This document has undergone peer-review by an independent group of scientific experts in the field. Readers wishing to comment on the substance of the paper are welcome to send comments to peerreview@protectingourhealth.org.

Infertility and Related Reproductive Disorders
Ted Schettler, MD, MPH, Science Director,
Science and Environmental Health Network
May 2003

Infertility is a term that is often used to describe the failure to have a child, despite unprotected intercourse. Demographers define infertility in terms of the absence of children. The American Society of Reproductive Medicine defines infertility as a condition that can be diagnosed when a couple fails to conceive within 12 months of unprotected intercourse. Approximately 10-15% of couples of reproductive age meet this definition of infertility. For the purposes of study, improved understanding, and descriptive demography, distinctions between fertility and fecundity are sometimes useful. Fecundity refers to the physiologic ability to have children and is sometimes defined as the probability of conceiving within one menstrual cycle, in the absence of contraception.

Trends in infertility are difficult to determine or to interpret for several reasons. Perhaps most important, many couples now choose to delay childbearing for a number of years after reaching reproductive maturity. Fertility trends may be influenced by this choice, since fertility naturally declines with age, particularly after age 35. About one-third of women who defer pregnancy until the mid to late 30’s, and at least half of women over age 40, will have an infertility problem (Speroff et al. 1994). This is not a new phenomenon. Recent data, however, indicate that rising rates of infertility are not entirely explained by voluntary delayed childbearing. One report, for example, shows that even within specific age groups, impaired fecundity is increasing (Chandra and Stephen 1998). While impaired fertility increased between 1982-1995 by about 25% in all women aged 15-44, the increase was only 6% in women aged 35-44, 12% in women aged 25-34, and 42% in the youngest group. These data suggest that delayed childbearing does not fully explain the apparent upward trend and that even younger women are experiencing fecundity problems.

Some of the apparent increased rates of infertility may result from increased reporting of fertility problems because of newly available treatments. Assisted reproductive technologies, including pharmaceuticals that stimulate ovulation and in vitro fertilization, result in successful pregnancies and increase the likelihood that a woman or couple will seek medical interventions. Access to medical care will also influence the likelihood that fertility problems will be identified and reported. As a result, trend analyses of infertility are subject to significant limitations and should be interpreted with caution.

Infertility or impaired fecundity does not necessarily imply lack of conception. A couple might conceive, for example, but the fertilized egg might not implant normally in the uterus, or the developing embryo or fetus might not survive after implantation. Typically, this results in a miscarriage. If the loss occurs early, it might go undetected or the woman might think that her period is simply a few days late. For some women, early pregnancy loss (spontaneous abortion or miscarriage) may be a single event or may be recurrent. In the general population, about 50% of fertilized eggs do not progress to a viable pregnancy, and about 30% of pregnancies are lost in the first six weeks (Warburton 1987; Wilcox et al. 1988).

Infertility may result from male factors (estimates range from 20-50% of cases), female factors (about 30% of cases), and the rest are attributable to couple-dependent factors or are unexplained (Evers 2002; Irvine 1998). For purposes of understanding or treatment, the distinction among the causes can be important.

Sperm Count Trends

The mature testis is comprised of developing sperm surrounded by a protective shield of Sertoli cells. Together, the immature sperm and Sertoli cells are arranged in tubules surrounded by a basement membrane. The cells responsible for testosterone production, Leydig cells, lie outside of the basement membrane. Until the onset of puberty the developing sperm and Sertoli cells are bathed in blood from the general circulation. Early in puberty, however, the Sertoli cells rearrange in tight formation along the inner aspect of the basement membrane, creating what is called the blood-testis barrier. This barrier provides partial protection to the developing sperm from exposure to some toxic substances circulating in the blood that might otherwise more readily enter the tubules. A normal sperm count is directly dependent on the appropriate number of healthy Sertoli cells. An adequate sperm count is important for successful reproduction. Semen quality is easier to measure than most female-related fertility factors, and as a result, some historical data are available that describe semen quality in men in various geographical locations.

Studies have reported sperm counts in the general population over many years with widely varying numbers. More than 25 years ago several hundred men undergoing vasectomy were reported to have an average concentration of 48 million sperm/cc of seminal fluid, and the authors wondered if this was evidence of a population-wide change (Nelson and Bunge 1974). Previous studies had reported an average sperm concentration of 100-145 million/cc (Hotchkiss 1938; Farris 1949; Falk and Kaufman 1950). It was, however, virtually impossible to draw any valid conclusions about trends from these data since it was unlikely that these individuals were representative of the general population or that the groups were comparable.

A widely publicized study in 1992 ignited general concerns about falling sperm counts (Carlsen et al. 1992). This report included an analysis of 61 scientific papers published between 1938-1990 and concluded that there was evidence of a decline of the average sperm count in the general population from about 113 million/ml to 66 million/ml over this time frame. It is now generally agreed that sperm counts below 20 million/ml are quite likely to be associated with reduced fertility and many of the studies showed an increased number of men whose counts fell below that threshold.

The 1992 study received considerable attention and sparked animated controversy. The limits of the analysis and the difficulties measuring sperm count trends were widely discussed. Subjects were not chosen in the same manner in each study; they differed in ways that might have affected semen quality. Different statistical and analytic methods had been used in the 61 studies. Variability of sperm counts within individuals, depending on the timing of the last ejaculation and other personal factors, made drawing conclusions difficult. Since 1992 several other studies have also documented a decline in sperm counts (Auger et al. 1995; Irvine et al. 1996; Van Waeleghem et al. 1996) while others have failed to find similar evidence (Vierula et al. 1996; Bujan et al. 1996; Paulsen et al. 1996). One study of a large number of men living in three areas in the US (Minnesota, New York, Los Angeles) found no general decline in sperm count but discovered marked geographical variations in sperm quality in these men (Fisch and Goluboff 1996; Fisch et al. 1996). Sperm counts were highest in New York, intermediate in Minnesota, and lowest in Los Angeles.

More recently, a reanalysis of the data from the 1992 report using a variety of statistical techniques and correcting for factors that might inappropriately influence the outcome concluded that statistically there has been a decline, on average, in sperm counts in the US and in Europe, but not in non-Western countries (Swan et al. 1997).

Conflicting results of human sperm count studies continue to spark controversy and discussion (Irvine 1999). However, the apparent decline in sperm counts has now been placed in a new context. Using standardized methods, two recent studies show that semen quality can differ appreciably from one area to another. (Swan et al.2003, Jorgenson et al. 2001) Swan et al. found, for example, that men from semi-rural and agrarian mid-Missouri had markedly reduced sperm concentration and motility compared to men from more urban environments. The authors noted that the 1974 survey (Nelson and Bunge 1974) reporting lower sperm counts had also been conducted in an agricultural area (Iowa) and suggest that exposure to agricultural chemicals (pesticides) may play a role in regional differences in semen quality. Moreover, a large number of reports document the unequivocal impacts of environmental contaminants on wildlife reproductive health, including reduced sperm counts, reproductive failure, birth defects of the reproductive tract, and behavioral abnormalities (National Research Council 1999). This evidence, along with consideration of other human health trends, including certain birth defects of the male reproductive tract and testicular cancer, suggest that a more fundamental disruption of male and female development may be at work and that sperm count changes are only one manifestation of that process.

Other Changes in Male Reproductive Health Endpoints

Some data suggest that the incidence of birth defects of the male reproductive system is increasing in some parts of the world. In the US, hypospadias, a malformation in which the urethral opening is on the underside of the shaft of the penis rather than at the tip is increasing according to an analysis of a national and state birth defects registry (Paulozzi et al.1997). The incidence of cryptorchidism, a condition also known as undescended testes, in which the testes do not descend into the scrotum during development, is increasing in some countries, though not in others (Paulozzi 1999; Toledano et al.2003). Variable methods of collecting data, variable inclusion criteria, and inconsistent reporting make it difficult to estimate trends of these conditions with precision. (Toppari et al. 2001) The incidence of testicular cancer has increased 2-4 fold over the past 30-40 years in some populations in the US, Canada, and Europe (Liu et al. 1999; McKiernan et al. 1999; Bergstrom et al. 1996). Danish researcher Skakkebaek coined the term “testicular dysgenesis syndrome” to describe how these various pieces of information may be related and have a common explanation (Shakkebaek 2002). This is further discussed below.

Causes of Infertility

In women or men, infertility can be caused by genetic or environmental factors, combinations of the two, or endocrine or immune system disorders (Nudell and Turek 2000; Foresta et al. 2002; Achermann et al. 2002; Hruska et al. 2000; Oliva et al. 2001; Sharpe 2000; Hatasaka 2000).

Female Factors

In the female, ovulation depends on a number of factors, including normal egg development during fetal life and complex interactions among hormones secreted from the brain, the pituitary gland, and the ovary after reproductive maturity. During the menstrual cycle, the concentrations of hormones, including estrogen and progesterone change dramatically, resulting in ovulation and preparation of the uterus for implantation of the fertilized egg. If this highly orchestrated and tightly controlled sequence of events is interrupted, infertility or reduced fertility may result.

In women, failure to ovulate normally can be caused by genetic or environmental factors, including toxic exposures. Endocrine problems such as thyroid disease can also interfere with normal ovulation. Based on the results of animal testing, considerable attention is now also focused on the role of fetal exposure to endocrine disrupting chemicals in ovulatory abnormalities. Failure to ovulate normally is responsible for approximately 40% of infertility in women (Speroff et al. 1994).

In addition to failure to ovulate, female-related causes of infertility include abnormalities of the Fallopian tubes or other reproductive organs. Endometriosis or previous infections can cause obstruction or scarring of the tubes and interfere with fertilization of the egg or movement of the fertilized egg into the uterus. Hormone imbalances or exposure to some toxic chemicals can interfere with normal implantation of the fertilized egg and failure to maintain pregnancy.

Pregnancy loss can also occur after successful implantation of the fertilized egg. Genetic studies indicate that as many as 50% of early miscarriages may be due to chromosomal abnormalities that may be of either maternal or paternal origin (Cramer and Wise 2000). Chromosomal abnormalities may be inherited or caused by environmental factors. Other causes of early miscarriages include direct damage to the developing embryo or fetus from toxic exposures, radiation, maternal genital tract abnormalities, maternal illness, and immunologic abnormalities.

Male Related Factors

Male causes of infertility include low sperm count, altered sperm motility, and abnormalities of seminal fluid. Reduced sperm counts can be caused by genetic factors, infections (e.g. mumps), anatomic abnormalities, heat, or exposure to toxic chemicals during fetal development or adulthood. Toxic chemical exposure can directly damage developing sperm. Fetal or pre-pubertal exposure to toxic chemicals can also permanently harm or reduce the number of Sertoli cells, resulting in a corresponding reduction in sperm count in the adult, since there is a direct relationship between Sertoli cell numbers and sperm count. Leydig cell damage can result in decreased testosterone production with indirect impairment of fertility.

Couple Dependent Factors

Couple-related causes of infertility include combinations of marginally reduced sperm counts, incompatibility between the sperm and cervical mucus, and sperm antibodies that interfere with normal sperm motility or egg penetration (Hatasaka 2000). Either the male or female may produce anti-sperm antibodies.

Toxic Exposures and Infertility

Table 1 (Hruska et al. 2000) lists some of the chemical substances that can impair fertility or fecundity in men or in women. No animal data are cited in Table 1, though numerous animal studies show reproductive impacts of many commonly encountered chemicals (Schettler et al.1999). In all of the studies cited in Table 1, the reproductive impacts were caused by exposures during adulthood. Many of the studies are published reports of men or women whose exposure occurred in the workplace. Occupational exposures are often, but not always, higher than exposures in the general population. For example, a pesticide applicator who works with pesticides daily is likely to be more heavily exposed than someone who uses pesticides only occasionally. Higher exposures are usually more likely to cause health effects than lower exposures. However, even short-term exposures during occasional use of pesticides or other potentially toxic substances can have health consequences if precautions are not taken. Exposures to solvents, for example, can be excessive in the home during renovation projects or hobbies without proper ventilation.

Table 1: Environmental factors reported associated with adverse outcomes related to infertility/decreased fecudity

* Dibromochloropropane—a soil fumigant no longer used in the US. DBCP was responsible for causing sterility of many chemical and farm workers in the US and other countries
** Ethylene dibromide—a pesticide and jet fuel additive. Contaminates groundwater in some areas of the country

Many studies have reported the adverse impacts of solvent exposure on fertility and fecundity. The reproductive impacts of lead exposure are more likely to be seen in people with blood lead levels that are well above the national average. see note Workers in lead-related industries are probably at the highest risk, but some people are also excessively exposed to lead from contaminated dust in a house painted with leaded paint or from various hobbies. Most of the studies of the impacts of pesticides have been carried out in agricultural workers. In many cases, actual exposure levels have not been measured, and the reproductive health of workers has been compared to that of non-agricultural workers in order to estimate the risks. Studies of this kind may underestimate the risk of exposures because of misclassification of study subjects into the exposed or unexposed group.

New Concerns About Early Life Environmental Exposures and Fertility

Between 1947 and 1971 millions of pregnant women in the US were given a synthetic estrogen, diethylstilbestrol (DES), with the hope that it might reduce the risk of miscarriage, despite the fact that it was ineffective. In 1971 a landmark study demonstrated that use of DES during pregnancy caused a rare cancer of the vagina and cervix in daughters who had been exposed in the womb. (Herbst et al. 1971). Subsequently other health consequences of fetal DES exposure became clear in both male and female offspring. In addition to vaginal and cervical cancer, girls exposed to DES during fetal life are at risk of developing other abnormalities of the reproductive tract and immune system disorders (Giusti et al. 1995). Boys who are exposed to DES during fetal development appear to be at increased risk of hypospadias, cryptorchidism, and decreased sperm counts, although sons have been much less widely studied and male effects are less certain.

Hypospadias, cryptorchidism, and decreased sperm counts can also be produced in experimental animals by the administration of estrogenic or anti-androgenic substances during fetal development (National Research Coucil 1999). These endocrine-disrupting substances include a variety of pesticides, components of commonly encountered plastics (e.g., bisphenol A, alkylphenols, phthalates), glues and resins, detergents, hormones used in pharmaceuticals for humans and farm animals, byproducts of waste incineration (e.g., dioxin), and are in many other consumer products.

Although the causes of testicular cancer are not well understood, young boys with cryptorchidism are at significantly increased risk of developing testicular cancer (Moller et al. 1996). Animal studies show that fetal exposure to some estrogenic substances can fundamentally disrupt testicular development, setting the stage for later emergence of testicular cancer (National Reserach Council 1999; Yasuda et al. 1985). The term “testicular dysgenesis syndrome” links these pieces of information together with a proposed common explanation (Shakkebaek 2002). In essence, studies of laboratory animals, wildlife, and humans show that fetal exposures to endocrine disrupting substances can fundamentally alter development. If the suggested trends in human male reproductive health are valid, they may result from alterations in male fetal development, reflected in multiple endpoints that appear at different times. That is, hypospadias or cryptorchidism may be present at birth, but testicular cancer and low sperm counts may not develop or become apparent for years. For these reasons, when considering the trends of infertility in humans and its potential environmental causes, it is important to consider 1) the fetal environment as well as the environment after birth and 2) the relationship of fertility factors, such as semen quality, to other measures of reproductive health and development.

Concerns about the reproductive-health consequences of chemical exposures during fetal and childhood development are based on animal testing, wildlife observations, and limited human data. Unfortunately, few human studies have examined the impacts of early life exposure to endocrine disruptors or other toxic chemicals on reproductive function later in life. DES is an exception. A long-term analysis of the impacts of early life exposures to a wide variety of other substances on male reproductive function is missing from the scientific literature. For example, the extent to which fetal exposure to dioxin or substances that damage Sertoli cells during development might be responsible for a decline in human sperm counts over the last 50 years is simply unknown, yet the evidence is strong in laboratory animals. Similarly, the extent to which ovulatory dysfunction in adult women is caused by fundamentally altered hormone levels or hormonal feedback control loops, as a result of early life exposure to endocrine disrupting chemicals is unknown.

Human studies designed to examine these questions are complex, difficult to carry out, and expensive. Studies that measure fetal exposures to substances of interest and then follow offspring as they mature and attempt to reproduce require decades and consistent, close follow up. A children’s health study, intended to measure a variety of contaminants in umbilical cord blood of a large cohort of children at birth and monitor a variety of health indicators as they grow and develop, is in the design phase (Children's Health Study). The results of that investigation will not be available for many years. Meanwhile, a central question requiring public discussion and policy decisions is the extent to which the results of animal testing, wildlife observations, and limited information about human health trends should be used now for protecting the reproductive health of humans and wildlife.