Pediatr Surg Int (2005) 21: 240–254 DOI 10.1007/s00383-005-1382-0

R EV IE W A RT I C L E

C. Ong Æ S. Hasthorpe Æ J. M. Hutson

Germ cell development in the descended and cryptorchid testis and the effects of hormonal manipulation

Accepted: 18 January 2005 / Published online: 23 February 2005 Ó Springer-Verlag 2005

Abstract Germ cell development is an active process in normal testes during the first 4 years after birth, with transformation of the neonatal gonocytes into adult dark spermatogonia and then primary spermatocytes. The hormonal regulation of these changes is not fully understood, with evidence both for and against a role for gonadotrophins and androgens. Early surgical intervention in infancy aims to prevent or reverse germ cell maldevelopment. Although hormonal treatment for maldescent has been shown to be ineffective, there is still controversy over whether it may be useful as an adjunct to surgery to stimulate germ cells. Current evidence suggests that hormonal therapy may not stimulate transformation of neonatal gonocytes but may trigger prepubertal mitosis of primary spermatocytes. Further studies are required to determine the role of hormone treatment on germ cell development. Keywords Testis Æ Descent Æ Germ cell Æ Gonocyte Æ Spermatogonia

Supported in part by grants from the National Health and Medical Research Council of Australia. C. Ong F. Douglas Stephens Surgical Research Laboratory, Murdoch Children’s Research Institute, Melbourne, Australia S. Hasthorpe Department of General Surgery, Royal Children’s Hospital, Melbourne, Australia J. M. Hutson Department of Paediatrics, University of Melbourne, Melbourne, Australia J. M. Hutson (&) General Surgery, Royal Children’s Hospital, Parkville, Victoria, 3052, Australia E-mail: [email protected] Tel.: +61-3-93455805 Fax: +61-3-93457997

Introduction Cryptorchidism is a common condition, as yet imperfectly understood, with significant risks of infertility and malignancy [1]. Histopathological changes in normal and cryptorchid testis have contributed greatly to the elucidation of normal germ cell maturation and factors leading to disordered development in cryptorchidism. This review aims to summarise current knowledge of germ cell development in the normal and cryptorchid testis, endocrine factors, and the effects of hormonal manipulation. Normal germ cell development The normal human testis contains germ cells at various developmental stages plus supporting Sertoli cells in the seminiferous tubules and interstitial Leydig cells in the peritubular connective tissue. Germ cell development is a multistaged process starting in the fetus and completed only at puberty (Fig. 1) [2]. Beginning at 4–6 weeks after conception, the genital ridges organise to form the indifferent gonad. The primordial germ cells arise from endodermal cells of the caudal end of the yolk sac, migrate into the medullary cords of the gonadal blastema, and differentiate into gonocytes. These gonocytes transform to fetal spermatogonia from 10 to 22 weeks postconception [2]. At 3 months postnatally, fetal spermatogonia begin to transform via transitional spermatogonia subtypes into adult dark (Ad) spermatogonia (normally complete by 12 months of age). This coincides with a rise in luteinising hormone (LH), testosterone, and Mu¨llerian inhibiting substance (MIS), as well as Leydig cell proliferation. By 3–4 years of age, the Ad spermatogonia further transform to B spermatogonia and primary spermatocytes. The testis then remains quiescent until the final steps of spermiogenesis occur at puberty [3, 4].

241

Fig. 1 Germ cell development (modified from [2, 4])

Concomitant development of the Sertoli cells occurs with fetal Sertoli cells developing at 7 weeks after conception from mesenchymal cells in the gonadal blastema. Postnatally they transform to Sa and Sb subtypes and subsequently to Sc at puberty. Total numbers of Sertoli cells decrease steadily from birth to puberty. Leydig cells are prominent in early life, peaking at 3– 4 months and thereafter degenerating to minimal numbers by the age of 2 years [5]. Subsequently at puberty, another population of Leydig cells differentiate to adult Leydig cells [6]. Cryptorchid testis Histology Abnormal germ cell development in the cryptorchid testis appears to be an acquired condition. It results from failure of two critical postnatal maturation steps. First, there is delay in the first postnatal transformation, demonstrated by persistence of gonocytes beyond 6 months and decreased numbers of Ad spermatogonia [4, 5, 7]. Beyond infancy, quality as well as quantity of spermatogonia are affected because the spermatogonia present are predominantly fetal and have bizarre nuclei [2]. Subsequently, there is failure of the second transformation step of Ad spermatogonia to primary spermatocytes at 3 years of age [4]. There is associated maldevelopment of other cells in the cryptorchid testis. Leydig cell numbers are markedly reduced, reflecting lack of gonadotrophin stimu-

lation; hence they fail to generate the normal surge in testosterone at 60–90 days [4, 5]. Sertoli cell transformation is similarly affected, with early loss of fetal Sertoli cells and only partial Sa-to-Sc transformation at puberty [2]. The volume, diameter, and total length of seminiferous tubules in the cryptorchid testis are diminished compared with the normal one, and there is increasing peritubular connective tissue starting from the 2nd year of life [2]. Epididymal abnormalities are also well described with cryptorchidism, possibly related to the reduced local paracrine secretion of testosterone [2, 8]. This has implications for fertility because restoration of normal germ cell development by timely orchidopexy may still not result in normal spermiograms by unassisted means. The secondary degeneration of the cryptorchid testis is presumed to be due to the higher temperatures to which it is subjected (2–4°C higher than the normal scrotal temperature of 33°C) [9]. The reader is directed to an excellent review by Setchell [10] that summarises most of the animal work on the effect of heat on germ cell maturation, spermatogenesis, and Sertoli cells. The mechanism of heat stress is via a combination of direct and indirect effects on germ cell apoptosis, defective germ cell maturation resulting from heat shock proteins, reactive oxygen species, Sertoli cell damage, and epididymal gene products [10–12].

Implications for fertility Based on the studies showing abnormal histology developing in infancy, the recommended age for orchidopexy has been gradually lowered over the years [9, 13]. The percentage of normal spermatogonia in biopsies falls with increasing age of orchidopexy [14, 15]. Cortes et al. [16] report that the youngest patient in their series with no germ cells on biopsy is 15 months, supporting early orchidopexy. The effect of age at orchidopexy on sperm counts in adulthood has been evaluated [17–22], but results have been inconclusive because the age at orchidopexy in all these studies was over 2 years. Only one study [23] concluded that early orchidopexy in patients younger than 2 years was not necessarily essential, as their patients had similar outcomes when operated prior to 6 years old. Several problems with this study included use of testis size as a surrogate for fertility, a poor response rate (only half of those eligible underwent testicular examination), and failure to mention the percentage of intraabdominal testes in their younger age group. Sperm analysis on some of that cohort, which had a response rate of 34%, was reported in an earlier paper [24]. The researchers found that in a cohort of 51 men with previous orchidopexy at ages ranging from 10 months to 12 years, age at orchidopexy did not affect sperm counts in unilateral cryptorchidism, whereas in bilateral cryptorchidism, surgery before 4 years was necessary to preserve normal semen quality.

242

There are drawbacks of using sperm counts as a surrogate marker for fertility because low sperm counts do not preclude paternity [25], and paternity studies indicate that unilateral cryptorchidism is not associated with diminished paternity [26]. However, most of the young men of this recent era of early orchidopexy have not yet attempted paternity, and paternity studies may also be confounded by difficulties in confirming biological fatherhood. Indirect evidence of the beneficial results of early orchidopexy has come from hormonal function studies that show an inverse relationship between age of orchidopexy and adult testosterone levels, suggesting subclinical impaired Leydig cell function [27]. Data from the same cohort studied have shown that while cryptorchid men had higher follicle stimulating hormone (FSH) levels and lower sperm density than age-matched controls, those in the cryptorchid group who underwent orchidopexy before age 2 years had more normal FSH and inhibin B levels [28]. The failure of transformation of gonocytes to spermatogonia from 3 months to 12 months of age has been shown to correlate with abnormal sperm counts in adulthood [29, 30]. The risk of poor sperm counts is independent of the age of surgery [31]. This risk can be predicted if the number of spermatogonia per tubule is less than 1% of that in age-matched controls [32]. Based on these findings, several authors have advocated adjuvant hormonal therapy after orchidopexy when there is an abnormal testicular biopsy (see later section on ‘‘Adjunctive treatment with LHRH and other therapies after orchidopexy’’). Finally, puberty does not induce antisperm antibodies in surgically treated cryptorchidism even with testicular fixation [33]. Hence, subfertility in cryptorchidism is not due to a trauma-induced autoimmune response.

dysgenesis syndrome has been proposed as a unifying condition that encompasses testicular and genital maldevelopment, rare genetic disorders, cryptorchidism, poor sperm quality, and testicular cancer [42]. The prevalence of CIS is estimated to be 2–3% of all patients with history of cryptorchidism [34, 43]. The relative risk of CIS is greatly increased (7.8) if there is testicular atrophy in the adult as well as history of cryptorchidism [44]. The relative risk of testicular cancer in each testis in bilateral cryptorchidism is higher than in unilateral cryptorchidism [34, 37]. The contralateral descended testis in unilateral cryptorchidism is also at slightly increased risk compared with the general population [37, 45]. Testicular neoplasia risk in cryptorchidism is higher in patients with intraabdominal testes, abnormal genitalia, or known abnormal karyotype [46]. All of these observations agree with the testicular dysgenesis syndrome hypothesis that an underlying genetic predisposition to abnormal germ cell development within an abnormal environmental or hormonal milieu results in malignant transformation. Recent evidence (Cortes, personal communication) suggests that the malignancy rate in ascended testis is no higher than in the normal population. This suggests that the crucial postnatal gonocyte to Ad spermatogonia transformation took place at normal scrotal temperatures in early infancy (Fig. 2). Men with a history of ascended testis still have subfertility with abnormal spermatogenesis, as subsequent heat-induced defective germ cell maturation in childhood is likely to occur with ascent. On the other hand, Rusnack et al. [47] found no difference in histology of ascended testis and primary undescended testis as both groups had equally depressed

Risk of testicular neoplasia The risk of testicular malignancy with cryptorchidism is well known and is estimated to be fourfold to eightfold [34–37]. Carcinoma in situ (CIS) is a common precursor of different testicular neoplasms [38]. One plausible hypothesis is that CIS develops during early development of the germ cell in the fetus. If gonocyte transformation to spermatogonia is inhibited, it results in pluripotent stem cells that develop into subsequent malignancy. Genetic aberrations increase the risk of malignancy, and environmental factors contribute via disturbance of the fetus’s hormonal milieu [38, 39]. This hypothesis is supported by epidemiological evidence, morphological studies, and comparison of cell-surface proteins in CIS and fetal germ cells [39, 40]. Using genome-wide gene expression profiling, substantial overlap has been found when comparing the gene expression profiles of CIS and embryonic stem cells [41], supporting the multipotency of CIS cells. Testicular

Fig. 2 Schematic diagram of hypothalamo-pituitary-gonadal axis

243

germ cell numbers. This could be explained by the relatively late age (7 years) of orchidopexy in their group. Of course, the controversy remains whether most ascended testes are misdiagnosed true cryptorchid testes [48]. Likewise for retractile testes that are reportedly associated with reduced sperm counts [49], abnormal spermatogonia per tubule was found in 40% of cases [50], but there are no reports of increased risk of malignancy. Swerdlow et al. [37] have suggested that biopsy itself is a risk factor for developing testicular malignancy. This is disputed by Moller et al. [35], who showed that their cohort of 830 routinely biopsied cryptorchid patients had no increased malignancy risk related to biopsy. The results of the earlier study, which had only a 9% biopsy rate, may have been confounded by selection bias in choosing which testes were biopsied. The evidence that an earlier age at orchidopexy reduces malignancy risk [36, 37,45, 51–53] is inconclusive. However, these studies all reviewed patients who underwent orchidopexy at more than 2 years of age, when defective germ cell maturation had already taken place with the potential formation of CIS cells. Because of the time lag before peak incidence of testicular tumours in the 3rd decade of life, it remains to be seen whether the much lower age at orchidopexy of the present patient cohort will change the prognosis. In a bid to identify CIS before invasion occurs, some [43] advise that all young men with a history of cryptorchidism be offered biopsy. Noninvasive screening methods for preinvasive testicular cancer using semen analysis are currently not specific. Testicular ultrasound, however, shows great promise as a screening test for men at higher risk of testicular malignancy before testicular biopsy [54]. CIS is associated with microcalcifications with a characteristic appearance on ultrasound. The positive predictive value for CIS with score 4 on ultrasound is 22%, while the negative predictive value is 98% [55]. A tenuous association has been reported between the presence of multinucleated spermatogonia (found in 8% of cryptorchid boys) and CIS [56]. The clinical significance of this is as yet unclear, but it suggests potential disordered germ cell maturation and subsequent malignancy risk.

Hormonal influence in normal testis Hypothalamus–pituitary–gonadal axis (Fig. 3) There is still controversy regarding the roles of various hormones and their target structures in the normal descent of the testis. A primary deficiency of the hypothalamus–pituitary–gonadal axis is proposed to affect normal descent [2]. This is supported by the normal postnatal surge in LH and testosterone at 3–4 months of age [5, 57, 58] that is deficient in cryptorchid boys [59]. Another view is that androgens influence testicular

Fig. 3 Genes involved in germ cell development based on rodent studies (modified from [111–116])

descent in a far more complex way and that androgen insufficiency in cryptorchidism is secondary to testicular maldevelopment. Because theories of endocrine factors involved in testicular descent are available in two recent reviews [1, 60], this discussion is confined to hormonal effects on germ cell development. Hypogonadotrophic hypogonadism is presumed to explain the reduced germ cell count and defective germ cell maturation seen in both the cryptorchid as well as, to a lesser degree, the contralateral descended testis [4]. Studies of anencephalic fetuses and experimental results of hypophysectomised animals demonstrate the development of a functional hypothalamo-pituitary-gonadal negative feedback system in utero [2]. Fetal human chorionic gonadotrophin (HCG) levels parallel maternal HCG levels with a peak at 17–19 weeks, while fetal LH levels peak later, at 20–25 weeks. The maturation of fetal Leydig cells and onset of testosterone secretion is stimulated by both gonadotrophins in turn. After birth, levels of LH and FSH remain high in the first 6 months of life then fall gradually, with another peak of LH/ testosterone at 4 years [2]. The specific effects of LH and testosterone are difficult to differentiate from FSH and

244

its target hormones due to complex interconnected pathways. Animal models created to separate the specific actions of FSH and LH [61] show that while FSHmediated pathways are important in postnatal mitotic and meiotic germ cell development, LH-mediated effects are required for the full complement of germ cells and final stages of spermatogenesis. Sertoli cells express both FSH receptors as well as androgen receptors. Because no androgen receptors have been identified in human germ cells, the LH/testosterone effect is thought to be mediated via Sertoli cells. De Gendt et al. [62] have created a mouse model with selective knockout of the androgen receptor on the Sertoli cell. This results in a mouse with normal male phenotype and descended testes, unlike the mouse with complete androgen receptor knockout that displays the androgen insensitivity phenotype. They have demonstrated that androgens act via the Sertoli cell in late spermatocyte to spermatid development. A similar mouse model demonstrated androgen involvement via Sertoli cells in later spermatogenesis, the round spermatid to elongation step [63].

Sertoli cells and germ cells [71, 72]. Early puberty is characterised by positive correlation of inhibin B, LH, and testosterone levels. In late puberty, this relationship diminishes, and inhibin B develops a negative correlation with FSH levels that is maintained in the adult, indicating a classic negative feedback control [73]. In adults, serum inhibin B levels directly correlate with sperm concentration and can indicate the integrity of seminiferous tubules in men with history of cryptorchidism [28]. Recently, serum inhibin B levels have been used to evaluate seminiferous tubule function in prepubertal cryptorchidism [58, 74–76]. Gonadotrophin stimulation has been found to increase serum inhibin B levels in one study [74] while having no influence in others that used more specific immunoassays [58, 76]. Christiansen et al. [75] propose that HCG increases serum inhibin levels only in the younger prepubertal testis because older mature Sertoli cells are unable to secrete both subunits of inhibin B.

Hormonal therapy in cryptorchid testis Mu¨llerian inhibiting substance MIS is a 140-kDa glycoprotein produced by the Sertoli cells and is responsible for regression of the Mu¨llerian ducts in the embryo. In the testis, MIS is also thought to be involved in early testicular differentiation, testicular descent, and germ cell development. Its role in germ cell development may be related to the postnatal transformation of gonocytes to type A spermatogonia [64, 65]. Using mouse testis organ cultures, it has been shown that MIS, not HCG, is required for this step [65]. There is a peak in serum MIS levels at 4–12 months in the normal male [66, 67], which is deficient in cryptorchidism [68]. Serum levels of MIS and testosterone in childhood have an inverse relationship after infancy, with MIS levels falling at puberty coinciding with the rise in serum testosterone levels [6]. Hence, during childhood when testosterone levels are too low to test reliably, MIS serves as a useful marker of testicular tissue, such as in the differential diagnosis of complex intersex disorders [69]. The paracrine effects of MIS on Leydig cell development are postulated to explain this inverse relationship [6]. MIS production is downregulated by FSH while the cleavage of MIS is modulated by HCG, LH, and testosterone at a posttranslational stage [70]. Although the clinical significance of the various MIS cleavage products has yet to be elucidated, it is possible that each has a specific biological action, explaining the multiple sites of MIS activity and interaction with gonadotrophins. Inhibin B Inhibin B is a peptide hormone with a and b subunits. In the child, both subunits are produced by Sertoli cells; in contrast, after puberty, inhibin B is jointly produced by

Efficacy The rationale for hormonal therapy in treating cryptorchidism is based on the hypothesis that cryptorchidism is related to a defect in the hypothalamic-pituitarygonadal axis [2]. HCG was first introduced in 1930. It has a similar action to LH in stimulating testosterone production. Problems with HCG administration include multiple intramuscular injections and significant side effects of virilisation and aggression. In 1974, intranasal gonadotrophin releasing hormone (GnRH) analogue became available for use. GnRH stimulates pituitary LH production to increase testosterone production from Leydig cells. GnRH also increases pituitary FSH production to stimulate Sertoli cells. Combined therapy with GnRH and HCG is suggested to improve results as FSH recruits LH receptors [77]. Enthusiasm for hormonal therapy has waned as initial promising results have not been verified. Pyorala et al. [78] reviewed the use of hormonal therapy for undescended testis in a systematic review of 33 trials from 1975 to 1990. The nonrandomised studies in this report showed higher success rates, suggesting selection bias. In the metaanalysis on the nine randomised controlled trials available, hormonal therapy was effective in only a fifth of cases: luteinising hormone releasing hormone (LHRH) 21% (CI 18–24%), HCG 19% (CI 13– 25%), and placebo 4% (CI 2–6%). If retractile testes were excluded, the success rate of LHRH was even lower (12%; CI 8–15%). Based on the data given for trials excluding retractile testis, the number needed to treat for LHRH was 14 and for HCG was 8. In long-term followup, 24% (CI 13–35%) of the testes that descended after LHRH treatment had relapsed. Subgroup analysis carried out on success rates based on the initial position of the testis found it less effective on abdominal and

245

inguinal compared with prescrotal and high scrotal testis. (This analysis included data from nonrandomised studies.) No significant difference in efficacy was found in patients older than or less than 4 years of age. Limitations of this metaanalysis were inclusion of studies with heterogeneous drug treatment protocols and suboptimal methodologies. A more recent study [79] reviewed all randomised trials until June 2003. This paper identified three randomised controlled trials [80–82] in addition to the nine trials described by Pyorala et al. [78]. Two metaanalyses were performed on the different hormonal treatment comparison trials. The metaanalysis on two randomised controlled trials comparing HCG versus GnRH reported success rates of 25% versus 18% and absolute risk reduction of 7% (CI 0.012–0.170). The metaanalysis on nine randomised controlled trials comparing GnRH versus placebo had success rates of 19% versus 5%. This review was also limited by the quality of available data, hence the very similar results to the earlier metaanalysis. Table 1 summarises randomised and nonrandomised trials published subsequent to Pyorala et al.’s metaanalysis [78], from 1991 to 2003, on the use of hormonal therapy for undescended testis. The overall success rates reported by the six randomised controlled trials range from 8% to 43%. The studies generally agree with Pyorala et al. [78] that therapeutic success is greater when the testis is in a lower position and age is not a factor. In addition, four trials reported that it is more effective in bilateral cases rather than unilateral ones [81–84]. Although the overall efficacy of hormonal treatment alone for cryptorchidism is poor, higher success rates of partial descent of undescended testis have prompted use of HCG to aid in preoperative localisation of the nonpalpable testis [85–87]. It has potential for decreasing the need for laparoscopy and vascular manoeuvres for intraabdominal testis such as the Fowler–Stephens procedure [85]. HCG has also been reported to be useful as an adjunct to identify retractile testis [88], although with too poor sensitivity (58%) and specificity (80%) to be reliable as a diagnostic test. Table 2 summarises the results of these four nonrandomised trials for the use of HCG in managing cryptorchid and retractile testes. The use of HCG stimulation tests to identify functioning testicular tissue in cases of intersex disorders has been superseded by development of reliable MIS assays, thus obviating the need for painful HCG administration and androgenisation problems [69]. Histopathological changes on HCG-treated testis A short-term inflammatory response after HCG associated with increased testicular blood vessel volume density has been noted [89, 90]. This effect disappears in biopsies taken 6 months or more after HCG treatment. This may translate to better vessel mobility and easier orchidopexy [85, 91], although it does result in increased

interstitial bleeding, especially with testicular biopsy [91]. The safety of prepubertal administration of HCG has recently been questioned by several investigators. Dunkel et al. [92] have shown that in the human testis, HCG causes increased apoptosis of spermatogonia. Their findings of smaller testis volume and raised FSH levels in adulthood suggest disruption of normal testicular development. Based on rodent studies, increased germ cell apoptosis is postulated to be due to androgen withdrawal due to the fall in testosterone after HCG treatment has ceased. Other investigators report similar findings of smaller testis size [23] and decreased spermatogonia per tubule [93]. Some conflicting studies report that the number of spermatogonia per tubule is increased [15, 90] or unchanged [80, 94] with HCG treatment. Cortes et al. [93] suggest that age of administration of HCG can explain these different observations. In their experience, hormonal treatment before 4 years of age did not cause descent of the testis. Because levels of gonadotrophins and testosterone in 1- to 3year-old boys are naturally low, hormone administration during this period may be detrimental to germ cells. Supporting evidence from rodent work suggests that HCG in the prepubertal rat results in decreased testosterone and haploid cell population in the adult [95]. In another study, HCG in prepubertal mice induces premature spermatogenesis with transformation of primary spermatocytes to round spermatids, a step that usually occurs at puberty [96]. It is uncertain what this portends for subsequent fertility, although instinctively it is worrisome to have this late step in spermatogenesis occurring in the prepubertal child. With poor efficacy of HCG for cryptorchidism and mounting evidence that it causes irreversible changes in germ cell maturation that are of unknown consequence, the routine use of HCG cannot be justified. Larger randomised controlled trials looking at the specific use of HCG for managing the nonpalpable testis will help decide whether the benefits outweigh the potential risks in this instance. Histopathological changes in LHRH-treated testis The histological evidence for LHRH’s effect on germ cells is less extensive than HCG as it has been available intranasally only since 1974 and was more expensive. Three trials that included testicular biopsies in recent years [80, 97, 98] (Table 1) concluded that LHRH treatment increased germ cell count per tubule. One nonrandomised trial [15] reported an improved spermatogonia number per tubule if combination LHRH/ HCG therapy and surgery were completed before 1 year of age. On the contrary, in another trial with boys under 4 years old [93], preoperative LHRH treatment decreased the number of spermatogonia per tubule. In addition, cultures of testicular biopsy with LH or FSH or a combination of both had reduced spermatogonia

Retractile testis excluded

Yes

No

Yes

Yes

Yes

Bica et al. [80]

Christiansen et al. [81]

Hoorweg-Nijman et al. [117]

Bertelloni et al. [118]

Esposito et al. [83]

Randomised control trials Olsen et al. [82] No

Study

Total HCG HCG+HMG GnRH GnRH + HCG Total

Total FSH + HCG Placebo+ HCG

Total HCG LHRH Placebo

Total LHRH + HCG Placebo + HCG Surgery

Total LHRH Placebo

Treatment groups

155 37 39 39 40 324 (418)

22 (28) 14 (18) 8 (10)

257 (420) 85 (133) 83 (140) 89 (147)

59 22 (25) 19 (19) 18 (21)

123 62 61

No. of patients (testes)

15 19 13 13 15 28

43 33 60

Bilateral 23 9 2

– 36 10 NAD

8 10 2

Success rate (%)

Table 1 Published trials on hormonal therapy for cryptorchidism 1991–2003

Unilateral 29 0 3

NR

23

NR

5

1

NR

Relapse rate (%)

X

No

Yes

No

X

Yes

X

X

X

No

Yes

X

Testis site

Factor

1. More effective in bilateral than in unilateral (38 vs. 23) 2. No difference in treatment groups except HMG alone ineffective

No difference among hormonal treatments

FSH no additional effect on testicular descent

Biopsy shows LHRH increases germ cell per tubule compared with placebo and surgery groups More effective in bilateral than unilateral

More effective in bilateral UDT 25%

Other study findings

1. No intention to treat analysis (18/ 141) excluded for protocol violation; however, corrected for in sensitivity analysis 2. NNT=12 1. No intention to treat analysis (4/63 excluded for protocol violation) 2. Complete biopsy rate No intention to treat analysis—73/ 330 excluded because of protocol violation 1. Small study, no power calculation 2. No control group without hormonal treatment 3. Noncomparable groups with regard to initial testis position 1. Only unilateral UDT evaluated 2. No control group without hormonal treatment 1. Randomisation suboptimal— according to six monthly periods 2. Nonpalpable testis excluded 3. No control group without hormonal treatment

Notes Age

246

Retractile testis excluded

Yes

Yes

Yes

Lala et al. [98]

Kaleva et al. [91]

Nane et al. [120]

Nonrandomised trials and case series Waldschmidt et al. [119] No

Study

GnRH + HCG

HCG No HCG

170 (202)

LHRH +HCG (5-year follow-up) LHRH (+ HCG for nonresponders)

48 (70)

182 (219) 29 (29)

238 (272)

79 (91)

113 35 85 27 64

HCG HMG LHRH HMG+HCG LHRH + HCG LHRH + HCG (7-year follow-up)

No. of patients (testes)

Treatment groups

53

52

38 (LHRH 11 + HCG 27)

72

89 (immediate) 50 (after 7 years)

35 0 29 26 30

Success rate (%)

16

NR

NR

25

43

Relapse rate (%)

Yes

X

Yes

X

Yes

Yes

X

X

Testis site

Factor

HCG causes increased interstitial bleeding and blood vessel volume density especially in UDT Subgroup analysis age 7–12 years most effective; abdominal/ inguinal location nonresponsive

Improved germ cell count on biopsy for hormonetreated under 12 months

Combination therapy more effective than LHRH monotherapy

Other study findings

Two groups based on length of follow-up (7 and 5 years, respectively). Paper compared success rates with historical trial with LHRH monotherapy Details lacking on this group Biopsy rate only about half of hormone-failed UDT, (82/169) 28 cases aged