| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Institute for Endocrinology, Reproduction and Metabolism, Vrije Universiteit, Amsterdam, The Netherlands; Division of Reproductive Endocrinology and Fertility, Department of Obstetrics and Gynecology (J.A.M.B., M.v.d.M., J.S.), and the Division of Endocrinology, Department of Internal Medicine (E.A.v.d.V.), Free University Hospital, 1007 MB Amsterdam, The Netherlands
Address all correspondence and requests for reprints to: Dr. Jacoba A. M. de Boer, Division of Reproductive Endocrinology and Fertility, Department of Obstetrics and Gynecology, Free University Hospital, P.O. Box 7057, 1007 MB Amsterdam, The Netherlands.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
IGF-I levels as well as IGFBP-3 levels significantly increased (P < 0.01) during GH substitution. Monofollicular growth was not achieved in the first cycles. In five of six GH-substituted cycles, monofollicular growth was obtained. FSH threshold levels decreased in all patients during GH substitution. The FSH area under the curve was negatively correlated to IGF-I (r = -0.6; P < 0.05) and IGFBP-3 (r = -0.6; P < 0.05).
The results of this study indicate that GH may play a role in the physiological growth of the follicle; most likely this occurs by influencing the IGF-I or IGFBP-3 levels. GH appears to selectively increase the sensitivity of the dominant follicle to FSH, facilitating monofollicular growth.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Although it is known that pubertal development is delayed in GH-deficient (GHD) children and that this can be overcome by GH treatment (4), little is known about the consequences of GH deficiency and subsequent GH treatment on follicular development. Menashe (5) reported spontaneous pregnancies in Laron-type dwarfism, showing that IGF-I is not an absolute requisite for follicular growth. Blumenfeld (6) reported higher conception rates in GH-cotreated cycles during ovulation induction in patients who did not respond with a sufficient GH peak to a clonidine test. This was not shown in patients who did respond sufficiently. In our follow-up study subfertility was demonstrated in GHD women treated for GH deficiency during childhood. After discontinuation of GH treatment, menstrual cycle disturbances occurred in 54% of the women with spontaneous pubertal development (7).
Van Weissenbruch (8) and Van der Meer (9) developed a technique by which the responsiveness of the ovaries to gonadotropins could be quantified by determining the FSH threshold for monofollicular growth during ovulation induction.
This technique was used in the present study, to elucidate the effect of GH on follicular development in hypogonadotropic women. The objective in this study was to test the hypotheses that 1) GH administration to GHD women increases the responsiveness of the ovaries by lowering the FSH threshold; and 2) that this increase is related to the degree of GH deficiency, expressed in terms of IGF-I and IGF-binding protein-3 (IGFBP-3) levels.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Six hypogonadotropic women visiting the out-patient department
for ovulation induction and who were willing to participate in the
study were included. GH responses to an insulin tolerance test (ITT)
were determined before the first treatment cycle. GH deficiency was
defined to be present when the response did not exceed 7 µg/L. No
woman was tested who showed a normal response to the ITT. Patients with
diabetes mellitus or a body mass index less than 18 kg/m2
were excluded. Thyroid hormone or corticosteroid deficiencies, when
present, were adequately substituted for. The dosage of thyroid hormone
replacement therapy was adjusted (if necessary) during GH substitution
by monitoring serum T3 levels. Patients 2 and 3 (Table 1
) received GH therapy for GH deficiency
during childhood.
|
. The study was conducted under the guidelines of the Declaration of Helsinki and was subject to approval by the human research ethics committee of the Free University Hospital of Amsterdam.
Study design
The GH response was determined by an ITT. Patients underwent two low dose step-up ovulation induction cycles with human menopausal gonadotropin (hMG). During both cycles the FSH threshold was determined. The second cycle was preceded by GH treatment for 2 weeks, which continued during stimulation.
ITT
Hypoglycemia was induced with 0.1 IU insulin/kg BW. Samples for
GH were taken at baseline (-15 and 0 min) and 10, 20, 30, 45, 60, 90,
and 120 min after the insulin bolus. Hypoglycemia was defined as a
serum glucose level below 2.2 mmol/L combined with symptoms of
hypoglycemia. Glucose levels were determined every 5 min. A sample for
IGF-I was drawn at baseline (-15 min). Results are given in Table 2.
|
Patients were stimulated with hMG (Pergonal, Ares-Serono, The Hague, Netherlands; 75 IU FSH and 75 IU LH/ampule); actual ovulation was induced with hCG (Profasi, Ares-Serono; 10,000 IU hCG/ampule). The first treatment cycle was started either on the third day of menstruation after a previous treatment cycle or on the third day of withdrawal bleeding induced by withholding sex steroid replacement therapy. hMG was given iv using a portable infusion pump. Every 30 min a small amount of hMG was released. In this way a stable level of serum FSH could be maintained during the day. In each patient stimulation was commenced with a half-ampule (37.5 IU FSH) daily. If no ovarian response occurred after 7 days (growing follicle <11 mm and/or serum estradiol not exceeding 200 pmol/L) the serum FSH level was increased by approximately 1 IU/L at weekly intervals. This was achieved by increasing the daily FSH dose by one quarter ampule (18.75 IU FSH). When an ovarian response was observed, the dosage was maintained at a constant level until the largest follicle reached a diameter of 18 mm. At that time hMG administration was discontinued. hCG was then given im 12–24 h later. However, if at that time more than three follicles more than 16 mm or more than six follicles more than 13 mm were present, hCG was withheld to reduce the chance of a multiple pregnancy. After induction of ovulation, luteal support was provided by 10,000 IU hCG, im, every third day, to a maximum dosage of 30,000 IU. If there were no signs of follicular growth after 6 weeks of ovarian stimulation, the cycle would be canceled.
The second cycle (in which the patient was co-treated with GH) was stimulated in exactly the same manner after 2 weeks of GH pretreatment.
Cotreatment with GH
GH (16 IU/ampule; Genotropin, Pharmacia-Upjohn,
Uppsala, The Netherlands) was administered sc from the second or third
day of menstruation after the first treatment cycle. During the first 2
weeks, the GH dosage was maintained at 0.125 IU/kg·week to minimize
possible side-effects. After 2 weeks the GH dose was doubled (0.250
IU/kg·week), and hMG administration was commenced as described above.
After ovulation induction by hCG, GH administration was discontinued,
as the effects of GH in the early stages of pregnancy are not well
known. Individual GH doses are given in Table 2. The GH dosage
administrated is comparable to the recommended adult dose for GH
substitution (10). This approximates the lowest level of GH used in a
multicenter study (3) in which enhancement of gonadotropin action on
follicular development was found to be related to the GH dose.
Monitoring
During hMG treatment blood samples were drawn daily for FSH, estradiol, IGF-I, and IGFBP-3 determinations. One week after hCG administration to induce ovulation, a blood sample was drawn for progesterone, IGF-I, and IGFBP-3 measurements. During the first 2 weeks of GH treatment, a blood sample was drawn once a week for IGF-I, IGFBP-3, and estradiol determinations.
Vaginal ultrasound was performed three times a week during hMG treatment. Once a growing follicle of more than 11 mm was observed, ultrasound was performed daily until hCG administration. During the luteal phase, ultrasound was performed once, or more frequently in case of symptoms of ovarian hyperstimulation. During the first 2 weeks of GH treatment, ultrasound was performed once a week.
Assays
FSH and estradiol were determined by commercially available assays (Amerlite, Amersham, Aylesbury, UK), immunometric for FSH and competitive for estradiol. For FSH, the interassay coefficient of variation (CV) was 9% at 4.6 IU/L and 8% at 11.4 IU/L; the intraassay CV was 6% at 4.7 IU/L and 5% at 11.5 IU/L. The lower limit of detection was 0.5 IU/L.
For estradiol, the interassay CV was 8% at 511 pmol/L and 7% at 1106 pmol/L. The intraassay CV was 3% at 490 pmol/L and 3% at 1082 pmol/L. The lower detection limit was 90 pmol/L.
IGF-I levels were measured by immunoradiometric assay after extraction (Diagnostic Systems Laboratories, Webster, TX). The detection limit was 1 nmol/L. The interassay CV was 13% at 5 nmol/L and 7% at 15 nmol/L. The lower limit of the normal range was 18 nmol/L.
IGFBP-3 was determined by RIA (Diagnostic Systems Laboratories). The detection limit was 0.25 mg/L. The lower limit of the normal range was 2.7 mg/L. The intraassay CV was 4% at 5 mg/L and 2% at 25 mg/L. The interassay CV was 15% at 1.5 mg/L and 7% at 3 and 7 mg/L.
GH was determined by RIA (HGHK-2, Sorin Biomedica, Saluggia, Italy). The detection limit was 0.5 µg/L. The intraassay CV was 9% at 1 µg/L and 8% at 5 and 10 µg/L.
Data analysis
Determination of the FSH threshold. The FSH threshold is the level that the plasma concentration of FSH must exceed to initiate the final stages of follicular development, i.e. from the early antral (2–5 mm) to the preovulatory stage in the most sensitive follicle (11). Because of stepwise increments in FSH levels during stimulation, the threshold is determined semiquantitatively. It is therefore expressed as the average of the below threshold value (BTV; the highest level at which follicular growth could not be induced) and the above threshold value (ATV; the subsequent level at which follicular growth was induced). By administering FSH iv and by increasing the plasma level stepwise in small increments, BTV and ATV can be determined for each patient. This has previously been shown in hypogonadotropic patients (12) and in patients with polycystic ovary syndrome (9) It has been found that the FSH threshold in hypogonadotropic patients is a relatively stable phenomenon from cycle to cycle (12). By marginally exceeding the FSH threshold, monofollicular growth can be obtained. Monofollicular growth is defined as being present when, at the time the largest follicle reaches 18 mm (and hCG is administered), no other follicles with a diameter between 13–18 mm are also present.
Determination of FSH area under the curve (AUC). The FSH AUC was used to show the relationship between FSH and the degree of GH deficiency in terms of IGF-I and IGFBP-3 levels. The FSH AUC was calculated from the serum levels of FSH on all 10 days preceding the day the ATV was reached. These days cover the serum levels of which the BTV is composed as well as all of the serum levels until the ATV is reached, which means that they cover all of the serum levels by which the FSH threshold is determined. In this way the correlation between the FSH threshold and GH deficiency could be determined, and the influence of the number of stimulation days or the severeness of hypogonadotropism could be ruled out. (Because the serum levels of FSH were increased by approximately 1 IU/L at weekly intervals, the latter would influence the duration of stimulation and would therefore influence the FSH AUC.)
Differences in FSH threshold, estradiol levels, IGF-I, and IGFBP-3 levels were analyzed by the Wilcoxon signed rank test. The Pearson correlation test was used to determine correlations between FSH AUC and the means of IGF-I and IGFBP-3. The mean IGF-I and IGFBP-3 levels were determined from every third blood sample drawn during stimulation.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
All patients completed two treatment cycles. No cycles were canceled,
and all patients received hCG. Fluid retention (a side-effect due to GH
treatment) was seen only in patient 5. The GH dose in this patient was
decreased to 2.0 IU/day. During GH treatment IGF-I and IGFBP-3 levels
significantly increased (P < 0.01) in all patients
(Table 3).
|
Although a low-dose step-up protocol was used, we were not able to
achieve monofollicular growth in the first treatment cycles. In the
GH-treated cycles monofollicular growth was seen in five of six cycles.
Figure 1 shows the number of follicles on
the day hCG was administered.
|
. For all cycles, the FSH AUC was
negatively correlated to the mean IGF-I level as well as to the mean
IGFBP-3 level (r = -0.6; P < 0.05; Fig. 3).
|
|
), although significance was not achieved
(P = 0.09). No pregnancies were achieved.
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Adashi et al. (1) and Erickson et al. (18) showed that IGF-I synergizes with FSH in stimulating follicular maturation and growth. With GH treatment, both serum levels of IGF-I and follicular fluid concentrations increase (19). Granulosa cells do not express IGF-I messenger ribonucleic acid themselves (20). Therefore, follicular IGF-I is mainly derived from serum, as proven by Jesionowska et al. (21).
Early antral follicles achieve their responsiveness to FSH at a size of 2–5 mm (22). During subsequent growth these follicles increase their sensitivity to FSH. With regard to this, it is apparent that different follicles exhibit differing levels of sensitivity (9). It stands to reason that if IGF-I levels increase, the sensitivity of the follicles to FSH during GH treatment increases, leading to lower FSH thresholds.
In all studies concerning the effect of GH treatment on follicular growth during ovarian stimulation, the total number of ampules per cycle or the daily effective dose necessary to obtain full maturation of at least one follicle has been used as an indication of ovarian responsiveness (23). As shown by Van der Meer (9), a high variability exists in the increase in the serum FSH concentration that is reached after administration of a standard dose. It is therefore advantageous to use the FSH threshold concentration rather than the threshold dose. This is because the former gives more exact information about the responsiveness of the most sensitive follicle. Because of its stability from cycle to cycle (12), the FSH threshold can be used to evaluate the effect of interventions on cycles within one patient. With this method we were able to show the negative correlation between FSH, and IGF-I and IGFBP-3. This negative correlation indicates that the lower the IGF-I and IGFBP-3 levels, the higher the FSH levels need to be for appropriate follicle growth. This indicates that the responsiveness of the ovaries to FSH is related to the degree of GH deficiency.
The higher FSH thresholds in the first treatment cycles cannot be explained by the fact that in these cycles the ovaries were still immature. Schoemaker et al. (24) showed that for a full response of the pubertal ovary to FSH, repeated stimulations are necessary. Of the six patients in the present study, three had already been treated with hMG im (2) or pulsatile GnRH (1) before entering the study protocol. The three remaining women had previously conceived and were all using sex steroid replacement therapy before entering the study.
We were surprised to find that although a low dose step-up schedule was used in all patients and in all treatment cycles, we were not able to achieve monofollicular growth in any of the first treatment cycles (25). In other studies the number of follicles exceeding 14 mm tended to be lower or did not change during GH treatment (3, 19, 26).
It is hypothesized that GH or IGF-I/IGFBP-3 plays a role in the recruitment of the dominant follicle from its cohort, leading to monofollicular growth in humans. Eden et al. (27) found that although each cohort of follicles is exposed to the same serum level of IGF-I, dominant follicles contain up to 3 times the levels of IGF-I compared to other follicles in the same cohort. Roussie et al. (28) also found higher levels of IGF-I in the follicular fluid of mature follicles compared to the follicular fluid of immature follicles. Apparently the dominant follicle is better able than its companion follicles from the cohort to concentrate IGF-I from the serum. The results of the current study suggest that in GH deficiency low serum levels of IGF-I prevent the dominant follicle from increasing its IGF-I level. This leads to only small differences in the sensitivity to FSH of the different follicles in the cohort, which, in turn, results in multifollicular growth. During GH treatment the difference in sensitivity to FSH between this follicle and its cohort is restored by higher IGF-I levels, leading to monofollicular growth under a low dose step-up stimulation regimen.
We hypothesize that part of the menstrual cycle disturbances, found in 54% of GHD women in our earlier study (7), had been caused by a relative deficiency of FSH. Due to GH deficiency, higher levels of endogenous gonadotropins are needed to induce follicular growth. GH treatment in this group might therefore be able to cure menstrual cycle disturbances. The finding that the ovarian response to gonadotropin stimulation increases is in accordance with the finding that delayed pubertal development in GHD girls can be overcome by GH treatment (4).
The results of this study indicate that the somatotropic axis has a function in the physiological growth of follicles. GH treatment in women with low GH levels increases the sensitivity of the ovaries to gonadotropin stimulation. Importantly, this increase in sensitivity to FSH occurs preferentially in the dominant follicle, resulting in the growth of a single mature follicle.
|
Acknowledgments |
|---|
Received January 30, 1998.
Revised July 30, 1998.
Revised October 22, 1998.
Accepted October 29, 1998.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J. Neulen, D. Wenzel, C. Hornig, E. Wunsch, U. Weissenborn, K. Grunwald, R. Buttner, and H. Weich Poor responder-high responder: the importance of soluble vascular endothelial growth factor receptor 1 in ovarian stimulation protocols Hum. Reprod., April 1, 2001; 16(4): 621 - 626. [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 |
