Based on the results of the history, physical examination, and laboratory studies, the physician should place the patient into an etiologic category. The clinician is then faced with two potential management options. The male patient may be treated with either specific or nonspecific therapies in an attempt to improve the semen parameters and subsequent fertility potential through intercourse.

The alternate management approach is to overcome the male problem by using his spermatozoa with the ART or bypass the problem by proceeding with either donor insemination or adoption. We believe that, as a general rule, it is preferable to treat the male to improve his fertility status rather than ignore the male factor and use high-cost, advanced technology ARTs, which places the burden of treatment and risk on the female partner for a male problem.

On the other hand, it is often not possible to improve spermatogenesis or attempted treatment may fail. In addition, there may be a concomitant female factor that requires treatment with ARTs. In these instances, it is very appropriate to proceed with ARTs. These two management approaches are not mutually exclusive, and, in some couples, attempts may be made to treat the male at the same time the couple proceeds with ARTs. This more aggressive approach may be used in the older couple when the female’s reproductive potential is diminishing because of age. Finally, both donor insemination and adoption need to be mentioned as alternatives in appropriate cases.


The results of the history, physical examination, and laboratory testing allow the clinician to place the patient into a diagnostic classification. These categories may imply an etiology for the subfertility, although a large percentage of the patients fall into the idiopathic category.

Endocrine Causes

Endocrine causes of male infertility are often referred to as pretesticular causes. Impairment of fertility in these cases is secondary to either a hormone deficiency or an excess.

Pituitary Disease

Pituitary function may be affected in cases of pituitary surgery, infarction, tumors, radiation, or infectious diseases. Patients with prepubertal onset of pituitary disease are usually diagnosed before a fertility evaluation as a result of growth retardation, delayed puberty, or adrenal and thyroid deficiency. Infertility, erectile dysfunction, visual field disturbances, and severe headaches may be presenting symptoms in the adult male with pituitary tumors.

Normal male secondary sexual characteristics are usually present in those patients with postpubertally acquired pituitary disease. Patients with congenital pituitary disease will undergo adrenarche and have small amounts of straight pubic hair, unless concomitant adrenal insufficiency exists. Small, soft testes may be demonstrated on physical examination. This is in contrast to cases of primary testicular failure with tubular and peritubular sclerosis, in which case the testes are small but firm to palpation.

Plasma testosterone levels are typically low or low normal and gonadotropin levels are low in most patients with pituitary disease. However, the normal range for gonadotropins, particularly LH, goes quite low. Thus, a normal LH value associated with a very low serum testosterone value should be considered suspicious and further evaluation of the pituitary gonadal axis is required. However, a borderline serum testosterone level associated with a normal or low LH is probably normal because testosterone is secreted in a pulsatile fashion and the low value may represent a nadir. Evaluation of other pituitary hormones and endocrine functions, adrenal and thyroid, should be performed only if there is clinical evidence of a specific endocrinopathy.

Isolated Hypogonadotropic Hypogonadism

Gonadotropin deficiency may occur in the presence of otherwise normal pituitary function. This condition may be due to Kallmann’s syndrome (congenital hypogonadotropic hypogonadism associated with anosmia) or idiopathic hypogonadotropic hypogonadism. Kallmann’s syndrome is a genetically heterogeneous disorder that may be inherited in an X-linked, autosomal dominant or autosomal recessive pattern (Duke et al, 1995; Maya-Nunez et al, 1998).

The most prevalent is an X-linked form that maps to theKAL1 gene, which encodes for a neuron adhesion molecule thought to be responsible for guiding migration of LH releasing hormone secreting neurons to the medial basal hypothalamus (Dacou-Voutetakis, 1992; Lutz et al, 1993). Complete or partial anosmia is a consistent finding in these patients. Cryptorchidism and gynecomastia are common, and micropenis occurs in approximately 50% of affected males.

The primary hormonal defect is a failure of GnRH secretion by the hypothalamus, leading to secondary testicular failure (Hoffman and Crowley, 1982).Multiple other congenital anomalies such as craniofacial asymmetry, cleft palate, harelip, color blindness, congenital deafness, and renal anomalies may be associated with this syndrome (Danish et al, 1980). A delay in pubertal development is the hallmark of the syndrome and most commonly causes the patient to present for medical evaluation.

The diagnosis may occasionally be made in childhood because of the presence of cryptorchidism or micropenis. As a result of a delay in the androgen-dependent closure of the epiphyseal plates, the length of the arms and legs may be greater than that of the trunk. In addition, the testes are prepubertal, usually being smaller than 2 cm in diameter. In the prepubertal male, differentiating between Kallmann’s syndrome and delayed sexual maturation can be difficult (Whitcomb and Crowley, 1993).

The presence of a family history of Kallmann’s syndrome or the presence of somatic midline defects and/or anosmia may help in the prepubertal diagnosis.

The first sign of puberty is testicular growth. Thus, if the testes are enlarged, the patient is experiencing delayed puberty rather than hypogonadotropic hypogonadism. Normal males experiencing delayed puberty will demonstrate LH pulses if frequent blood samples are obtained.

These pulses are not present in patients with Kallmann’s syndrome (Boyar et al, 1972). GnRH stimulation testing of these patients results in an absent or blunted rise in gonadotropins. However, repeated GnRH injections prime the pituitary gland, resulting in rises in levels of both gonadotropins.

This pattern of response may also be found in prepubertal boys (Snyder et al, 1979). Finally, after doses of 5000 IU of hCG, prepubertal and pubertal boys demonstrate larger rises in testosterone levels than patients with Kallmann’s syndrome. Androgen replacement with testosterone is adequate treatment for the virilization of teenagers with Kallmann’s syndrome.

However, treatment with exogenous androgens results in a suppression of intratesticular testosterone production. Thus, spermatogenesis and testicular growth are not stimulated by this treatment. Prior treatment with androgens does not impair subsequent testicular response to gonadotropin therapy (Ley and Leonard, 1985).

Androgen therapy has been most commonly given in parenteral form as testosterone enanthate or cypionate. IM injections of 200 mg every other week are usually sufficient to induce full virilization in most patients. Although oral androgens are available as fluoxymesterone and 17-methyltestosterone, they are less potent, owing to erratic absorption and hepatic metabolism, and result in a higher incidence of hepatic abnormalities.

Reversible intrahepatic cholestasis resulting in elevations of plasma transaminase, lactate dehydrogenase, and bilirubin may be noted. Transdermal testosterone skin patch treatment is an effective alternative but at a much higher cost. In addition, there has been a significant risk of dermatitis with some of the transdermal patches. This risk is significantly less with the topical gel.

The advantage of the transdermal testosterone preparations is that they maintain more constant serum testosterone levels as compared with the cyclical pattern observed with every-other-week depot injections. In addition, a lower rate of polycythemia is observed in patients receiving the transdermal preparations.

Gonadotropin therapy is required for the initiation of spermatogenesis. Given as 2000 IU subcutaneously three times per week, hCG initiates spermatogenesis in most patients with acquired hypogonadotropic hypogonadism. However, the completion of spermatogenesis in patients with congenital forms of hypogonadotropic hypogonadism usually requires the addition of FSH. FSH may be given in the form of human menopausal gonadotropin (hMG), which contains 75 IU of FSH and 75 IU of LH per vial.

The alternative to hMG is recombinant human FSH, which has pure FSH activity (Kliesch et al, 1995; Liu et al, 1999). The IM administration of FSH at a dose of 37.5 IU (½ vial) to 75 IU (1 vial) three times per week is most commonly started after 3 to 6 months of hCG therapy and usually results in the completion of spermatogenesis (Finkel et al, 1985). Stimulation of the testes with FSH and LH results in testicular growth, although the final testis volume often remains below normal.

Whereas the semen motility and morphology parameters are usually quite good, oligospermia with sperm counts below 10 million/ml are common. However, in contrast to patients with idiopathic oligospermia who are often infertile with sperm densities below 20 million/ml, many patients with treated hypogonadotropic hypogonadism are able to conceive despite very low sperm densities (Burris et al, 1988a).

GnRH supplied by means of intermittent subcutaneous injections or a pulsatile infusion pump with 90-minute pulses is an alternative to treatment with gonadotropin therapy in patients with hypogonadotropic hypogonadism who have intact pituitary glands.

GnRH therapy is effective only in patients with an intact native pituitary gland because it is ultimately dependent on pituitary secretion of gonadotropins. Therefore, GnRH therapy is contraindicated in patients with acquired hypogonadotropic hypogonadism caused by pituitary tumors, pituitary surgery, or head trauma. Studies have shown that pulsatile infusion is superior to intermittent injections of GnRH (Shargil, 1987). However, direct comparison of continuous-infusion GnRH therapy to gonadotropin therapy has not shown a significant difference to justify the increased expense and inconvenience of GnRH therapy (Liu et al, 1988).

Previous therapy with testosterone does not affect subsequent response to therapy (Ley et al, 1985), and the best predictor is baseline testicular volume (Burris et al, 1988b). We usually use hCG followed by recombinant human FSH as the initial treatment and reserve infusion pump therapy for those patients who do not respond adequately.

Fertile Eunuch Syndrome

Isolated LH deficiency occurs rarely in patients with normal FSH levels. These men demonstrate a variably eunuchoid habitus, large testes, and small-volume ejaculates that may contain a few spermatozoa (Faiman et al, 1968). Plasma testosterone and LH levels are low, but FSH levels are in the normal range. Testicular biopsy demonstrates maturation of the germinal epithelium.

However, Leydig cells may not be apparent because of insufficient LH stimulation. A rise in serum testosterone after hCG therapy documents normal Leydig function in these patients. Sufficient intratesticular testosterone appears to be produced to support a minimal degree of spermatogenesis. However, inadequate peripheral androgen levels result in a lack of virilization.

Isolated Follicle-Stimulating Hormone Deficiency

Patients with this rare disorder demonstrate normal virilization, normal LH and testosterone levels, and normal-sized testes. Because of a lack of FSH, oligospermia or azoospermia is present. Administration of hMG has been shown to effectively stimulate spermatogenesis in these patients (Al Ansari et al, 1984). Recombinant human FSH would be the preferred treatment today.

Other Congenital Syndromes

The Prader-Willi syndrome consists of obesity, hypotonic musculature, mental retardation, small hands and feet, short stature, and hypogonadism. There is a familial tendency. The locus for Prader-Willi syndrome has been mapped to chromosome 15q11-q13. The cause of the syndrome is a deletion or uniparental disomy. The DNA probe PW71 can be used to establish the diagnosis (Lerer et al, 1994).

Patients with Prader-Willi syndrome have LH and FSH deficiencies from a lack of GnRH secretion. Treatment is identical to that for Kallmann’s syndrome. As a result of the multiple anomalies in these patients, however, infertility is often not a clinical problem (Bray et al, 1983). A similar clinical picture is found in patients with Laurence-Moon-Bardet-Biedl syndrome, which consists of hypogonadotropic hypogonadism, retinitis pigmentosa, polydactyly, and hypomnesia.

Androgen Excess

Gonadotropin production is inhibited by negative feedback by both estrogens and androgens at the level of the hypothalamus and pituitary. A hypogonadal state, therefore, may be induced by androgen excess whether it is due to exogenous sources, such as anabolic steroids used by athletes, or to endogenous production, such as a metabolic abnormality or an androgen-producing tumor.

The intratesticular testosterone concentration is 50- to 100-fold higher than serum levels from local production (Turner et al, 1984). Thus, the seminiferous tubules are normally exposed to extremely high levels of testosterone when androgen synthesis takes place in the Leydig cells.

Introduction of sex steroids into the circulation from a source besides the testis has an inhibitory effect on spermatogenesis because of the resultant reduction of both intratesticular testosterone and FSH through feedback inhibition on the pituitary gland. Thus, the administration of exogenous testosterone is a male contraceptive and anabolic steroids have a similar effect.

CAH is the most common cause of endogenous androgen excess. A congenital deficiency of 21-hydroxylase is the most common of the five enzyme defects responsible for this syndrome. The diagnosis can be established in the first trimester of pregnancy by DNA analysis of biopsy samples of the chorionic villi (New, 1994). A deficiency of 21-hydroxylase results in a decrease in cortisone synthesis, which leads to an increase in pituitary production of ACTH.

Elevated levels of ACTH result in hyperstimulation of the adrenal gland and an increased production of adrenal androgens. The resultant serum excess of adrenal androgens has a negative feedback on pituitary gonadotropin secretion.

These patients often have short stature and may develop precocious puberty. As a result of androgen stimulation, premature enlargement of the penis may occur; however, because of a lack of gonadotropin stimulation, the testes remain small. Basal plasma 17-hydroxyprogesterone levels are often elevated 50 to 200 times above normal levels. In addition, elevated urinary 17-ketosteroid and pregnanetriol levels may occur.

Not all patients with this syndrome demonstrate fertility abnormalities. Urban and coworkers (1978) studied 20 patients with CAH. Almost all patients demonstrated normal serum gonadotropin and testosterone levels. Two patients were untreated, and 3 had been poorly treated; however, 4 of these patients had children.

Of the 15 treated patients, 8 had conceived. In some patients, the adrenal androgen production may not be sufficient to interfere with the normal hypothalamic-pituitary-gonadal axis, explaining the fertility of these patients.

Glucocorticoid therapy results in a reduction of ACTH levels, which induces a decrease in peripheral adrenal androgens, thus stimulating endogenous gonadotropin secretion and testicular steroidogenesis. This approach has been successfully employed in the treatment of men with infertility secondary to CAH (Augarten et al, 1991). Bilateral testicular adrenal rest tumors may develop in some patients with CAH.

These tumors often resolve with hormonal therapy (Cutfield et al, 1983). However, patients with 21-hydroxylase deficiency, complicated by adrenal rest tumors, may be permanently infertile because of testicular fibrosis. Men with a partial 21-hydroxylase deficiency may remain undiagnosed into adulthood because a mild elevation of adrenal androgens may not be evident in males and there is sufficient glucocorticoid production.

There have been case reports of infertile men with partial 21-hydroxylase deficiencies; however, most of these men would be expected to be fertile.

Excess androgen production may also be due to adrenal or testicular tumors. This results in the failure of testicular development when present in prepubertal patients. Tubular and peritubular sclerosis may occur in the postpubertal patient, and this may be irreversible. Leydig cell tumors are not always evident on physical examination. Therefore, testicular ultrasonography should be obtained as part of the evaluation of patients with androgen excess.

Estrogen Excess

Peripheral estrogens normally suppress pituitary gonadotropin secretion. A state of secondary testicular failure may be induced by estrogen-secreting tumors in the adrenal cortex or in the testis. Testicular Sertoli cell tumors or interstitial cell (Leydig cell) tumors may produce estrogen.

Excess peripheral estrogens may also result from hepatic dysfunction or obesity. Peripheral adipose tissue contains aromatase, which is an enzyme that converts androgen into estrogen. Elevated estrogen levels have been identified in morbidly obese patients; however, not all investigators have confirmed this finding (Hargreave et al, 1988; Jarow et al, 1993).

Erectile dysfunction, gynecomastia, and testicular atrophy may be present in patients with estrogen excess. Hormonal studies demonstrate low levels of FSH, LH, and testosterone in the presence of elevated estrogens. Urinary 17-ketosteroid levels may also be elevated. Treatment is directed at the underlying condition.

Prolactin Excess

Hyperprolactinemia may be caused by a pituitary tumor, stress, medications, medical illness, or idiopathic causes. Both erectile dysfunction and male infertility are associated with hyperprolactinemia.

Routine screening of infertile men for hyperprolactinemia has not been shown to be useful (Eggert-Kruse et al, 1991; Sigman and Jarow, 1997). In patients with prolactin-secreting pituitary adenomas, gonadotropin and testosterone levels are depressed whereas prolactin levels are markedly elevated.

A small percentage of men with prolactin-secreting pituitary adenomas have borderline normal serum testosterone levels (Carter et al, 1978; Spark et al, 1982). Men with prolactin-secreting pituitary adenomas are typically diagnosed later in the course of disease than women and have larger tumors with higher serum prolactin levels.

Patients with an elevated prolactin level should undergo imaging of the pituitary gland, preferably an MRI with gadolinium contrast. In addition, because thyrotropin-releasing hormone stimulates prolactin secretion, hypothyroidism should be ruled out.Although surgery and radiation therapy were used in the past to treat patients with prolactin-secreting pituitary tumors, the vast majority of patients respond to medical therapy.

The two agents most commonly used today are bromocriptine and cabergoline. Cabergoline has the advantages of fewer side effects and less-frequent dosing. Patients with idiopathic hyperprolactinemia are treated with medication as well, which may be withdrawn yearly to determine whether hyperprolactinemia persists (Dollar and Blackwell, 1986; Wang et al, 1987).

We do not recommend treatment of patients with isolated mild hyperprolactinemia because, in our experience, this has not resulted in improved spermatogenesis.

Thyroid Abnormalities

Although hyperthyroidism has been associated with male infertility, patients with thyroid disorders rarely present with infertility as their chief complaint (Clyde et al, 1976; Kidd et al, 1979). Testicular and pituitary abnormalities as well as elevated levels of circulatory estradiol have been identified in patients with hyperthyroidism. In addition, maturation arrest patterns have been identified on testicular biopsy specimens.

Glucocorticoid Excess

Glucocorticoid excess may suppress LH secretion, resulting in androgen deficiency and testicular dysfunction. Glucocorticoid excess may be secondary to endogenous production as in Cushing’s syndrome or secondary to exogenous intake from medical therapy.

Hypospermatogenesis or maturation arrest patterns have been found on testicular biopsy specimens of patients with Cushing’s syndrome (Gabrilove et al, 1974). However, this is rarely a problem in patients receiving therapeutic dosages of steroids. Therapy is directed at correction of the underlying glucocorticoid abnormality.

Abnormalities of Androgen Action

Androgen abnormalities may involve a deficiency in androgen synthesis, a deficiency in conversion of testosterone to dihydrotestosterone (5α-reductase deficiency), or androgen receptor abnormalities. Both defects in androgen synthesis and 5α-reductase deficiency commonly result in ambiguous genitalia and are therefore discussed in Chapter 68, Sexual Differentiation: Normal and Abnormal.

Defects in androgen receptors result in androgen-resistant syndromes. Testosterone is the major circulating androgen in man. The majority of circulating testosterone is bound to hepatic-derived SHBG and albumin. Once testosterone reaches a target cell, it diffuses into the cytoplasm, where it may be metabolized into a more active androgen (dihydrotestosterone), aromatized into estradiol, or converted into a weaker androgen.

The sex steroids testosterone, dihydrotestosterone, or estrogen then bind with their respective receptors, either the androgen receptor or the estrogen receptor. The steroid-receptor complex then interacts with the DNA and nuclear matrix to affect DNA transcription into messenger RNA.Abnormalities of the androgen receptor result in resistance to androgens proportional to the severity of the defect despite the presence of elevated testosterone levels.

These patients are 46,XY males with phenotypes ranging from pseudohermaphroditism to a normal male phenotype with infertility. Androgen-resistant syndromes have been identified in phenotypically normal men with azoospermia and severe oligospermia. Patients with partial androgen insensitivity have elevated serum testosterone and LH levels. FSH levels are typically normal or elevated.Select series have documented partial androgen insensitivity in phenotypically normal infertile men with the characteristic hormonal pattern just described (Aiman et al, 1979; Schulster et al, 1983), but this condition is not found very often in an unselected infertile patient population (Griffin and Wilson, 1980).


Moreover, more recent molecular studies of the androgen receptor gene have not found a high incidence of mutations in infertile men (Puscheck et al, 1994). Because the androgen receptor gene is located on the X chromosome at Xq11-12, this syndrome is inherited as an X-linked recessive trait. Several investigators have observed expansion of the polymorphic CAG repeat located on exon 1 in infertile men, which purportedly decreases the activity of the androgen receptor (Dowsing et al, 1999; Yoshida et al, 1999).

However, other investigators could not reproduce these findings (Dadze et al, 2000). Androgen receptor abnormalities resulting in partial androgen insensitivity should be suspected in patients with elevated serum levels of testosterone and LH but are extremely rare.

Disorders of Spermatogenesis

Chromosomal Disorders

Klinefelter’s Syndrome

The presence of an extra X chromosome is the genetic hallmark of Klinefelter’s syndrome. Nondisjunction of the meiotic chromosomes of the gametes of either parent leads to pure Klinefelter’s syndrome, whereas nondisjunction during mitotic cell division of the developing embryo leads to mosaicism.

This syndrome is identified in one of every 600 male births (Paulsen et al, 1968; Nielsen and Wohlert, 1991). A phenotypic male with small firm testes, gynecomastia, and elevated gonadotropins characterizes the classic form of Klinefelter’s syndrome (Klinefelter et al, 1942).

More recent studies have confirmed the presence of testicular atrophy in the vast majority of patients; however, gynecomastia and a female escutcheon pattern may be observed in less than half of patients (Okada et al, 1999). In addition, up to 50% of the patients may have normal testosterone concentrations, although gonadotropins are usually elevated in these cases (Okada et al, 1999).

Although secondary sexual characteristics begin developing at the appropriate time, the completion of puberty is often delayed, at which point eunuchism, gynecomastia, or sexual or erectile dysfunction may be noted. Virilization may be complete in some patients, and the diagnosis is not uncommonly delayed until adulthood, at which point the patient may present with infertility with associated gynecomastia and small, firm testes.

Mental retardation and various psychiatric disturbances have been identified in some of these patients (Becker et al, 1966; Theilgaard, 1984).Azoospermia is typically present. Seminiferous tubular sclerosis is commonly identified on testicular biopsy. However, occasional Sertoli cells or spermatozoa may be found. Plasma FSH levels are usually markedly elevated as a result of the severe seminiferous tubular injury, whereas LH levels are elevated or normal. Total plasma testosterone levels are decreased in 50% to 60% of patients (Paulsen et al, 1968; Okada et al, 1999).

The physiologically active free testosterone concentrations are usually decreased. In addition, plasma estradiol levels are usually increased, stimulating increased levels of testosterone-binding globulin and resulting in a decreased testosterone-to-estrogen ratio, which is believed to be responsible for gynecomastia (Chopra et al, 1973; Wang et al, 1975)

.The diagnosis may be made with a chromatin-positive buccal smear, indicating the presence of an extra X chromosome; however, karyotypes are usually performed, demonstrating 47,XXY or, less commonly, a mosaic pattern of 46,XY/47,XXY. Less severe abnormalities are present in patients with the mosaic form of Klinefelter’s syndrome, and occasional patients are fertile (Foss and Lewis, 1971; Laron et al, 1982). In addition, more than one extra X chromosome may uncommonly occur in this syndrome.

There have been many reports of various cancers developing in patients with Klinefelter’s syndrome. A 50-fold higher risk of breast cancer development has been reported in these patients (Hultborn et al, 1997). However, not all studies have noted an overall increase in cancer development (Hasle et al, 1995).There is no therapy to improve spermatogenesis in Klinefelter’s syndrome. For the patient with mosaic Klinefelter’s syndrome and severe oligospermia, ICSI combined with IVF is a possibility (Harari et al, 1995).

More recently, testicular sperm extraction has been performed in patients with azoospermia and nonmosaic Klinefelter’s syndrome. Retrieved sperm have been successfully used for IVF with ICSI, resulting in the birth of normal children (Palermo et al, 1998; Reubinoff et al, 1998; Ron-El et al, 1999).

There remains concern that the use of spermatozoa from these patients may result in transmission of this karyotypic abnormality to their offspring. Studies examining the chromosomal complement of the sperm cells from patients with mosaic and nonmosaic Klinefelter’s syndrome have revealed an increased prevalence of abnormal chromosomal complements in sperm from these patients.

Interestingly, despite this, the majority of sperm have been found to have normal chromosomal complements, which may account for why the few live births that have been reported have had normal karyotypes (Cozzi et al, 1994; Estop et al, 1998; Foresta et al, 1998). These data strongly suggest that 47,XXY cells are able to undergo meiosis and form spermatozoa. This makes it imperative that genetic counseling is offered to all such patients before proceeding with ARTs.

XX Male

Findings similar to those of Klinefelter’s syndrome are found in patients with the XX male syndrome (sex reversal syndrome). These patients demonstrate small, firm testes; frequent gynecomastia; small to normal-sized penises; and azoospermia. Testicular biopsy may demonstrate seminiferous tubule sclerosis, resulting in elevated gonadotropin levels and decreased testosterone levels (Perez-Palacios et al, 1981).

Unlike typical patients with Klinefelter’s syndrome, these individuals have shorter than normal average heights, show no higher prevalence of mental deficiency, and have an increased prevalence of hypospadias (de la Chapelle, 1981).
Karyotypes reveal 46,XX chromosomal complements. Although it has been presumed that portions of the Y chromosome are present, molecular studies have not always demonstrated this. Some patients have been found to have the testis-determining gene (SRY).

This has not been demonstrated in all patients (Lopez et al, 1995). Recently, the duplication of another Y chromosome gene to an autosome has been identified in anSRY-negative XX male (Huang et al, 1999). This demonstrates that XX males are genetically heterogeneous.

Because these patients do not have any of the AZF regions present, it is unlikely that sperm would be recovered from testicular tissue. As of yet, there have been no reports of sperm being isolated from these patients.

XYY Syndrome

Patients with the XYY syndrome are characteristically tall, and semen analyses typically reveal severe oligospermia or azoospermia. Although this karyotype has been linked to aggressive and criminal behavior (Jacobs et al, 1965; Walzer and Gerald, 1975; Freyne et al, 1992), the cause for this finding is controversial.

Some have believed that these behaviors are secondary to tall stature, which may predispose individuals to these behaviors (Hook, 1973), whereas others have suggested it may be related to lower intelligence (Gotz et al, 1999). The XYY karyotype occurs in 0.1% to 0.4% of newborns (Balodimos et al, 1966; Price et al, 1966; Walzer and Gerald, 1975).

As has been found in Klinefelter’s syndrome, an increased prevalence of chromosomal abnormalities involving the sex chromosomes have been found in the sperm from these patients (Han et al, 1994; Blanco et al, 1997; Lim et al, 1999). Testicular biopsy specimens reveal patterns ranging from maturation arrest to complete germinal aplasia, as well as occasional cases demonstrating seminiferous tubule sclerosis (Santen et al, 1970; Skakkebaek et al, 1973a; Baghdassarian et al, 1975).

Occasional patients have been fertile (Stenchever and Macintire, 1969). Plasma gonadotropins and testosterone levels are most often within the normal range in these patients (Lundberg and Wahlstrom, 1970). Elevations of plasma FSH levels may be found in association with more severe patterns of testicular dysfunction. Although there is no treatment to improve spermatogenesis, these patients are candidates for ARTs. Genetic counseling should be offered before beginning this form of therapy.

Noonan’s Syndrome

The phenotypic appearance of patients with Noonan’s syndrome is similar to that found in Turner’s syndrome (XO). Thus, these patients have short stature and demonstrate hypertelorism, webbed neck, low-set ears, cubitus valgus, ptosis, and cardiovascular abnormalities (Collins and Turner, 1973).

Chromosomal analysis reveals 46,XY karyotype. Cryptorchidism and testicular atrophy are commonly present with associated elevations of gonadotropins. Of interest, females may be affected with this disorder, carrying a karyotype of XX, and are usually fertile.

Although most cases of Noonan’s syndrome are sporadic, familial transmission has been reported. In addition, a gene on chromosome 12 has been linked to this defect (Ogata et al, 1998). Although androgens may be given to complete virilization, there is no treatment for the fertility abnormality in these patients.

Y Chromosome Microdeletions

The majority of Y chromosome microdeletions that have been associated with azoospermia or severe oligospermia occur in one of three nonoverlapping regions of the long arm of the Y chromosome designated as AZFa (proximal), AZFb (middle), and AZFc (distal) (Vogt et al, 1996).

The vast majority of these deletions occur de novo and are not inherited from the parents. Rare vertical transmission from father to son has been reported (Stuppia et al, 1996; Chang et al, 1999). Although most studies have examined patients with idiopathic azoospermia or severe oligospermia, a 7% prevalence of Y chromosome microdeletions has been reported in patients with nonidiopathic severe male factor infertility (Krausz et al, 1999).

Patients with these microdeletions are phenotypically normal, with the only apparent abnormality being a defect in spermatogenesis. There is no strict correlation between the deletion and the histologic phenotype on testis biopsy. Patients with what appear to be similar deleted intervals may have different testicular histologies. This may be due to the many genes involved in these intervals and the fact that many of the genes are present in multiple copies.

Deletions in AZFc are the most frequently identified microdeletions in azoospermic and severely oligospermic men. The deleted azoospermia gene (DAZ) is one of the genes thought to be responsible for spermatogenic defects in patients with deletions in this interval.DAZ is expressed exclusively in the testes and seems to produce an RNA-binding protein (Menke et al, 1997; Habermann et al, 1998).

Sperm have been reported to be retrieved in over 50% of men with azoospermia and Y chromosome microdeletions limited to AZFc undergoing testicular sperm extraction (Brandell et al, 1998; Silber et al, 1998).

A gene called RBMY (RNA-binding motif, Y chromosome; also called RBM for RNA-binding motif) is thought to be the candidate spermatogenic gene in the AZFb region. There are multiple copies of this gene, which is germ cell specific (Ma et al, 1993; Elliott et al, 1997).

This gene produces an RNA-binding protein localized to germ cell nuclei. Some have reported a decreased likelihood of finding sperm on testicular sperm retrieval in patients with AZFb deletions (Brandell et al, 1998), but all authors have not reported this (Ferlin et al, 1999).

It appears that larger deletions involving more than one AZF region are more likely to yield Sertoli cell–only histology and a lower likelihood of sperm retrieval (Silber et al, 1998; Ferlin et al, 1999). Deletions in AZFa are less common than the other deletions.

At least three genes have been identified in this region:USP9Y (also known as DFFRY), DBY, and UTY. Recent evidence suggests that most deletions in this region that affect spermatogenesis involve DBY, with a lesser number involving USP9Y. Deletions limited to UTY may not affect spermatogenesis (Foresta et al, 2000).

Recently, a fourth region of the Y chromosome, termedAZFd, has been suggested to affect morphology. This region, occurring between AZFb and AZFc, has not yet been well investigated (Kent-First et al, 1999). There is currently no treatment to improve spermatogenesis in patients with Y chromosome microdeletions; however, these patients are candidates for IVF with ICSI.

Sperm from semen may be used in oligospermic patients, whereas attempts at testicular sperm extraction may be employed in azoospermic patients. It is important to realize that these deletions will be transmitted to male offspring (Jiang et al, 1999; Page et al, 1999). Before embarking on a course of ART, these patients should be offered genetic counseling.

Other Chromosomal Abnormalities

Various other chromosomal abnormalities have been identified in infertile patients. Translocations involving the somatic chromosomes have been associated with some cases of oligospermia (Plymate et al, 1976; Sarto and Therman, 1976; Chandley, 1979).

In addition, testicular biopsy specimens of oligospermic men have demonstrated meiotic abnormalities of the germ cells but with normal peripheral karyotypes (Hulten et al, 1970; Pearson et al, 1970; Skakkebaek et al, 1973b; Koulischer et al, 1974). Karyotypic abnormalities have been identified in the male partners of women who have undergone recurrent abortions (Blumberg et al, 1982; Fortuny et al, 1988).

Therefore, karyotype analysis should be offered to male partners of women with recurrent miscarriages. Several syndromes with a genetic component are associated with male infertility. Prune-belly syndrome is associated with absence of the abdominal wall musculature, cryptorchidism, and urogenital tract abnormalities, and an autosomal dominant inheritance pattern is suggested (Riccardi and Grum, 1977). The Prader-Willi and Laurence-Moon-Biedl syndromes also have a genetic basis and are associated with infertility.

Bilateral Anorchia

Also known as vanishing testis syndrome, bilateral anorchia is found in genetic XY males with nonpalpable testes. Patients demonstrate prepubertal male phenotypes, indicating that testicular tissue, secreting both androgens and müllerian-inhibiting substance, must have been present in utero.

It is thought that the testes may have been lost in utero secondary to infection, vascular injury, or testicular torsion. Molecular analysis of DNA from these patients has shown no abnormalities in the testis-determining regions (SRY gene) (Lobaccaro et al, 1993).

Low plasma testosterone and elevated gonadotropin levels are present in these males (Aynsley-Green et al, 1976). Virilization should be induced with testosterone administration at puberty, and these patients require testosterone supplementation for life. However, in the absence of any testicular tissue, their infertility is not treatable.


Cryptorchidism is present in 3% to 4% of full-term boys (Scorer and Farrington, 1971; John Radcliffe Hospital Cryptorchidism Study Group, 1992). By 1 year of age, 1% to 1.6% of boys demonstrate undescended testes (Scorer and Farrington, 1971; John Radcliffe Hospital Cryptorchidism Study Group, 1992).

After 6 months of age, the undescended testis is unlikely to descend on its own. Two thirds of cases are unilateral, and one third of cases are bilateral. Sperm concentrations below 12 to 20 million/ml are found in 50% of patients with bilateral cryptorchidism and in approximately 25% of patients with unilateral cryptorchidism (Lipshultz, 1976; Lipshultz et al, 1976; Okuyama et al, 1989).

Testicular biopsy of the cryptorchid testis reveals decreased numbers of Leydig cells. Within the first 6 months of life, the number of germ cells in the cryptorchid testis is within the normal range; however, the normal increase in germ cell numbers seen in early infancy does not occur. By 2 years of age, 38% of unilateral and bilaterally cryptorchid testes will have lost germ cells. The descended testis, in cases of unilateral cryptorchidism, may also demonstrate abnormalities with lower numbers of germ cells.

In general, there is a direct relationship between testicular position and fertility potential: The higher the cryptorchid testis, the more severe the testicular dysfunction. Absence of germ cells is found in 20% to 40% of inguinal testes in contrast to 90% of intra-abdominal testes (Hadziselimovic, 1983).

Both mechanical and hormonal etiologic factors have been suggested to explain the mechanism of cryptorchidism. Increasing evidence points to a defect in the hypothalamic-pituitary-gonadal axis in patients with cryptorchidism (Canlorbe et al, 1974), but underlying testicular abnormalities in both the cryptorchid and the normally descended contralateral testis are frequently seen.

It is important to differentiate the cryptorchid testis from both retractile testis, which is caused by a hyperactive cremasteric reflex, and from an ectopic testis. The latter two cases are not typically associated with testicular dysfunction. The finding of histologic changes in cryptorchid testis within the first year of life has led to therapy directed at correction of cryptorchidism by 1 to 2 years of age (419 et al, 1984; Huff et al, 1987).

Retrospective studies report fertility rates from 78% to 92% in patients with surgically corrected unilateral cryptorchidism (Puri and O’Donnell, 1988; Cendron et al, 1989; Kumar et al, 1989). The fertility rate for men with corrected bilateral cryptorchidism is significantly lower, 30% to 50% (Chilvers et al, 1986).

Testicular Torsion

Testicular torsion occurs most commonly during adolescence, with an incidence that has been estimated to be 1 in 4000 males younger than age 25 years (Barada et al, 1989). It is thought that testicular torsion is caused by an anatomic abnormality of a narrow mesenteric attachment from the spermatic cord onto the testis.

However, there is thought to be an underlying abnormality of these testes as well. Biopsies of the uninvolved contralateral testis in boys with unilateral testicular torsion suggest the presence of an underlying testicular abnormality that may affect fertility (Hadziselimovic et al, 1986, 1998; Dominguez et al, 1994).

In fact, most follow-up studies of patients who had torsion reveal reduced testicular exocrine function even in the uninvolved testis (Thomas et al, 1984; Anderson et al, 1992). Several investigators had reported abnormalities in the normal contralateral testis of animals undergoing experimental unilateral testicular torsion, suggesting the presence of a humoral factor (Nagler and White, 1982; Heindel et al, 1990).

Yet other investigators fail to identify any abnormality in the contralateral testis even after prolonged torsion (Turner, 1987). In fact, the fertility of adults with prepubertal testicular torsion does not appear to be reduced (Puri et al, 1985).

The incidence of testicular torsion is so low and the time so remote from attempted paternity that study of this problem has been close to impossible. Therefore, firm conclusions regarding the role of unilateral testicular torsion are not available; however, bilateral testicular torsion is a definite cause of testicular failure.


A varicocele is an abnormal tortuosity and dilatation of the testicular veins within the spermatic cord. A clinical varicocele exists when these dilated veins are palpable on physical examination, and a subclinical varicocele is present when these dilated veins are detectable only by ancillary techniques.

Varicoceles are rarely detected before the age of 10 years, with the prevalence increasing to approximately 15% by early adulthood. In contrast, the prevalence of varicoceles in men presenting with infertility is 20% to 40% (Dubin and Amelar, 1977; Cockett et al, 1979; Aafjes and van der Vijver, 1985; Marks et al, 1986).

The varicocele is the most common correctable cause of male infertility. Approximately 90% of varicoceles are left sided. Most studies report an approximately 10% prevalence of bilateral varicoceles, although some recent reports have reported a higher prevalence of bilaterality.

Some studies have reported a significantly higher prevalence of varicoceles in patients with secondary infertility as compared with primary infertility (Gorelick and Goldstein, 1993; Witt and Lipshultz, 1993; Jarow et al, 1996). Differences in the venous drainage patterns of the right and left testicular veins may account for this left-sided predominance.

The left testicular vein normally drains directly into the left renal vein, and the right testicular vein drains into the inferior vena cava. In addition, an absence of the venous valves is more commonly found on the left side than on the right (Ahlberg et al, 1966). Finally, the left renal vein may be compressed between the superior mesenteric artery and the aorta.

This “nutcracker phenomenon” may result in increased pressure in the left testicular venous system (Coolsaet, 1980). Unilateral right-sided varicoceles are uncommon and raise the possibility of thrombosis or occlusion of the vena cava (as with a right-sided renal tumor with vena caval thrombosis) or situs inversus (Grillo-Lopez, 1971).

The mechanism by which varicoceles affect testicular function remains unclear (Howards, 1995). Intratesticular temperatures decrease by 0.5°C in men without varicoceles when they rise from a supine to a standing position, as opposed to an increase of 0.78°C in men with varicoceles (Yamaguchi et al, 1989).

In addition, oligospermic patients with varicoceles have been found to have intrascrotal temperatures of 0.6°C higher than oligospermic patients without varicoceles (Zorgniotti and MacLeod, 1973). Other studies have found that elevated intratesticular temperatures are common in oligospermic patients regardless of the cause of the spermatogenic defect (Mieusset et al, 1989).

Not all investigators have found an association between higher intratesticular temperatures and varicoceles (Tessler and Krahn, 1966; Stephenson et al, 1968).

Other suggested causes for the detrimental effect of varicoceles have included reflux of renal and adrenal metabolites from the renal vein (Comhaire and Vermeulen, 1974), decreased blood flow (Saypol et al, 1981), and hypoxia (Chakraborty et al, 1985).

Experimental animal models have demonstrated that after production of unilateral varicoceles, bilateral increases in testicular blood flow and temperature occur. Thus, it may be that an increase in blood flow leads to a secondary increase in testicular temperature, resulting in an impairment of spermatogenesis (Saypol et al, 1981). Subsequent repair of these experimental varicoceles has resulted in a normalization of blood flow and temperature (Green et al, 1984).

These data may explain the bilateral effect of unilateral varicoceles. Other factors may influence the effect of varicoceles on spermatogenesis. Several studies have demonstrated that smoking in the presence of a varicocele has a greater adverse effect than either factor alone (Klaiber et al, 1987; Peng et al, 1990).

Varicoceles are associated with smaller ipsilateral testes in both adolescents and adults. In addition, ipsilateral testicular growth is often impaired in adolescents with varicoceles (Lyon et al, 1982; World Health Organization, 1992).

Semen samples from men with varicoceles have demonstrated decreased motility in 90% of patients and sperm concentrations less than 20 million/ml in 65% of patients. In addition, specific abnormalities in sperm morphology have been described as a stress pattern consisting of increased numbers of amorphous cells and immature germ cells, as well as more than 15% tapered forms (MacLeod, 1965).

This pattern is not unique to varicoceles and has been found in infertile patients without varicoceles (Rodriguez-Rigau et al, 1981; Portuondo et al, 1983; Ayodeji and Baker, 1986). Thus, the presence of tapered forms is not adequate to make the diagnosis of a varicocele. Of note, with current usage of strict morphology, the presence of tapered forms may no longer be noted on the semen analysis report.

Varicocele repair has been demonstrated to result in catch-up growth in adolescents with varicoceles and ipsilateral loss of testicular volume (Kass and Belman, 1987). In addition, an improvement in semen parameters as well as testicular volume has been reported after repair of adolescent varicoceles (Okuyama et al, 1988). Infertile patients with varicoceles and testicular atrophy have worse semen parameters than those with varicoceles without atrophy (Sigman et al, 1997).

Varicocele repair should be considered in adolescents with grade II or III varicoceles associated with ipsilateral testicular growth retardation. Men with large varicoceles tend to have lower sperm counts and motility than those with smaller varicoceles (Steckel et al, 1993; Sigman et al, 1997). Not all investigators have found a relationship between the size of the varicocele and semen quality (Dubin and Amelar, 1977).

It is quite clear that varicoceles are detrimental to testicular growth and spermatogenesis. However, the majority of men with varicoceles are fertile because the effect is modest or they started with a high spermatogenic potential and remained within the fertile range despite the adverse effect of the varicocele.

Normal gonadotropin and testosterone levels are usually found on hormonal studies, whereas elevations of FSH may be found in some patients (Swerdloff and Walsh, 1975). Abnormal GnRH stimulation tests are often present in both adolescents and subfertile men with varicoceles (Hudson et al, 1981).

Treatment of varicoceles is directed at ligation or occlusion of the dilated testicular veins. Surgical, radiographic, and laparoscopic techniques have been used toward this end and are discussed in depth in Chapter 44, Surgical Management of Male Infertility and Other Scrotal Disorders. Although medical therapy has been proposed, there are no well-designed studies demonstrating a role for this approach (Netto Junior et al, 1984).

Improvement in seminal parameters is demonstrated in approximately 70% of patients after surgical varicocele repair. Improvements in motility are most common, occurring in 70% of patients, with improved sperm densities in 51% and improved morphology in 44% of patients.

Conception rates have averaged 40% to 50% (Tulloch, 1955; Brown, 1976; Glezerman et al, 1976; Dubin and Amelar, 1977; Cockett et al, 1979; Marks et al, 1986; Marmar and Kim, 1994).

Most studies examining the effect of the varicocele on fertility have been uncontrolled. A review of randomized and nonrandomized controlled studies reported an average pregnancy rate of 33% in the varicocele repair group compared with 16% in the control group (Schlegel, 1997).

An early study by Nilsson and coworkers (1979) reported no effect of varicocele repair; however, this study reported extremely low pregnancy rates in both the treated and the nontreated groups, suggesting the presence of significant untreated female factors.

Another negative study reported a 29% pregnancy rate in the treatment group and a 25.4% pregnancy rate in the nontreatment group over a 12-month observation period (Nieschlag et al, 1998). Of note, the nontreatment arm was not a true control because the spouse’s gynecologist was requested to optimize female reproductive functions and the couples were counseled on a regular basis.

Two additional prospective randomized studies have reported a significant positive effect of varicocele repair on fertility. Madgar and associates (1995) reported a 60% pregnancy rate at 1 year in the treatment group as opposed to a 10% pregnancy rate in the nontreatment arm.

After 1 year of observation, varicocele repair was performed on the patients in the nontreatment arm whose partners had not conceived. In this group, a 44% pregnancy rate was reported within 12 months after varicocele repair. A large study by the World Health Organization reported a 1-year pregnancy rate of 34.8% in the treatment group as compared with a 16.7% in the nontreatment group (Hargreave, 1997).

The presence of a varicocele alone is not an indication for varicocele repair, because the majority of men with varicoceles are fertile. The presence of a clinically detectable varicocele associated with an abnormal semen analysis in an infertile couple is an appropriate indication for treatment after the female partner has been evaluated. Azoospermic patients will generally not obtain normal semen parameters after varicocele repair.

However, significant numbers of these patients may develop low sperm densities in their semen samples, allowing them to proceed with IVF with ICSI after varicocele repair (Mehan, 1976; Matthews et al, 1998; Kim et al, 1999).

Sertoli Cell–Only Syndrome

Sertoli cell–only syndrome is a histologic diagnosis. The cause is likely multifactorial, with some patients having defined Y chromosome microdeletions and others having apparently normal genetic evaluations. Patients usually present with small to normal-sized testes and azoospermic semen specimens.

Phenotypically, these patients are normally virilized males. Sertoli cells line seminiferous tubules with a complete absence of germ cells and normal interstitium (Rothman et al, 1982).

Plasma FSH levels are often, but not invariably, elevated because of the absence of germ cells, whereas plasma testosterone and LH levels are normal. Despite an apparent absence of germ cells on standard histologic examination, spermatozoa may be recovered in up to 50% of patients undergoing attempts at testicular sperm extraction (Tournaye et al, 1997).


Postpubertal mumps results in orchitis in approximately 30% of patients with bilateral involvement in 10% to 30% of the cases (Beard et al, 1977). Permanent testicular atrophy may develop within several months to several years after infection. Pathologically, there is intense interstitial edema and mononuclear infiltration (McKendrick and Nishtar, 1966).

This may result in atrophy of the seminiferous tubules. Severe bilateral orchitis may result in hypergonadotropic hypogonadism and gynecomastia. The entity has become uncommon since the advent of a mumps vaccine, although clusters of cases have occurred in nonvaccinated patients (Casella et al, 1997). One randomized study demonstrated a quick recovery and absence of permanent testicular atrophy with treatment with interferon alfa 2b during the orchitis (Ku et al, 1999).

Treatment with the long-acting GnRH analogue has also been reported effective (Vicari and Mongioi, 1995). Testicular sperm extraction combined with IVF and ICSI may be attempted in patients made azoospermic by mumps orchitis (Lin et al, 1999). Orchitis may also develop in patients with syphilis, gonorrhea, leprosy, and mononucleosis.

Myotonic Dystrophy

Myotonic dystrophy causes myotonia, which is a condition of delayed muscle relaxation after contraction. Patients may also demonstrate posterior subcapsular cataracts, cardiac conduction defects, premature frontal baldness, and mental retardation.

Testicular atrophy develops in up to 80% of patients during adulthood (Drucker WD et al, 1963). Leydig cells typically are uninvolved, with biopsy specimens demonstrating severe tubular sclerosis. Serum FSH levels are elevated (Mahler and Parizel, 1982).

Of interest, before the development of testicular atrophy, some patients with apparently normal semen parameters are sterile. A recent study has suggested defects in the ability of sperm to undergo capacitation and the acrosome reaction (Hortas et al, 2000). The disease is transmitted as an autosomal dominant trait with variable penetrance. Myotonic dystrophy is due to an unstable CTG sequence in a protein kinase gene on chromosome 19 (Buxton et al, 1992; Jansen et al, 1994).

There is no therapy to prevent the testicular atrophy in these patients. There have been no reports of ARTs using sperm from patients with myotonic dystrophy.



The majority of chemotherapeutic agents adversely affect spermatogenesis (Parvinen et al, 1984). The most susceptible cells are those most actively dividing and consist of spermatogonia and spermatocytes up to the preleptotene stage. The specific combination of drugs used for therapy, the dose administered, and the age of the patient at the time of treatment are determinants of the specific effect on the gonads.

Studies of large groups of patients who survived cancers in childhood have demonstrated that fertility rates of patients treated with alkylating agents were 60% lower than in nontreated controls. As single drugs, alkylating agents and procarbazine seem to result in the greatest amount of testicular damage (Waxman, 1983).

The use of multidrug chemotherapeutic regimens has made it difficult to determine which specific agents are responsible for specific defects. Permanent sterility occurs in 80% to 100% of patients with Hodgkin’s disease treated with MOPP (mechlorethamine, vincristine [Oncovin], procarbazine, prednisone) and COPP (cyclophosphamide, vincristine [Oncovin], procarbazine, prednisone) regimens (Roeser et al, 1978; Kreuser et al, 1987).

More recent protocols such as NOVP (mitoxantrone [Novantrone], vincristine [Oncovin], vinblastine, and prednisone) have resulted in fertility being restored to pretreatment levels in Hodgkin’s disease patients (Meistrich et al, 1997). Less severe effects on fertility have also been found after treatment of non-Hodgkin’s lymphoma and leukemia, conditions in which less toxic chemotherapeutic regimens are used (Bokemeyer et al, 1994).

During chemotherapy, most patients demonstrate elevations of serum FSH levels that correlate with the development of azoospermia. Those patients in whom FSH levels decline demonstrate a return of spermatogenesis, whereas those in whom FSH levels remain elevated are unlikely to demonstrate spermatogenesis (Kader et al, 1991). Of interest, patients with leukemia and lymphoma often have some degree of spermatogenic impairment before treatment, with Hodgkin’s lymphoma patients having the lowest pretreatment semen quality (Lass et al, 1998; Hallak et al, 1999).

Preexisting spermatogenic defects in the contralateral testis are found in 25% of testicular cancer patients (Berthelsen and Skakkebaek, 1983). These defects tend to be worse than in patients with nontesticular malignancies (Lass et al, 1998).

A resumption of spermatogenesis occurs in 50% to 60% of these patients with the use of chemotherapeutic regimens such as PVB (cisplatin, vinblastine, bleomycin), PVP-16(cisplatin, etoposide), and POMB/ACE (cisplatin, vincristin [Oncovin], methotrexate, bleomycin/actinomycin, cyclophosphamide, etoposide) (Drasga et al, 1983; Hendry et al, 1983; Fossa et al, 1985; Kreuser et al, 1987; Nijman et al, 1987; Rustin et al, 1987).

The cumulative dose of cisplatin determines whether spermatogenesis is impaired irreversibly by chemotherapy (Pont and Albrecht, 1997). With cisplatin-based chemotherapy, most patients will become azoospermic; however, the majority will recover spermatogenesis within 4 years (Ohl and Sonksen, 1996).

There appears to be no increased risk of birth defects in children born to patients after chemotherapy (Senturia et al, 1985). Of note, sex chromosomal and autosomal aneuploidy has been noted in human sperm during chemotherapy (Robbins et al, 1997). Thus, patients should bank sperm before, not during, chemotherapy. In addition, contraception should be used during and for a period of time (3 to 24 months) after chemotherapy.

Attempts to protect spermatogenesis during chemotherapy have been disappointing. Down-regulation of spermatogenesis with GnRH agonists has not been successful (Johnson et al, 1985; Waxman, 1987). Future attempts may focus on cryopreservation and in vitro culture of testicular tissue before therapy and germ cell transplantation (Avarbock et al, 1996).

Radiation Exposure

Because of the high rate of cell division, the germinal epithelium is very radiosensitive. Spermatids are more resistant than spermatogonia or spermatocytes. Leydig cells are reasonably radioresistant; therefore, testosterone levels usually remain normal after radiation exposure. Serum FSH levels increase after radiation but may revert to normal after a return of spermatogenesis. Azoospermia usually results from doses of over 65 cGy (Hahn et al, 1982; Sandeman, 1966).

After doses less than 100 cGy, recovery takes 9 months to 18 months, with doses of 200 to 300 cGy, recovery may take 30 months; at doses of 400 to 600 cGy, more than 5 years may be required for spermatogenesis to return (Rowley et al, 1974). Because radiation directed at the abdomen scatters and affects the testes, most centers use gonadal shielding. Hahn and colleagues (1982) estimated the gonadal exposure at 78 cGy in the presence of gonadal shielding. Sperm counts decreased into the oligospermic range between 1 and 4 months after radiation. Most men were azoospermic between 2.5 and 7.5 months after radiation.

Semen quality will usually return to baseline within 2 years after radiation therapy for seminoma (Ohl and Sonksen, 1996). Approximately one fourth of patients may become permanently infertile from such radiation treatment (Fossa et al, 1989). After radiation therapy, most patients are advised to avoid conception for 2 years until the prognosis is more certain. Pregnancies after treatment have revealed no evidence of an increase in the prevalence of congenital anomalies in the offspring of these patients (Senturia et al, 1985; Nygaard et al, 1991).


There is experimental evidence that heat exposure may be detrimental to spermatogenesis. A meta-analysis examining occupations in which heat exposure occurs reported a detrimental effect on sperm morphology and time to conception (Thonneau et al, 1998).

A very small prospective randomized study found lower sperm counts in those wearing tight underwear as opposed to loose boxer shorts (Tiemessen et al, 1996). However, a larger study has demonstrated no effect (Wang et al, 1997). We currently recommend men wear whatever underwear they feel comfortable in.

Environmental Toxins and Occupational Exposures

Spermatogenesis has been shown to be adversely effected by exposure to lead, arsenic, hydrocarbons, cadmium, amebicide soil fumigants, as well as 2-bromopropane, a recently introduced substitute for chlorfluorocarbons (Lancranjan et al, 1975; Toth, 1979; Van Thiel et al, 1979; Lipshultz et al, 1980; Kim et al, 1999; De Celis et al, 2000; Telisman et al, 2000). Most controversial is the theory that sperm counts in men are declining over time, and the reason for this decline is prenatal exposure to environmental toxicants that have estrogenic effects on the embryo (Sharpe and Skakkebaek, 1993). Data are currently insufficient to support or refute this theory.

Other Gonadotoxins, Medications, and Drugs

Drugs. Harmon and Aliapooulios, 1972; Hembree et al, 1976). Cocaine use has been correlated with decreased sperm counts (Bracken et al, 1990).

Medications. A higher prevalence of epididymal cysts and cryptorchidism has been noted in patients with prenatal exposure to DES (Coscrove et al, 1977). Of significance, follow-up studies of men with prenatal DES exposure have revealed no adverse effects on fertility or sexual function (Wilcox et al, 1995). Early maturation arrest at the primary spermatocyte stage may be induced by high doses of nitrofurantoin (Nelson and Steinburger, 1952; Nelson and Bunge, 1957). Sulfasalazine treatment for ulcerative colitis induces reversible defects in sperm concentration and motility (Birnie et al, 1981; Toovey et al, 1981).

Patients with ulcerative colitis should be treated with 5-aminosalicylic acid, which does not affect semen parameters (Riley et al, 1987). There is evidence that calcium channel blockers cause a reversible functional defect in sperm, interfering with the ability of sperm to fertilize eggs but otherwise not interfering with sperm production or standard semen analysis parameters (Benoff et al, 1994). Although this effect may be reversible after cessation of the medication (Hershlag et al, 1995), not all investigators have found detrimental effects with calcium channel blockers (Katsoff and Check, 1997).

Alcohol. Both the testes and the liver are directly affected by ethanol. Testicular atrophy is commonly found in chronic alcoholics. Testicular specimens demonstrate peritubular fibrosis and a reduction in the number of germ cells. Free testosterone levels are often decreased, whereas total testosterone levels may be within the normal range secondary to elevated levels of testosterone-estradiol–binding globulin.

Patients, therefore, may demonstrate erectile dysfunction and gynecomastia as well as a decrease in general virilization (Mendelson et al, 1977). Gonadotropin levels may or may not be increased, because pituitary function may be suppressed. Studies on the acute consumption of alcohol in nonalcoholics demonstrate fallen testosterone levels with ingestion of alcohol (Gordon et al, 1976). No correlation has been found between sperm count or motility and the level of ethanol consumption in groups of infertile men (Marshburn et al, 1989; Dunphy et al, 1991).

In addition, the chance of couples conceiving (fecundity) has not been found to be associated with alcohol consumption in men (Goverde et al, 1995; Curtis et al, 1997; Olsen et al, 1997). Of interest, one study did find an association between female alcohol consumption and fecundity, whereas most studies have not (Jensen et al, 1998a).

Cigarettes. The data on the effect of cigarette smoking on semen parameters and fertility are conflicting. Studies have correlated smoking with adverse effects on parameters such as seminal volume, sperm count, motility, morphology, and increased numbers of white blood cells in the semen (Evans et al, 1981; Marshburn et al, 1989; Close et al, 1990; Vine et al, 1996; Merino et al, 1998; Chia et al, 2000). Other studies, however, have found no influence of cigarette smoking on semen parameters (Rodriguez-Rigau et al, 1982; Dikshit et al, 1987; Dunphy et al, 1991; Osser et al, 1992; Goverde et al, 1995). Smoking has been found to have an adverse effect on female fecundity, but most studies demonstrate no significant effect on male fecundity (Goverde et al, 1995; Bolumar et al, 1996; Jensen et al, 1998b). Of note, a study by Jensen, although finding no adverse effect of current smoking on male fecundity, did demonstrate an adverse effect of prenatal smoke exposure on subsequent male fertility in the offspring (Jensen et al, 1998b).

Caffeine. It has been suggested that caffeine affects sperm density and shape; this has not been confirmed (Fraser, 1984). Studies examining fecundity have not consistently demonstrated a direct effect of caffeine (Jensen et al, 1998c). Of note, caffeine applied to sperm in vitro results in increased sperm motility. We do not recommend that our patients eliminate caffeine from their diet.

Idiopathic Infertility

Despite the advancements in diagnostic methodology, up to 25% of patients exhibit abnormal semen analyses for which no etiology can be identified. This condition is referred to as idiopathic male infertility and likely is associated with a multitude of causes.

Not unexpectedly, semen parameters of these patients demonstrate a wide range of abnormalities. Although often referred to as idiopathic oligospermia, the vast majority of these patients have abnormalities of all semen parameters, or oligoasthenoteratospermia. Isolated abnormalities of sperm concentration, motility, and morphology are much less common (see Table 43–5).

History and physical examination are generally unremarkable, and hormonal determinations are typically normal. However, patients with mild elevations of FSH levels are also included, because this hormonal abnormality is a result of, rather than a cause of, abnormal spermatogenesis.

In the absence of an identifiable or correctable etiology, patients with idiopathic male infertility are managed with either empirical medical therapy or ART. A variety of medical therapies have been recommended to treat this group of patients. However, with few exceptions, none of these therapies has been shown to be effective in repeated controlled trials. A meta-analysis of all controlled studies for idiopathic male infertility has failed to reveal significant efficacy of currently available treatments (O’Donovan et al, 1993).

Yet, because of isolated case reports and small patient series demonstrating efficacy of some of these agents, there is continued hope that they may be effective in select subpopulations of men with idiopathic infertility. However, there is a significant background pregnancy rate (26%) for untreated couples with abnormal semen parameters (Collins et al, 1983). Because of the variability in semen analyses as well as the occurrence of treatment-independent pregnancies in couples, placebo-controlled, randomly assigned, double-blinded, crossover studies are required to determine the therapeutic efficacy of empiric medical therapies.

ARTs, which offer some hope to these patients, are discussed in a later section. These techniques include semen processing associated with IUI, IVF, and related technologies as well as micromanipulation of sperm and/or ova. For the majority of patients, initial attempts are directed at improving the quality of semen through treatment of specific abnormalities. If empirical pharmacologic therapy is going to be used, it should be administered for a minimum 3- to 6-month period so that at least one full spermatogenic cycle will be incorporated.

If this is unsuccessful, ARTs are employed. However, in view of the limited efficacy of empirical therapy, many couples wish to proceed with ARTs alone or a combination of both approaches simultaneously, which emphasizes the need for individualized therapy. Following is a discussion of the pharmacologic agents used for empirical therapy for idiopathic male infertility.

Testosterone-Rebound Therapy
Testosterone-rebound therapy involves large doses of exogenous testosterone, which are administered parenterally to suppress the activity of the patient’s pituitary gland. Suppression of pituitary release of LH, in turn, reduces the intratesticular level of testosterone.

The androgen therapy is then stopped in the hope that the system will rebound and improved spermatogenesis will result. There has been only one controlled study of this approach, and it was negative (Wang et al, 1983). There is currently no role for testosterone-rebound therapy, because there are other methods that are equally good or better and because some patients have persistent azoospermia after treatment.

Gonadotropin-Releasing Hormone
Several investigators have tested GnRH for idiopathic oligospermia with disappointing results. LH is normally released by the pituitary gland in a pulsatile fashion. Aulitzky and associates (1988), in an uncontrolled study, selected 14 men with idiopathic infertility who had an abnormal pulse pattern in their LH secretion.

They observed an improvement in sperm count in 8, and the partners of 3 became pregnant. Although GnRH therapy is efficacious in the treatment of patients with Kallmann’s syndrome, two controlled studies failed to find efficacy in patients with idiopathic infertility (Badenoch et al, 1988; Crottaz et al, 1992).

Therefore, we do not recommend GnRH therapy in patients with idiopathic infertility, owing to its high cost and lack of efficacy.

The two gonadotropins, FSH and LH, stimulate spermatogenesis and steroidogenesis, respectively. Historically, these hormones were only available as hMG and hCG, which were purified from the urine of menopausal and pregnant women, respectively.

Now, more purified as well as recombinant forms of these gonadotropins are available. Numerous uncontrolled studies have been performed with hCG and significantly less with hMG in men with idiopathic oligospermia. These studies have revealed limited efficacy, whereas the only published controlled study failed to find any efficacy over placebo (Knuth et al, 1987). As with GnRH, these treatments are expensive and of limited efficacy. We do not recommend these therapies in men without a demonstrable hormonal abnormality.

Clomiphene Citrate and Tamoxifen

Antiestrogen agents are the most commonly used medical therapy in the United States for idiopathic oligospermia. They increase pituitary gonadotropin secretion by blocking feedback inhibition, thus increasing serum FSH and LH levels as well as the testicular production of testosterone. Clomiphene citrate (25 mg/day) is the standard recommended treatment. Higher doses may cause down-regulation of the system, although they are occasionally indicated.

There are many uncontrolled studies reporting a high percentage of patients with improved semen quality and impressive pregnancy rates among their partners. However, the majority of investigators have found pregnancy rates lower than 30%.

Of the nine published controlled studies of clomiphene citrate dating back to the early 1970s, the majority failed to identify any efficacy over placebo (Foss et al, 1973; Paulson, 1979; Rönnberg, 1980; Abel et al, 1982; Micic and Dotlic, 1985; Sokol et al, 1988; World Health Organization, 1992). However, two studies did reveal a positive effect on both sperm counts and pregnancy rates (Wang et al, 1983; Check et al, 1989).

Controlled studies using tamoxifen (10 to 20 mg/day) have all had negative results (Willis et al, 1977; Török, 1985; AinMelk et al, 1987; Krause et al, 1992). Antiestrogens are relatively inexpensive and safe oral medications for the treatment for idiopathic male infertility, which explains their popularity. Because their efficacy is in doubt, prolonged courses of empirical therapy should not be used as a substitute for more effective modes of management.


Testolactone is an aromatase inhibitor that has a similar effect to an antiestrogen because it blocks the conversion of testosterone to estradiol. This oral agent is very expensive and has not been widely studied. A single controlled study failed to find a significant effect on fertility over placebo (Clark et al, 1989). There is a theoretical basis to recommend either this agent or antiestrogens in men with low testosterone-to-estradiol ratios, such as obese men, but this has never been formally studied.


Various androgens have been used in the treatment of idiopathic male infertility. However, in the presence of an intact hypothalamic-pituitary-gonadal axis, the effect of androgen administration is to lower the concentration of testosterone in the testis. Mesterolone is a synthetic androgen widely used in Europe to treat idiopathic male infertility in doses ranging from as low as 2 mg/day up to 150 mg/day.

The World Health Organization (1989) sponsored a double-blind study of men who received placebo or 75 or 150 mg of mesterolone daily. This study, as well as four other controlled studies, failed to show a positive effect of this drug on fertility (Aafjes et al, 1983; Wang et al, 1983; Comhaire, 1990; Gerris et al, 1991). Continuous androgen administration has a contraceptive effect on men by lowering intratesticular testosterone concentration and should never be used in the treatment of infertility.

Miscellaneous Treatments

A variety of vitamins, nutritional supplements, and anti-inflammatory agents have been used in the empirical therapy of infertile men. Thyroxine, arginine, corticosteroids, antibiotics, zinc, methylxanthines, bromocriptine, and vitamins A, E, and C have all been shown to be of little or no benefit in the treatment of infertile men without evidence of a specific deficiency.

Controlled studies of kallikrein, indomethacin, and glutathione have shown mixed results but nothing sufficient to promote their use (Barkay et al, 1984; Glezerman et al, 1993; Lenzi et al, 1993).l-Carnitine is a known component of epididymal secretions and is now available as an over-the-counter nutritional supplement for the treatment of idiopathic male infertility.

Studies have not shown a direct relationship between semen l-carnitine levels and fertility or that orally administered carnitine increases levels within the epididymis (Soufir et al, 1984). Uncontrolled studies demonstrate improvement in semen parameters but not fertility, and controlled studies have not been performed (Vitali et al, 1995).

Sperm Delivery Disorders

Ductal Obstruction

Genital duct obstruction is a potentially surgically curable cause of male infertility. Obstruction of the ductal system is found in 7% to 12% of all infertile men and is much more common in azoospermic men (Dubin and Amelar, 1971; Greenberg et al, 1978; Jarow et al, 1989a).

Obstruction may be bilateral or unilateral and may occur at multiple locations. For instance, concomitant epididymal obstruction may be present in patients with either vasal or ejaculatory duct obstruction, presumably caused by a backpressure phenomena (625, 1979, 1980). Unilateral ductal obstruction should not adversely affect fertility except when there is contralateral testicular pathology such as torsion, varicocele, or cryptorchidism.

However, unilateral obstruction may be a risk factor for the development of antisperm antibodies. Detection of unilateral ductal obstruction in a patient with oligospermia requires a high degree of clinical suspicion (Matsuda et al, 1992). The etiology of ductal obstruction may be congenital, because of malformation or absence of ductal structures, such as in CBAVD, or acquired secondary to infection, stricture, or vasectomy.

CBAVD is the most common cause of obstructive azoospermia in patients who have not undergone elective sterilization (Jarow et al, 1989b). Careful physical examination reveals a range of epididymal findings, but the most common finding is the presence of the caput epididymis without the remainder of the epididymis or the vas deferens being present. The seminal vesicles are often but not always absent or hypoplastic in these patients (Goldstein and Schlossberg, 1988).

Unilateral renal agenesis has also been noted in patients with CBAVD but is much more common when the vasal defect is unilateral (Mickle et al, 1995; Schlegel et al, 1996). As noted previously, low-volume azoospermic ejaculates are characteristic of these patients. Testicular biopsies generally demonstrate normal spermatogenesis. Surgical treatment with either alloplastic (Belker et al, 1986) or natural spermatoceles (Wen et al, 1993) has resulted in extremely low pregnancy rates of 0% to 4% and is not currently recommended.

The development of ICSI for IVF has greatly improved the treatment of couples with infertility due to this condition. Temple-Smith and colleagues (1985) were the first to report a pregnancy after IVF using sperm aspirated from the epididymis of a patient with obstructive azoospermia (Temple-Smith et al, 1985). Subsequent reports showed that this same technique could be used in men with obstructive azoospermia due to congenital absence of the vas deferens (Silber et al, 1990).

However, the poor fertilization rates achieved using standard IVF with epididymal sperm limited the overall success of this treatment until ICSI was introduced (Schlegel et al, 1995). The epididymis is not the only potential source of sperm in these patients. Sperm retrieval can be achieved by either open surgical or percutaneous aspiration from either the testis or the epididymis. If there is no epididymis or viable epididymal sperm are not identified, testicular biopsy with sperm harvesting may be used.

The currently recommended management of couples with infertility owing to CBAVD is sperm retrieval combined with IVF using ICSI after appropriate genetic testing and counseling of the couple regarding the risk of cystic fibrosis.

Bilateral complete ejaculatory duct obstruction has pathognomonic clinical findings of acidic, fructose-negative, low-volume ejaculate azoospermia. The main differential diagnosis for these findings is CBAVD, which can easily be distinguished by the presence of vasa on physical examination. In contrast, the diagnosis of partial ejaculatory duct obstruction is much more difficult to establish. Some authors believe that partial obstruction of the seminal vesicles can cause poor motility, poor forward progression, and decreased sperm count.

However, this is still a controversial diagnosis and further investigation is required (Jarow, 1996). If partial or unilateral obstruction is suspected because of a low-volume ejaculate in an oligospermic patient or because of a normal-volume ejaculate with low motility and poor forward progression, TRUS can be used to identify dilated seminal vesicles.

The diagnosis of ejaculatory duct obstruction is traditionally made by the combination of testicular biopsy and vasography. However, TRUS is currently the imaging modality of choice for the evaluation of patients with suspected ejaculatory duct obstruction (Belker and Steinbock, 1990).

The presence of dilated seminal vesicles is suggestive of ejaculatory duct obstruction, which can be confirmed by finding numerous sperm in the seminal vesicles (Jarow, 1994). The current standard treatment for ejaculatory duct obstruction is transurethral resection of the ejaculatory ducts. Balloon dilatation of the ejaculatory ducts has been reported but is still investigational (Jarow and Zagoria, 1995).

Ductal obstruction may also be present in either the epididymis or the vas deferens. The most common cause is elective sterilization. Semen volume is typically normal in these patients because the testis and epididymis contribute less than 5% to the ejaculate.

A variety of biochemical tests have been employed to detect the presence of epididymal enzymes, such as carnitine in the ejaculate, with very little success (Zopfgen et al, 2000). The diagnosis of either vasal or epididymal obstruction is established at the time of scrotal exploration for intended reconstruction in azoospermic patients with intact spermatogenesis documented on a prior testicular biopsy. The management of these patients is either surgical reconstruction or sperm retrieval combined with IVF using ICSI.

Ejaculatory Problems

Any process that interferes with the peristaltic function of the vas deferens and closure of the bladder neck may result in either failure of emission or retrograde ejaculation. Ejaculatory dysfunction should be suspected in any patient with low-volume (<1.0 ml) or absent ejaculate and should be distinguished from anorgasmia. Retrograde ejaculation is diagnosed by examining the postejaculate urine for sperm.

Although exact criteria have not been established for a positive postejaculate urinalysis, the finding of greater than 10 to 15 sperm per high-power field confirms the presence of retrograde ejaculation. In contrast, sperm will not be present in the urine of a patient with failure of emission, which must be diagnosed by clinical suspicion.

The causes of ejaculatory dysfunction are divided into anatomic and functional, because these categories determine effective management. Patients with retrograde ejaculation from anatomic causes including bladder neck surgery and transurethral resection of the prostate do not respond to medical therapies.

In contrast, patients with ejaculatory dysfunction from neurologic abnormalities such as diabetes mellitus, multiple sclerosis, and retroperitoneal surgery may respond to medical intervention.

Pharmacologic therapy for retrograde ejaculation is likely to be effective only in patients who do not have surgical changes of the bladder neck and for patients with failure of emission. Ephedrine sulfate (25 to 50 mg qid), pseudoephedrine (60 mg qid), and imipramine (25 mg bid) may induce ejaculation secondary to an increase in the sympathetic tone of the internal sphincter and vas deferens. Some patients with failure of seminal emission may be converted to retrograde ejaculation by oral medical sympathomimetic therapy.

Medical therapy for ejaculatory dysfunction is administered on a cyclical basis timed to the wife’s ovulatory cycle. These medications are more effective if given for a period of at least 7 to 10 days before planned ejaculation, and tolerance may develop if administered continuously over several cycles. Success is unlikely if no effect is observed within 2 weeks of treatment.

Those patients with retrograde ejaculation unresponsive to medical therapy or caused by ablation of the bladder neck may be treated by recovery of sperm from the bladder urine combined with IUI (Shangold et al, 1990). The urine pH and osmolarity should be optimized for sperm survival in the urine by adjusting the patient’s fluid intake and by the administration of sodium bicarbonate (650 mg qid), acetazolamide (250 mg qid), or baking soda (2 tablespoons the night before and 2 hours before ejaculation).

This may take several attempts before the best combination is found. Sperm are recovered from the urine by gentle centrifugation and then washed in an appropriate insemination media. Deposition of pH-adjusted media into the bladder before ejaculation will lessen the toxic effect of urine and may be used as an adjunct to protect the sperm if the other methods are not successful.

Surgical reconstruction of the bladder neck had been attempted in the past (Abrahams et al, 1975), but, with the advancement in semen processing techniques, this surgery is not indicated today.

Spinal cord injury is a common cause of ejaculatory dysfunction (failure of emission), and these patients rarely respond to oral medical therapy. Penile vibratory stimulation results in ejaculation in approximately 70% of spinal cord–injured men (Ohl et al, 1996; Brackett et al, 1998). Although specially designed equipment with specific vibration frequencies and amplitudes are available, many practitioners have found good results using readily available vibrators intended for general use.

This approach should be used in patients with upper motoneuron lesions such as spinal cord injuries above T10. Patients with spinal cord injuries involving the lower spinal cord or peripheral neural lesions, such as after retroperitoneal surgery, are not likely to respond to vibratory ejaculation.

Electroejaculation involves the application of pulsed electrical current applied to the periprostatic plexus with the use of either a rectal probe or needle electrodes. The ejaculation has been successfully induced in patients after spinal cord injury, retroperitoneal lymph node dissections, multiple sclerosis, transverse myelitis, and diabetes mellitus. Rectal probe ejaculation produces an ejaculation in approximately 75% of patients (Brindley, 1981; Shaban et al, 1988; Ohl et al, 1989).

In spinal cord–injured patients with complete cord lesions, the procedure may be performed in an office setting without anesthesia. However, in patients with incomplete cord lesions, or with intact pelvic sensation, general anesthesia is required (Bennett et al, 1987).

Electroejaculation should be attempted in only those patients who are not candidates for or who have failed to respond to penile vibratory stimulation. The results of electroejaculation may be optimized by simultaneous collection of bladder urine, because retrograde ejaculation frequently occurs, and by eradication of any urinary tract infections or colonization.

Autonomic dysreflexia may occur during either vibratory stimulation or electroejaculation. This occurs primarily in patients with lesions above the T4 level. Pretreatment of these patients with 10 to 20 mg of sublingual nifedipine, 15 minutes before the procedure, will usually allow the procedure to be performed safely (Steinberger et al, 1990).

The typical ejaculate obtained from patients with failure of emission has large numbers of poor-quality sperm. Initial studies suggested that the electroejaculation process itself might be harmful to the sperm (Rajasekaran et al, 1994; Sikka et al, 1994). However, other studies have not supported this theory and similar quality sperm is obtained after penile vibratory stimulation (Witt et al, 1992; Ohl et al, 1994).

Another potential explanation is testicular changes after spinal cord injury caused by either a heat effect or a chronic infection (Hirsch et al, 1991; Elliott et al, 2000). Yet the most likely explanation may be the simplest, that is, stasis within the genital tract from a lack of peristalsis (Ohl et al, 1999).

Pregnancies have been achieved through the combination of penile vibratory stimulation (for patients with spinal cord lesions above T10) or electroejaculation and IUIs on a sporadic basis. However, the majority of couples do not conceive through IUI and require IVF using ICSI (Hakim et al, 1995; Nehra et al, 1996).

Sperm Function Disorders

Immunologic Infertility

Two approaches have been used for the treatment of immunologically mediated infertility. In the first, an attempt is made to suppress antibody formation. The second approach uses the spermatozoa, with or without semen processing, to remove antibodies or select those spermatozoa not bound by antibodies for the ARTs.

Corticosteroids have been the medications most commonly employed used to attempt to suppress antisperm antibody formation (Shulman, 1976). Regimens have included high-dose cyclical therapy, in which the male takes the medication for 1 week during each menstrual cycle of the female partner, as well as intermediate- and low-dose, long-term therapy.

Changes in antibody titers, semen parameters, and pregnancy rates have been inconsistent (Kremer et al, 1978; Dondero et al, 1979; Haas and Manganiello, 1987; Sharma et al, 1995). Most studies have been uncontrolled and report pregnancy rates of 30% to 40%. A double-blind crossover trial by Hendry and coworkers (1990) using an intermittent intermediate-dose regimen reported a 31% pregnancy rate in the steroid-treated group as compared with a less than 10% pregnancy rate in those receiving placebo.

A double-blind, placebo-controlled study by Haas and Manganiello (1987) found no effect of steroid therapy on pregnancy rates. It appears that, in some patients, steroid treatment may result in improved fertility; however, this does not occur in the majority of patients.

The vast majority of pregnancies have occurred in partners of men with normal sperm densities. The chance of success must be weighed against the possible complications; the most devastating of which is aseptic necrosis of the hip, which appears to be a dose-dependent complication of corticosteroid treatment. Semen-processing techniques have not been able to completely remove antibodies from the surface of spermatozoa (Lenzi et al, 1988). Chymotrypsin has been used to digest off the Fc portion from the spermatozoa.

This has resulted in improved sperm motility and a decrease in sperm agglutination. Improved pregnancy rates have been reported in couples undergoing this type of semen processing when combined with IUI with superovulation (Bollendorf et al, 1994). Attempts to select sperm without antibody binding have not been successful.

IVF has been used in the management of couples with antisperm antibodies, and decreased fertilization rates have been documented when high levels of antisperm antibodies are present (Lahteenmaki, 1993). ICSI used in conjunction with IVF is a highly effective treatment for these couples (Clarke et al, 1997).

The advantages and risks of prednisone therapy should be weighed against the chances of pregnancy through IUI or IVF with ICSI. Although we present the patient the option of corticosteroid treatment, we encourage couples to consider IUI or IVF with ICSI if the semen is of adequate quality. For couples wishing to proceed with immunosuppressive therapy, we recommend an intermediate cyclical steroid regimen such as the one used by Hendry and coworkers (1990).

Ultrastructural Abnormalities of Sperm

Electron microscopy allows for the identification of ultrastructural abnormalities of spermatozoa. Defects in outer dense fibers, microtubules, and mitochondria may be detected. The most common axonemal defects involve abnormal numbers of the microtubules and their associated elements. Thus, aberration from the standard 9 + 2 microtubule organization may result in immotile spermatozoa.

The most common of these defects involves a complete absence of both inner and outer dynein arms. Other patterns consist of absence of either the inner or the outer arms. In addition, in some cases, not all spermatozoa demonstrate the same defects.

Complete absence of the axonemal elements is rare. Ultrastructural axonemal defects are commonly associated with identical defects in the cilia of the respiratory tract. This condition is known as the immotile cilia syndrome or primary ciliary dyskinesia (Eliasson et al, 1977). Chronic respiratory tract infections with bronchiectasis are common in these patients. When these clinical findings are combined with situs inversus, which is present in 50% of these cases, the patient has Kartagener’s syndrome.

Kartagener’s syndrome has an autosomal recessive inheritance pattern. However, studies to date have not been able to localize the gene to one specific chromosome. Other ultrastructural defects of the flagella, such as radial spoke abnormalities and transposition of microtubules, have been identified in patients with immotile cilia. Thus, this syndrome is heterogeneous, and motility may be maintained by some spermatozoa (Sturgess et al, 1979; Jouannet et al, 1983).

Defects in the outer dense fibers are less common than microtubular defects. Abnormalities in these structures result in dyskinetic spermatozoa with flagella movement that is not normal (Feneux et al, 1985). Defects in the sperm head have also been identified by electron microscopy. Round-headed spermatozoa (globozoospermia) are characterized by an absence of the acrosome as well as absence of a cytoskeletal protein (calicin). Two types of this condition have been identified.

Type I is characterized by a complete absence of the acrosome and acrosomal contents. This is the classic form of this disease in which spermatozoa lack the ability to fertilize human ova. Spermatozoa with the type II defect contain remnants of the acrosome and may retain some ability to fertilize ova (Singh, 1992). The inheritance pattern for these conditions remains unclear.

Although there have been a few pregnancies reported with ICSI of round-headed spermatozoa, most attempts have failed. The cause for this remains unclear, although it may involve premature chromosome condensation, owing to the oocyte not being activated after fertilization but remaining in metaphase II (Stone et al, 2000).
Finally, abnormalities of the connecting piece may be identified. Separation of the head from the tail is occasionally seen on the semen analysis, wherein headless motile flagella are noted.

Ultrastructural defects in the region of the junction of the sperm tail to the head have been identified; however, the genetics behind this remain unknown (Toyama et al, 2000). Although there is no cure for these ultrastructural conditions, the spermatozoa may be used for IVF with ICSI. Caution should be exercised when recommending this treatment in this population, because some of these conditions have a genetic basis.


Since about 1980, ARTs have gained a dominant role in the management of infertile couples. These techniques involve the manipulation of sperm or ova or both in an attempt to improve the chance of conception and resultant live birth rates.

Most ARTs are referred to by abbreviations rather than their full descriptive names ( Table 43–8). ART is frequently indicated in idiopathic male infertility, in unexplained infertility, or in cases in which no therapy is available or has effectively resulted in conception. The range of techniques varies from those that involve only manipulation of sperm to more sophisticated procedures involving the manipulation of sperm, ova, and/or embryos. Fertilization may occur inside the woman (in vivo) or in vitro.

Controlled ovarian hyperstimulation, hormonal stimulation of superovulation using gonadotropins, plays a critical part in most forms of ART. Because of the growth in the number of centers offering ARTs and the ability to perform IVF with ICSI in cases of severe male factor infertility, there has been a trend to pursue IVF as a first-line therapy. This approach either does not treat the male or ignores simpler, less expensive management options.

Because of the rising costs of medical care, the cost-effectiveness of this approach has been examined. Using this type of analysis, which commonly examines the cost per live delivery for a given technique, varicocele repair has been demonstrated to be more cost effective than IVF with ICSI (Schlegel, 1997).

Similarly, for patients after vasectomy, vasectomy reversal is more cost effective than IVF with ICSI; for patients who have failed one vasectomy reversal, a repeat vasectomy reversal has been found more cost effective than IVF with ICSI; and for patients with epididymal obstruction, vasoepididymostomy has been demonstrated to be more cost effective than IVF with ICSI (Kolettis and Thomas, 1997; Pavlovich and Schlaegel, 1997; Donovan et al, 1998).

In addition, IVF as an initial treatment for infertile couples in general is less cost effective than more standard approaches such as IUI (Van Voorhis et al, 1997; Guzick et al, 1998; Karande et al, 1999; Goverde et al, 2000). These data make it clear that ARTs should be used in appropriate settings and not as broad cure-all techniques.

For many couples in which infertility is caused by a mild to moderate male factor, IUI or IVF is an appropriate therapy. Because of the significantly lower cost of IUI, this is often the initial treatment, and couples failing to conceive may proceed to IVF.

If there is reason to believe the sperm may not fertilize ova without assistance, then IVF with ICSI may be the initial treatment. Such cases include severe male factor infertility in which inadequate numbers of motile sperm are available for regular IVF or IUI.

In addition, cases with abnormal sperm function testing, such as a zero score on the SPA, may proceed with IVF rather than IUI. If an initial IVF cycle demonstrates fertilization but pregnancy does not ensue, the couple may consider proceeding with IUI in lieu of further IVF cycles.

Semen Processing

Semen must be processed before use for ARTs. A variety of procedures are used, including simple sperm washing, swim-ups (allowing pelleted sperm to swim up into the supernatant), sedimentation, and centrifugation through density gradients (Pousette et al, 1986; Adeghe, 1987; Pardo et al, 1988; Tanphaichitr et al, 1988; McClure et al, 1989; Joshi et al, 1998). Seminal plasma is removed in all of these procedures, whereas other methods select motile sperm by removing nonmotile sperm and leukocytes.

Intrauterine Insemination

IUI involves using a small catheter to inject processed sperm through the cervix into the uterine cavity. It is hoped that by bypassing the cervical mucus, higher numbers of motile sperm will be able to reach to the fallopian tubes, thus increasing the chances of conception.

The injection of raw semen is contraindicated because seminal prostaglandins may cause severe uterine cramping and pelvic infection may be induced by bacterial contamination of the seminal fluid.

Male factor infertility, unexplained infertility, cervical mucus abnormalities, and anatomic abnormalities interfering with the deposition of sperm at the cervical os (severe hypospadias, retrograde ejaculation, and, in some cases, erectile dysfunction) are all indications for IUI. Female factors, such as cervical factor infertility, anatomic abnormalities interfering with vaginal intercourse, such as dyspareunia, and psychogenic sexual dysfunction are also indications for IUI.

Women undergoing IUI may be allowed to ovulate naturally (natural-cycle IUI) or may be given medication to induce maturation of multiple ova (superovulation or controlled ovarian hyperstimulation). Natural-cycle IUI is indicated primarily in cases in which proper deposition of semen into the vagina is not possible (hypospadias, ejaculatory dysfunction, sexual dysfunction).

The vast majority of controlled studies have shown no benefit to natural-cycle IUI for couples with male factor infertility from abnormal semen parameters (Hughes et al, 1987; Ho et al, 1989; te Velde et al, 1989; Martinez et al, 1990; Kirby et al, 1991). Natural-cycle IUI is also indicated when using cryopreserved sperm that demonstrated normal semen parameters before cryopreservation. This is the case with donor insemination or the use of sperm cryopreserved before chemotherapy or radiation therapy.

Because of the poor results of natural-cycle IUI in male factor couples, ovulation induction is often added to the regimen. A variety of controlled studies have been reported, the majority of which demonstrate improved pregnancy rates with this approach (Cruz et al, 1986; Bolton et al, 1989; Evans et al, 1991; Ho et al, 1992; Nulsen et al, 1993; Nan et al, 1994; Arici et al, 1994; Guzick et al, 1999).

Ovulation induction may be accomplished with oral medications such as clomiphene citrate or with the use of injectable gonadotropins. A review of current data clearly suggests that the baseline pregnancy rate in male factor couples is 1% to 3% per cycle with intercourse alone. This rate is not improved by natural-cycle IUI when the semen parameters are suboptimal.

However, IUI combined with ovulation induction improves the pregnancy rates above this baseline. Although the data with clomiphene citrate–stimulated cycles are not completely clear, pregnancy rates in the range of 5% to 8% per cycle are often obtained. IUI cycles with gonadotropin ovulation induction result in pregnancy rates of 10% to 15% per cycle on average.

In the majority of studies, most pregnancies occur within the first three cycles, with few pregnancies occurring beyond that point with a particular ovulation induction regimen. Thus, couples may often begin with clomiphene citrate cycles for several months. If no pregnancy ensues, they may move on to ovulation induction cycles using gonadotropins for up to three cycles.

IVF should be considered if pregnancy does not occur with IUI. As with pregnancy after intercourse, there is no minimum number of sperm that is required for pregnancy to occur after IUI. Significantly lower pregnancy rates occur with inseminating motile sperm counts of less than 1 million motile sperm. Pregnancy rates may increase with increasing motile sperm counts up to 10 to 20 million/ml but appear to plateau beyond that point (Berg et al, 1997; Centola, 1997).

It is our policy to offer IUI to couples with at least 1 million motile sperm after semen processing. Because semen processing has yields of only 10% to 20%, semen samples usually have at least 5 to 10 million motile sperm before processing. Complications of IUI include uterine cramping, which is generally self limited, pelvic infections in less than 0.5% of patients, and rare allergic reactions to the insemination media (Corson et al, 1989; Sonenthal et al, 1991).

Although the addition of ovulation induction has improved pregnancy rates, it has also resulted in multiple gestations in 15% to 30% of pregnancies. Approximately 80% of multiple gestations involve twins, with approximately 12% involving triplets and 7% involving more than three gestational sacs (Schenker et al, 1981; Dickey et al, 1992).

Occasionally, females whose male partners are HIV positive inquire as to whether HIV can be removed from the semen by processing. HIV has been found free in seminal plasma as well as in the white blood cells in the semen from HIV-1–infected men. Although a few centers have reported successful IUI after semen processing to remove viral particles, the safety of this approach requires further documentation (Semprini et al, 1992; Marina et al, 1998).

In Vitro Fertilization

IVF has become increasingly used to treat infertile couples. Most centers use gonadotropin superovulation to recruit multiple oocytes each cycle. Follicular development is monitored ultrasonically, and ova are harvested before ovulation with the use of ultrasound-guided needle aspiration. In vitro insemination is performed by mixing processed sperm with recovered oocytes.

In standard IVF, when fertilization occurs, the developing embryos are incubated for 2 to 3 days in culture and then placed transcervically into the uterus. Only 20% to 30% of transferred embryos will implant and produce clinical pregnancies. Recently, embryos have been cultured for 5 days to transfer them at the blastocyst stage. It appears that implantation rates may be higher with blastocyst transfer as compared with standard day 3 embryo transfer (Gardner et al, 1998).

This allows the transfer of fewer embryos, decreasing the risk of multiple gestations. Weakening of the zona pellucida (assisted hatching) has also been used to improve implantation rates, which may benefit a subgroup of IVF patients (De Vos et al, 2000).

More than 90% of inseminated oocytes are routinely fertilized when sperm function is normal. However, fertilization rates are reduced significantly when a male factor is present. With IVF combined with ICSI, single sperm are injected into individual ova.

This allows for fertilization with extremely low numbers of sperm (Van Steirteghem et al, 1993). This technique has essentially replaced prior micromanipulation procedures such as the subzonal insertion of sperm or other zona manipulation techniques.

IVF with ICSI is indicated in cases of severe male factor infertility, in couples with prior failed or poor fertilization during regular IVF cycles, or in cases in which the sperm demonstrate significant fertilizing ability defects (such as with round-headed sperm).

Clinical pregnancy refers to pregnancies in which at least one gestational sac is present within the uterus as documented by ultrasonography. This is in contrast to biochemical pregnancies, which may never reach the clinical pregnancy stage. Clinical pregnancy rates should be used when reporting IVF results. The clinical pregnancy rates by standard IVF average 20% to 30% per initiated cycle.

There is a significant effect of the female age on pregnancy rates with IVF. For instance, the Centers for Disease Control and Prevention’s survey for IVF cycles in 1997 reported pregnancy rates of 35.7% in women younger than 35 years and 13.2% in women older than 40 years (Centers for Disease Control and Prevention, 2000). Similar or better pregnancy rates are reported with ICSI (Society for Assisted Reproductive Technology, 1999).

It should be realized that pregnancy rates are not live birth rates because not all cycles in which women begin ovarian stimulation result in retrieval of oocytes and transfer of embryos and 15% of clinical pregnancies end in miscarriage. Thus, although the pregnancy rate for women younger than 35 years is 35.7%, the live birth rate is 30.7%.

Of note, in women older than 40 years, the pregnancy rate is 13.2%, whereas the live birth rate is 7.6% (Centers for Disease Control and Prevention, 2000).

With regular IVF, three to four embryos are commonly implanted, whereas with blastocyst transfer, only two embryos are usually transferred. Multiple gestations occur in 18% to 44% of pregnancies. Whereas the majority are twins, 3% to 14% are triplet or higher multiple gestations (Centers for Disease Control and Prevention, 2000).

Sperm Retrieval

IVF with ICSI still requires the presence of viable sperm. In those cases with azoospermia or only nonviable sperm in the semen, sperm retrieval may be considered. This may be employed in patients with either obstructive azoospermia or nonobstructive azoospermia. Both percutaneous and open techniques are currently used.

For patients with obstructive azoospermia, sperm may be retrieved from the ductal system or from the testicular parenchyma. In contrast, only testicular sperm retrieval is applicable for patients with nonobstructive azoospermia.

Microsurgical epididymal sperm aspiration (MESA) is commonly employed to retrieve sperm out of the ductal system in cases with obstructive azoospermia such as CBAVD (Silber et al, 1988). Some have advocated percutaneous epididymal sperm aspiration (PESA) as a less invasive technique that does not require microsurgical skills (Craft et al, 1995).

Pregnancy rates are likely comparable between open and percutaneous sperm retrieval techniques for patients with obstruction; however, many more sperm are retrieved by MESA than by PESA (Sheynkin et al, 1998). Because excess sperm may be cryopreserved and used for subsequent IVF cycles, patients are likely to only need one MESA procedure, whereas they may require multiple PESA procedures for subsequent cycles. When used for IVF with ICSI, there appears to be no difference in pregnancy rates between frozen and fresh epididymal sperm (Friedler et al, 1998; Garrels et al, 1998; Chang et al, 1999; Tournaye et al, 1999).

Other techniques that have been employed for retrieval of sperm in cases of obstructive azoospermia include seminal vesical aspiration and vasal aspiration (Hovatta et al, 1996; Jarow, 1996). These techniques are limited to those patients with either distal obstruction or anejaculation.

Testicular sperm retrieval techniques may be used in cases of obstructive as well as nonobstructive azoospermia. For obstructive azoospermia, there appears to be no significant difference in pregnancy rates regardless of the sperm retrieval technique employed, and sperm are usually retrieved with any of the techniques; however, open techniques tend to retrieve more sperm than percutaneous techniques.

In contrast, open surgical testicular sperm extraction techniques retrieve sperm in a greater percentage of patients and yield higher numbers of sperm than percutaneous techniques in patients with nonobstructive azoospermia (Friedler et al, 1997a; Ostad et al, 1998; Schlegel, 1999).

In one study that used optical magnification to identify larger seminiferous tubules for open testicular sperm extraction, an increase was reported in the percentage of patients in which sperm are retrieved of from 45% to 63% in cases of nonobstructive azoospermia (Schlegel, 1999). For patients with nonobstructive azoospermia, we recommend open testicular sperm extraction. For cases of obstructive azoospermia, either open or percutaneous techniques may be used.

Pregnancy rates in cases of nonobstructive azoospermia appear to be less than in cases of obstructive azoospermia (Palermo et al, 1999). It remains unclear whether pregnancy rates are the same for cryopreserved and fresh testicular sperm in cases of nonobstructive azoospermia (Friedler et al, 1997b; De Croo et al, 1998; Habermann et al, 2000).

For those cases of nonobstructive azoospermia in which no spermatozoa are retrieved, attempts have been made to use elongated spermatids as well as round spermatids. Elongated spermatids have been successfully used in several centers. However, the pregnancy rates and feasibility of using round spermatids remain highly controversial (Al Hasani et al, 1999; Ghazzawi et al, 1999; Gianaroli et al, 1999; Prapas et al, 1999; Schoysman et al, 1999; Zech et al, 2000).

Although this technology represents a major advance in the management of the infertile male, it must be remembered that these techniques are relatively new and their long-term safety has yet to be determined. There is evidence of an increase in sex chromosomal abnormalities in children born from ICSI cycles (Tarlatzis and Bili, 2000).

Additional evidence questioning increases in major congenital malformation rates as well as decreased cognitive development in children born after ICSI remains highly controversial (Kurinczuk et al, 1997; Bonduelle et al, 1998; Bowen et al, 1998; Hawkins et al, 1999; Sutcliffe et al, 1999; Givens, 2000). It is very important that the physician understands the appropriate role of these techniques and possible risks to properly counsel the infertile couple.