ACCORDING to Scorer (1) undescended testes are

Transcription

1 X/88/ $02.00/0 Journal of Clinical Endocrinology and Metabolism Copyright 1988 by The Endocrine Society Vol. 66, No. 1 Printed in U.S.A. Hormonal Evaluation of Boys Born With Undescended Testes During Their First Year of Life* S. M. P. F. DE MUINCK KEIZER-SCHRAMA, F. W. J. HAZEBROEK, S. L. S. DROP, H. J. DEGENHART, J. C. MOLENAAR, AND H. K. A. VISSER Department of Pediatrics, Division of Endocrinology (S.M.P.F.D.M.K.S., S.L.S.D., H.J.D., H.K.A.V.), and the Department of Pediatric Surgery (F. W.J.H., J.C.M.), Erasmus University Medical School, and University Hospital/Sophia Children's, Rotterdam, The Netherlands ABSTRACT. We studied pituitary-gonadal function during the first year of life in 48 boys born with 56 undescended testes in order to test the hypotheses that functional insufficiency of the hypothalamo-pituitary-gonadal axis and disorders of testosterone (T) biosynthesis occur in such boys. Cryptorchidism persisted for longer than 1 yr in 29 boys (30 testes; group I), whereas spontaneous descent occurred in 19 boys (20 testes; group II), in 6 after the sixth month. A control group (group III) included 160 boys. Basal and peak LHRH-stimulated serum LH and FSH and basal serum T values were determined at 3, 6, and 12 months. Serum T, dihydrotestosterone (DHT), progesterone (P), 1-hydroxypregnenolone, 1-hydroxyprogesterone, dehydroepiandrosterone sulfate, and androstenedione before and after hcg administration were determined at age 1 yr. Comparing the 3 groups, cross-sectional evaluation revealed no significant differences in basal or peak LHRH-stimulated serum LH and FSH levels, except that basal serum LH levels were slightly higher in group II than in group III. Comparing groups I and II, longitudinal evaluation revealed similar basal and peak LHRHstimulated serum LH and FSH values, with comparable changes with time. Basal serum T, DHT, and T precursor levels were similar in all three groups, with similar rises of T and DHT and variable minimal increases in androstenedione and dehydroepiandrosterone sulfate after hcg stimulation. We conclude that during the first year of life, boys with cryptorchidism have no functional insufficiency of the hypothalamo-pituitary-gonadal axis or disorders in T biosynthesis. (J CUn Endocrinol Metab 66: 159, 1988) ACCORDING to Scorer (1) undescended testes are found at birth in 4.3% of male infants (2.% for fullterm and 21% for preterm boys). During the first year of life, spontaneous descent may occur, so that approximately 1% of 1-yr-old boys have undescended testes. In normal male infants, there is a testosterone (T) surge in the first months of life, which subsequently subsides gradually (2). Serum T values during the first months of life have been reported to be lower in boys with undescended testes than in boys with either normal or delayed spontaneous testicular descent, leading to the hypothesis that cryptorchidism may be due to (transient) functional insufficiency of the hypothalamo-pituitary-gonadal axis (3, 4). Basal and LHRH-stimulated LH values also have been found to be lower in some cryptorchid boys than in boys with normal or delayed spontaneous testicular descent (5, 6). In contrast, Tapanainen and co-workers () found high postnatal serum T levels in cryptorchid boys, comparable to those in normal boys, and similarly aug- Received June 1,198. Address all correspondence and requests for reprints to: Dr. S. M. P. F. De Muinck Keizer-Schrama, Sophia Children's Hospital, P.O. Box 0029, 3000 LL Rotterdam, The Netherlands. * This work was supported by a grant from the Sophia Foundation for Medical Research (Project 40). mented serum LH responses to LHRH in the two groups of infants. These conflicting reports stimulated this study of boys born with undescended testes during the first year of life. The aims of the study were to ascertain whether there is a (transient?) functional insufficiency of the hypothalamo-pituitary-gonadal axis in infants with undescended testes and to investigate whether boys with undescended testes have any disorder(s) in the biosynthesis of T. Subjects and Methods Our study population consisted of 48 boys, with 56 undescended testes, referred to the Sophia Children's Hospital within their first year of life between October 1982 and January Forty-five of these boys were born after a fullterm pregnancy; the remaining 3 were born prematurely, 2 with a birth weight of 2000 g after, respectively, 33 and 34 weeks gestation, and 1 weighing 220 g after 36 weeks gestation. All 3 were unilaterally cryptorchid. In the entire group, cryptorchidism was unilateral in 40 boys (2 left-sided and 13 right-sided) and bilateral in 8. Twenty-four boys were less than 3 months old when first examined; of them, 20 had unilateral cryptorchidism ( left-sided and 9 right-sided), and 4 had bilateral cryptorchidism. All boys had a normally developed scrotum and penis. After referral, the boys were seen regularly in our out-patient 159 Downloaded from

2 160 DE MUINCK KEIZER-SCHRAMA ET AL. JCE & M 1988 Vol66«Nol clinic, if feasible, when they were 3, 6, 9, and 12 months old. At each visit, the spontaneous and most caudal testicular position was determined independently by the first 2 authors, as previously described (8). Delayed spontaneous testicular descent during the first year of life occurred in 19 boys (20 testes), as shown in Table 1. Of the 4 boys with bilateral cryptorchidism in this group, only 1 had spontaneous descent of both testes. The other 3 boys had hemidescent, i.e. only 1 testis descended (2 left-sided and 1 right-sided). Spontaneous delayed descent did not occur in any of the prematurely born boys. In 6 boys spontaneous descent occurred after the sixth month of life (4 left-sided, 1 rightsided, and 1 bilateral). Of the 24 boys entering the study within the first 3 months of life, 12 had spontaneous descent, including 2 with bilateral cryptorchidism who had hemidescent (1 leftsided and 1 right-sided). In 6 of these boys spontaneous descent occurred before the third month, including both boys with bilateral cryptorchidism who had hemidescent, while 3 others had spontaneous descent after the sixth month (2 left-sided and 1 right-sided). At the end of the first year, 10 of the 20 testes that descended spontaneously after birth had assumed a permanent scrotal position, including the six testes that had descended within the first 3 months of life and 2 of the 6 testes that had descended after the sixth month of life. The other 10 testes were retractile. Hormonal evaluation involved several tests, all carried out with the informed consent of the parents. At 3, 6, and 12 months the boys underwent an LHRH test; serum LH and FSH levels were determined before as well as 30 and 60 min after iv administration of 50 /zg LHRH. Basal serum T levels were determined at the same time. At the start of the second year of life, the boys underwent an hcg stimulation test; serum T, dihydrotestosterone (DHT), and the T precursors progesterone (P), 1a-hydroxypregnenolone (1OHPreg), 1a-hydroxyprogesterone (1OHP), dehydroepiandrosterone sulfate (DHEAS), and androstenedione (A 4 ) were measured before and 3 days after im administration of 1500 U hcg. A total of 160 boys with no endocrinological anomalies served as control subjects, with the informed consent of their parents. This group included 153 boys with normally descended testes who were in their first year of life. Basal serum T levels were TABLE 1. Delayed spontaneous testicular descent during thefirstyear of life Cryptorchidism Unilateral left Unilateral right Bilateral Boys 2 () 13(9) 8(4) Spontaneous descent Boys Testes 10(6) 5(4) 4 (2)* 10(6) 5(4) 5(2) Total 48 (24) 19 (12) 20 (12) d The numbers in parentheses refer to boys enrolled before the third month of life. 0 Three boys had hemidescent. 6 Both boys had hemidescent. c Six testes descended after the sixth month of life. d Six testes descended before the third month and three after the sixth month of life. determined in 144 of them from blood samples taken for routine blood tests. Nine boys who had an indwelling catheter for other reasons had a LHRH test. The remaining boys were in their second year of life and had retractile testes. They underwent an hcg test, as described above. For comparisons of the hormonal data, the study population was divided into 3 groups; group I consisted of 29 boys who remained cryptorchid, group II consisted of 19 boys who had delayed spontaneous descent (including hemidescents), and group III included all 160 control subjects. Laboratory determinations The serum hormonal analyses were carried out by the Pediatric Laboratory, headed by H. J. Degenhart, and the Department of Clinical Chemistry, headed by N. C. den Boer, both of Sophia Children's Hospital; and the Laboratory for Endocrinological Chemistry, headed by W. Schopman and W. H. L. Hackeng of Bergweg Hospital, Rotterdam. All measurements were made in duplicate. A 4 was measured by RIA after extraction with hexanetoluene (8:2, vol/vol) (9). The intra- and interassay coefficients of variation were 6.% and %, respectively, and the detection limit of the assay was 0.35 nmol/l. DHT was measured by RIA after extraction with Extrelut columns (Merck, Darmstadt, West Germany). The extract was purified by high pressure liquid chromatography, using a Waters high pressure liquid chromatographic apparatus (Waters Associates, Milford, MA) with a Merck Lichrosorb column (5 ixm; Si 60). Hexane-isopropanol (92:8, vol/vol) was used as eluent (0.8 ml/min). DHT was estimated using an antiserum obtained from Radioassay Systems Laboratories (RSL). The intra- and interassay coefficients of variation were 9% and 22%, respectively, and the detection limit of the assay was 0.13 nmol/l. 1OHPreg was measured by RIA after extraction and purification with Extrelut columns (Merck), using commercial kits (RSL). The intraand interassay coefficients of variation were % and 14%, respectively, and the detection limit of the assay was 0.9 nmol/ L. 1OHP was measured by RIA after extraction with Extrelut columns (Merck), using commercial kits from Institut National des Radioelements. The intra- and interassay coefficients of variation were 12% and 18%, respectively, and the detection limit of the assay was 1.2 nmol/l. DHEAS was measured by RIA, using kits obtained from RSL. The antibody used in the RIA was raised in rabbits against DHEA-S-3 monohemisuccinate-human serum albumin. The intra- and interassay coefficients of variation were 9% and 15%, respectively, and the detection limit of the assay was 0.02 /nmol/l. LH and FSH were measured by RIA using kits obtained from Byk-Mallinckrodt, Dietzenbach, West Germany. The standards employed in these kits have been calibrated for LH us. the Medical Research Council (MCR) standard 68/40 and for FSH us. the MRC standard 69/104, both expressed in international units per L. The intra- and interassay coefficients of variation for LH were less than 4% and 9%, respectively, and for FSH were less than 3% and 12%; the detection limits for LH and FSH were both 0.6 IU/L. P was measured by RIA, using kits obtained from Farinos Ltd. (Turku, Finland). The intra- and interassay coefficients Downloaded from

3 HORMONAL STATUS IN CRYPTORCHIDS' FIRST YEAR 161 of variation were 5% and %, respectively, and the detection limit of the assay was 0.5 nmol/l. T was measured by RIA after extraction by n-hexane-diethylether without chromatography. The antibody was raised against T-a-carboxyethylthioether bovine serum by the method of Pratt et al. (10). The intra- and interassay coefficients of variation were % and 22%, respectively, and the detection limit of the assay was 0.02 nmol/l. Statistical analysis To avoid making any assumptions concerning the distribution of the results, e.g. Gaussian distribution, distribution-free methods were used for significance testing. The cross-sectional hormonal data for the three groups were compared as follows. Serum T. Every six boys of group III, equal or successive in age, were taken together. Their median values with respect to age and T level were used as cross-median values (). These values were connected, forming the cross-median curve. The sign test was used for comparison of the T values of groups I and II with the cross-median curve of group III. In addition, the individual T values of the boys in group I were compared with the cross-median T curve of group II. The cross-median T values of group II were composed of four values determined at equal or successive ages. Serum LH and FSH. Cross-median serum LH and FSH values could not be determined for group III because of the small number of values. Consequently, the individual basal and peak LH and FSH values of groups I and III were compared with the cross-median curve of group II. Likewise, the individual LH and FSH values of boys in groups II and III were compared with the cross-median curve of group I. The sign test was used for these comparisons. The longitudinal data were compared as follows. Serum T. The Kruskal-Wallis test was used to compare the longitudinal T values of groups I and II and the cross-sectional T values of group III at 3,6, and 12 months of age. If the results were significant (P < 0.05), the Mann-Whitney U test was carried out as well. The changes in the longitudinal serum T values within groups I and II from 3 to 6 and from 6 to 12 months of age were compared using the Friedman test. In the case of significant results (P < 0.05), the Wilcoxon matched pairs signed rank test also was applied. Between-group results were compared using the Mann-Whitney U test. Serum LH and FSH. The longitudinal basal and peak LH and FSH values of groups I and II at 3, 6, and 12 months of age were compared with the Mann-Whitney U test. The changes in the longitudinal LH and FSH values within groups I and II from 3 to 6 and from 6 to 12 months of age were compared using the Friedman test. If the results were significant (P < 0.05), the Wilcoxon matched pairs signed rank test was applied. Between-group results were compared using the Mann-Whitney U test. Correlations. We investigated the correlations between basal or peak LH values and T as well as between the basal LH to T or the peak LH to T ratios and age for groups I and II. Spearman's correlation coefficient was used to calculate the correlations. hcg tests. The Kruskal-Wallis test was used to compare the results of the hcg tests for the three groups. In case of significance (P < 0.05), the Mann-Whitney U test was applied. In addition, the basal vs. peak values of the T precursors T and DHT were compared within each group using Wilcoxon matched pairs signed rank tests. Results There were no significant differences in hormonal values between boys with unilaterally or bilaterally undescended testes in groups I and II, possibly due to the small number of boys with bilateral cryptorchidism. Consequently, in the statistical analyses no distinction was made for unilateral or bilateral cryptorchidism in either group. Comparison of cross-sectional and longitudinal data Since cross-sectional values were based on one sample per patient, the first determination was used for boys in groups I and II. For the control group we had only single samples. Figure 1 shows the individual serum T values in the first year of life for 29 boys of group I, 18 boys of group II, and 144 boys of group III. Figure 2 shows the individual basal and peak LHRH-stimulated serum LH values in the first year of life for 29 boys of group I, 18 boys of group II, and 9 boys of group III. In both of these figures, boys with unilateral and bilateral cryptorchidism T nmol/l ,v * A 1 Persistently cryptorchid boys II Boys with delayed spontaneous descent III Controls." * A '.TA' 13 " A *A A uni A bi uni bi 0 hemi 0 A " > "Jkf ja» *»'AA, M20 Age [days] FIG. 1. Cross-sectional serum T concentrations of boys in groups I, II, and III. n-25 n-t n-lt n-l n-3 n-m 1 * Downloaded from

4 162 DE MUINCK KEIZER-SCHRAMA ET AL. JCE & M 1988 Vol 66 No 1 (A) LH IU/ eo * 1 Persis tently c ryptorchid boys II Boys with del desce III Controls ayed spontaneous uni A bi n-25 n-1 uni n-im o bi El hem n-3 n-9 longitudinal change in peak FSH values in either group, which remained the same throughout the first year of life. Correlations There were no significant correlations between crosssectional basal or peak serum LH and T levels in the first year of life from the 80th day in group I or II. The Spearman test revealed a highly significant positive correlation between the basal LH to T or peak LH to T ratios and age in the first year of life in groups I and II (P < 0.005). Both groups had an age-related increase in the basal LH to T and peak LH to T ratios. 60 LH IU/ Age [days] FlG. 2. Cross-sectional basal (A) and peak LHRH-stimulated (B) serum LH concentrations of boys in groups I, II, and III. are shown separately, as are boys with bilateral cryptorchidism who had spontaneous hemidescent. The individual T values of boys in groups I and II were compared with the cross-median T curve of group III. In the same way, the individual T values of boys in group I were compared with the cross-median T curve of group II. The sign test revealed no significant differences in the three groups from the age of 80 to 180 days, 180 to 360 days, or at any time during the entire study period. Likewise, there were no significant differences in basal or peak serum LH values in the 3 groups, except for a higher basal LH level in group II than in group III. Almost all (41 of 55) basal serum FSH values of boys in the 3 groups were below the detection limit of the FSH assay (0.6 IU/L). Consequently, no evaluation of basal FSH was carried out. There were no significant differences in peak serum FSH values in the 3 groups. Comparing groups I and II longitudinally, there were no significant differences in serum T, basal and peak LH, and peak FSH values. There was a significant fall in serum T values in both groups, with a marked drop from the 3rd to the 6th months of life. From the 6th to the 12th months there was little or no change in values. Both groups had significant declines in basal and peak LH values from the 3rd to the 6th months, and these declines continued up to the 12th month. There was no hcg tests Table 2 shows the basal and peak serum (median and range) T, DHT, and T precursor concentrations before and after im administration of 1500 IU hcg at 1 yr of age for boys in group I and boys each in groups II and III. For technical reasons, 1OHPreg was not determined in all boys of group II or in any boy of group III. After hcg stimulation, serum T and DHT levels increased significantly in all 3 groups (P < 0.05), whereas serum P and 1OHP did not increase in any group, and 1OHPreg did not increase in groups I and II. Serum DHEAS increased in groups I and III, and serum A 4 increased in groups I and II (P < 0.05). A comparison of the 3 groups revealed no significant differences in basal serum 1OHP, A 4, T, or DHT levels among the 3 groups or in basal 1OHPreg levels between groups I and II (small number of boys). Basal P and DHEAS were slightly (P < 0.05) higher in group II than in group I. Peak serum P, 1OHP, DHEAS, T, and DHT levels were similar in all groups. Peak serum A 4 levels were slightly (P < 0.05) higher in group II than in group III. Peak serum 1OHPreg levels were similar in groups I and II. Discussion All boys with spontaneous testicular descent during the first year of life were born after a fullterm pregnancy. Such delayed spontaneous descent of the testis occurred in half of the boys that were enrolled in the study within the first 3 months of life and in half of them before the third month. According to Scorer (1, 12) and Forest et al. (13), spontaneous testicular descent rarely occurs after the sixth month of life, but this did occur in 6 of 19 boys with spontaneous descent in the first year of life in our study. Villumsen and Zachau-Christiansen (14) were unable to time the occurrence of spontaneous descent in the first year of life, because they only saw the boys on the fifth day and at 1 yr of age. In agreement with Forest et al. (15,16), cross-sectional Downloaded from

5 HORMONAL STATUS IN CRYPTORCHIDS' FIRST YEAR 163 TABLE 2. Results of hcg tests for T precursors, T, and DHT in the three groups of boys P 10HPreg 1OHP DHEAS (nmol/l) (nmol/l) (nmol/l) (^mol/l) A 4 T DHT (nmol/l) (nmol/l) (nmol/l) Group 1 Basal Peak Group II Basal Peak Group III Basal Peak 1.1 < < ND ND 1.5 < < < < c O < ' d < <0.13 O <0.13 < <0.13 O n, Number of boys; ND, not done. " Significantly higher than basal values (by Wilcoxon matched pairs signed rank test, P < 0.05). 6 Significantly higher than basal P of group I (by Mann-Whitney U test, P < 0.05). c Significantly higher than basal DHEAS of group I (by Mann-Whitney U test, P < 0.05). d Significantly higher than peak A4 of group III (by Mann-Whitney U test, P < 0.05). serum T analysis revealed elevated T levels in the first 6 months of life in boys with descended testes. Noteworthy was the wide range, particularly in the first 3 months of life. However, in contrast with the findings of Gendrel et al. (6), serum T levels during the first year of life were similar in the boys who remained cryptorchid, the boys who had spontaneous descent, and the control boys. Likewise, basal and peak serum LH and peak FSH values were similar in the boys who remained cryptorchid and the boys who had spontaneous descent in the first year of life. The basal and peak LH and peak FSH levels in the small number of control boys were comparable with those in cryptorchid boys and boys with spontaneous descent, although the latter group had higher basal LH values than the control boys. Consequently, there were no signs of functional insufficiency of the hypothalamopituitary-gonadal axis in these boys with undescended testes. This lack of a pituitary-gonadal disturbance in the boys who remained cryptorchid through the first year of life is in conformity with our findings in a double blind, placebo-controlled study of LHRH nasal spray in cryptorchid boys aged 1-12 yr (8). Gendrel et al. (6) found positive correlations between serum T and LH levels in boys who remained cryptorchid and in those who had delayed spontaneous testicular descent. In contrast, we found no such significant correlations. Hammond et al. (1) found good correlations between serum T and gonadotropins in the male infant population, while Forest et al. (16) found no such correlations. We did find a statistically significant correlation between the basal LH to T ratio and age as well as the peak LH to T ratio and age in the first year of life for boys who remained cryptorchid as well as for boys who had spontaneous descent. Toward the end of the first year of life, the basal LH to T and peak LH to T ratios increased in the two groups. Serum T declined to prepubertal levels earlier than did basal and peak LH values. Forest et al. (2) reported similar results for basal LH and T. The following explanations can be offered. 1) High levels of maternal estrogens drop rapidly after birth. Likewise, serum T falls rapidly in the first week of life, probably due to the clearance of hcg. These two events may contribute to the rise of circulating gonadotropins, which may be responsible for the subsequent serum T rise (16). 2) The rise of serum T may also be due to postnatal reduction of estrogens, because there is ample evidence that estradiol directly inhibits T production by enzyme inhibition (18, 19). 3) The gradual decrease in serum T after the second month of life may be caused by a decreased response of the testes to endogenous gonadotropins. The decline of fetal Leydig cells in the infant after the first postnatal months may well be responsible for this decreased response. 4) The decreases in basal and LHRH-stimulated LH secretion in the first Downloaded from

6 164 DE MUINCK KEIZER-SCHRAMA ET AL. JCE & M 1988 Vol 66 No 1 year of life might be due to a sex steroid-dependent mechanism, whereby a low set-point for gonadostat sensitivity to circulating sex steroids results in suppression of gonadotropins due to low amounts of circulating steroids or a sex steroid-independent mechanism, whereby intrinsic central nervous system inhibitory influences suppress the secretion of gonadotropins. The lack of correlation between basal and stimulated serum LH and T values for boys who remained cryptorchid and boys who had spontaneous descent in the first year of life substantiates the second theory. In agreement, Winter (20) suggested that the reduced childhood secretion of gonadotropins might be the result of central inhibition of LHRH secretion. We found no clinical signs of a defect in T biosynthesis. All boys had normal external genital development, none had bilaterally impalpable testes, and none had any clinical signs of adrenal insufficiency. Basal serum T, DHT, and T precursor levels were similar in the boys who remained cryptorchid and the boys who had either normal or delayed spontaneous descent. hcg stimulation resulted in similar increases in serum T and DHT levels in all three groups. In adult men, serum 1OHP levels rise 24 h after a single injection of hcg and are still elevated at 2 h, indicating inhibition of C1,20-lyase activity (21). In the young boys in our study, T precursors increased minimally. Consequently, our hormonal data do not suggest the presence of enzymatic defects or testicular enzyme inhibition in boys born with undescended testes, while their physical appearance and the normal stimulated DHT values excluded 5«-reductase deficiency. Conclusions We conclude that 40-50% of testes that are undescended at birth will descend spontaneously during the first year of life. Such delayed spontaneous descent may even occur after the sixth month of life. Our findings do not indicate that boys with undescended testes have functional insufficiency of the hypothalamo-pituitarygonadal axis or any disorder of T biosynthesis. Acknowledgments The authors would like to thank Prof. R. van Strik for statistical advice, S. van Duin and A. Weterings for hormonal determinations, Mrs. P. G. van Brakel and Mrs. M. S. P. Zwakhals-Eldering for technical assistance, Mrs. A. M. C. Ribbink-Goslinga for editorial assistance, Miss H. Versprille and Mr. J. de Kuyper for preparing the tables and figures, and Miss T. 0. E. T. M. Russel for typing the manuscript. References 1. Scorer CG 1964 The descent of the testis. Arch Dis Child 39: Forest MG, De Peretti E, Bertrand J 196 Hypothalamic-pituitarygonadal relationships in man from birth to puberty. Clin Endocrinol (Oxf) 5: Gendrel D, Job JC, Roger M 198 Reduced post-natal rise of testosterone in plasma of cryptorchid infants. Acta Endocrinol (Copenh) 89:32 4. Gendrel D, Canlorbe P, Job JC, Roger M, Toublanc J E 199 Endocrine data in cryptorchid children. In Bierich JR, Giarola A (eds) Cryptorchidism. Academic Press, London, p Job JC, Gamier PE, Chaussain JL, Roger M, Scholler R, Toublanc JE, CanPorbe P 196 L'exploration des secretions gonadotropes et gonadiques. Arch Fr Pediatr 33:31 6. Gendrel D, Roger M, Job JC 1980 Plasma gonadotropin and testosterone values in infants with cryptorchidism. J Pediatr 9:21. Tapanainen J, Koivisto M, Huhtaniemi I, Vihko R 1982 Effect of gonadotropin-releasing hormone on pituitary-gonadal function of male infants during the first year of life. J Clin Endocrinol Metab 55: De Muinck Keizer-Schrama SMPF, Hazebroek FWJ, Matroos AW, Drop SLS, Molenaar JC, Visser HKA 1986 Double blind, placebocontrolled study of luteinising-hormone-releasing-hormone nasal spray in treatment of undescended testes. Lancet 1:86 9. Van Landeghem AAJ, Poortman J, Despande N, Di Martino L, Tarquini A, Thyssen JHH, Schwarz F 1981 Plasma concentration gradient of steroid hormones across human mammary tumor. J Steroid Biochem 14: Pratt JJ, Wiegman T, Lapphon RE, Woldring MG 195 Estimation of plasma testosterone without extraction and chromatography. Clin Chim Acta 59:33. Tukey JW 19 Explorative Data Analysis. Addison-Wesley, Boston, p Scorer CG 1981 The descent of the testis. In Davis JA, Dobbing J (eds) Scientific Foundations of Paediatrics, ed 2. Heinemann, London, p Forest MG, David M, Francois R, Treatment of cryptorchidism with HCG. National Symposium Cryptorchidism, Brussels, Belgium, 1984, p Villumsen AL, Zachau-Christiansen B 1966 Spontaneous alterations in position of the testes. Arch Dis Child 41: Forest MG, Cathiard AM, Bertrand JA 193 Evidence of testicular activity in early infancy. J Clin Endocrinol Metab 3: Forest MG, Sizonenko PC, Cathiard AM, Bertrand J 194 Hypophyso-gonadal function in humans during the first year of life. I. Evidence for testicular activity in early infancy. J Clin Invest 53: Hammond GL, Koivisto M, Kouvalainen K, Vihko R 199 Serum steroids and pituitary hormones in infants with particular reference to testicular activity. J Clin Endocrinol Metab 49: Yanaihara T, Troen P, Troen BR, Troen ML 192 Studies of the human testis. III. Effect of estrogen on testosterone formation in human testis in vitro. J Clin Endocrinol Metab 34: Jones TM, Fang VS, Landau RL, Rosenfield R 198 Direct inhibition of Leydig cell function by estradiol. J Clin Endocrinol Metab 4: Winter JSD 1982 Hypothalamic-pituitary function in the fetus and infant. Clin Endocrinol Metab : Forest MG, Roulier R 1984 Kinetics of the HCG induced steroidogenic reponse of the human testis V: 1-20 desmolase blockade is related to age or previous gonadotropin environment. Ann Endocrinol (Paris) 45:281 Downloaded from