r

Vol. 56, No.6, December 1991

FERTILITY AND STERILITY

Printed on acid·free paper in U.S.A.

Copyright © 1991 The American Fertility Society

Endocrinology of gonadotropin-releasing hormone induced cycles in hypothalamic amenorrhea: the role of the pulse dose Didi D. M. Braat, M.D.* Joop Schoemaker, M.D., Ph.D. Division of Reproductive Endocrinology and Fertility, Department of Obstetrics and Gynecology, Academic Hospital Vrije Universiteit, Amsterdam, The Netherlands

Objective: To find the treatment regimen giving a maximum chance of ovulation and a minimal chance of multiple follicular development in pulsatile gonadotropin-releasing hormone (GnRH) therapy in patients with hypothalamic amenorrhea. Design: We prospectively studied the endocrinology of cycles induced with 5, 10, and 20 p,g GnRH pulse doses, randomly assigned per patient, comparing this with the endocrinology of spontaneous menstrual cycles. Setting: All patients were treated at the Academic Hospital of the Vrije Universiteit, Division of Reproductive Endocrinology and Fertility. Patients: Fifteen patients with hypothalamic amenorrhea were treated for one to three cycles; 14 normally cycling volunteers were studied for one cycle. Main Outcome Measure: Number of ovulations per pulse dose; luteinizing hormone, folliclestimulating hormone, total urinary estrogens (Es), and pregnanediol were measured per cycle day and per stimulation day. Results: The endocrinology of all ovulatory cycles remained within the normal range. First treatment cycles showed significantly higher ovulation rates compared with subsequent cycles. Significantly more anovulation was observed in cycles with 5-p,g pulse doses. Luteal Es were significantly higher in induction cycles compared with controls. Conclusions: The optimum treatment regimen should be to start induction with 5 p,g/pulse in the first cycle and to raise the dose to 10 p,g/pulse in subsequent cycles, regardless of the outcome of the first cycle. After ovulation, the pulse interval should be changed to 240 minutes. Fertil Steril 1991;56:1054-9

Pulsatile gonadotropin-releasing hormone (GnRH) administration is effective in restoring ovulation in patients with hypothalamic amenorrhea. 1,2 Since 1978, different treatment regimens have been used. Pulse intervals used varied from 60 to 120 minutes without seriously influencing the results. 3 Compared with the subcutaneous route, intravenous (IV) administration leads to higher serum concentrations, as well as to earlier ovulations. 4 ,5 We have reported high conception rates in IV -treated patients, especially if no other infertility factors were demonstrable. 6 However, in that study, as in most retrospective studies, different pulse doses have been used. Received March 1, 1991; revised and accepted August 7,1991.

* Reprint requests: Didi D. M. Braat, M.D., Department of Obstetrics and Gynecology, Academic Hospital Vrije Universiteit, P.O. Box 7057,1007 MB Amsterdam, The Netherlands.

1054

Braat and Schoemaker

Retrospective evaluation of GnRH-induced pregnancies revealed a higher multiple pregnancy rate if higher pulse doses were used.? Santoro et al. B performed a prospective study with different pulse doses in which 5 to 6 p,g/pulse already appeared to be a supraphysiological dose. Leyendecker et al.,9 in a retrospective study, reported normal gonadotropin levels if 10 to 15 p,g/pulse was used. The aim of the present study was to find the treatment regimen giving a maximum chance of ovulation and at the same time a minimal risk of multiple follicular development. MATERIALS AND METHODS Patients

Fifteen women, 23 to 38 years of age, with amenorrhea of suprapituitary origin, who desired preg-

Endocrinology in GnRH treatment

Fertility and Sterility

nancy, participated in the study. The study was approved by the Committee on Ethics of Research on Human Subjects of the hospital. Informed consent to participate in the study was granted by all 15 women. All patients were hypoestrogenic (total urinary estrogen (E) excretion < 120 nmol/24 h). Median luteinizing hormone (LH) and follicle-stimulating hormone (FSH) concentrations were 4.0 (range: 1.6 to 7.0) and 5.2 (range: 1.5 to 7.9) IU/L, respectively. All patients were normoprolactinemic, normoandrogenemic and euthyroid. Four patients had primary amenorrhea, 1 of whom was diagnosed as having Kallman's syndrome; the other 3 had idiopathic hypogonadotropic hypogonadism. Eleven patients had had secondary amenorrhea for at least 6 months, 4 of whom had anorexia nervosa or weight loss-related amenorrhea. One patient developed amenorrhea after partial pituitary resection for a pituitary adenoma with suprasellar extension. This patient still showed sustained LH and FSH release under stimulation. The origin of amenorrhea in the other 6 patients remained unclear. Six women were of normal weight (Quetelet's Index [QI] = 20.4 to 22.8 kg/m2), 5 were underweight (QI = 16.7 to 19.1 kg/m2), 3 were overweight (QI = 28.9 to 29.8 kg/m2), and 1 was obese (QI = 36.6 kg/m 2). All were resistant to clomiphene citrate in a dose of 150 mg for 7 days or 200 mg for 5 days. Fourteen normal women, ages 20 to 33 (mean 27.4 ± 1.0), volunteered as controls. All had normal weight (QI 20 to 25 kg/m2). Their cycle length varied from 26 to 30 days. Study Protocol

Patients were randomized to receive either 5, 10, or 20 ~g/pulse during the first cycle. Patients were treated for one to three cycles, never receiving the same pulse dose in any two or three cycles. Therapy was discontinued in case of pregnancy. All patients were treated IV with pulse intervals of 120 minutes. Transabdominal ultrasonography (3.5-MHz transducer, Ultramark 4 ultrasound [US] system; Squibb Medical Systems, Seattle, W A) (first part of the study) or transvaginal US (5-MHz transducer, Combison 320; Kretz Technik, Zipf, Austria) (last part of the study) was performed three times a week or daily once the largest follicle had reached a diameter of 14 mm. The diameter was calculated as being the average of two perpendicular measurements in one plane. Plasma samples were collected at 9.00 A.M. 15 minutes before the next pulse. This was carried out three times a week or daily once sonography was Vol. 56, No.6, December 1991

performed daily as well. Twenty-four-hour urine samples were collected daily for determination of total Es and pregnanediol, the latter only from day 10 of treatment onward. The 14 volunteers were scanned by US daily from day one of their cycle until ovulation had taken place. Daily plasma samples were taken, and daily 24-hour urine samples were collected for the whole cycle. Hormone Measurements

Plasma LH and FSH of the first eight patients were measured in duplicate employing a double antibody solid-phase radioimmunoassay (RIA; Amerlex, Amersham, United Kingdom) with a sensitivity of 1.0 IU/L (standards: MRC 68/40 for LH and MRC 68/39 for FSH). Samples of the last seven patients and all volunteers were measured in duplicate employing an immunoradiometric assay (IRMA, LH- and FSH-magnetic immunoradiometric assay [Maia clone]; Serono Diagnostics Ltd, Woking, United Kingdom). In our clinic, a sound correlation was found between the RIA and the IRMA: for LH, IRMA = 0.563 X RIA + 0.047 (r = 0.9409, n = 113); for FSH, IRMA = 0.954 X RIA - 0.094 (r = 0.8757, n = 123). To enable comparison of data, all RIA measurements were converted into IRMA measurements. Total E excretion was determined by an automated rapid fluorometric assay based on the Ittrich reaction.lO The intra-assay coefficient of variation (CV) was 15%. Total urinary excretion of pregnanediol was determined by gas chromatography after mild acid hydrolysis, as described by Metcalf.ll A value > 3 ~mol/24 h is considered substantial evidence for ovulation. The intra-assay CV was 12%. Statistical Analysis

Luteinizing hormone, FSH, total Es, and pregnanediol values were compared by Kruskal-Wallis one-way nonparametric ANOVA and the MannWhitney U -test, if appropriate. Duration of different treatment phases was compared using Wilcoxon's rank sum test for unpaired observations. Numbers of follicles and number of anovulatory cycles were compared by X2 test or by Kendall's tau B test. RESULTS

Fifteen patients were treated for 38 cycles. Five patients started on 5 ~g/pulse, of which one was anovulatory,5 on 10 ~g/pulse, all ovulatory, and 5 on 20 ~g/pulse, all ovulatory also. In 13 cycles, 5-~g

Braat and Schoemaker Endocrinology in GnRH treatment

1055

pulses were used (group I), of which 6 cycles were ovulatory and 2 were conceptional as well (Table 1). In 13 cycles, 10-llg pulses were used (group II), of which 10 cycles were ovulatory and 2 were conceptional as well. In 12 cycles, 20-llg pulses were used (group III), of which 10 cycles were ovulatory and 1 was conceptional as well. No multiple pregnancies were observed. There was a significantly higher anovulation rate in group I (53.8%) compared with groups II (23.1%) and III (16.7%) (P < 0.05). Of 15 first cycles, only 1 was anovulatory (7%) compared with 11 in 23 subsequent cycles (48%) (P < 0.05). All 14 volunteers (group IV) demonstrated an ovulatory cycle.

10

I1I

6

" 2

0 2

Results per Day of Treatment During the 1st 10 Days

8

5

7

8

10

9

B

*

10

Table 1 Number of Ovulatory, Anovulatory, and Conceptual Cycles per Pulse Dose

8 6

" 2

0

-

"

3

2

OVULATORY

5

8

7

--I- ANOVULATORY

e

8

10

--- CONTROlS

360 Total EBtro,eDII

c

* * * * * * *

300 250 200 150 100 50

2

3

I!!im OVULATORY

5

"TREATMENT DAY 8

I!!im ANOVULATORY

7

8

e

10

IIIII!I CONTROlS

Figure 1 Mean LH (A), FSH (B), and total Es (C) of all ovulatory, anovulatory, and control cycles during the 1st 10 days of treatment. *, P < 0.05.

ference was found. A significantly higher level of total Es was observed in groups I, II, and III compared with group IV on days 3, 4, 5, and 9 (P < 0.05). Results per Day of Ovulatory Cycles

Ovulation

Anovulation

Pregnancies

6 10 10

7 3 2

2 2 1

Braat and Schoemaker

"

3

FSH

12

Figure 1 shows the mean FSH, LH, and total E levels of all ovulatory, all anovulatory, and all control cycles during the 1st 10 days of treatment. On day 2, a significant difference was found (P < 0.05) between FSH values in ovulatory cycles and controls and between anovulatory and ovulatory cycles (P < 0.05). Significant differences of LH levels were found on days 6, 8, and 10 (P < 0.05). On these days, LH values of anovulatory cycles were significantly lower than those of ovulatory cycles (day 6: P = 0.032; day 8: P = 0.002; and day 10: P = 0.004) and those of controls on day 8 (P = 0.006) and day 10 (P = 0.005». Comparing total E per day of treatment revealed significantly higher E levels on days 3 to 9 in ovulatory cycles than in controls as well as in anovulatory cycles (P < 0.05). Comparing ovulatory cycles with different pulse doses (Fig. 2), a significant difference in FSH levels was found only on day 2 (P < 0.05) between groups I and IV and between groups III and IV. The increase on day 2 under the influence of the beginning of therapy was not clearly dose-dependent, the peak after 5-llg pulses being higher than that induced with 10-llg pulses, whereas lower than with 20-llg pulses. Comparing the mean LH values of groups I, II, III, and IV per day of treatment, no significant dif-

1056

*

8

14

Pulse dose

A

*

*

The day of the LH peak was considered to be cycle day o. Ovulatory cycles were synchronized accordingly. Mean FSH, LH, and total E levels per cycle day were calculated per group. Because FSH and LH were measured only three times a week, these

Endocrinology in GnRH treatment

Fertility and Sterility

111

LII

1. 12 10

8 8

• 2 0 1

2

20 FSH

*



3

8

5

7

8

10

9

15

10

5

O+---~--~~~~~~---+---+---r--~ 1

2

-

5

GROUP I

'-00 Total E.troa8D11

350

8

-t- GROUP D

*

7

....... GROUP DI

8

9

10

-e- GROUP IV

c

*

*

*

300

250

200 150

100 50

o 5

TREATMENT DAY

liliiii GROUP

I

liliiii GROUP U

liliiii GROUP DI

_GROUP IV

Figure 2 Mean LH (A), FSH (B), and total Es (C) of ovulatory cycles of group I (5 ~g/pulse), group II (10 ~g/pulse), group III (20 ~g/pulse), and group IV (controls) during the 1st 10 days of treatment. *, P < 0.05.

values were pooled, and calculations were made every 3 days except for the day of the LH peak. In each 3-day bracket, each patient contributed only once; values were randomly chosen if more values of one patient were available in the bracket. No significant differences were observed in FSH levels between the four groups (Fig. 3). Only on days -7/-9 and 0, significant differences were found in LH levels (P < 0.05). On day 0, mean LH in all Vol. 56, No.6, December 1991

patient groups was lower than in group IV, whereas on day -7/-9 mean LH in group III was higher than in group IV. For every ovulatory cycle the maximum E excretion was calculated. Comparing the mean maximal E excretion per group, no differences were found between the different groups (group I: 610.0 ± 106.8; group II: 477.7 ± 167.0; group III: 510.0 ± 208.1; and group IV: 453.8 ± 40.3 nmoljd). No significant differences were found between the four groups in E measurements per cycle day in the follicular phase. In the luteal phase, however, total Es in the patient groups were significantly higher than in group IV on days 3, 5, 9, 11, and 12 (P < 0.05). For every ovulatory nonconceptional cycle the maximum pregnanediol excretion was calculated. Comparing the maximum pregnanediol excretion per group, no differences were found between the different groups (group I: 12.9 ± 3.1; group II: 11.2 ± 1.9; group III: 13.3 ± 4.5; and group IV: 11.1 ± 0.7 JLmol/d). Comparing pregnanediol excretion per cycle day of nonconceptional cycles, a significant difference was only observed on day 11 (P < 0.05). The pregnanedioljE excretion ratio was calculated on cycle days 3, 6, and 9. No differences were observed between groups I, II, III, and IV. A significant difference of this ratio, however, was found when conceptional cycles were compared with nonconceptional cycles: higher on cycle day 6 in conceptional cycles (P < 0.05). Of 26 ovulatory induction cycles, 5 exhibited two large (~14 mm) follicles (19.2%) on cycle day 0; the other 21 had only one large follicle on this day. Of those cycles with two follicles, 1 (of 5) came from group I, 1 (of 11) from group II, and 3 (of 10) from group III. In group IV, 1 of 13 had two large follicles (7.7%); in one patient, although being ovulatory according to pregnanediol measurements, no follicles could be detected at all. The distribution of extra follicles over all four groups did not differ significantly; this was the case in comparing induction cycles to controls. On cycle day 0, the median follicular diameter was 20.5 (range: 15 to 28) mm in the patient groups and 22.0 (range: 15 to 28) mm in controls (not significant). First Versus Subsequent Cycles

Comparing FSH of first cycles with FSH of later cycles only on treatment day 2, a significant difference was found (P < 0.05) (not shown). Comparing LH of first cycles with LH of later cycles, also a significant difference was found on day 2 only (P < 0.05). No differences were found in total E levels. Comparing FSH, LH, and total E levels of first

Braat and Schoemaker Endocrinology in GnRH treatment

1057

UI

30

*

*

cycles with those of subsequent cycles per cycle day, no significant differences were found (not shown). Ignoring the pulse dose, the comparison of only first and later treatment cycles showed that 3 of 14 first ovulatory cycles had two large follicles (21.4%), whereas 2 of 12 (16.7%) later cycles had two large follicles (not significant).

A

25 20 15 10

DISCUSSION -,.11-10-1 -I -7 _I -0 -t -I -I -1 0

10

1



I

4

I



7





10 11 11

B

FSH

8 8

2

-

GROUP I

-+- GROUP

D

---- GROUP

m

--GROUP IV

c

800 Total B.UDle...

'* *

* **

500 ~oo

300 200 100

o

14 P....Il.Il.dlol

D

'*

CYCLE DAY

!Il!III GROUP 1

!Il!III GROUP

D

11IIIIIIII GROUP m

_

GROUP IV

Figure 3 Mean LH (A), FSH (B), total Es (C), and pregnanediol (D) of ovulatory cycles of group I (5 ltg/pulse), group II (10 ltg/pulse), group III (20 ltg/pulse), and group IV (controls) per cycle day. *, P < 0.05.

1058

This study illustrates that on the average the endocrinology in IV GnRH-induced cycles is not abnormal using either 5-, 10-, or 20-~g pulse doses. Although several investigators3 ,8 advocate use of 5 ~g or less per pulse, we observed a significantly higher percentage (53.8%) of anovulation in the 5-~g group. Low LH levels during the 1st 10 days of anovulatory cycles suggest understimulation, opening the possibility that anovulation might have been prevented by a higher pulse dose. In patients with hypothalamic amenorrhea, decreased LH secretion is a frequent observation. 12 This undoubtedly is caused by low endogenous GnRH stimulation. This low stimulation may result, however, in some priming of the pituitary, causing an exaggerated FSH response after initiation of exogenous GnRH therapy. Reame et al. 13 found higher FSH levels in patients with hypothalamic amenorrhea compared with late luteal FSH levels in normal cycles. This FSH will lead to some degree of follicular maturation, which might be responsible, together with the exaggerated FSH response on the 2nd day of treatment, for the higher ovulation rate in first compared with subsequent treatment cycles. Although not significantly different in the present study, this may also lead to more multiple ovulations and thus more multiple pregnancies in first cycles compared with subsequent treatment cycles, as we have reported earlier. 7 In this study, we could not detect the higher peak E values in higher pulse doses as did Santoro et al. 8 and Leyendecker et al. 9 However, we did find significantly higher E levels during the luteal phase compared with normal cycles, whereas the pregnanediol excretion remained the same. After ovulation, the corpus luteum (CL) secretes progesterone (P) and estradiol (E 2). Alilo and Hansel14 showed in cows that P secretion of luteinized granulosa cells after LH receptors have been occupied is LH pulse-independent, whereas E2 and P secretion of luteinized theca cells remain dependent on LH pulsatility. Although normal luteal phase pulsatility is approximately once every 4 hours, we did not change

Braat and Schoemaker Endocrinology in GnRH treatment

Fertility and Sterility

the frequency postovulatory. This might explain the divergence in total E- and pregnanediol excretion. On the contrary, no differences in steroid production of the CL was observed by Schoemaker et al.,15 who compared eight cycles in which the pulse interval was changed from 120 to 240 minutes with four cycles in which a pulse interval of 90 minutes was maintained. Because we know now from the in vitro fertilization literature 16 that too high E levels and/ or too low P /E2 ratios can have deleterious effects on implantation, we suggest decreasing the pulse frequency after ovulation has been confirmed. Santoro et al. B and Miller et al. 3 used pulse intervals of 90 minutes. This means a higher total daily GnRH dose using the same pulse dose as we did. Although this might account for better ovulation rates at lower pulse doses, this is in contrast with earlier observations by Schoemaker et al. 15 In this study, cycles with 120-minute pulse intervals were compared with cycles with 90-minute pulse intervals (on 20-~gpulse doses). No differences were observed in LH, total Es, and pregnanediol levels, whereas significantly higher FSH levels were seen in 120minute pulse interval cycles between days -3 and +7. Before cycle day -3, no difference was observed. In this study, no comparisons were made per stimulation day. In conclusion, the endocrinology of all our ovulatory cycles, as well as the production of large follicles, remained within the normal range, so pulse doses of 5, 10, or 20 ~g all seem to be reasonable options. Because first treatment cycles had a higher ovulation rate, we recommend starting treatment with a pulse dose of 5 ~g. If in this cycle no conception occurs, then raise the dose to 10 ~g/pulse to overcome the high anovulation rate in subsequent cycles at the 5-~g pulse dose. After ovulation, the pulse frequency should be reduced to once every 4 hours for economic reasons and perhaps to optimize implantation conditions. Acknowledgments. The authors thank Ms. Danka Kucharska for her statistical support and Gerda Berkhout, R.N., and Ted De Vries Robles-Korsen, R.N., for their technical assistance. REFERENCES 1. Leyendecker G, Wildt L, Hansmann M: Pregnancies following chronic intermittent (pulsatile) administration of GnRH by means of a portable pump ('Zyklomat')-a new approach to the treatment of infertility in hypothalamic amenorrhea. J Clin Endocrinol Metab 51:1214, 1980 2. Schoemaker J, Simons AHM, van Osnabrugge GJC, Lugtenburg C, Van Kessel H: Pregnancy after prolonged pulsatile

Vol. 56, No.6, December 1991

administration of luteinizing hormone-releasing hormone in • a patient with clomiphene-resistant secondary amenorrhea. J Clin Endocrinol Metab 52:882, 1981 3. Miller DS, Reid RR, Cetel NS, Rebar RR, Yen SSC: Pulsatile administration of low dose gonadotropin-releasing hormone: ovulation and pregnancy in women with hypothalamic amenorrhea. JAMA 250:2937, 1983 4. Reid RL, Leopold G R, Yen SSC: Induction of ovulation and pregnancy with pulsatile luteinizing hormone releasing factor: dosage and mode of delivery. Fertil Steril 36:553, 1981 5. Handelsman DJ, Boylan LM: Pharmacodynamics of gonadotropin-releasing hormone (GnRH). II. Pattern of GnRH delivery alters pituitary luteinizing hormone secretion in women. J Clin Endocrinol Metab 67:175, 1988 6. Braat DDM, Schoemaker R, Schoemaker J: Life table analysis of fecundity in intravenously gonadotropin-releasing hormone-treated patients with normogonadotropic and hypogonadotropic amenorrhea. Fertil Steril 55:266, 1991 7. Braat DDM, Ayalon D, Blunt SM, Bogchelman D, Coelingh Bennink HJT, Handelsman DJ, Heineman MJ, Lappiihn RE, Lorijn RHW, Rolland R, Willemsen WMP, Schoemaker J: Pregnancy outcome in luteinizing hormone-releasing hormone induced cycles: a multicentre study. Gynecol Endocrinol 3:35,1989 8. Santoro N, Wierman ME, Filicori M, Waldstreicher J, Crowley WF, Jr: Intravenous administration of pulsatile gonadotropin-releasing hormone in hypothalamic amenorrhea: effects of dosage. J Clin Endocrinol Metab 62:109, 1986 9. Leyendecker G, Struve T, Plotz EJ: Induction of ovulation with chronic intermittent (pulsatile) administration of LHRH in women with hypothalamic and hyperprolactinemic amenorrhea. Arch GynecoI229:177, 1980 10. Van Alphen JM, Van Kessel H: Geautomatiseerde, snelle bepaling van oestrogenen in urine van nietzwangeren. Ned Tijdschr Geneeskd 121:1255, 1977 11. Metcalf MG: Hydrolysis and decomposition of urinary pregnanediol in acid. Steroids 21:193, 1973 12. Schoemaker J, Schoemaker H, Berkhout G, Schoemaker HC: Pulsatile luteinizing hormone secretion in hypothalamic amenorrhea. In Advances in Gynecological Endocrinology, Vol. 1, Proceedings of the 1st Congress of the International Society of Gynecological Endocrinology, Edited by AR Genazzani, F Petraglia, A Volpe, F Facchinetti. Carnforth, the Parthenon Publishing Group, 1989, p 43 13. Reame NE, Sauder SE, Case GD, Kelch RP, Marshall JC: Pulsatile gonadotropin secretion in women with hypothalamic amenorrhea: evidence that reduced frequency of gonadotropin-releasing hormone secretion is the mechanism of persistent anovulation. J Clin Endocrinol Metab 61:851,1985 14. Alilo HW, Hansel W: Origin of different cell types in the bovine corpus luteum as characterized by specific monoclonal antibodies. BioI Reprod 30:1015, 1984 15. Schoemaker J, Korsen TJM, Van Kessel H: Reflections on pulse dose and pulse interval. In Pulsatile GnRH 1985, Proceedings of the 3rd Ferring Symposium, Edited by HJT Coelingh Bennink, AA Dogterom, RE Lappohn, R Rolland, J Schoemaker. Haarlem, Ferring, 1986, p 131 16. Gidley-Baird AA, O'Neill C, Sinosich MJ, Porter RN, Pike IL, Saunders DM: Failure of implantation in human in vitro fertilization and embryo transfer patients: the effects of altered progesterone/estrogen ratios in human and mice. Fertil Steril 45:69, 1986

Braat and Schoemaker Endocrinology in GnRH treatment

1059

Endocrinology of gonadotropin-releasing hormone induced cycles in hypothalamic amenorrhea: the role of the pulse dose.

To find the treatment regimen giving a maximum chance of ovulation and a minimal chance of multiple follicular development in pulsatile gonadotropin-r...
856KB Sizes 0 Downloads 0 Views