Comparison of gonadotropin-releasing hormone and gonadotropin therapy in male patients with idiopathic hypothalamic hypogonadism*

Transcription

1 - FERTILITY AND STERILITY Copyright 1991 The American Fertility Society Vol. 56, No, 6, December 1991 Printed on acid-free paper in U.S.A. Comparison of gonadotropin-releasing hormone and gonadotropin therapy in male patients with idiopathic hypothalamic hypogonadism* Jochen Schopohl, M,D.t Gerhard Mehltretter, M.D. Rupprecht von Zumbusch, M.D. Thomas Eversmann, M.D. Klaus von Werder, M.D.:j: Medizinische Klinik, Klinikum Innenstadt, Miinchen, Germany Objective: To compare pulsatile gonadotropin-releasing hormone (GnRH) therapy with gonadotropin therapy in male patients with idiopathic hypothalamic hypogonadism. Design: Prospective study. Patients had free choice between the two forms of therapy. Setting: Patients were treated on an outpatient basis in our department. Patients: Eighteen patients of matched age (mean [±SDj age: 21.1 ± 3.0 years and 23.6 ± 7.3 years) and similar testicular volume were treated in each group. Interventions: Pulsatile GnRH therapy was started with 4 J.Lg GnRH subcutaneously every 2 hours using a portable pump and gonadotropin therapy with 3 X 2,500 IU human chorionic gonadotropin (hcg) weekly injected intramuscularly. After 8 to 12 weeks of hcg treatment, 150 IU human menopausal gonadotropin two to four times weekly were added. Results: Testosterone (T) and estradiol (E 2) levels increased significantly higher (T: P < 0.03; E 2 : P < 0.001) in the gonadotropin group than in the GnRH group (T: 22.5 ± 8.1 versus 16.8 ± 5.5 nmol/l; E 2 : 150 ± 70 versus 88. ± 59 pmol/l). Five patients developed gynecomastia during gonadotropin therapy. The rise of testicular volume was significantly more pronounced (P < 0.001) in the GnRH group (1; testicular volume = 8.1 ± 2.0 ml) than in the gonadotropin group (1; testicular volume = 4.8 ± 1.8 ml). Ten patients of the GnRH and 8 of the gonadotropin group had positive sperm counts, ranging from 1.5 to 26 X 10 6 spermatozoa/ml. The latter was achieved more rapidly in the GnRH group (12 ± 1.6 versus 20 ± 2.3 months; P < 0.02). Conclusions: Endocrine and exocrine testicular function can be normalized by both forms of therapy. Gonadotropin therapy has more side effects. Gonadotropin-releasing hormone leads to a higher testicular volume and a more rapid initiation of spermatogenesis compared with gonadotropin therapy. Fertil Steril 1991;56: Nowakowski et al. l demonstrated in 1961 that hypogonadism associated with anosmia (Kallmann's syndrome 2 ) is caused by a hypothalamic defect. Soon Received May 7, 1991; revised and accepted August 20, * Presented in part at the 31th Meeting of the German Endocrine Society, Munster, Germany, March 4 to 7, 1987 and at the 70th Meeting of the American Endocrine Society, New Orleans, Louisiana, June 8 to 11, t Reprint requests to: Jochen Schopohl, M.D., Medizinische Klinik, Klinikum Innenstadt, Ziemssenstr. 1, D-8000 Munchen 2, Germany. * Present address: Schlosspark-Klinik, Free University of Berlin, Germany. afterwards, attempts were made to treat these patients with gonadotropins. 3 Gonadotropin treatment led to good results especially in inducing spermatogenesis in a large number of patients with idiopathic hypothalamic hypogonadism. 4-6 A new therapeutic approach evolved through the studies of Belchetz et al. 7 in 1978 demonstrating that pulsatile gonadotropin-releasing hormone (GnRH) secretion is essential for normal function of the pituitary-gonadal axis in humans. Subsequently, pulsatile GnRH therapy was shown to have considerable advantages compared with gonadotropin therapy in women with hypothalamic amenorrhea avoiding ovarian overstimulation and multiple Vol. 56, No.6, December 1991 Schopohl et al. GnRH and gonadotropin therapy in males 1143

2 pregnancies.8-l0 Also in men pulsatile GnRH therapy was able to initiate puberty and achieve spermatogenesisy-i4 Although GnRH as well as gonadotropin therapy proved to be effective in the treatment of male idiopathic hypothalamic hypogonadism, only two reports exist comparing both forms of therapy.i5,i6 Liu et al,15 found no improvement in endocrine or exocrine testicular function of eight men with idiopathic hypothalamic hypogonadism previously treated with gonadotropin, although they observed a further increase in testicular volume. The same authori6 published a second study with 5 patients treated with GnRH and 11 with gonadotropins reporting no significant differences in testicular volume and sperm count between both forms of therapy. The aim of the present study was to compare the effectiveness of GnRH and gonadotropins in normalizing endocrine and exocrine testicular function in two groups of male patients with idiopathic hypothalamic hypogonadism. Patients MATERIALS AND METHODS Thirty-five men with hypogonadotropic hypogonadism were selected for the study (Tables 1 and 2). All had failed to undergo puberty by the age of 18, except two patients who were already included at age 17 because they belonged to the group of 16 patients with Kallmann's syndrome. The diagnosis was established by a low testosterone (T) level «3.5 nmol/l) in the presence of low basal serum gonadotropin levels, which showed no or only a slight increase after 100 JLg GnRH intravenously. Additional anterior pituitary insufficiency was excluded by appropriate testing. (In patient 14, complete anterior pituitary failure of hypothalamic origin was documented in childhood. This patient had received growth hormone treatment for several years and was under substitution therapy with 100 JLg L-thyroxine and 25 mg hydrocortisone per day). Furthermore, in all patients a human chorionic gonadotropin (hcg) test was performed to exclude primary testicular insufficiency. A pituitary mass lesion or a suprasellar tumor was excluded by skull x ray and by cranial computerized tomography. After establishing the diagnosis of hypothalamic hypogonadism, the patients were informed about the treatment possibilities, i.e., T replacement, pulsatile GnRH therapy, or gonadotropin therapy. All patients wanted to know if spermatogenesis could be induced. Two patients wanted to become fertile to father children. They were informed about the study comparing GnRH with gonadotropin therapy, and consent was obtained. Every patient could choose between gonadotropin and GnRH therapy. Ten patients had been pretreated, 9 with T and one with gonadotropins (5 years before entering the study). Testosterone replacement was discontinued at least 3 months before starting therapy with GnRH or gonadotropin. Before starting therapy, all patients were azoospermic or aspermic. Patient 8 was switched from the GnRH group to the gonadotropin group after 3 months of futile GnRH treatment because of lack of compliance; in patient 9, pulsatile therapy was discontinued for the same reason after 4 months. Minimal duration of treatment was 3 months. Three patients (4, 13, and 20) refused semen analysis, and 2 of them stopped therapy after 3 months and 1 after 6 months, respectively. Eighteen patients were treated with GnRH (10 with idiopathic hypothalamic hypogonadism and 8 with Kallmann's syndrome) (Table 1). Mean (±SD) age at the beginning of therapy was 21.1 ± 3.0 years, mean bone age was 15.1 ± 2.0 years (in 2 patients epiphyseal closing already was complete, their bone age was considered 19 years). Testicular volume was 3.0 ± 2.3 ml. Four patients had had orchidopexy because of unilateral cryptorchism. Mean T levels were 1.3 ± 0.6 nmol/l. Basal and stimulated luteinizing hormone (LH) were 3.0 ± 1.0 and 12.2 ± 7.8 lull, respectively. Basal and stimulated folliclestimulating hormone (FSH) were 1.2 ± 0.3 and 2.9 ± 1.9 lull, respectively. Eighteen patients were treated with gonadotropins (9 with idiopathic hypothalamic hypogonadism and 9 with Kallmann's syndrome) (Table 2) with a mean age of 23.6 ± 7.3 years, mean bone age was 15.8 ± 2.4 years (in 5 patients epiphyseal closing was already complete). Testicular volume was 3.6 ± 3.0 ml. Five patients had had orchidopexy because of unilateral cryptorchism. Mean T levels were 1.2 ± 0.8 nmol/l. Basal and stimulated LH were 2.2 ± 1.1 and 8.2 ± 5.1 lull, respectively. Basal and stimulated FSH were 1.2 ± 0.7 and 2.3 ± 1.0 lull, respectively. Protocols Patients of the GnRH group were treated with a Zyclomat pump or a BT 2 pump (Ferring, Kiel, Germany). Through a subcutaneously (SC) placed butterfly, which was changed at least every 2 days, GnRH was administered in 90 (2, 3, and 11) or 120- minute intervals. Therapy was started with 4-JLg GnRH/bolus. The dosage was doubled if after Schopohl et al. GnRH and gonadotropin therapy in males Fertility and Sterility

3 (0 ( /) 1:1" 0 'ti e: (D...!!!- ;:l ~ '".,.g <: Table 1 Data of the Patients of the GnRH Group ~ 01 Before therapy $1' On therapy Z 9 Patient Testicular LH LH after FSH FSH after Sperm Time of Testicular Sperm 0. no. volume Age Pretreatment Cryptorchidism Anosmia T E2 basal stimulation basal stimulation count treatment volume count t) (I) <> ml y nmolll pmolll lull mo ml XI(/' (I) No g Az (I) Yes As" No Az No As 3 10 NO' Yes Az No As Yes Az d 3 18 No As 3 3 As 9d 3 18 No As 4 3 As No As No As No As No As 3 12 NO Yes Az No As No As Yes Az No As Mean ± SO 3.0 ± ± ± ± ± ± ± ± ± ± 2.4 CI Az, azoospermia. b As, aspermia. 'NO, not done. dlack of compliance. 0 Table 2 Data of the Patients of the Gonadotropin Group Before therapy On therapy Patient Testicular LH LH after FSH FSH after Sperm Time of Testicular Sperm c;j no. volume Age Pretreatment Cryptorchidism Anosmia T E2 basal stimulation basal stimulation count treatment volume count ~ ::x:: ml y nmolll pmolll lull mo ml XI(/' No Az Yes Az 6 12 NO" ~ No As' Yes As Yes Az q No Az g No As S No As No As ; Yes Az No As '< Yes Az S No As No ;:! As No Az ~ Yes As No As No As Mean ± SO 3.6 ± ± ± ± ± ± ± ± ± ± 3.0 ""' J!:o. ""' C1 o Az azoospermia. "NO, not done. C As, aspermia.

4 weeks no pituitary response was seen. The highest dose needed was 16-~g GnRH/bolus. Before starting therapy, a 4-hour LH and FSH profile was performed with blood samples taken every 20 minutes. This profile was repeated on day 1 of therapy, after 2 and 4 weeks of therapy, thereafter in 2 or 3-month intervals if the GnRH dosage was stable and pituitary and testicular response satisfactory. When GnRH dose had to be changed, LH and FSH were monitored every 4 weeks. Testosterone and E2 were determined every 4 to 6 weeks. In the same intervals, the patients were examined physically with measurement of testicular volume by Prader orchidometer (Consult, Gern, Switzerland). Testicular volume was recorded as the mean volume from both testes. The testicular volume of each patient was always examined by the same investigator to avoid interobserver variations. The patients were treated on an outpatient basis after they had learned to place the SC butterflies and to refill the pump with GnRH. In the gonadotropin group, therapy was started with 2,500 IV hcg three times a week intramuscularly (1M). In case of a supraphysiological T or E2 increase, the hcg dose was reduced. The minimal hcg dose needed for maintaining T within the normal range was 1,000 IV two times a week. After 2 or 3 months oftreatment, 150 IV human menopausal gonadotropin (hmg) two times a week were added. The hmg dosage was increased up to three and four times a week if testicular growth and development of sperm count were unsatisfactory. Testosterone, estradiol (E 2), and testicular volume were checked every 4 weeks during the 1st 4 months of therapy and then in 2-month intervals if T and E2 remained within the normal range. When gonadotropin dosage was changed, examinations were continued in 4- week intervals. Semen analysis in both groups was performed before therapy, after 3 months of therapy, and then in 2 or 3-month intervals. Patients from both groups were examined by all investigators in a random manner. In 24 patients, the same physician performed always the examination. Laboratory Procedures Hormones were measured by radioimmunoassay (RIA) used for clinical routine. Luteinizing hormone FSH (Travenol-Gentech-Diagnostics, Cambridge, MA), T (Mallinckrodt Diagnostica GmBH, Dietzenbach, Germany), and E2 (Baxter Merz+Dade AG Vnterschleissheim, Germany) were determined by commercially available kits without prior extraction. The sperm counts were performed after a 3 to 5-day period of abstinence. The ejaculate was collected by masturbation in the hospital, and analysis followed immediately by standard methods.!7 Sperm density was measured using the hemocytometer method. Statistical Methods Data are given as means ± SD unless indicated otherwise. Data were analyzed using Mann-Whitney V-test and Mantel-Cox test. Multiple means were analyzed with factorial ANOV A. Correlation was tested by t-distribution. Differences were considered statistically significant when P < RESULTS Endocrine Testicular Function Before therapy, no significant differences in age, testicular volume, LH, FSH, and T levels existed between both groups of patients (Tables 1 and 2). Testosterone increased in both groups in all patients except two (8 and 9) in the GnRH group. The mean T during therapy in the GnRH group was 16.8 ± 5.5 and in the gonadotropin group 22.5 ± 8.1 nmol/l. Three patients of the GnRH group had T values between 10.5 and 12 nmol/l and remained slightly below the normal range. The T levels during therapy in the gonadotropin group were significantly higher (P < 0.03) compared with the GnRH group. In the two patients of the GnRH group, in whom no T increase could be observed, an adequate LH and FSH increase after pulsatile GnRH application was demonstrated during the 4-hour LH and FSH profile. Furthermore, the T increase after 1M administration of hcg was normal excluding an additional Leydig cell defect. Failure of GnRH therapy in these two patients could therefore only be explained by insufficient compliance. For this reason, the two patients were excluded from statistical analysis. In both groups, an increase of the E2 levels could be observed (Fig. 1). Whereas the mean E2 in the GnRH group increased only slightly from 46 ± 22 to 88 ± 59 pmol/l, the increase in the gonadotropin group was significantly more pronounced (P < ) from 41 ± 11 to 179 ± 102 pmol/l, when the highest E2 levels after gonadotropin treatment were compared. The E2 levels were still significantly (P < 0.001) more augmented when the hcg dose was reduced (41 ± 11 to 150 ± 70 pmol/l). Only in one patient (3) of the GnRH group, E2 rose to a supraphysiological value of 292 pmol/l. No patient of the GnRH group developed gynecomastia. In the gonadotropin 1146 Schopohl et al. GnRH and gonadotropin therapy in males Fertility and Sterility

5 GnRH Gonadotropins Increase After dose reduction 3-18 months 6-27 months Figure 1 Estradiol levels before and on therapy from each patient of the GnRH and the gonadotropin-treated group. Values before therapy are the means of two samples. Values in the middle section represent the highest levels of each patient of the gonadotropin group during therapy. Values on therapy in the left and the right section, respectively, are the mean of the three last E2 levels obtained during therapy from each patient of the GnRH and the gonadotropin group. The shaded area represents the normal range. Closed circles are the mean value in each group (±SD). Estradiol increase was significantly higher in the gonadotropin group (P < 0.001). a, the two patients with lack of compliance. group, however, eight patients had an exaggerated E2 increase up to 439 pmol/l, and five of them developed gynecomastia. After reduction of the hcg dosage, elevated E2 levels persisted in six of these patients (Fig. 1). Further reduction of hcg normalized E2 but was accompanied by a decrease of T levels < 7 nmol/l. Gynecomastia improved when E2 levels decreased but disappeared only when E2 levels had returned into the normal range. During the time period with the highest E2 levels, T levels were >35 nmol/l in all patients. Two patients with maximal E2 levels of 443 and 336 pmol/l during gonadotropin treatment, had later, while receiving T replacement, normal E2 levels of 91 and 102 pmol/l, respectively, with T levels> 35 nmol/l. Testicular Volume, Testicular Growth The increase of testicular size was significant in both groups. In the GnRH group, the increase of testicular volume (ll testicular volume 8.1 ml ± 2.0) was significantly higher (P < 0.001) than in the gonadotropin group (ll testicular volume 4.8 ml ± 1.8) (Fig. 2). Whereas in the GnRH group 10 patients had an increase in testicular volume of 8 ml or more, only 1 patient in the gonadotropin group showed an enlargement of 8 ml. All others had smaller increases in testicular size (Fig. 2). The same finding was made when patients with a pretherapeutic testicular volume :::0;; 3 ml (11 patients in the GnRH group and 10 patients in the gonadotropin group) were compared. Whereas the GnRH treated patients had an increase of testicular volume of 8.6 ml ± 1.7, the increase in the gonadotropin group was only 5.0 ml ± 1.3 (P < 0.001). The lower limit of 3 ml was chosen because it indicates complete lack of endogenous GnRH secretion,18 whereas a volume of 4 ml or more reflects residual endogenous GnRH comparable with GnRH secretion at the beginning of puberty. 19 The testes grew more rapidly during the 1st 6 months of therapy. The growth rate was maximal in the GnRH group during the 2nd month of treatment with 1.5 ± 0.9 ml per month and in the gonadotropin group during the 4th month of treatment with 0.7 ± 0.5 ml per month. Comparison of the growth rate over 15 months showed a significantly faster growth in the GnRH group (P < 0.001). Significant correlations were found between the initial testicular volume and the maximal testicular volume in both groups. The correlation coefficient in the GnRH group was 0.65 (P < 0.01) and in the gonadotropin group 0.82 (P < 0.001). Sperm Count In 14 patients of the GnRH group and in 17 patients of the gonadotropin group sperm counts were performed. In 10 patients (71 %) under pulsatile GnRH therapy sperm counts ranging from 1.5 to 14 X 10 6 spermatozoa/ml were achieved after 9.7 ± 5.6 months (range 4 to 18 months) of therapy (Table 1). In the gonadotropin group, 8 patients (47%) reached a positive sperm count ranging from 2 to 26 X 10 6 /ml after 14.9 ± 7.2 months (range 4 to 27 months) of treatment (Table 2). No correlation was GnRH Gonadotropins 12 (n) patients 12 :::i' 3-18 months (1) 6-27 months :::i' E n=16 n=18 E ;- 10 (4) 10 ;- E (3) E '0 " 8 (2) (1) 8 ~ " > (1) (2).Ii.Ii 6 (4) (4) 6 11 i (3) $I S S '0 4 (1) (3) 4 '0 II (3).. III ~ 2 (2) ".5 (2)8 0,,,, 0 before on before on _therapy therapy Figure 2 Individual increases of testicular volume of each patient from both groups. Numbers in parenthesis are numbers of patients with the indicated testicular increase. Testicular increase was significantly higher in the GnRH group (P < 0.001). a, the two patients with lack of compliance. Vol. 56, No.6, December 1991 Schopohl et al. GnRH and gonadotropin therapy in males 1147

6 found between the elevated E2 levels and the failure Percentage of positive sperm counts to induce spermatogenesis. 100 _ GnRH (n=14) In the GnRH group, the mean duration of treatment until positive sperm counts were achieved was 90 - Gonado1ropins (n=17) ± 1.6 months; in the gonadotropin group, however, mean duration was 20 ± 2.3 months. This was 60 Ip < 0.02 I a significant difference (P < 0.02) (Fig. 3). % Of the patients with a testicular volume ::;;3 ml 40 cr- ~- in the GnRH group 7 of 10 (70%) had a positive CP " sperm count, whereas in the gonadotropin group 20 only 3 of 10 (30%) had evidence of spermatogenesis. _~-a--i The difference in time until reaching a positive O~~~~--,--,---r--.--,--,--. sperm count was still more pronounced with 13.5 o ± 1.8 months for the GnRH group and 25.8 ± 0.8 months months for the gonadotropin group (P < 0.001). Two Figure 3 Percentage of positive sperm counts during treatment. patients (1 and 24) achieved pregnancy of their The mean time until initiation of spermatogenesis was significantly shorter (P < 0.02) in the GnRH group compared with the spouses and two children were delivered. In the other gonadotropin group (12 ± 1.6 versus 20 ± 2.3 months). patients, pregnancy was not a therapeutic goal at the time of therapy. DISCUSSION All patients studied had primary idiopathic hypothalamic hypogonadism, i.e., none of them had undergone normal puberty because the hypothalamic defect in GnRH secretion existed since birth.12 Only a few papers can be found in which results of treatment of a larger number (~8) of idiopathic hypothalamic hypogonadism patients are reported.3,6,12,13,16,20,21,22 Often a heterogeneous group of hypogonadotropic patients with delayed spontaneous puberty or onset of hypogonadotropic hypogonadism after puberty have been investigated.4,5,14 In the latter, normalization oftesticular function is easier to achieve.4 Our results concerning normalization of endocrine testicular function with GnRH are in good agreement with the literature. Thus Spratt et al 13 treated 22 patients successfully, and similar data are reported by others.14,16 Shargil22 reported higher mean T levels in 10 patients successfully treated with GnRH, which might be because of different patient selection. None ofthe patients had Kallmann's syndrome, and a large number of patients were pretreated with gonadotropins. Morris and colleagues12 found only normalization of T levels in 5 of 10 GnRH-treated patients with idiopathic hypothalamic hypogonadism. These differences with our results may also relate to patient compliance, which is of primary importance when GnRH is self-administered. There are few data of E2 changes during GnRH therapy. A small physiological increase was observed by Spratt.13 Liu et al16 found supranormal E2 levels in 2 of 5 patients after 12 months of GnRH therapy using a higher GnRH dose (median dose of 30 ~g/pulse) and documenting elevated LH levels in their patients, which might have been responsible for the elevated E 2. Testosterone and E2 responses in the gonadotropin group were significantly higher than in the GnRH group. Mean T levels of the former group were within the range reported in the literature.4,6,16 Liu et al16 also observed elevated E2 levels in 3 of 11 patients in the gonadotropin group, although mean E2 levels were only slightly elevated compared with the GnRH group. Burris and co-workers6 measured E2 levels during gonadotropin therapy, observing supraphysiological E2 values in 7 of 22 patients in agreement with our data. In contrast to Liu et al,16 they treated also patients with initial testicular volume> 3 ml. No data on the development of gynecomastia were reported. The higher T and E2 levels after gonadotropin therapy are probably because of testicular overstimulation. It is unlikely that elevated E2 levels are because of peripheral conversion of T because they were not observed in our 2 patients on T replacement leading to comparable elevation of T levels after T injection. Furthermore, Winters and Troen23 showed that T and E2 are secreted concordantly under LH control by the human testes. Because gonadotropin stimulates testicular aromatase activity,24 elevated E 2levels reflect an exaggerated increase in enzyme activity not usually observed during GnRH therapy. The high E 2levels may be an equivalent to ovarian overstimulation observed in women after gonadotropin, al Schopohl et al. GnRH and gonadotropin therapy in males Fertility and Sterility

7 though the latter has less clinical importance in males compared with female patients. The increase in testicular volume was markedly different in our GnRH- and gonadotropin-treated patients. Increase of testicular volume and velocity of testicular growth were significantly higher during GnRH therapy. This finding is confirmed by Liu et al.15 in four patients with sufficient compliance after previous gonadotropin therapy. The same group16 reported that the difference in testicular growth between GnRH and gonadotropin-treated patients was minimal. However, surprisingly mean testicular volume in the GnRH group decreased between 12 and 18 months of therapy, suggesting a possible intermittent lack of compliance in some of their GnRH-treated patients. The testicular increase of our GnRH group was in good agreement with the report of Spratt et al.13 Similar increases of testicular volume after gonadotropins compared with our gonadotropin group were seen by Ley and Leonhard5 and by Burris et al.6 after heg alone. The good correlation between the initial testicular volume and the testicular volume obtained during therapy was also seen by others.6,13 Our results showed a lower degree of sperm-positive patients, particularly in the gonadotropin group, compared with other authors.5,6,13,21 These differences may be because of longer duration of therapy in the other studies,5,6,13,21 because several of our patients finished therapy (Tables 1 and 2) before reaching the mean time of treatment until a positive sperm count appeared (Fig. 3). The reason for premature termination of therapy was mostly because of the young age of these patients in both groups. Because they did not intend pregnancy in their spouses or girl friends at present, they preferred T substitution after achieving a significant increase in testicular volume. However, there were some cases in both groups, in whom failure to initiate spermatogenesis remained unclear because they had reached a sufficient testicular size (3, 12, and 26). Probably only testicular biopsy could reveal the cause of failure of therapy in these patients and if there would be a chance to induce spermatogenesis with prolonged therapy. The most important difference between both therapies observed by us was the significantly faster initiation of spermatogenesis in the GnRH group (Fig. 3). This is in contrast to the results observed by Liu et au6 The comparison of the patients with an initial testicular volume of ~3 ml (comparable with the selection of patients by Liu et al.16) showed an even more pronounced difference in the time until spermatogenesis was achieved. Because no correlation between elevated E2 levels and the lack of sperm count in the gonadotropin group could be determined, this cannot be the reason for the difference between the two groups. Even though therapy in several patients was too short, this cannot explain the difference between both groups because the mean time of treatment in the gonadotropin group was considerably longer. The small number of patients reported by Liu et au6 who were treated with GnRH > 1 year (4 patients) and the fact that these patients were significantly older than the patients in their gonadotropin group may explain the differences with our results. New findings showing that 12 hours after 1M injection of 150 IU hmg no more FSH bioactivity could be found25 suggest that the injection of hmg two, three, or four times weekly may be inadequate and may explain the differences between gonadotropin and GnRH treatment. Summarizing our results, it can be stated that GnRH therapy led more rapidly to spermatogenesis compared with gonadotropin treatment, especially in patients with testicular volume ~ 3 ml before therapy. Furthermore, GnRH therapy did not lead to gynecomastia, which was frequently observed during gonadotropin therapy. On the other hand, GnRH therapy requires good patient compliance and technical understanding because patients have to manage most ofthe technical part ofthe therapy by themselves, which is not needed when patients are treated with gonadotropins. Because not all patients with idiopathic hypothalamic hypogonadism are able or willing to carry an infusion pump for approximately a year or more, gonadotropin therapy remains an effective alternative treatment. Acknowledgments. We gratefully acknowledge Wolf Bernhard Schill, M.D., and Eva Maria Parsch, M.D., (Dermatologische Klinik, Klinikum Innenstadt, Mtinchen, Germany) for performing part of the sperm counts and August Konig, M.D., (Medizinische Klinik, Klinikum Innenstadt, Mtinchen, Germany) for his help in the statistical evaluation. REFERENCES 1. Nowakowski H, Lenz W: Genetic aspects in male hypogonadism. Recent Prog Horm Res 17:53, Kallmann F, Schonfeld W A, Bariere SE: The genetic aspects of primary eunuchoidism. Am J Med 28:203, Lunenfeld B, Mor A, Mani M: Treatment of male infertility. 1. Human gonadotropins. Fertil Steril 18:581, 1967 Vol. 56, No.6, December 1991 Schopohl et al. GnRH and gonadotropin therapy in males 1149

8 4. Finkel DM, Phillips JL, Snyder PJ: Stimulation of spermatogenesis by gonadotropins in men with hypogonadotropic hypogonadism. N Engl J Med 313:651, Ley SB, Leonard JM: Male hypogonadotropic hypogonadism: factors influencing response to human chorionic gonadotropin and human menopausal gonadotropin, including prior exogenous androgens. J Clin Endocrinol Metab 61:746, Burris AS, Rodbard HW, Winters SJ, Sherins RJ: Gonadotropic therapy in men with isolated hypogonadotropic hypogonadism: the response to human chorionic gonadotropin is predicted by initial testicular size. J Clin Endocrinol Metab 66:1144, Belchetz PE, Plant TM, Nakai Y, Keogh EJ, Knobil E: Hypophysial responses to continuous and intermittent delivery of hypothalamic gonadotropin-releasing hormone. Science 202:631, Leyendecker G, Wildt L, Hansmann M: Pregnancies following intermittent (pulsatile) administration of GnRH by means of a portable pump (Zyklomat): a new approach to the treatment of infertility in hypothalamic amenorrhea. J Clin Endocrinol Metab 51:1214, Mason P, Adams J, Morris DV, Tucker M, Price J, Voulgaris Z, Van der Spuy ZM, Sutherland I, Chambers GR, White S, Wheeler MJ, Jacobs HS: Induction of ovulation with pulsatile luteinizing hormone-releasing hormone. Br Med J 288:181, Hurley DM, Brian R, Outch K, Stockdale J, Fry A, Hackman C, Clarke I, Burger HG: Induction of ovulation and fertility in amenorrheic women by pulsatile low-dose gonadotropinreleasing hormone. New Engl J Med 310:1069, Hoffman AR, Crowley WF: Induction of puberty in men by long-term pulsatile administration of low-dose gonadotropinreleasing hormone. New Engl J Med 307:1237, Morris DV, Adeniyi-Jones R, Wheeler M, Sonksen P, Jacobs HS: The treatment of hypogonadotropic hypogonadism in men by the pulsatile infusion of luteinising hormone-releasing hormone. Clin Endocrinol (Oxf) 21:189, Spratt D, Finkelstein JS, O'Dea LSL, Badger TM, Rao PN, Campbell JD, Crowley WF: Long-term administration of gonadotropin-releasing hormone in men with idiopathic hypogonadotropic hypogonadism. Ann Intern Med 105:848, Aulitzky W, Frick J, Galvan G: Pulsatile luteinizing hormonereleasing hormone treatment of male hypogonadotropic hypogonadism. Fertil Steril 50:480, Liu L, Chaudhari N, Corle D, Sherins RJ: Comparison of pulsatile subcutaneous gonadotropin-releasing hormone and exogenous gonadotropins in the treatment of men with isolated hypogonadotropic hypogonadism. Fertil Steril 49:302, Liu L, Banks M, Barnes KM, Sherins RJ: Two-year comparison of testicular responses to pulsatile gonadotropin-releasing hormone and exogenous gonadotropins from the inception of therapy in men with isolated hypogonadotropic hypogonadism. J Clin Endocrinol Metab 67:1140, World Health Organization: WHO Handbuch zur Laboruntersuchung des menschlichen Ejakulats und der Spermatozoen-Zervikalschleim-Interaktion. Schattauer Verlag, Spratt DI, Carr DB, Merriam GR, Scully RE, Rao PN, Crowley WF: The spectrum of abnormal patterns of gonadotropinreleasing hormone secretion in men with idiopathic hypogonadotropic hypogonadism: clinical and laboratory correlations. J Clin Endocrinol Metab 64:283, Zachmann M, Prader H, Kind HP, Haefliger H, Budliger H: Testicular volume during adolescence; cross-sectional and longitudinal studies. Helv Paediatr Acta 29:61, Lieblich JM, Rogol AD, White BJ, Rosen SW: Syndrome of anosmia with hypogonadotropic hypogonadism (Kallmann syndrome). Am J Med 73:506, Burris AS, Clark RV, Vantman DJ, Sherins RJ: A low sperm concentration does not preclude fertility in men with isolated hypogonadotropic hypogonadism after gonadotropin therapy. Fertil Steril 50:343, Shargil AA: Treatment of idiopathic hypogonadotropic hypogonadism in men with luteinizing hormone-releasing hormone: a comparison of treatment with daily injections and with the pulsatile infusion pump. Fertil Steril 47:492, Winters SJ, Troen P: Testosterone and estradiol are co-secreted episodically by the human testis. J Clin Invest 78:870, Forest MG, Lecoq A, Saez JM: Kinetics of human chorionic gonadotropin-induced steroidogenic response of the human testis. II. Plasma 17 a-hydroxyprogesterone, delta-4-androstenedione, estrone, and 17 ~-estradiol: evidence for the action of human chorionic gonadotropin on intermediate enzymes implicated in steroid biosynthesis. J Clin Endocrinol Metab 49:284, Jockenhovel F, Fingscheidt U, Khan SA, Behre HM, Nieschlag E: Bio and immunoactivity of FSH in serum after intramuscular injection of highly purified urinary human FSH in normal men. Clin Endocrinol (Oxf) 33:573, Schopohl et al. GnRH and gonadotropin therapy in males Fertility and Sterility