help button home button Endocrine Society JCEM
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Gill, S.
Right arrow Articles by Hall, J. E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gill, S.
Right arrow Articles by Hall, J. E.
The Journal of Clinical Endocrinology & Metabolism Vol. 86, No. 6 2428-2436
Copyright © 2001 by The Endocrine Society

Original Articles: Hormones and Reproductive Health

Specific Factors Predict the Response to Pulsatile Gonadotropin-Releasing Hormone Therapy in Polycystic Ovarian Syndrome1

Sabrina Gill2, Ann E. Taylor, Kathryn A. Martin, Corrine K. Welt, Judith M. Adams and Janet E. Hall

Reproductive Endocrine Unit and the National Center for Infertility Research, Department of Medicine, Massachusetts General Hospital, Boston, Massachusetts 02114

Address all correspondence and requests for reprints to: Janet E. Hall, M.D., Reproductive Endocrine Unit, BHX-5, Massachusetts General Hospital, Boston, Massachusetts 02114. E-mail: hall.janet{at}mgh.harvard.edu

Abstract

Ovulation induction is particularly challenging in patients with polycystic ovarian syndrome (PCOS) and may be complicated by multifollicular development. Pulsatile GnRH stimulates monofollicular development in women with anovulatory infertility; however, ovulation rates are considerably lower in the subgroup of patients with PCOS. The aim of this retrospective study was to determine specific hormonal, metabolic, and ovarian morphological characteristics that predict an ovulatory response to pulsatile GnRH therapy in patients with PCOS.

Subjects with PCOS were defined by chronic amenorrhea or oligomenorrhea and clinical and/or biochemical hyperandrogenism in the absence of an adrenal or pituitary disorder. At baseline, gonadotropin dynamics were assessed by 10-min blood sampling, insulin resistance by fasting insulin levels, ovarian morphology by transvaginal ultrasound, and androgen production by total testosterone levels. Intravenous pulsatile GnRH was then administered. During GnRH stimulation, daily blood samples were analyzed for gonadotropins, estradiol (E2), progesterone, inhibin B, and androgen levels, and serial ultrasounds were performed.

Forty-one women with PCOS underwent a total of 144 ovulation induction cycles with pulsatile GnRH. Fifty-six percent of patients ovulated with 40% of ovulatory patients achieving pregnancy. Among the baseline characteristics, ovulatory cycles were associated with lower body mass index (P < 0.05), lower fasting insulin (P = 0.02), lower 17-hydroxyprogesterone and testosterone responses to hCG (P < 0.03) and higher FSH (P < 0.05). In the first week of pulsatile GnRH treatment, E2 and the size of the largest follicle were higher (P < 0.03), whereas androstenedione was lower (P < 0.01) in ovulatory compared with anovulatory patients. Estradiol levels of 230 pg/mL (844 pmol/L) or more and androstenedione levels of 2.5 ng/mL (8.7 nmol/L) or less on day 4 and follicle diameter of 11 mm or more by day 7 of pulsatile GnRH treatment had positive predictive values for ovulation of 86.4%, 88.4%, and 99.6%, respectively. Ovulatory patients who conceived had lower free testosterone levels at baseline (P < 0.04).

In conclusion, pulsatile GnRH is an effective and safe method of ovulation induction in a subset of patients with PCOS. Patient characteristics associated with successful ovulation in response to pulsatile GnRH include lower body mass index and fasting insulin levels, lower androgen response to hCG, and higher baseline FSH. In ovulatory patients, high free testosterone is negatively associated with pregnancy. A trial of pulsatile GnRH therapy may be useful in all PCOS patients, as E2 and androstenedione levels on day 4 or follicle diameter on day 7 of therapy are highly predictive of the ovulatory response in this group of patients.

POLYCYSTIC OVARIAN syndrome (PCOS) is a complex disorder characterized by menstrual irregularity and hyperandrogenism, affecting 4–6% of reproductive-aged women (1, 2). Although the pathogenesis is unclear, PCOS has been associated with insulin resistance, androgen excess, neuroendocrine abnormalities, and polycystic ovarian morphology. Reproductive abnormalities have been attributed to hyperinsulinemia and hyperandrogenism (3, 4, 5). It has been estimated that infertility due to PCOS represents 17% of the annual incidence of infertility in the general population (6).

Treatment of infertility with ovulation induction in PCOS is challenging. Many patients ovulate and conceive in response to clomiphene citrate (7, 8). However, 25–40% of patients fail to conceive on clomiphene and proceed to other forms of therapy (9, 10, 11, 12, 13, 14, 15). In these clomiphene-resistant PCOS patients, gonadotropin therapy [human menopausal gonadotropin (hMG) or FSH] results in ovulation rates of 89%/cycle and conception rates of 12.5%/cycle (16), lower than those in women with hypothalamic amenorrhea (97%/cycle and 25%/cycle, respectively) (17). In addition, gonadotropin therapy in women with PCOS is frequently complicated by ovarian hyperstimulation (10.9%), multiple gestation (33.3%/pregnancy), and cycle cancellation (4.2%) (16). Despite improvements in complication rates using lower dose gonadotropin regimens, up to 8% ovarian hyperstimulation, 5–18% cycle cancellation, and 6% multiple gestation have been reported (13, 15).

Pulsatile GnRH is a physiological alternative to gonadotropins. It has been successfully used in patients with hypothalamic amenorrhea with similar rates of ovulation and conception, but lower rates of multiple folliculogenesis (18.9% vs. 47.6% with hMG) and multiple gestation (8.3% vs. 14.8% with hMG) (17). In clomiphene-resistant PCOS patients, ovulatory rates of 40–80%/cycle and conception rates of 10–40%/ovulatory cycle have been achieved using pulsatile GnRH with minimal rates of multiple gestation, ovarian hyperstimulation, and cycle cancellation (0–5%) (9, 10, 11, 12). Addition of a GnRH agonist before pulsatile GnRH has significantly improved ovulation and conception rates in patients with PCOS, but this regimen requires 4–6 weeks of agonist administration before each cycle of treatment (18).

Although the ovulatory response rate is imperfect in this patient population, pulsatile GnRH may be a worthwhile selection for ovulation induction in individual patients given its low complication rate. Our experience indicates that those who ovulate in response to pulsatile GnRH do so consistently, suggesting that specific patient characteristics may predispose to a favorable response. Thus, this study was designed to retrospectively review our clinical experience and define the specific hormonal, metabolic, and morphological characteristics that predict ovulation in women with PCOS before therapy begins and/or within the first week of ovulation induction with pulsatile GnRH.

Subjects and Methods

Subjects

PCOS was defined by chronic amenorrhea or oligomenorrhea (fewer than nine menses per yr) and clinical (hirsutism with Ferrimen-Gallway score >10, acne, and/or male pattern alopecia) and/or biochemical hyperandrogenism in the absence of any adrenal or pituitary disorder, as previously described (1, 19, 20). Biochemical hyperandrogenism was defined in relation to the upper 95% confidence limits for total testosterone (>0.85 ng/mL; 2.95 nmol/L) and free testosterone (>0.041 ng/mL; 0.14 nmol/L) from 30 carefully characterized normal women studied in our laboratory. All infertile PCOS subjects who presented to the Reproductive Endocrine Unit at Massachusetts General Hospital and underwent pulsatile GnRH therapy from 1983 to 1999 were included in the study (n = 41). Patients were between 23 and 42 yr of age and in good general health. Two women had well controlled hypothyroidism, two women had well controlled hyperprolactinemia, and one woman had stable type 1 diabetes mellitus. Subjects stopped all other medications for at least 3 months before the study except for thyroid hormone, bromocriptine, and insulin. All women had documented normal TSH and PRL levels. Late-onset congenital adrenal hyperplasia was excluded with normal 17-hydroxyprogesterone levels in the morning or after ACTH stimulation. Thirty-seven patients had previously failed to conceive with clomiphene (n = 36) and/or hMG (n = 12). Of those using gonadotropins, seven previously experienced hyperstimulation with therapy. Four women who participated in the study were not interested in pregnancy.

This study was approved by the institutional review board of the Massachusetts General Hospital. All women gave written informed consent before undergoing the study protocols.

Protocol

Baseline evaluation. All subjects underwent a medical history and physical examination, including determination of body mass index (BMI) and clinical assessment of hirsutism using the Ferriman-Gallwey scoring system (20, 21). Subjects had a transvaginal pelvic ultrasound examination to evaluate ovarian morphology (22) and volume, as well as to identify any dominant follicles or corpora lutea. From 1985 to 1999 (n = 35), the ultrasounds were performed by a single ultrasonographer (J.M.A.) using 1 machine [Sonolayer L, SAL-778 (Toshiba, Tokyo, Japan), with a transvaginal 3- to 5-MHz probe]. Before 1985 (n = 6), the ultrasounds were performed by a qualified ultrasonographer in our radiology department. Subjects were admitted to the General Clinical Research Center and underwent baseline frequent sampling studies for pulsatile LH. Blood was sampled through an indwelling intravenous catheter every 10 min for 12–24 h, with the exception of 1 patient who underwent only 4 h of sampling. Fasting insulin levels were measured in all but 2 subjects. Fifteen subjects also underwent the tolbutamide-modified, frequently sampled, iv glucose tolerance test (FSIVGTT) to evaluate insulin sensitivity (sI) and the acute first phase response of insulin to a glucose load (AIRg), as previously described (23). All subjects had a baseline total testosterone measurement. In addition, ovarian androgen responsiveness was assessed by measurement of 17-hydroxyprogesterone, androstenedione, and total testosterone before and 24 h after hCG (5000 U, im) administration in 16 subjects.3

In those desiring pregnancy, the semen analysis of the male partner was assessed for count, volume, and motility using WHO reference values. Two patients, whose partners had a low sperm count or abnormal morphology, underwent intrauterine insemination at ovulation.

Pulsatile GnRH regimen. Pulsatile GnRH was administered iv for all treatment cycles using the Zyklomat (Ferring Pharmaceuticals Ltd., Malmo, Sweden) pump, at pulse frequencies varying from every 60–240 min across the cycle, as previously described, to mimic the physiological frequency in normal women (24, 25). Therapy was initiated at a dose of 75 ng/kg·bolus. Serial ultrasound measurements of follicular diameter were used to document follicle growth and development of a preovulatory follicle (>=16 mm). If dominant follicle development did not occur within 14 days, the dose was increased up to 150 ng/kg (n = 17), and if follicle development was too exuberant [>15 follicles and/or estradiol (E2) >2000 pg/mL], the dose was decreased (n = 2) or treatment was discontinued (n = 1). Blood samples were collected daily, 45 min after a dose of GnRH at the same time of day, for subsequent determination of E2, LH, FSH, and progesterone (P) levels. Inhibin B, 17-hydroxyprogesterone, androstenedione, and total testosterone were also assayed for the first 5 days of the first treatment cycle (n = 28). E2 and inhibin B were used as indicators of folliculogenesis (26), and androgens were measured due to the potential adverse effect of hyperandrogenemia on follicle development (27). During the cycle, ovulation was determined by midcycle urinary LH monitoring, rise in basal body temperature, and/or ultrasound findings suggestive of a corpus luteum and was retrospectively confirmed by progesterone measurement in the daily blood samples.

Assays

E2 and progesterone were assayed by RIA as previously described (28, 29). LH and FSH were analyzed by RIA (29) or by a two-site monoclonal nonisotopic system calibrated with the same reference preparation used in the RIA (30). LH and FSH were extensively cross- referenced in the two assay systems and were not significantly different across a broad range of values and across the menstrual cycle within individual subjects (26). Gonadotropin levels are expressed in international units per L, as equivalents of the Second International Reference Preparation of hMG. Total testosterone, 17-hydroxyprogesterone, and androstenedione were measured by RIA as previously described (20, 31). All samples from an individual cycle were measured in the same assay. The sensitivity of the assay and the intra- and interassay coefficients of variation were similar to those described previously. Inhibin B was measured by enzyme-linked immunosorbent assay as previously described (32). The intra- and interassay coefficients of variation were less than 7% and less than 10%, respectively. Free testosterone was measured by RIA (Coat-A-Count Free Testosterone kit, Diagnostic Products, Los Angeles, CA). The sensitivity of the assay is 0.15 pg/mL, and the intra- and interassay coefficients of variation are less than 6% and less than 18%, respectively.

Glucose was measured by the glucose hexokinase method, and insulin by an automated immunometric assay with sensitivity of 1.0 µIU/L, as previously described (33), with both intra- and interassay coefficients of variation less than 5%.

Data analysis

Treatment cycles were considered ovulatory if luteal phase serum progesterone levels were greater than 6 ng/mL (19 nmol/L) (24). Pregnancies were defined by a positive serum hCG and were further documented by the presence of a fetal heartbeat on ultrasound approximately 4.5 weeks from ovulation. Subjects were classified as ovulatory or anovulatory based on their response to pulsatile GnRH therapy in the majority of cycles initiated. In the ovulatory group, the subjects who desired pregnancy were further classified into the pregnant or nonpregnant group. In subjects who conceived more than once, only data from their first treatment cycle was considered.

Baseline FSH and LH levels were determined by the mean value of the baseline frequent sampling studies for each subject. Analysis of pulsatile LH secretion was performed using a previously validated modification of the Santen and Bardin method (34, 35). Based on the results of the tolbutamide-modified FSIVGTT, insulin sensitivity was calculated using the minimal model analysis program (MINMOD) as previously described (36); the AIRg was calculated as the mean increment above basal insulin values measured between 0–10 min. The androgen response to hCG was determined as the difference between the androgen levels at baseline and 24 h after the hCG injection. The baseline characteristics of the two groups were compared using Student’s t tests or Wilcoxon rank-sign tests, as appropriate.

For analysis of the daily hormonal response to pulsatile GnRH, the first cycle after the subject’s baseline study was selected for each subject. Hormone levels on the first 5 days of the cycle were compared between ovulatory and anovulatory groups using ANOVA for repeated measures followed by post-hoc Newman-Keuls testing for individual differences. Logistic regression analysis of the daily hormone levels was performed to develop a multivariate model of ovulation predictors and to identify the optimal day for assessment of these predictors. The model was adjusted for BMI to account for the influence of body fat on hormonal abnormalities in PCOS (20). The area under the receiver operator curve (ROC) was used to define the predictive ability of the model. As follicle diameter after 1 week of therapy correlated with E2 levels, it was considered independently as a predictor of ovulation.

An exploratory review of the data was conducted to establish specific parameters for the predictive variables on the optimal day. Cut-off levels for the significant hormones (androstenedione and E2) with the optimal sensitivity, specificity, and positive predictive values for ovulation were selected. The cut-off value for follicle size was based on the minimum follicle diameter achieved by day 7 in the ovulatory group and was consistent with the value used in our clinical practice as indicative of a dominant follicle. The Statistical Analysis System Program (SAS Institute, Inc., Cary, NC) was employed for data analysis. Results are expressed as the mean ± SEM unless otherwise indicated. P < 0.05 was considered statistically significant.

Results

Clinical outcomes

Forty-one women with polycystic ovary syndrome underwent a total of 144 ovulation induction cycles with pulsatile GnRH. Thirty-seven patients had previously failed to conceive with clomiphene and/or gonadotropins. Ovulation occurred in 77 cycles in 23 patients. Fifty-six percent of women ovulated, yielding an ovulation rate of 53%/cycle (Fig. 1Go). All but 2 ovulated at doses of 100 ng/kg pulsatile GnRH or less. One cycle was discontinued due to development of 17 preovulatory follicles (11–26 mm in diameter) and a high E2 level (4,520 pg/mL; 16,593 pmol/L). Eight women successfully conceived, and 1 woman had 2 pregnancies with pulsatile GnRH treatment. Thus, of the 20 women who were ovulatory and also desired pregnancy, 40% conceived, with a conception rate of 12%/cycle. Two women had spontaneous miscarriages early in their pregnancies. All pregnancies were singletons.



View larger version (24K):
[in this window]
[in a new window]
  		Figure 1. Clinical outcomes for ovulation induction using pulsatile GnRH in women with clomiphene resistance and PCOS.

 
Baseline characteristics

Several distinct characteristics were noted to differentiate the ovulatory (n = 23) from the anovulatory (n = 18) patients before starting therapy. Ovulatory women had lower BMI and higher baseline mean FSH (Table 1Go). However, there was no difference in LH, the LH/FSH ratio, LH pulse frequency, LH pulse amplitude, E2, inhibin B, or ovarian volume between the ovulatory and anovulatory groups.


View this table:
[in this window]
[in a new window]
  		Table 1. Characteristics of ovulatory and anovulatory subjects

 
Women with lower fasting insulin were more likely to ovulate (Table 1Go). In the subset of subjects (n = 11 ovulatory; n = 4 anovulatory) who underwent FSIVGTT, lower AIRg and higher insulin sensitivity were seen in ovulatory women compared with anovulatory patients, consistent with the lower mean fasting insulin levels in the ovulatory group. Total testosterone levels were not different between ovulatory and anovulatory groups. Interestingly, in the 15 subjects who underwent hCG stimulation tests (n = 10 ovulatory; n = 5 anovulatory), ovarian 17-hydroxyprogesterone and total testosterone responses to hCG stimulation was significantly greater in the anovulatory group (Table 1Go).

Baseline characteristics predicting pregnancy

Of the 20 women who ovulated in response to pulsatile GnRH therapy and desired pregnancy, the 8 women who became pregnant were not different from the nonpregnant women with respect to age, BMI, or mean baseline FSH levels (Table 2Go). Lower mean free testosterone (P = 0.03) levels were significantly associated with successful conception in the pregnant compared with the nonpregnant ovulatory women.


View this table:
[in this window]
[in a new window]
  		Table 2. Characteristics of pregnant and nonpregnant subjects

 
Hormonal response to pulsatile GnRH

An abnormal increase in LH and FSH was seen with initiation of pulsatile GnRH. E2 levels in the follicular and luteal phases were higher than normal despite the development of a single dominant follicle in the vast majority. Luteal phase progesterone was indistinguishable from normal (Fig. 2AGo). The anovulatory subjects failed to demonstrate normal gonadotropin, E2, or progesterone dynamics (Fig. 2BGo).



View larger version (33K):
[in this window]
[in a new window]
  		Figure 2. Daily mean hormone (±SEM) levels in ovulatory (A) and anovulatory (B) patients in response to pulsatile GnRH therapy. Normal ranges (±1 SD; n = 109) are indicated by shading. To convert E2 levels to picomoles per L, multiply by 3.671. To convert progesterone levels to nanomoles per L, multiply by 3.180.

 
In the first 5 days of pulsatile GnRH therapy (15 ovulatory and 15 anovulatory women), LH and FSH levels increased in both groups within 24 h, with a parallel decrease in the 2 hormones after day 2 (main effect, P > 0.10; Fig. 3Go). There was an initial increase in E2 levels in both groups; however, by day 3, they diverged, with higher E2 levels in the ovulatory group (main effect, P < 0.02). Inhibin B exhibited a similar pattern as E2, with a decline in the anovulatory group beginning on day 4, but the change was not statistically significant overall (main effect, P < 0.20; Fig. 3Go).



View larger version (19K):
[in this window]
[in a new window]
  		Figure 3. Mean daily hormone results (±SEM) for gonadotropins, E2, and inhibin B in patients during the first 5 days of pulsatile GnRH therapy in ovulatory (n = 15 for LH, FSH, and E2; n = 13 for inhibin; •) and anovulatory (n = 15; {circ}) patients. Overall effect for E2, P < 0.02. **, P < 0.01. To convert E2 levels to picomoles per L, multiply by 3.671.

 
The androgen response to pulsatile GnRH differed in the ovulatory (n = 13) and anovulatory groups (n = 15; Fig. 4Go). Specifically, the ovulatory group had significantly lower androstenedione levels (main effect, P < 0.05) and a trend toward lower total testosterone levels (main effect, P < 0.15) by day 2 compared to the anovulatory group. The 17- hydroxyprogesterone levels were not different between the groups.



View larger version (22K):
[in this window]
[in a new window]
  		Figure 4. Mean daily hormone results (±SEM) for androgens during the first 5 days of pulsatile GnRH therapy in ovulatory (n = 13; •) and anovulatory (n = 15; {circ}) patients. Overall effect for androstenedione, P < 0.05. **, P < 0.01. To convert to Systeme International units, multiply androstenedione levels by 3.492, testosterone levels by 0.035, and 17-hydroxyprogesterone levels by 3.026.

 
By day 7, the patients destined to ovulate all had a dominant follicle on ultrasound, with a mean follicle size of 14.1 ± 1.0 mm (range, 11–20 mm) in the ovulatory group compared with 9.4 ± 0.6 mm (6–14 mm) in the anovulatory group (P < 0.001).

Logistic regression analysis revealed day 4 to be the optimal day for assessment of hormonal measures. E2 and androstenedione were independent predictors of ovulation. Use of the combination of these two variables resulted in a high sensitivity and specificity for prediction of ovulation, as indicated by an area under the ROC curve of 0.92 (Table 3Go). Adjustment for BMI had no effect on the predictive value of these factors. Due to the correlation between follicle size and E2 (r2 = 0.45; P < 0.02), follicle size was evaluated in an independent model. Results indicated that follicle diameter was a significant predictor of ovulation by day 7 of pulsatile GnRH therapy, with a high sensitivity and specificity and an area under the ROC curve of 0.94 (Table 3Go). Total testosterone, inhibin B, LH, and FSH were not independent predictors of ovulation in this population.


View this table:
[in this window]
[in a new window]
  		Table 3. Factors predicting ovulation with pulsatile GnRH in clomiphene-resistant PCOS patients based on logistic regression analysis

 
On day 4, E2 levels of 230 pg/mL (844 pmol/L) or more had a positive predictive value for ovulation of 86.4%, whereas androstenedione levels of 2.5 ng/mL (8.7 nmol/L) or less had a positive predictive value of 88.4% (Table 4Go). By day 7, a follicle size of 11 mm or greater was 99.7% predictive of ovulation (Table 4Go).


View this table:
[in this window]
[in a new window]
  		Table 4. Positive predictive value (PPV) for ovulation of estradiol and androstenedione on day 4 and follicle diameter on day 7 of pulsatile GnRH

 
Discussion

In this study we have systematically investigated the baseline subject characteristics and early hormonal and ultrasound responses to administration of pulsatile GnRH in women with PCOS to determine characteristics that predict successful ovulation induction with pulsatile GnRH. Factors associated with ovulation were lower baseline BMI, lower fasting insulin, and higher mean FSH levels. In our PCOS patient population, cycles with lower androstenedione (<2.5 ng/mL; 8.7 nmol/L) and higher E2 levels (>230 pg/mL; 844 pmol/L) by day 4 of pulsatile GnRH therapy were more likely to progress to ovulation. The presence of a follicle of 11 mm or greater by day 7 of therapy was virtually 100% predictive of ovulation. In addition, in ovulatory patients, lower baseline free testosterone was predictive of pregnancy.

In our series, the ovulation and conception rates of 53% and 12%/cycle, respectively, were compatible with those reported in clomiphene-resistant patients treated with pulsatile GnRH. Women with PCOS who failed ovulation or conception with clomiphene or gonadotropin treatment had ovulatory rates of 40–80% and pregnancy rates between 10–40%/cycle when treated with pulsatile GnRH (9, 10, 11, 12). Ovulation induction with exogenous gonadotropins in PCOS has been associated with a high risk of multiple gestation and ovarian hyperstimulation syndrome (13, 14, 15). In our subjects, none had multiple gestation, and one woman had mild ovarian hyperstimulation that was treated with discontinuation of the pump without further complications or hospitalization. Thus, pulsatile GnRH is a viable and safe method of ovulation induction in PCOS.

Previous studies that have attempted to identify patient characteristics that predict the ovulatory response to different ovulation induction therapies, including pulsatile GnRH, have suggested that higher BMI, testosterone, and insulin are markers of a poor prognosis (8, 10, 13, 15, 37, 38, 39, 40, 41, 42, 43, 44, 45). However, the majority of these studies have not focused specifically on PCOS, but have included a variety of anovulatory or oligoamenorrheic patients (37, 38, 40, 41, 42, 43). As higher BMI, testosterone, and insulin are features of PCOS, these studies have only confirmed that PCOS patients respond poorly to ovulation induction. In PCOS patients treated with clomiphene, high BMI, insulin, and testosterone were associated with lower ovulatory rates (8, 44, 45). Only an elevated BMI has been identified to be a negative prognostic factor of ovulation with exogenous gonadotropin therapy (13, 15, 39). In PCOS patients treated with pulsatile GnRH, anovulatory patients had higher BMI and testosterone levels, but insulin was not assessed (10). In the current study we have shown that in PCOS patients treated with pulsatile GnRH, patients with higher BMI and fasting insulin levels and lower baseline FSH levels had a poor ovulatory response. In the subgroup of patients who also had an IVGTT, lower insulin sensitivity was observed in the anovulatory compared with the ovulatory group, consistent with elevated fasting insulin levels as a marker of insulin resistance (46). A decrease in insulin sensitivity was associated with increased ovarian androgen secretion in response to hCG, as previously described (47).

E2 increases with follicular growth, as was demonstrated in our study with E2 levels diverging after 2 days of pulsatile GnRH therapy, separating the ovulatory and anovulatory groups. E2 levels after 4 days of therapy correlated with follicular diameter on day 7. Inhibin B is also secreted by the granulosa cells and marks early follicular growth in response to endogenous or exogenous FSH stimulation (26, 48, 49). Anderson et al. demonstrated a 2-fold increase in inhibin B levels within 2 days of administration of exogenous FSH in women with PCOS (50). In the current study inhibin B levels also increased with FSH stimulation, remaining elevated in the ovulatory group and declining after 2 days in the anovulatory group; however, variability was greater than for E2 and androstenedione, and overall results were not significant. The early abnormal increase in LH and FSH in response to administration of pulsatile GnRH in our study has been previously documented in women with PCOS (11, 18) and may be attributed to increased pituitary sensitivity to GnRH in PCOS (18, 51).

We found that higher levels of androstenedione early in the follicular phase predict anovulation in response to pulsatile GnRH treatment. Androstenedione production is higher in PCOS, and hyperandrogenemia has been previously associated with a poor follicular response to pulsatile GnRH in PCOS (37, 52, 53). Increased stroma in the PCOS ovary (54) may contribute to excess androgen secretion, resulting in an acceleration of the basal growth rate of early, gonadotropin-independent stages of follicle development (55); however, development of a dominant follicle and ovulation may be inhibited by hyperandrogenemia. Reduction of hyperandrogenemia using an androgen receptor blockade (flutamide) has been associated with improved folliculogenesis (56), supporting the role of androgens in abnormal folliculogenesis in PCOS. Pulsatile LH secretion has also been shown to stimulate ovarian androgen production (18, 31, 51). Pretreatment with a GnRH agonist to suppress gonadotropin secretion also lowers ovarian androgen production and improves the response to ovulation induction with pulsatile GnRH (18, 37).

In the current study lower free testosterone levels were associated with successful conception. A higher total testosterone level has been previously associated with significantly higher abortion rates in a series of anovulatory infertile women treated with pulsatile GnRH (37). In our study there was a trend toward higher BMI predicting lower pregnancy rates, but results did not reach significance. BMI has been shown previously to be a predictor of conception in PCOS using exogenous gonadotropins, with higher pregnancy rates observed in younger and leaner PCOS women (13, 15, 39).

This study confirms that pulsatile GnRH is a viable option for ovulation induction in patients with PCOS who are unresponsive to clomiphene, particularly in light of the complications associated with exogenous gonadotropin therapy. Although the ovulatory rate is 53%/cycle with patients resistant to other therapies, pulsatile GnRH may be most useful in a specific subset of patients with PCOS. We have now shown that patients with lower BMI and fasting insulin levels, lower responsiveness of 17-hydroxyprogesterone and testosterone to hCG, and higher baseline FSH are more likely to ovulate. In addition, a high likelihood of ovulation in response to pulsatile GnRH therapy can be identified within the first week of treatment in a given patient, using either hormonal criteria after day 4 or follicle size by day 7 of therapy. Although our prediction model for ovulation is developed in a relatively small number of subjects, it provides specific guidelines for monitoring patients on pulsatile GnRH.

Acknowledgments

We acknowledge the support of William F. Crowley, Jr., M.D., and Helene Lavoie, M.D., during these studies. We gratefully acknowledge the technicians of the RIA Core Laboratory of the Reproductive Endocrine Unit, under the direction of Patrick Sluss, Ph.D., for their superb technical contributions to this study, and the nurses from the General Clinical Research Unit for their excellent work. We also thank the women who participated in this study for their cooperation and commitment to this research.

Footnotes

1 This work was supported by NIH Grants U54-HD-29164, M01-RR-1066, and P30-HD-28138. Back

2 Recipient of fellowship support from the British Columbia Endocrine Research Foundation and Parke-Davis Canada. Back

3 Baseline gonadotropin and testosterone levels have previously been published [Refs. 19 (n = 6) and 20 (n = 17)]. Back

Received October 2, 2000.

Revised January 18, 2001.

Revised February 12, 2001.

Accepted February 12, 2001.

References

  1. Zawadzki JK, Dunaif A. 1992 Diagnostic criteria for polycystic ovary syndrome: towards a rational approach. In: Dunaif A, Givens JR, Haseltine FP, Merriam GR, eds. Polycystic ovary syndrome. Boston: Blackwell; 377.
  2. Knochenhauer ES, Key TH, Kahsar-Miller M, Wagoner W, Boots LH, Azziz R. 1998 Prevalence of the polycystic ovary syndrome in unselected black and white women of the southeastern US: a prospective study. J Clin Endocrinol Metab. 83:3078–3082.[Abstract/Free Full Text]
  3. Taylor AE. 1998 Polycystic ovary syndrome. Endocrinol Metab Clin North Am. 27:877–902.[CrossRef][Medline]
  4. Anttila L, Ding YQ, Ruutiainen K, Erkkola R, Irjala K, Huhtaniemi I. 1991 Clinical features and circulating gonadotropin, insulin, and androgen interactions in women with polycystic ovarian disease. Fertil Steril. 55:1057–61.[Medline]
  5. Acien P, Quereda F, Matallin P, et al. 1999 Insulin, androgens and obesity in women with and without polycystic ovary syndrome: a heterogenous group of disorders. Fertil Steril. 72:32–40.[CrossRef][Medline]
  6. Hull MG. 1987 Epidemiology of infertility and polycystic ovarian disease: endocrinological and demographic studies. Gynecol Endocrinol. 1:235–245.[Medline]
  7. Lobo RA, Gysler M, March CM, Goebelsmann U, Mishell DR. 1982 Clinical and laboratory predictors of clomiphene response. Fertil Steril. 37:168–174.[Medline]
  8. Polson DW, Kiddy DS, Mason HD, Franks S. 1989 Induction of ovulation with clomiphene citrate in women with polycystic ovary syndrome: the difference between responders and nonresponders. Fertil Steril. 51:30–34.[Medline]
  9. Burger CW, Korsen TJM, Hompes PGA, van Kessel H, Schoemaker J. 1986 Ovulation induction with pulsatile luteinizing releasing hormone in women with clomiphene citrate-resistant polycystic ovary-like disease: clinical results. Fertil Steril. 46:1045–1054.[Medline]
  10. Tan SL, Jarhi J, Homburg R, Jacobs HS. 1996 Induction of ovulation in clomiphene-resistant polycystic ovary syndrome with pulsatile GnRH. Obstet Gynecol. 88:221–226.[CrossRef][Medline]
  11. Corenthal L, Von Hagen S, Larkins D, Ibrahim J, Santoro N. 1994 Benefits of continuous physiological pulsatile gonadotropin-releasing hormone therapy in women with polycystic ovarian syndrome. Fertil Steril. 61:1027–1033.[Medline]
  12. Eshel A, Abdulwahid NA, Armar NA, Adams JM, Jacobs HS. 1988 Pulsatile luteinizing horone-releasing hormone therapy in women with polycystic ovary syndrome. Fertil Steril. 49:956–960.[Medline]
  13. Hamilton-Fairley D, Kiddy D, Watson H, Paterson C, Franks S. 1992 Association of moderate obesity with a poor pregnancy outcome in women with polycystic ovary syndrome treated with low dose gonadotropin. Br J Obstet Gynecol. 99:128–131.[Medline]
  14. Fluker MR, Urman B, MacKinnon M, Barrow SR, Pride SM, Ho Yuen B. 1994 Exogenous gonadotropin therapy in World Health Organization groups I and II ovulatory disorders. Obstet Gynecol;83 :189–196.
  15. White DM, Polson DW, Kiddy D, et al. 1996 Induction of ovulation with low-dose gonadotropins in polycystic ovary syndrome: an analysis of 109 pregnancies in 225 women. J Clin Endocrinol Metab. 81:3821–3824.[Free Full Text]
  16. Homburg R, Levy T, Ben-Rafael Z. 1995 A comparative prospective study of conventional regimen with chronic low-dose administration of follicle-stimulating hormone for anovulation associated with polycystic ovary syndrome. Fertil Steril. 63:729–733.[Medline]
  17. Martin KA, Hall JE, Adams J, Crowley Jr WF. 1993 Comparison of exogenous gonadotropins and pulsatile gonadotropin-releasing hormone for induction of ovulation in hypogonadotropic amenorrhea. J Clin Endocrinol Metab. 77:125–129.[Abstract]
  18. Filicori M, Campaniello E, Michelacci L, et al. 1988 Gonadotropin-releasing hormone analog suppression renders polycystic ovarian disease patients more susceptible to ovulation induction with pulsatile GnRH. J Clin Endocrinol Metab. 66:327–333.[Abstract/Free Full Text]
  19. Waldstreicher J, Santoro NF, Hall JE, Filicori M, Crowley WF Jr. 1988 Hyperfunction of the hypothalamic-pituitary axis in women with polycystic ovarian disease: indirect evidence for partial gonadotroph desensitization. J Clin Endocrinol Metab. 66:165–172.[Abstract/Free Full Text]
  20. Taylor AE, McCourt B, Martin KA, et al. 1997 Determinants of abnormal gonadotropin secretion in clinically defined women with polycystic ovary syndrome. J Clin Endocrinol Metab. 82:2248–2256.[Abstract/Free Full Text]
  21. Ferriman D, Gallwey JD. 1961 Clinical assessment of body hair growth in women. J Clin Endocrinol Metab. 21:1440–1447.
  22. Adams J, Polson DW, Franks S. 1986 Prevalence of polycystic ovaries in women with anovulation and idiopathic hirsutism. Br Med J. 293:355–359.
  23. Saad MF, Steil GM, Kades WW, et al. 1997 Differences between the tolbutabmide-boosted and the insulin-modified minimal model protocols. Diabetes. 46:1167–1171.[Abstract]
  24. Santoro N, Wierman ME, Filicori M, Waldstreicher J, Crowley Jr WF. 1986 Intravenous administration of pulsatile gonadotropin-releasing hormone in hypothalamic amenorrhea: effects of dosage. J Clin Endocrinol Metab. 62:109–116.[Abstract/Free Full Text]
  25. Martin KA, Santoro N, Hall JE, Filicori M, Wierman M, Crowley Jr WF. 1990 Clinical review 15: management of ovulatory disorders with pulsatile gonadotropin-releasing hormone. J Clin Endocrinol Metab. 71:1081A–1081G.[Abstract/Free Full Text]
  26. Welt CK, Adams JM, Sluss PM, Hall JE. 1999 Inhibin A and inhibin B responses to gonadotropin withdrawal depend on stage of follicle development. J Clin Endocrinol Metab. 84:2163–2169.[Abstract/Free Full Text]
  27. Greenblatt E, Casper RF. 1987 Endocrine changes after laparoscopic ovarian cautery in polycystic ovarian syndrome. Am J Obstet Gynecol. 156:279–285.[Medline]
  28. Crowley Jr. WF, Beitins IZ, Vale W, et al. 1980 The biologic activity of a potent analogue of gonadotropin-releasing hormone in normal and hypogonadotropic men. N Engl J Med. 302:1052–1057.[Abstract]
  29. Filicori M, Butler JP, Crowley Jr WF. 1984 Neuroendocrine regulation of the corpus luteum in the human. J Clin Invest. 73:1638–1647.
  30. Taylor AE, Khoury RH, Crowley Jr WF. 1994 A comparison of 13 different immunometric assay kits for gonadotropins: implications for clinical investigation. J Clin Endocrinol Metab. 79:240–247.[Abstract]
  31. Ibanez L, Hall JE, Potau H, Carrascosa A, Prat N, Taylor AE. 1996 Ovarian 17-hydroxyprogesterone hyperresponsiveness to GnRH agonist challenge in women with polycystic ovary syndrome is not mediated by gonadotropin hypersecretion: evidence from GnRH agonist and human chorionic gonadotropin stimulation testing. J Clin Endocrinol Metab. 81:4103–4107.[Abstract/Free Full Text]
  32. Seminara SB, Boepple PA, Nachtigal LB, et al. 1996 Inhibin B in males with gonadotropin-releasing hormone deficiency: changes in serum concentration after short term physiologic GnRH replacement. J Clin Endocrinol Metab. 81:3692–3696.[Abstract]
  33. Taylor AE, Hubbard J, Anderson EJ. 1999 Impact of binge eating on metabolic and leptin dynamics in normal young women. J Clin Endocrinol Metab. 84:428–434.[Abstract/Free Full Text]
  34. Santen RJ, Bardin CW. 1973 Episodic luteinizing hormone secretion in man: pulse analysis, clinical interpretation, and physiologic mechanism. J Clin Invest. 52:2617–2628.
  35. Hayes FJ, McNicholl DJ, Schoenfeld D, Marsh EE, Hall JE. 1999 Free {alpha}- subunit is superior to luteinizing hormone as a marker of gonadotropin-releasing hormone despite desensitization at fast pulse frequencies. J Clin Endocrinol Metab. 84:1028–1036.[Abstract/Free Full Text]
  36. Pacini G, Bergman RN. 1986 MINMOD: a computer program to calculate insulin sensitivity and pancreatic responsivity from the frequently sampled intravenous glucose tolerance test. Comput Methods Programs Biomed. 23:113–122.[CrossRef][Medline]
  37. Filicori M, Flamigni C, Dellai P, et al. 1994 Treatment of anovulation with pulsatile gonadotropin-releasing hormone: prognostic factors and clinical results in 600 cycles. J Clin Endocrinol Metab. 79:1215–1220.[Abstract]
  38. Homburg R, Eschel A, Armar NA, et al. 1989 One hundred pregnancies after treatment with pulsatile luteinising hormone releasing hormone to induce ovulation. Br Med J. 798:809–817.
  39. McClure N, Kovacs GT, McQuinn B, Healy DL, McDonald J, Burger. 1992 Body weight, body mass index, and age: predictors of menotropin dose and cycle outcome in polycystic ovarian syndrome. Fertil Steril. 58:622–624.[Medline]
  40. Imani B, Eijkemans MJC, Te Velde ER, Habbema JDF, Fauser BCJM. 1998 Predictors of patients remaining anovulatory during clomiphene citrate induction of ovulation in normogonadotropic oligoamenorrheic infertility. J Clin Endocrinol Metab. 83:2361–2365.[Abstract/Free Full Text]
  41. Imani B, Eijkemans MJC, De Jong FH, et al. 2000 Free androgen index and leptin are the most prominant endocrine predictors of ovarian response during clomiphene citrate induction of ovulation in normogonadotropic oligoamenorrheic infertility. J Clin Endocrinol Metab. 85:676–682.[Abstract/Free Full Text]
  42. Filicori M, Dellai P, Flamingi C, Michealacci, Cognigni G, Arnone R. 1994 Increased insulin secretion in patients with multifollicular and polycystic ovaries and its impact on ovulation induction. Fertil Steril. 62:279–285.[Medline]
  43. Armstrong AB, Hoeldtke N, Wiess TE, Tuttle RM, Jones RE. 1996 Metabolic parameters that predict response to clomiphene citrate in obese oligo-ovulatory women. Military Med. 161:732–734.
  44. Murakawa H, Hasegawa I, Kurabayashi T, Tanaka K. 1999 Polycystic ovary syndrome. Insulin resistance and ovulatory responses to clomiphene. J Reprod Med. 44:23–27.[Medline]
  45. Espinosa de los Monteros A, Ayala J, Sanabria LC, Parra A. 1995 Serum insulin in clomiphene responders and nonresponders with polycystic ovarian disease. Rev Invest Clin. 47:347–353.[Medline]
  46. Laakso M. 1993 How good a marker is insulin level for insulin resistance? Am J Epidemiol. 137:959–965.[Abstract/Free Full Text]
  47. Taylor AE, Hall JE. Ovarian 17-OH progesterone hyperresponsiveness predicts hyperinsulinemia in women with polycystic ovary syndrome. Proc of the 82nd Annual Meet of The Endocrine Soc. 2000; 406.
  48. Burger HG, Groome NP, Robertson DM. 1998 Both inhibin A and B respond to exogenous follicle-stimulating hormone in the follicular phase of the human menstrual cycle. J Clin Endocrinol Metab. 83:4167–4169.[Abstract/Free Full Text]
  49. Welt CK, Smith ZA, Pauler DK, Hall JE. 2001 Differential regulation of inhibin A and inhibin B by LH, FSH, and stage of follicle development. J Clin Endocrinol Metab. In press.
  50. Anderson RA, Groome NP, Baird DT. 1998 Inhibin A and inhibin B in women with polycystic ovarian syndrome during treatment with FSH to induce mono-ovulation. Clin Endocrinol (Oxf). 48:577–584.[CrossRef][Medline]
  51. Hayes FJ, Taylor AE, Martin KA, Hall JE. 1998 Use of a gonadotropin-releasing hormone antagonist as a physiologic probe in polycystic ovary syndrome: assessment of neuroendocrine and androgen dynamics. J Clin Endocrinol Metab. 83:2343–2349.[Abstract/Free Full Text]
  52. Filicori M, Flamigni C, Campaniello E, et al. 1989 The abnormal response of polycystic ovarian disease patients to exogenous pulsatile gonadotropin- releasing hormone: characterization and management. J Clin Endocrinol Metab. 69:825–831.[Abstract/Free Full Text]
  53. Rossmanith WG, Wirth U, Benz R, Wolf AS. 1989 Endocrine dynamics during pulsatile GnRH administration in patients with hypothalamic amenorrhea and polycystic ovarian disease. Gynecol Endocrinol. 3:21–34.[Medline]
  54. Adams J, Franks S, Polson DW, et al. 1985 Multifollicular ovaries: clinical and endocrine features and response to pulsatile gonadotropin releasing hormone. Lancet. 2:1375–1378.[Medline]
  55. Vendola KA, Zhou J, Adesanya OO, Weil SJ, Bondy CA. 1998 Androgen stimulates early stages of follicular growth in the primate ovary. J Clin Invest. 101:2622–2629.[Medline]
  56. De Leo V, Lanzetta D, D’Antona D, La Marca A, Morgante G. 1998 Hormonal effects of flutamide in young women with polycystic ovary syndrome. J Clin Endocrinol Metab. 83:99–102.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 J. Clin. Endocrinol. Metab.Home page
Y. L. Pagan, S. S. Srouji, Y. Jimenez, A. Emerson, S. Gill, and J. E. Hall
Inverse Relationship between Luteinizing Hormone and Body Mass Index in Polycystic Ovarian Syndrome: Investigation of Hypothalamic and Pituitary Contributions
J. Clin. Endocrinol. Metab., April 1, 2006; 91(4): 1309 - 1316.
[Abstract] [Full Text] [PDF]

Home page
 Hum ReprodHome page
I. E. Messinis
Ovulation induction: a mini review
Hum. Reprod., October 1, 2005; 20(10): 2688 - 2697.
[Abstract] [Full Text] [PDF]

Home page
 J. Clin. Endocrinol. Metab.Home page
C. K. Welt, A. Falorni, A. E. Taylor, K. A. Martin, and J. E. Hall
Selective Theca Cell Dysfunction in Autoimmune Oophoritis Results in Multifollicular Development, Decreased Estradiol, and Elevated Inhibin B Levels
J. Clin. Endocrinol. Metab., May 1, 2005; 90(5): 3069 - 3076.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Endocrinol. Metab.Home page
E. W. C. M. van Dam, F. Roelfsema, J. D. Veldhuis, S. Hogendoorn, J. Westenberg, F. M. Helmerhorst, M. Frolich, H. M. J. Krans, A. E. Meinders, and H. Pijl
Retention of estradiol negative feedback relationship to LH predicts ovulation in response to caloric restriction and weight loss in obese patients with polycystic ovary syndrome
Am J Physiol Endocrinol Metab, April 1, 2004; 286(4): E615 - E620.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Gill, S.
Right arrow Articles by Hall, J. E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gill, S.
Right arrow Articles by Hall, J. E.

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