, 11:171

Male hypogonadism


  • Andrea M. Isidori
    • Department of Medical Pathophysiology (DFM-Fisiopatologia Medica)‘Sapienza’ University of Rome
  • Elisa Giannetta
    • Department of Medical Pathophysiology (DFM-Fisiopatologia Medica)‘Sapienza’ University of Rome
    • Department of Medical Pathophysiology (DFM-Fisiopatologia Medica)‘Sapienza’ University of Rome

DOI: 10.1007/s11102-008-0111-9

Cite this article as:
Isidori, A.M., Giannetta, E. & Lenzi, A. Pituitary (2008) 11: 171. doi:10.1007/s11102-008-0111-9


The hypothalamic–pituitary–gonadal (HPG) axis regulates the development, endocrine and reproductive function of the gonads throughout all phases of life. Male hypogonadism is defined an inadequate gonadal function, as manifested by deficiency in gametogenesis and/or secretion of gonadal hormones. In most cases, male hypogonadism is diagnosed through detailed history, physical examination and a few basic hormonal evaluations. In selected cases, however, additional tests are needed to define the aetiology and the extent of HPG axis dysfunction. These include semen analysis, pituitary imaging studies, genetic studies, bone densitometry, testicular ultrasonography, testicular biopsy and hormonal dynamic testing. The stimulation tests of the HPG are of particular importance in the differential diagnosis of congenital delayed puberty versus pre-pubertal hypogonadism in children. This review will focus on the methods, indications and limitations of endocrine testing in the characterisation and differential diagnosis of male hypogonadism at various ages. A practical hands-on guide on how to perform these tests is also provided.


PubertyTesticular failureKallmanCryptorchidismInfertilityKlinefelter


The hypothalamic–pituitary–gonadal (HPG) axis regulates the development, endocrine and reproductive function of the gonads throughout all phases of life. In the hypothalamus, specialised neurons release pulses of a releasing hormone (GnRH) that modulates the secretion of gonadotropins from the pituitary gland. In turn, luteinising hormone (LH) and follicle-stimulating hormone (FSH) produced in the anterior pituitary gland stimulate steroid secretion and germ cell production in the testis. HPG axis activity is highly regulated by a complex integration of endogenous inputs, chronobiological signals and exogenous stressors. This reflects the profound changes occurring in gonadal function during life. At birth, gonadotropin levels are elevated and show a pulsatile pattern for the first few months, gradually declining through the first 10 years of life. In boys approaching puberty, the hypothalamic GnRH pulse generator primes the pituitary gland, resulting in increased LH secretion during sleep followed by an elevation of serum LH and FSH as puberty advances. In this phase, GnRH levels pulse approximately once an hour during sleep, and progressively dampen down to every 90–120 min as sexual steroids reach the normal range for men. In post-pubertal life, the GnRH pulse generator maintains the secretory tone of gonadotropins required for normal gonadal function. During senescence, the feedback mechanism of the HPG axis is altered, with decreased neuronal sex steroid signalling, a disorganised GnRH pulsatility and an increase in LH and FSH levels secondary to a primary defect of testicular responsiveness [1, 2].

Male hypogonadism is defined as “inadequate gonadal function, as manifested by deficiency in gametogenesis and/or secretion of gonadal hormones” [3].

It can be classified by age at presentation (pre- or post-pubertal) and according to the site primarily involved: the gonads, the hypothalamus or the pituitary gland.

Male hypogonadism is diagnosed through detailed history, physical examination and a few basic hormonal evaluations. In selected cases, additional tests are needed to define the aetiology and the extent of HPG axis dysfunction. These include semen analysis, pituitary imaging studies, genetic studies, bone densitometry, testicular ultrasonography, testicular biopsy and hormonal dynamic testing. This review will focus on the methods, indications and limitations of endocrine testing in the characterisation and differential diagnosis of male hypogonadism at various ages.

The hypothalamic–pituitary–gonadal axis


The hypothalamic–pituitary–gonadal (HPG) axis is regulated by complex feedback loops (Fig. 1). During foetal maturation, GnRH neurons originate in the olfactory placode and migrate through the cribriform plate of the ethmoid to localise in the hypothalamus. These neurons are linked by synaptic connections to form an integrated circuit: the influence of excitatory and inhibitory neurotransmitters coordinates the synchronous discharge of GnRH from neurons, resulting in discrete waves of GnRH release. This circuit is named the GnRH-pulse generator. The opioid peptides play a predominant role in regulating the frequency of GnRH discharge and the subsequent pulsatile release of LH and FSH from gonadotropic pituitary cells.
Fig. 1

The hypothalamus releases GnRH that stimulates the anterior pituitary to secrete the gonadotropins luteinising hormone (LH) and follicle-stimulating hormone (FSH). LH and FSH bind testicular receptors on the Leydig cells and Sertoli cells respectively, stimulating sex steroids, spermatogenesis and Inhibin production. Sex steroids and Inhibin exert a negative feedback on GnRH, LH and FSH secretion in the hypothalamus and pituitary gland respectively [2, 4, 5]. A further regulatory loop is provided by Activin, produced by Sertoli cells [6], peritubular cells [7] and interstitial calls [8]. Activin stimulates the hypothalamus to release gonadotropin releasing hormone (GnRH), to maintain pituitary FSH secretion [9]. Follistatin, produced by Sertoli and germ cells, inhibits Activin binding to its hypothalamic receptor [9]. Secretion and bioactivity of Activin and Follistatin is also regulated paracrinally by cytokines, in particular IL-1β (the exact mechanisms remain unclear) [10])

It is fascinating to observe that a single cell type—the gonadotrope—produces two different hormones (FSH and LH) under the control of a single hypothalamic hormone, GnRH. In animal models, it has been demonstrated that GnRH pulses modulate the release of the two gonadotropins under a specific time pattern of gene expression stimulation. LH and FSH are heterodimeric glycoproteins, consisting of a common α subunit (shared with TSH) and specific β subunits that confer specificity of action. Differential expression of the subunit genes depends on mRNA concentrations, while GnRH pulse amplitude and frequency activate intracellular signalling mechanisms, resulting in differential gene expression of the two subunits. Animal studies have demonstrated that α and LHβ mRNAs are expressed in response to rapid GnRH pulses every 8–30 min, while a slower frequency, every 120–240 min, maintains FSHβ mRNA transcription. In addition to the effect of the time pattern of GnRH pulses on subunit expression, FSH secretion is differentiated from LH thanks to other hormones, including Activin and Inhibin (Fig. 1). In general, the effects of GnRH on FSHβ transcription are limited, confirming the need of further mechanisms involved in FSHβ mRNA expression [11] (Fig. 1).

GnRH stimulates the release of LH and FSH through a calcium-dependent, cAMP-independent mechanism. Whereas LH interacts with cell surface receptors on Leydig cells to stimulate a membrane-bound adenylate cyclase. Cyclic adenosine monophosphate (cAMP) is subsequently release into the cytoplasm of Leydig cells, where it binds a protein kinase (PK). Activation of PK stimulates the delivery of cholesterol to the inner mitochondrial membrane. In Leydig cells, cholesterol is transported into the mitochondrial by the carrier protein StAR (steroidogenic acute carrier regulatory protein) and is transformed into testosterone by a series of enzymatic reactions, under cytochrome CYP-P450. Testosterone is partially aromatised into oestradiol (E2), and both T and E2 exert a negative feedback on the hypothalamus (GnRH) and pituitary gland (LH). Sertoli cells are the primary site of FSH action. After FSH binds its receptors, an increase in cAMP activates cAMP-dependent PK, which stimulates mRNA transcription of various genes, including androgen binding protein (ABP) and aromatase gene expression. Inhibin B is a glycoprotein synthesised in Sertoli cells and also involved in negative feedback of FSH pituitary secretion. Both FSH and androgens promote Inhibin B synthesis. Serum Inhibin levels are positively correlated with spermatogenesis and inversely with serum FSH levels. In adults, Inhibin B is a clinically useful marker of Sertoli cell function (gonadal reserve).

Development and age-related changes in the HPG axis

During foetal and perinatal life, hypothalamic–pituitary axis activity plays an important role in sexual differentiation. From about the eighth week of gestation, testosterone production by primordial Leydig cells and antimüllerian inhibitory factor (MIF) production by Sertoli cells are stimulated by placental human chorionic gonadotropin (hCG). Testosterone concentration peaks at around 12–13 weeks of gestation in a critical period for normal male sexual differentiation. By mid-gestation, testosterone secretion is under the influence of the foetal HPG axis that, at this stage, is sensitive to the negative feedback of sex hormones. At birth, gonadotropins increase to pubertal levels in response to falling maternal and placental oestrogens. In male infants, rebound in pituitary gonadotropin secretion is followed by a transient surge in testosterone production, which continues for the first 3 months, declining thereafter to very low levels until the onset of puberty [12]. By the age of 6 months, LH and FSH levels in boys start to fall, reaching a nadir at mid-childhood when response to GnRH is minimal. In childhood, LH is secreted in a clear circadian pulsatile pattern, with a greater pulse amplitude and frequency during sleep. As nocturnal pulses of GnRH increase, pituitary secretion of LH is further facilitated. In fact, studies on pituitary responsiveness to GnRH have found that mean serum LH and FSH levels and LH pulse amplitudes are greater after the induction of sexual maturation. These data demonstrate that initiation of puberty by GnRH secretion induces an auto-amplification of the circuitry at both the pituitary and gonadal levels, contributing to the rapid changes needed for sexual maturation in humans [1114]. With the onset of puberty in boys, nocturnal LH pulses increase further in amplitude and frequency. The enhanced LH release stimulates plasma testosterone to almost adult values during sleep, with the inhibitory feedback resulting in decreased GnRH secretion by the next morning [12]. During this phase LH secretion becomes predominant compared to FSH.

As puberty progresses the marked circadian pattern of LH secretion is tempered as daytime pulses increase in amplitude and duration, stimulating the production of testosterone by the Leydig cells. Reflecting the LH secretion pattern, testosterone levels are initially maximum at night while remain low by day; as puberty progresses the adult pattern of testosterone secretion is achieved.

FSH undergoes an increased basal secretion rather than a change in pulse frequency and amplitude. In the first pubertal stage, a sufficient concentration of FSH is necessary to induce proliferation and differentiation of Sertoli cells [15]. In the first few months of life, a peak in Inhibin B levels is observed which remains measurable throughout childhood, with a nadir at 6–10 years. At the beginning of puberty, increased FSH secretion acts synergically with intratesticular testosterone to stimulate the proliferation and maturation of seminiferous tubules, resulting in a more rapid secretion of Inhibin B by Sertoli cells [16]. Interestingly, at start of puberty, Inhibin B levels increase more rapidly than the other endocrine signals associated with reproduction. Our data suggest that there is a biphasic relationship between FSH and Inhibin B: in early puberty, FSH stimulates Inhibin B secretion and both cooperate in the maturation of the gonad; whereas, in late puberty and adult life, Inhibin B retains only its negative feedback on the pituitary gland [16]. Finally, in the adrenarche the adrenal axis and the growth hormone (GH)-insulin like growth factor (IGF-1) axis also contribute to the maturation of the reproductive system. The interaction between the three systems is complex: the increase of sex steroid primes the pituitary gland to produce GH and induces bone growth and maturation, necessary for an optimal pubertal growth spurt. In turn, IGF-I has its own receptors in the gonads and participates in their development as well as in steroid hormone production. Adrenarche is the first endocrine event of puberty and its onset is temporally related, even if its role remains unclear. Adrenarche appear to be independent of the hypothalamus-pituitary-gonadal axis maturation, as it may occur in the absence of gondarche (Turner’s syndrome) or fail to occur while gonadarche progresses normally (Addison’s disease). The surge of adrenal androgens, however, take part in the differentiation of pilosebaceous units and growth of pubic and armpit terminal hair.

Male hypogonadism

Signs and symptoms of hypogonadism depend primarily on the age of onset (Table 1). Hypogonadism is often unrecognised before puberty unless it is associated with growth retardation or other anatomic or endocrine abnormalities. In most cases the first sign is delayed puberty, defined as ‘absence of secondary sexual characteristics at an age more than 2SD above the population mean for the onset of puberty’ [1721](Table 2).
Table 1

Symptoms and signs of hypogonadism related to the onset

Pre-pubertal hypogonadism

Post-pubertal hypogonadism

Small testes, phallus and prostate

Progressive decrease in muscle mass

Scant pubic and axillary hair

Increase in visceral fat mass

Disproportionately long arms and legs (from delayed epiphyseal closure)


Reduced male muscolature

Loss of libido



Persistent high-pitched voice

Oligospermia or azoospermia


Occasionally, menopausal-type hot flushes (with acute onset of hypogonadism)


Poor ability to concentrate


Osteoporosis and increased risk of fractures

Post-pubertal loss of testicular function results in symptoms and signs due to testosterone deficiency, infertility or both conditions. In the ageing male, loss of testicular function is more subtle and sexual dysfunction may be the reason for presentation.

The diagnostic approach to male hypogonadism is based on history and physical examination: testis volume, weight, height, secondary sexual characteristics and bone age determination (according to the Greulich and Pyle atlas). Symptoms of anterior pituitary hormone deficiency or excess should be sought along with any possible mass effects (headache, visual disturbance, bitemporal visual field loss, cranial nerve palsies, cerebrospinal fluid rhinorrhoea). However, most cases of hypogonadotropic hypogonadism are isolated, without other endocrine, neurological or somatic defects.

Clinical data and a small set of basal hormones (Testosterone, LH, FSH oestradiol, SHBG, prolactin, DHT, Inhibin B) are most often sufficient for diagnosis. It should be remembered that FSH and LH are secreted in short pulses, and a single measurement may not be sufficient to clarify the diagnosis. FSH has a longer half-life than LH and is more likely to provide adequate results on a single assessment. Pooling of 2–3 samples taken 20–30 min apart offers a more accurate LH value [3].

The discrimination of hypogonadotropic hypogonadism (HH) and constitutional delayed puberty (CDP) remains difficult, especially when inherited somatic abnormalities such as anosmia, midline deformities, unilateral renal agenesis or micropenis are missing [17, 21, 22]. Several authors [2325] suggest that a basal testosterone >1.7 nmol/l has a good prognostic value in discriminating HH from CDP. In boys who reach this testosterone level, puberty has started spontaneously despite the absence of clinical signs. In contrast, further investigations are always recommended in patients of an appropriate age with a basal testosterone level below 1.7 nmol/l.

Persistent borderline (low/low-normal) values may be further evaluated with dynamic endocrine testing. These tests include GnRH stimulation test, the clomiphene stimulation test and the human chorionic gonadotropin (hCG) stimulation test. These dynamic studies should be prescribed by an experienced endocrinologist and may have limited value unless indicated and interpreted on the basis of solid clinical grounds.

Dynamic tests

Dynamic testing of the HPG axis is required in patients with inconsistent results at baseline evaluation or whose hormone values do not match the clinical presentation. The interpretation of these tests can be difficult, as the ranges of response (and respective cut-off values) are poorly defined and vary with the age of the subjects and several other variables. In the following paragraphs, we provide a practical guide on how to perform these tests, their most widely accepted indication, and hints on their interpretation.

GnRH stimulation test

Indications Investigation of gonadotropin deficiency in: (A) adult men with reduced testosterone levels and normal or low-normal gonadotropins (accepted indication); (B) children with delayed puberty for differential diagnosis between permanent (hypothalamic or pituitary) hypogonadotropic hypogonadism and temporary (hypothalamic) hypogonadism (accepted, but may require priming); (C) adolescents and young adults with varicocele to predict improvement after varicocelectomy (controversial). Contraindications Allergy. Precautions Combination with Insulin Tolerance Test (ITT) is not recommended. Procedure Classic GnRH test: An intravenous cannula must be applied to the patient. Fasting is recommended. The patient receives GnRH 100 mcg intravenously. Blood samples for LH and FSH are taken at 0, 15, 30, 45, 60 and 90 min [2628] (limited sampling can be performed at 0, 20 and 60 min [26, 27]). GnRH priming test: GnRH priming commences 36 h before the classic GnRH test is administered. An i.v. catheter is inserted and a peristaltic pump delivers 50 mcl of a 5 mcg GnRH solution every 90’ for 36 h. The priming is followed by a standard GnRH bolus test (100 mcg) and samples are taken at −30, 0, 15, 30, 60, 90 and 120 min [29]. GnRH analogue: GnRH test can also be performed with a single s.c. injection of 0.1 mg of a GnRH analogue (Buserelin [30, 31], Triptorelin [31], Leuprolide acetate [32], Nafaralin [33]). Literature data [30] suggest that Buserelin is twice as potent as Leuprolide acetate. GnRH analogues have a plasma half-life several times longer than native GnRH (elimination half-life of about 80 min) [34]. Blood for LH and FSH measurement is taken at 0 and 4 h after the injection [31]. The advantage of using the analogues is their priming effect on the pituitary gland. This results in a larger release of gonadotropins in patients with insufficient pituitary priming by endogenous GnRH, such as those with prolonged hypogonadotropic hypogonadism and constitutional delayed puberty [31, 3538].

Normal response In classic GnRH tests, LH and FSH both rise with a peak occurring between 15 and 60 min. On absolute values, LH increases 3–6 fold, whereas FSH increases about 20–50% above the baseline. In the GnRH priming test, LH has a lower peak than in the classic test. In the GnRH analogue test, LH levels peak 3–4 h after the injection, while FSH levels peak between 3 and 6 h. Most authors suggest that a single evaluation after 4 h (in addition to the baseline assessment) is sufficient to capture the response of both gonadotropins.
Table 2

Causes of delayed puberty in male

Hypogonadotropic hypogonadism

Hypergonadotropic hypogonadism

Constitutional delay of growth and puberty

Klinefelter syndrome

Chronic illness (celiac disease, Crohn’s disease, sickle cell anaemia, cystic fibrosis, diabetes mellitus)

Bilateral anorchia


Gonadal failure following chemotherapy or radiotherapy


Inactivating mutations of gonadotropin-releasing hormone or gonadotropin receptor





    Acquired: radiotherapy; CNS malignancy, infection, trauma


Isolated gonadotropin deficiency:


    Congenital: Kallman syndrome; mutation in Dax-1


    Acquired: intracranial neoplasm (craniopharingyomas, germinomas, gliomas and prolactinomas)




Syndromes with hypogonadotropism: Laurence–Moon Biedl syndrome; Bardet–Biedl subgroup; Prader–Willi, Alstrom, Rud, Bloom syndrome; mutation in leptin


Interpretation GnRH response can be normal, subnormal, delayed or exaggerated. A blunted rise in gonadotropins is generally seen in pituitary diseases or long-standing hypothalamic defects, while an exaggerated response is seen in primary testicular failure (but also in hypothalamic disorders of recent onset, Table 3). The GnRH test does not diagnose gonadotropin deficiency but rather the level of pituitary reserve of LH/FSH secretion.

(A) In adult men, the test may be indicated in the presence of a low serum testosterone with normal gonadotropin values. Men with secondary hypogonadism of hypothalamic or pituitary origin may have a reduced or normal response that in itself is insufficient to discriminate between the two sites possibly affected. Long standing hypothalamic dysfunction can lead to poor or absent pituitary responsiveness, and in these patients, a single injection of GnRH may produce a blunted response, even in the absence of intrinsic pituitary disease. Priming of the pituitary glands with repeated doses of GnRH may improve the test’s sensitivity.
Table 3

Summary of responses to dynamic test in hypogonadism

Cause of hypogonadism

Basal values

GnRH test

HCG test

Clomiphene test

Hypergonadotropic hpogonadism

5α-reductase deficiency


↑ LH

↑ T/DHT ratio

Not indicated

Exaggerated ↑↑↑ LH and FSH response

↑↑↑ (T/DHT) ratio before puberty

Not indicated

Vanishing testes syndrome

∼↑ LH

↓ T

Not indicated

Indicated if basal LH is normal. No response in testicular agenesia. A ↑T raises the possibility of intra-abdominal testes

Not indicated

Mixed hypogonadism


∼↓↑ LH

∼↓↑ FSH

≠ LH pulsatility

Possible ↑ LH and ↑ FSH response; may disclose a testicular deficiency and subtle abnormalities of hypothalamic function

Small ↑ T (a 20–40% lower response on absolute and percentage change compared to young men)

LH may ↑

Hypogonadotropic hypogonadism

Kallman syndrome

↓ LH


↓ T

LH and FSH are released in response to priming followed by classic GnRH-test

T does not rise normala range levels

Lack of LH and FSH response

Prader–Willi Syndrome and Laurence–Moon Biedl

↓ LH


↓ T

LH and FSH are released in response to priming followed by classic GnRH-test

T does not rise normala range levels

Lack of LH and FSH response

Hypothalamic defect

∼↓ LH

∼↓ FSH

↓ T

LH may have an exaggerate response if recent or a reduced response if long standing. In this case, priming may induce a normal response

T does not rise normala range levels in pre-pubertal boys; if hypothalamic dysfunction is recent or in post-pubertal boys T may increase normally

Lack of LH and FSH response

Acquired hypopituitarism and Congenital hypopituitarism

∼↓ LH

∼↓ FSH

↓ T

Reduced. After priming LH has a five-time lower response than normal; peak (FSH/LH) ratio >0.55

Normal response: T raise 2× within range (8–27 nmol/l) if disease presented in adulthood; in prepubertal hypogonadism T fails rise normallya

Lack of LH and FSH response

Chronic illness and malnutrition

∼↓ LH

∼↓ FSH

↓ T

Normal LH response after priming

T increases normallya if disease developed before puberty

Reduced LH and FSH response

Constitutional Delayed Puberty

↓ LH


↓ T

A ↑ LH is further enhanced using GnRH analogs or after priming. A LH peak >8 mUI/ml predicts spontaneous puberty within 1 year

T increases normallya

(T peak >8 nmol/l)

No response in early puberty


∼ LH

∼↓ FSH

∼ T

↓ Inhibin

An exaggerated response (↑ 2× FSH/↑ 5× LH over baseline) is positively related to a better sperm outcome after treatment

T increases normallya

(T peak >8 nmol/l)

↑ 2× of LH

↑ 20–50% of FSH

aNormal is referred to the age-adjusted reference range

In contrast, patients with complete pituitary failure do not respond even after pituitary priming. On a practical level, imaging studies are usually preferred for the differential diagnosis of pituitary or hypothalamic hypogonadism in men with a sudden onset of markedly reduced gonadotropins and testosterone levels. A possible role for GnRH test is to assess the pituitary reserve in view of possible treatment with pulsatile GnRH. However, the test in not indicated in men with elevated gonadotropin levels. Various degrees of primary testicular failure cause a higher than expected peak level of LH and/or FSH after stimulation.

(B) A clear distinction between hypogonadotropic hypogonadism and constitutional delayed puberty is not possible on basal assays of testosterone or analysis of the spontaneous nocturnal gonadotropin profile, as there is a substantial overlap in the results [37, 39, 40]. To ameliorate the diagnostic power of the test, a 36 h priming with GnRH can be used prior to the classic GnRH test.

After GnRH priming, the LH increment is five times lower in hypogonadotropic hypogonadism than in delayed puberty. Considering the ratio of the maximum delta of FSH over LH (Δ FSH max/Δ LH max), a cut-off value of >0.55 is suggestive of a hypogonadotropic hypogonadism, while <0.55 favours the diagnosis of constitutional delayed puberty [29]. GnRH priming plus an evaluation of the ratio of Δ FSH max/Δ LH max is reported to achieve 100% accuracy in the differential diagnosis of HH and CDP [17, 29, 35].

The use of GnRH analogues also offer a clear advantage over the classic test in the differential diagnosis of CDP. Stimulation with buserelin or triptorelin produces a significantly higher LH response in patients with Constitutional Delayed Puberty than in patients with HH, with no overlap between the two groups. Several studies have indicated that all patients with an LH response peak >8 mIU/ml entered spontaneous puberty within 1 year of the diagnostic procedure [32, 38].

(C) Several studies [4145] have correlated GnRH test response with fertility parameters (semen analysis and partner pregnancy rate) in adult men with varicocele. Some, but not all, concluded that the GnRH test response may have a role in predicting the improvement of fertility after varicocelectomy and therefore could be used to select those patients who may benefit from varicocele repair. The test is performed before and 3 months after the procedure. A doubling of FSH and/or a 5-fold increase of LH over the baseline is considered an exaggerated response and suggests a possible initial testicular impairment, which may be hypothetically reversed by treatment [45]. A normalisation of LH response to GnRH after varicocele repair is reported as directly correlated with an improvement in fertility potential within the next few years [41]. As the GnRH test can reveal subtle testicular damage, it could provide further indications for the surgical removal of damaging conditions (varicoceles, compressions, ischemia, infections, etc.).

Clomiphene stimulation test

Indication (A) Confirmation of suspected gonadotropin deficiency. (B) Differential diagnosis of gonadotropin deficiency from weight-related hypogonadism and idiopathic delayed puberty. Clomiphene acts as an anti-oestrogen centrally and as a weak oestrogen peripherally. The central anti-oestrogen effect induces a rise in LH and FSH, interrupting the negative feedback of oestrogen on GnRH release. Contraindications Depression. Side effects Visual phenomena (peripheral flickering or central haloes) have been recorded during the test, disappearing after administration: depression in males.

Procedure Hundred milligram of clomiphene citrate is given for 5–7 days as an evocative test of hypothalamic–pituitary axis. Sampling Serum LH and FSH are assessed at days 0, +4, +7, +10.

Normal response A doubling of LH and a 20–50% increase in FSH are normal result, indicative of an intact hypothalamic–pituitary response [46].

Interpretation Lack of LH and FSH response suggests gonadotropin deficiency due to pituitary or hypothalamic disease. Young children do not normally respond to clomiphene. Similarly, peripubertal children may fail to show a gonadotropin increase during the test. In later stages of pubertal development, gonadotropins begin to rise. Serum testosterone should not be measured during clomiphene stimulation, as it elevates SHBG and T levels, due to its peripheral (hepatic) oestrogenic effect. Sensitivity and specificity are lower than for the GnRH test. For this reason, together with the fact that the procedure is time-consuming (blood sampling for 10 days), the Clomiphene test is a second choice among dynamic HPG axis test.

Human chorionic gonadotropin (hCG) stimulation test

Indication (A) In infants, differential diagnosis of vanishing testis syndrome (agenesia versus cryptorchidism). (B) In peripubertal boys, differential diagnosis of hypogonadism versus delayed puberty. (C) In adults, differential diagnosis of combined testicular and pituitary failure versus secondary hypogonadism. Contraindications None. Precautions None.

Procedure Two different protocols are used for the hCG stimulation test. In post-pubertal boys, a single dose of hCG (5,000 IU intramuscularly) is administered and testosterone values are measured at the baseline and every 24 h up to day five (96 h) [47]. Some authors suggest the inclusion of an early phase measurement at 4 h post-injection. Others report a better response with repeated hCG injections: 1,000 UI/day for 3 days (children <2 years) or 2,000 UI/day on days 0 and 2. In selected cases, measurement of oestradiol, DHT or inhibin B may provide additional information. Sampling Serum testosterone at 0, +48 and +72 h (+24 and +96 h optional data points). Normal response Serum testosterone should rise to outside the normal reference range for the age group, between 48 and 72 h.

Interpretation (A) In infants with neither testis palpable in the scrotum, vanishing testis syndrome is hypothesised. If LH levels are normal or only marginally increased, the hCG test is performed. A flat response in testosterone values is consistent with testicular agenesia. If testosterone increases above or to the top of the normal range for age, this suggests intra-abdominal testes, which must be localised. Children with a normal response to the hCG test should undergo abdominopelvic ultrasonography, MRI or CT, to identify the cryptorchidism site.
  • (B) The serum testosterone increment after hCG stimulation is normally higher (2–3 fold increase in T) in constitutional delayed puberty than in hypogonadotropic hypogonadism patients. Several studies [23, 48, 49] have analysed the response to hCG test, concluding that patients with T values >8 nmol/l should be classified as having constitutional delayed puberty [23], while patients with T < 3 nmol/l are most likely to have hypogonadotropic hypogonadism. Using such criteria the predictive positive value of the test reaches 100% and the predictive negative value 80%. Between the two cut-off values, the results overlap and about 29% of patients remain unclassified. In this case, the diagnosis must be integrated with the GnRH test and diagnostic imaging: testis ultrasound and brain scans [38]. (C) In adult hypogonadism the absence of increased testicular testosterone after hCG suggests a lack of functioning testicular tissue. Conversely, a rise suggests an intact Leydig cell system. In gonadotropin deficiency with no primary testicular abnormality, the basal testosterone value should triple after hCG.


In this review, we reviewed the dynamic testing of hypothalamic–pituitary–gonadal axis to suggest a rational approach to male hypogonadism. It is stressed that these tests should be performed in a cost-efficient and clinically appropriate manner to allow pertinent treatment considerations. So far no consensus has emerged on a single, easy and reliable test with acceptable sensitivity and specificity that works for every case.

Interpretation of dynamic HPG axis tests carries significant limitations due to several issues: the various aetiologies of hypogonadotropic hypogonadism, the absence of clear cut-off criteria, different responses according to the age of the subjects (and whether pre- or post-pubertal), improper selection of patients, differences in the sensitivity of hormone assays, discrepancies in the dose (hCG) and power, half-lives, route of administration (see GnRH tests) of the stimulating agents [23].

The aim of this paper is to try to clarify the indications for dynamic testing in the diagnosis of male hypogonadism. It also gives practical details on how to perform these tests. The most important information for endocrinologists using these tests is that they should be interpreted on the basis of clinical signs and symptoms in order to address a specific hypothesis. A few considerations that may help in interpreting the results are given below. (1) No single test can reliably discriminate hypothalamic from pituitary dysfunction, therefore integration may be needed. (2) For all tests the main indication is represented by a reduced T level without a compensatory increase in gonadotropin levels (with few exceptions, see Table 3). (3) Follow-up of baseline hormones can provide solid data in a non-invasive way (for example, Inibhin B levels during puberty [16]). (4) The GnRH-analogue test is more specific and sensitive than other GnRH-stimulation tests in the differential diagnosis of delayed puberty in early teenage years. (5) The short GnRH analogue test with sampling at 0 and 4 h after injection (rather than the 24 h protocol) is convenient and easy to perform in the outpatient clinic [38]. (6) Dynamic testing for the diagnosis of delayed puberty is useful only when basal T levels are lower than 1.7 nmol/l. (7) In the differential diagnosis of hypogonadotropic hypogonadism and delayed puberty in patients >14 years, the GnRH test alone is not significant. To avoid the need for several days’ hospitalisation, the pituitary priming can be replaced by using the GnRH analogue test. However, the hCG test is also safe and reliable and require no hospitalization. (8) In summary, to discriminate boys with constitutional delayed puberty—in whom spontaneous pubertal onset is expected—from boys affected by gonadotropin deficiency—which could require replacement therapy—the results of hCG and GnRH analogue stimulation test provide the best combination. Patients with a peak >8 nmol/l for testosterone after hCG [38] are most likely affected by constitutional delayed puberty; subjects in whom T rise after hCG is between 4 and 8 nmol/l, the use of GnRH-analogue test could be complementary for the diagnosis (where a LH peak >8 UI/ml reassure for a spontaneous puberty occurring within 1 year). Finally, patients with basal T < 1.7 nmol/l, serum T increment after hCG <8 nmol/l and insufficient rise of gonadotropins after GnRH-analogue stimulation, must have an MRI investigation, because it may reveal a pituitary or supra-pituitary tumour in patients with hypogonadotropic hypogonadism. Most cases of hypogonadotropic hypogonadism with normal MRI correspond with idiopathic hypogonadotropic hypogonadism, for which hormonal investigations are still mandatory [23]. (9) In male aging, often presenting with a mild combined reduction of both LH and testosterone, dynamic testing are usually not necessary for the diagnosis; however, a marked reduction of LH and testosterone levels (<6 nmol/l) cannot be dismissed simply as aging, and rather indicate a potentially more serious aetiology. (10) Our final statement is that all tests are useless, or even misleading, if performed ignoring the particular case of each subject and its clinical context [23].

Copyright information

© Springer Science+Business Media, LLC 2008