IVF Implantation Failure
Difficulty arises in studying implantation failure due to the fact that not all scientists agree on a definition. Some doctors will call failure of a single in vitro fertilization cycle implantation failure. Others maintain that it is not implantation failure until a couple has failed to achieve pregnancy despite several in vitro fertilization failures. Still others look at the total number of embryos transferred as opposed to the number of treatment attempts. There are also many exceptions that have to be considered. For example, can a couple be said to have implantation failure if they have a history of a previous successful pregnancy? What about women with a tubal pregnancy? There is much that needs to be learned about human embryo implantation.
Implantation failure can be divided into three areas: Problems with the embryos, problems with the “host uterus” and problems in the interaction between the embryo and uterus.
How can you identify the best embryos?
Where is the best place to transfer an embryo?
Recently, the debate re-emerged as some maintained that GIFT or ZIFT may be superior for couples with IVF implantation failure. However, in 2005, a group from Israel studied couples with IVF implantation failure by dividing them into two groups. One group simply repeated a standard in vitro fertilization cycle. The second group underwent a treatment cycle using ZIFT. The results indicated that there was no difference in the implantation rate or pregnancy rate between the two methods. Thus, transferring embryos into the fallopian tube is not an effective treatment for implantation failure.
What is the best time to transfer embryos in an IVF cycle?
Assisted hatching was first performed in the early 1990s. Many different variations of the original technique have been described. However, to this day, the effectiveness of assisted hatching remains controversial. Several well designed randomized controlled studies have been published which have attempted to prove whether assisted hatching improves the chance for IVF implantation. The results have been mixed. Roughly one half of the studies showed an improvement in the IVF implantation rate while the other one half showed no improvement. In some cases, the groups that were subjected to assisted hatching techniques, had lower implantation rates.
There does appear to be some risks associated with assisted hatching. The zona pellucida helps protect the integrity of the embryo during early embryo development. By removing the zona pellucida, the embryo may have a greater chance of splitting. In fact, studies have shown an increase in the rate of identical twinning when assisted hatching is used. Identical twinning carries greater risks to the babies than fraternal twinning. A published report has also noted a case of Siamese twinning that occurred after assisted hatching.
At the current time, assisted hatching must be viewed as an uncertain method to treat IVF implantation failure.
The problem is in trying to determine who has an infection, which infections are important to treat and what the best method is for trying to treat them. Many physicians attempt to obtain uterine cavity samples by passing a thin catheter through the vagina and cervix and into the uterine cavity. This method is problematic, however, because the specimens can be contaminated with bacteria from the vagina and cervix and lead to over diagnosis.
Some physicians prefer to treat all patients with antibiotics without performing testing for infection. Aside from the occasional allergic reaction to the antibiotics, it is unknown whether the antibiotic being given is the correct one or whether the infection has been eradicated.
At best, treating for presumed intrauterine infections is uncertain as a method to increase the chances for IVF implantation.
Immune factors as a cause for IVF implantation failure
In the areas of IVF, infertility and recurrent pregnancy loss, there has been an extensive amount of investigation as to the role of immune factors in the reproductive process. Unfortunately, many of these studies have been poorly performed and this has led to many erroneous conclusions. A review of some of the areas that have been looked at follows:
Anti-phospholipid antibodies (APA)
A few years ago, a large study was conducted in couples going through in vitro fertilization at a large IVF program in the United States. Before the treatment was started, the females had blood drawn to look at a large number of different anti-phospholipid antibodies. The blood was sent to a lab which was very experienced in studying blood for anti-phospholipid antibodies. After the in vitro fertilization cycles were completed, the results of the antibody testing were released. In this way, neither the physicians nor the patients could be influenced in how they performed the in vitro fertilization cycle since they did not know the results of the antibody testing. The results showed conclusively that, the presence of high levels of anti-phospholipid antibodies did not influence the chance for embryo implantation. Stated another way, the implantation rate was the same whether women had high levels, low levels or undetectable amounts of anti-phospholipid antibodies.
The quality of medical evidence against a role of anti-phospholipid antibodies in implantation failure is so strong that it is considered unethical to obtain these tests in women with implantation failure unless it is part of a research protocol. In fact, the Practice Committee of the American Society for Reproductive Medicine has concluded that “the assessment of APA is not indicated among couples undergoing IVF. Therapy is not justified on the basis of existing data.”
Antinuclear antibodies (ANA)
To date, there have been no studies which have linked anti-nuclear antibodies with implantation failure. Obtaining them is not recommended.
Antibodies to alpha-2 glycoprotein 1
A recent study of over 600 women going through in vitro fertilization identified 13% with anti-thyroid antibodies . These women were compared to the remaining 87% who did not have elevated levels. There was no difference in the pregnancy rate between the two groups. Thus, anti-thyroid antibodies are not associated with implantation failure.
Natural Killer Cells (NK Cells, CD56 Cells)
There are reasons to suspect that uterine NK cells may be involved in reproduction. NK cells are the dominant type of immune cell in the uterine lining during formation of the placenta. These uterine NK cells are also present in the uterine lining of non-pregnant women. After ovulation, for example, uterine NK cells divide and replicate so rapidly that they eventually account for about 1/3 of the cells present in the uterine lining at that time. Uterine NK cells accumulate in large numbers at the embryo’s implantation site.
Furthermore, uterine NK cells come in close contact with the cells of the placenta called the trophoblast. These placental trophoblast cells are important for transforming the arteries in the uterus into a type that will accommodate the high blood flow necessary to support a developing fetus.
However, although uterine NK cells are in close proximity to important reproductive events, the function of these cells is completely unknown. In addition, it is important to realize that uterine NK cells are completely different in how they look and act compared to NK cells that circulate in the blood.
Based on the assumed similarities between NK cells in blood and uterine NK cells, it has become increasingly common for physicians to measure NK cells using blood tests in women with infertility and recurrent miscarriage. These blood tests are obtained based on the assumption that women with recurrent miscarriage and infertility also have abnormalities in uterine NK cell function. However, examination of blood NK cells is not representative of their cousins in the uterus.
There are more problems associated with the measurement of blood NK cells. The percentage of NK cells in blood in normal healthy individuals varies from 5% to 29%. Despite this well known fact, a finding of more than 12% NK cells in women with infertility or miscarriage has been arbitrarily defined as abnormal. The percentage of NK cells in the blood can be affected by many factors including sex, ethnicity, stress, and age, but there is no indication that concentrations in the upper end of the normal range are ever harmful.
Some physicians have come up with more elaborate ways to try to prove that blood NK cells are associated with reproductive problems. Physicians have attempted to measure the activity of blood NK cell by a using a wide range of tests. Consistency is a major problem with these test and the results will vary significantly in different laboratories. The most commonly used test of NK cell activity is called a cytotoxicity assay. This measure the ability of NK cells to kill other cells in a laboratory setting. The unproven assumption is that if blood NK cells are efficient at killing other cells in a test tube then, uterine NK cells must do the same thing in the uterus. Despite the limitations of this type of testing, there is little correlation between the cytotoxicity of uterine NK cells and blood NK cells.
Attempts have also been made to compare the number of NK cells in the non-pregnant uterine lining of women with recurrent miscarriage or infertility with that of women with normal fertility. Since the number of uterine NK cells changes rapidly after ovulation, precise timing is necessary to avoid miscalculation of the results. In other words, comparing uterine NK cell numbers in a woman 5 days after ovulation may give much different results than a woman who is 6 days after ovulation. Uterine NK cells must be obtained by performing a biopsy of the uterine lining. However, since the density of NK cells throughout the uterine lining varies, a “deep” biopsy will give different results than a “shallow” biopsy. It is impossible to measure the depth of the biopsy in a living uterus. Therefore, it is nearly impossible to determine whether a given patient with implantation failure or miscarriage has an “elevated number” of NK cells in the uterine lining compared to a woman without these problems.
Because of the technical limitations described above, studies attempting to demonstrate a correlation between NK cells and implantation failure produce wildly inaccurate and potentially misleading results.
Women with implantation failure are being given treatments such as steroids and intravenous immunoglobulin (IVIG) to “cure” their problem with elevated NK cells. Neither steroids nor the other treatments being offered to women with “raised” levels of NK cells in blood have been approved by the FDA for use in implantation failure. More importantly, the studies involving these treatments have concluded there is no therapeutic benefit to their use. For example, a large Canadian randomized controlled trial found no benefit to the use of IVIG. There are risks to these treatments and some, such as IVIG, are very expensive.
Data here is lacking. In an initial study in 2001, Italian researchers concluded that two genetic mutations causing thrombophilias were more commonly seen in women with repeated in vitro fertilization implantation failure. However, of the 18 women studied, almost half had recurrent miscarriage and not implantation failure. Subsequently, a much larger study involving 234 women who failed to achieve pregnancy with in vitro fertilization, did not find any difference in the incidence of thrombophilias.
Empiric use of anticoagulants for implantation failure
The best study to date was published in 2004 in a well respected journal called Fertility and Sterility. In this study, couples with implantation failure who had at least ten embryos transferred without a pregnancy were enrolled. Each couple was randomly assigned to receive either heparin and aspirin or a phony or placebo. Neither the patients nor the doctors knew who were receiving the real treatment or the phony. The study failed to find any difference in the number of positive pregnancy tests, fetal heart beats on ultrasound or live birth rates per embryo transferred. In other words, there was no difference in the chance for implantation. Given the high quality of this study, it is considered unethical to administer heparin and aspirin for implantation failure, unless it is part of a study protocol.