MONITORING OF OVULATION INDUCTION

Transcription

1 FERTIUTY AND 8TERIUTY Copyright 1978 The American Fertility Society Vol. 3, No. 6, December 1978 Printed in U.SA. MONITORING OF OVULATION INDUCTION CHUNG H. WU, M.D. Department of Obstetrics and Gynecology, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania 1914 Anovulation is one of the major causes of infertility. Prior to the recent availability of potent ovulation-inducing agents, the anovulatory infertile female had little chance of conception. In 1958 Gemzell et al., 1 using human pituitary gonadotropin, induced ovulation in hypogonadotropism; in 1961 Greenblatt et al.,2 using clomiphene citrate (Clomid; Merrell-National Laboratories, Cincinnati, Ohio), successfully induced ovulation in patients with secondary amenorrhea or oligomenorrhea. Since then, these two ovulationinducing agents have become widely used and are helpful in induction of ovulation and pregnancy in such cases. In spite of the relative simplicity of clomiphene therapy, it is still not recommended for anovulatory patients unless interested in pregnancy. Furthermore, menopausal gonadotropin therapy is dangerous and frequently associated with serious complications; its application is thus highly restricted. It has been generally recommended that induction of ovulation should be initiated only in patients desiring pregnancy at the time of therapy. Recently, other ovulation-inducing agents have also been used, including bromoergocriptine and gonadotropin-releasing hormone (GnRH). Since induction of ovulation is not without danger, careful observation and monitoring of this medication should be considered. The purposes of monitoring of ovulation induction are 2-fold: first, to prevent unwanted complication and risk; second, to improve the efficacy of therapy. In this article, biochemical and biophysical changes during the normal ovulatory cycle are reviewed, and blood hormone patterns in the gonadotropininduced ovulatory cycle and clomiphene-induced cycle are presented for comparison. Methods of ovulation induction available at the present time Received August 31, are summarized. Finally, monitoring of ovulation induction in gonadotropin therapy is described, and a possible monitoring technique in clomiphene therapy is suggested. HORMONE PA'ITERNS IN THE NORMAL OVULATORY CYCLE In the normal ovulatory cycle a distinct cyclic pattern of blood hormone concentrations has been observed. 3 4 The delicate balance in the hypothalamic-pituitary-ovarian axis and its mutual feedback mechanism controls normal follicular maturation and the eventual ovulatory phenomenon. Postovulatory corpus luteum function is also maintained by this delicate balance. Significant hormones related to the ovulatory cycle include pituitary follicle-stimulating hormone (FSH), luteinizing hormone (LH), and the ovarian steroid hormones-estrogens, androgens, and progestins. The mean values of plasma hormone levels from a group of women with normal ovulatory cycles are presented in Figure 1.5 In the normal cycle, a distinct LH peak can be recognized at midcycle, preceded by a slightly elevated FSH level in the early phase of the cycle. It has been suggested that elevated FSH levels in association with tonic LH levels in the early follicular phase of the cycle initiate follicular maturation. Once follicular maturation has been initiated, the follicles continue to mature autonomously, and actively secrete estradiol (E2). Estradiol then feeds back to the hypothalamic-pituitary area to suppress FSH; thus FSH gradually declines toward the end of the follicular phase. At midcycle, the E2 level in the blood reaches a significant peak level, usually above 2 pg/ml. At this level, E2 positively feeds back to the hypothalamic-pituitary area to trigger surges of LH and FSH. Following an LH surge, the preovulatory ovarian follicle responds to LH and the entire series of changes, culminating in ovulation.

2 618 wu December Q. 2 FIG. 1. Mean daily values(± standard error) of plasma LH, FSH, estrone (E 1), 17,8-estradiol (E 2), progesterone (P), androstenedione (A), and testosterone (T) during the normal ovulatory cycle (N = 8) (reproduced from Wu5 with permission). Within approximately 12 to 36 hours following the LH surge, the Graafian follicle ruptures and an ovum is expelled. Following ovulation, the Graafian follicle is luteinized and transformed into a corpus luteum. Luteal tissue then secretes large amounts of E 2 and progesterone (P). Thus during the luteal phase both E 2 and P levels continue to rise, reaching a peak level on the 7th postovulatory day. Thereafter, for unknown reasons, luteolysis is initiated; and as the corpus luteum goes through the regression phenomenon, E 2 and P levels gradually decline. Endometrial shedding occurs with a precipitous decline in E 2 and P levels. Urinary excretion of estrogen (E) has been shown to parallel the plasma E level, 6-8 while urinary pregnanediol excretion also correlates with the plasma P level during the normal menstrual cycle. 8 In addition, other protein and steroid hormones have been measured in plasma during the menstrual cycle. Plasma prolactin (PRL) has been shown to fluctuate throughout the cycle, with some evidence of elevation at midcycle, 9 1 but no consistent pattern has been observed. The growth hormone level is high during the late follicular phase and tends to decrease following ovulation for the remainder of the ovulatory cycle. 11 Thyroidstimulating hormone (TSH), on the other hand, appears to be stable at the tonic level without significant cyclic fluctuation. 11 Adrenocorticotropic hormone (ACTH) decreases slightly during the early follicular phase, rises suddenly to a peak at midcycle, precipitously decreases again during the next 2 days, and demonstrates a second but gradual elevation during the luteal phase. 11 However, these changes are neither associated with nor parallel to plasma cortisol (F) levels. Plasma F levels fluctuate less than ACTH, 11 although it is well known that ACTH and F levels in plasma show some correlation. Androgens, including androstenedione (A), testosterone (T), dihydrotestosterone (DHT), and dehydroepiandrosterone sulfate (DHA-S), have been evaluated during the menstrual cycle ; all seem to fluctuate within a certain range without a distinct pattern. It has been reported that 17 a-hydroxyprogesterone (17H-P) demonstrates a small peak at midcycle coinciding with the LH peak. There is another broad peak of 17H-P parallel with Pin the luteal phase of the cycle; the actual concentration, however, is much lower than that of P. 14 A pattern similar to that ofp is also shown by 2a-hydroxyprogesterone, although again the plasma level of this hormone is much lower than that of P. 15 Aldosterone levels and plasma renin activity have been reported in the normal menstrual cycle. 16 Both of these hormones are stable during the follicular phase; however, during the postovulatory phase they tend to spike frequently. This pattern is not consistent, nor is it seen in every ovulatory cycle. Daily DHA-S concentrations in plasma have been measured in an adrenalectomized person who was ovulatoryy DHA-S levels in plasma tend to rise continuously and reach a peak about 2 days after the LH surge, then gradually decline during the luteal phase, suggesting that the ovaries may secrete DHA-S, although in a far smaller amount than is secreted by the adrenals. In addition to protein and steroid hormones, it has been shown that plasma catecholamines such as epinephrine and norepinephrine spike at midcycle, when both E and LH levels are high. 17 It has also been suggested that the plasma GnRH concentration spikes during the ovulatory phase of the cycle. 18 On the other hand, in spite of evidence that prostaglandins (PGs) may initiate the ovulatory mechanism/ 9 the plasma concentration of PGF 2a seems stable and tonic throughout the cycle without a significant surge during the ovulatory phase. 2

3 Vol. 3, No.6 MONITORING OF OVULATION INDUCTION 619 Physical changes in the ovulatory cycle have been well investigated. The basal body temperature (BBT) usually shows a distinct, biphasic pattern with an elevation during the postovulatory phase of the cycle. This elevated temperature has been attributed to the thermogenic effect of P. However, some ovulatory cycles lack temperature elevation or a biphasic basal body temperature pattern. Thus, although the BBT can be utilized to suggest ovulation in the presence of a biphasic pattern, the reverse is not necessarily correct. Secretion of cervical mucus has been suggested to be E-dependent; during midcycle,when thee level is elevated, copious amounts of mucus are secreted from the cervix. 2 Throughout the second half of the follicular phase the volume of cervical mucus gradually increases, and its cellularity decreases. Mucus viscosity also declines during the preovulatory phase, and during the periovulatory phase the ferning pattern, spinnbarkheit, and sperm penetrability gradually increase. 21 These distinct findings correlate well with amount of E secretion. In addition, vaginal epithelium matures in response to E. During the late follicular phase the vaginal cytology scores more highly on the maturation index or karyopyknotic index, reaching its highest point on PREGNANCY ( + l DAY OF THERAPY FIG. 2. Daily plasma hormone concentration in hmg/hcginduced ovulation (reproduced from Wu24 with permission). this index during the periovulatory phase. 21 Endometrium responds toe with proliferation during the follicular phase, subsequently gradually changing to secretory tissue during the luteal phase under the influence of P. Both endometrial tissue 22 and corpus luteum 23 have been histologically dated with some accuracy during the luteal phase. In summary, during the normal ovulatory cycle, hormone production or secretion of FSH, LH, E, and progestins is distinctly patterned, while characteristic biophysical changes appear in response to the hormonal milieu. This is exemplified by changes in cervical mucus, which is accessible to external examination. HORMONE PATTERNS IN THE INDUCED OVULATORY CYCLE Induction of ovulation is simply an attempt to simulate the hormone production of the menstrual cycle by initiating follicular maturation, followed by rupture of a single follicle. It is important to evaluate the various hormone patterns during the induced ovulatory cycle for comparison with normal hormonal events. Gonadotropin-Induced Cycles Hormone levels during a typical ovulatory cycle in a patient treated with Pergonal* are presented in Figure When daily plasma hormone data from this particular cycle are compared with similar data from a normal spontaneous ovulatory cycle, it is clear that during human menopausal gonadotropin (hmg) therapy plasma FSH levels gradually increase and remain at higher levels throughout hmg treatment, whereas LH levels remain low throughout this medication period. This pattern is distinctly different from that of the normal cycle, in which FSH levels tend to be slightly elevated in the first part of the follicular phase and then decline and remain low. In the physiologic state this pattern is due to the negative feedback phenomenon exerted by E 2; however, in the induced cycle using exogenous gonadotropin, in spite of a gradual increase of E 2, there is no negative feedback mechanism controlling the FSH level. Therefore, the high level offsh during the late follicular phase may possibly stimulate multiple follicles to full maturation. Thus when human chorionic gonadotropin (hcg) is given fol- *Serono Laboratories, Inc., Braintree, Mass.; human menopausal gonadotropin preparation containing 75 IU offsh and 75 IU of LH per ampule.

4 62 wu December 1978 lowing hmg to achieve ovulation, it is possible that more than one mature follicle will ovulate, resulting in multiple ovulation and possible multiple pregnancy. The patterns of E2, estrone (E1), and P in the induced cycle are similar to those of the normal cycle, except that concentrations of both E and P are much higher than in the normal ovulatory cycle. Androgen secretion from the ovary, on the other hand, in spite of treatment with a high dose of exogenous gonadotropin, appears to be unaffected. A relatively stable tonic concentration is present prior to ovulation. However, following hcg administration, the T concentration suddenly rises to, and persists at, a high level. A similar phenomenon has also been observed in a patient with ovarian hyperstimulation syndrome.25 Therefore, hcg not only may initiate follicular rupture but also may luteinize a thecal component, stimulating greater androgen secretion than in the normal cycle. The effect of this high level of androgens following ovulation has not been fully evaluated. Clomiphene-Induced Cycles Clomiphene is a synthetic compound with antiestrogenic as well as estrogenic activity. When given to the patient, it can induce ovulation. Its mechanism of action appears to be through effects on the hypothalamic-pituitary level It has been postulated that clomiphene competes with E2 for the E receptor protein in the hypothalamic neurons These hypothalamic neurons in the gonadotropic center will perhaps recognize it as a hypoestrogenic state and initiate GnRH synthesis and release, subsequently stimulating pituitary release of gonadotropins. Recently it has been shown that clomiphene itself will sensitize the pituitary cell to respond to GnRH. 29 The purpose of this medication is to initiate gonadotropin release from the pituitary, and the final outcome is an elevated plasma gonadotropin level. Daily plasma hormone levels during one clomiphene-induced ovulatory cycle are illustrated in Figure 3.5 Clomiphene is usually given to the patient with an intact hypothalamic-pituitary-ovarian axis who shows evidence of endogenous E secretion. It is administered daily in the early part of the cycle for 5 consecutive days. As is shown in Figure 3, during the administration of clomiphene, plasma LH levels tend to rise gradually with occasional spikes to a high level. In the meantime, FSH levels, although relatively tonic in this particular cycle, FIG. 3. Daily plasma hormone concentration during clomiphene therapy (reproduced from Wu with permission). also gradually rise during clomiphene therapy.3 31 It is probable that elevated FSH in conjunction with elevated LH synergistically stimulates and initiates follicular maturation. Once follicles are stimulated, they develop autonomously and reach maturity approximately 2 weeks after starting medication. Thus plasma E2 levels gradually rise during the 2nd week of therapy, reflecting the maturation offollicles. Estradiol levels then rise gradually to a peak ( >3 pg/ml) which triggers an LH surge. This sequence is similar to the events during the normal ovulatory cycle. The Graafian follicle then goes through the ovulatory phenomenon and expels the ovum; the ruptured follicle eventually becomes a corpus luteum. The plasma P level is similar to that in the normal cycle, rising gradually and peaking about the 6th day after ovulation. The significant difference between the hormonal pattern seen during clomiphene treatment and that of a normal ovulatory cycle is in the elevated LH level during the early follicular phase. This pharmacologically induced elevated LH level has been named the "Clomid peak."3 31 It seems to stimulate thecal components of the ovary to produce excess androgens5 32; thus in the clomiphene-induced cycle the T level tends to rise significantly not only prior

5 Vol. 3, No.6 MONITORING OF OVULATION INDUCTION 621 to ovulation, but also during midcycle and even after ovulation. It has been suggested that this elevated androgen at the midportion of the cycle interferes with cervical mucus secretion, as unfavorable mucus has frequently been observed during clomiphene therapy. 32 The antiestrogenic effect of clomiphene has also been blamed for this poor mucus secretion METHODS OF OVULATION INDUCTION Gonadotropin Therapy hmg-hcg therapy is limited to specifically indicated patients because of its association with medical complications. It should be instituted only in hypophysectomized or hypogonadotropic patients who do not respond top injection with withdrawal bleeding. Some patients fail to ovulate after repeated high-dose clomiphene or clomiphene-hcg therapy and they are also considered for hmg-hcg therapy, such as the occasional patient with polycystic ovarian disease or patients with hypothalamic amenorrhea who are capable of endogenous estrogen secretion. In the latter cases careful monitoring of the patient is mandatory. hmg (Pergonal) is given by daily intramuscular injection, beginning with one or two ampules and gradually increasing the dosage while monitoring follicular maturation by either urinary or plasma E levels. Subsequently, hcg (a substitute for LH) is given intramuscularly (approximately 5, to 1, IU) at the time of full maturation of the follicle. Following an initial high dose, additional hcg is recommended in some cases on the 3rd or 6th day of the postovulatory phase to "support" luteal function. Occasionally low-dose adjunctive E medication prior to hmg therapy is recommended to "sensitize" or "prime" the hypothalamic-pituitary-ovarian axis. 35 Ovulation rates following hmg-hcg therapy are high (8% to 1%); however, the pregnancy rate remains within the range of3% to 5%, a discrepancy which is currently unexplained. Multiple ovulation and multiple pregnancy rates are also high (2% to 3%) in hmg-hcg therapy and the abortion rate following pregnancy is higher than that of routine pregnancy,3 &- 38 but the risk of congenital abnormalities is not increased However, the significant and serious complication of the hyperstimulation syndrome has been reported to be in the range of.5% to 2.% Careful monitoring of ovulation induction can reduce or prevent this syndrome. As a general rule, ifhcg is withheld in cases with overstimulated ovaries, severe hyperstimulation syndrome can be avoided, yet multiple pregnancy can hardly be eliminated, even with careful monitoring and clinical observation. Clomiphene Therapy Therapy begins with a low dose, 5 mg/day for 5 days for three cycles, followed by 1 mg/day for 5 days for another three cycles. A higher dose of clomiphene, up to 25 mg/day for 5 days, apparently does not induce serious complications. Generally, clomiphene is given following P withdrawal bleeding. Fewer examinations are required than for gonadotropin therapy, and the BBT chart and/or plasma P or endometrial biopsy in the 3rd week or therapy suffice to verify ovulation. Several adjunctive therapies have been recommended, such as (1) hcg at midcycle, to substitute for the LH surge to initiate ovulation This adjunctive regimen is recommended if the patient fails to ovulate at a higher dose of clomiphene. (2) Occasionally, supplementary hcg is given during the luteal phase to support luteal function, if a luteal phase defect has been documented in a previously induced cycle. (3) Low-dose adjunctive treatment withe from day 5 to day 14 of the cycle has been used to stimulate the cervix for mucus secretion 44 ' 46 since, as described earlier, cervical mucus secretions are frequently suppressed during clomiphene therapy. Alternatively (4), a high dose of E may be given at the immediate preovulatory phase (5 days prior to ovulation) to improve cervical mucus without interrupting the ovulatory phenomenony (5) In some patients E and progestin sequential therapy prior to clomiphene treatment may "enhance" or "sensitize" the hypothalamic-pituitary-ovarian axis, improving the response to clomiphene The latter is an empirical clinical observation for which definite endocrine documentation is not yet available. Clomiphene therapy rarely induces hyperstimulation of the ovary, so that patients are infrequently monitored. However, active monitoring of clomiphene therapy as considered later in this paper may both shorten the duration of use and improve its efficacy. Although high ovulation rates (7% to 9%) have been achieved in clomiphene therapy, the pregnancy rate is still low (3% to 5%) s-54 Although they have not been definitely identified, causes such as (1) unfavorable cervical mucus, (2)

6 622 wu "entrapped" ova, (3) luteinization without ovulation, and ( 4) induced luteal phase defect have been suggested. The complications of clomiphene therapy are mild; however, the incidence of early pregnancy wastage is higher than that in the normal population Other Ovulation-Inducing Agents Bromoergocriptine Treatment. Bromoergocriptine (BC) is a potent suppressor of PRL secretion from the pituitary gland, especially in patients with amenorrhea-galactorrhea and elevated plasma PRL levels. In addition, it effectively restores gonadotropic function and eventually reestablishes ovulation Since hyperprolactinemia has been frequently associated with pituitary adenoma, when the use of BC is considered, it is important to rule out pituitary adenoma prior to induction of ovulation. Some anovulatory patients with neither galactorrhea nor hyperprolactinemia have also been successfully treated with BC.58 Although its role in ovulation is not yet clear, PRL significantly influences ovulation, and "critical" PRL levels seem important for normal ovulatory function. BC is given in daily doses of2.5 to 1 mg continuously. An ovulatory cycle is usually established within 2 months. This medicine has been extensively utilized in European countries and Canada, and has only recently been made available in the United States. GnRH Therapy. GnRH is a hypothalamic hormone containing 1 amino acids; it has been isolated, identified, and synthesized. GnRH can be given to anovulatory patients to release gonadotropin and subsequently to induce ovulation However, it is considerably less efficacious than replacement therapy with gonadotropins, and hence is not widely accepted for clinical use. It has merit in cases which call for the exact timing of ovulation, such as in artificial insemination,62 where a large dose of GnRH is given at midcycle to release a surge of LH to coincide with insemination. hcg can also fill this particular function. Estrogen-Progestin Therapy. Estrogen-progestin cyclic therapy followed by discontinuation of steroids has reportedly initiated spontaneous recovery of hypothalamic-pituitary-ovarian function Spontaneous ovulation can occur in anovulatory patients following this simple cyclic therapy for a few months. A similar approach, but using estrogen and progestin combined medication (as in birth control pills), has been used to "sensitize" or induce "rebound" hypothalamic- December 1978 pituitary-ovarian function. Thus spontaneous ovulation has occasionally been observed, although not consistently, following discontinuation of this form of steroid therapy. Corticosteroid Therapy. In patients with amenorrhea and hirsutism, especially patients whose androgen excess has been proved to be of adrenal origin, continuous suppression therapy with corticosteroids can re-establish spontaneous ovulation Usually, however, the subsequent ovulation rate is relatively low. A combination of corticosteroids plus clomiphene has recently been reported to improve the ovulation rate of such patients. 67"69 In summary, several ovulation-inducing agents are currently available which can induce ovulation without significant medical complications during therapy; thus at this time monitoring of therapy with these particular agents seems unnecessary. MONITORING OF OVULATION INDUCTION hmg-hcg Therapy Since gonadotropin induction of ovulation is sometimes associated with significant medical complications, it is vitally important to conduct therapy according to a well-defined monitoring system. In the past, because laboratory support has not been available, clinicians have tended to monitor gonadotropin therapy by physical signs, i.e., observations of cervical mucus and vaginal cytology. Cervical mucus has seemed in the past to be the most sensitive physical parameter of E secretion during the menstrual cycle. It has been carefully evaluated, for example, by combining various measurements of cervical mucus in a composite "cervical score"7 which has then been utilized to monitor gonadotropin therapy. Some authors consider it a highly effective monitor7 71 but as we stated earlier, although cervical mucus is a good indicator of the hormonal milieu in the normal menstrual cycle, it is insufficient to pinpoint the time for hcg administration. Cervical mucus secretion becomes active during the midfollicular phase; after mucus production reaches a certain stage, further increase in plasma E concentration seems not to cause a proportional increase in cervical mucus secretion. 73 Cervical parameters of estrogen secretion from the ovary do not reflect the actual hormonal state after this point and it is easy, therefore, to be misled and either give hcg prematurely or delay hcg and continue hmg to the point of inducing hyperstimulation. A similar

7 Vol. 3, No.6 MONITORING OF OVULATION INDUCTION 623 difficulty can be encountered using vaginal cytology as a parameter. Vaginal cytology reflects thee level in the blood, but because of its variability and lack of linear correlation with plasma E at higher levels, is insufficiently sensitive for the critical monitoring required in gonadotropin therapy. For these reasons the biophysical signs such as cervical mucus secretion and vaginal cytology, although useful adjunctive parameters, are insufficient monitors if used alone. Since E secretion from the ovaries does reflect the follicular maturation process, direct monitoring of E secretion seems to be the answer. Urinary E has been utilized extensively for this purpose; plasma E parallels urinary E excretion, thus either plasma or urinary E is an adequate indicator for monitoring of gonadotropin therapy. Urinary Estrogen Monitoring. Daily urinary measurement and frequent, even daily, pelvic examination are recommended in monitoring gonadotropin therapy. The advantage of urinary E monitoring is that it reflects total ovarian E secretion better than the cervical or vaginal parameters often used; its disadvantage, however, is in requiring 24 hours for a complete urine collection, plus 1 day for laboratory determination, so that measurements are actually obtained 2 days after the fact. Despite this disadvantage, monitoring of gonadotroprin therapy with urinary E excretion can be very successful Urinary E is relatively low at the beginning of therapy (usually less than 25 JLg/24 hours), and as hmg medication is started and the doses are gradually increased, daily urinary E should rise. Ideally, urinary total E should increase in a gradual fashion, preferably reaching levels of1 to 15 JLg/24 hours within 8 to 15 days of the start of hmg therapy If this level of urinary E is reached after 1 or 2 weeks of hmg therapy, hcg is then given as described earlier. Approximately 7 to 1 days later, measurements of plasma P or urinary pregnanediol will verify ovulation. Figures 4 to 6 show various possible outcomes at this point. In successful ovulatory cycles following gonadotropin therapy, urinary E excretion gradually rises (Fig. 4). 75 In contrast, in patients receiving gonadotropin therapy who do not ovulate (Fig. 5), urinary E levels remain low 75 (except in rare cases where urinary E levels rise drastically on the 4th and 5th days of therapy; in those cases, gonadotropin would be withheld to avoid hyperstimulation). In patients who develop ovarian enlargement following gonadotropin therapy, urinary E levels (Fig. 6)1 5 rise drastically after approximately 5 to 8 days of therapy. This &;., N... "' 125 Ul z 1&.1 C) a:: Ul 1&.1 >- a:: a:: ::J...J DAYS OF HMG THERAPY FIG. 4. Total urinary estrogen excretion in ovulatory cycles during hmg therapy (reproduced from Karam et al. 75 with permission). suggests that a sudden large increase in E excretion indicates overstimulation and, ifpossible, hcg should be withheld to avoid hyperstimulation. Frequent, even daily, pelvic examination and &;., N... "' Ul z 1&.1 C) a::., 1&.1 >- a:: z a: ::J...J DAYS OF HMG THERAPY FIG. 5. Total urinary estrogen excretion in anovulatory cycles during hmg therapy (reproduced from Karam et al. 75 with permission).

8 624 wu December 1978 N ""...! Ill z "" C) a: I III "" > a: <l z a: ::;)..J <l IQ II DAYS OF HMG THERAPY FIG. 6. Total urinary estrogen excretion in cycles associated with ovarian enlargement during hmg therapy (reproduced from Karam et al.7 5 with permission). laboratory tests are recommended throughout hmg treatment in order to detect ovarian enlargement and hopefully terminate therapy in cases of overstimulation. Inconvenience of urine collection and expensive laboratory tests for urinary E have led some therapists to compromise. To minimize expense without increasing medical complication, they suggest monitoring therapy initially by observing the cervical mucus, shifting toe monitoring when the cervical mucus becomes copious and favorable. This approach cuts down the necessity for laboratory tests without compromising therapeutic outcome. However, in patients who have previously had surgery of the cervix, such as cauterization or cone biopsy, cervical mucus secretion even in the late follicular phase may not reflect the E level in the blood. Therefore, even scanty or absent mucus secretion may mislead to the point that the ovary can be overstimulated. In such patients, E monitoring is important from the very beginning of gonadotropin therapy. Plasma Estrogen Monitoring. Because of the relatively stable metabolic clearance rate ofe throughout the menstrual cycle, 76 E production rates correlate well with the plasma E concentration. Therefore, in monitoring follicular matura- tion during gonadotropin therapy, plasma E levels also reflect the patient's response. 39 " Although the E 2 concentration reflects follicular maturation, monitoring of total plasma E is generally considered adequate. Total plasma E has been reported to parallel the E 2 level, 44 and radioimmunoassay (RIA) of total E is simple compared with that of E 2 Plasma E monitoring has the advantage that measurements reflect the E status on the day the specimen is obtained, whereas the urinary E level represents that of the previous day. However, there is a practical disadvantage in plasma E monitoring: not every community hospital or commerciallaboratory is capable of providing, or willing to offer, a service to determine estrogen because it is costly and inconvenient. Although RIA of total plasma E is not a complicated procedure, the demand in clinical practice is so limited that it does not justify running daily plasma E determinations. The special demands on technician time and laboratory expense therefore make the cost of E determination prohibitive to both patient and therapist. In addition, the assay needs to be run on an emergency basis, with the clinician asking for a report on the same day as the sample is submitted, further disrupting laboratory work. Thus regular commercial laboratories generally do not offer this service. Nonetheless these daily tests are needed for good medical practice, and it is hoped that in the future the service will become available to clinicians. In our own laboratory we are relatively attuned to this type of service, and frequently accumulate patients requiring monitoring so that therapy may be concurrent and expense and time saved. Ideally, daily E determination should be run early in the morning, and it is wise to keep samples from previous days to rerun in the same assay. This step helps to eliminate interassay error, so that the actual trend of estrogen increase is more clearly recognizable. The dosage of gonadotropin required to induce ovulation varies from one person to another. Even in the same individual there is significant variation between cycles. Therefore individualization of each cycle ofpergonal therapy is necessary. A typical case of gonadotropin therapy monitored by plasma E is shown in Figure 7. Patient E. D. had hypogonadotropic hypogonadism. She received two ampules ofpergonal daily to begin treatment. For the first 3 days plasma E seemed to stay at the same level. Therefore, on the 4th day, the Pergonal dose was increased to three ampules/day. After 3 more days, a significant rise in plasma E was ob-

9 Vol. 3, No.6 MONITORING OF OVULATION INDUCTION 625 HCG Pergonal 1J =I Amp \i7.& anuinn'! Cl 3 fa q: 2 DAY OF THERAPY E.D. Pregnancy H+l FIG. 7. Plasma estrogen and progesterone levels and dosage of Pergonal-hCG during therapy in patient E. D. (reproduced from Wu 77 with permission). served. Pergonal was continued at the same dose for the next few days. On the 8th day of Pergonal therapy, the plasma E concentration reached 3 pg/ml, suggesting adequate maturation of follicle(s) Therefore after 2 more days ofpergonal therapy, hcg(lo,ooo IU) was given intramuscularly. One and two weeks later, P was elevated to 8 ng/ml and 3 ng/ml, respectively. A pregnancy test was positive, and the patient experienced a normal pregnancy and delivered twins near term. In patient C. E. (Fig. 8), Pergonal was started at two ampules/day. Since the estrogen level did not rise for the next several days, the dose ofpergonal was increased to three ampules on day 7 of therapy. Plasma E remained low and, on the 12th day, the Pergonal dose was increased to four ampules/day; following 3 days ofthis dose ofpergonal, plasma E finally rose gradually to a point above 3 pg/ml. After the E level had exceeded 3 DAY OF THERAPY Fra. 8. Plasma estrogen and progesterone levels and dosage of Pergonal-hCG during therapy in patient C. E. (reproduced from Wu 77 with permission). pg/ml for more than 2 days, hcg (1, IU) was given to initiate ovulation.. Eight and ten days later plasma P levels were elevated to 34 and 27 ng/ml, respectively, suggesting an ovulatory cycle. However, this patient failed to conceive and menstruation followed on the 15th day after hcg injection. Compromise Approach to Plasma Estrogen Monitoring. Daily office visits and plasma E determination are inconvenient and expensive, thus some experienced therapists have attempted to minimize them without drastically influencing therapy outcome. For example, some physicians obtain plasma daily for E determinations but run RIA only every 3rd or 4th day. In other words, the patient appears for Pergonal medication, a venous blood specimen, and a pelvic examination daily during the first few days, and on the 4th or 5th day E determinations are begun. On the basis of thee level over the previous 4 or 5 days, the Pergonal dosage may be adjusted or maintained. For the next 3 to 4 days, clinical observation without daily plasma E is continued, and at the end of the 3 to 4 days the RIA ofe is repeated. Thereafter, every 3 or 4 days, depending on the plasma E trend, the Pergonal dosage may or may not be increased. It may be convenient to run RIA Mondays and Fridays and on other days to draw blood and observe the patient clinically. Usually, once response is initiated, the plasma E level doubles each succeeding day, enabling exact prediction of the day when the critical E level will be attained and hcg may be given. In conjunction with clinical findings on pelvic examination, this method can avoid overstimulation. Other experienced therapists 78 have considered eliminating daily plasma E and basing their judgment on measurements of the E level every 3 or 4 days throughout Pergonal therapy. In such cases sometimes the local physician administers the Pergonal injection, obviating daily office visits. This approach is convenient for the patient, but in inexperienced hands is very risky, since in cases in which response to Pergonal is rapid, overstimulation is likely. This compromise approach is justified only ifit is economically or socially impossible for the patient to return to the office every day, and, in addition, the therapist is experienced and can predict or anticipate the patient's response to PergonaL Therefore, in the hands of the inexperienced therapist or in a patient with sensitive ovaries, this compromise approach should not be considered, and daily plasma E monitoring remains essential

10 626 wu December 1978 OF,_ 98 Ol Ol 97 e 8... ::> It Ill :I: 2 (/) LL "'. It: "'..:..: _J 11.. E /.\ OAY OF THERAPY 28 Preg. E.../ 12 -;;. <=.._j 8... CX. MUCUS: Amount, Ferning,Spinnborkheit,Cell PLAN: P-K Blood E a P R_ (P.R.N.) he,,,,,.., '''"''''"''' fl Endo.Bx ( p.r.n.) Blood Pa E 4 (/)..: 35;t Preg.Test Blood E a P HCG FIG. 9. A schematic representation of biophysical changes and plasma hormone levels during a Clomid-induced ovulatory cycle with suggested clinical studies and adjunctive therapy. Preg., pregnancy; E, estrogen; P, progesterone; Cx., cervical; P-K, postcoital test; Endo.Bx, endometrial biopsy; Rx, adjunctive therapy. Clomiphene Therapy Side effects of clomiphene therapy are less serious than those with hmg. Hence patients are frequently permitted to medicate themselves without frequent clinical observation. Clomiphene is usually given to the patient in a fixed-dose schedule for several months. Although progressively increased doses of clomiphene (up to 25 mg/day for 5 days) have been recommended in each successive therapy cycle, active supportive therapy or monitoring in this therapy is not considered. Basal body temperature is simply used to verify ovulation. In contrast, we suggest an active monitoring approach to clomiphene therapy. Hormonal and physical changes accessible to external examination are shown in Figure 9. A typical hormone pattern in a clomiphene-induced cycle, as described earlier, is summarized in the upper portion of Figure 9. Active laboratory monitoring and physical examination can be planned according to hormonal status and characteristics of the cervical mucus. On day 14, postcoital tests can be performed to evaluate cervical mucus and cervical mucus-sperm compatibility. On the same day, blood is drawn for total E (and alsop if desired). On this day, in the responsive cycle, plasma E should rise above 3 pg/ml; P levels should increase slightly, yet rarely above 2 ng/ml. The combination of these two parameters helps to identify the periovulatory phase. One week later blood E and P are again determined to verify ovulation and evaluate adequacy of steroid secretion from the corpus luteum to uncover luteal phase defect. Endometrial biopsy should also be performed on this day. A biopsy should be performed during an ovulatory cycle, preferably only once on each patient. Following biopsy, the patient is advised to return on day 35 of therapy for evaluation of menstrual flow, which usually occurs between days 28 and 35 of the cycle. If the patient fails to menstruate, a pregnancy test is necessary. Elevated plasma E and P levels on this day are also supportive evidence of pregnancy. On this day in a pregnant patient plasma E rises to over 3 pg/ml and P to over 1 ng/ml. In the meantime, theresults of plasma E and P determinations from days. ' I B.BJ. (I) p. I ( lll) fh I (:SZ:) flo I RTO! '! Blood E a P Blood Ea P P-K Endo. Bx(p.r.n) Ep a J"'"'' ' l lrh ; P-K (poor) Endo. Bx (Secretory) 2 3 P-K (good) Preg.T. (t) 4 5 SCHEDULE 5 (Week) Blood E a P ( p.r.n) Preg Test ( p.r.n.) Resp.(-) tciomid Resp.( t) HCG at Ov. (-) Mid-<:ycle Resp.( +> E-B at Ov. <+> Mid-<:ycle P-K(X) Resp.(t) Ov. (t) P-K (o) Resll (+! Ov.(t) Preg.(t) Some Clomid OB Core FIG. 1. Monitoring of Clomid therapy: clinical schedule, evaluation, and planning. See text for details. RTO, return to office;resp., response; Ov., ovulation; the remaining abbreviations are defined in legend to Figure 9.

11 Vol. 3, No.6 MONITORING OF OVULATION INDUCTION and 21 are available for evaluation, and doses of clomiphene in future cycles can be adjusted in the patient who fails to conceive. Adjunctive therapy may also be considered if clinically justified, such as hcg at midcycle, E during the follicular phase, or even supplemental P during the luteal phase for a luteal phase defect. For practical purposes, the monitoring of clomiphene therapy can be scheduled as summarized in the upper portion of Figure 1. In patients with oligomenorrhea or secondary amenorrhea, clomiphene therapy can be instituted at any time, but preferably following progesterone withdrawal bleeding. After clomiphene medication, the patient is advised to return in 2 weeks, when a postcoital test and blood fore and P should be obtained. The patient returns again 1 week later, meanwhile continuing to chart her BBT at home. In the 3rd week of therapy, blood fore and P determinations is again obtained. Endometrial biopsy is recommended only if there is evidence of ovulation. In the 5th week of therapy, when the patient returns, it is important to review the previous plasma E and P levels, since at this point various courses of action may be selected depending on the findings. These are summarized in Figure 1 under groups I to V. Should E and P both be low (group I), this suggests that the patient's ovary did not respond to clomiphene and therefore the dosage should be increased. If the E level in the 2nd week is over 3 pg/ml, as shown in group II, this suggests that the ovary responded with follicular maturation, although ovulation is not assured. In the 3rd week if blood hormone data indicate low P and medium or low E, the patient's ovary probably responded to clomiphene yet failed to ovulate. In this case menstruation may or may not occur following E withdrawal. In such a patient clomiphene dosage need not be increased, rather supplementary hcg at midcycle is indicated. Supplemental hcg thus can be instituted even in a patient on a low dose of clomiphene. In group III, plasma E obtained in the 2nd week shows elevation, suggesting an ovarian response, and in the 3rd week a value for P of over 5 ng/ml suggests ovulation, yet the postcoital test is poor. In the following cycle, ovulation should be induced using the same dose of clomiphene, supplemented withe (either a low dose from the early to the late follicular phase, or higher doses during the preovulatory phase) to improve cervical mucus. The postcoital test should be repeated in the next cycle to evaluate response. Another type of response is seen in group IV, where the patient ovulated and the postcoital test at midcycle seems adequate; here there is no need either to increase clomiphene or to institute supplemental therapy. If the P level in the 3rd week measures between 2 and 6 ng/ml, a luteal phase defect may be suspected and either increased clomiphene or supplemental hcg is recommended. In the last group (group V) in which ovulation is successfully induced and the pregnancy test and plasma {3-hCG levels suggest pregnancy, plasma E is usually high (over 3 pg/ml), as is plasma P (over 1 ng/ml) in the 5th week of therapy. Following this active therapy protocol, every patient can be routed individually and actively evaluated every month, avoiding the passive waiting which has been usual. In clomiphene-hcg therapy a similar concept has been proposed 79 and similar phenomena have been observed. This approach may not necessarily improve the pregnancy rate, but it is definitely supportive to the patient. This may be beneficial from psychologic, social, and medical viewpoints. SUMMARY The plasma hormonal patterns of the normal menstrual cycle have been reviewed. A consistent cyclic pattern of plasma hormone levels is observed in LH, FSH, estrogens, and progestins in the menstrual cycle. Other plasma hormones, such as ACTH, growth hormone, TSH, and PRL, as well as androgens and corticosteroids, fluctuate throughout the menstrual cycle without any consistent pattern during the ovulatory cycle. FSH, LH, E2, Et. P, T, and A levels during the induced ovulatory cycle are presented for comparison. In the gonadotropin-induced ovulatory cycle most hormones behave in a manner similar to that in the normal ovulatory cycle, except for FSH levels, which rise continuously throughout the follicular phase of the cycle. Following ovulation in the gonadotropin-induced cycle, T rises above normal levels. Early in the clomiphene-induced ovulatory cycle, unlike the normal cycle, LH is distinctly elevated. Levels of both LH and FSH in the rest of the cycle simulate those in the normal cycle. However, T and A levels rise from the very beginning of clomiphene therapy and continue to rise throughout the clomiphene-induced ovulatory cycle. Levels of E and P are higher than in the normal ovulatory cycle, but a similar pattern is preserved. Because of the potential dangers of gonadotropin therapy, monitoring by frequent examination

12 628 wu and laboratory tests is required. E monitoring is mandatory to evaluate follicular maturation, to time hcg administration, and to minimize hyperstimulation. Cervical mucus is an unreliable parameter for monitoring gonadotropin therapy alone. In addition to cervical mucus, plasma or urinary E should be monitored regularly. Clomiphene therapy is less dangerous than gonadotropin therapy. Because of its lesser risk, monitoring is rarely performed during clomiphene use. An active monitoring approach has been described. While this approach may not necessarily improve the outcome of clomiphene therapy, it may hasten the process of selecting the appropriate dose. Although other ovulation-inducing agents are available, their use is rarely associated with serious medical complications, and monitoring would seem unnecessary. REFERENCES 1. Gemzell CS, Diczfalusy E, Tillinger KG: Clinical effect of human pituitary follicle-stimulating hormone (FSH). J Clin Endocrinol Metab 18:1333, Greenblatt RB, Barfield WE, Jungck EJ, Ray A W: Induction of ovulation with MRL/41. Preliminary report. JAMA 178:11, Ross GT, Cargille CM, Lipsett MB, Rayford PL, Marshall JR, Strott CA, Rodbard D: Pituitary and gonadal hormones in women during spontaneous and induced ovulatory cycles. Recent Prog Horm Res 26:1, Vande Wiele RL, Bogumil J, Dyrenfurth I, Ferin M, Jewelewicz R, Warren M, Rizhallah T, Mikhail G: Mechanisms regulating the menstrual cycle in women. Recent Prog Horm Res 26:63, Wu CH: Plasma hormones in clomiphene citrate therapy. Obstet Gynecol 49:443, Lehman F, Neale CH, Bettendorf G: Ovarian response during gonadotropin therapy measured by plasma progesterone, 17a-hydroxyprogesterone, and plasma estrogens. In Clinical Application ofhuman Gonadotropins. Proceedings of a Workshop Conference, Hamburg 197, Edited by G Bettendorf, V Insler. Stuttgart, Georg Thieme Verlag, 197, p Robertson DM, Steele SJ: The measurement of plasma oestradiol and progesterone in women with amenorrhea treated with gonadotrophins. Acta Endocrinol (Kbh) 69:68, McGarrigle HHG, Radwanska E, Little V, Swyer GIM: The monitoring of gonadotropin therapy by plasma oestradiol and progesterone determinations. J Obstet Gynaecol Br Commonw 81:657, Ebara Y, Siler T, Vandenberg G, Sinha YN, Yen SSC: Circulating prolactin levels during the menstrual cycle: episodic release and diurnal variation. Am J Obstet Gynecol 117:962, Gordon J, Wu CH, Mikhail G: Daily plasma prolactin in various gynecologic endocrinopathies. Fertil Steril. In press December Genazzani AR, Lemarchand-Beraud Th, Aubert ML, Felber JP: Pattern of plasma ACTH, hgh, and cortisol during the menstrual cycle. J Clin Endocrinol Metab 41:431, Judd HL, Yen SSC: Serum androstenedione and testosterone levels during the menstrual cycle. J Clin Endocrinol Metab 36:475, Abraham GE, Chakmakjian ZH: Serum steroid levels during the menstrual cycle in a bilaterally adrenalectomized woman. J Clin Endocrinol Metab 37:581, Thorneycroft IH, Mishell DR Jr, Stone SC, Kharma KM, Nakamura RM: The relation of serum 17-hydroxyprogesterone and estradiol-1713 levels during the human menstrual cycle. Am J Obstet Gynecol111:947, Wu CH, Prazak L, Flickinger GL, Mikhail G: Plasma 2ahydroxypregn-4-ene-3-one in the normal menstrual cycle. J Clin Endocrinol Metab 39:536, Katz FH, Romfh P: Plasma aldosterone and renin activity during the menstrual cycle. J Clin Endocrinol Metab 34:819, Zuspan FP, Zuspan KJ: Ovulatory plasma amine ( epinephrine and norepinephrine) surge in the woman. Am J Obstet Gynecol117:654, Arimura A, Kastin AJ, Schally AV: Immunoreactive LHreleasing hormone in plasma: midcycle elvation in women. J Clin Endocrinol Metab 38:51, Behrman HR, Caldwell BV: Role of prostaglandins in reproduction. In Reproductive Physiology, Edited by RO Greep. London, Butterworths, 1974, p WilksJW, Wentz AC, Jones GS: Prostaglandin F2a concentrations in the blood of women during normal menstrual cycle and dysmenorrhea. J Clin Endocrinol Metab 37:469, Moghissi KS, Syner FN, Evans TN: A composite picture of the menstrual cycle. Am J Obstet Gynecol 114:45, Noyes RW, Hertig AT, Rock J: Dating the endometrial biopsy. Fertil Steril 1:3, Corner GW: The histological dating of human corpus luteum of menstruation. Am J Anat 98:337, Wu CH: Plasma hormones in human gonadotropin induced ovulation. Obstet Gynecol 49:38, Schumert Z, Spitz I, Diamant Y, Polishuk WZ, Rabinowitz D: Elevation of serum testosterone in ovarian hyperstimulation syndrome. J Clin Endocrinol Metab 4:889, Kato J, Kobayashi T, Villee CA: Effect of clomiphene on the uptake of estradiol by the anterior hypothalamus and hypophysis. Endocrinology 82:149, Vaitukaitis JL, Bernudez JA, Cargille CM, Lipsett MB, Ross GT: New evidence for an anti-estrogenic action of clomiphene citrate in women. J Clin Endocrinol Metab 32:53, Vandenberg C, Yen SSC: Effect of anti-estrogenic action of clomiphene during the menstrual cycle: evidence for a change in the feedback sensitivity. J Clin Endocrinol Metab 37:356, Hsueh AJW, Erickson GF, Yen SSC: Clomiphene citrate (Clomid)-induced sensitization of pituitary cells to luteinizing hormone releasing hormone (LRF) in vitro (abstr 14). Presented at the 25th Annual Meeting of the Society for Gynecologic Investigation, Atlanta Ga, March 15-18, Jacobson A, Marshall JR, Ross GT, Cargille CM: Plasma gonadotropins during clomiphene induced ovulatory cycles. Am J Obstet Gynecol 12:284, Yen SSC, Vela P, Ryan KJ: Effect of clomiphene citrate in

13 Vol. 3, No.6 MONITORING OF OVULATION INDUCTION 629 polycystic ovary syndrome: relationship between serum gonadotropin and corpus luteum formation. J Clin Endocrinol Metab 31:7, Dupon C, Rosenfield RL, Cleary RE: Sequential changes in total and free testosterone and androstenedione in plasma during spontaneous and Clomid-induced ovulatory cycles. Am J Obstet Gynecol115:478, Graff G: Suppression of cervical mucus during clomiphene therapy. Fertil Steril 22:29, Murray M, Osmond-Clarke F: Pregnancy results following treatment with clomiphene citrate. J Obstet Gynaecol Br Commonw 78:118, Logan GG, Warren JC, Johnson DC: Induction of ovulation with human menopausal gonadotropin: the effect of prior treatment with estrogen. Fertil Steril 21:774, Thompson CR, Hansen LM: Pergonal (menotropins): a summary of clinical experience in the induction of ovulation and pregnancy. Fertil Steril 21:844, Hack M, Brish M, Serr DM, Insler V, Lunenfeld B: Outcome of pregnancy after induced ovulation. JAMA 211:791, Katz M, Lunenfeld B, Insler V: Induction of ovulation with gonadotropins. S Afr Med J 47:1272, Shaaban MM, Klopper A: A study on the monitoring of gonadotrophin therapy by the assay of plasma oestradiol and progesterone. J Obstet Gynaecol Br Commonw 8:783, Black WP, Coutts JRT, Dodson KS, Rao LGS: An assessment of urinary and plasma steroid estimations for monitoring treatment of anovulation with gonadotrophins. J Obstet Gynaecol Br Commonw 81:667, Tredway DR, Goebelsmann V, Thorneycroft IH, Mishell DR Jr: Monitoring induction of ovulation with human menopausal gonadotropin by a rapid estrogen radioimmunoassay. Am J Obstet Gynecol12:135, Ellis JD, Williamson JG: Factors influencing the pregnancy and complication rates with human menopausal gonadotropin therapy. Br J Obstet Gynaecol 82:52, Kistner RW: Further observations on the effects of clomiphene citrate in anovulatory females. Am J Obstet Gynecol 92:38, Poliak A, Smith JJ, Romney SL: Clinical evaluation of clomiphene; clomiphene and human chorionic gonadotropin; and clomiphene, human chorionic gonadotropin, and estrogens in anovulatory cycles. Fertil Steril24:921, Seki M, Tajima C, Maeda HR, Seki K, Yoshihara T: Effect of quinestrol administration with clomiphene citrate on serum FSH and LH and on other clinical findings. Am J Obstet Gynecol 116:388, Insler V, Zakut H, Serr DM: Cycle pattern and pregnancy rate following combined clomiphene-estrogen therapy. Obstet Gynecol 41:62, Taubert HD, Dericks-Tan JSE: High doses of estrogens do not interfere with the ovulation-inducing effect of clomiphene citrate. Fertil Steril 27:375, Evans JH, Taft HP, Brown JB, Adey FD, Johnstone JW: Induction of ovulation by cyclical hormone therapy. J Obstet Gynaecol Br Commonw 74:367, Evans J, Townsend L: The induction of ovulation. Am J Obstet Gynecol 125:321, Kistner RW: Induction of ovulation with clomiphene citrate (Clomid). Obstet Gynecol Survey 2:873, Spellacy WN, Cohen WD: Clomiphene treatment of prolonged secondary amenorrhea associated with pituitary gonadotropin deficiency. Am J Obstet Gynecol 97:943, MacGregor AH, Johnson JE, Bunde CA: Further clinical experience with clomiphene citrate. Fertil Steril 19:616, MacLeod SC, Mitton DM, Parker AS, Tupper WRC: Experience with induction of ovulation. Am J Obstet Gynecol 18:814, Garcia J, Jones GS, Wentz AC: The use of clomiphene citrate. Fertil Steril 28:77, DelPozo E, Varga L, Wyss H, Tolis G, Friesen H, Wenner R, Vetter L, Vettwiler A: Clinical and hormonal response to bromocriptin (CB-154) in the galactorrhea syndromes. J Clin Endocrinol Metab 39:18, Thorner MO, Besser GM, Jones A, Dacie J, Jones AE: Bromocriptine treatment offemale infertility: report of 13 pregnancies. Br Med J 4:694, Canales ES, Forsbach G, Soria J, Zarate A: Infertility due to hyperprolactinemia and its treatment with ergocryptine. Fertil Steril 27:1335, Tolis G, Naftolin F: Induction of menstruation with bromocryptine in patients with euprolactinemic amenorrhea. Am J Obstet Gynecol 126:426, Zarate A, Canales ES, Soria J, Gonzales A, Schally AV, Kastin AJ: Further observation on the therapy of anovulatory infertility with synthetic luteinizing hormone-releasing hormone. Fertil Steril 25:3, Henderson SR, Bonnar J, Moore A, Mackinnon PCB: Luteinizing hormone-releasing hormone for induction of follicular maturation and ovulation in women with infertility and amenorrhea. Fertil Steril 27:621, Zarate A, Canales ES, Soria J, Forsbach G, Kastin AJ, Schally A V: Therapeutic use of gonadoliberin (folliclestimulating hormone/luteinizing hormone-releasing hormone) in women. Fertil Steril 27:1233, Nakano R, Katayama K, Mizuno T, Tojo S: Induction of ovulation with synthetic luteinizing hormone-releasing hormone. Fertil Steril 25:471, Jones GES, Howard JE, Langford H: Use of cortisone in follicular phase disturbances. Fertil Steril 4:49, Greenblatt RB, Barfield WE, Lampros CP: Cortisone in the treatment of infertility. Fertil Steril 7:23, JefferiesWMcK, WeirWC, WeirDR,ProutyRL:Theuseof cortisone and related steroids in infertility. Fertil Steril 9:145, FerrimanD,Purdie WE, Tendall WJ: Use of corticosteroids in infertility associated with hirsutism and oligomenorrhea. Br Med J 1:16, Fayez JA: Selection of patients for clomiphene citrate therapy. Obstet Gynecol 47:671, Chang RJ, Abraham GE: Effect of dexamethasone and clomiphene citrate on peripheral steroid levels and ovarian function in a hirsute amenorrheic patient. Fertil Steril 27:64, Check JH, Rakoff AE, Roy BK: Induction of ovulation with combined glucocorticoid and clomiphene citrate therapy in a minimally hirsute woman. J Reprod Med 19:159, Insler V, Melmed H, Eden E, Serr D, Lunenfeld B: Comparison of various methods used in monitoring gonadotropic therapy. In Clinical Application of Human Gonadotropins. Proceedings of a Workshop Conference, Hamburg 197, Edited by G Bettendorf, V lnsler. Stuttgart, Georg Thieme Verlag, 197, p Taymor ML, Sturgis SH, Lieberman BL, Goldstein DP: Induction of ovulation with human postmenopausal