FERTILITY AND STERILITY Copyright 0 1988 The American Fertility Society Printed in U.S.A. The importance of human chorionic gonadotropin support of the corpus luteum during human gonadotropin therapy in women with anovulatory infertility Ioannis E. Messinis, M.D. *t Torbjorn Bergh, M.D., Ph.D.** Leif Wide, M.D., Ph.D. Uppsala University, Akademiska Sjukhuset, Uppsala, Sweden One hundred ten women with anovulatory infertility (World Health Organization [] group In = 50, group II n = 60) were given 341 treatment courses with human menopausal gonadotropin (hmg) and human chorionic gonadotropin (hcg). Additional hcg was given as single or repeated injections during the luteal phase in 205 ovulatory cycles. In group I, the incidence of luteal phase defects was lower and the pregnancy rate higher in cycles with extra hcg administration during the luteal phase than in cycles with no extra hcg. In group II, there was no such difference after supplemental hcg. The abortion rate was the same after cycles with or without extra hcg administration. It is suggested that during ovulation induction with hmg/hcg in anovulatory women with no evidence of endogenous estrogen activity, the luteal phase should be supplemented with additional hcg. Fertil Steril 50:31, 1988 Treatment with human menopausal gonadotropin (hmg) and human chorionic gonadotropin (hcg) in women with anovulatory infertility usually results in a pregnancy rate which is much lower than the ovulation rate. 1 This difference can probably to some part be explained by defect hormonal effects during the luteal phase of the stimulated cycle. Several studies have shown that the luteal phase of cycles stimulated with hmg/hcg can be shorter compared with spontaneous cycles. 2-4 This phenomenon has been associated with high preovulatory serum estrogen levels 3 4 and inadequate dose ofmidcycle hcg. 4 5 To support corpus luteum, mul- Received July 7, 1987; revised and accepted March 28, 1988. * Department of Obstetrics and Gynecology. t Present address: Department of Obstetrics and Gynecology, University of Ioannina, loannina, Greece. :j: Reprint requests: Torbjorn Bergh, M.D., Ph.D., Department of Obstetrics and Gynecology, Uppsala University, Akademiska Sjukhuset, S-75185, Uppsala, Sweden. Department of Clinical Chemistry. tiple doses of hcg have been given empirically during the luteal phase of hmg/hcg-stimulated cycles. 2 5 6 However, apart from some case reports,2 5 no data exist to show whether supplementary HCG can increase the pregnancy rate in hmg/hcg cycles. Here we present the results of hmg/hcg treatment of women with anovulation and infertility. The effects of additional HCG, given during the luteal phase, on the incidence of luteal phase defects and pregnancy rate are evaluated. Patients MATERIALS AND METHODS One hundred ten women, aged 23 to 38 years (mean 29 years), with anovulatory infertility were given a total of 341 courses with hmg/hcg during a 5-year period. Ovulation was regarded to occur in 328 (96%) of these cycles. Follicle-stimulating hormone (FSH) was low or normal in all the women. Serum prolactin (PRL) levels were normal in all Messinis et al. hcg support during hmg/hcg therapy 31
but five patients who had PRL levels just above the normal upper limit (16-25 #tg/1, normal < 15 #tg/1). Three patients had substitution therapy with thyroxine and cortisol because of panhypopituitarism after previous hypophysectomy. Fifty of the women were amenorrheic with no evidence of endogenous estrogen activity (negative progesterone [P] withdrawal test, i.e., no uterine bleeding after an intramuscular injection of 50 mg of natural Pin oil). They were classified as World Health Organization () group I patients. 7 The remaining 60 patients had amenorrhea or oligomenorrhea with evidence of endogenous estrogen production (positive P withdrawal test) and belonged to group II. 7 Mean FSH, luteinizing hormone (LH), and estradiol (E 2) pretreatment values were significantly lower in group I than in group II patients (Table 1). Although most (85%) of the group II women had some symptoms or signs of the polycystic ovary syndrome (i.e., hirsutism or overweight or both), only 24 patients had all the main criteria for the diagnosis of this syndrome, i.e., androgen excess, increased LH:FSH ratio of at least 3:1, oligoamenorrhea, and bilateral ovarian enlargement. None of the group II patients had disorders of adrenal and thyroid function. All of them had previously failed to ovulate after treatment with clomiphene citrate (CC) in doses of 150 to 200 mg daily for 5 days. Gonadotropin Therapy An individually adjusted treatment schedule was used, which has been previously described. 1 8 HMG (Humegon ampules, each containing 75 IU FSH and 75 IU LH, Organon, Ltd, The Netherlands) was given to the women by intramuscular injections at an initial dose of 150 IU daily. In patients Table 1 Pretreatment Serum Gonadotropin (~g/1) and E 2 (pmol/1) Levels in Anovulatory Women Group I Group II (n = 50) (n = 60) p FSH, mean 0.70 1.02 <0.001 below normal 18% 3.4% <0.05 LH,mean 0.34 0.90 <0.001 below normal 78% 22% <0.001 above normal 2% 10% NS FSH andlh below normal 14% 1.6% <0.05 E2, mean 100 157 <0.001 (27 pg/ml) (43 pg/ml) 32 Messinis et al. hcg support during hmg/hcg therapy with evidence of endogenous estrogen production, the treatment started on day 2 or 3 after a spontaneous or P-induced menstrual bleeding. During treatment, each patient had frequent pelvic examinations with recording of cervical mucus quality. Daily blood samples were taken from each woman for serum E2 measurement. The response to treatment was considered significant when serum E2 levels showed a continuous rise so that 5 days after the onset of treatment they were at least twice as high as the pretreatment levels. If no estrogen response occurred, the hmg dose was increased stepwise by one ampule every fifth day until a significant rise in the serum E 2 level was obtained. The treatment continued until serum E 2 concentration reached a level of about 2000 pmol/1 (545 pg/ml). Ovulation was then induced by a single intramuscular injection of hcg (Pregnyl, Organon Ltd., The Netherlands). The dosage of hcg was arbitrarily chosen to be 6000 IU during the first 117 cycles and 9000 IU during the next 224 cycles. From the day of HCG injection to the day of menstruation or confirmation of pregnancy, blood ~amples were taken every other day for E 2 and P measurement. Additional hcg during the luteal phase was given during 205 cycles on different days after the ovulatory hcg. This was based on results in rats showing in vitro refractoriness of the preovulatory follicle to hcg for several days after the in vivo exposure to this hormone. 9 This has more recently been confirmed in human thecal cells. 10 Two different regimens of extra hcg were arbitrarily chosen. During the first 175 cycles a single injection of 3000 IU (28 cycles) or 4500 IU (147 cycles) was given 7 days after the ovulatory injection. The routine was then changed to be three repeated injections of 1500 IU every other day from the fourth day after the ovulatory injection ofhcg (30 cycles). No extra hcg was given during 123 cycles. Criteria of Ovulation and Corpus Luteum Function A nonconceptual cycle was considered as ovulatory with normal corpus luteum function if serum P values reached a level of at least 32 nmol/1 (10 ng/ml) 11 and the luteal phase was longer than 10 days calculated from the day of the presumed ovulation to the day of the next menstruation or 12 days from the hcg injection. If serum P was between 6 and 32 nmol/1 (2 and 10 ng/ml) or the duration of the luteal phase less than 10 days or both, the cycle was considered as ovulatory with luteal phase defect. 8 A cycle was considered as an- Fertility and Sterility
ovulatory when serum P was <6 nmol/1 ( <2 ng/ml). Hormone Assays Immunoreactive E2 in serum was measured by a radioimmunologic technique in which charcoal was used for separation.12 For progesterone measurement, a similar radioimmunologic technique was used.13 The normal midluteal P level in our laboratory is >32 nmol/1 (> 10 ng/ml). FSH and LH determination in serum was done by a radioimmunosorbent method with indirectly coupled antibodies.14 Statistical Methods The statistical analysis of the results was done by using the chi-square test and the unpaired Student's test. The hormone values were transformed into logarithms in the statistical calculations. RESULTS The pregnancy outcome of the hmg/hcg therapy in the two groups ( I and II) is shown in Table 2. All but one of the women (99%) ovulated at least once during the treatment, and 62 of them conceived (overall pregnancy rate 57%). The multiple rate pregnancy rate was 23% (Table 2), with no difference between the two groups. In the group I, the serum E2 values on the day of hcg administration in cycles that resulted in single pregnancies were similar (mean 2236 pmol/1, range 734-4800 = 609 pg/ml, range 200-1308) to those in the cycles with multiple pregnancies Table 2 Five Year Results of Human Gonadotropin (HMG/HCG) Therapy Patients Treatment cycles Ovulatory cycles Patients with ovulation Patients with pregnancy Pregnancies Term Multiple Abortions Patients who took home at least one baby Group I n 50 167 163 (98%) 50 (100%) 33 (66%) 39 29 9 (31%) 10 (26%) 28 (56%) 4 Group II n 60 174 165 (95%) 59 (98%) 29 (48%) 33 24 3 (13%) 9 (27%) 22 (37%) Total n 110 341 328 (96%) 109 (99%) 62 (57%) 72 53 12 (23%) 19 (26%) 50 (45%) "x 2 = 4.11, P < 0.05 (difference from corresponding value of group II). (mean 2468 pmol/1, range 1370-5908 = 672 pg/ml, range 373-1610). Also in the group II patients, the mean E2 values did not differ (2717 pmol/1, range 640-5900 = 740 pg/ml, range 174-1608 versus 3077 pmol/1, range 1140-11300 = 838 pg/ml, range 311-3079) between single and multiple pregnancies. Luteal phase defects occurred in 7.3% of the cycles. There was no difference in the incidence of luteal phase defects between cycles in which ovulation was induced with 6000 IU hcg and those with 9000 IU hcg in either group I (10.6% versus 7.0%, respectively) or group II women (7.9% versus 5.8%, respectively). Also, the number of anovulatory cycles did not differ between the administration of 6000 IU and 9000 IU of hcg in both groups of women (2.9% versus 2.5% and 6.1% versus 5.5%, respectively). In the group I women, the incidence of luteal phase defects was higher in cycles when no extra hcg was given during the postovulatory phase than in cycles in which extra hcg was given (P < 0.001) (Table 3). Also, the pregnancy rate w~s higher when supplemental hcg injections were given during the luteal phase in group I (P < 0.01) (Table 3). In the group II, there were no such differences. The incidence of luteal phase defects and pregnancy rate did not differ between cycles in which one hcg injection was given and cycles in which repeated hcg injections were given. Therefore, these cycles have been combined in Table 3. An example of the importance of extra hcg during the luteal phase is given in Figure 1, which shows three different treatment cycles (one with luteal phase defects, one with apparently normal luteal phase, and one resulting in pregnancy) from a hypophysectomized woman. The abortion rate did not differ between cycles with supplemental luteal phase hcg injections and cycles with no extra hcg either in group I (26% versus 20%) or in group II (28% versus 25%). No difference in the total number of hmg ampules and days of hmg administration was found between group I and group II or among conceptual, nonconceptual, ovulatory, and anovulatory cycles. Neither did these parameters differ among term pregnancies, abortions, and single or multiple pregnancies in both groups. The pretreatment values of FSH, LH, and E2 did not differ between patients with conceptual and nonconceptual cycles or with single and multiple pregnancies. There was no correlation between the pretreatment FSH and LH values and the total Messinis et al. hcg support during hmg/hcg therapy 33
Table 3 The Effect of Supplemental hcg During the Luteal Phase of hmg/hcg-stimulated Cycles on the Incidence of Luteal Phase Defects and Pregnancy Rate Incidence of luteal phase defects Pregnancy rate Group I Group II Group I Group II No extra hcg Extra hcg 11/53 (21 %) 2/110 (1.8%)" 6170 (9%) 5/95 (5.2%) 5/53 (9%) 34/110 (31.9%)b 13/70 (18%) 20/95 (21%) a X 2 = 17.4, p < 0.001. b X 2 = 9.06, P < 0.001 (difference from no extra hcg). number of hmg ampules or days of hmg administration. There was a negative correlation between the pretreatment E2 values and the total number of hmg ampules (group I r = -0.21, P < 0.01; group II r = -0.17, P < 0.05). DISCUSSION Our results show for the first time that supplementary hcg given during the luteal phase of hmg/hcg cycles supports the luteal function and increases the pregnancy rate significantly in patients with no evidence of endogenous estrogen activity. Since endogenous gonadotropin secretion in these patients is significantly reduced, the results support previous data in women and monkeys showing the importance of pituitary LH secretion after ovulation for maintenance of corpus luteum function. 15 16 Ovarian stimulation with human gonadotropins for multiple follicular development in normal women results in a significant suppression of endogenous LH secretion during the luteal phase, as compared with spontaneous cyclesp It is therefore possible that in our study the extra hcg given during the luteal phase compensated for the -9 31 years old DAY RELATIVE TO HCG INJECTION Figure 1 E 2 and progesterone levels in serum during three different hmg/hcg treatment cycles of a 31-year-old amenorrheic woman of group I who had undergone transfrontal hypophysectomy due to a pituitary tumor. In cycle (a) luteal phase defect occurred. In cycle (b) extra HCG was given during the postovulatory phase and a normal luteal phase was obtained. The patient conceived in cycle (c) when repeated supplemental hcg injections were given during the luteal phase. inadequate secretion of pituitary LH in the women with low endogenous LH (and E 2) secretion. Previous results have suggested that multiple injections of hcg during the postovulatory phase may reduce the abortion rate in hmg/hcg-stimulated cycles. 5 However, this assumption is not supported by our findings, which show similar rates of abortion between cycles with and cycles without extra hcg injections. Apart from the high preovulatory estrogen levels that may interfere with the implantation process, 18 19 no clear explanations for the high abortion rate during treatment with human gonadotropins exist. The recent development of in vitro fertilization has provided an opportunity for cytogenetic study of human pre-embryos and may give insight into the mechanisms that control early embryo development after use of exogenous hormones for ovarian stimulation. The inability of the hmg/hcg treatment to support the luteal phase and increase the pregnancy rate in group II patients suggests that ovulation failure in these women is not due to inadequate stimulus of endogenous gonadotropins. Recent data have shown that patients with polycystic ovary syndrome, i.e., 85% of the group II patients in this study, can be treated more successfully with other modalities of therapy. 20 21 In conclusion, our study shows that supplemental hcg administration during the luteal phase reduces the incidence of luteal phase defects and increases the pregnancy rate significantly in patients with no evidence of endogenous estrogen secretion ( group I). It also shows that treatment with hmg/hcg is more effective in that group than in the group of women with low endogenous estrogen production ( group II). REFERENCES 1. Bergquist C, Nillius SJ, Wide L: Human gonadotropin therapy. I. Serum estradiol and progesterone patterns during conceptual cycles. Fertil Steril 39:761, 1983 34 Messinis et al. hcg support during hmg/hcg therapy Fertility and Sterility
2. Townsend SL, Brown JB, Johnstone JW, Adey FD, Evans JH, Taft HP: Induction of ovulation. J Obstet Gynaecol Br Commw 73:529, 1966 3. Johansson EDB, Gemzell C: The relation between plasma progesterone and total urinary estrogens following induction of ovulation in women. Acta Endocrinol (Copenh) 62:89, 1969 4. Olson JL, Rebar RW, Schreiber JR, Vaitukaitis JL: Shortened luteal phase after ovulation induction with human menopausal gonadotropin and human chorionic gonadotropin. Fertil Steril 39:284, 1983 5. Brown JB, Evans JH, Adey FD, Taft HP, Townsend L: Factors involved in the induction of fertile ovulation with human gonadotropins. J Obstet Gynaecol Br Commw 76:289, 1969 6. Wang CF, Gemzell C: The use of human gonadotropins for the induction of ovulation in women with polycystic ovarian disease. Fertil Steril 33:4 79, 1980 7. Scientific Group. Agents stimulating gonadal function in the human. World Health Organization Technical Report Series No 154. Geneva, World Health Organization, 1973, p 7 8. Bergquist C, Nillius SJ, Wide L: Human gonadotropin therapy. II. Serum estradiol and progesterone patterns during nonconceptual cycles. Fertil Steril 39:766, 1983 9. Lamprecht SA, Zor U, Tsafriri A, Lindner HL: Action of prostaglandin E 2 and of luteinizing hormone on ovarian adenylate cyclase, protein kinase and ornithine decarboxylase activity during postnatal development and maturity in the rat. J Endocrinol 57:217, 1973 10. Dennefors BL, Hamberger L, Nilsson L: Influence of human chorionic gonadotropin in vivo on steroid formation and gonadotropin responsiveness of isolated human preovulatory follicular cells. Fertil Steril 39:56, 1983 11. Johansson E, Larsson-Cohn U, Nygren K-G, Jeppsson S, Victor A, Weiner E: Inhibition of ovulation by gestagens. Ups J Med Sci [Suppl] 22:16, 1978 12. Hotchkiss J, Atkinsson LE, Knobil E: Time course of serum estrogen and luteinizing hormone (LH) concentrations during the menstrual cycle of the rhesus monkey. Endocrinology 89:177, 1971 13. Thorneycroft IH, Stone SC: Radioimmunoassay of serum progesterone in women receiving oral contraceptive steroids. Contraception 5:129, 1972 14. Wide L, Nillius SJ, Gemzell C, Roos P: Radioimmunosorbent assay for folliclestimulating hormone and luteinizing hormone in serum and urine from men and women. Acta Endocrinol (Copenh) [Suppl]174:1, 1973 15. Vande Wiele RL, Bogumil J, Dyrenfurth I, Ferin M, Jewelewicz R, Warren M, Rizkallah T, Mikhail G: Mechanisms regulating the menstrual cycle in women. Recent Prog Horm Res 26:63, 1970 16. Hutchison JS, Zeleznik AJ: The Rhesus monkey corpus luteum is dependent on pituitary gonadotropin secretion throughout the luteal phase of the menstrual cycle. Endocrinology 115:1780, 1984 17. Messinis IE, Templeton AA: Disparate effects of endogenous and exogenous oestradiol on luteal phase function. J Reprod Fertil 79:549, 1987 18. Morris JM, van Wage nan G: Interception: The use of postovulatory estrogens to prevent implantation. Am J Obstet Gynecol 115:101, 1973 19. Oelsner G, Serr DM, Mashiach S, Blankstein J, Snyder M, Lunenfeld B: The study of induction of ovulation with menotropins: Analysis of results of 1897 treatment cycles. Fertil Steril 30:538, 1978 20. Fleming R, Haxton MJ, Hamilton MPR, McGune GS, Black WP, Macnaughton MC, Couts JRT: Successful treatment of infertile women with oligomenorrhoea using a combination of an LHRH agonist and exogenous gonadotropins. Br J Obstet Gynaecol 92:369, 1985 21. Franks S, Adams J, Mason H, Polson D: Ovulatory disorders in women with polycystic ovary syndrome. Clin Obstet Gynaecol 12:605, 1985 Messinis et al. hcg support during hmg/hcg therapy 35