Short-term luteinizing hormone-releasing hormone agonist treatment: prospective trial of a novel ovarian stimulation regimen for in vitro fertilization

FERTILITY AND STERILITY

Vol. 52, No.2, August 1989

Copyright <> 1989 The American Fertility Society

Printed in U.S.A.

Short-term luteinizing hormone-releasing hormone agonist treatment: prospective trial of a novel ovarian stimulation regimen for in vitro fertilization Michael C. Macnamee, Ph.D.*t Colin M. Howles, Ph.D.:!: Robert G. Edwards, Ph.D.§

Patrick J. Taylor, M.B., B.Ch.*11 KayT. Elder, M.B., Ch.B.*

Bourn Hall Clinic, Bourn, Physiological Laboratory, Cambridge, and Serono Laboratories (UK) Ltd., Welyn Garden City, Herts, United Kingdom

Over a period of 4 months, 262 infertile couples participated in a prospective pseudorandom trial of a novel short-term luteinizing hormone-releasing hormone/human menopausal gonadotropin (LH -RH/hMG) treatment; the short-Buserelin-gonadotropin (Hoechst, Hounslow, United Kingdom) regimen. Patients treated with the short-Buserelin-gonadotropin regimen had a significantly higher likelihood of achieving pregnancy than patients treated with the standard clomiphene citrate (CC)/hMG regimen (respectively, 35.5% and 18% per treatment cycle). A significantly higher number of eggs were collected after short-Buserelin-gonadotropin treatment than CC/hMG, but the proportion of patients having a given number of embryos replaced was similar in the two groups. The short-Buserelin-gonadotropin-treated patients were distinguished from the CC/ hMG-treated group by significantly lower levels of LH in the late follicular phase and a lower plasma level of estradiol. A detrimental relationship between elevated endogenous LH secretion and failure of implantation has been established. The nature of the shortBuserelin-gonadotropin regimen provokes high levels of endogenous gonadotropin secretion in the early follicular phase and induces a suppression of gonadotropin secretion in the late follicular phase. This may be the physiologic basis of the greater implantation rate after short-Buserelin-gonadotropin treatment than is seen with conventional CC/ hMG treatment. Fertil Steril52:264, 1989

We have been assessing the clinical value of novel therapies with luteinizing hormone-releasing hormone (LH-RH) for ovarian stimulation in patients attending Bourn Hall for various forms of assisted conception. The present work was stimulated by observations in patients excreting high

Received December 28, 1988; revised and accepted April 4, 1989. * Bourn Hall Clinic, Bourn. t Reprint requests: M. C. Macnamee, Ph.D., Deputy Scientific Director, Bourn Hall, Bourn, Cambridge CB3 7TR, United Kingdom. :j: Serono Laboratories (UK) Ltd. § Physiological Laboratory, Cambridge. 1/ Present address: Department of Obstetrics, Gynecology and Reproductive Sciences, The University of Manitoba, Winnipeg, Manitoba, Canada.

264

Macnamee et a1.

levels oftonic urinary LH when superovulated with clomiphene citrate (CC) and human menopausal gonadotropins (hMG). In a study of over 200 patients, high urinary LH in the late follicular phase was associated with failure of implantation and early pregnancy loss, while low levels of urinary LH were associated with the establishment of ongoing pregnancy. 1,2 An inverse relationship between basal LH secretion in stimulated cycles and embryo fitness is implied by the greater survival rate of extra embryos in culture from patients with low tonic urinary LH levels. 3 The detrimental link between LH and oocyte quality is further strengthened by an analysis of 62 patients stimulated with CCjhMG or CC alone, in which elevated plasma levels of LH predicted reduced fertilization rates associated with darkening of the cumulus mass. 4

Improved IVF success with LH-RH treatment

Fertility and Sterility

The cause of high tonic LH secretion is unknown, but may reflect either an underlying hypothalamo-pituitary defect or solely the influence of CC during stimulation cycles. 5 During ovarian stimulation using CC/hMG, both plasma and urinary levels of LH decline toward ovulation. I.6A The decrease is the result of the time-related differential sensitivity of the pituitary gonadotropes to the antiestrogenic effects of CC 5 coupled with the negative feedback effects of rising, supraphysiologic levels of estradiol (E 2) from the developing follicles. This pattern of LH secretion contrasts markedly with the steady levels maintained throughout the follicular phase of the natural cycle. The long-term use ofLH-RH agonists in ovarian stimulation has been shown to suppress endogenous gonadotropin secretion7 and decrease the production of ovarian androgens. 8 These treatments have been shown to be highly effective for in vitro fertilization (IVF).9,IO However, complete pituitary desensitization may take up to 15 days,l1 and larger doses of exogenous gonadotropins are required for longer than is normal to achieve an adequate follicular development. We have previously described the use of a novel short-term application of an LH-RH agonist (short-Buserelin-gonadotropin regimen, Hoechst, Hounslow, United Kingdom), which substantially improved the performance of 'poor responders' undergoing ovarian stimulation before IVF.2 This study extends the application of this form of stimulation as a prospective trial comparing the performance of patients treated with CC and hMG or with the short-Buserelin-gonadotropin regimen. MATERIALS AND METHODS Treatment Groups

From November 1, 1987 to March 1, 1988, all patients wishing to attend the clinic for their first, second, or third IVF attempt, except those diagnosed as having polycystic ovarian syndrome, were allocated to one of two stimulation regimens. Patients unable to attend the clinic on day 2 of their menstrual cycle or whose traveling time to the clinic was >2 hours were sent prescriptions for the standard CC/hMG regimen; CC (100 mg day 2 to day 6: Serophene, Serono, Welwyn Garden City, United Kingdom) and human menopausal gonadotropins (2 ampules per day from day 5 onwards; Pergonal, Serono). All other patients presented at the clinic on day 2 of their menstrual cycles. A Vol. 52, No.2, August 1989

blood sample taken from these patients was assayed for progesterone (P), and their ovaries were ultrasonographically scanned for the presence of cysts or regressing corpora lutea. If the plasma P level was <1.0 ng/mr I and no large cysts were present on scan, then they were treated with the short-Buserelin-gonadotropin regimen; Buserelin (500 J.tg subcutaneously [daily on day 2, day 3, and day 4; Suprefact, Hoechst, Hounslow, United Kingdom]) and human menopausal gonadotropins (3 ampules per day from day 3 onward). During the course of both stimulation regimens, the dose of hMG was titrated to the individual patients' response, i.e., up to a maximum of 6 ampules or down to a minimum of 1 ampule per day. The same criteria for dose adjustment were applied to both groups, based on the number of follicles> 12 mm in diameter, the optimum being taken as between 5 and 10. The mean day of cycle on which hCG was administered was not different between the two treatments: 10.6 ± 0.16 days (range, day 8 to day 15) for short-Buserelin-gonadotropin-treated and 10.6 ± 0.24 days (range, day 8 to day 14) for CC/ hMG. The mean number of ampules of hMG administered per cycle was 32 ± 4 for short-Buserelin-gonadotropin and 21 ± 3 for CC/hMG. All protocols for this study were ratified by the Bourn Hall Ethical Committee and the in-house scientific review body. Patient Data

Other than those criteria stated above, no attempt was made to select patients entering either treatment group. In all, 138 women were treated with CC/hMG and 124 with short-Buserelin-gonadotropin. The mean age and age range were similar for both groups, as was the distribution of ascribed cause of infertility (Table 1). Oligozoospermia of varying degrees accounted for the majority of male factor infertility in both groups. Tubal damage, ovulatory, and hormonal problems accounted for 56% ofthe infertility in the study population. Combined male and female factor infertility accounted for 34 (13%) of the couples entering the study and 50 (20%) couples were classified as idiopathic/unknown cause of infertility. Endocrine and Ultrasound Monitoring

From day 8 of short-Buserelin-gonadotropinstimulated cycles or day 9 of CC/hMG-stimulated cycles, the number and size of developing follicles was assessed every second day by either transvagi-

Macnamee et al.

Improved IVF success with LH·RH treatment

265

Table 1 Ages, Causes of Infertility, and Progress of Women Receiving Either Standard CC/hMG or Short-Buserelin-Therapy Treatment

Total no. entering study Age (mean ± SEa) Age range (yrs) Cause of infertility (n [%]) Male factor(s) only Female factor(s) only Male and female factor(s) Idiopathic/unknown Discharged preoocyte recovery (poor response) Failed fertilization No. of patients with embryos replaced Mean no. ± SE of embryos replaced/patient No. of patients (%) with: one embryo transferred two embryos transferred three embryos transferred four embryos transferred

CC/hMG

Short-BuserelinhMGtherapy

138 35.7 ± 0.6 28-43

124 35.3 ± 0.4 25-42

14 80 17 27

(10%) (58%) (12%) (20%)

15 (12%) 69 (56%) 17 (14%) 23 (18%)

12 (8%) 5 (4%)

11 (9%) 2 (2%)

121 (88%)

111 (89%)

3.32 ± 0.4

3.16 ± 0.4

15 (12%) 14 (12%) 20 (16%) 72 (60%)

15 (13.5%) 15 (13.5%) 18 (16%) 63 (57%)

Statistical Analyses Unless otherwise stated, all statistical testing was done using the unpaired Student's t-test. The technique of IVF embryo culture and embryo transfer (ET) were those described by Purdy.12 During the period of the study, a maximum of four embryos were replaced per patient 2 days after oocyte collection. Remaining embryos of sufficient quality were cryopreserved. Luteal phase support was given to all patients on the basis of the number of follicles aspirated. If fewer than ten follicles were aspirated, then hCG (5,000 IU) was administered on day 3 and day 8 after oocyte recovery. If more than ten follicles were aspirated, patients received daily injections of 50 mg of P (Gestone, Payne & Burn, London, United Kingdom) from the day of embryo replacement to lessen the risk of ovarian hyperstimulation following hCG administration. Clinical pregnancy was established by rising plasma titers of,3-hCG from 15 days after, and the presence of a fetal heartbeat 35 days after, induction of ovulation.

a SE, standard error.

RESULTS nal or transabdominal ultrasound scanning. Daily blood samples were taken, and patients supplied a minimum of eight timed and measured urine samples each day. Plasma samples were assayed on the same day as collection for E 2, P, and LH. Details of assay methodologies are reported elsewhere. 7 Interassay and intra-assay coefficients of variation were, respectively, 8.0% and 5.5% for E2 7.3% and 3.2% for P; 10.1% and 4.5% for urinary, and 8.2% and 3.0% for plasma LH. Levels of LH in urine were assessed in a minimum of three out of the eight daily samples. Criteria for Ovulation Induction Consistently rising plasma levels of E2 associated with a leading follicle of 17 to 18 mm (mean of three estimates of diameter in different planes) were the criteria for induction of ovulation by injection of human chorionic gonadotropin (hCG) (5,000 IU, Profasi, Serono). If an endogenous LH surge was detected, it was supported by 5,000 IU of hCG and oocyte recovery timed to its onset. Laparoscopic or, more commonly, transvaginal oocyte recovery was performed approximately 34.5 hours after hCG injection or 24 to 26 hours after the onset in urine of the endogenous LH surge. 266

Macnamee et aI.

Table 1 describes patient data and the progress of patients entering the study. Similar numbers were discharged from the two groups because of poor ovarian response. Failure of fertilization occurred in a total of seven cases. The mean number of embryos replaced in CCjhMG and short-Buserelin-gonadotropin treatment groups was the same, and similar numbers of patients had one, two, three, or four embryos replaced. Twenty-five (19%) patients in the CCjhMGtreated group had an endogenous LH surge. There was no evidence of an endogenous LH surge in any patients in the short-Buserelin-gonadotropintreated group. For the purpose of endocrine comparison, data from patients who surged is excluded from the analysis presented in Table 2. Both urinary and plasma levels of LH were significantly lower in the short-Buserelin-gonadotropin group than in the CCjhMG group (P < 0.05). Plasma levels of E2 were significantly lower (P < 0.001) in the short-Buserelin-gonadotropin group from their admission to the clinic on day 8 until hCG was administered. Plasma levels of P were indistinguishable between the two treatments. The mean number of oocytes recovered per patient was higher in the short-Buserelin-gonadotropin group than in the CCjhMG group (P < 0.05).

Improved IVF success with LH-RH treatment

Fertility and Sterility

Table 4 Incidence of Pregnancy Loss and Ectopic Pregnancy by Treatment

Table 2 Endocrine Comparison Between, and Oocyte Recovery, Fertilization, and Freezing Rates by Treatment Groups

CC/hMG No. of patients Mean urinary LH (IU/l/hr) Mean plasma LH (lU jl) Final E2 (pg/ml- 1) Final progesterone (ng/ml- 1) Mean no. of oocytes recovered per patient (±SEa) Fertilization/cleavage rate per patient No. of supernumerary embryos frozen per patient

Short-BuserelinhMGtherapy

101

111

0.22 ± 0.06 0.10 ± 0.05 7.9 ± 0.3 3.10 ± 0.3 923 ± 70 1352 ± 70

P<0.05 P
CC/hMG Total pregnancies Abortions < 11 weeks Ectopics Ongoing Ongoing pregnancies/ treatment cycle started

Short-BuserelinhMGtherapy

25 5 (20%) 1 (4%) 19 (76%)

44 8 (18%) 2 (5%) 34 (77%)

19/138 (14%)

34/124 (27.4%)

NS b

<0.3

<0.3

7.6 ± 0.6

9.5 ± 1.0

68% ± 11

64% ± 13

NS

1.21 ± 0.7

1.88 ± 0.6

NS

P<0.05

for each group is presented in Table 4. There was no difference between the two groups in terms of abortions and ectopic pregnancies, but a much higher ongoing pregnancy rate was achieved in the short-Buserelin -gonadotropin group.

a SE, standard error; NS, not significant.

DISCUSSION

The fertilization and cleavage rates for the two groups were the same. Also, the numbers of embryos remaining of suitable quality for cryopreservation after replacement was similar in both groups, although the trend was for increased numbers in the short-Buserelin-gonadotropin group (Table 2). The outcome of embryo replacement in terms of pregnancy is given in Table 3. When taken together, significantly more patients with three or four embryos replaced became pregnant in the short-Buserelin-gonadotropin group than in the CC/hMG group (pregnancy rates of 48.1 % [39/91] and 25% [23/92], respectively, P < 0.01, chisquare). The pregnancy rate per replacement for CC/hMG-treated patients in this study (20%) is typical of our previous results with this treatment and compares favorably with the national figures. 13 Although none of the pregnancies have yet delivered, an analysis of the abortion and ectopic rates Table 3 Pregnancies Following Embryo Replacement by Treatment Group

CC/hMG

Short-BuserelinhMGtherapy

(pregnancy/total)

(pregnancy/total)

1/15 1/14 6/20 17/72

2/15 3/15 9/18 30/63

25/121

44/111

No. of embryos replaced 1 2 3 4

Total

Vol. 52, No.2, August 1989

The short-Buserelin-gonadotropin regimen was originally developed for use in "poor responders".2 It is our belief that these patients suffered from prolonged hypersecretion of endogenous LH because of the sustained action of CC when treated with conventional therapy.5 While high levels of endogenous gonadotropins in the early follicular phase promote follicle recruitment, prolonged hypersecretion of LH caused by CC can lead to a degree of luteinization in the preovulatory follicle. 4 At worst, this leads to failure of fertilization in vitro, whereas milder overexposure of the developing follicle to LH can lead to failure of implantation ascribed to decreased embryo fitness. 1 This may be due to premature initiation of terminal growth changes in the follicle cells and oocyte, similar to those seen at midcycle following the LH surge,14 but not sufficient to provoke follicular rupture. 15 The rationale of the short-Buserelin-gonadotropin regimen is to provoke enhanced gonadotropin secretion by direct pituitary gland stimulation by the LH-RH agonist, and thereby to stimulate follicular recruitment during the critical period in the early stages ofthe menstrual cycle, but not sustaining the hypersecretion into the critical stages of preovulatory development. Although no microscopic differences in oocyte quality were recorded (i.e., subjective cumulus and viscous scoring being similar in short-Buserelin-gonadotropin and CC/ hMG groups), the suppression of LH levels in plasma and urine in the late follicular phase of short - Buserelin - gonadotropin - treated patients (Table 2) suggests that the degree of exposure of

Macnamee et al.

Improved IVF success with LH-RH treatment

267

the follicle to LH was reduced, so that oocytes of optimal maturity were collected from these patients. Presumably, optimal oocyte maturity provides fitter embryos' following in vitro fertilization with a better chance of implantation (Table 2). The high pregnancy rate (48.1%) in the short-Buserelin-gonadotropin group after replacement of three or four embryos (Table 3) substantiates this theory. Rutherford et al. 1O have also reported that Buserelin, given in a long protocol, also increased the chances of pregnancy after IVF compared with conventional therapy. However, the risk of multiple pregnancies also may be increased. Currently we do not have data to support or refute this. The suppression of LH secretion in the late follicular phase in short-Buserelin-gonadotropintreated patients may be attributable to several factors. Acute LH -RH agonist administration results in binding to and internalization of LH -RH receptors on the pituitary gonadotropes, and, after internalization, pituitary sensitivity to LH -RH can take up to 9 days to recover, even though the half-life of Buserelin in plasma is only 80 minutes. 16 In shortBuserelin-gonadotropin-treated patients, the receptor complement of the pituitary might be diminished for sufficiently long to appear as a mild suppression of endogenous LH output up to the induction of ovulation. Secondly, the massive release of gonadotropins produced by acute LH-RH administration may not only deplete pituitary stores of LH,16 but also may inhibit LH production and release by a short-loop feedback mechanismY-19 Thirdly, the lower levels of circulating E2 in shortBuserelin-gonadotropin-treated patients (Table 3) may be sufficient to initiate negative feedback control of LH secretion but insufficient to stimulate the positive feedback release of LH, which may occur in CC/hMG patients who have higher levels of E 2. The differential production of E2 between the two groups may be due to a direct ovarian effect resulting from the inhibition of androgen production, even though the LH -RH agonist was given for such a brief period.8 ,2o,21 It is likely that the observed mild suppression of LH secretion in the late follicular phase of short-Buserelin-gonadotropin treatment is due to a combination of these factors. Aside from the matter of 'fitter' embryos, the question also arises of direct and/or indirect effects of acute LH -RH therapy on the endometrium. There is no doubt that, in certain circumstances, the LH -RH agonist can be luteolytic. 22-24 This seems unlikely to occur after short-Buserelin-gonadotropin treatment, where a 1.5-mg dose of LH268

Macnamee et al.

RH is given 12 days before the expected time of implantation. Nonetheless, in case luteolysis was a possibility, both groups of patients received tried and tested forms of luteal support. Any influence of the LH-RH agonist in short-Buserelin-gonadotropin treatment on endometrial receptivity is more likely to be the protection of the endometrium from high levels of E2 than due to the direct ovarian effect. High levels ofE2 in the late follicular phase have been associated with failure of implantation and early pregnancy loss in patients treated with CC/hMG. A more appropriate steroid millieu therefore may pertain after short-Buserelin-gonadotropin treatment, which encourages secretory changes in the endometrium, facilitating implantation. 6 As well as the detrimental influence of LH hypersecretion, the inferior implantation rate seen in the CC/hMG-treated group may be due in part to the direct effect of CC causing endometrial retardation. 25 In conclusion, the use of the short-Buserelin-gonadotropin regimen significantly increases the chance of pregnancy after embryo replacement when compared with treatment with CC/hMG. The physiologic basis for this incremental success is attributable to the mild suppression of LH in the late follicular phase. The short-Buserelin-gonadotropin regimen also gives a substantial improvement over long-term down-regulation, as the length of the treatment cycle is shorter and the amount of exogenous gonadotropins required for adequate follicular development is less (56 ampules versus 32 ampules).3 The short-Buserelin-gonadotropin regimen has the advantage of suppressing the LH surge while permitting the full participation of endogenous gonadotropins in sustaining multiple follicle growth.

Acknowledgments. We would like to thank our colleagues at Bourn Hall Clinic for their time and effort, particularly Miss Caroline Walshe, Mr. Adam Burnley, and Miss Caroline Dawkin for her excellent secretarial work.

REFERENCES 1. Howles CM, Macnamee MC, Edwards RG, Goswamy R,

Steptoe PC: Effect of high tonic levels of luteinizing hormone on outcome of in-vitro fertilization. Lancet 2:521, 1986 2. Howles CM, Macnamee MC, Edwards RG: Short-term use of an LHRH agonist to treat poor responders entering an in vitro fertilization (IVF) programme. Hum Reprod 2:17, 1987 3. Macnamee MC, Edwards RG: Unpublished data

Improved IVF success with LH-RH treatment

Fertility and Sterility

4. Stanger JD, Yovich JL: Reduced in-vitro fertilization ofhuman oocytes from patients with raised basal luteinizing hormone levels during the follicular phase. Br J Obstet Gynaecol 2:85, 1985 _ 5. Adashi EY: Clomiphene citate: mechanism(s) and site(s) of action-a hypothesis revisited. Fertil Steril42:331, 1984 6. Howles CM, Macnamee MC, Edwards RG: Follicular development and early luteal function of conception and nonconceptional cycles after human in-vitro fertilization. Hum Reprod2:17,1987 7. Fleming R, Adam AH, Barlow DH, Black DH, Black WP, Macnaughton MC, Coutts JR: A new systematic treatment for infertile women with abnormal hormone profiles. Br J Obstet GynaecoI89:80, 1982 8. Ory ST: Clinical uses of luteinizing hormone-releasing hormone. Fertil SteriI39:577, 1983 9. Macnamee MC, Howles CM, Edwards RG: Pregnancies after IVF when high tonic LH is reduced by long-term treatment with GnRH agonists. Hum Reprod 2:569, 1987 10. Rutherford AJ, Suback-Sharpe RJ, Dawson KJ, Margora RA, Franks S, Winston RML: Improvement of in vitro fertilization after treatment with Buserelin, an agonist of luteinizing hormone releasing hormone. Br Med J 296:1,765, 1988 11. Fleming R, Yates RWS, Haxton MJ, Coutts JRT, Hamilton MPR, Conaghan C: Ovulation induction using the combination of Buserelin and exogenous gonadotrophins in women with functional pituitaries. Br J Clin Pract 41(Suppl. 48):34, 1987 12. Purdy JM: Methods for fertilization and embryo culture in vitro. In Human Conception In Vitro, Edited by RG Edwards, JM Purdy. London, Academic Press, 1982, p 135 13. Voluntary Licensing Authority for Human In Vitro Fertilization and Embryology. Third report. London, Voluntary Licensing Authority, 1988. (Available from the Secretariat, 20 Park Crescent, London WIN 4AL) 14. Winer-Sorgen S, Brown J, Tsutomu 0, Gale JA, Campeau JD, Marrs RP, Dizerega GS: Oocyte maturation inhibitor activity in human follicular fluid: quantitative determination in unstimulated and clomiphene citrate- and human menopausal gonadotrophin-stimulated ovarian cycles. J In Vitro Fert Embryo Transfer 3:218, 1986

Vol. 52, No.2, August 1989

15. Fowler RE, Fox NL, Edwards RG, Walters DE, Steptoe PC: Steroidogenesis by cultured granulosa cells aspirated from human follicles using pregnenolone and androgens as precursors. J Endocrinol 77:171, 1978 16. Sandow J, Fraser HM, Seidel H, Krauss B, Jerabek-Sandow G, van Rechenberg W: Buserelin: pharmacokinetics, metabolism and mechanics of action. Br J Clin Pract 41(Suppl. 48):6, 1987 17. Clayton RN, Catt KJ: Gonadotrophin-releasing hormone receptors: characterization, physiological regulation and relationship to reproductive function. Endocrinol Rev 2: 186,1981 18. Clayton RN: Gonadotrophin-releasing hormone modulation of its own pituitary receptors: evidence for a biphasic regulation. Endocrinology 111:152,1982 19. Clayton RN, Channabasavaiah K, Stewart JM, Catt KJ: Hypothalamic regulation of pituitary gonadotrophin releasing hormone receptors: effect of hypothalamic lesions and a gonadotrophin-releasing hormone antagonist. Endocrinology 110:1,108, 1982 20. Casper RF, Erickson GF, Yen SSC: Studies on the effect of gonadotrophin-releasing hormone and its agonist on human luteal steroidogenesis in vitro. Fertil Steril42:39, 1984 21. Popkin R, Bramley TA, Currie A, Shaw RW, Baird DT, Fraser HM: Specific binding of luteinizing hormone-releasing hormone to human luteal tissue. Biochem Biophys Res Comm 114:750, 1983 22. Casper RF, Yen SSC: Introduction of luteolysis in the human with a long-acting analog of luteinizing-hormone-releasing factor. Science 205:408, 1979 23. Clayton RN, Harwood JP, Catt KJ: Gonadotrophin -releasing hormone analogue binds to luteal cells and inhibits progesterone production. Nature 282:90, 1979 24. Schriock ED, Monroe SE, Martin MR, Henzel MR, Jaffe RB: Effect on corpus luteum function of luteal phase administration of potent gonadotrophin-releasing hormone analog (Nafarelin). Fertil Steril43:844, 1985 25. Sterzik K, Dallenbach C, Schneider V, Sasse V, DalenbachHellweg G: In vitro fertilization: the degree of endometrial insufficiency varies with the type of ovarian stimulation. Fertil Steril50:457, 1988

Macnamee et al.

Improved IVF success with LH-RH treatment

269