0021-972X/86/6201-0109$02.00/0
Journal of Clinical Endocrinology and Metabolism
Copyright© 1986 by The Endocrine Society
Vol. 62, No. 1
Printed in U.S.A.
Intravenous Administration of Pulsatile Gonadotropin
Releasing Hormone in Hypothalamic Amenorrhea:
Effects of Dosage*
N. SANTORO, M. E. WIERMAN, M. FILICORIf, J. WALDSTREICHER, AND
W. F. CROWLEY, JR.
Reproductive Endocrine Unit, the Departments of Medicine and Gynecology, Vincent Research Laboratories,
Massachusetts General Hospital, Boston, Massachusetts 02115
ABSTRACT. Eighteen women with well characterized hypothalamic amenorrhea underwent 30 cycles of pulsatile GnRH
treatment in an effort to examine the role of GnRH dosage in
pituitary and ovarian responses. GnRH was administered iv at
2 doses (25 and 100 ng/kg bolus) at a physiological range of
frequencies (90 and 60 min) in the follicular phase of the induced
cycles. After demonstration of ovulation by ultrasound and
clinical parameters, the frequency of GnRH administration was
progressively slowed from every 60 min to every 90 min and
then to every 240 min to mimic the slowing of endogenous LH
secretion that occurs during the luteal phase in normal women.
The results of these induced cycles were compared to those of
62 ovulatory cycles from normal women.
Overall clinical and biochemical results revealed the following. Patients receiving doses of 25 ng/kg GnRH successfully
ovulated only 80% of the time, with recruitment of a single
dominant follicle. Two of 5 patients became pregnant. Peak
estradiol levels were significantly lower than normal [261 ± 33
(±SE) US. 342 ± 11 pg/ml, respectively; P < 0.02]. Integrated
luteal phase progesterone production was also significantly re-
H
YPOTHALAMIC amenorrhea is the clinical syndrome that results from disordered secretion of
endogenous GnRH (1-3). Due to this impairment of
normal neuroendocrine function, these patients have
arrested folliculogenesis and, thus, are ideal candidates
for replacement with a GnRH regimen that simulates
the normal program of hypothalamic secretion of this
releasing factor. In addition, this disorder provides an
opportunity to test various regimens of GnRH treatment
and compare their abilities to reconstruct the normal
ovulatory process in a physiological manner. Such an
effort, however, assumes the prior acquisition of normaReceived May 23,1985.
Address requests for reprints to: Dr. W. F. Crowley, Jr., Department
of Medicine, Massachusetts General Hospital, Vincent Research Laboratories, Boston, Massachusetts 02115.
* This work was supported by NIH Grants HD-15080 and RR-1066
and the Vincent Research Fund.
t Current address: Department of Reproductive Medicine, University of Bologna, via Massarenti 13, Bologna, Italy.
duced in the 25 ng/kg group compared to normal (78 ± 17 vs.
145 ± 8 ng/ml/entire luteal phase, respectively; P < 0.02). All
women receiving bolus doses of 100 ng/kg GnRH ovulated;
maturation of multiple follicles occurred in 5 of 20 cycles, and 6
of 7 women conceived. Peak estradiol values were significantly
higher than those in either normal women or the 25 ng/kg group
(478 ± 48 pg/ml; P < 0.02 for both), with integrated luteal phase
progesterone levels significantly higher than those in patients
receiving the 25 ng/kg dose (196 ± 25 ng/ml/luteal phase; P <
0.02).
This study demonstrates that 1) ovulation and fertility can
be achieved with a physiological frequency regimen of pulsatile
GnRH administration using bolus doses of both 25 and 100 ng/
kg in women with hypothalamic amenorrhea; 2) the 25 rig/kg
dose of GnRH may represent a threshold of stimulation of the
pituitary-ovarian axis and recreates cycles with an inadequate
luteal phase; and 3) a 100 ng/kg dose of GnRH may well cause
a supraphysiological stimulation of the pituitary-gonadal axis.
(J Clin Endocrinol Metab 62: 109,1986)
tive data which can be employed to derive a regimen of
GnRH administration that mimics the pattern of GnRH
secretion that occurs during the normal menstrual cycle.
We have previously accumulated data from 62 ovulatory menstrual cycles in normal women, in whom gonadotropins and sex steroids were measured daily for a
complete cycle, and gonadotropins were measured at 10min intervals at 1 of 6 key periods of follicular or corpus
luteum function (4, 5). From these data, a program of
frequency of GnRH administration was constructed to
mirror that of normal endogenous GnRH secretion. Using this frequency program and the iv route of administration of GnRH (6, 7), we held these two critical variables constant and examined the impact of varying the
dosage of GnRH on the resulting pituitary and ovarian
responses.
Materials and Methods
Patient population
Normal subjects. The control series consisted of studies of 62
109
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 July 2015. at 09:35 For personal use only. No other uses without permission. . All rights reserved.
110
SANTORO ET AL.
ovulatory menstrual cycles in 39 paid normal women, aged 1839 (8). Each subject had 1) a history of regular 27- to 32-day
cycles, 2) a body weight within 1 SD of normal (9), 3) no history
of heavy exercise, 4) no evidence of hirsutism, galactorrhea, or
genital abnormalities; 5) normal serum PRL and T 4 levels, and
6) normal luteal function in the menstrual cycle preceding the
study, as determined by a biphasic basal body temperature
chart and a midluteal serum progesterone level greater than 6
ng/ml. None had received any medication for at least 6 months
preceding the study.
Amenorrheic patients. Eighteen women, aged 17-39 yr (mean,
28.1 yr), who were otherwise healthy, had the following clinical
and biochemical findings: 1) normo- or hypogonadotropic
amenorrhea of at least 6 months duration, 2) body weight
greater than or equal to 1 SD below average [mean weight 67.9
± 20.0 (±SD) kg; mean height, 165.5 ± 8.5 (±SD) cm]; 9), 3) an
absence of a history of intensive exercise, 4) a normal physical
examination, with no hirsutism or ovarian enlargement, 5)
normal serum TSH, T4, T3, and PRL levels, 6) no androgen
excess (normal 24-h urinary 17-ketosteroid excretion and/or
normal serum dehydroepiandrosterone sulfate and testosterone
levels), and 7) no clinical or rbentgenographic evidence of a
hypothalamic or pituitary defect. None of these subjects had
received any sex steroid therapy for at least 6 months before
participation. Informed consent was obtained from each subject
before participation. This study was approved by the Subcommittee on Human Studies of the Massachusetts General Hospital.
Nine of the study subjects had primary amenorrhea and,
thus, were considered to have idiopathic hypogonadotropic
hypogonadism (IHH); of these, three were anosmic (Kallmann's syndrome). The diagnosis of IHH was confirmed in all
women by normal cranial tomography of the hypothalamicpituitary area and normal basal and stimulated levels of other
anterior pituitary hormones in response to insulin (0.15 U/kg)
and TRH (200 /xg). The remaining nine patients had secondary
amenorrhea, and all had abnormal baseline gonadotropin secretory profiles indicative of GnRH deficiency, as previously
reported (3). In those patients desiring pregnancy, tubal patency was assured by either hysterosalpingography or laparoscopic dye studies, and postcoital tests were normal, as was
seminal fluid analysis of the partner.
Protocol
Monitoring. Normal subjects: The normal women were monitored with a single daily blood sample for gonadotropin and sex
steroid levels and basal body temperature charting.
Amenorrheic patients: All amenorrheic patients were monitored with daily blood sampling for LH, FSH, estradiol (E2),
and progesterone (P). Each blood sample was obtained 45 min
after a GnRH dose. GnRH was administered iv via portable
infusion pumps (Ferring), the reservoir of which contained a
heparinized (100 IU/ml) solution of GnRH. Intravenous lines
were inspected visually daily and were maintained for up to 14
days. The course of follicular maturation was followed by serial
ultrasonography, cervical mucus examinations, and basal body
temperature charting.
GnRH dosage and frequency. The frequency of GnRH admin-
JCE & M • 1986
Vol 62 • No 1
istration was adjusted throughout the cycle to conform as
closely as possible to the progression of gonadotropin pulsations
during a normal ovulatory cycle, as outlined below (5, 8).
Patients with hypothalamic amenorrhea were then divided into
3 groups on the basis of the GnRH regimen employed. One
patient underwent 4 cycles of GnRH therapy, 2 patients underwent 3 cycles of GnRH treatment, and the remainder of the
amenorrheic women contributed 1 or 2 cycles to the general
data analysis. Three subjects are represented in more than 1
group, as they underwent more than 1 attempt at ovulation
induction with a different regimen of GnRH administration;
thus, a total of 30 cycles were studied.
Group I (25 ng/kg): Group I consisted of 7 patients (10 cycles)
who received 25 ng/kg bolus doses of GnRH every 60 min in
the follicular phase. When ovulation was confirmed by the
above-mentioned clinical parameters and ultrasound examination, the GnRH frequency was slowed to every 90 min. Seven
days after ovulation, the frequency of GnRH administration
was decreased further to 240-min intervals consistent with the
progressive luteal slowing of GnRH-induced gonadotropin pulsations that occurs in normal women (4).
Group II (100 ng/kg): The 7 women in group II (10 cycles)
each received 100 ng/kg bolus doses of iv GnRH. Frequency
adjustments were identical to those in group I.
Group HI (100 ng/kg; altered frequency regimen): The 7
women in this group (10 cycles) were managed with an alteration in the frequency of GnRH administration during the first
week of therapy. These patients initially received GnRH (100
ng/kg) every 90 min for the first 7 days. Subsequently, GnRH
was given every 60 min until ovulation occurred. Subsequent
frequency adjustments in the luteal phase were made exactly
as described for groups I and II. Since patients who received
100 ng/kg bolus doses started on the slower (90-min) and faster
(60-min) frequencies (groups II and III) did not differ significantly from one another with respect to any of the parameters
studied, the data for these groups were combined and analyzed
as the 100 ng/kg group.
To detect a difference in results based on dosage and not due
to variable responsiveness of individuals with differing patterns
of hypothalamic amenorrhea (3), we attempted to provide a
similar mix of patients in both the 25 and 100 ng/kg groups.
The 25 ng/kg group (n = 7) consisted of five patients who had
no pulsatile gonadotropin secretion. Two additional patients
had variably disordered patterns of endogenous GnRH secretion. The results were similar in these two subgroups.
The 100 ng/kg group consisted of 12 women, 7 of whom were
apulsatile, and 5 of whom had abnormal, but detectable, LH
pulsations during the frequent sampling study. However, as in
the subjects receiving 25 ng/kg, no qualitative differences were
found between the subgroups in terms of time to ovulation,
peak sex hormone values, or occurrence of multiple follicular
recruitment.
Thus, the majority of women in both groups were apulsatile
in their pretherapy gonadotropin secretory patterns. When the
apulsatile women alone were analyzed, no significant differences were found between the lengths of follicular and luteal
phases with respect to 25 ng/kg, 100 ng/kg, or normal cycles.
Significant differences for the 25 ng/kg cycles with respect to
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 July 2015. at 09:35 For personal use only. No other uses without permission. . All rights reserved.
iv GnRH IN HYPOTHALAMIC AMENORRHEA
normal and 100 hg/kg cycles were as described for the larger
groups (see below).
Laboratory methods
All serum samples from each subject were processed in a
single RIA for LH and FSH determinations, as previously
described (4). Values were expressed as equivalents of the
Second International Reference Preparation of human menopausal gonadotropin (milliinternational units per ml). The
interassay and intraassay coefficients of variation for FSH were
12.3% and 7.8%, respectively; the corresponding values for LH
were 5.4% and 5.8%. Serum E2 and P values were determined
as previously described (4). The interassay coefficient of variation for E2 at 50% hormone binding was 5.8%; the corresponding value for the P assay was 5.8%.
Data analysis
Analysis of all results was standardized to the day of the
midcycle surge. Only ovulatory cycles, defined as those in which
luteal P rose to at least 3 ng/ml, were included in data analysis
of all patients and normal women. The day of the midcycle
surge, designated day 0, was required to contain three of the
following four criteria: 1) the day of the LH peak, 2) the day of
the FSH peak, 3) the day of or day after the peak E2 level, and
4) the day P levels doubled from baseline and/or reached 0.6
ng/ml (10). Using this system, data represented at the extremes
of the figures consist of observations from progressively fewer
patients. In addition, peak mean E2 levels (obtained by averaging individual peak values) may vary from the depicted mean
on the day of the midcycle surge due to the criteria used for
determination of the midcycle surge (see above). FSH levels
were also analyzed according to the first day of menses in
normal women or the first day of GnRH administration in the
patient groups.
To evaluate corpus luteum function, P levels were integrated
from day - 2 (i.e. 2 days before the midcycle surge) to the day
of subsequent menses using a computerized program which
yielded a sum of the area under the P curve (Psum). Data from
patients who conceived during GnRH therapy were not included in the luteal phase P data; however, gonadotropins and
E2 values from these patients were combined with those of the
other patients up to 7 days postmidcycle surge, i.e. before the
onset of detectable serum hCG levels. Inadequate luteal phases
were defined as those having a peak P value of 6 ng/ml or less
and/or a duration of less than 10 days. Ovulatory cycles manifesting inadequate luteal phases were included in data analysis.
Statistics
Results of the GnRH-induced cycles were compared to those
of normal women using two-tailed t testing. All gonadotropin
and sex steroid values are given as the mean ± SEM.
Results
Normal women
The normal profiles of LH, FSH, E2, and P in the
cycles from the normal women are portrayed as the
111
shaded background in Figs. 1 and 2 and have been reported previously (8). By the above criteria, all of these
cycles were ovulatory. Mean follicular phase length was
13.7 ± 0.3 days, and mean luteal phase length was 14.0
± 0.3 days. Individual peak mean follicular phase E2
levels were 342 ± 1 1 pg/ml, and luteal phase P sums
were 145 ± 8 ng/ml. Only one of these cycles had an
inadequate luteal phase.
Amenorrheic patients: clinical results
The pretreatment baseline evaluation of the amenorrheic women revealed the following characteristics.
Serum E2 levels were near assay detection limits (20 pg/
ml) in these subjects, with a mean of 23 ± 2 pg/ml. PRL
levels were normal (<15 ng/ml) in all subjects, with a
mean of 5.2 ± 0.8 ng/ml. Serum total and free testosterone levels were 52.9 ± 11.2 and 0.64 ± 0.12 ng/dl,
respectively, both well within normal limits (25-90 and
0.09-1.28 ng/dl). Twenty-four-hour 17-ketosteroid excretion, presumably due to the negligible contribution of
ovarian androgens in these hypogonadal subjects, was
4.0 ±1.0 mg/24 h, with a normal range of 4-16 mg/24 h.
The overall clinical results of the ovulation induction
cycles are summarized in Table 1. Since not all patients
who requested ovulation induction were attempting to
conceive, the percentage of cycles in which conception
occurred is adjusted for the number of women pursuing
this outcome.
Amenorrheic patients: effects of GnRH dose
25 ng/kg cycles. The 25 ng/kg regimen resulted in folliculogenesis and ovulation in 8 of 10 cycles. In 1 cycle,
folliculogenesis occurred, with growth of a single follicle
to 2.9 cm and a rise in serum E2 to 224 pg/ml; however,
P levels did not exceed 1.9 ng/ml after the initial E2
peak. The other anovulatory cycle was characterized by
a minimal steroid response, no discernible follicular development, and termination of the cycle after 25 days.
Two of 5 women attempting pregnancy conceived. Both
conceptions occurred during the second cycle of GnRH.
The mean durations of the follicular and luteal phases
in the ovulatory cycles were 13.9 ± 2.0 and 12.7 ± 0.9
days, respectively, neither of which differed from normal.
A single dominant follicle was seen by ultrasound examination in all of these ovulatory cycles, with circulating E2 and P levels consistent with a single ovulation in
every case. Gonadotropin and sex steroid patterns in
these 25 ng/kg cycles are contrasted with those in normal
women in Fig. 1. Since the data shown in this figure are
standardized to the day of the midcycle surge, as defined
above, values at the extremes of this graph are representative of progressively fewer observations. As can be
seen, E2 secretion followed a pattern similar to but
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 July 2015. at 09:35 For personal use only. No other uses without permission. . All rights reserved.
SANTORO ET AL.
112
JCE & M • 1986
Vol 62 • No 1
25 ng/kg
FlG. 1. Serum LH, FSH, E2) and P concentrations (mean ± SEM) during GnRH
administration (25 ng/kg) to women
with hypothalamic amenorrhea. Mean ±
SEM values in normal women are represented by the shaded areas. Values beyond —6 or +9 days on the x-axis represent observations on progressively fewer
subjects than the indicated number (n).
Unconnected points represent observations on a single patient.
DAYS
#•"; Normal Women, n = 6 2
* A 2 5 ng/kg, n=8
Ar* 100 ng/kg, n = 2 0
DAYS
TABLE 1. Overall clinical and biochemical results of GnRH replacement cycles
Dose
No.
of
patients
No.
of
cycles
Ovulation
Conception
25 ng/kg
100 ng/kg
7
12
10
20
80
100
40
86
Normal
39
62
100
201
Q.
(T
H
•b
c
15-
I
10-
a
Z>
H
E
Psum
261 ± 33"
478 ± 48fc
78 ± 17°
196 ± 25"
87.5
100
342 ± II* 6
145 ± 8*
98.4
P>6
The percentage of conceptions is adjusted for the number of women
attempting pregnancy in each group.
CVI
x
en
E2 peak
P<0.02vs. 25 ng/kg.
* P < 0.02 us. 100 ng/kg.
5-
0J
i
1
!
1
|
1
I
2
3
4
5
6
DAY OF CYCLE
FIG. 2. Comparison of serum FSH levels (mean ± SEM) during the
early follicular phase of the menstrual cycle in normal women (shaded
background; ±1 SE) and in women receiving 25 and 100 ng/kg GnRH.
Data are standardized to the onset of menses in normal women or the
first day of GnRH administration in women with hypothalamic aiiienorrhea.
slightly lower than normal. The individual mean peak E2
level was 261 ± 33 pg/ml, which was significantly lower
than that in normal women (P = 0.012). In addition, P
production by the subsequent corpus luteum was well
below that in the normal women, as demonstrated by the
luteal phase sums of P (78 ± 17 ng/ml/luteal phase; P =
0.01). The early follicular phase FSH levels in the 25 rig/
kg cycles also were below the lower limits of the normal
range. In Fig. 2, FSH levels are shown standardized to
day 1 of the normal menstrual cycle or the first day of
pulsatile GnRH administration. Mean FSH levels in the
25 ng/kg group increased at the beginning of the cycles,
but integrated FSH levels over the first 6 days of the 25
ng/kg cycles were slightly but not significantly lower
than those in normal women.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 July 2015. at 09:35 For personal use only. No other uses without permission. . All rights reserved.
iv GnRH IN HYPOTHALAMIC AMENORRHEA
100 ng/kg cycles. The 100 ng/kg cycles are compared to
those in the normal women in Fig. 3. All patients who
received this dose of GnRH ovulated, and 6 of the 7
women attempting pregnancy successfully conceived (all
within 3 cycles of therapy). There was 1 twin pregnancy.
Once again, the durations of the follicular and luteal
phases (12.8 ± 0.8 and 14.1 ± 0.6 days, respectively) did
not differ from those in normal women or the 25 ng/kg
group. In contrast to the situation in the 25 ng/kg group,
ultrasonic evidence of recruitment of multiple follicles
was common in the 100 ng/kg group (seen in 5 of 20
cycles), with E2 and P values in these cycles approximately twice normal, consistent with multiple ovulation.
As can be seen in Fig. 3, E2 secretion was greater than
normal, with an individual mean peak value of 478 ± 48
pg/ml (P = 0.011 us. normal). P secretion by the corpus
luteum of the 100 ng/kg cycles was slightly but not
significantly greater than normal (196 ± 25 ng/ml/luteal
phase; P = 0.073), due to the scatter of values engendered
by the presence of single and multiple corpora lutea.
Integrated luteal P sums in the 100 ng/kg cycles were
significantly increased compared to those in the 25 ng/
kg cycles (P < 0.01). The pattern of augmented FSH
secretion in the early follicular phase of the 100 ng/kg
cycles (Fig. 2) was not statistically different from that in
normal women or women receiving 25 ng/kg.
Five women in the 100 ng/kg group underwent their
initial and second cycles of GnRH replacement in im-
113
mediate succession. The follicular phase lengths, luteal
phase durations, peak E2 levels attained, and P sums
were not significantly different between the first and
second treatment cycles.
Side-effects of iv therapy
Maintenance of iv lines for periods up to 14 days in
ambulatory, otherwise healthy young women was relatively free of complications. One case of mild phlebitis
necessitating catheter removal and local therapy only
occurred. Blood and catheter tips were routinely cultured
whenever iv lines were changed. Aside from the occasional growth of skin contaminants (e.g. Staphylococcus
epidermidis) on the catheter tip (8 positive reports in the
30 cycles) or the rare blood culture (1 positive report in
the 30 cycles), all cultures were sterile. Patients were
instructed to record their oral temperatures twice a day,
and no subject had a febrile episode attributable to the
presence of the iv catheter.
Discussion
The advantages and disadvantages of various GnRH
dosage regimens and routes of administration have been
explored by several investigators (6, 7, 11-19). Recent
observations in women with hypothalamic amenorrhea
(11) revealed that iv GnRH doses as high as 20 ng induced
supranormal E2 and P responses, occasional mild ovarian
IOO ng/kg
C
CJ
20-
e
H
FIG. 3. Serum LH, FSH, E2, and P concentrations (mean ± SEM) during GnRH
administration (100 ng/kg) in women
with hypothalamic amenorrhea. Mean ±
SEM values in normal women are represented by the shaded areas. As in Fig. 1,
values beyond - 7 or +10 days from the
midcycle surge (day 0) represent observations on progressively fewer patients
than the indicated number (n).
E 400-
DAYS
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 July 2015. at 09:35 For personal use only. No other uses without permission. . All rights reserved.
114
SANTORO ET AL.
hyperstimulation, and multiple pregnancies compared to
a smaller group of normal women. We have employed a
considerably lower dose of 100 ng/kg (5-6 ^g) of GnRH
and demonstrated a similar, though milder, supraphysiological response. Although most investigators agree that
the iv route is considerably more effective in inducing
ovulation in women with hypothalamic amenorrhea (6,
11), others have studied both iv and sc administration
(7, 11, 13) or used the sc route alone for reasons of
convenience (14-19). Studies employing the sc route for
GnRH therapy have generally indicated that higher
doses are required for longer periods of time, lower rates
of ovulation are achieved, and unphysiological patterns
of gonadotropin secretion can result, since prolonged and
variable GnRH absorption may occur (6, 7, 11, 15). In
the present work, we chose to employ the iv method
exclusively due to its superior ability to deliver a discrete
dose of GnRH to the pituitary and to elicit a pattern of
LH release that most closely resembles that in normal
women (6, 7). Utilization of this route then allowed us
to compare the pituitary and gonadal responses of these
subjects with those of normal women, and thereby modify
our GnRH replacement regimen appropriately.
The optimal concentration of exogenous GnRH required for restoration of cyclic function in women with
hypothalamic amenorrhea is controversial. Since considerable interspecies differences in GnRH sensitivity exist,
previous animal experience is of limited value in establishing a dose of GnRH that mimics the normal amplitude of gonadotropin release during the menstrual cycle
(20-25). Measurements of GnRH in the peripheral circulation are not helpful, since they do not correlate with
the portal levels of GnRH (26), probably due to its short
half-life and relatively confined space of distribution
within the hypothalamic-pituitary blood supply. Furthermore, GnRH secretion may change in quantity during the course of the normal menstrual cycle, rendering
direct hypophyseal-portal blood measurements insufficient to provide the necessary data (27).
The 25 ng/kg dose of GnRH used in this study was
derived from previous information concerning the physiological range of GnRH doses used to restore normal
gonadotropin secretory patterns, sex steroid production,
and gametogenesis in GnRH-deficient men (28, 29).
However, in contrast to the experience in GnRH-deficient men, the 25 ng/kg GnRH dose in women appears
to represent a threshold dose for folliculogenesis, since
ovulation occurred in only 8 of 10 cycles. The overall
deficient P secretion by the subsequent corpus luteum in
those cycles that proved ovulatory suggests that the 25
ng/kg dose is marginal in its ability to restore the normal
amplitude of pituitary-gonadal responses in GnRH-deficient women. This observation taken together with our
previous experience in GnRH-deficient men implies that
JCE & M • 1986
Vol 62 • No 1
women in the follicular phase secrete a larger quantity
of GnRH per secretory episode than adult men.
Conversely, 100 ng/kg, while uniformly effective in
inducing ovulation in our small series of patients, is
probably a slightly supraphysiological dose of GnRH for
the follicular phase of the cycle. This observation is based
on the higher than normal early follicular phase FSH
levels, the significantly elevated midcycle E2 peaks, the
occurrence of multiple follicular development by ultrasonic detection, the high early luteal phase P levels, and
twinning in one of seven pregnancies. Clearly, in patients
with multiple folliculogenesis, E2 levels rose well above
the normal range, to individual levels as high as 1046 pg/
ml, and luteal P values during these cycles were twice
those in normal women. In concert, these findings suggest that endogenous negative feedback mechanisms of
E2 regulation of FSH secretion are intact during pulsatile
GnRH therapy, but overriding of this negative feedback
can occur with the use of supraphysiological levels of
GnRH. Of interest was the finding that the LH peak in
patients receiving 100 ng/kg bolus doses of GnRH was
significantly blunted compared to normal women (mean,
65 ± 8 vs. I l l ± 7 mlU/ml; P = 0.001), suggesting that
suppression of positive feedback on the pituitary gland
by these increased levels of E2 may have occurred. Alternatively, nonsteroidal substances produced by the overstimulated ovary may be responsible for suppressing the
LH surge, as has been suggested by others (30). Luteal
phase ultrasonography was also performed routinely 7
days postovulation. Although ascites was never detected
in any patient who received 100 ng/kg, mild ovarian
enlargement with multiple cysts compatible with corpora
lutea was observed in five women receiving the 100 ng/
kg dose. Thus, subtle adjustments of the GnRH dose
during the follicular phase might be important in avoiding multiple pregnancy while using the minimum effective amount that will induce ovulation. Conversely, the
use of higher doses of GnRH in the follicular phase of
the cycle may well have a role in ovulation induction
attempts associated with in vitro fertilization programs
where multiple folliculogenesis is a goal (8, 31).
In humans, GnRH exerts a self-priming effect by sensitizing the pituitary gland to subsequent GnRH exposure for up to 1 month (29, 32). From these findings, one
might anticipate that an initial cycle of exposure to
exogenous GnRH would be less than satisfactory, but
could lead to improved responses in following cycles. In
the limited context of these studies, however, it did not
appear that a previous cycle exerted either a positive or
a negative influence on succeeding cycles. The pituitary
response was neither increased nor decreased in the five
women who underwent initial and second cycles of
GnRH replacement without interruption. Furthermore,
gonadal responses did not differ between these two
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 July 2015. at 09:35 For personal use only. No other uses without permission. . All rights reserved.
iv GnRH IN HYPOTHALAMIC AMENORRHEA
groups, implying that the ovaries of women with longstanding hypothalamic amenorrhea are ready to respond
as soon as a physiological program of GnRH is instituted.
Cycles of conception were not included in evaluation
of luteal phase P secretion. Nonetheless, P secretion
during the 100 ng/kg cycles compared favorably with
normal. Since no difference occurred in initial and second
cycles, it is unlikely that priming cycles skewed comparisons of luteal phase P production. Furthermore, since
all conceptions occurred within three cycles of GnRH
administration and were approximately evenly distributed among the first, second, and third cycles, the nonconception cycles were unlikely to have been due to
unfavorable patterns of P secretion or other hormonal
factors. Indeed, this distribution of conceptions suggests
the pregnancy pattern that occurs in normal women.
The difference in frequency of GnRH administered in
the early follicular phase (90 vs. 60 min) to women who
received 100 ng/kg GnRH failed to cause different patterns of gonadotropin secretion, folliculogenesis, or luteal
function. The small numbers of cycles studied might be
insufficient to elucidate a subtle difference. On the other
hand, it is possible that the hypothalamic-pituitary-ovarian axis may respond identically to frequencies that fall
within a tolerable range (8, 33).
Intravenous pulsatile GnRH administration appears
to be an effective and feasible technique for restoring
fertility in women with IHH. The adoption of a physiological frequency of GnRH stimulation resulted in uniform ovulation in properly selected patients suffering
from a presumed deficiency of endogenous GnRH secretion. Inadequate quantities of GnRH stimulation sometimes caused attenuated folliculogenesis, with subsequent inadequate corpus luteum function. Slightly excessive doses of GnRH resulted in more exuberant follicular
growth and corpus luteum function, but with concomitant problems of overstimulation, multiple folliculogenesis, and multiple gestation. Further studies of different dosages of GnRH will be required to determine that
which simulates the physiology of the normal menstrual
cycle more closely.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
Acknowledgments
The authors wish to thank Raechel Katzin and Birgit Keller for
technical assistance; Carole Thompson, Nancy Perry, and Jacquelyn
Donnelly for their help in preparation of the manuscript; Dr. Robert
Neer and the staff of the General Clinical Research Center for their
expert assistance in patient care; and Steven Trigilio and Donna
Beardsworth for preparation of artwork.
References
1. Kletsky O, Davajan V, Nakamara RM, Mishell DR 1975 Classification of secondary amenorrhea based on distinct hormonal patterns. J Clin Endocrinol Metab 41:660
2. Wildt L, Schwilden H, Wesner G, Roll C, Brensing K-A, Luckhaus
18.
19.
20.
21.
22.
23.
115
J, Bahr M, Leyendecker G 1983 The pulsatile pattern of gonadotropin secretion and follicular development during the menstrual
cycle and in women with hypothalamic and hyperandrogenic amenorrhea. In: Leyendecker G, Stock H, Wildt L (eds) Brain and
Pituitary Peptides II: Pulsatile Administration of Gn-RH in Hypothalamic Failure: Basic and Clinical Aspects. Karger, Basel, pp
11-27
Santoro N, Filicori M, Wierman M, Hall J, Crowley Jr WF 1984
The spectrum of abnormal patterns of gonadotropin-releasing hormone (GnRH) secretion in hypothalamic amenorrhea. Clin Res
32:690A (Abstract)
Filicori M, Butler JP, Crowley Jr WF 1984 Neuroendocrine regulation of the corpus luteum in the human: evidence for pulsatile
progesterone secretion. J Clin Invest 73:1638
Filicori M, Hoffman AR, Mansfield MJ, Dunaif AE, Beardsworth
D, Trigilio S, Donnelly J, Crowley WF, The frequency modulation
of pulsatile release of gonadotropins in the human menstrual cycle.
64th Annual Meeting of The Endocrine Society, San Francisco
CA, 1982, p 280A (Abstract 803)
Reid RL, Leopold GR, Yen SSC 1981 Induction of ovulation and
pregnancy with pulsatile luteinizing hormone releasing factor: dosage and mode of delivery. Fertil Steril 36:553
Handelsman DJ, Jansen RPS, Boylan L, Spaliviero JA, Turtle JR
1984 Pharmacokinetics of gonadotropin-releasing hormone: comparison of subcutaneous and intravenous routes. J Clin Endocrinol
Metab 59:739
Crowley Jr WF, Filicori M, Spratt D, Santoro N 1985 The physiology of gonadotropin-releasing hormone (GnRH) secretion in men
and women. Recent Prog Horm Res 41:473
Sargent DW 1963 Weight-height relationship of young men and
women. Am J Clin Nutr 13:318
Hoff JD, Quigley ME, Yen SSC 1983 Hormonal dynamics at
midcycle: a reevaluation. J Clin Endocrinol Metab 57:792
Leyendecker G, Wildt L 1984 Pulsatile administration of GnRH
in hypothalamic amenorrhea. Up J Med Sci 89:19
Crowley Jr WF, McArthur JW 1980 Stimulation of the normal
menstrual cycle in Kallmann's syndrome by pulsatile administration of luteinizing hormone releasing hormone (LHRH). J Clin
Endocrinol Metab 51:173
Mason P, Adams J, Morris DV, Tucker M, Price J, Voulgaris Z,
van der Spuy ZM, Sutherland I, Chambers GR, White S, Wheeler
MJ, Jacobs HS 1984 Induction of ovulation with pulsatile luteinizing hormone releasing hormone. Br Med J 288:181
Mason P, Adams J, Tucker M, Jacobs HS 1984 The role of
subcutaneous luteinizing hormone releasing hormone in the induction of ovulation. Ups J Med Sci 89:33
Hurley DM, Brian R, Outch J, Stickdale J, Fry A, Hackman C,
Clarke I, Burger HG 1984 Induction of ovulation and fertility in
amenorrheic women by pulsatile low-dose gonadotropin-releasing
hormone. N Engl J Med 310:1069
Skarin G, Nillius SJ, Wide L 1982 Pulsatile low-dose luteinizing
hormone-releasing hormone treatment for induction of follicular
maturation and ovulation in women with amenorrhea. Acta Endocrinol (Copenh) 101:78
Nillius SJ, Skarin G, Wide L 1984 Subcutaneous pulsatile LH-RH
therapy of secondary amenorrhoea. Up J Med Sci 89:53
Seibel MM, Kamrava M, McArdle C, Taymor ML 1983 Ovulation
induction and conception using subcutaneous pulsatile luteinizing
hormone-releasing hormone. Obstet Gynecol 61:292
Hammond CB, Wiebe RH, Haney AF.Yancy SG 1979 Ovulation
induction with luteinizing hormone-releasing hormone in amenorrheic, infertile women. Am J Obstet Gynecol 135:924
Knobil E 1980 The neuroendocrine control of the menstrual cycle.
Recent Prog Horm Res 36:53
Carmel PW, Araki S, Ferin M 1976 Pituitary stalk portal blood
collection in rhesus monkeys: evidence for pulsatile release of
gonadotropin-releasing hormone (GnRH). Endocrinology 99:243
Clarke IJ, The relationships between GnRH and LH secretion. 7th
International Congress of Endocrinology, 1984, p 80 (Abstract
S.66)
Clarke IJ, Cummins JT 1982 The temporal relationship between
gonadotropin-releasing hormone (GnRH) and luteinizing hormone
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 July 2015. at 09:35 For personal use only. No other uses without permission. . All rights reserved.
SANTORO ET AL.
116
(LH) secretion in ovariectomized ewes. Endocrinology 111:1737
24. Clarke IJ, Cummins JT, deKretser DM 1983 Pituitary gland function after disconnection from direct hypothalamic influences in the
sheep. Neuroendocrinology 36:376
25. Karsch FJ 1980 Seasonal reproduction: a saga of reversible fertility.
Physiologist 23:29
26. Elkind-Hirsch K, Ravnikar V, Tulchinsky D, Schiff I, Ryan KJ
1984 Episodic secretory patterns of immunoreactive luteinizing
hormone-releasing hormone (IR-LH-RH) in the systemic circulation of normal women throughout the menstrual cycle. Fertil Steril
41:56
27. Antunes JL, Carmel PW, Housepian EM, Ferin M 1978 Luteinizing
hormone-releasing hormone in human pituitary blood. J Neurosurg
49:382
28. Hoffman AR, Crowley Jr WF 1982 Induction of puberty in men by
long-term pulsatile administration of low-dose gonadotropin-releasing hormone. N Engl J Med 307:1237
29. Valk TW, Corley KP, Kelch RP, Marshall JC 1980 Hypogonado-
30.
31.
32.
33.
tropic hypogonadism: hormonal responses to low dose pulsatile
administration of gonadotropin-releasing hormone. J Clin Endocrinol Metab 51:730
Sopelak VM, Hodgen GD, Porcine follicular fluid (pFF) induced
blockade of estrogen positive feedback: effect of pulsatile gonadotropin-releasing hormone (GnRH). 32nd Annual Meeting of the
Society for Gynecologic Investigation, Phoenix, AZ, 1985 (Abstract
3)
Liu JH, Durfee R, Muse K, Yen SSC 1983 Induction of multiple
ovulation by pulsatile administration of gonadotropin-releasing
hormone. Fertil Steril 40:18
Hashimoto T, Miyai K, Onishi T, Matsumoto K, Kumahara Y
1975 Comparison of short and long-term treatment with synthetic
LH-releasing hormone and clomiphene citrate in male hypothalamic hypogonadism. J Clin Endocrinol Metab 41:905
Wildt L, Hausler A, Marshall G, Hutchison JS, Plant TM, Belchetz
PE, Knobil E 1981 Frequency and amplitude of gonadotropinreleasing hormone stimulation and gonadotropin secretion in the
rhesus monkey. Endocrinology 109:376
Postgraduate Course in Clinical Nutrition by the American Society for Clinical
Nutrition
A postgraduate course in Clinical Nutrition, sponsored by the American Society for Clinical Nutrition,
will be held April 28 and 29, 1986 at the Sheraton Washington Hotel, Washington, DC. The course will
include sessions on nutrition and eating behaviour, dietary influences on atherogenic risk factors, nutritional disorders in the elderly, thermogenesis, and fetal and neonatal nutritional support. Category I credit
is offered.
For further information please contact:
American Society for Clinical Nutrition
9650 Rockville Pike
Bethesda, MD 20814
Telephone:
JCE & M • 1986
Vol 62 • No 1
(301) 530-7110
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 July 2015. at 09:35 For personal use only. No other uses without permission. . All rights reserved.