Comparison of pulsatile subcutaneous gonadotropin-releasing hormone and exogenous gonadotropins in the treatment of men with isolated hypogonadotropic hypogonadism

Comparison of pulsatile subcutaneous gonadotropin-releasing hormone and exogenous gonadotropins in the treatment of men with isolated hypogonadotropic hypogonadism

Vol. 49, No. 2, February 1988 Printed in U.S.A. FERTILITY AND STERILITY Copyright 0 1988 The American Fertility Society Comparison of pulsatile subc...

990KB Sizes 1 Downloads 43 Views

Vol. 49, No. 2, February 1988 Printed in U.S.A.

FERTILITY AND STERILITY Copyright 0 1988 The American Fertility Society

Comparison of pulsatile subcutaneous gonadotropinreleasing hormone and exogenous gonadotropins in the treatment of men with isolated hypogonadotropic hypogonadism

Linda Liu, M.D. *t Norma Chaudhari, R.N.* Donald Corle, M.S.:j: Richard J. Sherins, M.D.* National Institutes of Health, Bethesda, Maryland

Eight men with isolated hypogonadotropic hypogonadism were treated with pulsatile gonadotropin-releasing hormone (GnRH) after maximal testicular growth and function had already been achieved with human chorionic gonadotropin (hCG) and human menopausal gonadotropin (hMG). Only four subjects could normalize plasma testosterone (T) levels (group A). After 18 months of GnRH therapy, testicular size of group A increased by 53% (P < 0.01) over that previously attained with exogenous gonadotropins. However, despite further testicular growth, two men who were previously azoospermic on hCG/ hMG remained so on GnRH. In the other two patients, total sperm count increased minimally. Thus, pulsatile gonadotropin levels achieved with GnRH are more effective in stimulating testicular growth, but not necessarily sperm output, than are stable gonadotropin concentrations obtained with hCG/hMG. Fertil Steril 49:302, 1988

Pulsatile delivery of gonadotropin-releasing hormone (GnRH) is required for optimal stimulation of gonadotropin release, as either continuous or excessively frequent administration of GnRH results in a paradoxical suppression of pituitary secretion of luteinizing hormone (LH) and folliclestimulating hormone (FSH). 1 •2 However, whether pulsatile gonadotropic stimulation, as it occurs normally, is important for optimal testicular function remains unclear. This question is relevant, because conventional treatment of gonadotropin-de-

Received July 8, 1987; revised and accepted October 6, 1987. *Developmental Endocrinology Branch, National Institute of Child Health and Human Development. t Reprint requests: Linda Liu, M.D., NIH, Building 10, Rm 10N-262, 900 Rockville Pike, Bethesda, Maryland 20892. :j: Biometry Branch, Division of Cancer Prevention and Control, National Cancer Institute.

302

Liu et al.

Sequential hCG!hMG and GnRH therapies

ficient men with human chorionic gonadotropin (hCG) and human menopausal gonadotropin (hMG) results in a subnormal testicular size and reduced sperm output. 3 We postulate that the suboptimal testicular response obtained with exogenous gonadotropin administration might be related to the unphysiologic, continuous, nonpulsatile stimulation of the testes, 4 and that, consequently, testicular function might be improved further by providing pulsatile gonadotropic stimulation with GnRH therapy. To test this hypothesis, testicular function was assessed before and after chronic subcutaneous pulsatile GnRH administration in eight men with isolated hypogonadotropic hypogonadism (IHH) who previously had been treated with hCG/hMG. These subjects had achieved maximal testicular size and stable sperm output before they were switched from exogenous gonadotropins to subcutaneous GnRH therapy. Fertility and Sterility

MATERIALS AND METHODS Subjects

Eight men with IHH, aged 23 to 30 years, participated in the study after giving informed consent. The diagnosis of IHH had been made according to previously well-described criteria. 5 During their initial treatment with hCG/hMG and prior to the switch to GnRH therapy, the subjects had achieved normal plasma testosterone (T) concentrations (;;;.250 ng/dl) and maximal testicular size and sperm output, which were stable for more than 9 months. In six of the eight men, hCG 2000 IU and hMG (Pergonal, 75 IU FSH and 75 IU LH, Serono Laboratories, Inc., Randolph, MA) were co-administered three times per week intramuscularly from the inception of treatment for 31 to 46 months (mean ± standard deviation [SD] = 38 ± 2). The other two patients had received hCG alone for 15 and 58 months before hMG was administered with hCG for another 12 and 63 months, respectively. Study Protocol

The study was designed in two phases. During phase I, the dosage of GnRH required to stimulate plasma T to levels within the normal range (250 to 1000 ng/dl) was determined in each subject. Immediately before switching to pulsatile GnRH treatment, patients were admitted to the Clinical Center at the National Institutes of Health for physical examination and quantification of endocrine parameters following administration of the last hCG/hMG dose. Blood was withdrawn at 12, 18, 24, 30, 36, 42, 48, and 54 hours after hCG/hMG for measurement of plasma T and estradiol (E2) levels, and at 12, 36, and 54 hours for measurement of plasma sex hormone-binding globulin (SHBG) binding capacity. Then, upon completion of the blood sampling and 3 days after the administration of the last hCG/hMG dose, GnRH was instituted subcutaneously (SC) at 2 hourly intervals to mimic the LH pulse frequency of normal men. 6 •7 Because previous investigators have reported achievement of normal plasma T levels with a minimal GnRH dose of 25 ng/kg every 2 hours SC,8 •9 our starting dose of GnRH was 25 or 71 ng/kg every 2 hours (approximately 2 to 5 ~g/pulse for a 70-kg man). The GnRH dosage then was raised at monthly intervals by an increment of 71, 143, or 286 ng/kg every 2 hours (approximately 5, 10, or 20 ~g/pulse) until the plasma T level was within the normal range, or until patients discontinued GnRH treatVol. 49, No.2, February 1988

ment before plasma T concentrations could be normalized. Testicular examination, semen analysis, and blood sampling were performed in the clinic every month during phase I. Testicular size was measured with a Prader orchidometer (AB Alexander Graf, Halmstad, Sweden) by the same observer before the termination of hCG/hMG therapy and throughout GnRH treatment. Blood was collected every 20 minutes for 140 minutes starting at 1:00 to 2:00P.M. for measurement of plasma LH and FSH levels (during and following episodic GnRH release), and at 0, 60, and 120 minutes for measurement of plasma T and E2 concentrations. A single sample was obtained at 0 minutes for measurement of plasma SHBG binding capacity. At each visit, patients were specifically asked about compliance with the treatment and the presence of any symptoms. During phase II of the study, the effect of longterm GnRH therapy on testicular size and sperm production was further assessed. Each patient was maintained on an individualized, nonvarying GnRH dosage that had produced the highest plasma T level during phase I, and was seen every 2 months for physical examination, semen analysis, and blood sampling, as described previously. Following blood sampling, plasma was separated and stored at -20°C until assayed. To eliminate interassay variability, plasma T, E2, SHBG, LH, and FSH levels from blood samples obtained for each subject after the last hCG/hMG dose and throughout phase I of the study were measured in the same assays. This allowed comparisons of plasma T concentrations obtained during hCG/ hMG and GnRH therapies in the same patient, as well as plasma hormone concentrations achieved at various GnRH doses in the same subject. Gonadotropin-Releasing Hormone

Synthetic GnRH was purchased initially from Ayerst (Factrel, New York, NY) in the first 5 months of the protocol, and subsequently from Bachem (Torrance, CA) for the remainder of the study. Analysis by high-pressure liquid chromatography demonstrated that the two GnRH preparations had equivalent activity on a weight basis. In addition, the drugs had indistinguishable biologic activities in vivo at GnRH doses ;a. 429 ng/kg every 2 hours, as similar plasma gonadotropin and T levels were obtained when the same dose of each of the GnRH batches was given for 4 weeks (three subjects, data not shown). Liu et al.

Sequential hCG/hMG and GnRH therapies

303

GnRH was administered via a portable, batteryoperated pump (Pulsamat, from Ferring Laboratories, Inc., Suffern, NY) that delivers 50 ~-tl of solution during 1 minute through a needle placed SC in the arm, thigh, or abdomen. Patients were instructed to change the needle and needle placement site every 3 to 4 days. Semen

RESULTS Subject Selection

Patients were instructed to collect one semen sample each month during hCG/hMG and GnRH therapies, using a strict abstinence period of 36 to 48 hours. For patients living out of town, samples were kept frozen until analysis. Semen analysis was performed as described previously. 10 Hormone Assays

Plasma T 11 and E 2 12 were measured by specific radioimmunoassays (RIAs) following ether extraction and celite chromatography. Intra- and interassay coefficients of variation at a T dose of 330 ng/dl were 7.1 and 14.6%, respectively. The least detectable plasma E 2 concentration was 10 to 20 pg/ml. The normal upper limit for plasma E 2 for men is 40 to 58 pg/ml. Intra- and interassay coefficients of variation at an E 2 level of 30 pg/ml were 5.7 and 13.4%, respectively. The level of SHBG binding capacity was determined by a solid phase method using concanavalin A-Sepharose 4B and performed at a physiologic pH and temperature. 13 Sensitivity of the assay was 0.1 ~-tg/dl; intra- and interassay coefficients of variation were 9 and 14%, respectively. Free T levels were calculated using a computer program that determined the amount of free steroid based on the plasma SHBG binding capacity, total T, and albumin concentrations. 14 Plasma LH and FSH concentrations were measured using previously described RIAs. 15 Statistical Analyses

Pair-wise correlations were evaluated using the Pearson product-moment correlation coefficient. Because of their non -Gaussian distribution, values for plasma T and E 2 were log-transformed before analysis. Plasma levels of each hormone achieved at varying GnRH doses in groups A and B were compared using slope and intercept estimates from the analysis of covariance. The eight individuals were modeled as having the same linear dose-response slope, while each had his own intercept term. Comparisons of the mean dose-response slopes of each plasma hormone between the two 304

groups were made using an interaction term of dose with group. All other comparisons were made by means of two-tailed paired or unpaired Student's t-test, where appropriate. Differences were considered statistically significant when P < 0.05.

Liu et al.

Sequential hCG/hMG and GnRH therapies

Clinical and laboratory characteristics of the eight subjects with IHH before receiving GnRH treatment are shown in Table 1. Treatment duration of hCG/hMG averaged 38 ± 14 months (mean ± SD; range, 12 to 63). Two men remained azoospermic after 31 and 40 months, respectively, of therapy with combined exogenous gonadotropins. In one of them, a left varicocoele had been present, the repair of which had led to further ipsilateral testicular growth during additional hCG/hMG therapy for 9 months prior to our GnRH study. Testicular size and sperm output of all the subjects had reached a plateau for a mean of 21 ± 8 months (range, 9 to 29) before the switch to GnRH therapy was instituted. Phase 1: Gonadotropin-Releasing Hormone DoseResponse Study

While plasma T concentrations achieved previously with hCG/hMG were uniformly well within the normal range in all subjects, those attained during the GnRH dose-response study segregated the patients into two distinct groups. In four men (group A), the highest plasma T concentrations achieved were >250 ng/dl, whereas in the other four subjects (group B), the highest plasma T levels remained consistently <170 ng/dl at a similar GnRH dose range (Fig. 1). Except for age at the onset of GnRH treatment, the two groups did not differ in their clinical and laboratory characteristics prior to any treatment or during previous hCG/hMG therapy (Table 1). Except for one patient in group A treated with the highest GnRH dose (1094 ng/kg every 2 hours), the GnRH dose range tested in subjects of both groups was similar, ranging from 25 to 71 ng/kg to 429 to 714 ng/kg every 2 hours. In group A, all four subjects achieved plasma T within the normal range at minimal GnRH doses of 71 to 286 ng/kg SC every 2 hours. The plasma hormone dose-response leastsquares lines of individual patients within each of the two groups are depicted in Figures 2 and 3. When the data of all eight patients were analyzed Fertility and Sterility

Table 1 Clinical and Laboratory Characteristics of the Eight hCG/hMG-Treated Hypogonadotropic Men Prior to Receiving GnRH Therapy

Age (years) Testis size (ml) Pre-hCG/hMG Post-hCG/hMG Sperm output Azoospermic subjects Sperm-positive, n = 6 (X106 /ejaculate) Plasma hormone levels Testosterone (ng/dl) Estradiol (pg/ml) Treatment duration with hCG/hMG (months)

All patients (n = 8)

Group A (n = 4)

26.6 ± 3.0" (range, 23-30)

24.0 ± 1.4

28.3 ±

2.4c

2.6 ± 1.7 (range, 0.5-5.0) 8.4 ± 0.9 (range, 5.0-12.0)

2.4 ± 1.7

2.8 ±

1.9

7.6 ± 2.1

9.1 ±

2.8

2 2.3b (range, 0.4-12.1)

2 5.2b (range, 1.2-12.1)

521 ± 100 (range, 341-634) 54 ± 28 (range, 16-92) 38 ± 14 (range, 12-63)

• Values are expressed as mean ± SD, unless otherwise stated. b Median.

Group B (n = 4)

c

0 2.1b (range, 0.4-4.8)

538

± 89

504

± 122

60

± 26

52

± 29

30

± 12

46

± 13

P < 0.05 when comparing mean values of groups A and B.

together, increases in the levels of plasma T, E 2 , LH, and FSH were significantly (P < 0.05) correlated with increases in the GnRH dose from 25 to 1094 ng/kg every 2 hours. A strong positive correlation was found between plasma T and LH levels (r = 0.785, P < 0.00001). Plasma SHBG binding capacity was significantly (P < 0.05) and negatively correlated with increasing GnRH dosage. The mean dose-response slopes of plasma T, SHBG, LH, and FSH were found to be similar for both groups A and B because of the nonsignificance (P > 0.05) of a statistical interaction term. For plasma E 2 , the significance of this interaction (P < 0.05) indicated that increases in GnRH dose were associated with higher proportionate increases in plasma E 2 for patients in group A than for those in group B. There were no significant differences between the plasma SHBG binding capacities achieved at various GnRH doses in the two groups. Thus, the lower plasma T levels observed in patients of group B could not be attributed to lower plasma SHBG binding capacities. In fact, plasma calculated free T concentrations at various GnRH doses were significantly lower (P < 0.00001) in group B than in group A. Though mostly falling within the normal range, plasma LH and FSH achieved at varying GnRH doses in group B were significantly lower (P < 0.00001) than those obtained in group A. When

pulsatile characteristics of plasma LH achieved with the same GnRH doses (during one pulse interval) were compared between the two groups, the mean rises of plasma LH from nadir to peak of group B were half those of group A (data not shown). Each patient was asked repeatedly about compliance, which could contribute to the success or fail-

Figure 1 Plasma T concentrations achieved during prior hCG/hMG treatment (left) and highest plasma T levels obtained during the GnRH dose-response study (right) in each of the eight men with IHH. Values of subjects in group A are represented in closed circles, and those of group B in open circles. The gray area denotes normal male range.

Vol. 49, No. 2, February 1988

Liu et al.

800

~

E.'

600

UJ

z a:

0

~ 400

c.n

0 1--

r.n

~ 200

8 8 hCG/hMG

GnRH

Sequential hCG!hMG and GnRH therapies

305

GROUP A

GROUP B

GROUP B

GROUP A LH

LH

In T

FSH

In E,

GnRH Dose (ng/kg q 2 h s.c.)

SHBG

Figure 3 Mean plasma LH and FSH concentrations of subjects in group A (left panel) and group B (right panel) at varying doses of GnRH administered SC every 2 hours. Each symbol and number represents an individual patient. Each line represents the dose-response slope for an individual patient. OL;!;,.~..,-'-.,.,;\;-'----.l,.-----'-------;;;i,,*

n

a

~

26

143

n

367

&

~

sn

""'

GnRH Dose lng/kg q 2 h s.c.l

Figure 2 Plasma ln T, ln E 2 , and SHBG concentrations of subjects in group A (left panel) and group B (right panel) at varying doses of GnRH administered SC every 2 hours. The gray area denotes normal male range for each plasma hormone. Each symbol and number represents an individual patient. The same number designates the same patient in Figures 3 and 4. Each line represents the dose-response slope for an individual patient.

ure of GnRH treatment. Whereas each subject in group A claimed good compliance, three of the four men in group B admitted to intermittently interrupting GnRH treatment for 20% to 40% of the time. The reasons given by the patients for the lack of compliance include interference with daily physical activities or job requirements, and social embarrassment at wearing the device. Because of persistent symptomatic hypogonadism, all four patients in group B chose to discontinue GnRH therapy after 5 to 13 months (mean, 9.0 months) and before completing phase I of the study. At the time of discontinuing GnRH administration, the mean testicular size of group B was not statistically different from that previously achieved with exogenous gonadotropins (8.1 ± 2.3 versus 9.1 ± 2.8 ml, not significant [NS]). Median total sperm count decreased in three of the four men, though still remaining at a detectable level of 0.3 to 0.9 X 106 (Fig. 4, right panel). In the remaining subject 306

Liu et al.

Sequential hCG/hMG and GnRH therapies

(patient B-8), the 1.9-fold increase in median total sperm output observed during GnRH therapy was probably due to prolonged sexual abstinence, as the patient suffered from loss of libido and potency. Phase II: Testicular Response to Long-Term Gonadotropin-Releasing Hormone Therapy

Only the four patients in group A were available for assessment of testicular response during longterm GnRH therapy. In these men, GnRH was given for a total duration (phases I and II) of 18 to 24 months (mean, 22 months). During phase II, GROUP B

GROUP A

7030 ~ ifA-2 ..!! ~

8

, tl) ,I I '



• • • ~··

1 -

I

I

I

I

!

••• I

I

,.

ee

...

J

E~A-4

t

-24

,

J

I



5

I

I

I

·12

-8

t ';

I

..

!

18·8 .. . ..,•

40 1

-Ill

r

Itt

.

]

~

• • •

• .,.



• :·



f

I

I

I

I

I

~•

-12

~

o

a

12

I

m

I

~

TIME !months!

Figure 4 Total sperm count of subjects in group A (left panel) and group B (right panel) during the last 24 months of exogenous gonadotropin administration and during 5 to 24 months of GnRH therapy. Fertility and Sterility

plasma T levels were maintained within the normal range (mean ± SD = 390 ± 49 ng/dl) with the administration of GnRH 429 to 714 ng/kg every 2 hours SC. After 18 months of GnRH therapy, testicular size increased in all patients in group A by a mean of 53% over that previously achieved with exogenous gonadotropins (mean± SD = 11.6 ± 3.0 versus 7.6 ± 2.1 ml; P < 0.01). However, despite further testicular growth, the two men who were previously azoospermic on hCG/hMG remained so on GnRH. In the remaining two subjects who were previously sperm-positive while receiving exogenous gonadotropins, median total sperm count increased from 12.1 to 26.5 million in one man, and from 1.2 to 8.0 million in the other during GnRH administration (Fig. 4, left panel).

The sequential treatment of gonadotropin-deficient men with exogenous gonadotropins followed by pulsatile GnRH therapy allowed us to address the question of whether pulsatile gonadotropic stimulation might result in a superior testicular response compared with that previously obtained with continuous gonadotropic stimulation using exogenous gonadotropins. Additionally, we were able to compare the relative efficacy of both treatment methods in the same men. In each of the four subjects who maintained plasma T levels within the normal range on longterm GnRH therapy (group A), a substantial increase in testicular size over that previously attained with exogenous gonadotropins was noted. It is unlikely that this further testicular growth might represent a process that would have occurred had patients continued exogenous gonadotropins, because we previously have shown that testis size remains stable once it has plateaued for 9 months or longer. 16 We also cannot attribute a mean increase in testicular size of 4 ml to a measurement artifact because only one examiner was employed throughout the study, eliminating between-observer variability, and our within-observer error in measuring testicular volume is approximately 10%. Preliminary observations suggest that the additional testicular growth gained during GnRH therapy is at least partially reversible, since the mean testicular size of two men decreased 20% 5 months following cessation of GnRH treatment and reinstitution of hCG/hMG therapy (data not shown). Based on these observations, pulsatile gonadotropin levels appear to be a more potent stimulus to the growth

of the seminiferous tubules than the steady concentrations achieved with exogenous gonadotropins. Despite further testicular growth during GnRH treatment, sperm production failed to appear in the two azoospermic men previously treated with hCG/hMG. The causes of azoospermia in these two patients are unclear, but appear to be unrelated to the type of gonadotropic stimulation provided by either exogenous gonadotropins or GnRH. It is also unlikely that the previous presence of a varicocoele in one of the subjects contributed to his infertility, because azoospermia is rarely associated with varicocoele, and the causal relationship of varicocoele and infertility remains highly controversiat1 7 Given the dissociation between testicular growth and sperm output in these azoospermic subjects, we conjecture that GnRH therapy might increase total germ cell mass without completing spermiogenesis. It is conceivable that a period of GnRH therapy even longer than 18 to 24 months might be required for significant improvement in sperm production in some subjects. In the other two previously sperm-positive subjects, sperm output was augmented 2- to 7-fold. The biologic significance of this increase is, however, uncertain. In particular, the 2-fold increase in median sperm count in one of the men is unlikely to be important, since the range of values on both therapies is similar, and wide variation (up to 10- to 15-fold) in total sperm count or sperm concentration can be observed within any single individual, even if semen collections are performed after a fixed abstinence period. 10 •18 In the other four patients (group B), who suffered from symptomatic hypogonadism and in whom plasma T leves remained persistently below 170 ng/dl during GnRH therapy, testicular size and sperm output achieved during hCG/hMG therapy were maintained for a surprisingly long time on GnRH administration. This situation is quite reminiscent of the fertile eunuch syndrome 19 in which T production, in response to partial gonadotropic stimulation, is inadequate to maintain peripheral virilization, but is sufficient to stimulate testicular growth and germinal maturation. In our group B patients, low plasma T levels were associated with distinct LH pulses, but the pulses were lower and flatter than those measured in the good GnRH responders (group A). Such blunted hormonal responses could be accounted for by reduced SC absorption of GnRH and/ or excessive degradation of GnRH by tissue peptidases, 20 thus resulting in in-

Vol. 49, No. 2, February 1988

Liu et al.

DISCUSSION

Sequential hCG/hMG and GnRH therapies

307

sufficient bioavailability of the hormone at the pituitary. In addition, reduced patient compliance in administering the GnRH contributed to the inadequate gonadotropic and testicular responses, and probably led to reduced pituitary and gonadal priming. 21 It is possible that intravenous administration of pulsatile GnRH, which produces more physiologic plasma GnRH and gonadotropin pulses, 22•23 might result in a better testicular response than that obtained with GnRH delivered subcutaneously. Improvement in ovulation induction and pregnancy rate has been demonstrated in hypogonadotropic women treated with intravenous GnRH as compared with SC GnRH. 24 However, intravenous administration of GnRH in the male is impractical because the hormone must be given for extended time periods to effect adequate stimulation of the germinal tissue for the appearance of sperm in the ejaculate. Additionally, the increased risks of venous thrombosis and infection entailed by chronic indwelling intravenous catheters remain major concerns in the implementation of long-term intravenous GnRH administration in otherwise healthy subjects.

Acknowledgments. We gratefully acknowledge the expert help of the outpatient and inpatient nursing staff and the NIH Pharmaceutical Development Service throughout the study.

REFERENCES 1. Belchetz PE, Plant TM, Nakai Y, Keogh EJ, Knobil E: Hypophyseal responses to continuous and intermittent delivery of hypothalamic gonadotropin releasing hormone. Science 202:631, 1978 2. Spratt DI, Finkelstein JS, Butler JP, Badger TM, Crowley WF Jr: Effects of increasing the frequency of low dose of gonadotropin-releasing hormone (GnRH) on gonadotropin secretion in GnRH-deficient men. J Clin Endocrinol Metab 64:1179, 1987 3. Sherins RJ: Hypogonadotropic hypogonadism. In Current Therapy of Infertility (2nd edition), Edited by CR Garcia. St Louis, CV Mosby, 1984, p 147 4. Sherins RJ: Unpublished data 5. Sherins RJ, Howards SS: Male infertility. In Campbell's Urology (5th edition), Edited by PC Walsh, RE Gittes, AD Perlmutter, TA Stamey. Philadelphia, WB Saunders, 1985, p 640 6. Santen RJ, Bardin CW: Episodic luteinizing hormone secretion in man. J Clin Invest 52:2617, 1973 7. Merriam GR, Brody SA, Collins RL, Evans WS, Klibansky A, Koppelman MC: Episodic gonadotropin secretion in normal and hyperprolactinemic men and women. In Gonad-

308

Liu et al. Sequential hCG/hMG and GnRH therapies

8.

9.

10.

11.

12. 13.

14.

15.

16. 17. 18.

19.

20.

21.

22.

23.

24.

otropins and Prolactin, Edited by C Dotti. Milano, Becton Dickinson, 1984, p 61 Hoffman AR, Crowley WF Jr: Induction of puberty in men by long-term pulsatile administration of low-dose gonadotropin-releasing hormone. N Eng! J Med 307:1237, 1982 Crowley WF Jr, Spratt Dl: The results of long-term GnRH administration in idiopathic hypogonadotropic hypogonadism. In LHRH and Its Analogues, Edited by F Labrie, A Belanger, A Dupont. Quebec, Elsevier Science Publishers BV, 1984, p 479 Sherins RJ, Brightwell D, Sternthal PM: Longitudinal analysis of semen of fertile and infertile men. In The Testis in Normal and Infertile Men, Edited by P Troen, HR Nankin. New York, Raven Press, 1977, p 473 Abraham GE: Radioimmunoassay of plasma steroid hormones. In Modern Methods of Steroid Analysis, Edited by E Heftman. New York, Academic Press, 1973, p 451 Jiang N, Ryan RJ: Radioimmunoassay for estrogens: a preliminary communication. Mayo Clin Proc 44:461, 1969 Nisula BC, Dunn JF: Measurement of the testosterone binding parameters for both testosterone-estradiol binding globulin and albumin in individual serum samples. Steroids 34:771, 1979 Dunn JF, Nisula BC, Rodbard D: Transport of steroid hormones: binding of 21 endogenous steroids to both testosterone-binding globulin and corticosteroid-binding globulin in human plasma. J Clin Endocrinol Metab 53:58, 1981 Odell WD, Rayford PL, Ross GT: Simple partially automated method for radioimmunoassay of human thyroid stimulating, growth, luteinizing and follicle stimulating hormones. J Lab Clin Med 70:973, 1967 Sherins RJ: Unpublished data Vermeulen A, Vandeweghe M: Improved fertility after varicocele correction: fact or fiction? Fertil Steril 42:249, 1984 Clark RV, Sherins RJ: Use of semen analysis in the evaluation of the infertile couple. In Male Reproductive Dysfunction, Edited by RJ Santen, RS Swerdloff. New York, Marcel Dekker, 1986, p 253 McCullagh EP, Beck JC, Schaffenburg CA: A syndrome of eunochoidism with spermatogenesis, normal urinary FSH and low or normal ICSH (fertile eunuchs). J Clin Endocrinol 13:489, 1953 Handelsman DJ, Swerdloff RS: Pharmacokinetics of gonadotropin-releasing hormone and its analogs. Endocr Rev 7:95,1986 Barkan AL, Reame NE, Kelch RP, Marshall JC: Idiopathic hypogonadotropic hypogonadism in man: dependence of the hormone responses to gonadotropin-releasing hormone (GnRH) on the magnitude of the endogenous GnRH secretory defect. J Clin Endocrinol Metab 61:1118, 1985 Spratt Dl, Crowley WF Jr, Butler JP, Hoffman AR, Conn PM, Badger TM: Pituitary luteinizing hormone responses to intravenous and subcutaneous administration of gonadotropin-releasing hormone in men. J Clin Endocrinol Metab 61:890, 1985 Handelsman DJ, Jansen RPS, Boylan LM, Spaliviero JA, Turtle JR: Pharmacokinetics of gonadotropin-releasing hormone: comparison of subcutaneous and intravenous routes. J Clin Endocrinol Metab 59:739, 1984 Reid RL, Leopold GR, Yen SSC: Induction of ovulation and pregnancy with pulsatile luteinizing hormone releasing factor: dosage and mode of delivery. Fertil Steril 36:553, 1981

Fertility and Sterility