Berek Novak's Gyn 2019. Chapter 36 Infertility

CHAPTER 36

Infertility

KEY POINTS

The physician’s initial encounter with an infertile couple is the most

important one because it sets the tone for subsequent evaluation and

treatment. Factors from either or both partners may contribute to difficulties

in conceiving. Therefore, it is important to understand male and female

reproductive physiology and consider all possible diagnoses before pursuing

invasive treatment.

1 The main causes of infertility include male factor, decreased ovarian reserve,

ovulatory disorders (ovulatory factor), tubal injury, blockage, or paratubal adhesions

(including endometriosis with evidence of tubal or peritoneal adhesions), uterine

factors, systemic conditions (including infections or chronic diseases such as

obesity, autoimmune conditions, or chronic renal failure), cervical and immunologic

factors, and unexplained factors (including endometriosis with no evidence of tubal

or peritoneal adhesions).

2 Basic investigations that should be performed before starting any infertility treatment

are semen analysis, confirmation of ovulation, and documentation of tubal patency.

Ovarian reserve screening is often performed.

3 Male factor is the only cause of infertility in 20% of infertile couples, but it may be a

contributing factor in as many as 50% of cases. Within this category are same sex

female couples desiring to conceive; they would not necessarily be considered

traditionally infertile, but they would need donor sperm. Treatment of reversible

anatomic, endocrine or infectious causes of subfertility, such as varicocele repair,

cessation of suppressive medications or toxic behaviors, and treatment of sexually

transmitted diseases and thyroid disorders, may be efficacious. Intrauterine

insemination (IUI) is the best studied and most widely practiced of all the

insemination techniques. Intracytoplasmic sperm injection (ICSI) has allowed

couples with male factor infertility to achieve assisted reproductive technology

(ART) pregnancy outcomes that are comparable with those of couples with nonmale

factor infertility using conventional in vitro fertilization (IVF) treatment.

4 An association between the age of the woman and reduced fertility has been welldocumented. The decline in fecundability begins in the early 30s and accelerates

during the late 30s and early 40s. Chemotherapeutics may hasten loss of ovarian

reserve and may prompt fertility preservation counseling. Similar counseling should

be undertaken for male cancer patients.

5 Disorders of ovulation account for about 30% to 40% of all cases of female

infertility. These disorders are generally among the most easily diagnosed and most

treatable causes of infertility.

6 The most common cause of oligo-ovulation and anovulation—both in the general

population and among women presenting with infertility—is polycystic ovarian

syndrome (PCOS).

7 Tubal and peritoneal factors account for 30% to 40% of cases of female infertility.

2218Cervical factor is estimated to be a cause of infertility in no more than 5% of

infertile couples. Uterine pathologies contribute to infertility in as many as 15% of

couples seeking treatment. Leiomyomas have not been shown to be a direct cause of

infertility. If the evaluation is negative (i.e., unexplained) then superovulation with

IUI may be considered, though with less success than ART.

8 ART techniques include gametes from autologous or third-party sources, IVF, ICSI

using ejaculated or surgically retrieved sperm that is either fresh or cryopreserved,

gamete intrafallopian transfer (GIFT), zygote intrafallopian transfer (ZIFT), oocyte

and embryo cryopreservation, and use of gestational carriers. Because of improved

success rates associated with IVF embryo transfer, the performance of GIFT and

ZIFT has declined dramatically. Embryo screening using morphokinetic methods,

PGD, and PGS are also utilized.

9 Multiple gestation, especially higher-order multiple gestation, is a serious

complication of infertility treatment and has tremendous medical, psychological,

social, and financial implications.

10 Other complications of ART can occur, but stress is the most common reason for

patients to discontinue fertility therapy. Improved protocols can minimize ovarian

hyperstimulation syndrome (OHSS) risks, and recent studies have not shown an

increased risk for breast, uterine, or ovarian cancer secondary to medications used

for superovulation in the treatment of infertility. Although concerns have been raised

about congenital abnormalities in offspring, absolute risk seems to be low.

11 Information on the Society for Assisted Reproductive Technology (SART) and

registered ART clinics are accessible by the public at

https://www.sartcorsonline.com/rptCSR_PublicMultYear.aspx?reportingYear=2015.

The Centers for Disease Control (CDC) by law collects ART data and this is

publicly available at ftp://ftp.cdc.gov/pub/Publications/art/ART-2015-Clinic-ReportFull.pdf#page=4.

Assessing the scope of infertility is hampered in part by the lack of

standardized definitions (1). Infertility is variably defined as no pregnancy

with unprotected intercourse for 1 year (2) or 2 years (1). This condition may

be further classified as primary infertility, in which no previous pregnancies

have occurred, and secondary infertility, in which a prior pregnancy,

although not necessarily a live birth, has occurred (3). About 90% of couples

should conceive within 12 months of unprotected intercourse (4). The term

subfertility is often used interchangeably with infertility (5). Fecundability

refers variously to the probability of pregnancy per cycle or to the number of

cycles to achieve a pregnancy (6). Fecundity refers to the probability of

achieving a live birth (7). The diagnosis of impaired fecundity has been

proposed to include couples with ê36 months without conception or physical

inability or difficulty in having a child (7).

2219EPIDEMIOLOGY

The prevalence of infertility in the United States has been variously

estimated from 7.4% (8) to 15.5% (9) depending on the methodologic

approach used. However, utilization of infertility services in the United States

remains low at 17% or 6.9 million women aged 25 to 44 from 2006 to 2010 with

higher utilization noted among women age ≥30, non-Hispanic Whites, and higher

socioeconomic strata (10).

INITIAL ASSESSMENT

The physician’s initial encounter with an infertile couple is the most important

one because it sets the tone for subsequent evaluation and treatment. It cannot be

overemphasized that in partnered patients, infertility is a problem of the

couple. The presence of the partner, beginning with the initial evaluation,

involves him or her in the therapeutic process. [1] This essential involvement

demonstrates the physician’s receptivity to the partner’s needs and those of the

patient, allowing both members of the couple an opportunity to ask questions and

voice any concerns.

[2] The physician should obtain a complete medical, surgical, and

urologic/gynecologic history from both partners. Specifically, information

regarding menstrual cycle regularity, pelvic pain, and previous pregnancy

outcomes is important. Risk factors for infertility, such as a history of pelvic

inflammatory disease (PID) or pelvic surgery, should be reviewed. A history of

intrauterine exposure to diethylstilbestrol (DES) is significant. In addition, a

review of systems relevant to pituitary, adrenal, and thyroid function is useful.

Questions regarding galactorrhea, hirsutism, and changes in weight are

particularly relevant. A directed history, including developmental defects such as

undescended testes, past genital surgery, infections (including mumps orchitis),

previous genital trauma, and medications should be obtained from the male

partner. A history of occupational exposures that might affect the reproductive

function of either partner is important, as is information about coital frequency,

dyspareunia, and sexual dysfunction. Finally, information should be obtained on

any family history of infertility, premature ovarian failure, congenital physical or

intellectual impairments, and hereditary conditions relevant to preconception

planning, such as cystic fibrosis, thalassemia, and Tay–Sachs disease.

[1] The initial interview provides the physician with the opportunity to assess the

emotional impact of infertility on the couple. It gives the physician a chance to

emphasize the emotional support available to the couple as they proceed with the

diagnostic evaluation and suggested treatments. In most cases, referral to a trained

social worker or psychologist may be beneficial.

2220The physical examination of the woman should be thorough, with

attention given to height, weight, body habitus, hair distribution, thyroid

gland, and pelvic examination. Referral of the male partner to a urologist for

examination often is beneficial if historic information or subsequent evaluation

suggests an abnormality. This initial encounter provides an opportunity to outline

the general causes of infertility and discuss subsequent diagnostic and treatment

plans (Figs. 30-1 to 30-3).

[3] The basic investigations that should be performed before starting any

infertility treatment are semen analysis, confirmation of ovulation, and

documentation of tubal patency (Figs. 36-1 to 36-3). If a patient with severe

systemic illness such as renal failure, liver failure, or cancer wishes to conceive,

careful preconception assessment and counseling is advisable because the risks of

fertility treatment and pregnancy can be substantial.

CAUSES OF INFERTILITY

The main causes of infertility include:

1. Male factor

2. Decreased ovarian reserve

3. Ovulatory factor

4. Tubal factor

5. Uterine factor

6. Pelvic factor

7. Unexplained

The relative prevalence of the different causes of infertility varies widely

among patient populations (Table 36-1). Worldwide, male factor accounts for

51.2% (11) and tubal blockage for 25% to 35% (12) of infertility and subfertility

(conception after attempting for 1 year). In Europe, ovulatory dysfunction

accounts for 21% to 32%, male factor 19% to 57%, tubal factor 14% to 26%,

unexplained 8% to 30%, endometriosis 4% to 6%, and combined male and female

factors 34.4% of infertility (13–15). [4] Male factor is the only cause of

infertility in about 20% of infertile couples, but it may be a contributing

factor in as many as 50% of cases (15–17). Very few couples have absolute

infertility, which can result from congenital or acquired irreversible loss of

functional gametes in either partner or the absence of reproductive structures in

either partner. [2] Spontaneous conception is less likely to occur with age >42,

infertility duration >4 years, severe endometriosis, or severe tubal disorder (18),

so patients with those features should be strongly encouraged to consider

2221treatment rather than expectant management.

FIGURE 36-1 Diagnostic and treatment algorithm: Infertility. HSG,

2222FIGURE 36-2 Diagnostic and treatment algorithm: Anovulation. FSH, folliclestimulating hormone; LH, luteinizing hormone; E2, estradiol; TSH, thyroid-stimulating

hormone; T4, thyroxine; GH, growth hormone; ACTH, adrenocorticotropic hormone;

BMI, body mass index; MRI, magnetic resonance imaging; GnRH, gonadotropin-releasing

2223FIGURE 36-3 Diagnostic and treatment algorithm: Ovarian disorders. FSH, folliclestimulating hormone; LH, luteinizing hormone; CCCT, clomiphene citrate challenge test;

ART, assisted reproductive technology. (From Yao M. Clinical management of infertility.

Impact of Lifestyle on Fertility

[2] Women with class II obesity particularly when partnered with similarly obese

men experience longer time to pregnancy compared to normal-weight couples,

and female obesity is associated with lower pregnancy rates with IVF compared

to normal-weight women (21–23). Potential etiologies for obesity-associated

infertility include disrupted hypothalamic–pituitary–ovary axis, diminished

fertilization rate, and impaired endometrial gene receptivity. In mildly obese body

2224mass index (BMI) <35 kg/m2 women randomized to weight loss of ∼ 20 pounds

versus no weight loss, IVF live birth rates did not significantly differ, but

spontaneously conceived live birth rates were higher in the weight-loss group

(24). Obese men have higher rates of hypogonadotropic hypogonadism and sperm

DNA or mitochondrial damage compared to normal-weight men (25,26).

Substance (drug and alcohol) use in men is discussed in “Male Factor.” A survey

of 12,800 women in the United States undergoing IVF revealed that nearly a

quarter of them consumed alcohol, 7% smoked, and two-thirds used caffeine

during treatment (27). Studies report that compared to nonsmokers, maternal and

paternal tobacco smokers have lower live birth rates with intercourse (28), lower

pregnancy rates using donor IUI (29) and partner IUI (30), and higher rates of

IVF cycle cancellation prior to oocyte retrieval even among female former

smokers (31). Reassuringly, maternal low to moderate daily intake of alcohol

(<12 g) and caffeine (<200 mg) in the 1 year prior to ART did not impact ART

outcomes (32).

[2] TABLE 36-1 Causes of Infertility

Relative Prevalence of the Etiologies of Infertility (%)

Male factor 17–28

Both male and female factors 8–39

Female factor 33–40

Unexplained infertility 8–28

Approximate Prevalence of the Causes of Infertility in the Female (%)

Ovulatory dysfunction 21–36

Tubal or peritoneal factor 16–28

Miscellaneous causes 9–12

Prevalence of etiologies of infertility (1,15,19,20)

Male Factor

The concept of a global decline in sperm counts is controversial (33,34). A

decline in sperm density has been observed in the United States, Europe, and

Australia, while decreased motility and semen volume have been reported in India

2225(33). Given that decreases in sperm parameters have been noted in fertile men

(35) and in U.S. sperm donors (36), the clinical relevance for fecundability is

unknown. However, one simulation model has suggested that if sperm

concentrations decline by 21% to 47%, fecundability would decrease by 7% to

15% (37).

Physiology

Spermatogenesis

The male reproductive tract consists of the testis, epididymis, vas deferens,

prostate, seminal vesicles, ejaculatory duct, bulbourethral glands, and urethra.

Gonadotropin-responsive cells in the testes include Leydig cells (the site of

androgen synthesis) and Sertoli cells, which line the seminiferous tubules (the

site of spermatogenesis) (38). The [1] pituitary gland secretes luteinizing

hormone (LH), which stimulates the synthesis and secretion of testosterone

by the Leydig cells, and follicle-stimulating hormone (FSH), which acts with

testosterone on the Sertoli cells to stimulate spermatogenesis (38). In [1]

humans, a new cohort of spermatogonia enter the maturation process every 16

days, and the development from spermatogonia stem cells to the mature sperm

cells takes about 75 days (39). Spermatogonia undergo mitotic division to give

rise to spermatocytes (38). These diploid spermatocytes subsequently undergo

meiosis to produce haploid spermatids, which contain 23 (rather than 46)

chromosomes (38). Maturation of spermatids is called spermiogenesis and

involves condensation of the nucleus, formation of the flagellum, and the

formation of the acrosome (a structure derived from the Golgi complex covering

the tip or head of the sperm nucleus) (40). The resultant spermatozoa are released

into the seminiferous tubule lumen and enter the epididymis, where they continue

to mature and become progressively more motile during the 2 to 6 days that is

required to traverse this tortuous structure and reach the vas deferens (41).

Sperm Transport

[1] During ejaculation, mature spermatozoa are released from the vas

deferens along with fluid from the prostate, seminal vesicles, and

bulbourethral glands (42). The released semen is a gelatinous mixture of

spermatozoa and seminal plasma; however, this thins out 20 to 30 minutes after

ejaculation. This process, called liquefaction, is the direct result of proteolytic

enzymes within the prostatic fluid (42). [1] Following ejaculation, the released

spermatozoa must undergo capacitation to become competent to fertilize the

oocyte (43). Capacitation occurs within the cervical mucus and involves removal

of inhibitory mediators such as cholesterol from the sperm surface, tyrosine

phosphorylation, and calcium ion influx, all of which allow the sperm to

2226recognize additional fertilization cues during travel through the female

reproductive tract (43). After the sperm reach the tubal isthmus they are slowly

released into the ampulla, further reducing the number of sperm that reach the

oocyte (43). Sperm transport from the posterior vaginal fornix to the fallopian

tubes occurs within 2 minutes during the follicular phase of the menstrual cycle

(44).

Fertilization

[1] As the capacitated sperm near and pass through cumulus cells

surrounding the oocyte, hydrolytic enzymes are released from the acrosome

via exocytosis in a process called the acrosome reaction (43). Capacitation and

the acrosome reaction can be induced in vitro (40,43). Following the acrosome

reaction, the sperm binds to and penetrates the zona pellucida (the extracellular

coat surrounding the oocyte). This allows fusion of the sperm with the plasma

membrane of the oocyte, an event that promotes changes in the oocyte that

prevent entry by additional sperm (43). As the first sperm penetrates, cortical

granules are released (the cortical reaction) from the oocyte into the

perivitelline space. This stops the oocyte’s zona pellucida from binding new

sperm and inhibits penetration by previously bound sperm, further reducing

the possibility of polyspermy (45).

Recreational, Iatrogenic, and Environmental Male Reproductive Toxins

Decreased sperm concentration and motility have been noted in areas of the

United States with heavy agriculture and pesticide use (35), but occupational

exposures have not been linked to infertility (46). [4] Alcoholism negatively

affects all semen analysis parameters (47,48) and either smoked (48,49) or

chewed tobacco (50) is associated with decreased sperm density and motility.

Compared to nonusers, marijuana used greater than once per week was associated

with diminished sperm concentration and count by nearly 30% (51). Concurrent

use of other illicit substances such as cocaine further decreased sperm quantity

and negatively impacted motility (51). Certain medications may reduce sperm

numbers or function or may cause ejaculatory dysfunction (Table 36-2). Vaginal

lubricants such as Astroglide, KY Jelly, saliva, and olive oil inhibit sperm

motility in vitro, while no adverse effects are seen with hydroxethyl cellulose

(Pre-Seed), mineral oil, or canola oil (44). [4] However, Astroglide and KY do

not appear to affect fecundability in noninfertile couples engaging in procreative

intercourse (52).

[1] TABLE 36-2 Drugs that Can Impair Male Fertility

Impaired spermatogenesis—sulfasalazine, methotrexate, nitrofurantoin, colchicine,

2227chemotherapy

Pituitary suppression—testosterone injections, gonadotrophin-releasing hormone

analogs

Antiandrogenic effects—cimetidine, spironolactone

Ejaculation failure—a blockers, antidepressants, phenothiazines

Erectile dysfunction—b blockers, thiazide diuretics, metoclopramide

Drugs of misuse—anabolic steroids, cannabis, heroin, cocaine

From Hirsh A. Male infertility. BMJ 2003;327:669–672, with permission.

Semen Analysis

[3] The basic semen analysis measures semen volume, sperm concentration,

sperm motility, and sperm morphology (42). Recently revised, the normal

values suggested by the World Health Organization (WHO) in 2010 are listed

with the previously published guidelines in Table 36-3 (42,53). Both criteria were

developed using fertile men whose semen parameters were in the lowest 5th

centile of the group studied, but values above the reference ranges do not

guarantee male fertility. Because infertile men were not used to develop the

criteria, values below the cutoffs may not necessarily indicate infertility (42).

However, significant deviations from the reference limits are generally classified

as male factor infertility (17). Given regional differences between laboratories and

in semen quality, laboratories are encouraged to develop their own reference

ranges (42). Typically, semen is assessed manually, but computer-aided sperm

analysis (CASA) may be used (42). Limitations of CASA include a lack of

standardization among instruments, an inability to differentiate intact from

nonintact sperm, possible bias from artifacts during preparation, and a paucity of

studies on fertility outcomes in large populations (42).

Abstinence

[3] Previous recommendations for prolonged abstinence time up to 7 days

(42) have little supporting evidence, and likely there is benefit to shorter

abstinence duration regardless of baseline parameters. [1] The epididymis

stores the equivalent of three ejaculations (41), and therefore short abstinence is

unlikely to significantly deplete sperm quantity in most men. A retrospective

United States study of 1939 oligozoospermic subfertile men found that >2 days

abstinence was associated with more samples with >20 M total motile counts and

higher ejaculate volume but no statistically significant impact on concentration,

2228motility, or morphology (54). With >5 days of abstinence, oligozoospermic men

had reduced total motile sperm count and normal morphology (55). Although

longer abstinence >2 days in normozoospermic men increased sperm

concentration and total count, it was globally detrimental with respect to lower

ejaculate volume, total and progressive motility, viability, and tail morphology

(54). With prolonged abstinence >10 days, even normozoospermic men will have

reduced total motile sperm count and reduced proportions of sperm with normal

morphology (55), as sperm overflow into the urethra and are flushed out into the

urine (42). Prolonged epidydimal storage with heightened oxidative stress

negatively affects quality in terms of elevated sperm DNA fragmentation (54),

and may explain the improved pregnancy rates associated with shortening

abstinence times to ≤2 days prior to specimen collection for intrauterine

insemination (IUI) (56).

[3] TABLE 36-3 Semen Analysis Terminology and Normal Values

Terminology (42)

Normozoospermia All semen parameters normal

Oligozoospermia Reduced sperm numbers

Mild to moderate: 5–20 million/mL

Severe: <5 million/mL

Asthenozoospermia Reduced sperm motility

Teratozoospermia Increased abnormal forms of sperm

Oligoasthenoteratozoospermia Sperm variables all subnormal

Azoospermia No sperm in semen

Aspermia (anejaculation) No ejaculate (ejaculation failure)

Leucocytospermia Increased white cells in semen

Necrozoospermia All sperm are nonviable or nonmotile

Normal Semen Analysis—World Health Organization (42)

1992 Guidelines 2010

Guidelines

Volume >2 mL ≥1.5 mL

2229Sperm concentration >20 million/mL ≥15

million/mL

Sperm motility >50% progressive or >25%

rapidly progressive

≥32%

progressive

Morphology (strict criteria) >15% normal forms ≥4% normal

forms

White blood cells <1 million/mL <1

million/mL

Immunobead or mixed

antiglobulin reaction test

<10% coated with antibodies <50%

Terminology from Hirsh A. Male infertility. BMJ 2003;327:669–672, with permission.

Specimen Collection

The specimen should be obtained by masturbation and collected in a clean

container kept at ambient temperature (42). [3] The patient should report any loss

of the specimen, particularly the first portion of the ejaculate, which contains the

highest sperm concentration (42). Collection may be performed either at home or

in a private room near the laboratory (42). The sample should be taken to the

laboratory within one-half to one hour of collection to prevent dehydration and

degradation (42). If masturbation into a container is not possible, condoms

specially designed for semen analysis should be used rather than latex

condoms which are toxic to sperm (42). Intercourse to collect the sample is

discouraged because of contamination risk (42). [3] Even when the specimen is

obtained under optimal circumstances, interpretation of the results of the semen

analysis is complicated by variability within the same individual and overlap in

infertile and fertile reference semen parameters (42). Semen parameters may vary

widely from one man to another and among men with proven fertility (42). In

many circumstances, several specimens are necessary to verify an

abnormality (42).

Sperm Volume and pH

The lower limit of normal semen volume is ê1.5 mL and the pH should be

ê7.2 (42). These parameters are affected mainly by the balance between the acidic

secretions of the prostate gland and the alkaline fluid from the seminal vesicles

(42). Low volume along with pH <7 suggests obstruction of the ejaculatory ducts

or absence of the vas deferens (42). [3] Difficulties with collection, retrograde

ejaculation, or androgen deficiency can contribute to low volume (42). Consider

further workup with urine voided immediately after ejaculation in patients with

2230documented low semen volume <1 mL, after verifying with the man that no spill

occurred during collection (17). Calculate the total sperm in the urine divided by

the (total sperm in urine plus total sperm in ejaculate) multiplied by 100. A value

>7.1% to 8.3% will be found in patients with partial retrograde ejaculation (57).

High volumes >5 mL suggest inflammation of the accessory glands (42).

Sperm Concentration

Sperm concentration or density is defined as the number of sperm per milliliter in

the total ejaculate (42). [3] The normal lower limit is ê 15 million/mL, revised

from ê 20 million/mL (42,53). Only intact sperm are counted in determining

sperm concentration (42). Fifteen percent to 20% of infertile men are azoospermic

(no sperm) and 10% have a density of <1 million/mL (42,58).

Sperm Motility and Viability

Sperm motility is the percentage of progressively and total (progressive plus

nonprogressive) motile sperm in the ejaculate (42). [3] The normal lower limit is

ê 32%, which was revised from ê 50% (42,53). Viability should be at least

58% (42). Progressive motility refers to movement either linearly or in a large

circle regardless of speed. Nonprogressive motility describes sperm that display

only small movements/twitching or no movement at all (immotile) (42).

Assessments of speed of progression, either rapid or slow, have been removed

from the revised guidelines as a result of difficulty in unbiased measurement of

this parameter (42). A reduction in sperm motility is referred to as

asthenozoospermia (42). Leukocytes can impair sperm motility through oxidative

stress (42). When many immotile sperm are present or when progressive motility

is <40% viability studies should be performed (42). Viable immotile sperm may

have flagellar defects while the presence of nonviable immotile sperm

(necrozoospermia) suggests epidydimal pathology (42). [3] Viable sperm have

intact plasma membranes, which will not stain (dye exclusion) but will swell in

hypoosmotic solutions (hypoosmotic swelling test) (42). Sperm motility in-vitro

is decreased in the presence of women’s serum neutrophils, which release

extracellular “traps” that enmesh sperm in filaments elaborated by these

inflammatory cells.

Sperm Morphology

Morphology testing assesses anatomic malformations of the sperm. [3] The lower

limit for normal morphology is ê 4% using strict criteria, a change from

previous guidelines using a more lenient assessment and a cutoff of ê 30%

(42,53). Assessment of sperm morphology involves fixing and staining a portion

of the specimen (42). The strict Tygerberg criteria were introduced by Kruger et

al. in 1986 to assess sperm morphology (42,59,60). Using this system, the entire

2231spermatozoon—including the head, midpiece, and tail—is assessed, and even

mild abnormalities in head forms are classified as abnormal (42). Most sperm

from normal men exhibit minor abnormalities when subjected to Tygerberg

standards (42). An abnormality of sperm morphology is known as

teratozoospermia, and these sperm have poor fertilizing potential and may have

abnormal DNA (42). Low strict sperm morphology <9% was a better

discriminator for infertility than either count or motility in a large cross sectional

multicenter study (61). A disadvantage to any morphology assessment is that

reproducibility may be hampered by the subjective nature of the assessment (42).

Round Cells, Bacteria, and Leukocytes

[3] Round cells include immature germ cells and leukocytes (42). The

prevalence of round cells in a large IVF cohort was 5.4%, of which 73% were

found to be immature germ cells (62). Immature germ cell elevation is associated

with lower sperm concentration and normal morphology, compromised sperm

DNA integrity, lower fertilization, and higher pregnancy loss rates with

Intracytoplasmic sperm injection (ICSI) (62), while leukocytes (predominantly

neutrophils) are associated with inflammation (42). Leukocytes can be

distinguished by positive peroxidase staining, and normal leukocyte

concentrations should be <1 million/mL (42). However, the prognostic

significance of leukocytes in the semen is controversial; even low levels can be

associated with reactive oxygen species and sperm DNA damage, but their

presence has inconsistent effects on IVF/ICSI outcomes (42,62–64).

Bacteriospermia prevalence among infertile men is ∼ 50%, with detection rates

likely affected by semen collection technique and specific test parameters, and

prevalence is not necessarily associated with leukocytospermia; it may represent

asymptomatic colonization without an inflammatory response (62,65). The most

common pathogens associated with bacteriospermia are Chlamydia trachomatis

(41.4%), Ureaplasma urealyticum (15.5%), and Mycoplasma hominis (10.3%),

with M. hominis and U. urealyticum having higher prevalence in men with

leukocytospermia compared to those without (65).

Antisperm Antibodies

Antisperm antibodies are often associated with sperm agglutination, may be

present in seminal plasma, on sperm surfaces and in maternal serum, and risk

factors include history of ductal obstruction, prior genital infection, testicular

trauma, and prior vasectomy reversal (17). Using the immunobead test, washed

spermatozoa are exposed and assessed for binding to labeled beads (42). In the

mixed agglutination reaction, human red blood cells sensitized with human IgG

are mixed with partner’s semen (42). Spermatozoa that are coated with antibodies

2232form mixed agglutinates with the red blood cells (42).[3] Antisperm antibody

testing is limited by the lack of specific antibody binding, which is important

because only a subset of antibodies is thought to have cytotoxic or agglutinating

effects (66). Although the presence of antisperm antibodies has typically been

managed with ICSI, it is uncertain whether that treatment is beneficial (66), and

antibody testing would not be indicated if ICSI were a priori planned (17).

Postcoital Testing

[3] The purpose of the postcoital test, in which the cervical mucus is

microscopically examined shortly after intercourse during the periovulatory cycle

phase, is to assess the effect of cervical mucus on sperm viability and function

(2). The test is considered positive if one or more progressively moving sperm per

high power field is observed, negative if no sperm or only nonprogressive or

immotile sperm are seen (3). However, its prognostic value and impact on

management are limited by poor specificity, poor reproducibility, or controversial

interpretation of the results (2). Nevertheless, a large recent study of infertile

couples with <3 years of infertility from the Netherlands found that a positive

postcoital test was associated with improved spontaneous pregnancy rates

compared to negative test results, while a negative postcoital test was associated

with improved IUI and IVF pregnancy rates compared to positive test results (3).

Sperm DNA Integrity

[3] Sperm DNA fragmentation has been associated with impaired pregnancy rates

during IVF and is associated with an elevated risk of miscarriage, perhaps related

to paternal age with its associated increase in double strand DNA damage (67).

Sperm DNA fragmentation does not appear to correlate with blastocyst

aneuploidy in a small study (68). Sperm chromatin structure assays, terminal

deoxynucleotidyl transferase–mediated deoxyuridine triphosphate nick end

labeling (TUNEL) assays, sperm chromatin dispersion tests, and single gel

electrophoresis assays have varying thresholds for DNA damage diagnosis and

varying positive predictive values (PPVs) for pregnancy with assisted

reproduction; TUNEL assay results have the best correlation (67). These tests

reflect the aggregate sperm DNA integrity of a heterogeneous sperm population

rather than an individual fertilizing sperm, which limits the utility of this testing

in the context of fertility treatment (67). [3] Nonetheless it may add etiologic

information particularly in otherwise unexplained cases, and may prompt

recommendations for reducing sperm DNA damage with interventions such as

antioxidant supplementation (67).

Epigenetic Sperm DNA Assessment

Epigenetics is the next frontier in sperm functional biology assessment, and

2233epigenetic changes may be acquired transgenerationally or from conditions such

as obesity (69). In couples with unexplained infertility, poor embryo development

during IVF and men with abnormal semen analysis parameters, the sperm exhibit

genome-wide alterations in genome packaging (i.e., histone modifications) and

DNA methylation using chip microarrays as compared to those from fertile men

(70,71). Assessment is available with the Episona test, but similar limitations as

with the sperm integrity assays exist, including lack of applicability to the

individual fertilizing sperm, and uncertainty as to treatments. In the case of

obesity, weight loss may ameliorate epigenetic sperm changes through

remodeling (69).

Differential Diagnosis of Male Factor

An underlying etiology for male factor was found using diagnostics in two-thirds

of infertile men (72). Table 36-4 lists the differential diagnoses for male factor

infertility (73). Several groups have attempted to assess the distribution of male

infertility diagnoses; two such distributions are shown in Table 36-5 (74,75). The

first is the result of a WHO study of 7,057 men with complete diagnoses based on

the WHO standard investigation of the infertile couple (74). The figures include

data from cases in which the male partner was normal and the presumed cause of

the couple’s infertility was a female factor. The second distribution is the result of

a study of 425 subfertile male patients (74). Although the two studies represent

different populations (one is from a study of couples, the other from a urologic

practice) and differ in their distribution of male infertility diagnoses, idiopathic

male factor and varicocele predominate. Other anatomic and endocrine causes

occur less frequently. Particularly among men with oligozoospermia, 75% of

cases were idiopathic (16). Factors which definitively lead to male infertility

include secondary hypogonadism, genetic causes, seminal tract obstruction,

oncologic disease therapies, and severe sexual dysfunction, while plausible

causative factors include congenital urogenital anomalies and acquired testicular

damage (unilateral cryptorchidism, testis cancer, and orchitis) (16). [2] Although

not considered as primary etiologies for male factor infertility, varicocele,

leukocytospermia, obesity, and uncontrolled chronic disease are more prevalent in

infertile compared to fertile men (16). Same sex female couples or single women

without a male partner who desire pregnancy may be considered as part of this

category.

Male Age

[2] Men have reportedly fathered children into their 90s, but pregnancy rates

are decreased with paternal age ê 40 to 45 and particularly ê age 50 (76).

Increasing paternal age is associated with a higher frequency of disomic sex

2234chromosomes and structural chromosomal abnormalities in the sperm (76). With

respect to the offspring, paternal age confers higher rates of autosomal dominant

diseases such as achondroplasia and craniosynostosis and somewhat higher rates

of trisomy 21 (76). In mice, offspring of older fathers have decreased survival

(77). Increased paternal age is associated with recurrent pregnancy loss (78).

Psychiatric conditions such as schizophrenia have been associated with increased

paternal age, but this may reflect trends of men with psychiatric conditions to

delay fathering children until later ages rather than de-novo genetic mutations

(79).

[2] TABLE 36-4 Etiologic Factors in Male Infertility

Pretesticular Testicular

Endocrine Genetic

Hypogonadotropic hypogonadism Klinefelter syndrome

Coital disorders Y chromosome deletions

Erectile dysfunction Immotile cilia syndrome

Psychosexual Congenital

Endocrine, neural, or vascular Cryptorchidism

Ejaculatory failure Infective (orchitis)

Psychosexual Antispermatogenic agents

After genitourinary surgery Heat

Neural Chemotherapy

Drug related Drugs (especially testosterone)

Posttesticular Irradiation

Obstructive Vascular

Epididymal Torsion

Congenital Varicocele

Infective Immunologic

2235Vasal Idiopathic

Genetic: cystic fibrosis

Acquired: vasectomy

Epididymal hostility

Epididymal asthenozoospermia

Accessory gland infection

Immunologic

Idiopathic

Postvasectomy

From De Kretser DM. Male infertility. Lancet 1997;349:787–790, with permission.

Treatment of Male Factor Not due to Azoospermia

Medical treatment of reversible infectious or endocrine causes of male

subfertility, such as sexually transmitted diseases and thyroid disorders, tends to

be efficacious (80). Although injections of exogenous FSH have been reported to

improve pregnancy rates, the benefit is less clear for clomiphene citrate, an

estrogen antagonist at the hypothalamus and pituitary that promotes gonadotropin

release (80). [4] Exogenous testosterone is not recommended in the treatment

of male subfertility because of negative feedback inhibition at the pituitary

that results in reduced intratesticular testosterone and decreased

spermatogenesis (81). The prevalence of men receiving prescribed testosterone

replacement therapy is 7% to 9%, of whom one-third have normal or

supraphysiologic testosterone levels prior to treatment and half do not receive

appropriate follow-up (81,82). Antioxidant food supplements have been evaluated

(80). [4] Glutathione, carnitine, and vitamin E do not appear to affect semen

parameters, but administration of zinc and folic acid were associated with

improved sperm concentration and morphology (83). The diagnosis and treatment

of azoospermia (no sperm on semen analysis) will be discussed separately from

other forms of male factor infertility.

Varicocele Repair

A varicocele is an abnormal dilation of the veins within the spermatic cord

(16). Clinically palpable varicoceles are present in 13.5% of men who recently

fathered a pregnancy and 31% of men with otherwise idiopathic oligozoospermia

2236(16). The pathophysiologic effects of a varicocele appear to be mediated by an

associated rise in testicular temperature, reflux of toxic metabolites from the left

adrenal or renal veins, or higher reactive oxygen species (84,85). [4] While

treatment is associated with improved semen parameters in some studies, it

is not clear whether varicocele repair improves fertility (84,86,87). Treatment

via surgery or percutaneous embolization is typically considered if the varicocele

is palpable and the semen analysis is abnormal, and may allow some couples to

use IUI instead of in vitro fertilization (IVF) (88) and improve IVF outcomes in

the setting of oligozoo or azoospermia (89). Complications of treatment include

infection, varicocele persistence, recurrence, and hydrocele formation (84).

[4] TABLE 36-5 Frequency of Some Etiologies in Male Factor Infertility

Cause Percentage Cause Percentage

No demonstrable cause 48.5 Varicocele 37.4

Idiopathic abnormal semen 26.4 Idiopathic 25.4

Varicocele 12.3 Testicular failure 9.4

Infectious factors 6.6 Obstruction 6.1

Immunologic factors 3.1 Cryptorchidism 6.1

Other acquired factors 2.6 Low semen volume 4.7

Congenital factors 2.1 Semen agglutination 3.1

Sexual factors 1.7 Semen viscosity 1.9

Endocrine disturbances 0.6 Other 5.9

Total 103.9a 100

aMore than 100% because of multiple factors.

From The ESHRE CAPRI Workshop Group. Male sterility and subfertility: guidelines

for management. Hum Reprod 1994;9:1260–1264; and Burkman LJ, Cobbington CC,

Franken DR, et al. The hemizona assay (HZA): Development of a diagnostic test for the

binding of human spermatozoa to the human hemizona pellucida to predict fertilization

potential. Fertil Steril 1988;49:688–697, with permission.

Artificial Insemination

Artificial insemination has mainly been used to treat unexplained infertility

2237(usually combined with superovulation) and male factor infertility (including

same sex female couples). [4] All artificial insemination procedures involve

the placement of whole semen or processed sperm into the female

reproductive tract, which permits sperm–ovum interaction in the absence of

intercourse. The placement of whole semen into the vagina as a mode of fertility

treatment is now rarely performed except in cases of severe coital dysfunction.

All the common forms of artificial insemination involve processed sperm

obtained from the male partner or a donor. Many techniques for artificial

insemination have been described, but only intracervical (ICI) and IUI have been

routinely employed.

Insemination Processing

During and after intercourse, seminal fluid usually is prevented from reaching the

intrauterine cavity and intra-abdominal space by the cervical barrier. [4] The

introduction of seminal fluid past this barrier may be associated with pelvic

infection and severe uterine cramping or anaphylactoid reactions, possibly

mediated by seminal factors such as prostaglandins (90). Thus, protocols for

processing whole semen include the washing of specimens to remove seminal

factors and to isolate pure sperm (90). Additional processing may include

centrifugation through density gradients, sperm migration protocols, and

differential adherence procedures (90). Finally, phosphodiesterase inhibitors, such

as pentoxiphylline, have been used during semen processing to enhance sperm

motility, fertilization capacity, and acrosome reactivity for IVF procedures

(83,91).

Intrauterine Insemination

IUI is the best studied and most widely practiced of all the insemination

techniques. It involves placement of approximately 0.3 to 0.5 mL of washed,

processed, and concentrated sperm into the intrauterine cavity by

transcervical catheterization (90). [4] Patients should remain immobile for

approximately 15 minutes following the procedure (92). It is difficult to identify

an optimal protocol or definitively determine pregnancy rates with IUI because of

patient and protocol heterogeneity, with multifactorial influences on pregnancy

rates including couple age, fertility diagnosis, stimulation type, and insemination

timing relative to ovulation (93). Ideally, the partner’s total motile sperm

count in the IUI specimen should be ê5 million (94–96) or ê10 million (97),

although a French study found similar pregnancy rates with all values ê1

million (98). Pregnancy rates with semen meeting those thresholds have been

reported to be 10.5% per cycle (97) and 38% after 4 to 6 cycles (96). Delivery

rates (17% vs. 8%) are higher when ≥2 follicles ≥15 mm are present as compared

2238to 1 follicle, particularly when GnRH antagonists are employed (98). Although

teratozoospermia using older criteria has been associated with IUI failure (94),

low strict normal morphology of 2% to 4% has not been found to be predictive of

IUI failure as compared to morphology >4% in two recent retrospective studies

(99,100), but effects of more severe teratozoospermia on IUI outcomes remain

nebulous. No benefit has been found with respect to double IUI versus single IUI

during a single cycle (101). The technique of fallopian tube sperm perfusion

(FSP) has been proposed to improve IUI pregnancy rates, whereby the washed

sperm is instilled slowly over several minutes through an inflated balloon catheter

to prevent reflux, but has not been found in a large randomized trial (102) or in a

2013 Cochrane systematic review (103) to be associated with improved

pregnancy rates. Unexplained infertility and IUI will be discussed separately later

in this chapter.

Timing of IUI

IUI typically is performed 24 hours following the LH surge (104). However,

given data that ovulation can occur much later than this, it is not surprising that no

differences in pregnancy rates were found when IUI was performed between 24

and 60 hours following the LH surge (105). IUI pregnancy rates doubled when

human chorionic gonadotropin (hCG) trigger was used instead of natural LH

surge (98). [4] Although timing of IUI is typically performed 36 hours following

hCG administration to coincide with follicle rupture, no significant differences in

pregnancy or live birth rates are seen whether IUI is performed at 24 or 36 hours

following the hCG trigger (106).

Donor IUI

For men with azoospermia, couples with significant male factor infertility who do

not desire ART, and women without a male partner who are seeking pregnancy,

therapeutic donor insemination offers an effective option (107). In earlier

prospective randomized or crossover trials, IUI was shown to be superior to ICI

for donor insemination (108). The success rates with ICI are lower than those

with IUI, particularly when using frozen sperm (90,108–110). Women receiving

donor sperm because of same sex partnership or being single had natural cycle

IUI ongoing pregnancy rates that were similar to ICI at 10% per cycle and 40%

cumulative over six treatment cycles (111). Lesbian women who achieved a live

birth required a (mean ± SD) 3 ± 1.1 donor IUI cycles to conceive (112). [4] In

patients younger than 30 years of age who have no other infertility factors,

delivery rates approach 90% after 12 cycles of IUI treatment with frozen sperm

(113), so patients who do not conceive within 6 to 12 months should be assessed

for female factors and encouraged to terminate treatment or proceed with

2239alternative forms of therapy. The concomitant use of clomiphene citrate or human

menopausal gonadotropin (hMG) for controlled ovarian hyperstimulation (COH)

did not result in higher fecundity rates in these patients (113). [4] Psychological

counseling should be offered because of the potential repercussions of using

donor gametes (107). Higher pregnancy rates have been demonstrated when

donor IUI is performed 1 day after natural cycle serum LH rise (20%) compared

to those randomized to 2 days postrise (11%) (114).

Donor Sperm Screening

Although the use of fresh donor semen is associated with higher pregnancy rates

than the use of frozen specimens (115,116), both the Centers for Disease Control

and Prevention (CDC) and the American Society for Reproductive Medicine

recommend the use of frozen samples (107). This recommendation stems from

the increasing incidence of human immunodeficiency virus (HIV) infection in

the general population and the lag time between HIV infection and

seroconversion (107). Semen donors are screened for HIV infection, hepatitis B,

hepatitis C, syphilis, gonorrhea, chlamydia, and cytomegalovirus infections, all of

which may be transmitted through semen (107). All cryopreserved samples are

quarantined for 6 months, and the donor is retested for HIV before clinical use of

the specimen (107). Donors are likewise questioned about any family history of

genetically transmitted disorders, including both mendelian (e.g., hemophilia,

Tay–Sachs disease, thalassemia, cystic fibrosis, congenital adrenal hyperplasia,

Huntington disease) and polygenic/multifactorial conditions (e.g., mental

retardation, diabetes, heart malformation, spina bifida) (107). [1] Those with a

positive family history of these conditions are eliminated as donor candidates

(107).

Intracytoplasmic Sperm Injection

ICSI in the United States has been utilized for 93% of ART cycles for male factor

infertility to achieve pregnancy outcomes that are comparable with those of

couples with nonmale factor infertility using conventional IVF treatment (117).

[9] The technique involves direct injection of a live sperm into the oocyte, thereby

theoretically bypassing limitations imposed by sperm motility and defects in

capacitation, the acrosome reaction, and/or sperm binding the zona pellucida

(117). This microsurgical procedure involves stripping the aspirated cumulus

complex of all surrounding granulosa cells, inserting a single viable sperm into

the cytoplasm (ooplasm) of the oocyte that optimally is metaphase II (117) though

less mature oocytes may still be considered for injection (118). ICSI should be

offered if the semen analysis shows <2 million motile sperm, <5% motility, or if

surgically recovered sperm is used (58), though there is wide latitude for utility

2240depending on use of semen normal value thresholds (119). [9] Higher pregnancy

rates have been noted with fresh when compared to frozen specimens and with

ejaculated when compared to surgically retrieved sperm (58). Success rates are

affected by the age of the female partner and oocyte quality (58). [9]Nonmale

factor indications for ICSI are contentious (120) but are utilized in 66.9% of

United States ART cases (117) and with much higher frequency worldwide,

and include teratozoospermia (121), unexplained infertility, previous

fertilization failure with conventional IVF, diminished ovarian reserve,

fertility preservation, and the fertilization of oocytes before preimplantation

genetic diagnosis (PGD) (117).

Risks of ICSI

Oocyte degeneration following ICSI is a relatively common occurrence with rates

ranging from 5% to 19% (122). Prior reports have not supported the idea that

oocyte degeneration is a function of technician skill; though, it is suggested that

inherent oocyte quality, ovarian stimulation itself, unexpected random technical

problems, or mechanical removal of cumulus and corona cells may contribute to

this phenomenon (122). While accumulated data indicate that there is an

overall increased risk of congenital abnormalities with ART, ICSI does not

appear to be associated with a higher risk compared to conventional IVF

(123). [11] The prevalence of sex chromosome abnormalities and translocation

appears to be slightly higher with ICSI versus IVF; however, it is unclear whether

these relate to the ICSI procedure or inherent gamete defects, as men with

abnormal semen parameters have higher rates of sperm aneuploidy (124). There

may be an increased risk of imprinting disorders with ICSI (125). Although prior

studies have not found an association with impaired intellectual or motor

development among ICSI children, one study reported a higher incidence of

autism when ICSI was used compared with IVF (126). In the setting of specific

genetic abnormalities such as Y chromosome microdeletions, abnormal

karyotypes, cystic fibrosis mutations or congenital absence of the vas

deferens, genetic counseling should be offered to address the possible risk of

infertility or other abnormalities in the offspring (124).

Azoospermia: Classification and Treatment

[4] Azoospermia describes the absence of spermatozoa in the ejaculate and is

found in 1% of all men (127) and up to 15% to 20% of infertile men (58,127).

Causes are categorized into pretesticular (nonobstructive), testicular

(nonobstructive), and posttesticular (includes obstructive and nonobstructive)

etiologies, but in some cases the condition is idiopathic (58).

Pretesticular Azoospermia

2241Pretesticular azoospermia from organic causes is relatively rare and results

from gonadotropin deficiency, which leads to loss of spermatogenesis

(58,127,128). The physician should perform a full endocrine history that includes

information on puberty and growth and check for low serum levels of LH, FSH,

and testosterone (58,127,128). Prolactin levels and pituitary imaging are indicated

in cases of hypogonadotropic hypogonadism (58). [4] Hormonal treatment

includes pulsatile GnRH, hCG, and exogenous gonadotropins (58,127,128).

The best predictors of good response are postpubertal onset of gonadotropin

deficiency and testicular volume >8 mL (58). Iatrogenic azoospermia

commonly results from medical testosterone use. Treatment includes

discontinuation of exogenous testosterone and recovery therapy with medications

such as clomiphene citrate, hCG, and/or follicle stimulating hormone (FSH) with

frequent follow-up semen analyses and endocrine labs (81). Azoospermia

recovers to a median concentration of 6.5 million/mL 4.5 months following

cessation of testosterone use; full recovery of sperm quantity was more likely in

those men who used testosterone for less than 1 year, while those use who used it

for several years were unlikely to recover (81).

Testicular Azoospermia

Gonadal failure is the hallmark of testicular azoospermia. Causes of this

condition may be genetic, acquired (e.g., radiation therapy, chemotherapy,

testicular torsion, varicocele, or mumps orchitis), or developmental (e.g.,

testicular maldescent) (58,127,128). Testicular atrophy is often present

(58,127,128). Because of the low chance of obtaining sperm, testicular biopsy is

generally not recommended with hypergonadotropic hypogonadism (elevated

levels of LH and FSH with low serum levels of testosterone) and consideration

should be given to using donor sperm (58,127,128). Diagnostic testicular biopsies

may be indicated in the setting of normal hormonal testing (58,127,128). If sperm

are present in diagnostic testing, consideration can be given to using surgically

retrieved spermatozoa for ICSI (58,127,128). If no sperm are present,

consideration can be given to correcting acquired conditions such as varicocele,

which may restore sperm to the ejaculate to permit ICSI or spontaneous

pregnancy (129,130). [4] Chromosomal abnormalities by peripheral

karyotype testing are present in about 7% of infertile, 5% of oligospermic,

and 10% to 15% of azoospermic men (131). Sex chromosome aneuploidies

such as Klinefelter syndrome (47,XXY) encompass two-thirds of these infertilityassociated chromosome abnormalities (131). Microdeletions in the Y

chromosome have been identified in 10% to 20% of men with idiopathic

azoospermia or severe oligospermia with concentration <5 million/mL (Table 36-

6) (131). These microdeletions can be transmitted to the male offspring, who may

2242suffer from infertility (131). Therefore, screening for genetic causes is indicated

in nonacquired cases of testicular azoospermia so that genetic counseling can be

provided before treatment (124). The two most commonly implicated candidate

gene families are the RNA-binding motif (RBM) and the “deleted in azoospermia”

(DAZ) families, but microdeletions at various loci on the Y chromosome have

been described (131,132). For example, microdeletions in Yq11.23 can occur in

one or more of three regions: AZFa (proximal), AZFb (central), and AZFc (distal)

(131).

[4] TABLE 36-6 Genetics and Male Infertility

Clinical Diagnosis Genetic Tests Most Common

Defects

Incidence

(%)

Congenital bilateral absence of

vas deferens (CBAVD)

Cystic fibrosis

(CFTR gene)

ΔF508, R117H 66

Nonobstructive azoospermia Karyotype 47, XXY AZFa,

AZFba, AZFc

15–30

10–15

Severe (<5 M/mL)

oligozoospermia

Karyotype

Translocation

Y chromosome

microdeletions

47, XXY

Partial AZFb,

AZFc

1–2

0.2–0.4

7–10

aAZFb the most severe (DAZ gene—deleted in azoospermia) causes the most severe

defects of spermatogenesis; AZFc causes the mildest defects of spermatogenesis

CFTR, cystic fibrosis transmembrane conductance regulator; AZF, azoospermia factor.

From Hirsh A. Male infertility. BMJ 2003;327:669–672, with permission.

Posttesticular Azoospermia

Posttesticular or obstructive etiologies are associated with normal gonadotropin

and testosterone levels and are present in up to 40% of azoospermic men

(58,127). Ejaculatory dysfunction is associated with oligospermia or aspermia but

rarely azoospermia (127). [4] Obstructive causes include congenital absence or

obstruction of the vas deferens or ejaculatory ducts, acquired obstruction of

these ducts, or ductal dysfunction, including retrograde ejaculation (58,127).

In the congenital bilateral absence of the vas deferens (CBAVD) or

hypogonadism, men with low-volume ejaculate should have a postejaculatory

urinalysis to check for retrograde ejaculation, which is associated with diabetics

and surgery to the bladder or prostate (58,128). Sperm may be isolated from the

neutralized urine of men with retrograde ejaculation and processed for

2243insemination or for ART (42). Transrectal ultrasound (US) may be of use to

diagnose ejaculatory duct obstruction or unilateral vasal agenesis (to demonstrate

contralateral atresia) but is generally not needed for CBAVD (127). [3] Renal

imaging is necessary when either unilateral or bilateral vasal absence is

diagnosed as a result of the 10% to 25% incidence of renal agenesis (127).

Most men with CBAVD will also have seminal vesicle agenesis, so they will

almost all have low semen volume, pH, and low fructose (127). Spermatogenesis

can be expected to be normal in CBAVD, so diagnostic biopsy is generally not

indicated (133). In some cases, testicular biopsy may be indicated to differentiate

between testicular and posttesticular causes (127). At least two-thirds of men with

CBAVD have mutations of the cystic fibrosis transmembrane conductance

regulator (CFTR)gene (127). [3] However, because many CFTR mutations are

undetectable, CBAVD patients should be assumed to have a mutation and

thus testing of the female partner’s carrier status should be performed (127).

Vasectomy Reversal and Treatment of Obstructive Azoospermia

Vasectomy can be reversed effectively using microsurgical vasovasostomy or

vasoepididymostomy. The latter technique can be used for epidydimal

obstructions (134). Patency and subsequent pregnancy rates after vasectomy

reversal approach are 100% and 80% respectively (135). Pregnancy typically

occurs within 24 months of reversal (135). [4] Rates of patency and pregnancy

vary inversely with the length of time from vasectomy, particularly for

reversal procedures performed after ê15 years (134,135). Although 60% of

reversal patients develop antisperm antibodies, these do not appear to affect

fecundability (135). Following surgery, periodic semen analyses can identify

reobstruction, which can range from 3% to 21% depending on which segments

were anastomosed (135). For those patients with azoospermia 6 months after

reversal, the procedure is considered a failure and testicular sperm aspiration with

ICSI could be considered (135). However, repeat vasovasostomy is associated

with patency rates of 75% and pregnancy rates of 43% (135).

Surgical Sperm Recovery for Intracytoplasmic Sperm Injection

Among the many surgical methods for sperm recovery, the most widely described

are microsurgical epidydimal sperm aspiration (MESA), percutaneous epidydimal

sperm aspiration (PESA), testicular sperm extraction (TESE), and percutaneous

testicular sperm fine-needle aspiration (TESA, also called fine needle aspiration,

or FNA) (133). Both MESA and TESE are open surgical procedures performed

with an operating microscope and general or regional anesthesia, whereas the

percutaneous procedures need only local anesthesia (133). The optimal method

for surgical sperm recovery has not been determined and certainly varies based on

2244patient history (133). [9] Hematoma risk appears to be low regardless of method

(133). Testicular atrophy is a rare complication of TESE and TESA even when

biopsies are obtained from multiple testicular sites (133).

With obstructive azoospermia, pregnancy rates using sperm retrieval and ICSI

are 24% to 64% and outcomes using either frozen thawed or fresh sperm are

comparable (133). Because MESA allows for diagnosis and possible

reconstruction of ductal pathology and usually yields very large numbers of

sperm, sperm cryopreservation and avoidance of repeat surgery may be possible

(133). If repeat sperm retrievals are needed, the minimum interval between

procedures is 3 to 6 months to allow for adequate healing (134). Surgical sperm

retrieval in the setting of nonobstructive azoospermia is addressed in the

“Testicular Azoospermia” section.

Female Age and Decreased Ovarian Reserve

Decreased Fecundability and Diminished Ovarian Reserve

Most women will experience an age-related decline in fecundability that is

physiologic rather than pathologic. [9] This decline begins in the early 30s and

accelerates during the late 30s and early 40s, reflecting declines in oocyte

quantity and quality, commonly referred to as ovarian reserve (136).

Attribution of age-related decline of fecundability to ovarian reserve is supported

by research findings among women using a wide range of methods to conceive,

including natural conception, donor sperm through IUI for male factor infertility,

use of donor egg in IVF, and age-related declines in IVF success rates when using

a woman’s own eggs. Among populations that do not practice contraception,

fertility peaks at age 20, declines somewhat at age 32, steeply declines after the

age of 37, and is rare after age 45 (136). Female partners of azoospermic

husbands who received donor IUI had cumulative pregnancy rates over 12 cycles

of 74% (age <31 years), 62% (age 31 to 35), and 54% (age >35) (137).

Chronologic aging of the endometrium does not seem to play an appreciable role

in reduced fertility, given the excellent (∼ 40% to 60%, depending on whether

frozen oocytes were utilized) live birth rates using donor oocytes regardless of

recipient age (138,139). These examples of reproductive aging relate to the

decline of the stock of primordial follicles, established early in fetal life, to near

zero at menopause (140). [5] “Diminished ovarian reserve” has been variously

defined based on age, biochemical parameters, sonography, poor follicular

response to ovarian stimulation, and number of oocytes retrieved during IVF

(141). The prevalence of diminished ovarian reserve diagnoses in a U.S. IVF

cohort increased to 26% in 2011, but the diagnosis has not consistently predicted

IVF or pregnancy outcomes (141). Etiologies attributed to prematurely

2245diminished ovarian reserve include iatrogenic (i.e., cytotoxic agents, radiation,

and surgical removal of reproductive tissues), pelvic endometriosis or adhesions,

immunologic, and genetic factors. [5] Although premature ovarian

failure/menopause prior to age 40 has been associated with an increased number

of premutation-level trinucleotide repeat lengths in the FMR1 gene, a similar

increase has not been consistently confirmed in premenopausal diminished

ovarian reserve women (142,143).

Spontaneous Pregnancy Loss

[5] Reproductive aging is associated with abnormalities in the oocyte meiotic

spindles that lead to chromosome alignment errors and increase rates of

conceptus aneuploidies, particularly trisomies (136). This serves to increase

the risk for spontaneous pregnancy loss and thereby decrease live birth rates in

older women (136,144). A large study based on the Danish national registry

estimated the rates of clinically recognized spontaneous pregnancy loss for

various age groups to be 13.3% (12 to 19 years), 11.1% (20 to 24 years), 11.9%

(25 to 29 years), 15.0% (30 to 34 years), 24.6% (35 to 39 years), 51.0% (40 to 44

years), and 93.4% (>45 years) (145). Using sensitive hCG assays in women

during their reproductive years, 22% of all pregnancies were found to be lost

before they could be clinically diagnosed (146).

Ovarian Reserve

[1] Ovarian reserve refers to the size of the nongrowing, or resting,

primordial follicle population in the ovaries. This ovarian reserve, in turn,

presumably determines the number of growing follicles and the “quality” or

reproductive potential of their oocytes (144). Although ovarian reserve and

fecundability decrease with aging, they may not be consistently reflected by a

woman’s chronologic age because ovarian reserve does not decline at the same

pace at a given age for all women. The prognostic value of anti-mullerian

hormone (AMH) levels and antral follicle count (AFC) among IVF patients,

independent of age, has established them as standard ovarian reserve tests (147–

150).

Serum Anti-Mullerian Hormone

AMH is produced by the granulosa cells of preantral and small antral follicles

(151,152). The serum level of AMH in women with normal cycles declines with

age (151) and becomes undetectable after menopause (153). [3] Unlike other

serum markers, AMH can be measured at any time in the menstrual cycle

(153) and levels are best interpreted using age-specific reference ranges

(154). Evolving assay platforms from Diagnostic Systems Laboratories to

2246Beckman Coulter AMH Generation II (154), modifications to the generation II

assay protocol (155) and further use of automated platforms (156) limits

interpretation of serial AMH results. In women engaging in procreative

intercourse without a history of infertility, AMH level does not correlate with

monthly fecundability regardless of age at least in the short term (6 to 12 months)

(6,157). [5] Anti-mullerian hormone was found to be a good predictor of both

excessive (>3.5 ng/mL) and poor (<1 ng/mL) IVF stimulation response and is

strongly correlated to the AFC (152,153,158).

Antral Follicle Count

Using transvaginal US in the early follicular phase, all 2 to 10 mm ovarian

follicles are counted and the total for both ovaries is called the basal AFC (140).

The AFC correlates very well with chronologic age in normal fertile women and

appears to reflect what remains of the primordial follicular pool (140). Decreases

in AFC with age are gradual rather than sudden (159). [3] A total AFC <4 is

predictive of poor response and higher cancellation rates with IVF (160,161).

With the use of machine learning, IVF success prediction models using AMH

and/or AFC can be used together with age and other clinical factors such as BMI,

reproductive history, and clinical infertility diagnoses, to provide highly accurate

IVF live birth probabilities. [3] For example, the use of AMH and/or AFC

reclassified over 60% of patients to have higher IVF live birth probabilities and

over 10% of patients found to have lower probabilities compared to the agecontrol model (150). Some providers use other ovarian reserve tests in addition to

AMH and AFC.

Serum Day 3 FSH

[5] As women age, FSH physiologically rises in the early follicular phase (cycle

day 3), with levels of 5.74 IU/L at ages 35 to 39 and 14.34 IU/L at ages 45 to 59

(162). In a study of 181,536 IVF cycles in the United States, a basal FSH ≥12

constituted the 90th centile of the dataset and was selected as the threshold for

elevated level (141). Higher levels are seen after unilateral oophorectomy (162).

In women in their 40s, levels >20 IU/L are predictive of menopause (162).

Because the incidence of abnormal values is lower in younger women, testing is

typically performed for women aged 35 or older (162). In subfertile women with

an FSH ≥8 IU/L, spontaneous pregnancy rates decrease by 7% per unit of FSH

increase, with a 40% reduction at 15 IU/L and 58% at 20 IU/L (163). FSH levels

vary widely by assay, laboratory, and population (164). Basal FSH ≥12 mIU/mL

has a PPV of 30.9% when used alone and 48% when combined with female age

≥40 years for predicting cycle cancellation for poor response or fewer than four

oocytes obtained at retrieval, such that IVF therapy should not necessarily be

2247withheld from patients diagnosed with diminished ovarian reserve (141).

Compared to women with normal levels, elevated basal serum (>10 mIU/mL) did

not affect monthly fecundability in women aged 30 to 44 without a history of

infertility attempting to conceive spontaneously, regardless of age (157).

Basal Estradiol Level

Basal day 3 FSH is often combined with estradiol (E2) testing. E2 levels on day 3

reflect follicular growth rather than the number of antral follicles (151).

Elevations in FSH and decreases in inhibin B (see below) that accompany aging

result in advanced follicular growth at the end of the preceding luteal phase. In

response, early follicular E2 levels are typically higher in older women and in

women with advanced reproductive aging (140).

Clomiphene Citrate Challenge Test

Clomiphene citrate is thought to have antiestrogenic effects on the hypothalamic–

pituitary axis, resulting in a decrease in E2-mediated suppression of FSH

production by the pituitary (165). The clomiphene citrate challenge test

(CCCT) involves the measurement of serum FSH and E2 on day 3 of the

menstrual cycle, and again on day 10 after administration of clomiphene

citrate (100 mg orally each day) from days 5 to 9 (165). [3] The CCCT has

been reported to be more sensitive than basal FSH alone in identifying poor

response to exogenous gonadotropins, but others report that the predictive values

of the tests do not differ substantially (164,165).

Serum Inhibin B

Serum inhibin B is secreted by ovarian granulosa cells starting at the preantral

follicle stage and therefore reflects the size of the growing follicular cohort (152).

Reduced inhibin B levels are seen with aging even in normal fertile women (151).

Inhibin B alone has poor predictive value for ovarian response (152,164) but

improves the predictive value when added to the CCCT (165). Unlike basal

testing, levels of inhibin B measured on the 5th day of ovarian stimulation were

predictive of live birth following IVF/ICSI (166).

Options for Diminished Ovarian Reserve

[5] Treatments of diminished ovarian reserve include autologous IVF, use of

donor oocytes or embryos, and adoption. Multiple stimulation protocols have

been proposed for autologous IVF in poor responders including judicious use of

gonadotropins with LH activity, estrogen pretreatment, use of GnRH antagonists

during and prior to ovarian stimulation, microdose flare GnRH agonists, and

selective estrogen receptor modulators such as clomiphene and letrozole along

with gonadotropin injections (167,168). All have varying degrees of success, but

2248counseling for diminished ovarian reserve patients in whom donor gametes are

not of interest must include disclosure that per-cycle pregnancy rates are low,

particularly in older women (169), such that multiple IVF cycles may be

necessary and the prognosis may be poor or futile. However, isolated case reports

of success do provide some hope in this regard (170,171), so it may be

appropriate to offer autologous therapy to motivated, well-counseled patients.

Adjuncts During Autologous IVF

[5] Adjunctive autologous IVF therapies include androgens such as DHEA

and testosterone, growth hormone, Coenzyme Q-10 (CoQ10), vitamin D, and

omega-3 fatty acids. Pretreatment of women with diminished ovarian reserve for

2 to 5 months with dehydroepiandrosterone (DHEA; 25 mg three times daily) has

been described to improve oocyte yield and pregnancy rates with IVF in case

series (172–174). Putative mechanisms for this action include increasing serum

concentrations of insulin-like growth factor 1 (IGF-1), which may potentiate

gonadotropic effects on the follicle, and providing the ovary with additional

prohormone for additional ovarian steroidogenesis (174). Pretreatment with

DHEA led to higher ovarian volumes and AFCs in women with premature

ovarian insufficiency (175), however some prospective cohort and randomized

trials have not found statistically significant improvements for stimulation

parameters or pregnancy rates in poor (176,177) or normal responders (178). A

similar rationale has prompted the use of transdermal testosterone 10 to 12.5 mg

for 14 to 28 days pretreatment or during stimulation with varied results in terms

of number of oocytes retrieved and pregnancy rate (179,180). A Cochrane review

of 17 randomized controlled trials found moderate quality evidence of live birth

benefit for DHEA and testosterone in poor responders (181).

Growth hormone, by enhancing the granulosa cell response to gonadotropins

through increases in local IGF-1 production, has been found to improve

pregnancy and live birth rates for poor responders in some meta-analyses

including the Cochrane database (182–184). Suggested dosing for growth

hormone is 2.5 mg subcutaneously from day 6 of stimulation to day of trigger

(182). The cost of growth hormone may be prohibitive for some patients; it is

generally not covered by insurance for off-label use, and such use may be

restricted by the 1990 Anabolic Steroids Control Act (185).

Nutritional supplementation may provide important adjunctive benefits. CoQ10

is found in higher concentrations in follicular fluid of oocytes that produce good

quality embryos by morphokinetic assessment and pregnancies (186). Oocytes

and early embryos obtain energy solely from CoQ10-mediated mitochondrial

oxidative phosphorylation which allows proper oocyte maturation and

fertilization, but aging diminishes CoQ10 production mitochondrial oxidative

2249phosphorylation (187). These changes can be reversed by CoQ10

supplementation in mice and humans (188). Women randomized to 600 mg of

daily CoQ10 for 2 months prior and during their IVF cycle had a nonstatistically

significant increase in good quality embryos and embryo euploidy compared to

the placebo group (187). Omega-3 fatty acids such as those found in fish oil

extends reproductive lifespan in murine models, and this may be through effects

on the pituitary. Normal-weight reproductively healthy women taking 4 g of

eicosapentaenoic acid and docosahexaenoic acid daily for 4 weeks had reduced

baseline and GnRH-stimulated serum FSH levels (189). Vitamin D deficiency has

been inconsistently associated with AMH levels and AFC, so it is difficult to

derive clinical recommendations for vitamin D supplementation solely for the

indication of diminished ovarian reserve (190). Supplementation for vitamin D

deficient patients has produced inconsistent outcomes with respect to pregnancy

rates in the context of IVF (191,192).

Ovulatory Factor

[6] Ovulatory factor accounts for 30% to 40% of all cases of female

infertility (193). Initial diagnoses among women with ovulatory factor infertility

may include anovulation (complete absence of ovulation) or oligo-ovulation

(infrequent ovulation). [1] Menstrual history may be suggestive if

oligomenorrhea, amenorrhea, polymenorrhea, or dysfunctional uterine bleeding

are present (193). Menstrual dysfunction is present in 18% to 20% of the general

population (194). Figure 36-4 shows the fluctuations of E2, progesterone, FSH,

and LH in a normal, 28-day ovulatory cycle. The normal length of the

menstrual cycle in reproductive-age women varies from 21 to 35 days, with a

mean of 27 to 29 days (195). Most of the variability in cycle length occurs during

the follicular phase (196) but the luteal phase, often considered to be fixed at 14

days, can range from 7 to 19 days (197). [1] Even women with regular monthly

menses may have anovulation (194), although the presence of moliminal

symptoms such as premenstrual breast swelling, bloating, and mood changes are

much more suggestive of ovulatory cycles (194). [3] In women with absent or

infrequent ovulation, serum FSH, prolactin, and thyroid-stimulating hormone

(TSH) testing should be performed (193).

Methods to Document Ovulation

The “Fertile Window”

[1] The fertile window is the 6-day interval ending on the day of ovulation,

and does not extend past ovulation (44,196). Sperm can survive for up to 6 days

in well-estrogenized cervical mucus, but the egg may be fertilizable for less than a

2250day (196). Daily intercourse during this window may increase the probability of

conception (44,196). [1] The average woman is fertile between days 10 and 17 of

the menstrual cycle, but many women can conceive outside of this range (197).

[4] Therefore, if timed intercourse (TI) is too cumbersome, intercourse two to

three times weekly throughout the menstrual cycle will likely result in at least

some of those occasions falling within the fertile window (198). Duration of

abstinence prior to the fertile window has not been established, although one

author suggests 5 days (196).

Basal Body Temperature

This inexpensive method involves daily recording of oral or rectal temperature

using a basal body temperature (BBT) thermometer before the patient arises, eats,

or drinks (199). The secretion of progesterone following ovulation causes a

temperature increase of about 0.5° to 1°F over the baseline temperature of 97° to

98.8°F that is typically recorded during the follicular phase of the menstrual cycle

(199). Ovulation is assumed after 3 consecutive days of raised temperatures

(199). Charting of daily BBTs produces a characteristic biphasic pattern in

women with ovulatory cycles (199). [1] Limitations to BBT include its inability

to prospectively predict ovulation and its frequent false negative results

(200). Smoking and irregular sleep patterns can interfere with accurate BBT

testing (199).

FIGURE 36-4 Relative hormonal fluctuations in a normal, ovulatory, 28-day menstrual

cycle.

2251Cervical Mucus

During the fertile window, cervical secretions at the vaginal introitus are

slippery and clear, while secretions at other times of the menstrual cycle are

dry and sticky (44,199). [1] The volume of cervical mucus peaks 2 to 3 days

prior to ovulation, thus identifying higher day-specific probabilities of conception

(44).

Luteinizing Hormone Monitoring

At a mean time of 2 hours following the peak of the serum LH surge, urinary LH

can be detected (201). Commercially available kits for documenting the LH surge

are generally accurate, quick, convenient, and relatively inexpensive enzymelinked immunosorbent assays (ELISA) that use 35 to 50 mIU/mL as their

threshold for detection (201,202). When the LH surge is detected, ovulation

may occur within the next 48 hours (44,201,202). The positive predictive and

negative predictive values for these kits have been described to be 92% (201) and

95% (203), respectively. Because the duration of the surge may be <12 hours,

twice daily testing may increase detection rates (202). However, the 2 days of

highest probability of conception are the day of and the day prior to the LH surge,

so this may lead to abstaining from intercourse during a potentially fertile time

(196). False positive rates occur in 7% of cycles (44,204), which may reflect

urinary clearance of unsustained premature LH surges (200). [1] The tests cannot

be used in patients with irregular cycles (196).

Midluteal Serum Progesterone

When used to document ovulation, serum progesterone measurement should

coincide with peak progesterone secretion in the midluteal phase (typically on

days 21 to 23 of an ideal 28-day cycle or 7 days following the LH surge) (193).

The lower limit of progesterone levels in the luteal phase varies among

laboratories, but a level above 3 ng/mL (10 nmol/L) typically confirms

ovulation (193). Interpretation of isolated luteal phase measurements of

serum progesterone is complicated by the frequent pulses that characterize

the secretion of this hormone (193). [1] Although ovulatory levels are often

considerably higher than 3 ng/mL, low midluteal serum levels of progesterone are

not necessarily diagnostic of anovulation (193).

Ultrasound Monitoring

Ovulation is characterized by a decrease in the size of a monitored ovarian follicle

and by the appearance of fluid in the cul-de-sac using transvaginal US (193). [6]

Follicles reach a preovulatory diameter of 17 to 19 mm in spontaneous cycles

or 19 to 25 mm for clomiphene-induced cycles (205,206). A combination of LH

testing and US can be used, with LH kit testing starting when the US-measured

2252follicle size reaches 14 mm (200). Ten percent of the cycles of normally fertile

women may have a luteinized unruptured follicle, whereby progesterone is

released and the luteal phase progresses normally without visible signs of follicle

rupture when daily USs are performed from cycle days 10 to 20 (207). This

incidence is increased to 25% in women with unexplained infertility (207).

Because of the inconvenience and expense of serial measurements, US

monitoring should be reserved for patients who fail less expensive methods

for detecting ovulation or for certain types of ovulation induction (193).

Polycystic Ovarian Syndrome

[7] The most common cause of oligo-ovulation and anovulation—in the general

population and among women presenting with infertility—is PCOS (208). The

diagnosis of PCOS is determined by exclusion of other medical conditions such

as pregnancy, hypothalamic–pituitary disorders, or other causes of

hyperandrogenism, for example, androgen-secreting tumors or nonclassical

congenital adrenal hyperplasia and the presence of two of the following

conditions (209):

Oligo-ovulation or anovulation (manifested as oligomenorrhea or

amenorrhea)

Hyperandrogenemia (elevated levels of circulating androgens) or

hyperandrogenism (clinical manifestations of androgen excess)

Polycystic ovaries detected by US

Documentation of elevated serum LH: FSH ratios and hyperinsulinemia

are not required for either diagnosis or treatment of PCOS (208,209).

Patients with PCOS should be counseled and screened regarding potential

metabolic disease and obstetric complications prior to fertility treatment (208).

Ovulation Induction in Women With Polycystic Ovarian Syndrome

Despite the use of similar medications, the indications and goals of ovulation

induction should be distinguished from those of superovulation. The goal of

ovulation induction refers to the therapeutic restoration of the release of one

egg per cycle in a woman who either has not been ovulating regularly or has

not been ovulating at all (205). In contrast, the explicit goal of superovulation

for women with unexplained infertility is to cause more than one egg to be

ovulated, thereby increasing the probability of conception (205). Overall,

60% to 85% of women with PCOS will ovulate when treated with clomid or

letrozole. [6] Of the women who ovulate, 15% to 20% will be pregnant per

cycle and 50% will be pregnant by 6 months of treatment (210–212).

2253Weight Loss

Obesity in PCOS patients is associated with poor infertility treatment outcomes

(208,213–215), although the impact on pregnancy loss rates is less clear

(208,213,216). [7] Given that even a 5% weight loss may improve pregnancy

rates, weight loss should be encouraged in all overweight and obese infertility

patients (208). In general, lifestyle modification is the first-line therapy, followed

by pharmacologic treatment and weight-loss surgery (208). Lifestyle

recommendations include a decrease in daily caloric consumption by 500 kcal

and regular physical exercise, although the optimal regimen for the latter is

unknown (208,217). Weight-loss interventions should be undertaken prior to

attempting conception in patients with excess body weight (208).

Ovulation Induction Agent Pharmacology (Clomiphene Citrate [Clomid, Serophene],

Letrozole [Femara], Anastrozole [Arimidex])

Clomiphene citrate is a weak synthetic estrogen that mimics the activity of an

estrogen antagonist when given at typical pharmacologic doses for the induction

of ovulation (205). It is cleared through the liver and excreted into the stool, with

85% clearance in 6 days (205). [7] A functional hypothalamic–pituitary–

ovarian axis is usually required for appropriate clomiphene citrate action

(205). More specifically, clomiphene citrate is thought to bind and block estrogen

receptors in the hypothalamus for prolonged periods, thereby decreasing the

normal ovarian–hypothalamic estrogen feedback loop (205). This blockade

increases GnRH pulsatility, leading to increased pituitary secretion of

gonadotropins which promote ovarian follicular development (205). Letrozole

and anastrozole are potent, reversible nonsteroidal aromatase inhibitors that have

been used off-label for the treatment of patients with anovulatory infertility since

2001 (218). Letrozole is administered orally and cleared by the liver with a

shorter half-life (48 hours) than clomiphene citrate (2 weeks). By reducing

ovarian E2 secretion, letrozole disrupts E2’s negative feedback effect on pituitary

gonadotropin release. The consequent rise in FSH promotes ovarian follicular

development.

Ovulation Induction Outcomes

Ovulation induction is first-line treatment for anovulatory infertility (219).

Over the course of 6 months, clomiphene is associated with 49% ovulation,

23.9% pregnancy, and 22.5% live birth rates in women with anovulatory

infertility (213). [7] Clomiphene effectiveness is decreased by obesity,

increased age, and hyperandrogenic states (213,220,221). Side effects of

clomiphene citrate include vasomotor flushes, mood swings, breast tenderness,

pelvic discomfort, and nausea (205). In a minority of individuals, the

2254antiestrogenic effects of clomiphene citrate at the level of the endometrium or the

cervix may have adverse effects on fertility (205). In the presence of visual

abnormalities, clomiphene citrate should be discontinued promptly (205).

Multiple gestation rates with clomiphene citrate are approximately 8%, most of

which are twins (205). Though historically clomiphene citrate has been the drug

of choice for the treatment of women with PCOS-associated anovulation, current

evidence suggests that letrozole results in higher ovulation and live birth rates

compared with clomiphene, particularly for obese (BMI >30) women. In a

multicenter trial of 750 women with PCOS-associated infertility randomizing to

letrozole (2.5 mg) or clomiphene citrate (50 mg) from cycle day 3 to 7 for up to

five cycles, the cumulative live birth rate was significantly higher in the letrozole

arm (27.5% vs. 19.1%; RR 1.44, 95% CI 1.10–1.87) (222). Twin pregnancy rates

were similar in the letrozole (3.4%) and clomiphene (7.4%) groups (222). A

subsequent meta-analysis of six trials comparing clomiphene and letrozole for

ovulation induction in anovulatory women with PCOS observed higher live birth

rates with letrozole (OR 1.79; 95% CI 1.38–2.31) (223). Side effects with

letrozole include dizziness and fatigue. [7] Although concerns have been raised

regarding possible associations between fetal congenital anomalies and the use of

aromatase inhibitors, no such increase was observed when letrozole was

compared to clomiphene (224).

Ovulation Induction Dosing

Letrozole is supplied in 2.5 mg tablets, and the typical starting dose is 2.5 mg per

day. Clomiphene citrate drug is supplied in 50 mg tablets; the usual starting

dosage is 50 mg/day, but patients who are very sensitive may respond to 12.5 to

25 mg/day (205). Either agent is typically begun within the first 5 days after the

onset of a spontaneous or progesterone-induced menses and is continued for 5

days (i.e., treatment on days 3 to 7 or 5 to 9 of the menstrual cycle) (205). Prior to

initiating letrozole, patients should be counseled that ovulation induction

represents an off-label use of the drug. Given the pregnancy category X

designation of both clomiphene and letrozole, care should be taken to ensure that

the patient is not pregnant prior to administration of these agents. If ovulation

does not occur at the initial dosage of 2.5 mg letrozole or 50 mg clomiphene

citrate, the dosage is increased in each subsequent cycle by 2.5 mg or 50 mg per

day, respectively (205). Seventy-four percent of women will ovulate with 100

mg/day (205), the maximum dose approved by the U.S. Food and Drug

Administration (209). However, some patients need higher doses, which have

been safely given up to 250 mg/day (205). A stair-step method increases the dose

within a single cycle without intervening menses if no follicular response is

documented by US 4 to 5 days after the last dose (225). [7] Ovulation induction

2255treatment should be limited to six ovulatory cycles or 12 total cycles

(205,226,227).

Monitoring Ovulation Induction Therapy

If preovulation monitoring is not performed, patients should be instructed to have

intercourse every 2 to 3 days following the last day of therapy and check serum

progesterone weekly × 5 weeks before inducing a withdrawal bleed or increasing

the dose of the ovulation induction agent (213). Although no clear advantage

has been demonstrated for any ovulation monitoring technique, regular

contact should be maintained with patients to review response to therapy

(205). The urinary LH surge may be detected 5 to 12 days after treatment is

completed (205). When clomiphene or letrozole is given on cycle days 5 to 9, the

surge typically occurs on cycle days 16 to 17 and can be confirmed by midluteal

serum progesterone testing 7 days later (205). With US monitoring, treatment

should be withheld if large cysts are seen on baseline testing (205). Following

ovulation induction use, follicles typically reach a preovulatory diameter of 19 to

25 mm by US, but may be as large as 30 mm (205,206). [7] A combination of LH

testing and US can be used, with LH kits starting when the largest US-measured

follicle reaches 14 mm in diameter (200).

Human Chorionic Gonadotropin

If a dominant follicle develops, but there is no spontaneous LH surge, hCG can be

used to induce final follicular maturation (205), with ovulation occurring

approximately 40 hours following administration (228). Although

administration of hCG at mid-cycle does not appear to improve conception

chances in most infertility patients using clomiphene citrate (104,229,230), it

may be useful for patients with known ovulatory dysfunction (229) or for

IUI. [7] The medication may be derived from urine (5,000 to 10,000 international

units intramuscularly) or manufactured with recombinant technology (250 μg

subcutaneously, equivalent to 5,000 to 6,000 IU urinary) (228).

Insulin Sensitizers

Insulin resistance contributes to the pathogenesis of PCOS. Metformin is an oral

biguanide that is approved for the treatment of noninsulin-dependent diabetes and

has been used in PCOS to increase the frequency of spontaneous ovulation (231).

Metformin acts by several mechanisms, including inhibition of gluconeogenesis

in the liver and increasing the uptake of glucose in the periphery (231). Although

the literature is conflicting, larger studies have suggested that the live birth

rate with metformin alone (7.2%) is lower than that achieved with

clomiphene, and overall the combination does not confer additional benefit

over clomiphene alone (213,232). Subgroups of patients including those who are

2256obese may benefit from combination therapy, however (220,233). Though

prepregnancy use of metformin does not appear to reduce miscarriage rate, there

is insufficient evidence to recommend metformin use during pregnancy to reduce

the risk of miscarriage (234). Risks of metformin include gastrointestinal upset

and rare lactic acidosis, so it should be avoided in settings of hepatic/renal

dysfunction and prior to surgery or use of contrast radiologic dye (231). Reported

effective doses include 500 mg three times daily (231), 850 mg twice daily (231),

and 1,000 mg twice daily (213). The medication is best tolerated when started at

the lowest dose and increased gradually. Extended release formulations are

associated with less gastrointestinal upset (213). Because regular ovulation may

not occur with metformin alone, patients may need medications to induce

withdrawal bleeding to reduce the risk of continued anovulation and

endometrial hyperplasia (208). [7] The use of thiazolidinediones, including

the PPAR-f agonists, rosiglitazone and pioglitazone, for ovulation induction in

PCOS patients has become less widespread as a result of their association

with weight gain and, in the case of rosiglitazone, increased cardiovascular

risk (208).

Dexamethasone

Adjunctive oral dexamethasone may improve ovulation rates in patients resistant

to clomiphene alone (219). These improvements have been reported even in the

absence of adrenal hyperandrogenemia (235,236), so the mechanism of action

remains unclear. Dose regimens have included 0.5 mg × 5 days starting on day of

clomiphene (237), 0.5 mg × 6 weeks prior to starting clomiphene (238), and 2 mg

× 10 days starting on day of clomiphene (235,236).

Oral Contraceptive Pretreatment

[7] Administration of oral contraceptives (OCs) × 2 months prior to beginning a

cycle of clomiphene may improve ovulation and pregnancy rates in clomipheneresistant patients, perhaps by improving a pre-existing hyperandrogenic

environment (219,239).

Gonadotropin Therapy

Anovulatory PCOS patients who fail to ovulate or conceive with oral agents may

be considered for ovulation induction with exogenous gonadotropin injections

(240). For women with clomiphene-resistant PCOS, treatment with gonadotropins

results in ovulation rates of 43% to 83% and pregnancy rates of 21% to 29%

(241–243). Evening medication administration allows for morning monitoring

and mid-day decision-making. Monitoring involves serum E2 levels and

transvaginal US measurements of follicle development. Typical protocols monitor

2257at baseline, 4 to 5 days after treatment initiation, and every 1 to 3 days until

follicular maturation. Expected follicle growth is 1 to 2 mm daily after achieving

10 mm diameter (240). Given the goal of promoting growth of a single mature

follicle, low initial gonadotropin doses of 37.5 to 75 international units

(IU)/day are generally recommended, with increases in doses by 50% of the

previous dose after 7 days if no follicle >10 mm is observed (208,240). Typical

treatment duration is 7 to 12 days, but some patients require longer medication

regimens for adequate stimulation (240). The maximum required gonadotropin

dose seldom exceeds 225 IU per day (240). Ovulation triggering with hCG is

recommended for gonadotropin cycles and is used when 1 to 2 follicles are 16 to

18 mm diameter and the E2 level per dominant follicle is 150 to 300 pg/mL (240).

Ovulation is expected 24 to 48 hours after the hCG trigger. GnRH agonists such

as leuprolide (500-µg subcutaneously) can be used to trigger ovulation to

minimize risks associated with ovarian hyperstimulation syndrome, but require

progesterone supplementation following administration (240). Intercourse should

be recommended within 24 to 48 hours of ovulation triggering or IUI 24 to 36

hours after triggering (240). Testing for pregnancy is performed within 15 to 16

days after ovulation triggering and the cycle reviewed if pregnancy testing is

negative. [7] Gonadotropin dosage in future cycles should be altered if the prior

response was inadequate or excessive.

Gonadotropin Preparations

Several gonadotropin preparations are available. hMG is derived from human

urine and includes approximately equivalent FSH and LH (derived from hCG)

activity of 75 IU each (228). Formulations of hMG and FSH can be administered

either subcutaneously or intramuscularly (228). FSH-only preparations may be

derived either from urine or via recombinant methods, and are packaged either as

lyophilized powder or premixed liquid cartridges/pens (228). Except for highly

purified urinary FSH, which has an FSH activity of 82.5 IU per ampule, all other

products contain 75 IU of gonadotropin when supplied in ampules (228). All

preparations of FSH are highly purified, with minimal to no batch to batch

variation and a high level of safety regardless of the derived source (228).

Despite being differentially marketed as follitropin alpha and beta, these

recombinant FSH preparations still contain combinations of 1 alpha and 1 beta

glycoprotein chain. Rather, they differ in their posttranslational modifications and

processes for purification (228). [7] For PCOS patients, FSH alone appears to be

sufficient in the gonadotropin preparation, although hMG is not harmful (Table

36-7) (240). Contraindications to gonadotropin therapy are listed in Table 36-8.

Gonadotropin Outcomes

2258Cumulative live birth rates are similar when gonadotropins are compared to

clomiphene for ovulation induction when the goal is monofollicular ovulation and

a maximum of six ovulatory cycles is similarly recommended (208). [7] When

compared to other anovulatory patients, PCOS patients using gonadotropins

are at higher risk for multiple gestations (36%), ovarian hyperstimulation

syndrome (4.6%), and cycle cancellation (10%) as a result of their high

numbers of baseline antral follicles (208,212,240). Cancellation should be

strongly considered in patients who reach E2 levels of 1,000 to 2,500 pg/mL,

have ≥3 follicles ≥16 mm, or ≥2 follicles ≥16 mm plus ≥2 follicles ≥14 mm (208).

Sequential use of gonadotropins and either clomiphene or aromatase inhibitors

has been associated with lower gonadotropin requirements and lower cancellation

rates and treatment duration without compromising pregnancy rates. The addition

of aromatase inhibitors is associated with a lower number of dominant follicles

and lower maximum E2 levels (244–246).

[8] TABLE 36-7 Relative and Absolute Contraindications of Gonadotropins for the

Treatment of Infertility in Women

Surgical Treatment

For clomiphene resistant patients, surgical ovarian drilling has been performed as

an alternative to the outdated ovarian wedge resection to decrease ovarian

androgen-producing tissue and to promote ovulation without the risk of multiple

pregnancy seen with gonadotropin administration (208,247). Drilling 3 to 15

puncture sites per ovary is typically performed via laparoscopy using

electrocautery/diathermy (248–251) or laser (252–254), although transvaginal

US-guided (255,256) and vaginal hydrolaparoscopy (257) procedures have been

reported. Successful drilling has been performed with the harmonic scalpel (252).

Within 12 months after ovarian drilling, cumulative ovulation, clinical pregnancy,

and live birth rates are 52%, 26% to 48%, and 13% to 32%, respectively (247).

These outcomes are similar to those using gonadotropins, but ovarian

drilling carries a lower multiple gestation rate (247). Outcomes are not

significantly different when diathermy versus laser are used for drilling (247), but

they are compromised with patient age >35 years or basal FSH >10 mIU/mL

2259(249). [7] The risks of ovarian drilling include surgical complications, adhesions,

recurrence of anovulation, and a theoretical risk of ovarian failure (247,258).

[8] TABLE 36-8 Relative and Absolute Contraindications of Gonadotropins for the

Treatment of Infertility in Women

1. Primary ovarian failure with elevated follicle-stimulating hormone levels

2. Uncontrolled thyroid and adrenal dysfunction

3. An organic intracranial lesion such as a pituitary tumor

4. Undiagnosed abnormal uterine bleeding

5. Ovarian cysts or enlargement not caused by polycystic ovary syndrome

6. Prior hypersensitivity to the particular gonadotropin

7. Sex hormone–dependent tumors of the reproductive tract and accessory organs

8. Pregnancy

Adapted from Physicians’ desk reference. Micromedex (R) Healthcare Series Vol. 107.

Thompson PRD and Micromedex Inc., 1974–2004.

Ovulation Induction for Other Anovulatory Disorders

Hyperprolactinemia

Hyperprolactinemia can be associated with oligomenorrhea or amenorrhea and

should be evaluated with a pituitary MRI to exclude macroadenoma or other

intracranial pathology (259). [6] Dopamine agonists are first-line agents in

symptomatic patients with hyperprolactinemia to restore ovulation (259).

Bromocriptine normalizes prolactin levels and induces ovulation in 80% to 90%

of patients (260). It is taken two to three times daily, and most patients will

respond to a total dose <7.5 mg daily (260). Side effects can be significant and

include nausea, vomiting, postural hypotension, and headache (260). For this

reason, bromocriptine is begun with a low dose of 1.25 mg (half of a 2.5-mg

tablet) daily with food in evenings and prolactin level can be repeated 2 weeks

later to adjust the dose as appropriate (261). The 1.25-mg dose may be adequate

for those with prolactin level <50 ng/mL, while those <100 ng/mL may need 2.5

mg daily (261). In patients unable to tolerate oral administration, bromocriptine

has been used vaginally 5 mg daily (262). Cabergoline has similarly high efficacy

and the advantage of a 0.25 mg twice-weekly dosing schedule and fewer side

effects (260). [6] In a review of over 700 cases of cabergoline use at the time of

2260conception, the risk of miscarriage and malformations was no higher than the

baseline rate in the general population (263).

Hypogonadotropic Hypogonadism

[6] Anovulation in the presence of low serum LH, FSH, and E2 levels defines

hypogonadotropic hypogonadism and reflects dysfunction within the

hypothalamic–pituitary axis (259). Causes of hypogonadotropic hypogonadism,

including craniopharyngiomas, pituitary adenomas, arteriovenous malformations,

or other central space-occupying lesions, should be excluded using magnetic

resonance imaging (259). Stress, extreme weight loss, anorexia, excessive

exercise, and low BMI are all associated with functional hypothalamic

suppression, so good nutrition and optimal body weight should be encouraged to

restore ovulation (259,264). Leptin is a hormone produced by peripheral

adipocytes that reflects energy stores and is deficient in women with diet or

exercise-induced amenorrhea (265). Exogenous leptin has been reported to restore

ovulation in these women (265,266). Other conditions of hypothalamic

dysfunction, such as congenital hypothalamic failure (Kallmann syndrome),

can be treated using pulsatile GnRH therapy or gonadotropins. In these

patients, FSH and LH should be administered (259,267). Pulsatile GnRH

agonist therapy (25 ng/kg every 60 to 90 minutes) simulates normal physiology

and offers some advantages over gonadotropin injections, including fewer

multiple gestations and less ovarian hyperstimulation syndrome (OHSS) while

maintaining excellent pregnancy rates (267).

Hypothyroidism

The prevalence of hypothyroidism among mid-reproductive aged women is 2% to

4% and is mostly caused by autoimmune factors (268). Menstrual abnormalities,

including those from anovulation, are present in 23% to 68% of overtly

hypothyroid women and can be corrected with levothyroxine replacement (268).

The presence of antithyroid antibodies (even if euthyroid) are associated with

increased rates spontaneous pregnancy loss and lower live birth rates (269).

However, among euthyroid Chinese women with elevated thyroid peroxidase

antibodies, levothyroxine supplementation 25 to 50 mcg daily did not affect

miscarriage or live birth rates with IVF in a 2017 randomized trial (270). [2] In

any case, because even very mild or subclinical hypothyroidism can have adverse

effects on fetal brain development and subsequent intelligence quotient, it is

prudent to screen and treat women with thyroid hormone abnormalities before

commencing infertility treatment (268).

Tubal Factor

2261Tubal factor accounts for 25% to 35% of infertility (12). Noninfectious causes

for tubal factor include tubal endometriosis, salpingitis isthmica nodosa, tubal

polyps, tubal spasm, and intratubal mucous debris (12). [8] The incidence of

tubal infertility has been reported to be 8%, 19.5%, and 40% after one, two,

and three episodes of PID, respectively (271). Live birth rates are negatively

affected by the severity of a single episode of PID (272). C. trachomatis and

Neisseria gonorrhoeae are common pathogens associated with PID and infertility

(273). M. hominis and U. urealyticum have been implicated in PID but their

contribution to infertility is less clear (273). Many patients with documented tubal

damage have no history of PID and are presumed to have had subclinical

chlamydial infections (274,275).

Hysterosalpingography

Hysterosalpingography (HSG) is performed after menses but prior to ovulation

between cycle days 7 and 12 to avoid potential pregnancy and take advantage of

the thinner proliferative phase endometrium (276). The patient is typically

premedicated 30 to 60 minutes prior to the procedure with ibuprofen or related

medication (276). Lidocaine injected intracervically may provide further pain

relief (277). With the patient in the dorsal lithotomy position, either a metal

cannula or a balloon catheter is inserted through the cervix and past the

internal cervical os. [3] Contrast dye is injected under fluoroscopy to

visualize the uterine cavity, fallopian tube architecture, and tubal patency

(276,278). Certain disease processes, such as salpingitis isthmica nodosa, have a

characteristic appearance on HSG, with honeycombing as a result of contrastfilled diverticular projections (279). Compared to laparoscopy, HSG has a

sensitivity and specificity of 65% and 83% for the diagnosis of tubal patency

(280). By flushing inspissated mucus and debris, the HSG procedure may have

therapeutic value, particularly with use of oil-based contrast. A large multicenter

RCT that randomized 1,119 women to oil- or water-based contrast found a

significantly higher live birth rate after use of oil-based contrast at HSG (38% vs.

28%; RR 1.38, 95% CI 1.17–1.64) (281).

Hysterosalpingography Risks

Although use of oil-based contrast is thought to carry potential risk of

intravasation-associated oil embolism, this complication has not been observed in

large prospective studies (282). Patients should be screened for and pretreated

with glucocorticoids if iodine allergy is found (278). The risk of PID after HSG is

0.3% to 3.1% overall but is >10% in the setting of hydrosalpinges (283,284).

Therefore, HSG should be avoided in the setting of known hydrosalpinges

and/or current or suspected PID (276,284). The role of routine antibiotic

2262prophylaxis for HSG is controversial (285), but in high-risk patients doxycycline

could be considered. Recommended dosing is 100 mg twice daily, beginning the

day before HSG and continuing for 3 to 5 days. If prophylaxis is not used and

hydrosalpinges are noted on examination, postprocedure doxycycline treatment is

recommended. Other rare complications of HSG include vascular intravasation,

cervical laceration, uterine perforation, hemorrhage, vasovagal reactions, severe

pain, and allergic response to the contrast dye (276).

Sonohysterography

Many infertility practices routinely use sonohysterography to assess tubal patency

because it can be performed in a typical fertility clinic and involves no ionizing

radiation exposure for patient or staff. [3] This advantage over HSG is at least

partially offset by an inability to accurately define tubal architecture when

performed using standard 2D approaches. The use of contrast media during

sonohysterography is preferred to improve accuracy in documenting fallopian

tubal patency, but these contrast agents are not available in the United States. As a

substitute, the use of agitated saline (air-saline) during sonohysterography

provides good negative predictive value when compared to HSG or laparoscopy

(286). Large studies comparing the sensitivity and specificity of

sonohysterography and HSG in the detection of tubal occlusion are not available.

Chlamydia Serology

Chlamydia antibody testing appears to have comparable sensitivity and specificity

to HSG, but does not localize pathology and testing has no therapeutic benefit

(279). It has been proposed that positive serologies in the setting of normal HSG

should still prompt laparoscopy to rule out peritubal adhesions (275).

Laparoscopy

[3] Laparoscopy is considered the gold standard for diagnosing tubal and

peritoneal disease (287). It allows visualization of all pelvic organs and permits

detection and potential concurrent treatment of intramural and subserosal uterine

fibroids, peritubal and periovarian adhesions, and endometriosis (287). Abnormal

findings on HSG can be validated by direct visualization on laparoscopy using

chromopertubation, which involves the transcervical installation of a dye such as

indigo carmine to directly visualize tubal patency and fimbrial architecture (287).

[3] Even laparoscopy has been reported to have a false positive rate of 11%

for proximal tubal occlusion when resected tubal segments are examined

pathologically (288).

Other Diagnostic Modalities

Falloposcopy is used in conjunction with hysteroscopy and allows direct

2263fiberoptic visualization of tubal ostia and intratubal architecture for identification

of tubal ostial spasm, abnormal tubal mucosal patterns, and even intraluminal

debris causing tubal obstruction (288,289). Instrumentation availability and

technical complications, including tubal perforation, limit its routine use (289).

Treatment of Tubal Factor Infertility

[8] As success rates for ART continue to improve, the indications for surgical

approaches in the treatment of tubal infertility have become increasingly limited

(290). Surgery can be effective in several situations and may be the optimal

approach in some patients.

Proximal Tubal Occlusion

Proximal tubal catheterization and cannulation performed either via HSG or

hysteroscopy can restore tubal patency in up to 85% of obstructions, although the

reocclusion rate approaches 30% (12). The best candidates for proximal tubal

catheterization/cannulation have suspected muscle spasm, stromal edema,

amorphous debris, mucosal agglutination, or viscous secretions, while

nonresponders include those with luminal fibrosis, failed tubal

reanastomosis, fibroids, congenital atresia, or tuberculosis (12). Occlusion

caused by salpingitis isthmica nodosa, endometriosis, synechiae, salpingitis, and

cornual polyps will only occasionally respond to catheterization/cannulation (12).

Catheterization involves passage of a soft catheter into the tubal ostia, while

cannulation passes a guidewire through the ostia and injects contrast media or

colored dye (12,279). Tubal perforation, typically minor, occurs in 1.9% to 11%

of cases (12,279,288). Catheterization under fluoroscopy during HSG is

referred to as selective salpingography (291). Visualization of patency with

the hysteroscopic approach can be accomplished using laparoscopy (12) or

US (288). Ongoing pregnancy rates following proximal tubal

catheterization/cannulation are 12% to 44% regardless of hysteroscopic or HSG

approach (12). If occlusion persists or recurs, IVF is usually recommended.

Microsurgical tubocornual anastomosis is an option with small studies reporting

pregnancy rates of up to 68% (12). This procedure is typically performed via

laparotomy and involves excision of the tubal isthmus, followed by

reimplantation of the residual tube into a new opening made through the uterine

cornua (279).

Distal Tubal Occlusion (Excluding Sterilization or Hydrosalpinx)

[8] Distal tubal disease and occlusion are causal in 85% of all tubal infertility

and can be secondary to a variety of inflammatory conditions including

infection, endometriosis, or prior abdominal or pelvic surgery (279,292).

2264Patients younger than 35 years of age with mild distal tubal disease, normal tubal

mucosa, and absent or minimal pelvic adhesions are the best candidates for

corrective microsurgery (292). IVF should be considered for older patients or

those with diminished ovarian reserve, combined proximal and distal tubal

disease, severe pelvic adhesions, tubal damage that is not amenable to

reconstruction, or additional infertility factors (292,293). Fimbrioplasty involves

lysis of fimbrial adhesions or dilation of fimbrial phimosis, whereas

salpingostomy (also known as salpingoneostomy) involves the creation of a new

tubal opening in an occluded fallopian tube (292). In well-selected patients,

pregnancy rates are reported to be 32% to 42.2%, 54.6% to 60%, 30% to 34.6%,

and 55.9% for adhesiolysis, fimbrioplasty, salpingostomy, and nonsterilizationrelated anastomosis, respectively (279,292). As a group, these procedures are

associated with a 7.9% rate of subsequent ectopic pregnancy (292).

Sterilization Reversal

Twenty percent of women express regret following sterilization, and 1% to

5% of those will request reversal (294), often as a result of a change in

marital status (293). The technique for sterilization reversal involves

microsurgical dissection of the occluded ends of the fallopian tube followed by a

layered reapposition of the proximal and distal tubal segments (293). Surgical

approaches include minilaparotomy, laparoscopy (293), and robotic-assisted

laparoscopy (294). Pregnancy rates following microsurgical tubal reanastomosis

for sterilization reversal are 55% to 81%, with most pregnancies occurring within

18 months of surgery (293). Ectopic pregnancy rates following the procedure are

generally <10% but may approach 18% (292,295). The main predictors of

success are age <35 years (293,295), isthmic–isthmic or ampulo-ampullar

anastomosis (295), final anastomosed tubal length >4 cm (293), and less

destructive sterilization methods such as rings or clips (293,295). Unlike

vasectomy reversal, the length of time between fallopian tubal sterilization and

reversal does not seem to affect outcome (293). [8] IVF should be considered in

lieu of sterilization reversal for older patients or those with diminished ovarian

reserve, severe pelvic adhesions, additional infertility factors or prior unsuccessful

reanastomosis (292,293).

Hydrosalpinx

Distal occlusion may lead to fluid buildup in the fallopian tube causing a

hydrosalpinx (279). Hydrosalpinx fluid impedes embryo development and

implantation (279). A meta-analysis of 14 studies and 1,004 patients with

hydrosalpinges concluded that IVF pregnancy rates were significantly lower in

the presence of hydrosalpinges (296). [9] Salpingectomy for hydrosalpinx prior

2265to IVF significantly improves both pregnancy and live birth rates when

compared to IVF performed with the fallopian tubes in situ (297,298),

although laparoscopic tubal occlusion appears to be a reasonable alternative

(299). There is significantly less outcome data on the use of transvaginal needle

drainage and salpingostomy for treatment of hydrosalpinges prior to IVF (279).

Uterine Factors

[8] Pathologies within the uterine cavity are the sole cause of infertility in as

many as 15% of couples seeking treatment (300) and are diagnosed in >50%

of infertile patients (300). Therefore, the evaluation of the couple with infertility

should consistently include an assessment of the endometrial cavity. Uterine

cavity abnormalities include endometrial polyps, endometrial hyperplasia,

submucous myomas, intrauterine synechiae, and congenital uterine anomalies

(301).

Diagnostic Imaging for Uterine Pathology

Hysteroscopy

[3] Hysteroscopy is considered the gold standard for uterine cavity

evaluation because it permits direct visualization (300–302). The procedure

involves insertion of an endoscope through the cervical canal into the uterine

cavity and instillation of distention media to facilitate visualization (300–302).

Diagnostic hysteroscopy may be performed in the office using a smalldiameter hysteroscope and saline distention, often without need for

anesthesia (301). To optimize visualization of the endometrial cavity and avoid

performing the procedure during early pregnancy, hysteroscopy is typically

scheduled during the early- to mid-follicular phase of the cycle (300).

Disadvantages to the procedure include poor visualization when uterine bleeding

is present and the inability to evaluate structures outside the uterine cavity,

including those in the myometrium and adnexa (300). Office hysteroscopy is

reported to have 72% sensitivity for cavity abnormalities when compared to

operative hysteroscopy using general anesthesia (300).

Hysterosalpingogram

Insofar as it allows assessment of tubal and uterine pathology, HSG (“Tubal

Factors”) is a reasonable initial imaging technique to use in the basic infertility

evaluation (301). Hysterosalpingogram shows the general configuration of the

uterine cavity and indicates endometrial lesions as filling defects or irregularities

of the uterine [3] wall (301). Excessive contrast may lead to false negative

findings (301), which may account for the 50% sensitivity of HSG compared

2266to hysteroscopy for endometrial polyps (302). Inability to discriminate air

bubbles, mucous, and debris from true intracavitary pathology may account for

HSG’s high false positive rate when compared with hysteroscopy (301,302).

Other drawbacks include patient discomfort, use of iodinated contrast, and

radiation exposure (301).

Transvaginal Ultrasound

Compared to hysteroscopy, standard 2D transvaginal US has a 75% PPV and

96.5% negative predictive value for intracavitary polyps but a 0% PPV for

intrauterine adhesions (302). Similar to HSG, it has a sensitivity of 44% for

uterine malformations (302).

Sonohysterography

Saline infusion sonography (SIS), synonymous with sonohysterography, involves

the transcervical instillation of saline, often via a balloon catheter, during

transvaginal US to distend the uterine cavity and delineate the endometrium

(303). As with office hysteroscopy and HSG, SIS is performed during the

follicular phase of the cycle and anesthesia is typically not required (303).

Endometrial polyps appear as hyperechogenic pedunculated lesions, submucous

fibroids have mixed echogenicity, and adhesions contain densely echogenic and

cystic areas (303). Compared to hysteroscopy, SIS has 100% sensitivity,

specificity, positive and negative predictive values for uterine polyps (302).

As a result of the smaller volume of distention media used, SIS is generally

better-tolerated than HSG or hysteroscopy (300,303). Another advantage of SIS is

the ability to evaluate the adnexa, and the myometrium for fibroids or

adenomyosis. When combined with 3D technology, SIS is particularly good at

assessing the overall uterine contour and delineating congenital anomalies such as

septate uteri (303). Standard SIS has a somewhat lower sensitivity of 77.8% in

detecting uterine congenital anomalies when compared to hysteroscopy, but this

is higher than 2D transvaginal US or HSG (302). [3] Hysterosalpingogram and

SIS perform similarly for intrauterine adhesions, each having approximately 50%

PPV and >90% negative predictive value (302).

Magnetic Resonance Imaging (MRI)

Although transvaginal US, HSG, SIS, and hysteroscopy may suggest congenital

uterine anomalies, pelvic MRI is considered the gold standard for imaging and is

particularly useful for diagnosing rudimentary uterine horns (304). Pelvic MRI

has the best sensitivity and specificity for intramural and submucous myomas

compared to pathologic examination (305), and is especially useful for large or

multiple fibroids (306). [3] MRI has been suggested as a tool to differentiate

fibroids from adenomyosis, but routine use over US is not recommended

2267(306,307).

Congenital Anomalies of the Uterus

Congenital uterine anomalies occur in 3% to 4% of women. This increases to 5%

to 10% in women with early pregnancy loss and up to 25% in those with second

and third trimester pregnancy losses (304). During female embryonic

development the paired paramesonephric or mullerian ducts elongate toward each

other and fuse in the midline. This is followed by resorption of the intervening

septum to form the upper vagina, cervix, uterus, and fallopian tubes by week 20

of gestation (308). Failure of any of these steps leads to absent uterine

development or development of unicornuate, bicornuate, arcuate, didelphys, or

septate uteri (308). Because of the proximity of the paramesonephric ducts to

the urinary system, renal anomalies often coexist with mullerian anomalies.

Appropriate urologic imaging should be performed whenever a mullerian

anomaly is diagnosed (304,308). [2] Uterine anomalies are more closely

associated with pregnancy wastage and poor obstetric outcomes than

infertility, as the prevalence of congenital uterine defects is generally similar

among fertile and infertile women (304). The exception to this is mullerian

agenesis. Patients with mullerian agenesis can have genetically related children

only through use of IVF and a gestational carrier (304). The arcuate uterus is the

mildest congenital uterine anomaly and typically live birth rates are comparable to

those in women with normal uteri (304). Surgical uterine repair to improve

obstetric outcomes remains controversial for most anomalies. [8] However,

rudimentary uterine horns require removal on diagnosis and hysteroscopic

metroplasty of the septate uteri significantly reduces the rates of pregnancy

loss, but not infertility (304,309). The number of reproductive-age patients

presenting with in utero exposure to DES is declining rapidly and will continue to

decline, as the substance was banned in 1971 (304). Women whose mothers were

exposed to DES have higher rates of uterine malformations (e.g., T-shaped

uterus) and associated obstetric complications (304).

Acquired Abnormalities of the Uterus

Leiomyomas

Leiomyomas, also called myomas or fibroids, are benign monoclonal uterine

myometrial tumors that affect 25% to 45% of all reproductive-age women; this

number is even higher in African Americans (310). The mechanisms by which

fibroids cause infertility are unknown, but may involve altered uterine

contractility, impaired gamete transport, or endometrial dysfunction (311).

Among women with infertility and uterine leiomyomas, pregnancy rates are

primarily affected by leiomyoma location (305,311). [8] Subserosal fibroids do

2268not appear to affect fertility or obstetric outcomes, while cavity-distorting

intramural, noncavity-distorting >5 cm, and submucosal myomas are

associated with lower implantation and live birth rates (305,311,312).

Myomectomy

In women desiring fertility who require treatment for fibroids, myomectomy is

the preferred approach and uterine artery embolization is relatively

contraindicated (311). Removal of cavity-distorting intramural and

submucous myomas is generally recommended prior to proceeding with

infertility treatment (305,311). [8] The utility of surgical removal of

noncavity-distorting intramural fibroids is unknown (305,311,312).

Myomectomy can be performed hysteroscopically (306), via laparotomy

(310,313), laparoscopically (alone [310,314], with robotic assistance [313,314]),

or vaginally (315). Hysteroscopic removal is generally preferred for small

submucous fibroids (306), while use of the other methods generally depends on

patient preference, operator skill, or the presence of other pelvic pathologies

(310,313,314). The utility of pretreatment with GnRH agonists prior to surgery is

debatable. GnRH agonists may shrink larger fibroids (5 to 6 cm) enough to allow

hysteroscopic resection (306) and may decrease the risk of intraoperative blood

loss and postoperative anemia (310). Fluid overload and uterine perforation

are the most common complications of hysteroscopic myomectomy (306),

while bleeding and adjacent organ injury are more often associated with

alternative approaches (310). Fibroids that are located low on the uterus and

posteriorly are less amenable to laparoscopic resection (310). Transmyometrial

approaches raise concern for uterine rupture during pregnancy, although this risk

appears to be very low (310).

Endometrial Polyps

The incidence of asymptomatic endometrial polyps among women with infertility

has been reported to range from 6% to 8% (316,317) but may be as high as 32%

(318). Risk factors for polyp development include obesity, unopposed estrogen

exposure, and polycystic ovary syndrome (319). The mechanisms by which

endometrial polyps may impair fertility are incompletely described, but may

relate to disordered endometrial receptivity (319). One report localized 32% of

endometrial polyps in infertile women to the posterior uterine wall, indicated that

40.3% of patients had multiple polyps, and stated a 6.9% hyperplasia rate (320).

Polypectomy is generally performed via curettage, blind avulsion, or

hysteroscopic removal (318). [8] Although the efficacy of polypectomy prior to

infertility treatment has not been clearly established (318), a prospective

randomized trial showed a 2.1-fold higher rate of pregnancy among women

2269who underwent the procedure prior to IUI (321). Higher pregnancy rates have

been noted for polyps removed from the uterotubal junction when compared to

those removed from other locations (320). Smaller nonrandomized studies

provide conflicting data on the negative fertility effects of polyps <1.5 to 2 cm

(320,322,323).

Intrauterine Synechiae or Asherman Syndrome

Severe trauma to the basalis layer of the endometrium with subsequent tissue

bridge formation leads to intrauterine synechiae or Asherman syndrome (324).

Symptoms of severe disease include amenorrhea, menstrual irregularities,

spontaneous abortion, and recurrent pregnancy loss (324). [1] The causes of

intrauterine adhesions are often iatrogenic, with patients typically reporting

intraoperative or postoperative complications of uterine evacuations for

incomplete pregnancy loss, pregnancy termination, or postpartum

hemorrhage [1] (324). Myomectomy, hysterotomy, diagnostic curettage,

cesarean section, tuberculosis, caustic abortifacients, and uterine packing are less

common causes in Western countries (324). In developing countries, Asherman

syndrome caused by genital tuberculosis is quite common (325). Hysteroscopic

resection of synechiae is the preferred treatment to restore fertility in women

with Asherman syndrome. Patients with genital tuberculosis have a very

poor prognosis (324–326). Postoperative prevention of adhesion reformation

disease may involve estrogen therapy alone × 1 month or in combination with

intraoperative placement of an intrauterine device (such as a small Malecot

catheter or pediatric Foley catheter) for 1 to 2 weeks (324,326). There is no

standard regimen for estrogen therapy, but oral conjugated estrogens 2.5 mg daily

overlapping with progestin (324) or E2 valerate 2-mg injections daily (326) have

been suggested.

Luteal Phase Defect and Progesterone Supplementation

Mechanisms

The luteal phase is normally characterized by progesterone secretion by the

corpus luteum and appropriate endometrial secretory transformation that allow for

embryonic implantation in the endometrium and support of early pregnancy for

the first 7 to 8 weeks of gestation (327,328). Luteal phase deficiency (LPD) is a

condition in which endogenous progesterone is not sufficient to maintain a

functional secretory endometrium during the implantation window (329,330)

and is thought to account for 4% of infertility (329). Proposed mechanisms for

LPD include inadequate production of progesterone following ovulation (331),

improper GnRH pulsatility causing insufficient gonadotropin production during

the LH surge (329,330), and inadequate endometrial responsivity to progesterone

2270(329). Assisted reproductive technologies (ARTs) or gonadotropin ovulation

induction medications may induce iatrogenic LPD via disruption of

granulosa cells from follicular aspiration and suppression of endogenous LH

secretion through a combination of supraphysiologic E2 levels and GnRH

agonist/antagonist therapy (328,331).

Diagnosis

Diagnostic criteria for LPD have been variably defined, but have included a low

mid-luteal phase serum progesterone level, a BBT rise lasting less than 11 days,

and a shortened luteal phase of less than 14 days (332). Unfortunately, the

characteristic pulsatile secretion of progesterone during the luteal phase of the

menstrual cycle combines with wide temporal variations (even within a 60- to 90-

minute time span) to make interpretation of mid-luteal progesterone levels

difficult (327). Similar rates of shortened luteal phase are found in fertile and

infertile women and there is significant variability of luteal phase length

from cycle to cycle in an individual woman (330). The notion of a universal

optimal duration of progesterone to achieve embryo implantation has been

questioned (333). [1] There is significant interobserver variability in

pathologic interpretation of endometrial biopsies from infertile women (334)

and out-of-phase biopsy results poorly discriminate between fertile and

infertile women (335).

Treatment

[8] Progesterone therapy is considered standard practice during ART cycles,

but is more controversial during non-ART fertility treatments (336,337).

When used, progesterone supplementation can be administered via oral, vaginal,

or intramuscular routes, with highest serum levels associated with daily 50- to

100-mg intramuscular progesterone (336,337). Most products are delivered in oil,

and caution should be used to ascertain the presence of sesame or peanut

allergies. Vaginal micronized progesterone, 200 to 600 mg daily in divided doses,

is associated with the highest endometrial concentrations of progesterone. The

side effects of this delivery route can include vaginal discharge and irritation.

Other vaginal preparations include a once-daily gel and a 100-mg insert that is

given two to three times daily. As a result of erratic absorption and decreased

bioavailability, oral micronized progesterone is less effective for progesterone

supplementation (336). There remains no consensus regarding the superiority

of vaginal versus intramuscular administration. When hCG is used for trigger,

hCG levels peak at the time of egg retrieval, followed by a rise in serum

progesterone levels that peak 5 days postretrieval and then decline (336). There

are no differences in pregnancy or live birth rates when progesterone is initiated

2271on the same day following retrieval versus 1 day after retrieval (338). A

systematic 2015 review indicated lower pregnancy rates when progesterone

support was administered prior to hCG-triggered egg retrieval or delayed until 6

days postretrieval, and suggested an initiation window for intramuscular

progesterone of postegg retrieval day 0 to day 3 postretrieval (336). This initiation

window is consistent with descriptions of luteal phase support during GnRHagonist–triggered cycles (339). Because of the more rapid endometrial

advancement compared to intramuscular route, delayed vaginal progesterone

administration until 48 hours postretrieval has been advocated, but better-quality

randomized trials are needed before drawing definitive conclusions (336). [9]

Though the optimum duration of supplementation has not been determined,

progesterone is typically continued until 8 to 10 weeks of gestation (336,337).

Intravaginal luteal phase supplemental progesterone may increase pregnancy rates

in women with PCOS treated with letrozole (340). [8] A 2017 systematic review

and meta-analysis indicated a benefit for unexplained infertility using primarily

vaginal progesterone initiated 0 to 2 days post-IUI in cycles using clomiphene

alone or with gonadotropins with 1 live birth for every 11 patients treated (337).

Pelvic Factor

Endometriosis

Endometriosis affects 6% to 10% reproductive age women (341) but is present in

25% to 50% women with unexplained infertility (287). [8] It is characterized by

the presence of endometrial glands and stroma located outside of the uterine

cavity and is found primarily on the peritoneum, ovaries, and rectovaginal septum

(341). Fecundability rates in affected patients are estimated at 2% to 10% per

month (342). Possible mechanisms for infertility among women with

endometriosis include anatomic distortion from adhesions or fibrosis and the

impact of chronic inflammation on gametes, embryos, tubal fimbria, and eutopic

endometrium (341,342). Laparoscopic visualization and histologic confirmation

of biopsied lesion(s) remains the gold standard for the diagnosis of endometriosis.

[8] The disease is staged laparoscopically according to the Revised American

Society for Reproductive Medicine’s classification, with stage III to IV

(moderate-severe) characterized by ovarian endometriomas, dense tubal or

ovarian adhesions, and/or cul-de-sac obliteration (341). Vaginal US has excellent

specificity for endometriomas and other deep disease (343). It is unclear whether

the mere presence of endometriosis negatively affects IVF outcomes.

Endometriosis patients have similar rates of embryonic aneuploidy (344) and live

birth with ARTs (345,346) as compared to unaffected patients. [9] However,

more severe disease (stages III and IV) and previous endometriosis surgery are

2272associated with fewer oocytes retrieved and reduced pregnancy/live birth rates (by

30% to 40%) compared to patients without endometriosis (345,346).

Endometriosis Infertility

Management

Hormonal suppression of endometriosis typically has minimal benefit for

endometriosis-related infertility (341,347). For minimal to mild disease,

laparoscopic ablation appears to significantly improve pregnancy rates when

compared to diagnostic laparoscopy alone as evidenced by a large randomized

trial reporting 31% (treated) versus 17% (untreated) pregnancy rates (348,349). A

subsequent meta-analysis supported the effectiveness of laparoscopic treatment in

stages I to II endometriosis-associated infertility (341,342,350,351). [8] Number

needed to treat analysis suggests that eight laparoscopies involving treatment

of mild or minimal endometriosis would need to be performed for each

pregnancy gained. A Cochrane review found moderate evidence in favor of

particularly excisional laparoscopic treatment of minimal-mild endometriosis for

spontaneous pregnancy in the 9 to 12 months postoperatively (183,347). Although

removal of endometriomas may be indicated prior to IVF, when they would

interfere with oocyte retrieval (349), endometrioma resection during IVF/ICSI

treatment is associated with decreased ovarian function in up to 13% of cases

(352,353), reduced quantity of dominant follicles, and fewer retrieved oocytes

compared to unoperated patients in a meta-analysis of 2,649 IVF cycles (354).

Furthermore, 40% of endometriomas recur postoperatively (349). [8] Therefore,

IVF is considered a reasonable first-line therapy for endometriosis-associated

infertility because of the short time to pregnancy and avoidance of surgery (290).

Sclerotherapy, whereby the endometrioma is drained via transvaginal puncture

and a sclerotic agent such as ethanol is instilled into the cyst cavity, was

associated with higher numbers of retrieved oocytes and similar pregnancy rates

as compared to laparoscopic management, in a meta-analysis of 18 studies, and

may be an alternative to surgery (345).

Adhesions

Adhesions may result from sharp, mechanical, or thermal injury, infection,

radiation, ischemia, desiccation, abrasion, or foreign body reaction (355).

Adhesiolysis improves pregnancy rates by 12% at 1 year and by 29% at 2 years in

infertile women with adnexal adhesions (355). Use of adhesion barriers reduces

adhesion formation following laparoscopy and laparotomy, but there is no

consistent evidence for improvement in pregnancy rates (355,356).

Unexplained Infertility

2273Thirty percent of couples are diagnosed with unexplained infertility, in

which the basic infertility evaluation reveals normal semen parameters,

evidence of ovulation, patent fallopian tubes, and no other obvious cause of

infertility (357). Patients with unexplained infertility may feel reassured that

even after 12 months of attempting, 20% will conceive in the following 12

months and over 50% in the following 36 months (4). In couples with good

prognostic factors of female age <30, <24 months of infertility, and a previous

pregnancy in the same partnership (4), unexplained infertility may merely reflect

the lower extreme of normal fertility (357). It is likely that technology is limited

in terms of diagnosing all causes for infertility (357). [2] The utility of evaluations

other than basic testing in an infertile couple has yet to be proven (358).

Proposed Mechanisms for Unexplained Infertility

Luteinized Unruptured Follicle Syndrome

This condition involves luteinization of a follicle that has failed to rupture and

release its oocyte, leading to a normal menstrual cycle but infertility (207). It is

thought to occur in up to 25% of patients with unexplained infertility, more than

twice the incidence in fertile women (207). The diagnosis may justify the use of

IVF whereby follicles are aspirated and oocytes are retrieved and fertilized in

vitro.

Immunologic Factors

The impact of the immune system on reproductive function is an area of intense

study and debate. Among the potential immune causes of reproductive failure are

imbalances in the T-lymphocyte population. By increasing the secretion of antiinflammatory TH2 cytokines and inhibiting the secretion of proinflammatory TH1

cytokines, progesterone shifts the balance to TH2 dominance to support embryo

implantation and early pregnancy (359). The absence of a reliable screening test

for TH1:TH2 balance has handicapped studies designed to influence the balance

toward improved outcomes. Although serum antiphospholipid antibodies and

antithyroid antibodies are more prevalent among patients with unexplained

infertility than among fertile women (360), the presence of antiphospholipid

antibodies has not been found to adversely affect IVF outcomes, so screening is

discouraged (361). Furthermore, a randomized double-blinded placebo controlled

trial found no difference in in vitro fertilization-embryo transfer (IVF-ET) live

birth rate with administration of heparin 5,000 units SC and aspirin 100 mg daily

in patients with detectable antiphospholipid antibodies (362). The relationship of

thyroid antibodies and reproductive outcomes are discussed in the Ovulatory

Factor section.

2274Unexplained infertility has been associated with antisperm antibodies (363),

described further in the Male Factor section. Uterine natural killer cells (UNKs)

account for over 70% of leukocytes in the secretory endometrium and appear to

play a primary role in regulating early placentation (364). UNKs express

specialized cell surface receptors known as killer cell immunoglobulin receptors

(KIRs), encoded by highly polymorphic genes classically divided into inhibitory

(KIR A) and activating (KIR B) haplotypes (365). The ligand for KIRs is the

highly polymorphic HLA-C molecule expressed by extravillous trophoblast cells.

In a study of 668 euploid single embryo transfers with annotation of maternal

KIR and embryonic HLA-C haplotypes, maternal KIR haplotype was found to

influence the risk of pregnancy loss and the HLA-C ligands modified the risk

(366). High-risk combinations such as KIR A/homozygous C2 resulted in a 51%

increase in the risk of pregnancy loss with euploid embryo transfer (366). [2]

These studies support a potential role for immune factors contributing to

unsuccessful implantation and pregnancy.

Infection

C. trachomatis and related clinical and subclinical infections have been discussed

in the “Tubal Factor” section. No consistent associations have been reported

between chlamydial species, M. hominis and unexplained infertility in men or

women (275,367,368). U. urealyticum and Mycoplasma genitalium may be of

more concern (369). Prophylactic doxycycline (100 mg twice daily for 4 weeks)

given to infertile couples improved pregnancy rates only among couples in whom

the male partner cleared ureaplasma infections (370). Antimicrobial prophylaxis

is often given for ART, but it is not clear the extent to which clearance of these

organisms improves pregnancy rates (369,371). Chronic endometritis is an

inflammatory condition characterized by the presence of plasma cells in the

endometrium found in 56.8% of patients with unexplained infertility, 57.6% of

patients with repeat implantation failure, and 42.3% of women with endometriosis

(372–374). A significantly higher rate of spontaneous pregnancies and live births

were observed in patients successfully treated with antibiotics as compared to

patients with persistent endometritis after antibiotics (374). [8] Although the

uterine cavity has classically been designated to be sterile, different bacterial

communities have been detected that appear to influence implantation following

IVF embryo transfer (375).

Undiagnosed Pelvic Pathology

[8] Following a negative infertility workup, laparoscopy has been proposed to

evaluate for peritubal adhesions that may distort tubo-ovarian relationships and

endometriosis (376,377) but there is a lack of consensus as to the frequency of

2275these abnormalities and many practitioners will forego laparoscopy in lieu of

empiric fertility treatment in such patients. In young patients, laparoscopy may be

cost-effective compared to fertility therapy (378).

Occult Male or Oocyte Factors

[2] Occult male factor, such as impaired sperm DNA integrity (see Male Factor),

despite normal semen analysis and oocyte factors, such as premature zona

hardening, mitochondrial dysfunction, and/or aberrant spindle formation have

been suggested as putative mechanisms for unexplained infertility (379).

Treatment of Unexplained Infertility

Historically, unexplained infertility was often treated with superovulation using

clomiphene, letrozole, or gonadotropins in a stepwise approach, combined with

IUI (e.g., CC/IUI, letrozole/IUI, and FSH/IUI), followed by ART. However, a

randomized clinical trial (the FASTT trial) showed that for patients aged 21 to 39,

CC/IUI and FSH/IUI were inferior compared to IVF based on their lower ongoing

pregnancy rates (CC/IUI 7.6%, FSH/IUI 9.8%, and IVF 30.7%), a shorter timeto-pregnancy period and higher costs (380). These results are consistent to those

reported by Guzick et al. in which the per-cycle pregnancy rates are 9% for

gonadotropins/IUI, 4% for superovulation, or 5% for IUI alone (381). Metaanalysis of 1,159 participants with unexplained infertility involving seven

randomized controlled trials of CC alone or CC/IUI indicated no

improvement in pregnancy or live birth rates when compared to no

treatment or placebo (227). Another review of over 3,000 patients showed that

CC or CC/IUI or TI were similar to expectant management (382). Therefore,

CC/IUI or FSH/IUI is not the treatment of choice for unexplained infertility

(383,384). Similarly, in the Forty and Over Treatment Trial (FORT-T trial),

patients aged 38 to 42 were found to have higher live birth rates from IVF than

from CC/IUI or FSH/IUI (385). [9] Despite research findings supporting the

use of IVF as first-line treatment for unexplained infertility,

superovulation/IUI treatments have continued to be widely used, presumably

because they are less expensive than IVF on a per-cycle basis and in some

countries, superovulation/IUI may be much more accessible than IVF. (Risks with

clomiphene, letrozole, and non-ART gonadotropins, and gonadotropin

preparations, are discussed in the “Ovulatory Factor” section.)

ART options for unexplained infertility include conventional IVF, split

IVF/ICSI, and ICSI, depending on whether the patient and her provider accept the

risk of failed fertilization. The use of ICSI or splitting sibling oocytes between

IVF and ICSI has been proposed as an approach to unexplained infertility patients

to diagnose and treat underlying occult male or oocyte factors (379,386). While

2276ICSI is often suggested or recommended to patients with unexplained

infertility to minimize the occurrence of failed fertilization caused by occult

male or oocyte factors (379,386), IVF and ICSI have comparable pregnancy

or live birth rates in patients with unexplained infertility (386,387).

For young couples with unexplained infertility, especially those with a desire to

have more than one child, diagnostic laparoscopy may be a good option for the

evaluation and possible treatment of endometriosis (378,388).

Superovulation (Controlled Ovarian Hyperstimulation) with IUI

Unlike ovulation induction, in which the goal is to stimulate the release of a

single oocyte in anovulatory women, the explicit goal of superovulation (for nonART or ART purposes) is to cause more than one egg to be ovulated, thereby

increasing the probability of conception in women with unexplained infertility

(205). [8] In most superovulation protocols, a baseline scan is performed on

day 2 of the menstrual cycle to assess AFC and the presence or absence of

ovarian cysts. Baseline estrogen and progesterone levels are typically

obtained on this day (389). Starting daily doses in superovulation cycles are

higher for clomiphene (100 mg × 5 days) and gonadotropins (150 to 300 IU

daily) than those used for ovulation induction (152,381,389). Superovulation

with clomiphene is otherwise conducted as described above for ovulation

induction. When gonadotropins are used in normal responders with

unexplained infertility, FSH doses of 225 IU and 300 IU lead to similar

outcomes with IVF (152). The maximal gonadotropin dosage is typically 450 IU

per day because higher dosages do not increase ovarian response (389). In most

superovulation protocols, gonadotropins are started on day 2 or 3 of menses

(381,389,390). The starting dose of gonadotropins is maintained each day

until cycle day 6 or 7 (stimulation day 4 or 5), when the serum E2 level and

transvaginal US are first measured to document ovarian response

(152,381,390). Gonadotropin dosage is increased by 50 to 100 IU per day

every 2 to 4 days until a response is evident (391). E2 levels typically double

every 48 hours, with follicle growth of 1.7 mm daily after 10-mm diameter is

reached (392). Triggering of ovulation typically occurs when at least 2

follicles have reached an average diameter of ê17 to 18 mm and the

endometrial thickness is ê8 mm (152,381,390,393–395). For non-ART cycles,

patients should receive counseling about the risk of high-order multiple gestation

and offered the option for cancellation for E2 levels 1,000 to 2,500 pg/mL, for

ê3 follicles ê16 mm, or for ê2 follicles ê16 mm plus ê2 follicles ê14 mm (208).

In a randomized trial, previously untreated couples with unexplained

infertility had similar cumulative live birth rates using three cycles of

gonadotropin/IUI compared to a single cycle of IVF (5).

2277Cost-Effectiveness

According to computerized decision-tree modeling, laparoscopy followed by

expectant management for unexplained infertility may be cost-effective when

compared to no intervention, non-ART treatment, or IVF (378). Given that 15

gonadotropin/IUI cycles are needed to produce one additional pregnancy when

compared to ICI alone (357), IVF seems a reasonable option. [8] In

Massachusetts, where insurance companies are required to cover infertility

treatment, the cost-effectiveness of IVF in the treatment of unexplained infertility

was determined (380). Couples were randomized to accelerated treatment with

three cycles of clomiphene citrate/IUI followed by up to six cycles of IVF if no

pregnancy occurred (n = 256) versus conventional treatment with clomiphene/IUI

× 3 cycles, then injectable gonadotropins/IUI × three cycles, then IVF × six cycles

(n = 247) (380). Median time to pregnancy was 8 months in the accelerated

versus 11 months in the conventional treatment groups. The authors used an

average cost per cycle with clomiphene/IUI of $500, with gonadotropin/IUI of

$2,500, and assuming IVF costs were <$17,749/cycle. They reported that the

accelerated treatment protocol resulted in a savings of $2,624 per couple and

lowered the per delivery charges by $9,856 compared to conventional treatment

regimens (380). The study concluded that couples with unexplained infertility

who have not achieved pregnancy after three cycles of clomiphene/IUI

should proceed directly to IVF (380).

Assisted Reproductive Technologies (ART) Process

Process of ART

ARTs include IVF, ICSI, gamete intrafallopian transfer (GIFT), zygote

intrafallopian transfer (ZIFT), cryopreserved embryo transfers, and the use of

donor oocytes (396). Because of improved success rates associated with IVF

embryo transfer, the performance of GIFT and ZIFT has declined in the United

States (396), so this review will focus primarily on IVF and ICSI. Both processes

involve the following:

Prevention of a premature LH surge

Follicle growth

Pretreatment

Adjunctive medications

Oocyte maturation/ovulation triggering

Oocyte retrieval

Luteal support

Fertilization by IVF or ICSI

2278In vitro embryo culture

Transfer of fresh embryos

Cryopreservation surplus embryos

First trimester pregnancy monitoring

Baseline Ovarian Cysts

[9] Prior to beginning therapy, a baseline US may be performed on cycle day 2 or

3 during menses (or following GnRH agonist suppression) to confirm an

optimally thin (<4 mm) endometrium and quiescent ovaries. With clomiphene

citrate cycles, the presence of ovarian cysts on a baseline scan was associated

with decreased rates of ovulation but not of pregnancy, although cyst size was not

predictive of response (397). Ovarian cysts on baseline evaluation have been

associated with decreased pregnancy rates in ovulation induction cycles

employing gonadotropins (398). In IVF patients, functional (estrogenproducing) cysts are seen in 9.3% of women following GnRH agonist

suppression (399). Although nonfunctional ovarian cysts up to 5.3 cm did not

affect IVF outcomes (400), functional ovarian cysts (mean diameter 2 cm and

baseline E2 180 pg/mL) have been associated with increased gonadotropin

requirements (dosing and duration), higher cancellation rates, fewer retrieved

oocytes, poorer embryo quality, and lower pregnancy rates per cycle (9.6% vs.

29.7% in cyst-free cycles) (399). Ovarian cysts >20 mm that are seen on the

baseline scan tend to resolve spontaneously within 1 to 2 months and OC

administration does not appear to hasten their resolution (401). However, OC or

progesterone/progestin pretreatment prior to GnRH agonist cycles is associated

with decreased risk of cyst formation (402,403). Functional cyst aspiration prior

to gonadotropin stimulation, which is commonly performed to expedite initiation

of treatment, did not improve IVF outcomes in one large study (399). US-guided

transvaginal cyst aspiration prior to IVF stimulation led to similar pregnancy

outcomes as use of antagonist for cyst resolution though prestimulation antagonist

was associated with slightly longer duration and dose of stimulation (404).

Prevention of Premature Luteinizing Hormone Surge

LH Surge and Premature Luteinization

[9] Without GnRH agonist or antagonist suppression (Fig. 36-5), LH surges occur

in IVF cycles as a result of high E2 levels in the early follicular phase. This, in

turn, results in a lower oocyte yield and reduced pregnancy rates (405).

Spontaneous ovulation prior to oocyte retrieval is reported to occur in 16%

of nonsuppressed IVF cycles (406). The premature LH rise typically occurs after

5 to 7 days of stimulation (405). [9] In contrast, premature luteinization is a

somewhat misleading term that refers to a rise in serum progesterone

2279(cutoffs vary from 0.8 to 2 ng/mL) observed on the day of hCG

administration and occurs in the setting of low LH levels secondary to

gonadotropin-suppressive medications (407). The incidence of premature

luteinization has been estimated at 6% to 7% when the progesterone cut-off level

is set at >1.5 ng/mL (408), but it may be as high as 35% (407). Mechanisms

underlying premature luteinization may include: incomplete pituitary

desensitization, increased granulosa cell receptor sensitivity to LH secondary to

aggressive COH or innately poor responders, and/or the presence of multiple

follicles each producing a normal amount of progesterone consistent with the late

follicular phase (407,408). Although the literature on this subject remains

inconsistent, the negative effects of premature luteinization (409), have been

attributed to endometrial advancement rather than oocyte or embryo dysfunction

(407,408). Methods proposed to address premature luteinization have included

earlier triggering with hCG, less aggressive stimulation protocols,

cryopreservation of embryos (freeze all) (408) and administration of the

antiprogestin mifepristone at the time of hCG (410).

2280FIGURE 36-5 In vitro fertilization protocols using gonadotropin-releasing hormone

(GnRH) agonist or antagonist with gonadotropins in controlled ovarian hyperstimulation.

hCG, human chorionic gonadotropin.

GnRH Agonists

Native GnRH is rapidly degraded in the circulation (411). Commercial

preparations of GnRH agonists consist of decapeptides similar to GnRH but for

modification at two amino acid residues, which increase the half-life and receptor

binding affinities (411). Over the course of 10 to 14 days, agonists initially

2281bind to and upregulate pituitary GnRH receptor activity, leading to a “flare

response” of increased gonadotropin secretion. [9] This is followed by

receptor desensitization (depleted gonadotropin pools along with rapid

uncoupling of the GnRH receptor from its regulatory protein and loss of

signal transduction) which suppresses circulating levels of pituitary

gonadotropins and, with high doses and prolonged use, eventually decreases

GnRH receptor numbers (389,405,411–413). Therefore, prolonged use of

GnRH agonists induces a menopause-like state characterized by low E2 levels

and accompanied by common side effects such as hot flashes and moodiness

(389). The flare effect may cause ovarian cyst formation as described in the

“Unexplained Infertility” section of this chapter (389). GnRH agonists are

commercially available for either depot or daily use and can be administered

intranasally (buserelin and nafarelin) or by intramuscular or subcutaneous

injection (leuprolide, triptorelin, or buserelin). Intranasal preparations have lower

absorption rates when compared to injectable agonists and are associated with

milder suppression (414). Typical starting daily doses of leuprolide are 1 mg, 0.5

mg, or 25 mcg (microdose) (412).

GnRH Agonist Protocols

Some of the more commonly used ART medications and protocols are

summarized in Figure 36-5 and Table 36-7. [9] In the “long protocol,” a

GnRH agonist is started in the luteal phase (day 21) of the previous cycle.

This diminishes the GnRH agonists flare effect and suppresses endogenous

FSH and dominant follicle selection to promote synchronous follicular

growth (405). After 10 to 14 days of GnRH agonist administration, a pelvic

US and E2 level are used to confirm suppression and gonadotropin

stimulation begins. The GnRH agonist is continued (the dose may be halved

or unchanged) throughout the cycle until the hCG trigger (405,412).Mean

serum progesterone levels on the day of hCG administration using the long

protocol are reportedly 0.84 ng/mL (408). [9] The long protocol provides for

better oocyte yields and pregnancy rates in normal responders when

compared with shorter protocols that use later administration or early

cessation of agonists (405). Shorter protocols using lower GnRH agonist

doses have been advocated for poor responders in whom excessive

suppression may be undesirable, perhaps because of direct negative effects of

the agonist on the ovary (412). Long protocols, particularly those using

single-dose depot rather than daily GnRH agonist formulations (415), are

associated with higher gonadotropin dose and duration, which themselves

have been associated with lower pregnancy rates (416,417). Alternatively,

GnRH microdose flare protocols have been developed that may improve

2282oocyte yield in poor responders. Microdose flare regimens involve

pretreatment with 14 to 21 days of combination OCs. Four days following

the cessation of the OC pills, a microdose (25-mcg leuprolide) of agonist is

added in the early follicular phase to take advantage of the agonist flare

effect. Gonadotropin stimulation is initiated 1 to 2 days later while

continuing the agonist (412).

GnRH Antagonists

GnRH antagonists (cetrorelix and ganirelix) were developed by modifying the

GnRH decapeptide at six positions. They compete with endogenous GnRH for

binding to pituitary GnRH receptors (411). [9] Because they have no agonistic

activity, GnRH antagonists lead to almost immediate suppression of FSH

and LH and do not require the additional time for pituitary downregulation

that characterizes the GnRH agonists (411). With prolonged use, GnRH

antagonists downregulate GnRH receptors (411). The only delivery method

available for GnRH antagonist is subcutaneous injection, although orally active

agents are in development (413). GnRH antagonists can be given as 0.25 mg daily

doses or as a single 3-mg dose with no differences in outcomes (413). The time

between daily injections should not exceed 30 hours (389). With single-dose

regimens, avoidance of multiple injections is attractive but additional small doses

starting 4 days after initial dose are required in 10% of cycles (413). Use of

GnRH antagonists in ART protocols is typically begun on days 4 to 7 of

stimulation. [9] This timing balances the risk for premature LH surges with

the need for initiation of endogenous FSH-mediated follicular recruitment

and endogenous E2 production prior to administration (389,405,413).

GnRH Antagonists Fixed Versus Flexible Protocols

Several GnRH antagonist protocols have been developed for use in ART cycles.

Fixed protocols involve starting the antagonist on days 4, 5, 6, or 7 of

stimulation regardless of follicular response (389,395,405,413). Flexible

protocols were developed to reduce gonadotropin stimulation dose and duration

(405). In flexible protocols, varying thresholds have been described for

antagonist initiation. [9] These include addition of the antagonist when the

leading follicle has reached 12 to 16 mm in average diameter or when the E2

level has risen above 600 pg/mL (390,394,405,412,413). When flexible

protocols are used, pregnancy rates are similar when antagonist is initiated on

days 4 or 5 of stimulation but drop significantly when antagonist is initiated after

day 6. This suggests that rapid follicular growth may be more important in

preventing LH surges and improving pregnancy rates than the day the antagonist

is initiated (390). Fixed initiation of a GnRH antagonist on day 6 of stimulation

2283has been associated with higher pregnancy rates when compared to flexible

protocols in one meta-analysis involving four randomized trials, however the

true superiority of one approach over the other remains to be defined

(390,412,413). Diminished E2 levels can be expected following GnRH

antagonist administration, but this does [9] not appear to affect follicular

growth (395). Dose increases in exogenous FSH (389,412) or LH (412,413) do

not appear to be necessary.

GnRH Agonists Compared With Antagonists

Because there is no pituitary desensitization period required when GnRH

antagonists are used in ART, cycles can begin more quickly than those that

employ GnRH agonists (389). Antagonists are associated with lower duration

of stimulation, lower stimulation dose, and reduced rates of ovarian

hyperstimulation syndrome (418). Antagonists are often used in the absence

of pretreatment with OCPs. Initiation of antagonist cycles relies on the start

of spontaneous menses often making scheduling easier for GnRH agonist

cycles (419). Study and protocol heterogeneity make it difficult to detect

consistent differences in IVF pregnancy outcomes after GnRH agonist and

[9] antagonist cycles (418,420). A Cochrane review of 27 randomized

controlled trials indicated that, while the number of good quality transferred

embryos produced are similar, clinical pregnancy rates are higher by 4.7%

in agonist compared to antagonist cycles. They concluded that for every 21

couples, 1 additional pregnancy would be gained by using a GnRH agonist

protocol (418). The number of oocytes retrieved and the live birth rates also

favored agonist usage (418).

Follicular Growth and Stimulation Protocols

Follicular Recruitment

Primordial follicles are 0.03 mm in size and reach the point of ovulation only after

150 days (five menstrual cycles) during which they grow into preantral (primary

and secondary follicles), early antral, small antral, and large antral follicles

(421,422). The resting follicle pool includes primordial and preantral follicles

(421). Resting follicles continuously enter either growth or apoptotic phases from

birth until menopause, a process which is gonadotropin independent (421). After

90 days, activated resting follicles contain a fluid-filled antrum, marking their

responsiveness to gonadotropins and are referred to as early antral follicles of 0.2-

to 0.5-mm size (421,423). Over the next 70 days these early antral follicles grow

into 2 mm small antral follicles, with this slow pace of growth mediated by AMH

(421). Small antral follicles size 2 to 5 mm, which are recruitable and

gonadotropin dependent, begin to slowly enlarge in the late luteal phase of the

2284cycle preceding ovulation while concurrently LH-induced theca cell androgen

production increases (421,424). In the early follicular phase during the menses of

a natural cycle, the follicle that will ultimately become dominant has a diameter

of 5 to 8 mm (425). Growth of the follicle to 10 mm indicates selection, with the

next largest follicle typically 8 mm (424,425). When the antral follicle grows

beyond 10 mm, follicular AMH decreases and FSH induces the appearance of LH

receptors on the granulosa cells (414). [1] At that point, somewhat contrary to the

2-cell-2-gonadotropin model, LH can exert FSH-like actions on the granulosa

cell, including stimulation of aromatase with associated intrafollicular decreasing

androgen and increasing E2 levels (414,421). With complete absence of

endogenous LH, such as that seen with hypothalamic hypogonadism,

exogenous FSH produces fewer preovulatory follicles, inadequate E2, lower

ovulation rates, and thinner endometrium when compared to ovaries

stimulated with agonist downregulation and exogenous hMG (FSH and LH)

(414). [6] Therefore, in the setting of normal hypothalamic function, it seems

likely that even with suppressive therapy, enough endogenous LH is present to

synergize with exogenous FSH to provide adequate stimulation for recruitment

(414).

Follicular Waves and Timing of Stimulation

The above referenced 2 to 5 mm small antral follicles develop synchronously,

cyclically, in cohorts, or “waves,” during each menstrual cycle (423). The

emergence of a wave is detectable ultrasonographically when the lead follicle

reaches 4 to 6 mm, which occurs in association with at least two follicles reaching

≥6 mm (423). The wave is considered major (ovulatory) when over the next ∼ 2

days the lead follicle is ≥10 mm and exceeds other follicles by ≥2 mm, while in

minor (anovulatory) waves the largest follicle is <10 mm and is similar size to

other follicles (423,425,426). In major waves of unstimulated cycles, the

nonovulated follicles become atretic, while in minor waves all follicles in the

cohort become atretic (426). The majority of unstimulated human menstrual

cycles have two waves of follicular development (423,425). In two-wave cycles,

the first wave is minor and has been variously described to occur the day before

ovulation (423,425) or the day after ovulation (426), with progesterone elevations

facilitating follicle atresia (426). [1] The second wave (in two-wave cycles) is

major and occurs 14 to 15 days after ovulation (or the day after menses onset)

(423,425,426). Of the 32% of menstrual cycles with three waves, most of those

involved two minor waves occurring the day of ovulation and 12 to 13 days after

ovulation (or 1 day prior to menses onset) and one major wave occurring 17 days

after ovulation (or 6 days after menses onset) (423,425). However, 19% of the

three-wave cycles had three major waves occurring at each of the same time

2285points (425). For simplification, day 1 of menses (if onset occurred in evening

then day 1 is considered to be the next day) will likely correspond to the

emergence of the ovulatory wave in women with two waves, while day 1 of

menses will correspond to emergence of an anovulatory wave in women with

three waves (423).

Synchronization of Follicle Growth

A small randomized trial in 2012 found women who began stimulation on cycle

day 1 of spontaneous menses had two additional follicles ≥15 mm at time of

trigger compared to the group starting stimulation on cycle day 4; nonsignificant

trends favoring the day 1 start were observed for cycle cancellation rates, number

of total and mature oocytes retrieved (423). [9] During the hormone-free interval

on combination OC pills, the dominant follicles emerged at the 4 to 5 mm

diameter ∼ 3 to 5 days after pill cessation (427) providing support for the practice

of starting gonadotropin stimulation at this time. These findings are intriguing

and suggest a path toward greater synchrony during stimulation. Based upon

findings in cows, one group hypothesized that in women, aspirating the dominant

follicle at ∼ menstrual cycle day 6 to 7 should induce emergence of a new wave

2 days later at which point gonadotropin stimulation can begin to precisely

synchronize the follicle cohort (426). Pharmacologic induction of wave

emergence has been proposed to include high-dose E2 (i.e., a single 8-mg E2V

injection) followed immediately by 5 days of progesterone 100 mg daily given at

any time of the cycle to induce luteolysis and suppress FSH to regress the

dominant follicle (426). Upon cessation of the progesterone, a new wave should

begin with serum FSH surge at which point gonadotropin stimulation can start;

the authors suggest that OCs would not affect wave dynamics but a citation is not

given (426).

Monitoring and Timing of Trigger

Monitoring during ART is helpful to detect the synchronous growth of dominant

follicles, though practitioners vary in specific monitoring elements. [9] Survey

respondents of consistently high-performing United States IVF clinics all used

both E2 and US follicle size for stimulation monitoring, and 40% to 50%

additionally measured progesterone and LH during the stimulation, with

monitoring performed four to six times during the stimulation (428); however, a

Cochrane study in 2014 found that US monitoring alone without serum testing

may be reasonable for monitoring (429).

In one report, women who were predicted to be normal responders using 225 to

300 IU of FSH required 11 days of stimulation, had 11 to 13 follicles ≥15 mm on

the day of hCG trigger, had a peak E2 level of approximately 2,100 pg/mL, and

2286had 10 to 11 oocytes retrieved, of which 82% to 83% were mature (152). [9] The

optimal level of E2 at the time of hCG administration is between 70 and 140

pg/mL per oocyte or follicle (393). Oocytes have been recovered from follicles

with US mean diameters between 13 and 24 mm corresponding to fluid volumes

of >1 mL and ≤7 mL, with optimal recovery rates for 3 mL (430). In a worldwide

survey of IVF practices, of which 18% were in the United States, the majority

administered trigger when 2 to 3 lead follicles (81% of respondents) reached an

US mean diameter of 16 to 20 mm (92% of respondents) (431). High-performing

United States IVF clinics administer trigger after a mean of 9 to 10 days of

stimulation, when at least two lead follicles reach mean diameter ≥18 mm, with

that study reporting a range of 17 mm to >20 mm (428). See the “Superovulation”

section of “Unexplained Infertility” for additional information on gonadotropin

stimulation.

Follicle Stimulating Hormone Versus Human Menopausal Gonadotropin

Concern has been raised that exogenous LH might be needed to address the

sudden decrease in endogenous LH that is associated with GnRH antagonist use

and the possible excess endogenous LH suppression in long GnRH agonist

protocols (414). In a 2010 survey of high-performing United States clinics none

chose an FSH-only protocol for young normal responders (428). An analysis of

>21,000 IVF/ICSI cycles that used antagonist suppression performed in 15 Latin

American countries found hMG and LH protocols were associated with fewer

oocytes retrieved, while pregnancy and live birth rates as compared to FSH-only

protocols were not statistically significantly different (432). [9] This suggests that

less expensive FSH-only protocols may be reasonable in unselected women

(432). The authors did acknowledge, however, that the hMG and LH treated

women were older and had higher rates of ovarian insufficiency than the FSHonly patients (432), so individualizing protocols given extremes of age and

ovarian response is appropriate. In support of individualized protocols is a

study showing hMG in women with elevated AMH >5.2 ng/mL is associated in

agonist and antagonist protocols with fewer instances of retrieving >15 oocytes

and higher fresh transfer live birth rates as compared to FSH only, although

OHSS rates were similar (433). Higher androgen and lower progesterone levels

on the day of hCG trigger are observed when hMG is used for stimulation rather

than FSH alone, indicating a more favorable endocrine profile with hMG

(433,434). The ratio of LH to FSH must be evaluated cumulatively over the entire

stimulation rather than just at the beginning, as there appears to be a “sweet spot”

of 0.3 to 0.6 for the lowest risk of serum progesterone premature elevation ≥1.5

ng/mL (434).

Stimulation Dose and Optimal Number of Retrieved Oocytes

2287Concerns have been raised about premature luteinization and lower pregnancy

rates with higher-dose FSH stimulation protocols (408); each additional 100 IU of

gonadotropins has been associated with a 2% lower rate of clinical pregnancy and

live birth (416), and 300 units daily FSH did not increase number of retrieved

oocytes or live birth rate compared to 225 units in a randomized controlled trial of

“normal responders” (435). A possible explanation stems from comparative

evaluations of unstimulated cycles in which the dominant follicle contains higher

follicular E2, androgen, AMH, and LH concentrations compared to those

measured in stimulated dominant follicles and may indicate follicle quality is

inversely proportional to quantity (436,437). Therefore reduced-dose stimulation

to obtain <8 or ≤4 oocytes using clomiphene citrate 50 to 100 mg for either 5 days

or more extended durations through the day of trigger along with gonadotropin

doses of 75 to 150 units ranging from a few days to longer duration have been

described (437–439). A meta-analysis of >300 minimal stimulation cycles found

that a median of five oocytes retrieved was optimal for pregnancy and live birth

rates (417). These protocols are less expensive and multiple cycles may be

noninferior in terms of live birth rates to higher-dose protocols (440).

Higher-stimulation doses are understood to be a function of patient

characteristics (genetically determined ovarian sensitivity to gonadotropins

or age-related aneuploidy) rather than exerting a direct adverse effect on

IVF outcome and a more nuanced portrait of different ovarian reserve

categories is needed in order to draw conclusions between dose and outcome

(441). [9] Other publications provide compelling support for protocols that

regardless of responder status aim for increased retrieved oocytes in a single

retrieval cycle and has led some to disavow previous rationales for minimal

stimulation (437,442). Sixty percent of high-performing United States IVF

centers in 2010 reported that for a young normal responder with healthy BMI they

would use a relatively high-starting gonadotropin dose of 300 units per day (428).

There appears to be no benefit of exceeding a total daily gonadotropin dose of

450 IU in any patient (389). A previous trial indicating no benefit with higher

gonadotropin doses was notable for nonsignificant trends to higher age, higher

BMI, lower AMH, and lower antral count in the high FSH dose compared to the

lower-dose group (435). A randomized trial of minimal stimulation versus

conventional higher-dose stimulation in 564 first cycles with women age <39

found that while minimal stimulation was associated with 49% cumulative (over

6 months) live birth rates, the conventional stimulation live birth rates were

significantly higher at 63% (439). In fresh oocyte donor cycles, recipient live

birth rates were maximized when >10 oocytes were retrieved (443); each 3.4

mature (M2) oocytes were associated with 1 euploid embryo (444). Among

oocyte donors, embryo aneuploidy rates increased with decreased ovarian

2288sensitivity (i.e., gonadotropin doses per oocyte increased) but increased overall

gonadotropin doses were not associated with aneuploidy (444). In autologous

cycles 15 to 20 oocytes was optimal for fresh transfer live birth (445,446). [9]

Higher numbers of oocytes 15 to 25 were associated with more usable blastocysts

and >1 live birth from a single autologous IVF cycle and each additional oocyte

was associated with an 8% increase in the chance for >1 live births (446). The

likelihood of a single autologous retrieval leading to live birth over subsequent

fresh and frozen/thaw transfers was highest with >15 retrieved oocytes in a

randomized trial of >1,000 women allocated to either antagonist or long agonist

suppression (447). A second retrospective paper reporting on >1,000 women who

underwent a single retrieval cycle similarly found that >15 oocytes retrieved was

associated with the highest cumulative live birth rates with single embryo transfer

compared with fewer numbers of oocytes (448). In >400,000 autologous IVF

cycles, increased number of oocytes >10 were associated with a 3% lower

miscarriage rate compared to cycles with <4 oocytes even after adjusting for age

(449). In poor responders of mean age 38 with AFC 5 to 6 and AMH 0.7 ng/mL,

those randomized to high gonadotropin doses had nearly 10% lower cycle

cancellation rates, 8% lower rates of no oocytes retrieved, two additional mature

oocytes retrieved (4.8 vs. 2.7), more embryos available for transfer (2.7 vs. 1.8),

and a nonsignificant trend to higher pregnancy rate per initiated cycle (438).

Oocyte donors in whom ≥17 oocytes were obtained had 2 to 3 additional euploid

embryos per retrieval cycle compared to cycles with <17 oocytes, and aneuploidy

rate did not increase with gonadotropin dose (444). Although minimal

stimulation protocols by design are simpler with less up-front medication

and freezing services which may contribute to cost, these should be weighed

against the advantages of protocols that aim for safely maximizing per-cycle

live birth efficiency (442). Caution should be exercised when deciding whether

to prolong stimulation duration to obtain additional embryos when trigger-day

serum progesterone levels are >1.49 or >2 ng/mL as these thresholds correlate

respectively with fewer morphologically top-quality blastocyst or day 3 embryos.

Pretreatment

[9] Combined OCs are commonly taken for 14 to 28 days prior to GnRH

analogs to ease cycle scheduling (419), synchronize follicular development

(450), further prevent LH surges (450), reduce the incidence of ovarian cysts

(450), and reduce cancellation rates caused by hyperstimulation (451).

Patients can begin OCs anytime between days 1 and 5 of menses (450). For

antagonist cycles, COH begins 2 to 5 days after OCs (irrespective of menses)

(419). During long GnRH agonist protocols, the agonist overlaps the final 5 days

of OC use, followed by initiation of COH on the second or third day of

2289withdrawal bleeding (402,451). Microdose flare protocols involve pretreatment

with 14 to 21 days of OC, followed 4 days later by a microdose of agonist. COH

typically starts 1 to 2 days later (412). The progestins norethindrone acetate 10

mg orally daily, medroxyprogesterone acetate 10 mg orally daily, or a single

intramuscular dose of progesterone (not specified) can be used in place of an

OC in ART cycles. Recommended duration and the timing of treatment

initiation vary widely. A duration of 5 to 20 days and initiation anytime

between days 1 and 19 of menses have been reported (450). Four mg of daily

micronized 17β E2 or E2 valerate, has been used in lieu of OCs in ART cycles,

with initiation between cycle days 15 and 21 and a duration of 10 to 15 days

(450). OC pretreatment for antagonist cycles has been shown to increase the

duration and number of medication doses during stimulation. There is “moderate

quality” evidence from a recent Cochrane review for reduced ongoing pregnancy

and live birth rates with a possible improvement in pregnancy loss rates (403).

However, OC pretreatment during GnRH agonist cycles is associated with higher

pregnancy rates than those in cycles without pretreatment (402). Progestin and E2

pretreatment do not conclusively affect live birth rates in either agonist or

antagonist cycles (403). They may be useful to synchronize stimulated follicle

growth (see “Follicular Growth and Stimulation Protocols”).

Adjunctive Medications

Prenatal vitamins should be given to all infertility patients beginning at least

1 month prior to initiation of infertility treatment. Low-dose aspirin 80 to 100

mg daily commencing at downregulation and given for variable durations (up to

13 weeks gestation or if recurrent loss to 38 weeks) is commonly used during IVF

regimens to improve ovarian stimulation response and endometrial implantation.

In some recurrent loss patients low-dose aspirin is given in conjunction with

heparin (452). A 2016 Cochrane review of “low to moderate quality” RCT

evidence comprising >2,600 patients found no evidence of benefit for routine IVF

use in terms of pregnancy, miscarriage, or live birth (453). Aspirin use confers a

fourfold increased risk (50% vs. 13.6% in nonusers) of first trimester

subchorionic hemorrhage in non-IVF and IVF pregnancies, which was not

observed with heparin use (452). These findings prompted the authors to

discontinue the use of aspirin in routine infertility and IVF patients, reserving it

for antiphospholipid syndrome and recurrent loss patients. They will discontinue

the aspirin if subchorionic bleeding is detected (452). Glucocorticoids given to

women during the peri-implantation period may improve pregnancy rates in

women undergoing IVF (rather than ICSI). A Cochrane study found this result of

marginal statistical significance and did not observe improvement in terms of

pregnancy or live birth rates for glucocorticoids given to an overall sample of

22901,800 patients. More study is needed before recommendations to specific

subgroups (i.e., autoantibodies, unexplained infertility, or recurrent implantation

failure [RIF]) (454). In 2015 one group discontinued their routine use of oral

doses of methylprednisolone (16 mg daily) and doxycycline (100 mg twice daily)

for 4 days commencing either on the day of oocyte retrieval or 4 days prior to

frozen ET, because they did not find that these medications statistically beneficial

in terms of pregnancy, miscarriage, or live birth rates regardless of whether ICSI

and assisted hatching were performed (455). Metformin may limit OHSS in

PCOS patients (456). An updated Cochrane review found that metformin

improved clinical pregnancy rates with IVF, but showed no conclusive benefit for

live birth rates (457).

Physiology of Oocyte Maturation

Prior to maturation, oocytes are arrested in the prophase stage of meiosis I, also

known as the germinal vesicle stage (458,459). Meiosis I oocytes must reach at

least the early antral follicle stage to respond to FSH and be competent to resume

meiosis (460). In vivo, LH receptors on the follicle are induced by FSH during

later stages of follicular development (460). Therefore, only fully grown oocytes

respond to the LH surge in vivo to begin the cytoplasmic and nuclear maturation

that are required for developmental progression toward the metaphase stage of

meiosis II. At this point, the developmentally competent oocyte will extrude the

first polar body, the oocyte–cumulus complex will detach from the ovarian wall

(461), ovulation will occur, and fertilization is possible (458–460).

Trigger During ART Cycles

Because spontaneous LH surges occur inconsistently during non-ART

gonadotropin cycles and are suppressed in ART cycles, hCG has been used

to trigger ovulation. In combination with its long half-life, the homology

between hCG and LH (identical α subunits) permits cross-reactivity with the

LH receptor and induction of final oocyte maturation and ovulation (462).

[9] hCG is derived from urine (5,000 to 10,000 IU intramuscularly) or through

recombinant technology (250 μg subcutaneously, equivalent to 5,000 to 6,000 IU

of intramuscular urinary product) (228). The half-life of hCG is 2.32 days,

compared to 1 to 5 hours for LH (462). Ovulation is typically triggered when at

least 2 follicles are ê17 to 18 mm in average diameter (but <24 mm) and the

endometrial thickness is ê8 mm (152,381,390,393–395,463). Similar clinical

outcomes have been noted when 5,000 IU or 10,000 IU of urine-derived hCG

(464) and urinary or recombinant preparations (463,465) are used for triggering.

If there is concern for OHSS, GnRH agonists can substitute for hCG to trigger

ovulation in antagonist protocols.

2291Oocyte Retrieval

Oocyte retrieval is performed via transvaginal US–guided needle puncture into

each follicle and aspiration of follicular fluid. Either general anesthesia or

intravenous conscious sedation may be used (466). Prophylactic antibiotics such

as ceftriaxone are recommended at the time of retrieval (467). The vaginal prep

can be performed either with sterile saline alone or with povidone-iodine and

vigorous saline flushing (468,469). The highest oocyte yield is obtained when

oocyte retrieval is performed 36 to 37 hours after the hCG injection (463).

Earlier retrieval (35 hours) is associated with a much lower oocyte yield, and later

retrieval risks ovulation. Spontaneous follicular rupture appears to occur at a

mean of 38.3 hours following hCG administration (463).

Luteal Support

The rationale and regimens for luteal phase support with progesterone are

discussed in the “Uterine Factor” section. Luteal phase E2 supplementation is not

necessary for fresh embryo transfers unless GnRH agonist trigger is utilized

(339,470,471). In the case of GnRH agonist trigger, E2 0.1-mg patches starting 1-

day postretrieval coadministered with intramuscular progesterone has been

proposed to offset corpus luteum dysfunction (339). [9] Luteal support with lowdose hCG 1,000 to 1,500 units given at egg retrieval and again up to 5 days

postretrieval has been used successfully (339).

Fertilization by IVF or ICSI

Following semen collection and sperm processing (described in the “Male Factor”

section), sperm are incubated in media for 3 to 4 hours to promote sperm

capacitation and the acrosome reaction. Before fertilization, retrieved oocytes are

cultured in media. Conventional IVF involves insemination concentrations of

100,000 to 800,000 motile sperm/mL per oocyte with each oocyte in a small

droplet of media under oil (11,472,473). [9] For every three cycles done for

severe male factor, the use of ICSI prevents one case of fertilization failure when

compared to conventional IVF (11). The indications for, procedure, and risks of

ICSI are discussed under “Male Factor.”

In Vitro Embryo Culture

Embryo Development

Initial embryo development is typically assessed 15 to 20 hours after

insemination or ICSI, when fertilization is characterized by the presence of

two pronuclei and the extrusion of the second polar body (11,473,474). [9]

Embryos are examined again for cleavage after 24 to 30 hours of culture (11). The

2292first embryo cleavage occurs approximately 21 hours after fertilization, and

subsequent divisions occur every 12 to 15 hours up to the 8-cell stage on the 3rd

day of embryo development (475). Compaction to form the 16-cell morula occurs

on the 4th day of embryo development, and differentiation of the inner cell mass

and trophectoderm to form a blastocyst (containing a fluid-filled area called a

blastocoel) is completed by the 5th to 6th days (476,477).

Culture Environment

Prior to compaction, while the embryo is under genetic control of the oocyte, it

uses a pyruvate-based metabolism. It requires at least a few amino acids, and

prefers a relatively oxygenated environment (though much lower than

atmospheric oxygen) similar to that found in the fallopian tube (475,478). After

compaction, glucose and amino acid needs increase, the embryonic genome is

activated, and metabolism requires a very low oxygen environment similar to

that found in the uterus (475,478). Supplementation of culture media with

hyaluron and albumin is beneficial in the postcompaction period (475). Culture

media systems are classified as either single or sequential. Single culture media

was developed to provide all the necessary nutrients required (day 1 through day

5/6) in a single media formulation, reducing stress on the embryo by keeping a

stable environment and allowing any embryo-generated autocrine/paracrine

factors to remain (479). Single culture media uses stable dipeptide glutamine to

avoid toxic ammonium buildup from the breakdown of amino acids (479).

Sequential culture media systems were designed to mimic the changing metabolic

and nutritional requirements of the developing embryo by culturing embryos in

media designed for the cleavage-stage embryo (days 1 to 3) and on day 3

switching to culture media designed to support blastocyst development. Although

both culture systems have advantages, there is insufficient evidence to

recommend one over the other, particularly as to whether any benefit of

increased blastocyst development translates into improved pregnancy or live

birth rates (479).

Extended Culture to Blastocyst

Although precompaction human embryos can survive when placed in the uterus,

the uterine cavity is a nonphysiologic location for them and there is greater

uterine pulsatility during this period that may cause the embryos to be expelled

(480). Therefore, the blastocyst stage represents a more physiologic time for

embryo transfer (480). Because nearly 60% of morphologically normal cleavage

embryos, but only 30% of blastocysts, are chromosomally abnormal, extended

culture allows for better selection of embryos with improved quality (480,481).

Notably, if a patient is a carrier of a balanced translocation, nearly 80% of

2293cleavage embryos and 60% of blastocysts may be chromosomally abnormal

(482). The euploidy rate of blastocysts is higher with completed development by

postretrieval day 5 compared to day 6 or day 7.

Blastocyst Versus Cleavage Transfer Outcomes

Comparisons involving equal numbers of transferred embryos demonstrate

that fresh blastocyst transfer is associated with up to 10% higher pregnancy

rate and up to 13% higher live birth rate than cleavage-stage transfer, which

can be helpful information if considering elective single embryo transfer.

Drawbacks to extended culture include the small possibility that no embryos will

survive to fresh blastocyst transfer and a 23% reduction in women having

additional embryos for cryopreservation. However, ovarian reserve with AMH >1

ng/mL strongly positively influences the presence of supernumerary blastocysts

available for cryopreservation. Monozygotic twinning rates may be higher with

blastocyst culture, although this has not been a consistent finding.

Criteria for Extended Culture

There are no established guidelines or criteria that determine when to utilize

extended culture (483). Varying suggestions include: maternal age ≤42 with ≥

five 2-pronuclear (2PN) stage embryos on postretrieval day 1; maternal age of

≤40 and ≥ three good-quality day 3 embryos having 4 to 10 cells with <15%

fragmentation; maternal age of ≤41 to 42 and ≥ four good-quality day 3 embryos

having 4 to 10 cells with <15% fragmentation; and age <37 with ≥4

morphologically good embryos on day 3 or ≥4 embryos with 6 cells and <10%

fragmentation (474,476,483).

Embryo Transfer

Embryo Morphology

[9] Embryo morphology guides the choice of embryo for transfer. Pronuclear

embryos are assessed by their distribution and number of nucleoli, the position of

the second polar body relative to the first, and cleavage rates (abnormal rates are

too fast, too slow, or arrested) (477). Preferred cleavage-stage embryos have a

normal developmental pattern characterized by early cleavage on day 1, 4 cells on

day 2, and 8 cells on day 3. Embryo fragmentation should be ≤10%, the

blastomere size should be regular and there should be no multinucleation (474).

The Gardner and Schoolcraft system for scoring blastocysts uses a scale from 1

(worst) to 6 (best), with grades 1 to 3 indicating growth of the blastocele until it

completely fills the embryo. Grade 4 blastocysts are expanded with a larger

blastocele volume and a thinning zona pellucida. The trophectoderm in a grade 5

blastocyst is starting to hatch though the zona and the grade 6 blastocyst has

2294completely escaped or hatched from the zona (473). The inner cell mass is graded

A–C based on tightness and cellularity (A is best), and the trophectoderm is

assessed from A–C based upon cohesiveness and cellularity (A is best) (473).

Conventional embryo morphology grading remains the most common

methodology in choosing embryos to transfer but it has limitations. In order to

assess development and morphology of embryos, handling of the embryos and

removing them from the controlled environment of the incubator is required. This

exposure leads to differences in oxygen levels, temperature and pH which may

have adverse effects on the development of the embryos (484). Understandably

conventional morphology assessment is limited to specific points in time to avoid

repeated exposure to the atmospheric environment (484). Time-lapse imaging has

been developed by installing a camera in the incubator allowing for continuous

undisturbed evaluation of the embryos. In addition to the ability to assess

morphology at multiple time points, morphokinetic monitoring, which assesses

the rate at which embryos reach developmental events, can be accomplished with

time-lapse imaging. Prior results of randomized studies comparing clinical

outcomes of time-lapse monitoring versus conventional morphology have been

mixed (484–486). Although there appears to be a potential benefit in the

application of time-lapse imaging, more high-quality, well-designed randomized

control trials need to be performed before any definitive conclusions can be

drawn (484–486).

Number of Embryos to Transfer

Multiple gestation pregnancies increase complications for mothers and fetuses, so

guidelines have been developed to minimize this adverse outcome (487). [10]

Single embryo transfer should be the standard if an embryo is euploid

regardless of the age of the patient or stage of the embryo (487). Single

embryo transfer is recommended for patients <age 38 undergoing a fresh cycle, if

they have had a prior live birth or there is an expectation that at least one highquality embryo will be available for cryopreservation (487). Patients < age 38

undergoing frozen embryo transfer (FET) should have a single embryo transferred

if vitrified blastocysts are available, if it is the patient’s first FET, or if the patient

had a previous live birth (487). In all other cases, transfer of embryos should be

limited to two embryos in women under the age of 35. In older women, the

maximum number of transferred cleavage-stage embryos should be: 3 in women

aged 35 to 37, 4 in women aged 38 to 40, and 5 in women >40 years of age (487).

Because of their high implantation potential, no more than three blastocysts

should be transferred to any woman regardless of her age (487). [10] Limits on

the number of embryos transferred when the embryos were created from donor

oocytes should be based on the age of the donor, rather than the recipient (487).

2295Transfer Procedure

[9] The goal of transcervical embryo transfer is to atraumatically deliver the

embryos to an optimal intrauterine location for implantation. Prior to

performing the embryo transfer the cervix and vagina are cleansed using media or

saline. The utility of removing remaining cervical mucus appears to be beneficial

in improving live birth and clinical pregnancy rates (488). Embryo transfer can be

accomplished using a number of transcervical techniques. Direct transfer is

performed by loading the transfer catheter with embryo(s) and transferring

immediately without a trial transfer. Trial transfer permits assessment of the

cervical canal prior to performing the actual transfer. Trial transfer can be

combined with an afterload technique if the trial transfer is difficult. The outer

sheath of the trial transfer catheter is left in place while removing the inner

catheter and a new inner transfer catheter loaded with the embryo(s) is threaded

through the trial catheter sheath into the uterus. Afterload technique can be

planned and utilized intentionally without a trial transfer. Transabdominal US

should be utilized to assess the endometrial cavity and other pelvic structures

prior to and during the transfer to assure deposit of the embryo(s) in the upper or

middle area of the uterine cavity but no closer than 1 cm from the fundus (488).

[9] Implantation is more likely when using a soft catheter and when fundal

contact is avoided (488). When the embryos are deposited, the entire catheter

should be removed immediately as there does not appear to be any benefit in

delaying catheter withdrawal (488). After the transfer, the catheter is checked for

retained embryos. If present, retained embryos should be immediately

retransferred as there is no detriment in pregnancy rates (488). When the embryo

transfer is completed, the patient may leave immediately without a period of bed

rest (488). Although intrauterine infections decrease pregnancy rates, the efficacy

of antibiotic administration at the time of transfer is not supported (488).

Cryopreservation of Embryos

Embryo cryopreservation at the pronuclear, cleavage, and blastocyst stages has

permitted multiple transfer cycles from a single oocyte retrieval. Because transfer

of cryopreserved embryos is less expensive than a second fresh cycle, overall

fertility treatment costs can be optimized. Techniques for embryo

cryopreservation include slow freezing and rapid freezing or vitrification. Slow

freezing protocols use lower concentrations of cryoprotectants but are more timeconsuming when compared to vitrification, which uses high-concentration

cryoprotectants for rapid cooling and is less expensive (489). Embryo thawing is

accomplished by brief exposure to air and warm water followed by rehydration

(490). Although pregnancy rates for frozen embryo transfer (FET) cycles using

the two cryopreservation methods are similar, vitrification is associated with

2296higher postthaw embryo survival (93% vs. 76% with slow freezing) (489). [9]

Favorable postthaw embryo survival rates after vitrification has contributed

to the increasing success of FET, with live birth rates that are no different

than fresh embryo transfer (491). The success of FET has led to a growing

trend toward freezing entire embryo cohorts for transfer at a later date. This

strategy has the added advantage of greatly reducing the risk of ovarian

hyperstimulation syndrome with reassuring perinatal and obstetric outcomes

(492). Although a number of studies report a favorable live birth rate per transfer

when using a freeze-all strategy compared with fresh transfer, cumulative birth

rates have not been shown to be different (493). Particularly in PCOS patients, a

freeze-all strategy with subsequent warmed embryo transfer improved live birth

rates compared to fresh transfer, though an increase in preeclampsia and singleton

infant birth weight were noted. Well-designed randomized controlled trials are

needed before recommending routine use of a freeze-all strategy (493).

Endometrial Preparation for Frozen Embryo Transfer

When FET is combined with a recipient’s natural cycle, no exogenous treatment

is given and transfer is timed to spontaneous ovulation, performed 7 days

following detection of the serum LH surge (333). [9] In medicated FET cycles, E2

supplementation begins in the early follicular phase and is continued for up to 13

to 15 days prior to initiating progesterone (490). Multiple E2 preparations have

been described for use in FET cycles, but none have been proven superior (494).

Transvaginal US is used to assess endometrial thickness during estrogen therapy

and estrogen administration continues until an optimal thickness of ≥6.5 to 8 mm

is reached (333,490). Progesterone supplementation begins 48 to 72 hours prior to

transfer when cleavage-stage embryos are used and 6 to 7 days prior to transfer

when blastocysts will be thawed (490,494) unless different progesterone timing is

indicated (333). Several progesterone preparations have been described for use in

FET cycles, but none have been proven superior (“Uterine Factor”) (494).

Medicated FET cycles offer control and flexibility of transfer timing while

maintaining low cancellation rates; however, adequate pituitary suppression

may not always be achieved (495). [9] A GnRH agonist may be used in

conjunction with medicated FET cycles as an added measure to prevent

premature LH surges that may adversely affect endometrial maturation

(495). Coordination between the embryo and endometrium is critical to successful

implantation (333). For some women, this remains elusive resulting in RIF

defined as failure of at least three IVF cycles in which one or two

morphologically high-grade embryos are transferred (333). Prior studies of the

transcriptome of the endometrium have shown the ability to accurately

catalog the endometrium in all phases of the menstrual cycle and have

2297challenged the assumption that the window in which the endometrium is

receptive to embryo implantation is universal in all women [9] (333). The

address the optimal window of implantation in patients with RIF through the

use of a molecular diagnostic tool (333). While the initial reproductive outcomes

incorporating a larger population are needed to validate these results (496).

First Trimester Pregnancy Monitoring

hCG including the a subunit can be detected in spent culture media from the

2PN-stage onward. The production of hCG by the blastocyst can be detected

in the serum as early as 7 days posttransfer, and serum quantitative hCG

levels may be obtained 11 to 14 days following embryo transfer (490,497). [9]

A serum threshold of 200 mIU/mL of hCG measured 12 days after transfer is

92% and 80% predictive of ongoing pregnancies for day 3 and day 5

embryos respectively, with levels normally rising approximately 40% per

day (497). If normally rising hCG levels are detected, transvaginal US is

planned at 6 to 7 weeks of gestation to determine the location of the

pregnancy, the number of gestational sacs, and pregnancy viability (497). [9]

In those at high risk for ectopic pregnancy, early diagnosis using serial hCG

and US examinations are important to reduce the risk of a ruptured ectopic

pregnancy and to improve the success of medical management (498). These

patients may benefit from transvaginal US at 5.5- to 6-weeks gestation as a

gestational sac should become visible in a normal pregnancy during this time

(498). When concern arises as to pregnancy viability, serial US findings of

growth arrest with fetal pole length <20 mm at time of fetal demise correlate

with aneuploidy of conceptus.

[12] TABLE 36-9 Live Birth Rates from IVF Cycles Using Patient’s Own Eggs and

Uterus

ART SUCCESS RATES

IVF can be a highly effective treatment for many patients, but fewer than 200,000

2298IVF cycles are performed per year. The low IVF utilization rates are thought to be

a result of the high cost of IVF and low retention rates among patients after an

unsuccessful first IVF cycle. Providers have recommended counseling patients to

consider IVF as a course of treatment, if needed, in order to maximize each

patient’s probability of having a live birth from IVF (499). A patient’s decision to

pursue medical treatment for infertility is often impacted by personal, financial,

and medical factors. The provider’s role is to establish the clinical diagnoses,

discuss the risks and benefits of various treatment options, and recommend

treatment. If the treatment options include COH and IVF (such as in unexplained

infertility), providers would typically counsel the patients about the success rates

of IUI (with clomid or gonadotropins) and IVF in the context of the patient’s

clinical data and diagnoses. Patients who are paying out-of-pocket for IVF

treatment can benefit from a thorough discussion about the cost of each treatment

option in the context of the per-cycle cost and the number of treatments that may

be needed for each option. [1] Because the latter is directly related to the patient’s

per-cycle treatment success, understanding the relationship between the

personalized IVF success probability and potential cost of IVF can help patients

plan and maximize their chances of having a live birth (baby). Public reporting of

each IVF center’s success rates has been available for more than two decades.

There have been significant efforts to incorporate predictive modeling and

technology to provide patient-centric probabilities of IVF success based on each

patient’s own health data.

National IVF Outcomes Registry

Since 1992, all clinics performing ART in the United States have been required to

submit IVF cycle data and outcomes annually to the CDC via the National

Assisted Reproductive Technology Surveillance System (NASS) which publishes

reports of IVF outcomes analyses (500–504). The Society for Assisted

Reproductive Technology (SART) via the Clinic Outcome Reporting System

(CORS) publishes a summary of each registered ART clinic’s IVF outcomes at

https://www.sartcorsonline.com/rptCSR_PublicMultYear.aspx?

reportingYear=2015 (139). For most purposes, ART cycles using ICSI and IVF

are combined and analyzed as IVF cycles because their success rates are similar.

IVF success rates, largely dependent on maternal age, are conventionally

summarized from NASS/CDC and SART/CORS in an age-based format as shown

in Table 36-9. SART asks that its reports not be used to compare IVF centers,

presumably because success rates can be impacted by patient demographics and

clinical attributes which vary among IVF centers. SART has been reporting, since

2014, the parameter of cumulative live birth rate, which reflects the chance of

achieving a live birth after a fresh or FET within a year of cycle initiated for egg

2299retrieval (139). [12] This allows tracking individual outcomes over time,

accounting for fresh and frozen transfers stemming from the retrieval (139).

Prediction Models Personalize IVF Success

The ability to provide highly accurate and personalized probabilities of IVF

success benefits the individual patient and gives providers insight about

prognostic groups and profiles in their patient population. IVF success prediction

models are not new, but the use of machine learning to develop and validate

prediction models has dramatically improved model performance as defined by

prediction accuracy, the ability to rank patients based on the difference in success

probabilities using area-under-the-curve (AUC) in a receiver-operating curve

analysis, and the percentage of patients who would receive a significantly

different probability compared to conventional age-based prognostic classification

(505).

Of the clinical data available prior to treatment, the patient’s age is a key

predictor of COH or IVF success and accounts for over 50% of relative

importance of quantifiable predictors of IVF success (150,505,506). However,

many other clinical predictors—ovarian reserve, BMI, reproductive history,

clinical diagnoses, sperm parameters, total amount of gonadotropins used, oocyte

parameters, and embryo scores—impact and predict treatment success. Fertility

providers are interested in using data and technology to improve the patient’s

experience by offering personalized IVF success prognostics. First, providers

should define their goals. The intended goal of the prediction model determines

the types of clinical predictors to be used by the model. For example, a prediction

model that aims to inform a patient’s decision to do her first IVF cycle should not

use clinical predictors that can only be known after she starts IVF, such as the

amount of gonadotropin used or oocyte and embryo parameters (150,506). A

prediction model that aims to inform a patient’s decision to do eSET should use

clinical predictors—such as the number of blastocysts—that are available up to

the day of embryo transfer (507).

When choosing a prediction model, it is important to consider whether the

dataset used to develop and test the model reflects the patient population to which

the model will be applied. Specifically, several researchers continue to raise the

concern that a model developed and tested in one center may not have the same

performance or accuracy when applied to another center, or rigorous steps may be

required to test and recalibrate the model (508–511). This problem can be

partially addressed by using a prediction model developed from data pooled from

multiple centers or the national registry, though the applicability of the model to

each individual center still needs to be tested (506,512–514). Alternatively,

center-specific prediction models can be developed and tested based on each

2300center’s own data (150,505,507,515).

Cessation of Therapy

Patients must be accurately informed of estimated success rates and reasonable

expectations for all therapeutic interventions (516). Patients with a very poor

prognosis have a 2% to 5% chance of achieving a live birth with fertility therapy

and those with a futile prognosis have a ≤1% chance (516). Refusing or limiting

therapy may be justified if the risk of intervention outweighs the potential benefits

(516).

Third Party Reproduction

When gametes and/or the ability to gestate a pregnancy are compromised through

circumstances or disease, other reproductive options can be considered. These

include use of donor sperm (“Male Factor”), donor oocytes, donor embryos,

a gestational carrier, or a combination of these approaches. In contrast to

donor gametes (sperm or oocytes), the decision to donate embryos is typically

made after the embryos have been generated and there is a known surplus. A

gestational carrier receives and gestates birth embryos created from the intended

mother’s oocytes. Patients who choose to utilize a gestational carrier may have

irreparable uterine factor infertility or suffer from medical conditions that

contraindicate pregnancy. In true surrogacy, the birth mother is also the genetic

mother but not the mother who will raise the child. Legal and psychosocial

counseling are suggested for all parties embarking on any form of third party

reproduction.

Donor Oocyte

[9] Patients with premature ovarian insufficiency, poor oocyte quality, poor

ovarian response to stimulation, or failed fertilization or implantation after

multiple ART cycles may be candidates to receive donated oocytes. A female

same-sex couple may choose to have one partner undergo IVF and place the

resulting oocytes fertilized with donor sperm into the other partner (517). With

carefully selected donors using nonfrozen oocytes, live birth rates per cycle of

donor oocyte IVF are 50% to 60% or higher regardless of the recipient’s age

(138,139). Recipients must be aware that advanced recipient age is associated

with higher risk for preeclampsia, diabetes, and cesarean section (518). Oocyte

donors must endure all the interventions and risks of the ART process, except for

embryo transfer and luteal support. Because of the intensity of therapy and the

potential infectious disease and genetic risks for donor, recipient, and the resulting

offspring, oocyte donors must be screened for infectious and heritable disorders

2301similar to those performed for sperm donors (“Male Factor”), and undergo

meticulous informed consent and a comprehensive psychosocial evaluation (519).

Oocyte donors may be anonymous or known to the recipient (519). Oocyte

recipients undergo endometrial preparation as described in the “ART Process”

section, which may be synchronized with the donor for a fresh transfer (Fig. 36-6)

or may be dissociated from the donor retrieval if using cryopreserved oocytes or

embryos. Other topics, such as methods for donor recruitment and financial

compensation for the donor are challenging issues (520).

Oocyte Cryopreservation

Historically, oocyte donation was restricted to the use of fresh oocytes, requiring

coordination between the donor and recipient (519). [9] With the continued

evolution of cryopreservation technology, oocyte cryopreservation is a viable

alternative and is no longer considered an experimental technique (519).

Oocyte cryopreservation techniques have resulted in excellent oocyte survival

after vitrification and warming (519). Subsequent fertilization and pregnancy

rates have been similar to patients undergoing IVF/ICSI with fresh oocytes (519),

though live birth rates may be somewhat reduced (139). The utilization of

cryopreserved oocytes has the potential to offer greater flexibility in pregnancy

timing, more choices in the selection of a donor, and reduced cost (519). Multiple

applications of this technology have been proposed to include fertility

preservation in patients receiving gonadotoxic therapies, patients with genetic

conditions, patients who are unable to cryopreserve embryos, and patients who

want to electively defer childbearing (519). Oocytes can be cryopreserved when

there is insufficient sperm for fertilization at the time of oocyte retrieval (519).

Although preliminary data on perinatal outcomes are reassuring, more data are

needed before routine oocyte cryopreservation and universal donor oocyte

banking can be recommended (519).

Complications of Assisted Reproductive Technology

Risks of ICSI

Risks of ICSI are detailed in “Male Factor.”

Cycle Cancellation

Criteria for cycle cancellation vary; high-performing United States clinics

reported a threshold of ≤3 follicles (up to <5 follicles) (428). [9] Cycle

cancellation in normal responders occurred in up to 6% of the cycles as a

result of inadequate response and 1.5% of cycles for excessive response (152),

2302and risk increases with older age and lower ovarian reserve (416). In 0.2% to

7% of retrievals, no oocytes will be obtained (461). Two proposed explanations

include human error during the administration of hCG and early oocyte atresia

despite normal follicular response (461).

2303FIGURE 36-6 Regimens of ovarian stimulation and hormone replacement used to

synchronize the development of ovarian follicles in the oocyte donor and the endometrial

cycle in the recipient. hCG, human chorionic gonadotropin; OCP, oral contraceptive pill;

GnRH-a, gonadotropin releasing hormone agonist; TVA, ultrasound-guided transvaginal

aspiration of oocytes. (Adapted from Chang PL, Sauer MY. Assisted reproductive

techniques. Stenchever MA, ed. Atlas of Clinical Gynecology, Mishell DR, ed.

Reproductive Endocrinology. Vol. 3. Philadelphia, PA: Current Sciences Group; 1998,

with permission.)

2304Oocyte Retrieval

The risks of oocyte retrieval include bleeding requiring transfusion, injury to

adjacent structures requiring laparotomy, formation of a pelvic abscess leading to

loss of reproductive function despite prophylaxis, and risks related to anesthesia

(521).

Multiple Gestation

As the majority of ART cycles involve the transfer of more than one embryo,

multiple gestation occurs at higher rates than for spontaneous conception (3%).

This has social, medical, emotional, and financial ramifications (501,522). [10]

Although the majority of complications of multiple gestations occur with

high-order (ê3) multiples, twins have increased risks for low birth weight,

preterm birth, and neurologic deficits when compared to singletons

(522,523). [10] Despite this, 20% of infertile patients view multiple pregnancy as

a desired outcome (501). Because most patients lack insurance coverage for IVF,

patients may push their physicians to take greater risks when deciding on the

number of embryos to transfer (501). The multiple pregnancy risk is higher in

patients under the age of 35 who are undergoing IVF since their embryos are

typically of better quality and implantation rates are higher, but multiple

pregnancy can occur at any reproductive age (138,139). With increasing

adherence to guidelines suggesting limitations on the number of embryos to

transfer in a given ART cycle and improved implantation rates that allow single

embryo transfer, the multiple infant live birth rate has decreased in most age

groups from 2006 through 2015, occurring in 24% in women aged <35, 23% aged

35 to 37, and 20% age 38 to 40 (138). However, the multiple live birth rate

increased in that same time period slightly to 18% aged 41 to 42 and 12% aged

>42 (138). [10] Most multiple pregnancies arising from ART are dizygotic, but

monozygotic twinning occurs in 3.2% of IVF cycles (compared to a background

rate of 0.4%) (501). Concerns have been raised that monozygotic twinning might

increase after blastocyst culture, but this finding has not been consistent

(476,524).

Selective Reduction

Ten percent of multifetal pregnancies spontaneously lose at least one gestational

sac during the first trimester. This loss rate increases to 21% in women >35 years

of age (525). The spontaneous reduction rate to twins or singletons is much higher

for triplets (14%) (526) than for quadruplets (3.5%) [10] (525). Selective

pregnancy termination or multifetal reduction may be an option for some

patients in whom spontaneous reduction does not occur by 11 to 13 weeks

(527). This involves first karyotyping each fetus through transabdominal

2305chorionic villous sampling to preferentially reduce those that are abnormal, then

injecting potassium chloride into the heart of the targeted fetus (527). Although

selective reduction carries a 3% to 7% risk of losing the entire pregnancy when

performed prior to 19 weeks, this is still lower than the 15% chance of

spontaneously losing an entire triplet pregnancy (523,527). Selective fetal

reduction is typically considered for triplet or higher pregnancies, but even

reduction from twins to singletons may have benefits (523).

Ectopic and Heterotopic Pregnancy

ART pregnancies effectively bypass the fallopian tubes to achieve pregnancy but

ectopic pregnancy (implantation outside the uterus) is still possible. Historically

rates of ectopic pregnancy were higher in ART when compared to spontaneous

conception, although data have shown [11] rates of ectopic pregnancy in ART to

be similar to the general population rate of 2% (528). The absence of an

intrauterine pregnancy on transvaginal US evaluation in conjunction with a

maternal serum hCG level above a threshold of 1,500 to 2,500 mIU/mL

suggests an abnormal gestation (498). This threshold or discriminatory zone is

individualized to each institution depending on the characteristics of the hCG

assay used and clinical expertise. A more conservative discriminatory zone

decreases the risk of incorrectly diagnosing an ectopic pregnancy and terminating

a viable pregnancy (498). [11] This is especially important in multiple gestation

which will produce higher hCG levels at an earlier stage (498). Heterotopic

pregnancy involves concurrent intrauterine and ectopic pregnancy, usually within

the tube (529) but ovarian implantations have been reported (530). [11] The

incidence of heterotopic pregnancy, which is extremely rare in spontaneous

conceptions, is notably higher after IVF treatment with an incidence as high as

1% (529). Risk factors for heterotopic pregnancy include multiple gestation,

smoking, previous tubal surgery, and prior PID, in addition to ART (529). As

with standard ectopic pregnancies, pain and bleeding are the most common

presenting findings with heterotopic pregnancies (529). Heterotopic pregnancies

are most often diagnosed in the first 5 to 8 weeks of gestation using laparoscopy

or laparotomy (529). Only 26% of heterotopic cases can be diagnosed with

transvaginal US, possibly because of difficulties in sonographic

interpretation in the presence of concomitant ovarian hyperstimulation

(529). After treatment of a heterotopic gestation with laparoscopy, laparotomy, or

US-guided injection of potassium chloride into the extrauterine pregnancy, the

overall delivery rate for the intrauterine pregnancy is nearly 70% (529).

Ovarian Hyperstimulation Syndrome

OHSS is a medical complication that is completely iatrogenic and unique to

2306stimulatory infertility treatment (531). Its symptoms are the result of ovarian

enlargement and fragility, extravascular fluid accumulation, and

intravascular volume depletion (532). Proposed mechanisms for the

characteristic fluid shifts that accompany OHSS include increased protein-rich

fluid secretion from the stimulated ovaries, increased renin and prorenin within

follicular fluid, and increased capillary permeability mediated by angiotensin

(532). [11] Vascular endothelial growth factor (VEGF), whose expression in

granulosa cells and serum is augmented by hCG, and a variety of other

inflammatory cytokines have been implicated in the pathogenesis of this

disease (532). Two distinct patterns of OHSS onset have been described. Early

OHSS occurs 3 to 7 days following the hCG trigger and is associated with the

administration of exogenous hCG (533). [11] Late-onset disease occurs 12 to 17

days after the hCG trigger; it is the result of endogenous hCG secretion from the

pregnancy and tends to be more severe with multiple gestation (533). Pregnancy

outcomes are inconsistently affected by the presence of OHSS, with higher rates

of biochemical losses but similar rates of clinical losses when compared to

patients without OHSS (533).

Severity Classification

[11] A RCT (534) described mild, moderate, and severe OHSS classifications

considering criteria from Golan (531) and Navot (535) as follows: mild OHSS

(grade 1 to 2) is associated with abdominal distention and discomfort with grade 2

including nausea, vomiting or diarrhea accompanied by ovary diameter of at least

5 mm. Moderate or grade 3 includes grade 2 plus sonographic subclinical ascites

near the liver or in the pelvis if pocket >9 mm. Severe OHSS (grade 4 to 5) is

associated with grade 3 plus clinical ascites, hydrothorax, or dyspnea, with grade

5 including hemoconcentration, renal insufficiency or oliguria, elevated

transaminases, venous thromboembolism, or respiratory distress syndrome.

Risk Factors

Mild OHSS occurs commonly in both stimulated non-ART and ART cycles, but

severe disease is rare in non-ART cycles (536). Severe disease is more common

with ART, occurring in 5% of antagonist and nearly 9% of agonist protocols

(534). Polycystic ovary syndrome, polycystic ovarian morphology with attendant

elevated anti-mullerian hormone levels (>3.36 ng/mL), and previous episodes of

OHSS are major risk factors for the disease (532,534,536,537). E2 concentrations

of >3,500 pg/mL and >6,000 pg/mL at the time of hCG trigger were associated

with severe OHSS in 1.5% and 38% of patients, respectively (536). More than 20

preovulatory follicles were associated with a 15% incidence of severe OHSS

(536). When 20 to 29 oocytes were collected at retrieval, 1.4% of patients

2307developed severe OHSS; this rose to 22.7% with ≥30 oocytes (536).

Management

If outpatient management is appropriate, the patient should be instructed to limit

her activity, to weigh herself daily, and to monitor her fluid intake (at least 1

L/day of mostly electrolyte-balanced fluid) and output (532). Daily follow-up by

telephone or visit is important and the patient should be reassessed if she notes

worsening of the symptoms or if her weight gain increases to more than 2 lb per

day (532). Indications for hospitalization include inability to tolerate oral

hydration, hemodynamic instability, respiratory compromise, tense ascites,

hemoconcentration, leukocytosis, hyponatremia, hyperkalemia, abnormal

renal or liver function, and decreased oxygen saturation (532). Fluid intake

and urine output need to be carefully measured, and admission to an intensive

care setting can be considered if the patient has hyperkalemia, renal failure,

respiratory failure, or thromboembolic disease (532). Although IV fluids may

worsen ascites, they are essential to correct hypovolemia, hypotension, electrolyte

abnormalities and oliguria (532). Albumin 25% can be dosed 50 to 100 mg IV

every 4 to 12 hours if further intravascular volume expansion is needed (532).

Diuretics can be considered to improve weight gain and oliguria only after

hypovolemia has been corrected (532). Thromboembolic prophylaxis should be

given (532). Single or repeated transvaginal or transabdominal US–guided

paracentesis may relieve pain, hydrothorax, or persistent oliguria (532). [11]

Rapid large-volume fluid removal can be considered, as compensatory fluid shifts

are unlikely to occur in this typically young healthy population, so long as the

patient is carefully monitored (532).

OHSS Prevention

No method will prevent OHSS completely, but recognition of high-risk patients is

the first step in decreasing that risk (538). PCOS, elevated AMH (>3.4 ng/mL),

high peak E2 (>3,500 pg/mL), multifollicular development (≥25 follicles), or a

high number of oocytes retrieved (≥24 oocytes) are associated with an elevated

risk of OHSS (538).

COH and GnRH Analogs

Careful gonadotropin stimulation for monofollicular development is discussed

under “Ovulatory Factor.” For ART, stimulation protocols for high-risk

patients include lower initial COH doses of 150 IU to 225 IU and GnRH

antagonists for LH surge prevention, which reduce the total dosage and

duration of gonadotropin stimulation (534). [11] GnRH antagonists when

introduced following oocyte retrieval appear to accelerate regression of severe

early OHSS without any adverse effects on pregnancy and live birth rates (539).

2308Ovulation Triggering

Decreasing the dose of hCG to reduce the incidence of OHSS is controversial

(538). GnRH agonists can be used instead of hCG during antagonist cycles to

induce an endogenous LH surge. The very short half-life of endogenous LH

may reduce the incidence and/or severity of OHSS (538). As GnRH agonist

triggers are associated with lower pregnancy rates, it may be a preferable option

for patients who are not planning a fresh embryo transfer (538). [11] For those

wanting to continue with a fresh cycle, low-dose hCG coadministered with a

GnRH agonist trigger may prevent the lower pregnancy rates observed when a

GnRH agonist is used alone (538).

Coasting

Coasting may be considered when E2 levels are >4,500 pg/mL and/or there are 15

to 30 mature follicles present (537). During coasting, gonadotropin stimulation

is withheld and E2 levels are checked daily (537). An initial rise of E2 is

typically observed within the first 48 hours of the coast, but the levels should

subsequently plateau or decrease (540). The patient may be triggered when serum

E2 levels fall to <3,500 pg/mL (537). [11] If GnRH agonists are used for initial

LH surge suppression, switching to an antagonist during the coast has been

associated with improved outcomes (537).

Cycle Cancellation

[11] Cycle cancellation may be recommended if there are >30 mature follicles,

the coast duration is >4 days, or if E2 levels rise to >6,500 pg/mL during coasting

(537,540).

Adjunctive Medications

The adjunctive use of metformin is associated with decreased OHSS rates in

PCOS patients (541). Aspirin and calcium use appear to decrease the risk of

OHSS as separate adjuncts (538). Evidence suggests that the use of letrozole

following hCG triggering is effective in preventing moderate to severe early-onset

OHSS (542). Cabergoline, a dopamine agonist that inhibits VEGF

production, decreased OHSS rates when given at 0.5 mg daily for 8 days

from the day of hCG administration (538,543). Long-term use of cabergoline

should be avoided as it may be associated with valvular heart disease

(537,540,544). [11] The benefit of albumin administration in decreasing rates of

OHSS is unclear (538).

Embryo Cryopreservation

[11] Cryopreservation of all embryos without transfer appears to reduce late-onset

2309OHSS although early-onset OHSS may still occur (540). With improved freezing

techniques, live birth rates following frozen/thaw transfers are similar to fresh

transfer (491). More information on embryo cryopreservation can be found in the

“ART Process” section.

In vitro Oocyte Maturation

In vitro oocyte maturation (IVM) completely obviates the need to stimulate the

ovaries with gonadotropins. During IVM cycles, immature follicles are aspirated

following hCG administration, and the retrieved oocytes are grown in vitro until

they mature. Mature oocytes are fertilized by insemination or ICSI (545).

Randomized trials are still needed to further evaluate this technique (545). IVM is

considered an experimental procedure and should only be performed in

specialized centers (546). Randomized trials are still needed to further evaluate

this technique before it can be recommended as an alternative to conventional

ART techniques (547).

Risk of Cancer After Fertility Therapy

Infertility by itself is a predisposing factor for ovarian cancer and breast cancer

(548). Although treatments that promote incessant ovulation and elevated

estrogen levels offer biologic plausibility for further increased cancer risk,

data regarding the impact of infertility therapy on malignancy are

inconsistent and limited by methodologic issues (548). The practice committee

for the American Society of Reproductive Medicine completed an analysis of

accumulated data of prior studies addressing the potential risk of cancer with

fertility drug use (548). No association between fertility drug use and an increased

risk of invasive ovarian cancer, breast cancer, or endometrial cancer could be

established when evaluating the available data (548). Although several studies

have shown a small absolute risk of borderline ovarian tumors there is a lack of

consistent evidence that any particular fertility drug increases the risk of

borderline ovarian tumors (548). When considering other cancers, there does not

appear to be a risk of invasive thyroid cancer, colon cancer, or cervical cancer

(548). [11] There is insufficient evidence to determine if there is an increased risk

of melanoma or lymphoma (548).

Stress

Stress, as manifested by anxiety or depression, is thought to be increased among

women experiencing infertility (549). Stress is the most common reason for

patients, even those with insurance coverage, to terminate fertility treatment

(550). [11] Although the preponderance of the evidence suggests that stress does

not adversely affect IVF outcome, additional studies are needed before any

2310conclusions can be made (551–553). If psychological disorders are present,

treatment should be offered regardless of fertility status.

Neonatal and Child Development

With the ever-growing number of births resulting from ART, it is important to

consider potential perinatal risks and long-term childhood outcomes. Multifetal

gestation and its subsequent effect on maternal and fetal morbidity and mortality

represent the most significant risk of ART (554). Risks of multiple gestation

include preterm birth, low birth weight, small for gestational age, congenital

anomalies, and perinatal mortality. These risks appear to have an independent

association with IVF regardless of fetal number (554). The risk of negative longterm outcomes of children born after IVF are primarily associated with inherent

risks of multiple gestation pregnancies. When multiple gestation is excluded the

risk is uncertain and underscores the need for further studies (554). [10] The most

important step in mitigating neonatal and long-term pediatric risk associated

with IVF is continued emphasis on elective single embryo transfer (487). Patients

should have a thorough history and physical performed, addressing any health

problems or inherited conditions that could impact future pregnancy (554).

Preimplantation Genetic Diagnosis and Screening

[9] The primary indication for PGD is to improve the chances of having

healthy infants in families at high risk for a specific genetic disease (555).

Following embryo biopsy, genetic testing can be performed on a blastomere (cell

from day 3 embryo), polar body, or on blastocyst trophectoderm prior to

transferring the embryo (556) (Fig. 36-7). Data favor biopsy at the blastocyst

stage over the cleavage stage (557). Aneuploidy in embryos most commonly

affects chromosomes X, Y, 13, 14, 15, 16, 18, 21, and 22 (556). Fluorescence in

situ hybridization (FISH) is a technique that assesses aneuploidy, translocation,

other structural chromosomal defects, and sex chromosome content (556). FISH

is technically limited by the number of distinct chromosomes that can be

evaluated, giving the technique an inherently high false negative rate. Newer

genetic technologies have been developed that have the ability to assess the entire

genome using array comparative genomic hybridization (aCGH), single

nucleotide polymorphism (SNP) arrays, and next-generation sequencing (NGS)

(558). [9] One-quarter of the cases of PGD are performed for single gene

disorders, most commonly myotonic dystrophy, Huntington disease, cystic

fibrosis, fragile X syndrome, spinal muscular atrophy, tuberous sclerosis, Marfan

syndrome, thalassemia and sickle cell anemia (556). Because polymerase chain

reaction (PCR) is required for single gene disorder diagnosis, ICSI is performed

2311during ART to avoid contamination from sperm bound to the zona pellucida

(556). [9] PGD can be used for HLA tissue matching in an effort to produce a

child whose cord blood or stem cells could help an existing affected child (556).

Disadvantages to PGD include decreased postbiopsy embryo survival,

requirements for extended culture with an associated possibility that no embryos

will be available for transfer or cryopreservation, false positive and false negative

testing results, and controversies regarding disposition of nontransferred embryos

(555,556).

[9] FIGURE 36-7 Embryo biopsy procedure. A small opening is made in the zona

pellucida and depending on the stage of embryonic development the removal of a polar

body (fertilized or unfertilized oocyte), blastomere (cleavage stage 8 cell), or

trophectoderm (blastocyst) is performed.

Preimplantation Genetic Screening

PGD is not synonymous with PGS, which is performed in couples without

known chromosomal anomaly, mutation, or other genetic abnormalities

(555). [9] PGS focuses on identifying euploid embryos for transfer, with the

goal of increasing implantation rates and decreasing miscarriage rates (558).

[10] When embryos are known to be euploid, single embryo transfer is

universally recommended to decrease the risk of multiple gestation (487).

Available genetic testing platforms for PGS are the same as those used in PGD

and will equally detect whole chromosome aneuploidy. Newer genetic

technologies (Figs. 36-8 to 36-11) have added more genetic capabilities to include

assessment of segmental aneuploidy, mosaicism, and mitochondrial DNA

(mtDNA) copy number (558). Work on mtDNA has suggested that it may have

value as a potential biomarker of embryo viability (559). Earlier studies indicated

2312that elevated mtDNA content is associated with implantation failure, although

data have not corroborated these findings, suggesting the need for further

evaluation of the technology (559–562). Although it seems intuitive that replacing

only euploid embryos should improve pregnancy and live birth rates in patients

with advanced age, recurrent pregnancy loss, or implantation failure, study

outcomes have not been consistent (555,556,563–566). A significant challenge to

the application of PGS is chromosomal mosaicism (Figs. 36-12 and 36-13),

which arises from mitotic errors during embryo development, and has raised

concerns of how well the biopsy sample represents the embryo as a whole, and

whether subsequent embryonic self-correction can restore euploidy (567). Mosaic

embryos can be considered potentially viable but should be deprioritized for

transfer compared to embryos with nonmosaic euploid results as a result of

limited data on neonatal outcomes, diminished implantation, pregnancy, and live

birth rates and higher miscarriage risk (568,569). [9] Mosaicism involving two or

more chromosomes and/or higher amount of aneuploidy (involving more affected

cells) could be considered for transfer (568,569) but one study suggests they be

further deprioritized compared to embryos with single mosaic chromosome

and/or lower degrees of aneuploidy (569). [11] Thorough genetic and physician

counseling is necessary for the patient to proceed with testing embryos using

PGD/PGS and to understand the results if PGS/PGD is performed, in order for an

informed decision to be made regarding whether and which embryos to transfer.

2313FIGURE 36-8 Detection of abnormalities with different PGS platforms. Using single

nucleotide polymorphisms (SNP), diploid chromosome 21 is seen on the left, trisomy 21

on the right. (Adapted with permission from Brezina et al. [558].)

Preservation of Fertility in Cancer Patients

Improved cancer treatments such as chemotherapy, surgery, and radiotherapy

have greatly enhanced survival, such that many cancer survivors contemplate

parenthood (570). Unfortunately, those life-saving treatments can diminish

fertility potential in men and women (570). Cancer itself does not usually affect

oocytes, but certain chemotherapeutic drugs or radiation damage may adversely

affect ovarian reserve and uterine function, particularly in older women (570). [5]

If time allows, prior to commencement of cancer treatment, women can

2314undergo an ART cycle with embryo or oocyte cryopreservation, although

ovarian tissue preservation and in vitro maturation of oocytes remain

experimental (546). Fertility sparing treatments may be considered in certain

situations including radical trachelectomy for cervical cancer, progestin therapy in

lieu of hysterectomy for uterine cancer, and unilateral oophorectomy for some

ovarian tumors (570). Ovarian function may be preserved when ovarian

transposition is performed prior to radiation therapy (546). It is unclear whether

ovarian suppression with GnRH analogs during chemotherapy or radiation is

protective and it is still considered investigational (546). In men, cancer directly

affects gametogenesis and cancer treatments cause more fertility damage when

given at younger ages (570). [5] Semen and sperm cryopreservation prior to

cancer treatment, when feasible, is the standard fertility preservation method

in men (546).

FIGURE 36-9 Detection of abnormalities with different PGS platforms. Using

comparative genomic hybridization (CGH), the left panel indicates euploid/diploid

results with a relatively equal ratio of green/red fluorescence in all 23 pairs of

chromosomes. The top right shows chromosome 2 monosomy with downward deviation

indicating a relative lack of green, as compared to red, signal intensity. The bottom right

shows chromosome 13 trisomy with upward deviation indicating a relative increase of

green, as compared to red, signal intensity. (Adapted with permission from Brezina et al.

[558].)

2315FIGURE 36-10 qPCR. Detection of abnormalities with different PGS platforms. This

depicts the low number of regions of a given chromosome evaluated by quantitative PCR

(qPCR). (Adapted with permission from Brezina et al. [558].)

2316FIGURE 36-11 NGS. Detection of abnormalities with different PGS platforms. This

depicts a euploid (diploid) 46XY embryo using next-generation sequencing (NGS),

sequenced using MiSeq (Illumina). (Adapted with permission from Brezina et al. [558].)

2317FIGURE 36-12 Embryonic chromosomal mosaicism. Permutations at cleavage and

blastocyst stage. (Adapted with permission from Capalbo et al. [567].)

2318FIGURE 36-13 Embryonic chromosomal mosaicism. Mosaicism between inner cell

mass and trophectoderm in disaggregated embryos. (Adapted with permission from

Capalbo et al. [567].)

ObGyncases (Healthcare Touch)

Tìm kiếm Blog này

Berek Novak's Gyn 2019. Chapter 36 Infertility

Nhận xét

Đăng nhận xét