| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Institute of Obstetrics and Gynecology, Catholic University of Sacred Heart (M.G., V.P., M.C., F.M., A.M.F., R.A., A.C., S.M.), Rome; OASI Institute for Research (A.L.), Troina, Italy
Address all correspondence and requests for reprints to: Antonio Lanzone, M.D., Institute of Obstetrics and Gynecology, Catholic University of Sacred Heart, L.go A. Gemelli 8, 00168 Rome, Italy.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Patients were randomly divided into three groups characterized by different treatments: group A (nine patients) was treated with GnRH analog (one ampule every 28 days for 2 months), group B (eight patients) was treated with naltrexone (an oral opioid antagonist, 50 mg/day, orally) for 8 weeks, and group C (nine patients) was treated with GnRH analog plus naltrexone for 2 months. After continuation of treatment, all patients repeated the basal study in a second hospitalization.
Naltrexone treatment significantly reduced the insulin response to OGTT, whereas GnRH analogue administration did not significantly change the insulin secretion after the glucose load. The GnRH analog/naltrexone cotreatment was not able to influence the insulin secretory pattern; in fact, the insulin area under the curve was superimposable before and after therapy. These data could lead to the hypothesis that the opioidergic regulation of insulin secretion requires a normal steroidogenic pattern, thus suggesting that ovarian steroids modulate opioid activity also at peripheric districts.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Many studies have found a relationship between hyperinsulinism and elevated circulating androgen levels in PCOS (2, 3, 4), suggesting that insulin may cooperate with LH in inducing chronic hyperandrogenemia in PCOS patients. Previously, we showed that despite a marked and prolonged decrease in androgen ovarian production obtained by GnRH agonist, insulin secretion in response to an oral glucose tolerance test (OGTT) did not change in PCOS patients, thus indicating that androgens do not cause hyperinsulinism (5).
Our recent data demonstrated that treatment with naltrexone (a long acting opioid antagonist) is able to reduce the LH response to GnRH stimulus in PCOS patients (6).
Moreover, increasing evidence indicates that endogenous opiates are involved not only in the mechanism of gonadotropin release, but also in the regulation of many other compartments such as pancreatic islet function and thus in the pathogenesis of hyperinsulinemia and insulin resistance in subjects suffering from PCOS (7). Both acute (naloxone) and chronic (naltrexone) inhibition of opioidergic system significantly decreased the insulinemic response to an OGTT in a group of hyperinsulinemic PCOS patients without any change of plasma androgen levels (8, 9).
Several in vivo and in vitro studies suggested that the effect of endogenous opioid peptides is dependent on the steroid hormone milieu, as demonstrated by a different response to opioid antagonist treatment during the menstrual cycle (10, 11).
As we have demonstrated a significant involvement of opioid tone on endocrine and metabolic features of PCOS, the aim of the present study was to evaluate the possible impact of ovarian steroids on the opioid-mediated disorders of insulin in patients affected by PCOS.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Polycystic ovary disease was diagnosed by the presence of clinical findings (presence of amenorrhea or oligoamenorrhea with intervals ranging from 2–4 months and hirsutism), plasma androgen levels at the upper limit of the normal range (androstenedione, 0.57–1.6 ng/mL; testosterone, 0.17–0.58 ng/mL), and bilaterally normal or enlarged ovaries with the presence of at least 7–10 microcystis (<5 mm in diameter) at the time of ultrasonography and/or laparoscopy. A normal LH/FSH ratio was not considered an exclusion criterion (7). Obesity was defined as a body mass index greater than 25 kg/m2 (normal range, 19–25); the hirsutism score was calculated according to the criteria of Ferriman and Gallway (12). Informed consent was obtained from each woman. This study was previously approved by our ethical institutional board.
Women were hospitalized the day before the beginning of the basal study; this last one was performed, after a 3-day standard carbohydrate diet (300 g) and fasting overnight, 5–7 days after spontaneous or progestin-induced menses. At 0700 h, blood samples were taken for evaluation of LH, FSH, 17ß-estradiol, PRL, 17-hydroxyprogesterone, sex hormone-binding globulin (SHBG), testosterone, dehydroepiandrosterone sulfate, androstenedione, and cortisol plasma levels, and then each woman underwent an OGTT (75 g). Blood samples were collected every 30 min for 4 h after glucose ingestion and promptly centrifuged and stored at -20 C until assayed. Insulin and glucose were assayed in each sample. All patients left the hospital, and based on their insulin response to OGTT, they were classified as normoinsulinemic (14 patients) and hyperinsulinemic (26 patients) using a cut-off value of 15,000 µIU/mL·240 min for the insulin area under the curve (AUC), as previously established (13). Only hyperinsulinemic subjects continued the protocol study. Randomly, all hyperinsulinemic patients were divided into 3 groups. On day 21 of the cycle, 9 patients were submitted to treatment with GnRH analog (GnRH-A; Enantone depot, Takeda, Italy; 1 ampule im every 28 days) for 2 months (group A). The second group constituted 8 patients who received an oral opioid antagonist (naltrexone, Antaxone, Simes, Vicenza, Italy; 50 mg/day) for 8 weeks (group B); the remaining 9 patients on day 21 of the cycle began treatment with GnRH-A (1 ampule every 28 days for 2 months); after 20 days of treatment, naltrexone (50 mg/day) was administered until the end of the study (group C). After the continuation of treatment, all patients repeated the basal study in a second hospitalization.
The solid phase immunoradiometric assay (coated tube) based on the monoclonal double antibody technique was used for LH, FSH, SHBG, and insulin detection. Steroids were assayed using a direct RIA method in human serum or plasma. Intra- and interassay coefficients of variation were below 7% and 15%, respectively, for all determinations. Glucose levels were determined by the glucose oxidase technique; a normal glycemic response to the OGTT was defined according to the criteria of National Diabetes Data Group (14). Plasma insulin levels were also expressed as the AUC after the glucose ingestion, calculated by the trapezoidal rule, and expressed as microinternational units per mL/240 min. The insulin incremental area was calculated by the difference between AUC and basal AUC (basal AUC = area of the curve due to fasting insulin value x 240 min).
Data are presented as the mean ± SD. Data distribution was tested using the Kolmogorov-Smirnov test. Statistical analysis was performed using the nonparametric Kruskal-Wallis test for multiple comparisons; within-group comparisons were made using the nonparametric Wilcoxon, matched pairs, signed rank test. The significance level was set at P < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
shows
the plasma glucose and hormone fasting levels before and after therapy
in the three groups studied. There was no difference in SHBG, PRL,
basal insulin, and glucose plasma levels in all three groups of
patients before therapy. All studied subjects showed a normal glycemic
response to OGTT, and no difference in the glycemic response to glucose
load was seen after therapy.
|
Figure 1 shows the insulin response to
OGTT as well as the integrated secretory area and the insulin AUC
before and after therapy in all three treated groups of patients.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The mechanism of insulin resistance in this disease is of considerable interest and remains as yet partially unexplored. The hypothesis that endogenous opioid peptides may play an important role in the glyco-metabolic imbalance characteristic of PCOS is suggested by our previous data showing that long term opioid system inhibition obtained with naltrexone treatment was able to reduce the exaggerated insulin response to OGTT in hyperinsulinemic PCOS patients independently from body mass index (9).
Many data indicate that the endocrine milieu, particularly gonadal steroid levels, is involved in opioidergic control of endocrine and metabolic pathways and may contribute to peripheral and central opioid receptor activation (16, 17, 18). Berga and Yen (19) hypothesized that the opioidergic impairment in PCOS women could be due to a functional state secondary to ovarian acyclicity, thus suggesting a possible regulatory role of ovarian steroids in central opioid activity. More recently, Armeanu and colleagues (20), studying patients with hypothalamic amenorrhea and anorexia nervosa, showed that these subjects did not respond to naltrexone treatment, suggesting that the responsiveness to opioid antagonist administration depends on activation of the endogenous opioid system by ovarian steroid production.
Data from the present study clearly indicate that a marked and prolonged decrease in androgens and gonadotropins after long term GnRH-A therapy did not modify the insulinemic response to a glucose stimulus in hyperinsulinemic PCOS patients, confirming a previous report in which it was stated that hyperandrogenism does not produce insulin resistance (5). On the contrary, as previously demonstrated (9), chronic opioid blockade by naltrexone significantly decreased insulin secretion in the same subjects. In our series, opioid blockade reduced insulin levels in PCOS, but serum androgen levels did not change; these data are in keeping with our previous reports about the endocrine effects of naltrexone treatment (7, 9, 13, 21). However, other studies, using the insulin sensitizers metformin and troglitazone (22, 23, 24, 25) showed a decrease in serum androgen in concert with the reduction in serum insulin. A possible explanation for this disagreement could consist of the fact that the action of metformin and troglitazone is mediated by an improvement of insulin sensitivity, whereas pharmacological opioid blockade does not influence the insulin-receptor relationship (13); it could be argued that a reduction in the peripheral insulin resistance is necessary to obtain a decrease in androgen.
Interesting data have been raised from the analysis of group C; in these patients, the GnRH-A/naltrexone cotreatment was not able to influence the insulin secretory patterns. In fact, the insulin AUC was superimposable before and after therapy. These results do not seem to be dependent on the different basal characteristics in the studied groups, as we found a similar degree of obesity as well as similar hormonal and metabolic parameters; in particular, estradiol levels were superimposable in the three groups. Also, the different lengths of treatment do not represent the main explanation for our results; in fact, the present data for group B are superimposable to our previous data obtained with a shorter period of naltrexone therapy (13, 21). The most probable hypothesis to explain our results is that the impaired ovarian steroid production induced by GnRH-A treatment is directly responsible for the lack of opioid antagonist action on the insulin response to OGTT in such hyperinsulinemic patients.
The relationship between gonadal steroids and opioid peptides at the brain level is well established, as shown by data from literature previously discussed. However, the influence of opioids on glycoregulation was mediated by peripheral, rather than central, brain receptors, as highlighted by several findings of opioid receptors in the gastrointestinal tract and endocrine pancreas (26). Therefore, it is possible to hypothesize that ovarian steroids are able to modulate the opioid system not only at the central level, but also at the peripheral one.
Considering that circulating insulin levels depend on insulin secretion by ß-cell, hepatic insulin clearance, and peripheral insulin resistance, it could be argued that one or more of these three compartments represent the site of action of the steroid-opioid system. However, we have recently shown that a long term naltrexone treatment reduced the hyperinsulinemia of PCOS patients without affecting ß-cell secretion or insulin sensitivity, but acted chiefly on hepatic insulin clearance (13, 21). In light of the important hepatic involvement in steroid metabolism, it could be argued that the complex insulin/steroid/opioid interrelationships occur at the hepatic level.
In conclusion, our data could lead to the hypothesis that the opioidergic regulation of insulin secretion requires a normal steroid pattern, thus suggesting that ovarian steroids also modulate opioid activity in peripheric districts.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
activity and serum free testosterone after
reduction of insulin secretion in polycystic ovary syndrome. N
Engl J Med. 335:617–623.This article has been cited by other articles:
 |
|
 |
|
  M.I. Ahmed, A.J. Duleba, O. El Shahat, M.E. Ibrahim, and A. Salem Naltrexone treatment in clomiphene resistant women with polycystic ovary syndrome Hum. Reprod., July 18, 2008; (2008) den273v1. [Abstract] [Full Text] [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 |
) and after
(•) the specific treatment for each group. Right,
Insulin AUC (
) and insulin integrated secretory area (
) in
hyperinsulinemic PCOS patients before and after the specific treatment
for each group. Data are expressed as the mean ± SD.
For more details, see Materials and Methods. *,
P < 0.05; **, P < 0.001
(vs. values after treatment).