Why don't women get pregnant every month they have intercourse?
Adapted from a talk written for physicians
Why don't women get pregnant every month they have intercourse? Sperm are just DNA package delivery systems. We know that on average 90% of all sperm studied are chromosomally normal. It is rare to have a normal man with a large quantity of chromosomally abnormal sperm. If you do an IUI with a decent sperm count, you can be assured that at least some chromosomally normal sperm will get to the egg. Most men will have at least some sperm that can bind to and penetrate the zona pellucida around the egg and enter the egg. If not, we can always inject a sperm directly into an egg (ICSI). Why can't we make pregnancy always occur? Old data suggests that 60% of naturally fertilized eggs fail to become pregnancies. In our IVF program last year, more than 70% of patients (all ages) had positive pregnancies tests, but only half of those had ongoing or delivered pregnancies.
Perhaps women don't get pregnant once an egg is fertilized because of the uterine problems. Certainly this can occur. The best example is the closed hydrosalpinx containing fluid, which runs back into the uterus and changes the uterine environment. Lessey showed that removing the tube even changed the expression of a uterine protein, integrin, in the endometrial lining. Others showed that this surgery (removing the tube) doubled or tripled the pregnancy rate with IVF. Some fibroids and some polyps likely cause problems in achieving pregnancy. The best study on fibroids out of Johns Hopkins was randomized and showed a fertility benefit from myomectomy. Most reproductive endocrinologists would recommend myomectomy only for the larger fibroids and those fibroids located near and especially distorting the endometrial lining. One significant cost of treating fibroids is the time such treatment takes. With my concerns about age, I try to avoid most myomectomies. There is some data suggesting a fertility benefit by removing polyps greater than 1/2 cm. The data is weak, but the surgery is easy and with low likelihood of complications. There is also recent data that suggests that treatment of a polyp close to an IVF cycle enhances the implantation rate. A recent randomized study on the diagnostic benefit of endometrial biopsy for luteal phase defect has hopefully eliminated it from the routine infertility work-up. Recent data also suggests that endometriosis can cause an implantation defect which can be easily treated medically. Clearly, there can also be other problems impacting on implantation, but they do not appear to commonly be isolated causes of infertility. Another argument suggesting that the uterus is not commonly a problem is that older infertile women have a high pregnancy rate with donor eggs.
Perhaps women don't get pregnant because their eggs don't develop the biochemical capacity to implant and grow. Certainly this does occur. For example, the impact of prolactin elevation can range from amenorrhea to anovulation, to subfertile eggs. Dropping the prolactin level can lead to regular cycles and to pregnancy. Many women with PCO will get pregnant on glucophage. Glucophage does not induce ovulation or directly impact on ovulatory mechanisms, but likely changes the environment in the ovary enabling a better follicle and egg to develop. Our experience with PCO patients is the best argument that this problem does occur and can be treated. The pregnancy rate with gonadotropins also increases much more than that which can be attributed to the increased number of eggs produced. This suggests that gonadotropins improve the environment in which eggs are developing and produces better eggs. Older eggs also contain less of many important biochemical elements, such as ATP, than younger eggs. In an IVF setting, older eggs appear to be less able to overcome environmental stressors then younger eggs. Generally, the cytoplasm of older eggs looks different (dark and grainy) than younger eggs. The rate of positive pregnancy tests with IVF in our practice is similar between older and younger women, but younger women have a much higher ongoing pregnancy rate. In summary, lack of biochemical competence can lead to subfertility. The causes of that lack of competence are varied. Some causes of that incompetence can be easily simply treated. Gonadotropins and the IVF procedure can also optimize follicular development. However, biochemical abnormalities provide a poor (soft) explanation of why 60% of normal fertilized eggs don't become babies or why 2/3rds of IVF procedures do not produce babies in young women.
Much of the subfertility we see is associated with increasing age. From age 35 to 40 the likelihood of having a baby from a single cycle of IVF drops by 50%. It drops again by 50% from age 40 to 42 and again by 50% from age 42 to 43.
In many ways, the decrease in fertility associated with age makes a lot of sense. We did not evolve as a species with the objective of maintaining fertility as we aged. In 1900, the average women lived to about 50 years of age. Three hundred years ago the average woman lived to about age 35. Expecting to get pregnant at age 40 is a relatively new phenomenon. Subfertility due to age is a natural human event; it is not pathologic. Almost all women will develop infertility in this sense prior to menopause.
To help answer our question about why pregnancy does not always occur, it is useful to consider what is happening in the ovary as women age. A women starts life with about one million primordial follicles containing eggs in the ovary. Fifty years later there are essentially no eggs left in the ovary. Three to four hundred of these eggs are ovulated, but the rest leave their protected primordial state and wither away (apoptosis) when they do not receive adequate hormonal stimulation. The rate at which eggs leave their protected state is intrinsic to that woman. I like to refer to it as her biological clock. We do not understand how her biological clock is regulated although it is at least partly genetic.
The average woman has normal fertility until about age 38 and reduced fertility until age 43. It is rare to have pregnancies after age 43. On average, at age 38 there are about 25,000 eggs left in a woman's ovaries. This is a useful way of thinking about subfertility due to age. It is reasonable to say that subfertility due to age begins as the number eggs in the ovaries drops below 25,000.
Now 25,000 eggs still sounds like a lot of eggs. If you ovulate one of these, why doesn't pregnancy occur? Are these eggs different from eggs that women ovulate earlier in life? We believe that many of these eggs if ovulated become chromosomally abnormal. The best data for this comes from the clinical technique of PGD. With PGD, a single cell is removed from a day 3 8-cell embryo, the cell is fixed to a slide and the slide is stained with fluorescent tags that attach to 5 to 10 different chromosomes (FISH technique). In women above age 40, 60 to 90% of their embryos will be abnormal. Abnormal embryos are found in women of all ages with a significant increase in women above age 35 and those women with repetitive IVF failure.
Presumably in their primordial state, these eggs are as normal chromosomally as any others. In the process of development they become abnormal. The primordial egg contains 4 times too many chromosomes. With ovulation, the egg is supposed to get rid of half of its chromosomes, which go into the first polar body and with fertilization get rid of the remaining extra chromosomes. These are the steps at which the process fails and this failure increases with age. These abnormal eggs can fertilize, yield positive pregnancy tests, and develop for a while. (Some such as trisomy 21 may become babies).
Although there are many other issues that contribute to subfertility with age, the primary problem- and the one we can least directly fix- is the production of chromosomally abnormal embryos. Genetically abnormal embryos are a common part of human reproduction. This issue is hard for many people to accept. Again PGD makes this easier to believe. We recently had a patient who achieved four easy pregnancies, all boys, with her last only 2 years ago. She was 33 years old and requested PGD for sex selection. She produced 10 embryos. PGD showed that 7 were chromosomally abnormal.
Why are the eggs that women ovulate later in life more frequently chromosomally abnormal? We don't know for sure, but we do have an appealing theory. This theory hypothesizes that the eggs that are made last in fetal life are the last to be ovulated. These eggs are known to be slightly different from the eggs that were produced first in oogenesis. The chromosomes of the eggs created last have shorter telomeres than those on eggs that were created earlier.
Telomeres are non-coding repetitive DNA sequences of TTAGGG that cap the ends of chromosomes. As cells divide, the length of their telomeres gets shorter unless the cells have the enzyme to lengthen them. Telomeres play a role in helping identical chromosomes pair with each other so that the correct chromosomes can be passed out of the cell into a polar body. It is believed that shorter telomeres are less effective in enabling this chromosome pairing to be error free. Short telomere length in egg chromosomes is also correlated with low success rates in IVF. This also provides an explanation as to why it takes more cycles (gonadotropin or IVF) to achieve pregnancy in an older compared to a younger woman.
The telomere theory may or may not be true, but the last eggs that a woman ovulates, independent of her age, are more frequently chromosomally abnormal. Clinically, subfertility begins about age 38 when there are about 25,000 eggs left in a woman's ovaries. The statistics of this situation contains some very important messages.
As you all know better than I from your patient populations, menopause, the time at which there is almost no eggs left in the ovary occurs around age 51. This does not mean that all women become menopausal at age 51. Rather there is a normal distribution of ages with mean age 51, when women have no eggs left. In some women loss of all eggs occurs at age 40; in others it doesn't occur until age 60. Similarly there is a "normal" distribution of ages centered around age 38, when women have 25,000 eggs left in their ovaries. Given the mathematical properties of normal distributions, of key clinical importance is that 10% of women age 32 will have less than 25,000 eggs left in their ovaries. Although 10% of randomly selected women is a frequent occurrence, when infertility is the presenting complaint, the frequency of this being a problem is much higher. This is why length of infertility is so useful in pointing to patients with more severe problems. These women still have remaining fertility potential, but if their problems are not quickly recognized and they are not given aggressive/effective therapy, this pregnancy potential will be lost.
Over the last 10 years, a number of methods for assessing decreased ovarian reserve have been suggested. These include early follicular FSH and estradiol measurements, lupron stimulation tests, clomiphene challenge tests, inhibin measurements, anti-Mullerian inhibiting hormone measurements, and antral follicle counts.
Antral follicle counts have been shown to be superior to all of the other methods of assessing ovarian reserve. We find them extremely useful clinically. Our management of patients is always integrated with that patient's antral follicle count. It is a useful predictor of response to infertility tools and can also be used to counsel a patient about her plans to postpone pregnancy. I think that antral follicle counts will eventually have a role to play in general gynecological practices.
The only problem with antral follicle counts is that you likely have to do them yourself for them to be reliable, in part because some patients will always require clinical judgment to interpret what is being seen. One can do antral follicle counts anytime the ovaries can be clearly visualized. Developing follicles can obscure the ovary and suggest a lower antral follicle count than really exists.
The idea of an antral follicle count is simple. Measure the cysts between 3 mm and 10 mm in each ovary. The number of antral follicles that a normal patient has should total between 10 and 20 follicles. Greater than 20 suggests a PCO pattern. In the more severe PCO pattern patients you would want to avoid gonadotropin use without IVF due to increased risk of multiple birth. If there are only 5 to 10 follicles there is some degree of ovarian compromise and more aggressive therapy is indicated. Those patients with less than 5 follicles will be lucky to get pregnant. (In women with only one ovary, I would half these cutoffs and recognize that fewer eggs will be available with any therapy. However, such patients are likely to do better than others with the same total antral follicle count using two ovaries).
Anti-Mullerian Hormone (AMH) is a protein produced primarily in antral follicles. It can be measured anytime during a woman's menstrual cycle and correlates to the antral follicle count. It can be used in a manner similar to the antral follicle count and it provides information that is supplementary to it.
When I started training in gynecology, people suggested that the big increase in infertility was due to the sexual revolution and sexually transmitted diseases. I think they were partly right. The sexual revolution gave women more control over their bodies and enabled them to defer marriage and reproduction. In our complex society, this extra time for maturation, training, and career development is a good thing. However, it results in infertility due to age. This is a natural phenomenon and not a pathological process. It is a tricky problem, because it does not occur at any particular age and does not correlate with looking or feeling old.
Generally, we can't know with certainty if infertility care enabled a pregnancy to occur that otherwise would not have occurred, given enough time, but we do know that on average that pregnancy occurred in a shorter time period than it otherwise would have occurred. Since we also cannot clearly know the underlying fertility curve, i.e., the rate at which that patient is losing fertility potential as she ages, it is important to move expeditiously and aggressively. If we do not, there will surely be some people, who could have gotten pregnant and who lost their potential for pregnancy because we didn't.
If I feel that a couple has infertility primarily on an age related basis, then there are two approaches I take depending on how severe I believe the problem is. I would use gonadotropins (with IUIs) in anyone with age above 35 (without a PCO ovarian pattern) or younger with a moderately decreased antral follicle count (say 7). In both of these settings, failure to get pregnant after three cycles of gonadotropins provides additional support for the reasonableness of IVF as a therapy. Patients with more severely decreased ovarian reserve, especially under age 40, should optimally go directly to IVF. Success rates are lower with decreased ovarian reserve compared with IVF in patients with normal ovarian reserve, but the relative benefit of using IVF to achieve pregnancy is higher.
Most infertility problems have some underlying pregnancy rate. Even problems that have a dismal 1%/cycle pregnancy rate have a 21% pregnancy rate over 2 years and greater then a 50% pregnancy rate over 10 years. However, for the subset that doesn't get pregnant in those 10 years, one may have discarded the opportunity to use technology to ever achieve pregnancy. The failure to act earlier in that person's life has permanently consigned her to a life without her biological children. Generally avoiding this outcome is why we do infertility care. IVF is indicated when the potential for success with lesser therapies is low or the time interval to enable those therapies to be successful may prevent the patient from ever being successful with IVF.