and hypertension (22, 28, 29, 30, 31) as well as in normoglycemic subjects with a parental history of diabetes and in individuals with impaired glucose tolerance (32). Studies estimating ET-1 levels in women with PCOS, to our knowledge, have not been reported, notwithstanding the assessment of endothelial function by dynamic vascular reactivity tests has given controversial results. Recent dynamic studies by Mather et al. (33) reported normal endothelium vascular response in PCOS women compared with age-matched controls. On the other hand, preliminary results by Paradisi et al. (34) using the same techniques have shown abnormal endothelium vascular response in obese PCOS (PCOS-OB) women.
The present study was undertaken to investigate whether young obese and nonobese women with PCOS have abnormal plasma ET-1 levels compared with age-matched normal cycling women; the relationship of ET-1 levels with hormonal and metabolic abnormalities was also assessed. Additionally, the metformin effect on ET-1 levels was investigated in a subgroup of the PCOS women studied, because this insulin sensitizer has been shown to have beneficial effects on this dysmetabolic syndrome (35).
Subjects and Methods
Subjects
The study group consisted of 43 women with PCOS and 17 control subjects. The age of the enrolled women in this protocol ranged between 17–37 yr old. Polycystic ovary syndrome was diagnosed in women with oligomenorrhea and hyperandrogenemia; related disorders with similar clinical presentation were excluded (1990 NIH Consensus Conference on PCOS). Hyperandrogenism was defined as free T (FT) levels above the 95th percentile of the levels detected in the group of the control subjects. Oligomenorrhea was defined as less than 6 cycles/yr. The polycystic ovarian morphology detected by ultrasound was not considered an essential criterion for the diagnosis of the syndrome, since it is reported that approximately 20% of normal women could present the above ultrasonographic appearance (36). Additionally, serum PRL (nanograms per ml), TSH (units per ml) and 17-hydroxyprogesterone (nanograms per ml) levels were determined in blood samples of women with oligomenorrhea to exclude other causes of menstrual disorders (37). The control group was selected to have regular menstrual cycles, defined as 26–34 d in length.
All of the women studied were clinically healthy, without suffering from chronic or acute disease and were studied in the early follicular phase (1–8 d) of the menstrual cycle or after 3 months of amenorrhea. Possible disorders were excluded by routine laboratory studies and by thorough physical examination. All of them presented normal blood pressures (systolic, <120 mm Hg; diastolic, <85 mm Hg) after two repeated measurements with a 15-min interval between them. None of the women was currently smoking or receiving oral contraceptive therapy; those who were, had to have been off oral contraceptive therapy at least for 3 months before entering the study. This approach was also applied to other drugs that could interfere with the hormonal and metabolic studies. To gain entry to the study, body weight had to remain stable for at least 2 months before the study, and subjects participating in dietary or exercise program for weight reduction were excluded. Obesity was assessed by estimating body mass index [BMI; weight (kilograms)/height (meters)2; nonobese, <26; obese, >26.1]. Body fat distribution was assessed by measurements of the waist to hip girth ratio (38). The protocol was approved by the institutional review committee of Laiko General Hospital, and written informed consent was obtained from each subject before study.
Materials and Methods
After a 10- to 1-h overnight fast, blood samples were collected from each subject. Plasma endothelin-1, FT, total T (TT), insulin, and glucose were measured. Soon thereafter, the insulin sensitivity of the total PCOS group (n = 43) and that of eight control women was assessed using the hyperinsulinemic euglycemic clamp (De Fronzo, 1979, Am J Physiol 237:E214–E223). An iv catheter was inserted retrogradely into a dorsal vein on the left and kept warm at 65 C via a heated blanket for intermittent sampling of arterialized venous blood. A second catheter was inserted into an antecubital vein in the contralateral arm for administration of glucose and insulin infusions. After collection of three baseline blood samples over 30 min, a primed continuous infusion of crystalline human insulin was started at a rate of 40 mU/m2·min via an infusion pump for 120 min to increase the plasma insulin level to approximately 80 U/ml while maintaining plasma glucose at the basal level. Plasma glucose was maintained between 80–89 mg/dl by sampling every 5 min and was clamped at this level by periodically adjusting a variable infusion of 20% dextrose via an Abbott Lifecare infusion pump (Abbott Laboratories, Inc., Chicago, IL).
It has been previously demonstrated that hepatic glucose production is suppressed by 90% at an insulin concentration of approximately 300 pmol/liter (39). Under these conditions, peripheral glucose utilization is equal to the rate of glucose infusion to maintain euglycemia. The final 30 min of the infusion period was used for determination of peripheral glucose utilization. During that steady state, euglycemia was obtained, and the assumption was made that glucose disposal reflects glucose utilization by the peripheral tissue.
Metformin protocol
Twenty women with PCOS [10 PCOS-OB and 10 nonobese (PCOS-NOB)] of the total 43 women were accepted into the metformin protocol. Metformin was administered at a dose of 1700 mg daily for 6 months, (Lipha Sante, Aron Medicia Division, Lyon, France). All women followed a weight maintenance diet and were checked monthly. No side-effects were reported during the study. After 6 months of treatment, ET-1, TT, FT, SHBG, glucose, and insulin fasting levels were measured. Insulin sensitivity was reassessed by euglycemic clamp, as described above.
Assays
Plasma glucose was determined by the glucose oxidase method (glucose analyzer, Beckman Coulter, Inc., Palo Alto, CA). Blood samples were centrifuged immediately, and serum was stored at -20 C until assayed. Serum insulin levels were measured using the RIA INSULIN-CT kits (CIS-Bio International, Gif-sur-Yvette, France). Duplicate plasma samples were analyzed for FT using the commercially available Coat-A-Count FT kit (Diagnostic Products, Los Angeles, CA). SHBG serum levels were measured by an immunoradiometric assay using the SHBG 125I (Radim S.A., Liege, Belgium). PRL was measured using the PRL immunoradiometric assay kit from Medgenix Diagnostics (Nivelles, Belgium).
17-Hydroxyprogesterone levels were measured using an RIA kit (Diagnostics Systems Laboratories, Inc.). TSH was measured using the human TSH immunoradiometric assay kit (INCSTAR Corp., Stillwater, MN). Androstenedione was measured using the DSL-4200 RIA kit (Diagnostics Systems Laboratories, Inc., Webster, TX). TT was measured using the DSL-4000 RIA kit (Diagnostics Systems Laboratories, Inc.). The intra- and interassay coefficients of variances for FT were 4.3% and 5.5% for low levels and 3.2% and 3.4% for high levels; for SHBG, they were 5.1% and 5.1%, and 5.6% and 4.6%, respectively; for insulin, they were 8.2% and 8.8%, and 5.4% and 6.4%, respectively; for PRL, they were 4.0% and 7.1%, and 6.4% and 6.8%, respectively; for 17-hydroxyprogesterone, they were 9.3% and 9.7%, and 9.5% and 10.8, respectively; for TSH, they were 3.2% and 5.7%, and 3.3% and 4.9%, respectively; for androstenedione, they were 4.3% and 6.3%, and 2.7% and 4.8%, respectively; and for TT, they were 9.6% and 8.6%, and 8.1% and 9.1, respectively.
ET-1 measurement
Blood samples (5 ml) for ET-1 determination were collected into tubes containing EDTA after a 12-h fast and a 30-min resting period in the supine position. They were immediately centrifuged at 4 C for 20 min at 1600 x g, and serum extractions were stored at -70 C. After extraction, ET-1 was measured by ELISA (Biomedica Gesellschft, Vienna, Austria; intraassay coefficient of variation, 3.3% and 4.8% for high and low values, respectively; interassay coefficient of variation, 3.5% and 5.6%, respectively). Normal ET-1 levels ranged between 0.2–0.7 pmol/liter.
Statistical analysis
Statistical analysis, regarding the total groups of women with PCOS and the controls, was performed by analysis of covariance (ANCOVA), using BMI as the covariate. Additionally, BMI and fasting glucose and insulin levels were used as covariates for ET-1. Results are reported as the mean ± SD; the adjusted values are not shown. The subjects were divided into obese and nonobese groups, and ANOVA was performed. Particularly in the comparison of ET-1 levels in the obese subgroups, ANCOVA with covariate BMI was performed because of the disparity of the mean BMI values in the studied obese subgroups. ANOVA (repeated measures) was used for comparing basal and 6 month values in the metformin protocol. For multiple regression analysis, Pearson’s correlation coefficient was used. Analysis was performed using STATISTICA for Windows ’99.
Results
Statistical analysis was performed in the 2 groups studied (43
women with PCOS and 17 control women) and also after dividing them into
obese (BMI, >26 kg/m2) and nonobese subgroups
(23 PCOS-OB and 20 PCOS-NOB women, 7 obese and 10 nonobese controls).
Mean values of age, BMI, and waist to hip
girth ratio did not differ between the 2 groups, (Table 1
) and in the
nonobese subgroups (Table 2). In
the obese subgroups, BMI was higher in the PCOS group
compared with the control group (P = 0.04; Table 2),
and ANCOVA analysis should take into account this disparity.
|
|
Women with PCOS, as expected by the definition of the syndrome
(36), had higher levels of TT (0.88 ± 0.4
vs. 0.48 ± 0.16 ng/ml; P =
0.0005) and FT (2.64 ± 1 vs. 1.12 ±
0.7; P = 0.00001) than controls (Table 1
).
Similar results were observed for the obese and nonobese subgroups
(Table 2).
Metabolic profile
All studied women had normal fasting glucose levels (60–110
mg/dl). Fasting glucose (FG) levels did not differ between PCOS and
control group (FG, 86.9 ± 11 vs. 81.6 ± 7.6
mg/dl; P = NS; Table 1). Fasting insulin
(FI) levels were increased in the PCOS group compared with the controls
(FI, 24.5 ± 19.6 vs. 11.2 ± 3.4 U/liter;
P < 0.03; Table 1). During clamp studies, M40 was
lower in the PCOS group (n = 43) compared with the control (n
= 8; M40, 3.4 ± 2.4 vs. 5.6 ± 1.75 mg/kg·min;
P < 0.045, one-tailed; Table 1).
Additionally, mean FG levels in the obese and nonobese subgroups did
not differ between PCOS women and controls, whereas mean FI levels were
higher in women with PCOS (PCOS-NOB, 17.1 ± 10; nonobese
controls, 10.5 ± 3 U/liter; P < 0.05; PCOS-OB,
31 ± 23.6; obese controls, 12.2 ± 4 U/liter;
P < 0.05; Table 2).
ET-1 levels
ET-1 levels in each studied woman are illustrated in Fig. 1. ET-1 levels were significantly higher
in the studied women with PCOS compared with the controls (ET-1,
2.25 ± 1.75 vs. 0.43 ± 0.51 pmol/liter;
P = 0.0004), using BMI as covariate (Table 1).
Additionally, regarding ET-1, the mean values of the two studied
groups were adjusted with FG and FI. The statistical result using the
additional covariate (FG, FI) remained unchanged (Table 1). Similar
results were observed between nonobese subgroups (PCOS-NOB, 1.95
± 1.6; controls, 0.43 ± 0.65 pmol/ml; P = 0.009)
and obese subgroups (PCOS-OB, 2.52 ± 1.87; controls, 0.44 ±
0.23 pmol/ml; P = 0.008; Fig. 2 and Table 2). Although the statistical
analysis in the obese subgroups for ET-1 levels was also performed with
ANCOVA (covariate BMI), ET-1 levels remained significantly higher in
the PCOS group (P < 0.02; Table 2).
|
|
Multiple regression analysis was performed in the total group of
subjects (PCOS and control women). ET-1 was positively correlated with
FT levels (r = 0.4; P = 0.002).
Additionally, a negative significant
correlation was observed between ET-1 and M40 (r = -0.3;
P = 0.033; Figs. 3 and 4).
|
|
Ten PCOS-OB and 10 PCOS-NOB with PCOS of the group under study
were given metformin for a 6-month period. The results are shown in
Table 3. No significant differences were
observed in the mean BMI of the studied women before and after the
6-month period (PCOS-OB: BMI before, 32.4 ± 4.1; BMI after,
31.3 ± 3.1 kg/m2; P =
NS; PCOS-NOB: BMI before, 22.7 ± 2.7; BMI after,
22.6 ± 3.1 kg/m2; P =
NS; Table 3).
|
). Significant reductions were observed
in FI levels (PCOS-OB: FI before, 22.2 ± 12.1; FI after,
11.6 ± 6 U/liter; P = 0.006; PCOS-NOB: FI before,
15.5 ± 7.6; FI after, 9 ± 3.8 U/liter; P =
0.02) and in FG levels in PCOS-OB (FG before, 88.5 ± 15.4; FG
after, 74.3 ± 8.9 mg/dl; P = 0.007; Table 3). In
accordance with the amelioration of hyperinsulinemia, M40 and SHBG
levels were increased after metformin therapy (PCOS-OB: M40 before,
2.15 ± 0.5; M40 after, 4.7 ± 1.4 mg/kg·min;
P = 0.002; SHBG before, 35 ± 14; SHBG after,
44 ± 18.6 nmol/liter; P = 0.03; PCOS-NOB: M40
before, 3.4 ± 0.7; M40 after, 6 ± 1.7 mg/kg·min;
P = 0.0008; SHBG before, 47 ± 29; SHBG after,
68 ± 32 nmol/liter; P = 0.03; Table 3).
|
). Discussion
There are two main findings of the present study: 1) women with PCOS, obese and nonobese, have increased levels of ET-1 compared with age-matched normal cycling women; and 2) 6 months of metformin administration reduces ET-1 levels and ameliorates insulin resistance and hyperandrogenism, with no changes in body weight. This is the first report, to the best of our knowledge, of elevated ET-1 levels in PCOS.
Increased ET-1 levels are unfavorably linked with endothelial injury and have been broadly described in patients with atherosclerotic lesions and in individuals at high risk for developing cardiovascular disease (40, 41). Previous studies demonstrated increased ET-1 levels in different groups of obese (22, 23), hyperinsulinemic patients with DM (24, 25, 26, 42) and vascular complications (27, 43). Raised plasma ET-1 levels were also considered to be an additional risk factor in nonobese hypertensive subjects with metabolic abnormalities (28). Furthermore, obesity and insulin resistance are associated with endothelial dysfunction independently of other risk factors, although the exact mechanisms have not been determined.
A rather high percentage of women with PCOS (
60% of PCOS-OB and
40% of PCOS-NOB) present with insulin resistance and hyperinsulinemia.
Moreover, they seem to be at high risk for developing impaired glucose
tolerance, diabetes mellitus type 2 (4), and
cardiovascular disease (7, 8).
The total of the studied young women with PCOS had elevated ET-1 levels compared with age-matched normal cycling women. Elevated ET-1 levels were also demonstrated when obese and nonobese subjects were separately compared with the corresponding control groups. Despite the fact that there was a statistical difference in BMI between the obese groups, the statistical significance of the elevated levels of ET-1 between the PCOS-OB and controls remained after correcting for this disparity with ANCOVA (P = 0.0008). This finding suggests that high ET-1, a potent vasoconstrictor peptide, is a distinct abnormality in women with PCOS, independently of the presence of obesity. Interestingly, raised plasma ET-1 levels and endothelial dysfunction were demonstrated not only in patients with DM type 2 (25) and glucose impaired tolerance (32), but also in the first degree relatives of DM type 2 with normal glucose tolerance and no evidence of insulin resistance (32). As women with PCOS are insulin resistant and at risk for developing DM type 2 and cardiovascular disease, the observed elevated ET-1 levels may represent an early sign of abnormal vascular reactivity, which precedes that detectable by dynamic tests of endothelial dysfunction. However, the data about endothelial dysfunction in PCOS are very limited and conflicting (18, 33, 34). A recent study failed to detect any abnormality of endothelial reactivity, assessed by dynamic vascular reactivity tests, in women with PCOS (33). In contrast, preliminary data have shown abnormal vascular reactivity, assessed by similar tests, in PCOS-OB (34). However, it should be mentioned that in none of the above studies were ET-1 levels estimated. Our finding may prove to be of clinical significance, as elevated ET-1 levels may represent a very early sign of abnormal vascular reactivity.
Insulin resistance and endothelial function
The coexistence of insulin resistance and hyperinsulinemia with endothelial dysfunction and elevated ET-1 levels is supported by in vitro and in vivo studies (44, 45, 46, 47). In diabetic patients it has been shown that plasma ET-1 levels were rapidly increased during euglycemic hyperinsulinemic clamps (26). However, the cause-effect relationship between insulin resistance and endothelial dysfunction is under investigation. In the present study women with PCOS had hyperinsulinemia and reduced glucose utilization and exhibited elevated ET-1 levels. This finding seems to be in accordance with the results of other studies of individuals with a variety of insulin-resistant states, such as obesity (48), hypertension (29), and DM with micro/macrovascular complications (32). Furthermore, in the present study a negative relationship was shown between ET-1 levels and glucose utilization. This may be indicative of the involvement of hyperinsulinemia and insulin resistance in the abnormal endothelial status, manifested in this study by the elevated ET-1 levels.
The observed improvement of insulin sensitivity with the significant reduction of ET-1 levels postmetformin administration suggests an additional beneficial effect. This is in accordance with the report of a clinical study that weight reduction in obese individuals ameliorated hyperinsulinemia and reduced ET-1 levels (44). Interestingly, in this study the observed reduced ET-1 level does not seem to be linked with BMI changes, as no significant statistical difference was detected during the treatment.
This is the first report of pharmaceutic intervention by an insulin sensitizer that reduces ET-1 levels. The reduction of ET-1 levels reported here may contribute to the mechanism by which metformin exerts a blood pressure-lowering effect in women with PCOS, as was reported by Velasquez et al. (49).
Finally, the negative correlation of glucose utilization and ET-1 levels indicates the interaction of metabolic aberrations and endothelial status. Thus, therapeutic interventions that increase insulin sensitivity may offer an additional benefit, possibly by protecting and/or restoring the endothelial barrier.
Androgens and endothelial function
Male gender is an independent risk factor for coronary artery disease. Gender difference in vascular dynamics has commonly been attributed to the protective effects of E rather than to any possible adverse effect of androgens (50). The majority of studies that examine the effects of androgens on arterial physiology are performed with men. Studies of experimental animals have shown interesting results; Hutchinson et al. (51) demonstrated androgen-induced endothelial dysfunction in a rabbit model of atherosclerosis. Adams et al. (52) found that T administration increases atheroma lesions in cholesterol-fed female monkeys. In humans, the long-term effect of androgens on vascular physiology have not been explored. Studies investigating the role of androgen excess in vascular function in women are scanty. An interesting work in a relatively uncommon group of subjects has recently been published. In female to male transsexuals (16), endothelium- and smooth muscle-dependent vascular responses were studied after long-term treatment with high doses of androgens. It was shown that androgen administration to genetic females is associated with impaired vascular reactivity, consistent with a deleterious effect of androgen excess on arterial physiology.
Conflicting results have been provided by two recent reports regarding the assessment of endothelial dysfunction by dynamic tests and its relation to hyperandrogenemia in women with PCOS. The preliminary results by Paradisi et al. (34) have shown positive correlation of the abnormal dynamic tests of endothelial function and T levels. On the other hand, the study by Mather et al. (33), using similar dynamic testes, failed to demonstrate any endothelial abnormality in PCOS women compared with normal women and did not show any correlation with androgen levels. Intriguing findings were also reported by Caballero et al. (32) in normoglycemic individuals with a positive family history of diabetes. A gender difference in vascular reactivity was demonstrated; thus, women had a better overall response than men. In the same study it was demonstrated that the endothelial dysfunction was associated with increased levels of markers of endothelial cell damage: adhesion molecules, ET-1, von Willebrand factor, etc. (53).
In the current study it is shown for the first time that a well recognized vasoconstrictive peptide, ET-1, is elevated in women with PCOS and has a positive correlation with T levels. Reduction of androgens postmetformin therapy may add to the beneficial effect of therapy by reducing ET-1 levels.
The mechanism by which hyperandrogenemia might affect vascular reactivity is unknown. It is possible that T has direct effects on the vessel wall, as indicated by the presence of steroid receptor (54), although an indirect effect via alteration of the lipoprotein profile cannot be excluded. More studies are needed to elucidate the sequence of events leading to injured endothelium in this group of women with an array of metabolic and hormonal abnormalities, following them from puberty through reproductive age to menopause.
Study limitations
This is a cross-sectional study of ET-1 levels. The metformin protocol is a nonplacebo and nonrandomized therapeutic trial. Although the studied groups were closely matched for age, smoking habits, blood pressure, and absence of other known vascular risk factors, it is still possible that unmeasured differences were present, e.g. passive smoking, homocysteine levels, or abnormal lipid profile. Additionally, the lack of assessment of endothelial function by dynamics testing and its correlation with ET-1 levels are limitations of this study.
Conclusion
Women with PCOS often present a number of risk factors, all of which potentially contribute to increased atherogenesis and cardiovascular disease. Impairment of vascular reactivity and endothelium abnormalities are important early events in atherogenesis, and early detection of the abnormal signs is very critical. Thereafter, the need to identify an early marker of endothelian dysfunction, such as ET-1, in young women with this syndrome is justified. In obese and nonobese women with this syndrome, ET levels were found to be elevated compared with levels in the control groups. This finding may be considered a prognostic factor for premature atheromatosis.
Metformin administration for 6 months in our patients improved the metabolic and hormonal milieu; at the same time a significant reduction of ET-1 levels was observed. This observation suggests a possible link of insulin resistance and hyperandrogenemia with endothelial dysregulation. Finally, this therapeutic intervention could be justified if additional findings prove the clinical significance of this observation.
Acknowledgments
Footnotes
Abbreviations: ANCOVA, Analysis of covariance; BMI, body mass index; DM, diabetes mellitus; ET-1, endothelin-1; FG, fasting glucose; FI, fasting insulin; FT, free T; M40, glucose utilization; PCOS, polycystic ovary syndrome; PCOS-NOB, nonobese PCOS women; PCOS-OB, obese PCOS women; TT, total T.
Received October 24, 2000.
Accepted June 15, 2001.
References
This article has been cited by other articles:
 |
|
 |
|
  D. Heutling, H. Schulz, I. Nickel, J. Kleinstein, P. Kaltwasser, S. Westphal, F. Mittermayer, M. Wolzt, K. Krzyzanowska, H. Randeva, et al. Asymmetrical Dimethylarginine, Inflammatory and Metabolic Parameters in Women with Polycystic Ovary Syndrome before and after Metformin Treatment J. Clin. Endocrinol. Metab., January 1, 2008; 93(1): 82 - 90. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Schindler Review: The metabolic syndrome as an endocrine disease: is there an effective pharmacotherapeutic strategy optimally targeting the pathogenesis? Therapeutic Advances in Cardiovascular Disease, October 1, 2007; 1(1): 7 - 26. [Abstract] [PDF]
|
|
|
|
 |
|
 J. F. Reckelhoff Polycystic Ovary Syndrome: Androgens and Hypertension Hypertension, June 1, 2007; 49(6): 1220 - 1221. [Full Text] [PDF]
|
|
|
|
 |
|
 A. Lteif, P. Vaishnava, A. D. Baron, and K. J. Mather Endothelin Limits Insulin Action in Obese/Insulin-Resistant Humans Diabetes, March 1, 2007; 56(3): 728 - 734. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. DIAMANTI-KANDARAKIS, T. PATERAKIS, and H. A. KANDARAKIS Indices of Low-Grade Inflammation in Polycystic Ovary Syndrome Ann. N.Y. Acad. Sci., December 1, 2006; 1092(1): 175 - 186. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. DIAMANTI-KANDARAKIS and F. ECONOMOU Stress in Women: Metabolic Syndrome and Polycystic Ovary Syndrome Ann. N.Y. Acad. Sci., November 1, 2006; 1083(1): 54 - 62. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Persico, V. Savasi, E. Ferrazzi, M. Oneta, A.E. Semprini, and G. Simoni Detection of human immunodeficiency virus-1 RNA and DNA by extractive and in situ PCR in unprocessed semen and seminal fractions isolated by semen-washing procedure Hum. Reprod., June 1, 2006; 21(6): 1525 - 1530. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Diamanti-Kandarakis, T. Paterakis, K. Alexandraki, C. Piperi, A. Aessopos, I. Katsikis, N. Katsilambros, G. Kreatsas, and D. Panidis Indices of low-grade chronic inflammation in polycystic ovary syndrome and the beneficial effect of metformin Hum. Reprod., June 1, 2006; 21(6): 1426 - 1431. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Orio Jr., S. Palomba, T. Cascella, B. De Simone, F. Manguso, S. Savastano, T. Russo, A. Tolino, F. Zullo, G. Lombardi, et al. Improvement in Endothelial Structure and Function after Metformin Treatment in Young Normal-Weight Women with Polycystic Ovary Syndrome: Results of a 6-Month Study J. Clin. Endocrinol. Metab., November 1, 2005; 90(11): 6072 - 6076. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. F. Reckelhoff, L. L. Yanes, R. Iliescu, L. A. Fortepiani, and J. P. Granger Testosterone supplementation in aging men and women: possible impact on cardiovascular-renal disease Am J Physiol Renal Physiol, November 1, 2005; 289(5): F941 - F948. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. B. Strawbridge and J. S. Elmendorf Phosphatidylinositol 4,5-Bisphosphate Reverses Endothelin-1-Induced Insulin Resistance via an Actin-Dependent Mechanism Diabetes, June 1, 2005; 54(6): 1698 - 1705. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E Diamanti-Kandarakis, K Alexandraki, A Protogerou, C Piperi, C Papamichael, A Aessopos, J Lekakis, and M Mavrikakis Metformin administration improves endothelial function in women with polycystic ovary syndrome Eur. J. Endocrinol., May 1, 2005; 152(5): 749 - 756. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Orio Jr., S. Palomba, T. Cascella, B. De Simone, S. Di Biase, T. Russo, D. Labella, F. Zullo, G. Lombardi, and A. Colao Early Impairment of Endothelial Structure and Function in Young Normal-Weight Women with Polycystic Ovary Syndrome J. Clin. Endocrinol. Metab., September 1, 2004; 89(9): 4588 - 4593. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. DIAMANTI-KANDARAKIS and H. A. KANDARAKIS Conservative Management of Gynecologic Diseases: Insulin Sensitizing Agents in Polycystic Ovary Syndrome Ann. N.Y. Acad. Sci., November 1, 2003; 997(1): 322 - 329. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Morin-Papunen, K. Rautio, A. Ruokonen, P. Hedberg, M. Puukka, and J. S. Tapanainen Metformin Reduces Serum C-Reactive Protein Levels in Women with Polycystic Ovary Syndrome J. Clin. Endocrinol. Metab., October 1, 2003; 88(10): 4649 - 4654. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Diamanti-Kandarakis, J.-P. Baillargeon, M. J. Iuorno, D. J. Jakubowicz, and J. E. Nestler A Modern Medical Quandary: Polycystic Ovary Syndrome, Insulin Resistance, and Oral Contraceptive Pills J. Clin. Endocrinol. Metab., May 1, 2003; 88(5): 1927 - 1932. [Full Text] [PDF]
|
|
|
|
|
|
M. Schachter, A. Raziel, S. Friedler, D. Strassburger, O. Bern, and R. Ron-El Insulin resistance in patients with polycystic ovary syndrome is associated with elevated plasma homocysteine Hum. Reprod., April 1, 2003; 18(4): 721 - 727. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
) and normal cycling control women (n = 17;
).
