The Regulation of Testicular Descent and the …The Regulation of Testicular Descent and the Effects...

The Regulation of Testicular Descent and the Effectsof Cryptorchidism

John M. Hutson, Bridget R. Southwell, Ruili Li, Gabrielle Lie, Khairul Ismail,George Harisis, and Nan Chen

F. Douglas Stephens Research Group (J.M.H., B.R.S., R.L., G.L., K.I., G.H., N.C.), Murdoch Children’s Research Institute,and Urology Department (J.M.H.), Royal Children’s Hospital, Melbourne 3052, Victoria, Australia; and Department ofPaediatrics (J.M.H., B.R.S., G.L., K.I., G.H., N.C.), University of Melbourne, Parkville 3010, Victoria, Australia

The first half of this review examines the boundary between endocrinology and embryonic development, withthe aim of highlighting the way hormones and signaling systems regulate the complex morphological changesto enable the intra-abdominal fetal testes to reach the scrotum. The genitoinguinal ligament, or gubernaculum,first enlarges to hold the testis near the groin, and then it develops limb-bud-like properties and migrates acrossthe pubic region to reach the scrotum. Recent advances show key roles for insulin-like hormone 3 in the first step,with androgen and the genitofemoral nerve involved in the second step. The mammary line may also be involvedin initiating the migration.

The key events in early postnatal germ cell development are then reviewed because there is mountingevidence for this to be crucial in preventing infertility and malignancy later in life. We review the recent advancesin what is known about the etiology of cryptorchidism and summarize the syndromes where a specific molecularcause has been found. Finally, we cover the recent literature on timing of surgery, the issues around acquiredcryptorchidism, and the limited role of hormone therapy. We conclude with some observations about thedifferences between animal models and baby boys with cryptorchidism. (Endocrine Reviews 34: 725–752, 2013)

I. IntroductionII. The Anatomical Steps in Testicular Descent

III. Hormonal Control of Testicular DescentA. Insulin-like hormone 3 (INSL3)B. Mullerian-inhibiting substance/anti-Mullerian

hormone (MIS/AMH)C. Androgen, genitofemoral nerve (GFN), and calci-

tonin gene-related peptide (CGRP)IV. The gubernaculum

A. Cremaster muscle theoriesB. Does the gubernaculum grow like a limb bud?C. The role of the mammary line

V. Germ Cell DevelopmentVI. Clinical Aspects of Cryptorchidism

A. EtiologyB. Cryptorchidism in congenital syndromesC. FrequencyD. Timing of surgery and the risk of infertility and

malignancyE. Acquired cryptorchidismF. Roles of hormone therapy

VII. Extrapolation of Animal Models to HumansVIII. Conclusions

I. Introduction

To understand the regulation of testicular descent andthe effects of cryptorchidism, it is first necessary to

examine the embryological remodeling that brings theintra-abdominal testis into a sc scrotum. In this review, wefocus initially on the links between the anatomical stepsand the governing hormones and downstream signalingpathways. We then describe in detail the key anatomicalstructure, the genitoinguinal ligament, or gubernaculum,and what we have learned recently about how it migratesto the scrotum. Before discussing cryptorchidism, we re-view the recent work in postnatal germ cell developmentbecause this looks to be central to why infertility and can-cer occur later in life.

ISSN Print 0163-769X ISSN Online 1945-7189Printed in U.S.A.Copyright © 2013 by The Endocrine SocietyReceived November 19, 2012. Accepted April 26, 2013.First Published Online May 10, 2013

Abbreviations: AD-S, adult dark-spermatogonia; AER, apical ectodermal ridge; AMH, anti-Mullerian hormone; AR, androgen receptor; Cdc42, cell division control protein 42; CGRP,calcitonin gene-related peptide; CREB-BP, cAMP response element binding protein-bind-ing protein; CSL, cranial suspensory ligament; DRG, dorsal root ganglia; FGD1, faciogenitaldysplasia protein 1; FGF, fibroblast growth factor; GFN, genitofemoral nerve; hCG, humanchorionic gonadotropin; Hox, homeobox transcription factor; INSL3, insulin-like hormone3; JNK, c-Jun N-terminal kinase; MIS, Mullerian-inhibiting substance; PMDS, persistentMullerian duct syndrome; PTPN11, protein-tyrosine phosphatase non-receptor type 11;PV, processus vaginalis; Ror2, receptor tyrosine kinase-like orphan receptor 2; RS, Robinowsyndrome; RTS, Rubinstein-Taybi syndrome; Tbx-3, T-box transcription factor 3; TS,trans-scrotal; UDT, undescended testis/testes; UMS, ulnar-mammary syndrome; Wnt,wingless/integrated.

R E V I E W

doi: 10.1210/er.2012-1089 Endocrine Reviews, October 2013, 34(5):725–752 edrv.endojournals.org 725

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 17 December 2015. at 03:24 For personal use only. No other uses without permission. . All rights reserved.

The clinical aspects of cryptorchidism are described,including a brief review of syndromes where a molecularmechanism has been determined, the timing of surgery,and the role of hormone therapy. We describe the recentrecognition of acquired undescended testes (UDT) andhow this is changing our approach to management. Weconclude with some comments on the use of animal mod-els and how these can be extrapolated to the humancondition.

II. The Anatomical Steps in Testicular Descent

The mechanism of testicular descent had been studied forcenturies, with intense interest in the anatomical conceptsin the 18th and 19th centuries and gradual understandingof the hormonal regulation in the 20th century. Currentconcepts of testicular descent generally describe testiculardescent as a 2-stage model, regulated by different anatom-ical and hormonal factors (1). The transabdominal stageoccurs at 10–15 weeks gestation in humans and gesta-tional days 13–17 in rodents. Before transabdominal mi-gration of the testes, the testes develop on the anteromedialsurface of the mesonephros in the urogenital ridge, whichis anchored in position by the cranial suspensory ligament(CSL) cranially and the genitoinguinal ligament or guber-naculum caudally (2). The “gubernaculum” (Latin forhelm or rudder) was first named in the 18th century by theScottish surgeon John Hunter, who hypothesized that itcould steer the testis into the scrotum (3). As the mesone-phros regresses at the onset of sexual differentiation, thegubernaculum becomes attached more directly to the testisand the Wolffian duct, which will form the epididymis andvas deferens.

Between 10 and 15 weeks gestation, the testes in thehuman fetus stay closer to the future inguinal canals thanthe ovaries, which move further from the groin (4). Therelative descent of testis close to the groin is achieved bytesticular enlargement as the mesonephros regresses,along with the gubernacular “swelling reaction.” The gu-bernaculum in the male undergoes cell division and anincrease in extracellular matrix, mainly glycosaminogly-cans and hyaluronic acid, to become bulky and gelatinous(5, 6). Its proximal connection to the testis, the guber-nacular “cord,” remains short. The enlarged distal guber-nacular end or bulb, which is embedded in the anteriorabdominal wall, exerts traction through the gubernacularcord on the urogenital ridge to help anchor the testis inposition during fetal abdominal growth (4) (Figure 1).Furthermore, the increase in gubernacular volume, undercontrol of insulin-like hormone 3 (INSL3) (see SectionIII), causes dilatation of the inguinal canal (7). Subse-

quently, CSL regression occurs under the influence of an-drogen (8, 9). Both of these factors facilitate subsequentmigration of the testes through the inguinal canal. TheCSL is a more important anatomical factor in rodent mod-els than in humans, where in males it is rudimentary (10).In females, it develops into the suspensory ligament of theovary.

After completion of the first phase of descent, there isa pause in testicular migration until about 25 weeks inhumans. The inguinoscrotal phase then occurs from25–35 weeks gestation in humans and from birth to post-natal week 3–4 in rodents. The overall process is similarin most mammals, although there are important differ-ences in gubernacular remodeling, migration, and elon-gation across the pubis to allow testicular descent into thescrotum. In rodents, the bulk of the gubernaculum is re-sorbed with extensive remodeling to allow the gubernac-ulum to evert before inguinoscrotal migration, whereas inpigs and humans there is no obvious eversion and resorp-tion occurs after scrotal migration (2, 11).

Many propositions have been made on the role of theprocessus vaginalis (PV) in the second part of descent. ThePV is a specialized peritoneal diverticulum that developswithin the gubernaculum. Several authors suggested thatPV formation occurred passively due to a weakness in the“inguinal triangle,” which allowed herniation of the testisalong with the PV (4, 12). This was thought to be aided inpart by intra-abdominal pressure. Investigations on maleinfants with abdominal wall defects demonstrated lowerintra-abdominal pressure during intrauterine develop-ment and higher incidence of cryptorchidism (13, 14),consistent with abdominal pressure having a role.

By contrast, there is evidence in rodents that the PVdevelops from specialized peritoneal epithelium that ini-tially covers the urogenital ridge in the early rodent em-bryo (15). The cells of the PV form a simple cuboidalepithelium, rather than the simple squamous epitheliumelsewhere in the peritoneal cavity. When the rodent gu-bernaculum remodels at the onset of the inguinoscrotalphase, most of the undifferentiated cells within the bulbmigrate out beyond the external oblique muscle, allowingthe gubernaculum to evert, from a solid, internal cone intoa hollow everted cone. A small, highly proliferative area ofmesenchyme at the cranial end of the cone becomes thedistal, growing end of the externally protruding and elon-gating gubernaculum.

It has been found that the gubernaculum is far from aninert ligament as previously thought, but actively elon-gates to the scrotum during the inguinoscrotal phase. Theprocess of inguinoscrotal descent in rodents can be de-scribed in a number of distinct steps (16, 17). This involves

726 Hutson et al Regulation of Testicular Descent Endocrine Reviews, October 2013, 34(5):725–752


initial gubernacular bulb eversion out of the abdominalwall followed by elongation and directional migration ofthe gubernacular tip toward the scrotum, resulting ineventual migration of the testis from the inguinal canal,across the groin and pubic bone, and into the scrotum.Development of the cremaster muscle is also observed inthe outer rim of the gubernaculum during this stage (18). Ithas been shown that the cremaster muscle in rodents hascontractile properties that resemble those of cardiac or em-

bryonic skeletal muscle and that allowrhythmic contraction, which appearsto facilitate testicular descent (4).This will be described in more de-tail in Section IV.A.

III. Hormonal Control ofTesticular Descent

A. Insulin-like hormone 3 (INSL3)In 1999, a novel testicular factor,

INSL3, was discovered as the keyhormone in the transabdominalphase of descent (1, 19–22). It is amember of the insulin superfamily ofstructurally related hormones andgrowth factors and is a secretoryproduct of Leydig cells expressed in adifferentiation-dependent manner.The protein is synthesized as a 131-amino-acid preprotein, which con-tains a 24-amino acid signal peptide(23). To confirm the relationship be-tween gubernacular developmentand INSL3, experiments were con-ducted in which fetal rat gubernac-ula were harvested and maintainedin organ cultures with dihydrotestos-terone, INSL3, Mullerian-inhibitingsubstance (MIS), or anti-Mullerianhormone (AMH) or controls (testiscoculture). The combination ofINSL3 and dihydrotestosterone ex-hibited the greatest effect on fetal gu-bernacular growth. Together, MIS/AMH and INSL3 also contributed togubernacular development, albeit toa lesser extent (24, 25). In addition,Nef and Parada (22) also demon-strated in INSL3 mutant mice bilat-eral cryptorchidism and abnormalgubernacula, which lacked a centralcore of mesenchyme at embryonic

day 16.5. This was further supported by a study that dem-onstrated the greatest INSL3 mRNA expression at gesta-tional day 17 in the fetal rat testis (25). Transgenic over-expression of INSL3 in female mice caused the swellingreaction to occur in the gubernaculum, leading to ovariandescent to near the bladder neck, similar to transabdom-inal testicular descent in males (27). Recent studies on thedownstream signaling pathways activated by INSL3 in the

Figure 1.

Figure 1. The 2 stages of testicular (T) descent in rodents and humans. A, Transabdominal phaseand inguinoscrotal phase in humans. The gubernaculum (G) swells in the first phase and thenmigrates to the scrotum in the second phase. Meanwhile, it becomes hollowed out by aperitoneal diverticulum, the PV (arrowhead). B, Testicular descent in rodents. The initial swellingreaction is very similar to the transabdominal phase in humans. However, just before the start ofthe inguinoscrotal phase, the mesenchyme of the rodent gubernaculum migrates out of bulb,allowing eversion. At the onset of the inguinoscrotal phase, the undifferentiated mesenchyme atthe apex of the intra-abdominal cone (crosshatch) is now at the distal tip after eversion andfunctions similar to a “progress zone” in a limb bud. The distal undifferentiated tip (which hasthe growth characteristics of the tip of a limb bud) becomes smaller as the rodent gubernaculummigrates to the scrotum.

doi: 10.1210/er.2012-1089 edrv.endojournals.org 727


gubernaculum show roles for the NOTCH and Wnt/�-catenin pathways (11, 28, 29).

INSL3 mutations in cryptorchid humans may be un-common because interruption of the transabdominalphase is an uncommon form of cryptorchidism (�10%)and documented INSL3 mutations in boys with UDT arerelatively rare (although the tested populations remainsmall). Recently, the putative receptor of INSL3 has beenidentified. LGR8 receptor, now known as relaxin familyreceptor 2 (RXFP2), was consistently expressed in Leydigcells of the testis and is responsive to INSL3 in fetal ratgubernacula at embryonic day 16. Mice lacking in RXFP2receptor were also found to have a similar phenotype toINSL3 knockout mice (23). Furthermore, molecularbiological studies were able to isolate a mutant Great (Gprotein-coupled receptor) gene, which prevented the en-coding of the LGR8 receptor, leading to bilateral intra-abdominal cryptorchidism in rodents (30, 31). On theother hand, mutation analysis of human Great gene in 61cryptorchid patients notably showed only a single mis-sense mutation among these patients (32).

Despite this, recent findings by Bay et al (7) demon-strated measurable levels of INSL3 in amniotic fluid ofhuman male fetuses at 15 weeks gestation, which wereabsent in female fetuses. The results from this study sug-gest that INSL3 plays a significant role in the gubernacularswelling reaction that is essential for transabdominal re-location of the testes in the first stage of descent of mam-mals. The role of androgen in this phase is minimal (apartfrom regression of the CSL) because previous studies onthe porcine gubernaculum failed to show an effect of an-drogen (34). INSL3 is required for expression of androgenreceptors (ARs) in the gubernaculum because these arerequired later for the gubernacular eversion at the onset ofthe inguinoscrotal phase (11).

The possible role of estrogen in cryptorchidism has alsobeen explored. In utero exposure of diethylstilbestrol topregnant rats is capable of inducing cryptorchidism in thefetus. Two likely explanations are that estrogens have ledto the suppression of INSL3 expression in fetal Leydig cellsand also caused feedback inhibition of the hypothalamic-pituitary-gonadal axis, leading to decreased LH concen-tration, and in turn affecting Leydig cell differentiation(19, 35).

B. Mullerian-inhibiting substance/anti-Mullerianhormone (MIS/AMH)

Mullerian inhibitory substance, also known as anti-Mullerian hormone, is a member of the TGF-� multigenefamily of glycoproteins and is produced by Sertoli cells(36). MIS/AMH was initially proposed to be one of themain factors involved in the transabdominal phase by

causing regression of embryonic Mullerian ducts in maleembryos (37). Failure of this event to occur in humansresulted in persistent Mullerian duct syndrome (PMDS)whereby the affected males have a persisting uterus andFallopian tubes, and frequently intra-abdominal testes.Although some authors suggested that the cause of cryp-torchidism in the PMDS patients was primarily the me-chanical restraint from the abnormal Mullerian ducts,others attributed it to other effects of the abnormal MIS/AMH gene or its receptor. The mutation was hypothesizedto interfere with the normal swelling and shortening of thecord of the gubernaculum because the gubernacular cordis found to be very long and thin, analogous to the roundligament of the female (38–40). However, many studieson transgenic rodents with MIS/AMH mutations failed todemonstrate UDT, suggesting other molecular involve-ment in cryptorchidism, at least in rodents (17, 40). It stillremains possible that MIS/AMH may have a role in keep-ing the gubernacular cord short in humans, which is dif-ferent from rodents.

Although MIS/AMH has limited action on testiculardescent, it can be used as a marker of Sertoli cell functionin the evaluation of children with cryptorchidism. Multi-ple research on boys aged 0 to 18 years have demonstratedhigh serum levels of MIS/AMH after birth, with a peakduring early infancy and a gradual decline by puberty (41,42). This corresponds with the timing of maturation ofSertoli cells in the developing testes. The MIS/AMH levelsare highest in normally descended testes, lower in cryp-torchid testes, and virtually absent in anorchia (43).Hence, in prepubertal boys with nonpalpable gonads, de-tectable levels of serum MIS/AMH suggest that the testesare potentially present (44).

C. Androgen, genitofemoral nerve (GFN), and calcitoningene-related peptide (CGRP)

The notion that androgen regulates the inguinoscrotalstage of descent of the testis came in the 1980s. Mice withcomplete androgen resistance (Testicular feminizingmouse) and absence of external virilization also had com-plete failure of the inguinoscrotal phase of testicular de-scent (1). Prenatal treatment (embryonic d 16–17) of ratswith the antiandrogen (flutamide) resulted in derangedgubernacular migration during the postnatal inguinoscro-tal period, with failure of downward growth of the PV,resulting in cryptorchidism in most rodents (45–47).

Although androgens regulate inguinoscrotal descent,the relative physiological role of testosterone vs dihy-drotestosterone remains poorly understood. The role of5-� reductase is much more important in early sexual dif-ferentiation (8–12 wk), to allow the masculinization of theexternal genitalia. At this time, the peripheral levels of



testosterone are not adequate without conversion to themore active metabolite, dihydrotestosterone—hence, theambiguous genitalia in babies with 46, XY disorders of sexdevelopment with mutations in 5-� reductase (48). Bycontrast, by the time of inguinoscrotal descent (25–35 wk)the requirement for conversion to dihydrotestosterone isless critical, so descent may be only partially affected (48).

Gonadotropins from the placenta (eg, human chorionicgonadotropin [hCG]) and the fetal pituitary (eg, LH andFSH) are important in regulating the testicular productionof androgens as well as INSL3 from the Leydig cells andMIS/AMH from the Sertoli cells (48), respectively. Recentevidence suggests that INSL3 is dependent on Leydig celldevelopment, which during the transabdominal phase(10–15 wk) is regulated by placental hCG. Once the Ley-dig cells have differentiated and become functional, INSL3production appears constitutive (35). Pituitary gonado-tropins are more important in the androgen-mediated in-guinoscrotal phase (25–35 wk), where anomalies of thehypothalamic-pituitary-gonadal axis may lead to andro-gen deficiency manifested by micropenis and cryptorchid-ism. In those boys with bilateral UDT, there are subgroupswith high gonadotropins, suggesting the lack of andro-genic negative feedback, but also some patients with lowgonadotropins, consistent with a primary anomaly of thehypothalamic-pituitary axis, as proposed by Hadziseli-movic (50) and Thorup et al (51).

MIS/AMH production is presumed to be under controlof hCG and/or FSH, but there have not been many studiesof this in the literature (36).

The timing of flutamide administration has turned outto be quite critical because it is only effective during alimited time window, which in the rat fetus is day 15 to 19(52). Beasley and Hutson (53, 54) first proposed that theeffect of androgen is indirect by stimulating the GFN torelease a specific neurotransmitter, CGRP. CGRP wasproposed to be a “second messenger” involved in ingui-noscrotal descent because transection of the GFN in neo-natal rats led to subsequent failure of the gubernaculum tomigrate into the scrotum. Secondly, studies conducted oncryptorchid pigs with the testes arrested in the line of de-scent demonstrated that exogenous CGRP stimulated mi-gration of the undescended inguinal testes, whereas ecto-pic and intra-abdominal testes remain arrested despite theintroduction of the neurotransmitter (55).

The GFN is a sexually dimorphic nerve with its sensorynucleus found in the L1–L2 dorsal root ganglia (DRG) ofthe spinal cord (56). CGRP-positive neurons are largerand more numerous in male animals, suggesting their im-portance in sexual differentiation, including possibly gu-bernacular migration and differentiation (56, 57). In asearch for the neuropeptide involved in this phase, CGRP,

which is a 37-amino acid peptide with numerous functions(58), was eventually identified in the sensory nucleus of theGFN (59).

Today, there is abundant evidence from experimentsin various rat models that demonstrates the key role ofCGRP in inguinoscrotal descent. Ablation of the sensorybranches of the GFN by a neurotoxin, capsaicin, leads todecreased production of CGRP. The lack of CGRP supplycauses up-regulation of CGRP binding sites in the guber-naculum, which in turn sensitizes the gubernaculum toexogenous CGRP (60, 61). Exogenous CGRP introducedto gubernacula in culture from rats pretreated with cap-saicin demonstrated a higher degree of proliferation in thegubernacular tip as compared to control rats at postnataldays 0–2 (60). By contrast, exogenous CGRP added togubernacula from flutamide-treated rats showed no in-crease in gubernacular proliferation, suggesting that an-drogens are required not only to masculinize the GFN butalso to preprogram the gubernacular proliferative re-sponse to CGRP (62). CGRP released from the sensorynerve terminals of the GFN induces rhythmic contractilityof the developing cremaster muscle of the gubernaculum.The rhythmic contraction may be important to orientatethe gubernacular tip toward the scrotum and assist gu-bernacular migration in the appropriate direction (63).Cultured neonatal mouse gubernacula demonstrated en-dogenous contractility as well as increased rhythmic con-tractions in response to exogenous CGRP that is mostprominent from embryonic day 17 to the first week post-natally (64, 65). This is in keeping with the timing of nat-ural inguinoscrotal migration in vivo.

In addition, CGRP released from sensory nerve termi-nals of GFN also provides a chemotactic gradient to stim-ulate the gubernacular migration toward the scrotum.Trans-scrotal (TS) rats are naturally occurring mutant ratsthat have excessive GFN fibers with increased CGRP-im-munoreactive neurons on the ipsilateral side of the UDT.In 85% of these rats, the gubernacula elongated laterallytoward the superficial inguinal pouch (66, 67). The mostlikely explanation is that the imprecise and excessiveCGRP release misdirected the gubernaculum away fromthe scrotum. The potent chemotactic effect of CGRP wasalso shown when exogenous CGRP deliberately injectedinto the suprapubic region led to reorientation of the gu-bernacular tip in a cranial direction (68). CGRP has alsobeen shown to reorient the tip of the rat gubernaculum invitro (63). A summary of the effects of CGRP is shown inTable 1.

Are the GFN nerve and its sensory neurotransmitterCGRP relevant for humans, or are the studies described inTable 1 only relevant to rodents? The evidence for a rolein humans is 2-fold. First, there is the general principle of



conservation of key developmental pathways during evo-lution. Given the evidence (listed in Table 1) for a key rolefor CGRP release from the GFN in both rats and mice, itis very reasonable to expect a similar process in humans.Certainly, elsewhere in medical research, extrapolationfrom rodent models to humans is the norm, unless there isspecific evidence of a species difference. Secondly, there isspecific evidence in a study we did some years ago showingthat exogenous CGRP can cause obliteration (in vitro) ofthe human inguinal hernia membranes (ie, patent PV) ex-cised at inguinal herniotomy in infants. Because the nor-mal closure of the PV is timed to occur after testiculardescent is complete (because the testis descends inside it),we wondered whether hernia closure might be controlledby the same mechanism controlling testicular descent. Therodent studies suggested that the GFN and CGRP wereimportant, so we tested the effect of CGRP on the patentPV of humans with a hernia. To our amazement, the PVobliterated within 1–2 days in organ culture (70). This isthe only study implicating CGRP in human testicular de-scent, but it is very intriguing and supportive.

IV. The Gubernaculum

A. Cremaster muscle theoriesIt has long been accepted that the gubernaculum is in-

dispensable in testicular descent. The relatively simple gu-bernacular swelling reaction facilitates transabdominaldescent of the testis, whereas the inguinoscrotal phase is amuch more complex process. It involves gubernacular re-modeling and eversion (at least in rodents), proliferation,

reorientation, and contractile responses to allow the testisto reach the scrotum inside the peritoneal diverticulum,the PV. The PV was initially recognized as an evaginationof embryonic coelomic cavity into the mesenchyme of thegubernaculum (3, 71). The PV divides the gubernaculuminto 3 parts: the outer rim containing cremaster muscle, theinner gubernacular cord that is attached to the testis andepididymis, and the distal gubernacular tip/bulb (18, 72).

The rhythmic contractility and reorientation of the gu-bernacular tip is largely influenced by the developing cre-master muscle. The origin of cremaster muscle has beendebated over decades. Many anatomical textbooks stilldescribe the cremaster muscle as an extension of the in-ternal oblique muscle (73, 74). Numerous theories havesuggested that rising intra-abdominal pressure during fe-tal development induced stretching of the internal obliquemuscle around the gubernaculum out from the abdominalwall to form the inguinal canal and cremaster muscle (71).However, Harnaen et al (18) demonstrated that the cre-master muscle was more differentiated proximally andmyoblasts more distinct distally near the gubernacularbulb and cord. Again, similar trends were observed byHrabovszky et al (75), who showed the distal gubernacu-lar bulb to have the highest rate of mitotic activity as com-pared to the gubernacular cord and cremaster muscle, andcremaster muscle proliferation was greater distally thanproximally. This implied that myogenesis and mesenchy-mal cell differentiation occurs distally from the guber-nacular tip to form cremaster muscle.

Furthermore, immature muscle fibers (myosin heavychain [Myh]-7, slow myosin, and cardiac troponin) pre-

Table 1. Effects of the GFN and CGRP on Testicular Descent in Rodents

Effect Refs.

Normal rat Neonatal GFN transaction blocks descent Lewis, 1948 (26); Beasley and Hutson, 1987 (54)GFN sensory nerves contain CGRP Schwindt et al, 1999 (61); Hrabovszky et al, 2000 (56)Gubernaculum with CGRP receptors Yamanaka et al, 1993 (33)Exogenous CGRP stimulates gubernacular contractions Park and Hutson, 1991 (57)CGRP chemotactic to gubernaculum in vitro Yong et al, 2008 (69)CGRP stimulates growth of gubernaculum Ng et al, 2009 (88)CGRP-stimulated growth augmented by androgen Shenker et al, 2006 (62)

Flutamide rat 50% UDT Shono et al, 1994 (47)Gubernaculum sensitized to CGRP in vitro Momose et al, 1992 (49)

Testicular feminizing mouse 100% UDT with no migration Griffiths et al, 1993 (59)Gubernaculum sensitized to CGRP in vitro Momose et al, 1992 (49)

Capsaicin rat Sensory nerve block with UDT1 Shono and Hutson, 1994 (69)UDT1 with flutamide Abe and Hutson, 1994 (68)Exogenous CGRP disturbs migration Al-Shareef et al, 2007 (105)Restores CGRP response in TS rat Hrabovszky and Hutson, 1999 (60)

TS rat UDT 80% with ectopic testes Park and Hutson, 1991 (67)GFN with excess axons and excess CGRP Terada et al, 1995 (125)CGRP receptors in gubernaculum down-regulated Terada et al, 1994 (142)Unilateral UDT with ipsilateral CGRP1 Hrabovszky et al, 2001 (66)Exogenous CGRP disturbs migration Clarnette and Hutson, 1999 (191)

Reproduced with permission J. M. Hutson et al: Cryptorchidism. Semin Pediatr Surg. 2010;19:215–224 (118), with permission. © Elsevier Inc. 456/200/0/36/539/



dominated in the rodent cremaster muscle during the in-guinoscrotal stage, whereas the abdominal musclesmainly expressed fast-twitch adult fibers (Myh1, 2, and 4)during this period. By contrast, if the cremaster musclewere to arise from passive stretching and eversion of theabdominal muscles, it would be expected to have similarmaturity and fiber type expression over its entire length(76).

Another reason to refute older theories of the origins ofcremaster muscle is its distinct neurovascular supply. Thecremaster muscle is supplied by the cremasteric artery andgenital branch of the GFN, whereas the abdominal mus-cles receive multiple arterial supply from the lower pos-terior intercostal arteries, superficial and deep circumflexarteries, and posterior lumbar arteries. The latter are alsoinnervated by the terminal branches of the lower inter-costal nerves, subcostal nerves, and iliohypogastric andilioinguinal nerves (73, 77).

B. Does the gubernaculum grow like a limb bud?The outgrowth of the gubernaculum from the inguinal

wall is analogous in many respects to embryonic limb buddevelopment. The formation of a limb from a limb budoccurs in 3 dimensions: anterior-posterior axis, dorsal-ventral axis, and proximal-distal axis, regulated by a spe-cific expression pattern of genes (78, 79). The limb budconsists of a core of mesodermal cells, also known as the

growth center, surrounded by a layerof ectoderm. The mesoderm is madeup of undifferentiated mesenchymalcells that are progenitors of myo-cytes, chondrocytes, and connectivetissues that eventually form muscles,bones, and tendons (79). The initiallimb bud proliferation and mainte-nance are controlled by specific sig-naling factors. In broad terms, thesegenes are the fibroblast growth fac-tors (FGFs) and homeobox tran-scription factor (Hox) genes (80),which have key roles in segmenta-tion of the embryo to control mor-phological development at differentlevels of the body. To determine theirinfluence on limb bud development,Hox genes were knocked out in chickembryos, which demonstrated de-creased overall cell division within thelimb bud and absence of limbs (81).

The apical ectodermal ridge(AER), which is located on the dor-soventral boundary of the caudal tipof the ectoderm covering a limb bud,

provides a source of signals to control the patterning of theproximal-distal axis of the limb bud. The expression ofFGF4, -8, and -10 from the AER promotes mesenchymalcell proliferation in the “progress zone” of the limb bud,with progressive differentiation of proximal structures,resulting in limb elongation to form muscles and bones(82, 83). The progress zone model of the limb bud is de-scribed as the region of rapidly proliferating mesenchymalcells, adjacent to the AER, where cells acquire positionalinformation. There are 2 distinct lineages, mainly the cellsthat give rise to connective tissue (ie, tendons, ligaments),and cells that will eventually become muscles after migra-tion (84, 85). Removal of the AER resulted in mesenchy-mal cell apoptosis and failure of the limb to form distalstructures.

During the inguinoscrotal phase, the fetal human gu-bernaculumisanelongatedconicalor cylindrical structurecomposed of loose extracellular matrix and mesenchymalcells comprised mainly of fibroblasts and muscle cells (86).In the rodent, maximal rate of mesenchymal cell divisionoccurs distally in the oval-shaped region of the gubernacu-lar bulb (Figure 2). This acts like a limb bud progress zonethat contributes to elongation of the gubernaculum andmigration into the scrotum (63). CGRP controls the gu-bernacular proliferation, and when it is absent in capsai-cin-treated rats (capsaicin kills the sensory nerves that con-

Figure 2.

Figure 2. A, Sagittal section of day 2 wild type (�/�) mouse showing bulb of the gubernaculum(arrow) about halfway in migration from inguinal abdominal wall to the scrotum (S), which iscaudal to the penis (P) (Desmin immunohistochemistry) (magnification, �100). B, High power(�200) of the same gubernacular bulb showing a core of undifferentiated mesenchymal cells(arrow) that behave similar to a limb bud progress zone. [S. Nagraj et al: The development andanatomy of the gubernaculum in Hoxa11 knockout mice. J Pediatr Surg. 2011;46:387–392 (90),with permission © Elsevier Inc.].



tain CGRP), cells in the gubernacular bulb undergoapoptosis (87). This is similar to the function of the AERin the limb bud model.

Like the progress zone model of the limb development,undifferentiated cells in the gubernacular bulb continuallyundergo proliferation to form a number of different typesof cells. Cells destined for myogenesis migrate laterallyand proximally out of the growth center of the bulb toform myogenic precursor cells that differentiate into cre-master muscles (82). This results in reduction in the size ofthe gubernacular bulb and formation of a thin rim of mus-cle around the gubernaculum. On the other hand, the un-differentiated mesenchymal cells of the gubernaculummay be responsible for the direction of gubernacular elon-gation. The gubernacular tip can be excised and graftedsimilar to the embryonic limb bud (88).

Regulatory factors involved in co-ordinating the 3 axes of limb bud de-velopment are also found in the mi-grating gubernaculum. FGF10 andHoxa10 were distinctively expressedin the caudal end of the gubernacu-lum at embryonic day 15, suggestingthat proximodistal elongation of thegubernaculum may be activated bythese genes before inguinoscrotal de-scent (89). Furthermore, Hoxa11knockout mice demonstrate cryp-torchidism (90) (Figure 3). Body walloutgrowths such as genital tubercle,limb bud, and branchial arches arefundamental developmental ap-pendages of embryos and appear tobe regulated by similar genetic pro-grams. The exact mechanism bywhich these transcription factorsregulate gubernacular proliferationand remodeling is unclear and re-mains circumstantial, but the conser-vation of the key regulatory pro-cesses through evolution suggeststhat it is likely to be similar to limbbud development.

C. The role of the mammary lineRecently, a new hypothesis impli-

cating the role of mammary glands intesticular descent has been proposed(91). An inguinal breast bud is nor-mally present outside the future exter-nal inguinal ring in rodents, immedi-ately adjacent to the gubernaculum at

embryonic day 15. By postnatal day 2, the proliferating gu-bernacular tip is on its way toward the scrotum, accompa-nied by a zone of remodeling mammary fat pad adjacent toit (15). To understand the potential role of this close ana-tomical relationship between the embryonic gubernaculumandthemammaryline,whichiscloselyrelatedtotheinguinalcanal inbothrodentsandhumans (Figure4),weneed to lookat how androgens regulate the inguinoscrotal phase ofdescent.

Administration of the androgen antagonist, flutamide,to pregnant rats at embryonic days 16–19 (before ingui-noscrotal migration and during the critical time windowfor androgenic programming of genital development) wasobserved to cause cryptorchidism in the male offspringdue to its proposed effect on the GFN. This is a critical timeperiod whereby androgen acts to masculinize the GFN and

Figure 3.

Figure 3. Role of Hoxa10, Fgf10, and Hoxa11 genes in mouse gubernaculum and breast buds. A,Hoxa10 expression in wholemount of mouse embryo at embryonic day 14.5. Hoxa10 expressionis present throughout the gubernaculum (g), with labeling in the sixth mammary bud (arrow) inthe anterior abdominal wall, just superficial to the site of the future inguinal ring. The darkstructure centrally located at the bottom of the panel is the phallus. B, Fgf10 expression in awholemount of a mouse embryo at embryonic day 14.5 is absent in the proximal gubernaculumbut present in the mammary bud (arrow). [Panels A and B were reproduced from S. S.Nightingale et al: The migrating gubernaculum grows like a “limb bud.” J Pediatr Surg. 2008;43:387–390 (89), with permission. © Elsevier Inc.] C, Wild type (�/�) (day 2) mouse gubernacularbulb containing swirl of primitive mesenchyme (trichrome staining). D, Mutant Hoxa11 (�/�)(day 2) mouse gubernacular bulb with persisting “swelling reaction” of extracellular matrix andabsence of mesenchymal swirl (and associated loss of progress zone-like properties). [Panels Cand D were reproduced from S. Nagraj et al: The development and anatomy of thegubernaculum in Hoxa11 knockout mice. J Pediatr Surg. 2011;46:387–392 (90), with permission.© Elsevier Inc.].



stimulate the gubernacular proliferative response. Aftertreatment with antiandrogens during this prenatal period,the amount of CGRP in the GFN is reduced. Despite up-regulation of CGRP receptors in the gubernaculum tocompensate for this reduction in endogenous CGRP, ex-ogenous CGRP addition to the organ culture failed to in-duce gubernacular proliferation (62, 92). In addition, ex-posure to flutamide during this programming windowresulted in abnormal prostate and seminal vesicle growthas well as high rates of hypospadias in these cryptorchidrats. By contrast, flutamide administration after embry-onic day 19 had no effect on the male reproductive tractdevelopment (52).

The mechanism of androgenic action and the location ofARs that control the inguinoscrotal phase remains unclearbecause few studies have been conducted on AR in the past.A recent study from our laboratory showed no AR expres-sion during the critical prenatal time window in the rat DRGof the L1–L2 spinal cord where the sensory cell bodies of theGFN reside (93). AR was only expressed perinatally in thegubernaculum itself and DRG (94) (Figure 5). Hence, thegubernaculum is unlikely to be the primary target organ ofandrogenic action that controls inguinoscrotal descent. An-drogen is involved in the development and maintenance ofthe nervous system postnatally (93). Interestingly, AR wasfound in the sc inguinal fat pad destined to form the mam-mary fat pad in females at embryonic days 17–19, which isduring the critical time for testicular descent as well as otherfeatures of sexual development.

Gubernacula of flutamide-treated rats show failure ofgubernacular eversion and persistence of mammary tissue

postnatally (91). This suggests that the mammary linemight be involved in regulating gubernacular eversion.Cremaster muscles in gubernacula of normally descendedtestes are thinner and longer, whereas muscle in guber-nacula of UDT induced by flutamide appear shorter andthicker, with increased extracellular matrix and lesser col-lagen concentration in both rats and human models (4,95). ARs have been detected in gubernacular fibroblasts,and their binding stimulates metalloproteinase to degradeextracellular matrix during gubernacular involution (96).Hence, remodeling of the cremaster muscle component isrequired to decrease gubernacular bulb size to allow forgubernacular eversion and descent of the testis. Inguinalfat pad may respond to androgens by providing trophicsignals to the gubernaculum and sensory fibers of theGFN, which then stimulate CGRP release from the GFNto control migration of gubernaculum and testis.

The presence of AR in the inguinal fat pad of neonatalrodents and the close proximity of breast buds that regressduring the critical time window for androgens suggestedthat there may be a link to the mammary line. Despite thenovelty of this proposal for mammals, such as the rodentand human, in marsupials it is already proven fact! TheGFN in female marsupials, such as the Tammar wallaby,supplies the gubernaculum and its contained developingcremaster muscle (called ilio-marsupialis in marsupials),which has an intimate relationship to the developingbreast buds in the pouch. In fact, the ilio-marsupialis be-comes attached to the developing breast tissue to becomethe suspensory muscle of the nipples and breasts (97, 98).The sc mesenchyme just outside the future external ingui-

Figure 4.

Figure 4. The developmental mammary line in humans and rodents. A, Mammary line in the human. Polymastia and polythelia can occur anywherealong this line. B, Rat mammary line and mammary fat pads. The sixth mammary bud is directly over the inguinal canal during embryonicdevelopment.



nal ring in rodents, which is destined to form the mam-mary fat pad in the absence of androgens, therefore showshigh expression of AR (99) (Figure 6).

Recent work from our laboratory suggests that this AR-positive mesenchyme along the mammary line may be thesite of androgen signaling that induces sexual dimorphismof the GFN (100) because there are no ARs in the GFNitself during the critical window of androgenic action. Theinguinal fat pad, which contains AR during the criticalwindow and is also supplied by the mammary branch ofthe GFN in rodents, could provide trophic factors thatmasculinize the GFN (100) (Figure 7.) There is a precedentfor peripheral androgenic control of neuronal develop-ment in the nerves supplying the bulbocavernosus muscle

(101, 102). In this case, the peripheral tissues produce aneuronal growth factor that regulates the masculinizationof the nerve (103, 104), and it is possible that a similarprocess is occurring in the GFN.

In summary, the inguinoscrotal phase is a very complexmorphological process with an apparently equally com-plex regulatory mechanism. The relatively inert gubernac-ulum develops limb-bud-like properties and grows out ofthe inguinal abdominal wall, thereby creating the inguinalcanal. The mammary line is quite likely to be involved inthe local signaling to initiate this growth, which is partlycontrolled by ARs in the gubernaculum itself as well asindirectly via CGRP from the GFN, which may guide themigrating gubernaculum to the scrotum. Although these

Figure 5.

Figure 5. A, AR receptor immunoreactivity in gubernaculum of Sprague-Dawley rats at embryonic (E) day 16 to postnatal (P) day 4. Histogramshows density of AR-positive cells/100 �m2 in the gubernacula of control and flutamide-treated rats. Note the absence of AR labeling in thegubernaculum during the critical window for androgenic effects on testicular descent (E16 to E19). B, AR receptor immunoreactivity in the sctissue adjacent to the gubernaculum (that forms the sixth mammary fat pad). Histogram shows density of AR-positive cells/100 �m2 between E16and P4 in control and flutamide-treated rats. [Panels A and B were reproduced from T. R. Nation et al: The effect of flutamide on expression ofandrogen and estrogen receptors in the gubernaculum and surrounding structures during testicular descent. J Pediatr Surg. 2011;46:2358–2362(94), with permission. © Elsevier Inc.] C, Histogram of nuclear AR staining density in DRG cells in control and flutamide-treated Sprague-Dawleyrats at embryonic (E) and postnatal (P) timepoints. The data are expressed as means � SEM. Note the absence of AR immunoreactivity during thecritical window of androgen sensitivity (E16–E19) for effects on testicular descent. [Panel C was reproduced from T. Nation et al: Androgen andestrogen receptor expression in the spinal segments of the genitofemoral nerve during testicular descent. J Pediatr Surg. 2011;46:1539–1543 (93),with permission. © Elsevier Inc.]



ideas remain speculative and controversial, there is evi-dence for CGRP in closure of human inguinal hernia, con-sistent with the possibility that the GFN plays an impor-tant role in human testicular descent similar to that inrodent models (70).

V. Germ Cell Development

The primitive gonocytes migrate from the caudal yolk sacstalk into the urogenital ridge and colonize the developinggonads around 6 weeks gestation in the human (106).There has been extensive research into the regulatory fac-tors involved in this process (106), but little attention towhat happens to germ cells postnatally, apart from thework of Hadziselimovic and colleagues (50, 107), whohave studied postnatal germ cell development extensively.They have suggested a key role for early postnatal trans-formation of neonatal gonocytes into adult dark-sper-matogonia (AD-S) in subsequent fertility (108). AD-S havea dark nucleus containing a vacuole and dark cytoplasmwhen examined under standard magnification and hema-toxylin and eosin staining. They are located in the periphery

of the testicular tubule in contact withthe basement membrane, unlike theneonatal gonocyte, which is located inthe center of the testicular tubule.AD-S are thought to differentiateinto adult pale-spermatogonia,which have much paler cytoplasmand nucleus and are considered tobe committed for subsequent sper-matogenesis (109).

The number of germ cells in the tes-tis may be reduced even prenatally insome babies with cryptorchidism be-cause 23% of fetuses who died withcryptorchidism in the third trimesterwere found to have lower germ cellcounts compared with fetuses whodied with descended testes (110). Thissuggests that testiculardysgenesismaybean important factor ingermcell lossin the nonviable fetus. Whether this isan important factor in live-born in-fants with cryptorchidism remains tobe seen, although in these babies germcells are present in all cases, but thenumbers may be reduced from 28weeks gestation.

Shortly after birth in normal boys,there is a brief surge in gonadotro-

pins, followed by a transient rise in testosterone and asustained rise in MIS/AMH levels (111, 112) (Figure 8).Inhibin B is also elevated in the first few months after birth.This hormonal surge at around 3 months is known as“mini-puberty” (113, 114). The role of this transient burstof hormones is not fully known, but it has been suggestedthat it may be involved in masculinizing the brain (115), aswell as causing obliteration of the PV after testicular de-scent is complete (116). It is also likely to be critical fornormal germ cell development because transformation ofneonatal gonocytes into AD-S is proposed as a key func-tion of mini-puberty (108, 113, 114).

The AD-S is now thought to be the stem cell for sper-matogenesis, and its appearance at 3–9 months of age islinked to the potentially optimal time for orchidopexy(see Section VI.D.). We now know that the germ cellsundergo a series of developmental steps in early child-hood, with the appearance of primary spermatocytesaround 3– 4 years of age. The cells are then in a restingphase until the onset of puberty, when the physiologicalblood-testis barrier develops in Sertoli cells and the pri-mary spermatocytes begin differentiating into sperma-tids (117) (Figure 9).

Figure 6.

Figure 6. AR distribution in the gubernaculum (Gub), inguinal fat pad (IFP), and scrotum in asagittal section of a Sprague-Dawley rat at embryonic day 17 (in the middle of the critical timewindow for androgenic regulation of testicular descent). The gubernaculum contains no AR (red)and ends in the muscle of the anterior abdominal wall above the penis and pelvic bones. The IFPcontains numerous AR-positive fibroblasts (red) from just anterior to the gubernaculum anddown to the inside of the scrotum. [Reproduced from B. Allnutt et al: The common fetaldevelopment of the mammary fat pad and gubernaculum. J Pediatr Surg. 2011;46:378–383(99), with permission. © Elsevier Inc.]



Theearly stepsofgermcell developmentwereoverlookedfor a long time because the neonatal gonocyte was probablymistaken foraprimary spermatocyte, asbothare located in thecenter of the testicular cords (Figure 10). As will be described in

SectionIV.D.,gonocytetransformationintoAD-S,theputativestem cell, is the step inhibited by cryptorchidism.

Interestingly, only about half the neonatal gonocytestransform into AD-S, which might be an adaptation to select

the optimal cells for conversion intostem cells because the stem cells arelikely to be under significant evolu-tionary pressure since they will carrythe genome into the next generation.Those gonocytes that do not trans-form normally undergo apoptosis, sothat no gonocytes remain. It is quitelikely, although yet unproven, thatcryptorchidism may disrupt the apo-ptotic pathway as well as transforma-tion, which would lead to some gono-cytes persisting in the tubule (118).Because seminomas carry many of thesame cellular markers as gonocytes,it is possible that persisting abnor-mal gonocytes may be the origin ofcarcinoma-in situ and eventual overtmalignancy.

Figure 8.

Figure 8. Diagram of serum hormone levels vs age for testosterone (T), inhibin �, and MIS/AMH,showing the surge of androgens around 3 months of age, known as “mini-puberty.”

Figure 7.

Figure 7. Our current hypothesis about how androgens act mostly indirectly on the gubernaculum via the GFN and its neurotransmitter, CGRP. A,The original hypothesis suggested testosterone from the testis (T) acted on the AR of the gubernaculum to trigger it to simply bulge into theapparently adjacent scrotum. B, Our current hypothesis shows testosterone from the testis (T) acts on the AR in the sc mesenchyme of themammary line, immediately superficial to the gubernaculum, which ends in the inguinal abdominal wall muscles. AR stimulation of themesenchyme is proposed to cause production of a neurotropin, which is taken up by the GFN sensory fibers and masculinizes the nerve. The GFNthen produces CGRP, which is released from the sensory nerve endings to guide the migration of the gubernaculum to the scrotum, which is inthe perineum. C, During inguinoscrotal migration, the PV hollows out the gubernaculum, which remains attached to the testis (T) and epididymisby the gubernacular cord. The distal, solid tip of the rodent gubernaculum contains undifferentiated mesenchyme, which has properties similar toa “progress zone” (PZ) in an embryonic limb bud. Enzymes dissolve the extracellular matrix so that the tip of the gubernaculum is loose under theskin during migration.



VI. Clinical Aspects of Cryptorchidism

A. EtiologyTesticular descent involves a complex interaction be-

tween the hormonal control and a major anatomical re-modeling and movement (1). Thus, the causes of UDT aremultifactorial (Table 2) (4). Deficiency in prenatal andro-gen secretion secondary to insufficient pituitary gonado-tropin stimulation or low production of placental gonad-otropin is considered to be the commonest cause of UDT(119). The concept of primary or secondary testicular dys-genesis has gained interest in recent years, with the pro-posal that the testicular dysgenesis might be increasing infrequency secondary to hormonal disruptors in the envi-ronment (120).

Testicular dysgenesis is considered to be a possiblecause for an increasing frequency of hypospadias, cryp-torchidism, and testicular cancer and may account for lowprenatal androgens that are likely to cause inadequate in-guinoscrotal testicular descent (121, 122).

The common location for UDT to occur is at the neckof the scrotum or just outside the external inguinal ring,

just lateral to the pubic tubercle andconfined anteriorly by the superficialabdominal wall fascia (of Scarpa)and posteriorly by the externaloblique muscle.

There are many inherited syn-dromes with multiple anomalies thatareassociatedwithUDT.It iscommonto observe cryptorchidism in boyswith malformations of the brain suchas microcephaly because an intact hy-pothalamic-pituitary-gonadal axis isneededfornormaltesticulardescent tooccur (4, 123, 124). This can be due toimpaired production of testosteroneor failure of arachidonic acid forma-tion, which is important in neural tubedevelopment. Hadziselimovic et al(123) have also demonstrated the as-sociation of cryptorchidism with ex-omphalosoromphaloceleandmalfor-mations of the brain, emphasizing theimportance of considering the pres-ence of brain malformations in anychild found to have omphalocele andUDT. But Aliotta et al (124) could notfind any clear relationship betweengastroschisis or omphalocele andcryptorchidism. This study is in con-trast with studies done by Kaplan (13)and Frey and Rajfer (126) where they

managed to show an association between abdominal walldefects, such as gastroschisis, omphalocele, and umbilicalhernia with cryptorchidism. They have postulated that thesedefects can significantly decrease the intra-abdominal pres-sure, and this can disrupt descent of the testes. Studies byCortes et al (110, 127) have shown a close relationship be-tween cryptorchidism and abnormal caudal developmentsuch as malformation or dysplasia of the kidneys, ureters, orspine (T10–S2 segments), where 34% of these were presentin cryptorchid fetuses. Urological disorders such as prunebelly syndrome and posterior urethral valves have been as-sociatedwithahighfrequencyofcryptorchidism(128–131).Prune belly syndrome is a rare and complex malforma-tion—a clinical triad of deficiency of abdominal muscle, bi-lateral cryptorchidism, and a dilated urinary tract. The pa-thology is likely to be due to in utero bladder outletobstruction, with the overdistended bladder disrupting thenormal development of abdominal wall musculature andalso preventing testicular descent. Boys with severe hypos-padias are also at risk of having cryptorchidism, which islikely to be due to prenatal androgen disruption (132).

Figure 9.

Figure 9. Postnatal germ cell development in boys. The neonatal gonocytes transform into AD-Sbetween 3 and 9 months, during “mini-puberty.” Some AD-S differentiate into adult pale (A-pale) spermatogonia, then intermediate, B-spermatogonia, and finally primary spermatocytes at3–4 years. After 10 years, the blood-testis barrier develops, and primary spermatocytes begindifferentiating into spermatids for full spermatogenesis at puberty.



B. Cryptorchidism in congenital syndromesIn the following short summary of several congenital

syndromes associated with a high incidence of UDT, wehave only included syndromes in which one or more genemutations have been identified. However, in the vast ma-jority of cases, little is known about how the affected mo-lecular pathways lead to the characteristic phenotype ofthe particular syndrome. Nevertheless, in the sections en-titled “Pathophysiology of UDT,” we speculate on possi-

ble mechanisms by which theaffected genes may lead to cryp-torchidism based on current under-standing of the normal processes oftesticular descent (Figure 11).

1. Robinow syndrome (RS)a. Clinical features. RS is a rare humancondition characterized by me-somelic dwarfism (truncation of theforelimbs), facial abnormalities,brachydactyly, and hypoplastic ex-ternal genitalia, and it exists in bothdominant and recessive forms (133,134). In the most comprehensivestudy characterizing the clinical fea-tures of patients with RS, Mazzeu etal (135) noted a high frequency ofcryptorchidism in both dominant(72%) and recessive (68%) forms ofthe syndrome. Both forms of RS ex-hibit a large spectrum of potentialabnormalities, and 100 cases of RShave been definitively identifiedworldwide (134).

b. Genetics. RS results from a pertur-bation of the noncanonical “wingless/integrated” (Wnt)signaling pathway (136). In the noncanonical Wnt path-way, Wnt-5a activates the important transcription factorc-Jun N-terminal kinase (JNK), via the receptor tyrosinekinase-like orphan receptor 2 (Ror2). JNK modulates cellmigration and is necessary for spatially organizing cellswithin developing tissues, as well as stimulating cell pro-liferation (137). The recessive form of RS is caused byvarious mutations in the human Ror2 receptor (134, 138).Mutations in Wnt-5a itself have recently been described inseveral patients with dominant RS; however, similar mu-tations were not observed in other sufferers of this con-dition, suggesting that RS may be genetically heteroge-neous, resulting from mutations in several genes involvedin Wnt-5a/Ror2 signaling (139).

c. Pathophysiology of UDT. In Wnt-5a knockout mousemodels, the gubernaculum is hypoplastic, and the testesare found in a high intra-abdominal position. Experimentsutilizing microarray technology have shown that INSL3stimulates expression of several genes involved in Wntsignaling, including Wnt-5a (140, 141). Recent work fromour own laboratory has revealed that Wnt-5a and its re-ceptor Ror2 are strongly expressed within the mesen-chyme of the rodent gubernaculum early during its devel-

Table 2. The Potential Causes of UDT

Androgen deficiency/blockagePituitary/placental gonadotropin deficiency (may be common)

Gonadal dysgenesisAndrogen synthesis defects (rare)AR defects (rare)

Mechanical anomaliesPrune belly syndrome (bladder blocks inguinal canal)Posterior urethral valve (bladder blocks inguinal canal)Abdominal wall defects (low abdominal pressure/gubernacularrupture)Chromosomal/malformation syndrome (connective tissue defectsmay block migration)

Neurological anomaliesMyelomeningocele (GFN dysplasia)GFN/CGRP anomalies

Acquired anomaliesCerebral palsy (cremaster spasticity)Ascending/retractile testes (persisting fibrous remnant of processusvaginalis) may be secondary to androgen deficiency perinatally

Figure 10.

Figure 10. Germ cell development shown in the tubule. The gonocyte (white) begins in thecenter of the cord. At 3–9 months, it migrates between Sertoli cells (gray) to reach the basementmembrane, where it differentiates into an AD-S (dark gray), which sits on the membrane and hasdense cytoplasm and a prominent nucleolus. The AD-S is thought to be the stem cell forsubsequent development into adult-pale spermatogonia (AP-S; white), intermediatespermatogonia, type-B spermatogonia (B-S; white), and then primary spermatocytes (10 S; white)at 3–4 years. Cryptorchidism (UDT) blocks the first step. (Colors of germ cells refer to theirappearance in standard histological sections).



opment (142). Therefore, noncanonical Wnt signaling inthe cells of the gubernaculum may be requisite during thefirst phase of gubernacular development and transabdom-inal testicular descent. Failure of this process, due to mu-tations in Ror2 or Wnt-5a, may explain the high frequencyof UDT in patients with RS.

2. Aarskog syndrome

a. Clinical features. Aarskog syndrome (also known as Aar-skog-Scott syndrome) comprises a cluster of clinical signsresembling those of RS, including hypertelorism, shortstature (due to vertebral abnormalities), phalangeal de-fects, and urogenital anomalies including shawl scrotumand cryptorchidism (143, 144). The frequency of cryp-torchidism is high, with approximately 70% of male pa-tients affected, strikingly similar to the reported frequencyin both dominant and recessive RS (144).

b. Genetics. Aarskog syndrome results from various muta-tions in the faciogenital dysplasia protein 1 (FGD1) gene

situated on the X-chromosome (145,146). FGD1 is a guanine-exchangefactor that activates the signalingprotein cell division control protein42 homolog (Cdc42), which subse-quently serves to co-ordinate diversecellular functions dependent on cel-lular type and context (146, 147).

c. Pathophysiology of UDT. It has beenshown that Cdc42 is involved in ac-tivation of JNK (the downstream ef-fector of noncanonical Wnt signal-ing), and may promote cell polarityand migration via effects on the cy-toskeleton (148, 149). In fact, Cdc42and noncanonical Wnt pathways(via Wnt-5a) may work in synergy tomodulate cell polarity and migrationvia activation of JNK (150), possiblyaccounting for the similarities inphenotypes between Aarskog andRobinow syndromes. Thus, muta-tions in FGD1, with a subsequent de-crease in Cdc42 activity, may lead toa suppression of signaling throughthe JNK pathway that, like muta-tions in Wnt5a or Ror2, may result inabnormal cell organization and mi-gration within the gubernaculumand defective testicular descent.

3. Noonan syndrome/LEOPARD syndromea. Clinical features. Noonan syndrome describes a set ofclinical abnormalities including stenosis of the pulmonaryvalve (or other congenital cardiac defects); dysmorphicfacial features including low-set ears, pectus excavatum,or carinatum; short stature; webbed neck; and cryp-torchidism (151). The frequency of cryptorchidism ap-pears to be between 70 and 80% (151). LEOPARD syn-drome is a related condition comprising multiplelentigenes, electrocardiogram anomalies, ocular hyperte-lorism, pulmonary stenosis, genital abnormalities, growthretardation, and deafness (152). Cryptorchidism is also afeature in LEOPARD syndrome and has been reported inapproximately 50% of cases (153).

b. Genetics. Noonan syndrome is a genetically heteroge-neous condition, with various mutations affecting the ratsarcoma (Ras)/MAPK pathway (154–156). Most patientsexhibit an activating mutation in the protein-tyrosinephosphatase non-receptor type 11 (PTPN11) gene, which

Figure 11.

Figure 11. Summary of proposed mechanisms of UDT in selected congenital syndromes. NS,Noonan syndrome; HS, HITCH syndrome (hearing impairment, undescended testes,circumferential skin creases, and mental handicap); ECM, extracellular matrix.



encodes a protein tyrosine phosphatase Src homology re-gion 2 domain-containing phosphatase-2 (SHP-2) (157).This protein is involved in modulating many signalingpathways involved in cell differentiation and proliferationduring development. Although Noonan syndrome iscaused by a gain-of-function mutation in PTPN11, LEOP-ARD syndrome is the result of inactivating mutations inPTPN11 (158).

c. Pathophysiology of UDT. The molecular pathways in-volved in the phenotype of Noonan and LEOPARD syn-dromes remain to be elucidated, and given the diverse rolesof the Ras/MAPK pathway, the UDT is most likely due toperturbation of several molecular processes. Interestingly,SHP-2, the product of the PTPN11 gene, is known to ac-tivate HOXA-10, knockout of which results in bilateralcryptorchidism in mice (159, 160).

4. Cornelia de Lange syndrome (CdLS)a. Clinical features. CdLS is a rare disorder characterized bymultiple abnormalities including limb growth, facial dys-morphia, developmental delay, heart defects, severe gas-troesophageal reflux, and synophrys (161). Cryptorchid-ism occurs frequently with approximately 73% of boysaffected (162).

b. Genetics. Although thought to be a genetically hetero-geneous condition, most cases of CdLS are due to muta-tions in the nipped-B-like protein (NIPBL) gene, whichencodes the protein delangin (163). Delangin is involved inthe regulation of the chromosomal adhesion complex (fa-cilitating separation of sister chromatids during cell divi-sion) (164).

c. Pathophysiology of UDT. It isnotknownhowmutations inthe delangin protein cause the characteristic phenotype ofCdLS. However, in addition to its role regulating the chro-mosomal adhesion complex, delangin has been shown tocontrol expression of the distal-less homeobox (Dlx) fam-ily of genes, which, along with the Hox gene cluster, reg-ulate patterning of the limb (164, 165). Because the ho-meobox genes Hox-A10 and Hox-A11 are necessary forproper gubernacular development in mice, perturbation inthe related Dlx genes due to a mutation in delangin mayaffect gubernacular development and testicular descent(166).

5. Rubinstein-Taybi syndrome (RTS)a. Clinical features. RTS affects approximately 1 in 125 000live births and is characterized by its distinctive facial fea-tures (high arched eyebrows, characteristic grimace), mi-crocephaly, mental retardation, growth abnormalities,

genital anomalies, and broad thumbs and great toes (167).The rate of UDT in males is reported to be between 78 and100% (168, 169). There is also an increased risk of de-veloping several malignancies, including tumors of thenervous system and leukemias (170).

b. Genetics. Most cases of RTS are caused by mutations inthe cAMP response element binding protein-binding pro-tein (CREB-BP) gene, the product of which is CREB-bind-ing protein (171). CREB-BP is a transcription factor ex-pressed ubiquitously during embryonic development,which activates expression of many different genes in-volved in growth and homeostasis (172). A much smallernumber of cases are caused by mutations in the closelyrelated gene EA-1 binding protein 300 (EP300), whichfunctions in the same regulatory pathways as CREB-BP(173).

c. Pathophysiology of UDT. It is unknown how mutations inCREB-BP or EP300 result in the characteristic clinicalmanifestations of RTS. Interestingly, it has been shownthat CREB-BP is a coactivator of the AR and may be nec-essary for proper signaling through the androgen pathway(174). Thus, the high incidence of cryptorchidism in RTSmay result from the abnormal action of androgens in stim-ulating inguinoscrotal testicular descent.

6. Ulnar-mammary syndrome (UMS)a. Clinical features. UMS is characterized by variable ab-normalities of the ulnar aspect of the distal upper limb,including hypoplasia of the little finger and aplasia of thecarpal bones (175). In addition, gross hypoplasia or apla-sia of the mammary and apocrine glands is usually present(175, 176). Cryptorchidism is frequently seen in male pa-tients, although the precise incidence has not been de-scribed (175).

b. Genetics. UMS is caused by various mutations in T-boxtranscription factor 3 (Tbx-3) (176, 177). Tbx-3 is in-volved in regulating mesenchymal stem cell fate and pro-liferation during embryonic development (178, 179).

c. Pathophysiology of UDT. It has been established thatTbx-3 is a direct downstream target of canonical Wnt/�-catenin signaling (179). Recent evidence suggests an im-portant role of this signaling pathway in testicular descent.Mouse models with conditional �-catenin knockdown ex-hibit bilateral UDT with grossly underdeveloped guber-nacula (28), whereas recent work from our laboratorysuggests that canonical Wnt signaling may mediate theeffects of androgen during the remodeling process re-quired for the second phase of testicular descent (11).



Therefore, it may be that UDT in patients with UMS maybe due to perturbations in the downstream effects of Wnt/�-catenin due to mutations in Tbx-3. Furthermore, recentwork in the rodent has identified the inguinal mammarybud as a potential source of signals initiating testiculardescent (180). Thus, it may be that the mechanisms ofUDT and mammary gland aplasia in UMS are inextricablylinked.

7. HITCH syndromea. Clinical features. HITCH syndrome is an extremely raredisorder with only a handful of cases reported in the lit-erature and is an acronym of the characteristic features,namely, hearing impairment, UDT, circumferential skincreases, and mental handicap (181).

b. Genetics. The genetic etiology of this disorder is un-known; however, skin biopsies from affected individualshave shown degenerative collagen fibers, suggesting thatthe clinical manifestations may be related to abnormalcollagen synthesis (181).

c. Pathophysiology of UDT. The importance of extracellularmatrix remodeling during testicular descent has been re-cently highlighted in both animal models and humans. Ithas been proposed that remodeling of the inguinal fat padis vital for gubernacular migration in the rat (182). Fur-thermore, recent studies of human fetal gubernacula con-cluded that extensive remodeling of the extracellular ma-trix occurs in and around the gubernaculum between the15th and 29th weeks after conception (95). Interestingly,gubernacula from children with abnormal testicular po-sitions showed a lower collagen concentration (95). UDTin HITCH syndrome may therefore be the result of ab-normal collagen in the gubernaculum and inguinal fat pad,which may hinder the normal gubernacular remodelingand migration required for normal testicular descent tooccur.

C. FrequencyApproximately 2–4% of male infants are born with

UDT, thus making cryptorchidism one of the commonestcongenital anomalies in male genitalia (17, 118). Almost27 000 orchidopexies are done in the United States eachyear (183). The rate of cryptorchidism decreases to0.7–1% when the children reach the age of 1 year (184,185). It has also been reported that there was an increasingincidence of UDT in recent decades. For example, in theUnited Kingdom, the number of children with UDT at age1 year appeared to increase from 0.96–1.58% in 1984(186). From 5 to 7% of UDT boys do not have satisfactorytesticular position in the scrotum, but most of these boys

will have descended testes when they reach puberty (187).Many believed the increased incidence of UDT was due toincreased numbers of operations for “retractile” or “as-cending” testes (4) because the retractile testis is frequentlymisdiagnosed as cryptorchidism. In a study by Ward andHunter (188), which showed the problems in making thediagnosis of UDT, a higher prevalence of cryptorchidismwas found later in childhood compared to the previousprevalence the authors had found on the same boys atbirth.

Some studies suggested that environmental, social fac-tors, and geographical differences can affect the incidenceof UDT; in Denmark, 9% of newborn males were diag-nosed with UDT compared to only 2.4% in Finland (187,189). This supports the proposal that genetic factors maybe involved in some cases.

However, this perceived difference in frequency ofUDT may be related to the definitions used because theDanish study included a large number of boys with testesin a high scrotal position. When the definition of cryp-torchidism is restricted to lack of a testis in the scrotum, theincidence was 2% rather than 9%. These differences high-light the need to be explicit about the definitions used indifferent studies. In the same period, in 2002, in anotherarea of Copenhagen a similar study of 1012 consecutivesingleton newborn boys found 2.4% with cryptorchidism,defined as the lack of at least 1 testis in the scrotum (190).

It was also reported that UDT was higher (4%) in in-dustrialized Western societies (185). Cases of UDT are 5-to 7-fold higher in premature and low-birth-weight(�2500 g) infants (19, 186, 192, 193) because the processof testicular descent is completed around 35 weeks gesta-tion, just before birth. The frequency of UDT for prema-ture babies varies from 1.1 to 45.3%, with higher rates(50–70%) for bilateral UDT (185), showing a strong re-lationship between cryptorchidism and low birth weight.This may be related to impaired placental function (187)because testicular descent is dependent on hCG from theplacenta (194). Jensen et al (195) hypothesized that bothintrauterine environment and maternal inheritance cancontribute to cryptorchidism; ie, concordance rates of24.1% UDT in dizygotic twin brothers, 27.3% in mo-nozygotic twin brothers, and 3.4% in paternal half-broth-ers vs 6.0% in maternal half-brothers. Seventy percent ofUDT occur on the right side of the body, compared to 30%on the left side.

D. Timing of surgery and the risk of infertility andmalignancy

The main reasons why UDT should be treated are toallow normal fertility and prevent malignancy in the fu-



ture and to overcome the cosmetic and secondary psycho-logical problem.

The optimal time for intervention of UDT is related topostnatal germ cell maturation because studies haveshown early testicular injury in cryptorchidism. Since the1950s, the recommended timing of surgery has decreasedfrom early adolescence to recently suggested managementat 6–12 months (118). There is a view that orchidopexy inthe first year might be unnecessary because about half ofthe testes undescended at birth will descend during the firstyear. However, for term babies, delayed descent beyond12 weeks of age is quite rare, so that by 6 months of ageelective operation is justified because spontaneous descentis then rare (118). Studies have shown progressive effectsof cryptorchidism on morphological changes in the testis(118, 197–203), starting with electron microscopicchanges at 1–2 years of age, followed by degeneration oftesticular cells detected by light microscopy and macro-scopic atrophy of testes seen at 5–7 years of age. Based onthe morphological changes by electron microscopy, or-chidopexy was recommended in the 1980s at 1–2 years ofage.

Recent findings of the failure of gonocyte transforma-tion into AD-S at the age of 3–6 months (114, 204) suggestthat the recommended age of surgery should be about 6months. This is based on the assumption that early relo-cation of the testis into the scrotal environment (with itslower temperature) will allow the germ cells to developnormally, and thus lower the risk of oligospermia andcancer. A further assumption about the optimal timing ofsurgery is that there is no intrinsic endocrinopathy causinginadequate testicular maturation, which may be present insome boys (51).

As mentioned previously, it has been shown that theprimitive germ cells in the seminiferous cords undergo dif-ferentiation from neonatal gonocytes into AD-S—whichare now thought to be the stem cells for spermatogenesis(205, 206). This can only occur normally if the environ-mental temperature is lower than the core body temper-ature, ie, 33°C within the scrotum. Thus, this step will besignificantly impaired in UDT that are residing at a highertemperature, and this leads to the lack of stem cells forpostpubertal spermatogenesis, poor semen quality, and ahigher risk of infertility. In a study by Bremholm Rasmus-sen et al (207) on the fertility status of bilateral UDT thatdescended spontaneously after the age of 10 years, 53%had subnormal or severely low sperm concentration, and75% of these men had borderline testicular size indicatingimpaired spermatogenesis. Boys with UDT have a reducednumber of germ cells in the testes compared to the normalpopulation (187); germ cells may be lacking after 15months of age (208).

UDT is also known to increase the risk of having tes-ticular tumors, mainly seminomas (198, 209–211). In arecent review, we hypothesized that persistent exposure tohigh temperature in the UDT could allow mutation of theneonatal gonocytes that had failed to either mature intospermatogonia or undergo apoptosis (118, 212). Thesecells may persist in the testes for years, eventually becom-ing carcinoma in situ cells with a high risk of testicularmalignancy later in life, ie, 20–40 years of age (198, 199,210, 213). The high temperature in the UDT is thought toprevent both transformation of neonatal gonocytes toAD-S at 3–9 months of age and apoptosis of remaininggonocytes, and thus, over time, to predispose maturationof persisting gonocytes because of impaired DNA repairmechanisms.

In the 1940s and 1950s, the risk of having testiculartumors among those who had UDT was thought to be35–50 times greater than normal (209, 214), but withmore recent studies, the risk is now estimated to be 5–10times more than normal (215). This change is mostly re-lated to changing definitions and methods of estimatingthe link between UDT and cancer. The risk of testicularcancer is much higher in bilateral than in unilateral cryp-torchidism (216). In men with a testicular tumor and ahistory of unilateral cryptorchidism, 5–10% may developtesticular tumor in the contralateral testis, which mightsuggest a genetic predisposition (209).

Aiming to prevent germ cell death or dysplasia and toensure normal maturation of testicular germ cells (217),the current guideline for orchidopexy for cryptorchidismis about 6 months, and at most before 18 months of age(218, 219). Furthermore, studies by Kollin et al (220) andRitzen et al (221) show improvement of testicular growthwith early orchidopexy (9 mo) vs orchidopexy at the ageof 3 years (219). Studies by Taskinen et al (222) and En-geler et al (223) similarly show benefit of early orchi-dopexy. It appears that the longer cryptorchid boys waitto have orchidopexy, the higher the risk of having no germcells in their testes (224), increasing the risk of infertilityfrom 79 to 100% (225). The risk is much higher in menwho had bilateral compared to unilateral UDT (226). Or-chidopexy does lower the risk of having testicular cancer(227, 228) but has not been shown to completely eliminateit. However, the current paucity of studies showingwhether or not early orchidopexy can prevent germ cellcancer is a result of the lag time between recent changes intiming of surgery and assessing the effects of this 30–40years later in adulthood. Clearly, we need a more imme-diate marker for subsequent cancer risk so that we canoptimize management in infancy. Nevertheless, early or-chidopexy at least permits ready physical examination of



the previously UDT throughout childhood and early adultlife.

Although there is now significant evidence for orchi-dopexy under the age of 18 months to preserve fertility andto prevent testicular tumors (209, 215), most childrenwith UDT still do not have orchidopexy until a later age.In a study by McCabe and Kenny (219), only 20% of boysin the United Kingdom underwent orchidopexy below 18months of age over the period 1995 to 2005, but the studyalso showed that there was a trend toward earlier orchi-dopexy. In another study by Jensen et al (226) in Denmarkbetween 1995 and 2009, the authors observed a low or-chidopexy rate at the recommended guidelines, but theyalso concluded that there was a shift toward a younger agefor operation. Barthold and Gonzalez (185) reported thatonly 40% of boys had orchidopexy at � 4 years of age. Thereasons for this poor adherence to the guidelines might bethat some centers and surgeons are not keen to operate onvery young children, as well as ignorance by some sur-geons on the evidence base for management of UDT. Al-ternatively, some doctors may prefer to manage UDT con-servatively, hoping that spontaneous descent wouldoccur. In some cases, the boy may have developed retrac-tile testes, where the testes were initially located in thescrotum in early childhood. Rarely, there may be failure todetect UDT on the newborn baby check or delayed referralto the general surgeon, pediatric surgeon, or urologist(124).

E. Acquired cryptorchidismFor many years, cryptorchidism was regarded as a

purely congenital anomaly, and the assumption was thata testis not in the scrotum later in childhood had alwaysbeen undescended. Recently, there has been a change inthis view, with the gradual recognition that in many boysthe testicular malposition is only apparent later in life,rather than after birth.

This phenomenon was difficult to explain until it wasappreciated that for a testis to remain in the scrotum afterbirth, the spermatic cord must elongate progressively withage (229). Between birth and 10 years of age, the cord mustdouble in length from 5 to 10 cm, and acquired cryp-torchidism will occur when this process is perturbed (118).Possible causes include cerebral palsy, with an upper mo-tor neuron anomaly affecting cremaster muscle growth(229) or failure of the PV to obliterate and completelydisappear (116). The latter anomaly leaves a fibrous rem-nant of the PV in the spermatic cord, which prevents itsnormal elongation with age. The underlying reason forthis anatomical hindrance to normal elongation may below androgen levels and/or low CGRP levels in the GFNduring the perinatal period. There certainly is evidence

that CGRP is involved in regulating PV obliteration inhumans (70), which links inguinal hernia and hydrocelewith acquired UDT (116).

Recognition of acquired UDT by many authors (230–232) has forced a re-evaluation of the timing of surgeryand the prognosis in cryptorchidism.

For the age at operation, for instance, there are now 2peaks in frequency, with orchidopexy in the first 2 yearsnow becoming more common for congenital cryptorchid-ism, and a second, flatter peak at 5–10 years of age whenacquired UDT is recognized.

Whether or not acquired UDT requires surgery remainscontroversial, with emerging evidence that there is im-paired fertility but no increased malignancy risk (233). Wehave proposed that acquired UDT will not interfere withearly postnatal germ cell development because during thefirst year the testis will still be in the scrotum, and hencestill at its optimal postnatal temperature (ie, 33°C) (118).Gonocytes should, therefore, have transformed or under-gone apoptosis, so that subsequent malposition willmerely inhibit stem cell survival, but there should be noabnormally persisting gonocytes that might be the originof malignancy after puberty.

One of the confusions for clinicians is the differencebetween retractile and cryptorchid testes. The cremastericreflex pulls the descended testis back out of the scrotum inresponse to sensory stimuli (ie, local trauma or lower tem-perature). This reflex is less active when circulating an-drogen levels are elevated, such as in the first 6 months oflife and at puberty. By contrast, in the middle of childhoodwhen androgen secretion is minimal, the cremasteric re-flux is quite brisk, and there is commonly some overlap inthe appearance between a testis that is very retractile andone that is developing acquired cryptorchidism (118).With the gradual recognition in recent years of acquiredcryptorchidism, the so-called “retractile” testis has be-come less well defined and may represent a slightly shortspermatic cord. Clinically, the normal retractile testisshould return quickly to the bottom of the scrotum afterretraction into the inguinal region (232). When it onlyreturns to the upper scrotum, surveillance is required toensure that its position does not become higher to becomean acquired cryptorchid testis.

F. Roles of hormone therapyHadziselimovic et al (113) have shown that at 3 months

after birth, there is a hormonal surge of gonadotropins(LH and FSH) and testosterone, also called “mini-pu-berty,” which is thought to be needed for the maturationof germ cells to AD-S, as described in Section V. In cryp-torchid boys, this process is impaired, and this increasesthe risk of future infertility (234). There may be low levels



of LH or no LH surge, with a subsequent deficiency oftestosterone in cryptorchid boys (111). In the absence of ahormonal surge to stimulate the development of AD-S, ithas been hypothesized that hormonal treatment may beneeded to prevent the infertility problems among cryp-torchid boys, which was first recommended by BernhardSchapiro (235). Hormonal treatment is commonly recom-mended in Europe as part of UDT management, togetherwith orchidopexy (236). But hormonal therapy is not rec-ommended in Scandinavia (221), due to its poor efficacyand the possible side effects, such as accelerated secondarysexual development (due to the initial massive dose) andpremature epiphyseal closure, which can impact bonegrowth development (237). A study by Cortes et al (238)demonstrated an even lower quantity of germ cells due toapoptosis of germ cells after the hormonal therapy (239).

Studies by Hadziselimovic et al (204), Jallouli et al(240), Schwentner et al (241), Huff et al (242), and Lala etal (196) have demonstrated a higher number of spermato-gonia after GnRH treatment with a synthetic GnRH an-alog among boys with UDT. However, in randomizedcontrolled trials, the success rate of hormonal therapy wasonly 10 to 20% (197, 199), and some studies report that15% of the testes had subsequently reascended to a su-prascrotal position (198, 216). Although cryptorchidismreduces fertility, it does not completely block the paternitypotential (224). The differences in results between authorsabout the success of GnRH treatment may relate to theheterogeneity of etiology of cryptorchidism in the popu-lation treated because recent recognition of a subgroup ofboys with bilateral UDT and impaired gonadotropins sug-gests that hormone treatment may be required in specificcircumstances (51).

VII. Extrapolation of Animal Models toHumans

One issue that has bedeviled research into testicular de-scent and cryptorchidism is extrapolation from animalmodels to humans. The minor differences in anatomy andthe way the gubernaculum migrates have meant animalmodels need to be interpreted with care and some limits.However, the fundamental processes of descent are thesame. A curious difference yet to be fully accounted for isthe presence of cryptorchidism and an abnormally longgubernacular cord in boys with PMDS, whereas MIS/AMHgeneknockout in themouse showsnoobvious effecton the gubernacular cord.

Another difference is the timing of the 2 stages; bothstages are prenatal in humans, but only the first stage isprenatal in rodents, despite the fact that the critical win-

dow for androgenic action in the second, postnatal stagein rodents occurs prenatally. Remodeling of the guber-nacular bulb occurs at a different time, but all the keyfeatures of gubernacular migration during inguinoscrotaldescent seem the same.

One other difference—the timing of a stable scrotalposition of the testis—has important implications for ex-trapolating from rodent experiments to humans. Becausethe rodent testis is not stably located in the scrotum untilthe onset of puberty, models of cryptorchidism in rodentsmore closely mimic acquired human cryptorchidism thancongenital UDT. More importantly, the change in testic-ular temperature from body core to scrotal, which occurson the day of birth in normal boys, is delayed until 2–4weeks of age in rodents. This means that early postnatalgonocyte maturation normally occurs at 37°C in rats com-pared with 33°C in humans, and that models of UDT inrats will not impair early gonocyte maturation or apopto-sis. This may be the reason that rodent models of UDTcause impaired fertility, with temperature-dependent lossof stem cells but no risk of malignancy, because unlikehumans, there are no persisting, abnormal gonocytes.

VIII. Conclusions

The anatomical steps in testicular descent appear evermore complex as the mechanisms of gubernacular migra-tion to the scrotum yield gradually to intense study. Therole of INSL3 is now well documented in the transabdom-inal phase, whereas the mechanism of androgenic regula-tion of the inguinoscrotal phase remains controversial.Our own preliminary evidence suggests a role for themammary line in triggering inguinoscrotal migration,with possible trophic stimulation of the GFN to produceCGRP, which appears to be a key factor in stimulatinggubernacular growth, differentiation, and orientation to-ward the scrotum.

The causes of cryptorchidism continue to be elusive butmay be in suboptimal androgenic production leading toinadequate trophic stimulation of the GFN, which has thepotential to cause unilateral cryptorchidism. A review ofthe recently described molecular causes of cryptorchidismin recognized syndromes reveals a number of key signalingpathways.

As we learn more about germ cell development in thepostnatal testis, evidence is accumulating to support earlyorchidopexy in the first year to allow normal transforma-tion of the neonatal gonocyte into an AD-S. The latter isnow thought to be the stem cell for spermatogenesis. Wespeculate that inadequate transformation, secondary to anabnormally high temperature of the UDT, leads to a de-



ficient stem cell pool and later infertility. Furthermore,inhibited apoptosis of remaining gonocytes may allowsome to persist to eventually become carcinoma in situ andfrank malignancy in adulthood.

Acquired cryptorchidism is now accepted in surgicalcircles as the reason for many boys coming to orchidopexylater in childhood, although whether this is necessary re-mains controversial. Animal models of cryptorchidism areactually similar to human acquired UDT, but the testiculartemperature in rodents with UDT is not abnormal untilwell after normal formation of spermatogenic stem cells.

Finally, hormonal therapy remains extremely contro-versial, with evidence both for and against it. There maybe specific cases where hormone treatment may facilitategerm cell maturation after early orchidopexy, althoughidentifying these subgroups is a task for future research.

Given the fact that recent research has caused everyoneto rethink their views about normal descent and cryp-torchidism, we can look forward to the next few yearsbeing exciting and full of surprises.

Acknowledgments

Address all correspondence and requests for reprints to: Dr J. M. Hutson,Urology Department, Royal Children’s Hospital, Parkville 3052, Vic-toria, Australia. E-mail: [email protected].

This project was supported by the Victorian Government’s Opera-tional Infrastructure Support Program and National Health & MedicalResearch Council (Australia) Grant 607365.

Disclosure Summary: The authors have nothing to declare.

References

1. Hutson JM, Hasthorpe S. Testicular descent and cryp-torchidism: the state of the art in 2004. J Pediatr Surg.2005;40:297–302.

2. Hutson JM, Baker M, Terada M, Zhou B, Paxton G. Hor-monal control of testicular descent and the cause of cryp-torchidism. Reprod Fertil Dev. 1994;6:151–156.

3. Backhouse KM. The gubernaculum testis Hunteri: testic-ular descent and maldescent. Ann R Coll Surg Engl. 1964;35:15–33.

4. Hutson JM, Hasthorpe S, Heyns CF. Anatomical and func-tional aspects of testicular descent and cryptorchidism. En-docr Rev. 1997;18:259–280.

5. Heyns CF, Human HJ, Werely CJ, De Klerk DP. The gly-cosaminoglycans of the gubernaculum during testicular de-scent in the fetus. J Urol. 1990;143:612–617.

6. Heyns CF, De Klerk DP. The gubernaculum during testic-ular descent in the pig fetus. J Urol. 1985;133:694–699.

7. Bay K, Cohen AS, Jorgensen FS, et al. Insulin-like factor 3levels in second-trimester amniotic fluid. J Clin EndocrinolMetab. 2008;93:4048–4051.

8. Amann RP, Veeramachaneni DN. Cryptorchidism in com-

mon eutherian mammals. Reproduction. 2007;133:541–561.

9. van der Schoot P. Doubts about the ‘first phase of testisdescent’ in the rat as a valid concept. Anat Embryol (Berl).1993;187:203–208.

10. Barteczko KJ, Jacob MI. The testicular descent in human.Origin, development and fate of the gubernaculum Hunt-eri, processus vaginalis peritonei, and gonadal ligaments.Adv Anat Embryol Cell Biol. 2000;156:III-X, 1–98.

11. Chen N, Harisis GN, Farmer P, et al. Gone with the Wnt:the canonical Wnt signaling axis is present and androgendependent in the rodent gubernaculum. J Pediatr Surg.2011;46:2363–2369.

12. Hutson JM, Terada M, Zhou B, Williams MP. Normaltesticular descent and the aetiology of cryptorchidism. AdvAnat Embryol Cell Biol. 1996;132:1–56.

13. Kaplan LM, Koyle MA, Kaplan GW, Farrer JH, Rajfer J.Association between abdominal wall defects and cryp-torchidism. J Urol. 1986;136:645–647.

14. Koivusalo A, Taskinen S, Rintala RJ. Cryptorchidism inboys with congenital abdominal wall defects. Pediatr SurgInt. 1998;13:143–145.

15. Buraundi S, Balic A, Farmer PJ, Southwell BR, Hutson JM.Gubernacular development in the mouse is similar to therat and suggests that the processus vaginalis is derived fromthe urogenital ridge and is different from the parietal peri-toneum. J Pediatr Surg. 2011;46:1804–1812.

16. Hutson J, Nation T, Balic A, Southwell BR. The role of thegubernaculum in the descent and undescent of the testis.Ther Adv Urol. 2009;1:115–121.

17. Tomiyama H, Sasaki Y, Huynh J, Yong E, Ting A, HutsonJM. Testicular descent, cryptorchidism and inguinal her-nia: the Melbourne perspective. J Pediatr Urol. 2005;1:11–25.

18. Harnaen EJ, Na AF, Shenker NS, et al. The anatomy of thecremaster muscle during inguinoscrotal testicular descentin the rat. J Pediatr Surg. 2007;42:1982–1987.

19. Ivell R, Hartung S. The molecular basis of cryptorchidism.Mol Hum Reprod. 2003;9:175–181.

20. Adham IM, Agoulnik AI. Insulin-like 3 signalling in tes-ticular descent. Int J Androl. 2004;27:257–265.

21. Zimmermann S, Steding G, Emmen JM, et al. Targeteddisruption of the Insl3 gene causes bilateral cryptorchid-ism. Mol Endocrinol. 1999;13:681–691.

22. Nef S, Parada LF. Cryptorchidism in mice mutant for Insl3.Nat Genet. 1999;22:295–299.

23. Ivell R, Hartung S, Anand-Ivell R. Insulin-like factor 3:where are we now? Ann NY Acad Sci. 2005;1041:486–496.

24. Kubota Y, Nef S, Farmer PJ, Temelcos C, Parada LF, Hut-son JM. Leydig insulin-like hormone, gubernacular devel-opment and testicular descent. J Urol. 2001;165:1673–1675.

25. Kubota Y, Temelcos C, Bathgate RAD, et al. The role ofinsulin 3, testosterone, Mullerian inhibiting substance andrelaxin in rat gubernacular growth. Mol Hum Reprod.2002;8:900–905.

26. Lewis LG. Cryptorchism. J Urol. 1948;60:345–356.27. Adham IM, Steding G, Thamm T, et al. The overexpres-

sion of the insl3 in female mice causes descent of the ova-ries. Mol Endocrinol. 2002;16:244–252.



mailto:[email protected]

28. Kaftanovskaya EM, Feng S, Huang Z, et al. Suppression ofinsulin-like 3 receptor reveals the role of �-catenin andNotch signaling in gubernaculum development. Mol En-docrinol. 2011;25:170–183.

29. Bay K, Main KM, Toppari J, Skakkebaek NE. Testiculardescent: INSL3, testosterone, genes and the intrauterinemilieu. Nat Rev Urol. 2011;8:187–196.

30. Hsu SY, Nakabayashi K, Nishi S, et al. Activation of or-phan receptors by the hormone relaxin. Science. 2002;295:671–674.

31. Tomiyama H, Hutson JM, Truong A, Agoulnik AI. Trans-abdominal testicular descent is disrupted in mice with de-letion of insulin-like factor 3 receptor. J Pediatr Surg. 2003;38:1793–1798.

32. Gorlov IP, Kamat A, Bogatcheva NV, et al. Mutations ofthe GREAT gene cause cryptorchidism. Hum Mol Genet.2002;11:2309–2318.

33. Yamanaka J, Metcalfe SA, Hutson JM, Mendelsohn FA.Testicular descent. II. Ontogeny and response to denerva-tion of calcitonin gene-related peptide receptors in neona-tal rat gubernaculum. Endocrinology. 1993;132:280–284.

34. Visser JH, Heyns CF. Proliferation of gubernaculum cellsinduced by a substance of low molecular mass obtainedfrom fetal pig testes. J Urol. 1995;153:516–520.

35. Bay K, Andersson AM. Human testicular insulin-like fac-tor 3: in relation to development, reproductive hormonesand andrological disorders. Int J Androl. 2010;34:97–109.

36. Lee MM, Donahoe PK. Mullerian inhibiting substance: agonadal hormone with multiple functions. Endocr Rev.1993;14:152–164.

37. Josso N, Fekete C, Cachin O, Nezelof C, Rappaport R.Persistence of Mullerian ducts in male pseudohermaphro-ditism and its relationship to cryptorchidism. Clin Endo-crinol (Oxf). 1983;19:247–258.

38. Lane AH, Donahoe PK. New insights into Mullerian in-hibiting substance and its mechanism of action. J Endo-crinol. 1998;158:1–6.

39. Hutson JM, Baker ML. A hypothesis to explain abnormalgonadal descent in persistent Mullerian duct syndrome.Pediatr Surg Int. 1994;9:542–543.

40. Bartlett JE, Lee SM, Mishina Y, et al. Gubernacular devel-opment in Mullerian inhibiting substance receptor-defi-cient mice. BJU Int. 2002;89:113–118.

41. Tuttelmann F, Dykstra N, Themmen AP, Visser JA, Ni-eschlag E, Simoni M. Anti-Mullerian hormone in men withnormal and reduced sperm concentration and men withmaldescended testes. Fertil Steril. 2009;91:1812–1819.

42. Misra M, MacLaughlin DT, Donahoe PK, Lee MM. Mea-surement of Mullerian inhibiting substance facilitates man-agement of boys with microphallus and cryptorchidism.J Clin Endocrinol Metab. 2002;87:3598–3602.

43. Yamanaka J, Baker M, Metcalfe S, Hutson JM. Serumlevels of Mullerian inhibiting substance in boys with cryp-torchidism. J Pediatr Surg. 1991;26:621–623.

44. Lee MM, Misra M, Donahoe PK, MacLaughlin DT. MIS/AMH in the assessment of cryptorchidism and intersexconditions. Mol Cell Endocrinol. 2003;211:91–98.

45. Husmann DA, McPhaul MJ. Reversal of flutamide-in-duced cryptorchidism by time-specific androgens. Endo-crinology. 1992;131:1711–1715.

46. Shono T, Suita S, Kai H, Yamaguchi Y. The effect of aprenatal androgen disruptor, vinclozolin, on gubernacularmigration and testicular descent in rats. J Pediatr Surg.2004;39:213–216.

47. Shono T, Ramm-Anderson S, Goh DW, Hutson JM. Theeffect of flutamide on testicular descent in rats examined byscanning electron microscopy. J Pediatr Surg. 1994;29:839–844.

48. Hutson JM, Warne GL, Grover SR. Disorders of sex de-velopment: an integrated approach to management. Berlin:Springer; 2012.

49. Momose Y, Griffiths AL, Hutson JM. Testicular descentIII. The neonatal mouse gubernaculum shows rhythmiccontraction in organ culture in response to calcitonin gene-related peptide. Endocrinology 1992;131:2881–2884.

50. Hadziselimovic F. Cryptorchidism. Berlin, Heidelberg,New York: Springer-Verlag; 1983.

51. Thorup J, Petersen BL, Kvist K, Cortes D. Bilateral unde-scended testes classified according to preoperative andpostoperative status of gonadotropins and inhibin B in re-lation to testicular histopathology at bilateral orchiopexyin infant boys. J Urol. 2012;188:1436–1442.

52. Welsh M, Saunders PT, Fisken M, et al. Identification inrats of a programming window for reproductive tract mas-culinization, disruption of which leads to hypospadias andcryptorchidism. J Clin Invest. 2008;118:1479–1490.

53. Beasley SW, Hutson JM. The role of the gubernaculum intesticular descent. J Urol. 1988;140:1191.

54. Beasley SW, Hutson JM. Effect of division of genitofem-oral nerve on testicular descent in the rat. Aust N Z J Surg.1987;57:49–51.

55. Hutson JM, Watts LM, Farmer PJ. Congenital unde-scended testes in neonatal pigs and the effect of exogenouscalcitonin gene-related peptide. J Urol. 1998;159:1025–1028.

56. Hrabovszky Z, Farmer PJ, Hutson JM. Does the sensorynucleus of the genitofemoral nerve have a role in testiculardescent? J Pediatr Surg. 2000;35:96–100.

57. Park W-H, Hutson JM. The gubernaculum shows rhyth-mic contractility and active movement during testiculardescent. J Pediatr Surg. 1991;26:615–617.

58. Choi RC, Ting AK, Lau FT, et al. Calcitonin gene-relatedpeptide induces the expression of acetylcholinesterase-as-sociated collagen ColQ in muscle: a distinction in drivingtwo different promoters between fast- and slow-twitchmuscle fibers. J Neurochem. 2007;102:1316–1328.

59. Griffiths AL, Momose Y, Hutson JM. The gubernaculumin adult female, adult male and TFM male mice. Int J An-drol. 1993;16:380–384.

60. Hrabovszky Z, Hutson JM. Capsaicin restores gubernacu-lar contractility in TS rats. J Pediatr Surg. 1999;34:1769–1772.

61. Schwindt B, Farmer PJ, Watts LM, Hrabovszky Z, HutsonJM. Localization of calcitonin gene-related peptide withinthe genitofemoral nerve in immature rats. J Pediatr Surg.1999;34:986–991.

62. Shenker NS, Huynh J, Farmer PJ, Hutson JM. A new rolefor androgen in testicular descent: permitting gubernacularcell proliferation in response to the neuropeptide, calci-tonin gene-related peptide. J Pediatr Surg. 2006;41:407–412.



63. Yong EX, Huynh J, Farmer P, et al. Calcitonin gene-relatedpeptide stimulates mitosis in the tip of the rat gubernacu-lum in vitro and provides the chemotactic signals to controlgubernacular migration during testicular descent. J PediatrSurg. 2008;43:1533–1539.

64. Tomiyama H, Hutson JM. Contractility of rat gubernaculaaffected by calcitonin gene-related peptide and �-agonist.J Pediatr Surg. 2005;40:683–687.

65. Terada M, Goh DW, Farmer PJ, Hutson JM. Ontogeny ofgubernacular contraction and effect of calcitonin gene-re-lated peptide in the mouse. J Pediatr Surg. 1994;29:609–611.

66. Hrabovszky Z, Farmer PJ, Hutson JM. Undescended testisis accompanied by calcitonin gene related peptide accumu-lation within the sensory nucleus of the genitofemoralnerve in trans-scrotal rats. J Urol. 2001;165:1015–1018.

67. Park W-H, Hutson JM. A new inbred rat strain (TS) withsuprainguinal ectopic testes: a model for human cryp-torchidism. Pediatr Surg Int. 1991;6:172–175.

68. Abe T, Hutson JM. Calcitonin gene-related peptide in-jected ectopically alters gubernacular migration in the flu-tamide-treated rat with cryptorchidism. Pediatr Surg Int.1994;9:551–554.

69. Shono T, Hutson JM. Capsaicin increases the frequency ofcryptochidism in flutamide-treated rats. J Urol. 1994;152:763–765.

70. Hutson JM, Albano FR, Paxton G, et al. In vitro fusion ofhuman inguinal hernia with associated epithelial transfor-mation. Cells Tissues Organs. 2000;166:249–258.

71. Gier HT, Marion GB. Development of mammalian testesand genital ducts. Biol Reprod. 1969;1:1–23.

72. Ramasamy M, Di Pilla N, Yap T, Hrabovszky Z, FarmerPJ, Hutson JM. Enlargement of the processus vaginalis dur-ing testicular descent in rats. Pediatr Surg Int. 2001;17:312–315.

73. Moore KL, Agur A. Essential clinical anatomy. Baltimore,MD: Lippincort Williams and Wilkins; 2007.

74. Standring S. Gray’s anatomy. The anatomical basis of clin-ical practice. 40th ed. Edinburgh: Elsevier Church Living-stone; 2005.

75. Hrabovszky Z, Di Pilla N, Yap T, Farmer PJ, Hutson JM,Carlin JB. Role of the gubernacular bulb in cremaster mus-cle development of the rat. Anat Rec. 2002;267:159–165.

76. Sanders N, Buraundi S, Balic A, Southwell BR, Hutson JM.Cremaster muscle myogenesis in the tip of the rat guber-naculum supports active gubernacular elongation duringinguinoscrotal testicular descent. J Urol. 2011;186:1606–1613.

77. Lie G, Hutson JM. The role of cremaster muscle in testic-ular descent in humans and animal models. Pediatr SurgInt. 2011;27:1255–1265.

78. Summerbell D, Lewis JH, Wolpert L. Positional informa-tion in chick limb morphogenesis. Nature. 1973;244:492–496.

79. Mariani FV, Martin GR. Deciphering skeletal patterning:clues from the limb. Nature. 2003;423:319–325.

80. Huynh J, Shenker NS, Nightingale S, Hutson JM. Signal-ling molecules: clues from development of the limb bud forcryptorchidism? Pediatr Surg Int. 2007;23:617–624.

81. Goff DJ, Tabin CJ. Analysis of Hoxd-13 and Hoxd-11misexpression in chick limb buds reveals that Hox genes

affect both bone condensation and growth. Development.1997;124:627–636.

82. Na AF, Harnaen EJ, Farmer PJ, Sourial M, Southwell BR,Hutson JM. Cell membrane and mitotic markers show thatthe neonatal rat gubernaculum grows in a similar way to anembryonic limb bud. J Pediatr Surg. 2007;42:1566–1573.

83. Casanova JC, Uribe V, Badia-Careaga C, Giovinazzo G,Torres M, Sanz-Ezquerro JJ. Apical ectodermal ridge mor-phogenesis in limb development is controlled by Arid3b-mediated regulation of cell movements. Development.2011;138:1195–1205.

84. Dudley AT, Ros MA, Tabin CJ. A re-examination of proxi-modistal patterning during vertebrate limb development.Nature. 2002;418:539–544.

85. Wolpert L. Vertebrate limb development and malforma-tions. Pediatr Res. 1999;46:247–254.

86. Costa W, Sampaio FJB, Favorito L, Cardoso LE. Testicularmigration: remodeling of connective tissue and muscle cellsin human gubernaculum testis. J Urol. 2002;167:2171–2176.

87. Chan JJ, Farmer PJ, Southwell BR, Sourial M, Hutson JM.Calcitonin gene-related peptide is a survival factor, inhib-iting apoptosis in neonatal rat gubernaculum in vitro. J Pe-diatr Surg. 2009;44:1497–1501.

88. Ng Y, Farmer P, Sourial M, Southwell B, Hutson J. Growthof the rat gubernaculum in vitro and localisation of itsgrowth centre. J Pediatr Surg. 2009;44:422–426.

89. Nightingale SS, Western P, Hutson JM. The migrating gu-bernaculum grows like a “limb bud.” J Pediatr Surg. 2008;43:387–390.

90. Nagraj S, Seah GJ, Farmer PJ, et al. The development andanatomy of the gubernaculum in Hoxa11 knockout mice.J Pediatr Surg. 2011;46:387–392.

91. Nation T, Balic A, Buraundi S, et al. The antiandrogenflutamide perturbs inguinoscrotal testicular descent in therat and suggests a link with mammary development. J Pe-diatr Surg. 2009;44:2330–2334.

92. Terada M, Goh DW, Farmer PJ, Hutson JM. Calcitoningene-related peptide receptors in the gubernaculum of nor-mal rat and 2 models of cryptorchidism. J Urol. 1994;152:759.

93. Nation T, Buraundi S, Balic A, Southwell B, Newgreen D,Hutson J. Androgen and estrogen receptor expression inthe spinal segments of the genitofemoral nerve during tes-ticular descent. J Pediatr Surg. 2011;46:1539–1543.

94. Nation TR, Buraundi S, Balic A, et al. The effect of flut-amide on expression of androgen and estrogen receptors inthe gubernaculum and surrounding structures during tes-ticular descent. J Pediatr Surg. 2011;46:2358–2362.

95. Soito IC, Favorito LA, Costa WS, Sampaio FJ, Cardoso LE.Extracellular matrix remodeling in the human gubernac-ulum during fetal testicular descent and in cryptorchidicchildren. World J Urol. 2011;29:535–540.

96. Vigueras RM, Reyes G, Moreno-Mendoza N, Merchant-Larios H. Gubernacular fibroblasts express the androgenreceptor during testis descent in cryptorchid rats treatedwith human chorionic gonadotrophin. Urol Res. 2004;32:386–390.

97. Coveney D, Shaw G, Hutson JM, Renfree MB. Effect of ananti-androgen on testicular descent and inguinal closure in



a marsupial, the tammar wallaby (Macropus eugenii). Re-production. 2002;124:865–874.

98. Renfree MB, O WS, Short RV, Shaw G. Sexual differen-tiation of the urogenital system of the fetal and neonataltammar wallaby, Macropus eugenii. Anat Embryol (Berl).1996;194:111–134.

99. Allnutt B, Buraundi S, Farmer P, Southwell BR, HutsonJM, Balic A. The common fetal development of the mam-mary fat pad and gubernaculum. J Pediatr Surg. 2011;46:378–383.

100. Su S, Farmer PJ, Li R, et al. Regression of the mammarybranch of the genitofemoral nerve may be necessary fortesticular descent in rats. J Urol. 2012;188(4 suppl):1443–1448.

101. Breedlove SM, Arnold AP. Sexually dimorphic motor nu-cleus in the rat lumbar spinal cord: response to adult hor-mone manipulation, absence in androgen-insensitive rats.Brain Res. 1981;225:297–307.

102. Sengelaub DR, Forger NG. The spinal nucleus of the bul-bocavernosus: firsts in androgen-dependent neural sex dif-ferences. Horm Behav. 2008;53:596–612.

103. Yang LY, Verhovshek T, Sengelaub DR. Brain-derivedneurotrophic factor and androgen interact in the mainte-nance of dendritic morphology in a sexually dimorphic ratspinal nucleus. Endocrinology. 2004;145:161–168.

104. Forger NG, Wong V, Breedlove SM. Ciliary neurotrophicfactor arrests muscle and motoneuron degeneration in an-drogen-insensitive rats. J Neurobiol. 1995;28:354–362.

105. Al Shareef Y, Sourial M, Hutson JM. Exogenous calcitoningene-related peptide perturbs the direction and length ofgubernaculum in capsaicin-treated rats. Pediatr Surg Int.2007;23:305–308.

106. Bernard P, Ryan J, Sim H, et al. Wnt signaling in ovariandevelopment inhibits Sf1 activation of Sox9 via the Tescoenhancer. Endocrinology. 2012;153:901–912.

107. Hadziselimovic F, Huff D. Gonadal differentiation—nor-mal and abnormal testicular development. In: Zderic SA,Canning DA, Carr MC, Snyder HM, eds. Pediatric GenderAssignment: A Critical Reappraisal. New York, NY: Klu-wer Academic/Plenum; 2002:15–23.

108. Hadziselimovic F, Hadziselimovic NO, Demougin P, KreyG, Oakeley EJ. Deficient expression of genes involved inthe endogenous defense system against transposons incryptorchid boys with impaired mini-puberty. Sex Dev.2011;5:287–293.

109. Hutson JM, Li R, Southwell BR, Petersen BL, Thorup J,Cortes D. Germ cell development in the postnatal testis: thekey to prevent malignancy in cryptorchidism? Front En-docrinol (Lausanne). 2012;3:176.

110. Cortes D, Thorup JM, Beck BL. Quantitative histology ofgerm cells in the undescended testes of human fetuses, ne-onates and infants. J Urol. 1995;154:1188–1192.

111. Job JC, Toublanc JE, Chaussain JL, Gendrel D, Roger M,Canlorbe P. The pituitary-gonadal axis in cryptorchid in-fants and children. Eur J Pediatr. 1987;146(Suppl 2):S2–S5.

112. Baker ML, Metcalfe SA, Hutson JM. Serum levels of Mul-lerian inhibiting substance in boys from birth to 18 years,as determined by enzyme immunoassay. J Clin EndocrinolMetab. 1990;70:11–15.

113. Hadziselimovic F, Zivkovic D, Bica DT, Emmons LR. The

importance of mini-puberty for fertility in cryptorchidism.J Urol. 2005;174:1536–1539.

114. Hadziselimovic F, Emmons LR, Buser MW. A diminishedpostnatal surge of Ad spermatogonia in cryptorchid in-fants is additional evidence for hypogonadotropic hypo-gonadism. Swiss Med Wkly. 2004;134:381–384.

115. Hrabovszky Z, Hutson JM. Androgen imprinting of thebrain in animal models and humans with intersex disor-ders: review and recommendations. J Urol. 2002;168:2142–2148.

116. Clarnette TD, Rowe D, Hasthorpe S, Hutson JM. Incom-plete disappearance of the processus vaginalis as a cause ofascending testis. J Urol. 1997;157:1889–1891.

117. Puri P, Barton D, O’Donnell B. Prepubertal testicular tor-sion: subsequent fertility. J Pediatr Surg. 1985;20:598–601.

118. Hutson JM, Balic A, Nation T, Southwell B. Cryptorchid-ism. Semin Pediatr Surg. 2010;19:215–224.

119. Husmann DA, Levy JB. Current concepts in the patho-physiology of testicular undescent. Urology. 1995;46:267–276.

120. Sharpe RM. Pathways of endocrine disruption during malesexual differentiation and masculinization. Best Pract ResClin Endocrinol Metab. 2006;20:91–110.

121. Sharpe RM, Skakkebaek NE. Testicular dysgenesis syn-drome: mechanistic insights and potential new down-stream effects. Fertil Steril. 2008;89(2 suppl):e33–e38.

122. Skakkebaek NE, Rajpert-De Meyts E, Main KM. Testic-ular dysgenesis syndrome: an increasingly common devel-opmental disorder with environmental aspects. Hum Re-prod. 2001;16:972–978.

123. Hadziselimovic F, Duckett JW, Snyder HM III, SchnauferL, Huff D. Omphalocele, cryptorchidism, and brain mal-formations. J Pediatr Surg. 1987;22:854–856.

124. Aliotta PJ, Piedmonte M, Karp M, Greenfield SP. Cryp-torchidism in newborns with gastroschisis and omphalo-cele. Urology. 1992;40:84–86.

125. Terada M, Hutson JM, Farmer PJ, Goh DW. The role ofthe genitofemoral nerve and calcitonin gene-related pep-tide in congenitally cryptorchid mutant TS rats. J Urol.1995;154:734–737.

126. Frey HL, Rajfer J. Role of the gubernaculum and intraab-dominal pressure in the process of testicular descent.J Urol. 1984;131:574–579.

127. Cortes D, Thorup JM, Beck BL, Visfeldt J. Cryptorchidismas a caudal developmental field defect. A new descriptionof cryptorchidism associated with malformations and dys-plasias of the kidneys, the ureters and the spine from T10to S5. APMIS. 1998;106:953–958.

128. Diao B, Diallo Y, Fall PA, et al. Prune belly syndrome:epidemiologic, clinic and therapeutic aspects [in French].Prog Urol. 2008;18:470–474.

129. Hensle TW, Deibert CM. Adult male health risks associ-ated with congenital abnormalities. Urol Clin North Am.2012;39:109–114.

130. Chen IL, Huang H-C, Lee S-Y, et al. Urachal catheter pro-vides new choice for long-term urinary diversion in prunebelly syndrome. Urology. 2011;77:466–468.

131. Holmdahl G, Sillen U. Boys with posterior urethral valves:outcome concerning renal function, bladder function and



paternity at ages 31 to 44 years. J Urol. 2005;174:1031–1034.

132. Tasian GE, Zaid H, Cabana MD, Baskin LS. Proximalhypospadias and risk of acquired cryptorchidism. J Urol.2010;184:715–720.

133. Robinow M, Silverman FN, Smith HD. A newly recog-nized dwarfing syndrome. Am J Dis Child. 1969;117:645–651.

134. van Bokhoven H, Celli J, Kayserili H, et al. Mutation of thegene encoding the ROR2 tyrosine kinase causes autosomalrecessive Robinow syndrome. Nat Genet. 2000;25:423–426.

135. Mazzeu JF, Pardono E, Vianna-Morgante AM, et al. Clin-ical characterization of autosomal dominant and recessivevariants of Robinow syndrome. Am J Med Genet A. 2007;143:320–325.

136. Minami Y, Oishi I, Endo M, Nishita M. Ror-family recep-tor tyrosine kinases in noncanonical Wnt signaling: theirimplications in developmental morphogenesis and humandiseases. Dev Dyn. 239:1–15.

137. Mikels AJ, Nusse R. Purified Wnt5a protein activates orinhibits �-catenin-TCF signaling depending on receptorcontext. PLoS Biol. 2006;4:e115.

138. Afzal AR, Jeffery S. One gene, two phenotypes: ROR2mutations in autosomal recessive Robinow syndrome andautosomal dominant brachydactyly type B. Hum Mutat.2003;22:1–11.

139. Person AD, Beiraghi S, Sieben CM, et al. WNT5A muta-tions in patients with autosomal dominant Robinow syn-drome. Dev Dyn 239:327–337.

140. Warr N, Siggers P, Bogani D, et al. Sfrp1 and Sfrp2 arerequired for normal male sexual development in mice. DevBiol. 2009;326:273–284.

141. Johnson KJ, Robbins AK, Wang Y, McCahan SM, ChackoJK, Barthold JS. Insulin-like 3 exposure of the fetal ratgubernaculum modulates expression of genes involved inneural pathways. Biol Reprod. 83:774–782.

142. Terada M, Goh DW, Farmer PJ, Hutson JM. Calcitoningene-related peptide receptors in the gubernaculum of nor-mal rat and 2 models of cryptorchidism. J Urol. 1994;152(2 Pt 2):759–762.

143. Porteous ME, Goudie DR. Aarskog syndrome. J MedGenet. 1991;28:44–47.

144. Fryns JP. Aarskog syndrome: the changing phenotype withage. Am J Med Genet. 1992;43:420–427.

145. Pasteris NG, Cadle A, Logie LJ, et al. Isolation and char-acterization of the faciogenital dysplasia (Aarskog-Scottsyndrome) gene: a putative Rho/Rac guanine nucleotideexchange factor. Cell. 1994;79:669–678.

146. Schwartz CE, Gillessen-Kaesbach G, May M, et al. Twonovel mutations confirm FGD1 is responsible for the Aar-skog syndrome. Eur J Hum Genet. 2000;8:869–874.

147. Zheng Y, Fischer DJ, Santos MF, et al. The faciogenitaldysplasia gene product FGD1 functions as a Cdc42Hs-specific guanine-nucleotide exchange factor. J Biol Chem.1996;271:33169–33172.

148. Hou P, Estrada L, Kinley AW, Parsons JT, Vojtek AB,Gorski JL. Fgd1, the Cdc42 GEF responsible for faciogeni-tal dysplasia, directly interacts with cortactin and mAbp1to modulate cell shape. Hum Mol Genet. 2003;12:1981–1993.

149. Coso OA, Chiariello M, Yu J-C, et al. The small GTP-binding proteins Rac1 and Cdc42 regulate the activity ofthe JNK/SAPK signaling pathway. Cell. 1995;81:1137–1146.

150. Schlessinger K, McManus EJ, Hall A. Cdc42 and nonca-nonical Wnt signal transduction pathways cooperate topromote cell polarity. J Cell Biol. 2007;178:355–361.

151. Tartaglia M, Gelb BD. Noonan syndrome and related dis-orders: genetics and pathogenesis. Annu Rev GenomicsHum Genet. 2005;6:45–68.

152. Sarkozy A, Conti E, Digilio MC, et al. Clinical and mo-lecular analysis of 30 patients with multiple lentiginesLEOPARD syndrome. J Med Genet. 2004;41:e68.

153. Sarkozy A, Digilio MC, Dallapiccola B. Leopard syn-drome. Orphanet J Rare Dis. 2008;3:13.

154. Roberts AE, Araki T, Swanson KD, et al. Germline gain-of-function mutations in SOS1 cause Noonan syndrome.Nat Genet. 2007;39:70–74.

155. Schubbert S, Zenker M, Rowe SL, et al. Germline KRASmutations cause Noonan syndrome. Nat Genet. 2006;38:331–336.

156. Razzaque MA, Nishizawa T, Komoike Y, et al. Germlinegain-of-function mutations in RAF1 cause Noonan syn-drome. Nat Genet. 2007;39:1013–1017.

157. Tartaglia M, Mehler EL, Goldberg R, et al. Mutations inPTPN11, encoding the protein tyrosine phosphataseSHP-2, cause Noonan syndrome. Nat Genet. 2001;29:465–468.

158. Kontaridis MI, Swanson KD, David FS, Barford D, NeelBG. PTPN11 (Shp2) mutations in LEOPARD syndromehave dominant negative, not activating, effects. J BiolChem. 2006;281:6785–6792.

159. Lindsey S, Huang W, Wang H, Horvath E, Zhu C, EklundEA. Activation of SHP2 protein-tyrosine phosphatase in-creases HoxA10-induced repression of the genes encodinggp91(PHOX) and p67(PHOX). J Biol Chem. 2007;282:2237–2249.

160. Rijli FM, Matyas R, Pellegrini M, et al. Cryptorchidismand homeotic transformations of spinal nerves and verte-brae in Hoxa-10 mutant mice. Proc Natl Acad Sci USA.1995;92:8185–8189.

161. Liu J, Krantz ID. Cornelia de Lange syndrome, cohesin,and beyond. Clin Genet. 2009;76:303–314.

162. Kline AD, Krantz ID, Sommer A, et al. Cornelia de Langesyndrome: clinical review, diagnostic and scoring systems,and anticipatory guidance. Am J Med Genet A. 2007;143A:1287–1296.

163. Borck G, Redon R, Sanlaville D, et al. NIPBL mutationsand genetic heterogeneity in Cornelia de Lange syndrome.J Med Genet. 2004;41:e128.

164. Tonkin ET, Wang TJ, Lisgo S, Bamshad MJ, Strachan T.NIPBL, encoding a homolog of fungal Scc2-type sisterchromatid cohesion proteins and fly Nipped-B, is mutatedin Cornelia de Lange syndrome. Nat Genet. 2004;36:636–641.

165. Ferrari D, Harrington A, Dealy CN, Kosher RA. Dlx-5 inlimb initiation in the chick embryo. Dev Dyn. 1999;216:10–15.

166. Stock DW, Ellies DL, Zhao Z, Ekker M, Ruddle FH, WeissKM. The evolution of the vertebrate Dlx gene family. ProcNatl Acad Sci USA. 1996;93:10858–10863.



167. Hennekam RC. Rubinstein-Taybi syndrome. Eur J HumGenet. 2006;14:981–985.

168. Hennekam RC, Van Den Boogaard MJ, Sibbles BJ, VanSpijker HG. Rubinstein-Taybi syndrome in The Nether-lands. Am J Med Genet Suppl. 1990;6:17–29.

169. Stevens CA, Carey JC, Blackburn BL. Rubinstein-Taybisyndrome: a natural history study. Am J Med Genet Suppl.1990;6:30–37.

170. Miller RW, Rubinstein JH. Tumors in Rubinstein-Taybisyndrome. Am J Med Genet. 1995;56:112–115.

171. Petrij F, Giles RH, Dauwerse HG, et al. Rubinstein-Taybisyndrome caused by mutations in the transcriptional co-activator CBP. Nature. 1995;376:348–351.

172. Goodman RH, Smolik S. CBP/p300 in cell growth, trans-formation, and development. Genes Dev. 2000;14:1553–1577.

173. Roelfsema JH, White SJ, Ariyurek Y, et al. Genetic heter-ogeneity in Rubinstein-Taybi syndrome: mutations in boththe CBP and EP300 genes cause disease. Am J Hum Genet.2005;76:572–580.

174. Fronsdal K, Engedal N, Slagsvold T, Saatcioglu F. CREBbinding protein is a coactivator for the androgen receptorand mediates cross-talk with AP-1. J Biol Chem. 1998;273:31853–31859.

175. Schinzel A. Ulnar-mammary syndrome. J Med Genet.1987;24:778–781.

176. Klopocki E, Neumann LM, Tonnies H, Ropers HH,Mundlos S, Ullmann R. Ulnar-mammary syndrome withdysmorphic facies and mental retardation caused by anovel 1.28 Mb deletion encompassing the TBX3 gene. EurJ Hum Genet. 2006;14:1274–1279.

177. Bamshad M, Lin RC, Law DJ, et al. Mutations in humanTBX3 alter limb, apocrine and genital development in ul-nar-mammary syndrome. Nat Genet. 1997;16:311–315.

178. Lee HS, Cho HH, Kim HK, Bae YC, Baik HS, Jung JS.Tbx3, a transcriptional factor, involves in proliferationand osteogenic differentiation of human adipose stromalcells. Mol Cell Biochem. 2007;296:129–136.

179. Renard CA, Labalette C, Armengol C, et al. Tbx3 is adownstream target of the Wnt/�-catenin pathway and acritical mediator of �-catenin survival functions in livercancer. Cancer Res. 2007;67:901–910.

180. Balic A, Nation T, Buraundi S, et al. Hidden in plain sight:the mammary line in males may be the missing link regu-lating inguinoscrotal testicular descent. J Pediatr Surg.2010;45:414–418; discussion 418.

181. Kondoh T, Eguchi J, Hamasaki Y, et al. Hearing impair-ment, undescended testis, circumferential skin creases, andmental handicap (HITCH) syndrome: a case report. Am JMed Genet A. 2004;125A:290–292.

182. Churchill JA, Buraundi S, Farmer PJ, et al. Gubernaculumas icebreaker: do matrix metalloproteinases in rodent gu-bernaculum and inguinal fat pad permit testicular descent?J Pediatr Surg. 2011;46:2353–2357.

183. Trussell JC, Lee PA. The relationship of cryptorchidism tofertility. Curr Urol Rep. 2004;5:142–148.

184. Inan M, Aydiner CY, Tokuc B, et al. Prevalence of cryp-torchidism, retractile testis and orchiopexy in school chil-dren. Urol Int. 2008;80:166–171.

185. Barthold JS, Gonzalez R. The epidemiology of congenital

cryptorchidism, testicular ascent and orchiopexy. J Urol.2003;170:2396–2401.

186. Cryptorchidism: an apparent substantial increase since1960. John Radcliffe Hospital Cryptorchidism StudyGroup. Br Med J (Clin Res Ed). 1986;293:1401–1404.

187. Thorup J, McLachlan R, Cortes D, et al. What is new incryptorchidism and hypospadias—a critical review on thetesticular dysgenesis hypothesis. J Pediatr Surg. 2010;45:2074–2086.

188. Ward B, Hunter WM. The absent testicle: a report on asurvey carried out among schoolboys in Nottingham. BrMed J. 1960;1:1110–1111.

189. Boisen KA, Kaleva M, Main KM, et al. Difference in prev-alence of congenital cryptorchidism in infants between twoNordic countries. Lancet. 2004;363:1264–1269.

190. Cortes D, Kjellberg EM, Breddam M, Thorup J. The trueincidence of cryptorchidism in Denmark. J Urol. 2008;179:314–318.

191. Clarnette TD, Hutson JM. Exogenous calcitonin gene-related peptide can change the direction of gubernacularmigration in the mutant trans-scrotal rat. J Pediatr Surg.1999;34:1208–1212.

192. Berkowitz GS, Lapinski RH, Dolgin SE, Gazella JG, Bo-dian CA, Holzman IR. Prevalence and natural history ofcryptorchidism. Pediatrics. 1993;92:44–49.

193. Buemann B, Henriksen H, Villumsen AL, Westh A,Zachau-Christiansen B. Incidence of undescended testis inthe newborn. Acta Chir Scand Suppl. 1961;suppl 283:289–293.

194. Thorup J, Cortes D, Petersen BL. The incidence of bilateralcryptorchidism is increased and the fertility potential isreduced in sons born to mothers who have smoked duringpregnancy. J Urol. 2006;176:734–737.

195. Jensen MS, Toft G, Thulstrup AM, et al. Cryptorchidismconcordance in monozygotic and dizygotic twin brothers,full brothers, and half-brothers. Fertil Steril. 2010;93:124–129.

196. Lala R, Matarazzo P, Chiabotto P, de Sanctis C, CanaveseF, Hadziselimovic F. Combined therapy with LHRH andHCG in cryptorchid infants. Eur J Pediatr. 1993;152(suppl 2):S31–S33.

197. Henna MR, Del Nero RG, Sampaio CZ, et al. Hormonalcryptorchidism therapy: systematic review with metanaly-sis of randomized clinical trials. Pediatr Surg Int. 2004;20:357–359.

198. Huff DS, Fenig DM, Canning DA, Carr MG, Zderic SA,Snyder HM III. 2001 Abnormal germ cell development incryptorchidism. Horm Res 55:11–17.

199. Ong C, Hasthorpe S, Hutson JM. Germ cell developmentin the descended and cryptorchid testis and the effects ofhormonal manipulation. Pediatr Surg Int. 2005;21:240–254.

200. Mengel W, Hienz HA, Sippe WG 2nd, Hecker WC. Studieson cryptorchidism: a comparison of histological findings inthe germinative epithelium before and after the second yearof life. J Pediatr Surg. 1974;9:445–450.

201. Hadziselimovic F, Herzog B, Seguchi H. Surgical correc-tion of cryptorchism at 2 years: electron microscopic andmorphometric investigations. J Pediatr Surg. 1975;10:19–26.

202. Saito S, Kumamoto Y. The number of spermatogonia in



various congenital testicular disorders. J Urol. 1989;141:1166–1168.

203. Huff DS, Hadziselimovic F, Snyder HM III, Blythe B,Ducket JW. Histologic maldevelopment of unilaterallycryptorchid testes and their descended partners. Eur J Pe-diatr. 1993;152(suppl 2):S11–S14.

204. Hadziselimovic F, Hocht B, Herzog B, Buser MW. Infer-tility in cryptorchidism is linked to the stage of germ celldevelopment at orchidopexy. Horm Res. 2007;68:46–52.

205. Hadziselimovic F, Herzog B. The importance of both anearly orchidopexy and germ cell maturation for fertility.Lancet. 2001;358:1156–1157.

206. Hadziselimovic F, Hoecht B. Testicular histology relatedto fertility outcome and postpubertal hormone status incryptorchidism. Klin Padiatr. 2008;220:302–307.

207. Bremholm Rasmussen T, Ingerslev HJ, Høstrup H. Bilat-eral spontaneous descent of the testis after the age of 10:subsequent effects on fertility. Br J Surg. 1988;75:820–823.

208. Cortes D. Cryptorchidism—aspects of pathogenesis, his-tology and treatment. Scand J Urol Nephrol Suppl. 1998;196:1–54.

209. Wood HM, Elder JS. Cryptorchidism and testicular can-cer: separating fact from fiction. J Urol. 2009;181:452–461.

210. Giwercman A, Muller J, Skakkeboek NE. Cryptorchidismand testicular neoplasia. Horm Res. 1988;30:157–163.

211. Giwercman A, Grindsted J, Hansen B, Jensen OM, Skak-kebaek NE. Testicular cancer risk in boys with malde-scended testis: a cohort study. J Urol. 1987;138:1214–1216.

212. Mieusset R, Fouda PJ, Vaysse P, Guitard J, Moscovici J,Juskiewenski S. Increase in testicular temperature in case ofcryptorchidism in boys. Fertil Steril. 1993;59:1319–1321.

213. Giwercman A, Muller J, Skakkebaek NE. Carcinoma insitu of the testis: possible origin, clinical significance, anddiagnostic methods. Recent Results Cancer Res. 1991;123:21–36.

214. Whitaker RH. Neoplasia in cryptorchid men. Semin Urol.1988;6:107–109.

215. Forman D, Pike MC, Davey G, et al. Aetiology of testicularcancer: association with congenital abnormalities, age atpuberty, infertility, and exercise. BMJ. 1994;308:1393–1399.

216. Pyorala S, Huttunen NP, Uhari M. A review and meta-analysis of hormonal treatment of cryptorchidism. J ClinEndocrinol Metab. 1995;80:2795–2799.

217. Moller H. Epidemiological studies of testicular germ cellcancer. Thames cancer registry. London: Kings CollegeLondon; 2001.

218. Guidelines. European Association of Urology. http://www.uroweb.org/guidelines. 2007.

219. McCabe JE, Kenny SE. Orchidopexy for undescended tes-tis in England: is it evidence based? J Pediatr Surg. 2008;43:353–357.

220. Kollin C, Hesser U, Ritzen EM, Karpe B. Testicular growthfrom birth to two years of age, and the effect of orchi-dopexy at age nine months: a randomized, controlledstudy. Acta Paediatr. 2006;95:318–324.

221. Ritzen EM, Bergh A, Bjerknes R, et al. Nordic consensus

on treatment of undescended testes. Acta Paediatr. 2007;96:638–643.

222. Taskinen S, Hovatta O, Wikstrom S. Early treatment ofcryptorchidism, semen quality and testicular endocrinol-ogy. J Urol. 1996;156:82–84.

223. Engeler DS, Hosli PO, John H, et al. Early orchiopexy:prepubertal intratubular germ cell neoplasia and fertilityoutcome. Urology. 2000;56:144–148.

224. Lee PA, Coughlin MT, Bellinger MF. Paternity and hor-mone levels after unilateral cryptorchidism: associationwith pretreatment testicular location. J Urol. 2000;164:1697–1701.

225. Cortes D, Thorup JM, Visfeldt J. Cryptorchidism: aspectsof fertility and neoplasms. A study including data of 1,335consecutive boys who underwent testicular biopsy simul-taneously with surgery for cryptorchidism. Horm Res.2001;55:21–27.

226. Jensen MS, Olsen LH, Thulstrup AM, Bonde JP, Olsen J,Henriksen TB. Age at cryptorchidism diagnosis and or-chiopexy in Denmark: a population based study of508,964 boys born from 1995 to 2009. J Urol. 2011;186:1595–1600.

227. Thomas JW, Marc AD, Mary SC, Peter RC, Paul JT. Pre-pubertal orchiopexy for cryptorchidism may be associatedwith lower risk of testicular cancer. J Urol. 2007;178:1440–1446.

228. Pettersson A, Richiardi L, Nordenskjold A, Kaijser M,Akre O. Age at surgery for undescended testis and risk oftesticular cancer. N Engl J Med. 2007;356:1835–1841.

229. Smith JA, Hutson JM, Beasley SW, Reddihough DS. Therelationship between cerebral palsy and cryptorchidism.J Pediatr Surg. 1989;24:1303–1305.

230. Wohlfahrt-Veje C, Boisen KA, Boas M, et al. Acquiredcryptorchidism is frequent in infancy and childhood. Int JAndrol. 2009;32:423–428.

231. Hack WW, Goede J, van der Voort-Doedens LM, MeijerRW. Acquired undescended testis: putting the pieces to-gether. Int J Androl. 2011;35:41–45.

232. Keys C, Heloury Y. Retractile testes: a review of the currentliterature. J Pediatr Urol. 2012;8:2–6.

233. Cortes D, Petersen BL, Thorup J. Testicular histology incryptorchid boys—aspects of fertility. J Pediatr Surg Spec.2007;1:32–35.

234. Hadziselimovic F, Huff D, Duckett J, et al. Treatment ofcryptorchidism with low doses of buserelin over a6-months period. Eur J Pediatr. 1987;146(suppl 2):S56–S58.

235. Borgwardt G. Bernhard Schapiro (1888–1966): Talmudicscholar - andrologist - pioneer of hormonal treatment forcryptorchidism [in German]. Wurzbg Medizinhist Mitt.2004;23:393–411.

236. Ludwikowski B, Gonzalez R. The controversy regardingthe need for hormonal treatment in boys with unilateralcryptorchidism goes on: a review of the literature. Eur J Pe-diatr. 2013;172:5–8.

237. Paasch U, Thieme C, Grunewald S, Glander HJ. Electronicdata base systems support the evaluation of male infertilityfactors, example cryptorchidism. Urol Int. 2004;72:154–161.

238. Cortes D, Thorup J, Visfeldt J. Hormonal treatment may



http://www.uroweb.org/guidelines

http://www.uroweb.org/guidelines

harm the germ cells in 1 to 3-year-old boys with cryp-torchidism. J Urol. 2000;163:1290–1292.

239. Dunkel L, Taskinen S, Hovatta O, Tilly JL, Wikstrom S.Germ cell apoptosis after treatment of cryptorchidism withhuman chorionic gonadotropin is associated with im-paired reproductive function in the adult. J Clin Invest.1997;100:2341–2346.

240. Jallouli M, Rebai T, Abid N, Bendhaou M, Kassis M,Mhiri R. Neoadjuvant gonadotropin-releasing hormone

therapy before surgery and effect on fertility index in uni-lateral undescended testes: a prospective randomized trial.Urology. 2009;73:1251–1254.

241. Schwentner C, Oswald J, Kreczy A, et al. Neoadjuvantgonadotrophin-releasing hormone therapy before surgerymay improve the fertility index in undescended testes: aprospective randomized trial. J Urol. 2005;173:974–977.

242. Huff DS, Snyder HM III, Rusnack SL, Zderic SA, Carr MC,Canning DA. Hormonal therapy for the subfertility ofcryptorchidism. Horm Res. 2001;55:38–40.

Make sure your patients are getting the best medical care. Learn more about The Endocrine Society’s The Evaluation of Thyroid Nodules Practice

Improvement Module(PIM).www.endoselfassessment.org



The Regulation of Testicular Descent and the …The Regulation of Testicular Descent and the Effects...

Documents

Transcript of The Regulation of Testicular Descent and the …The Regulation of Testicular Descent and the Effects...