by Daniel Shapiro M.D.
When a couple present themselves for fertility care, their obvious goal is to leave the care of their physician with a healthy pregnancy. For most couples where the woman is under 40, pregnancy is the ultimate outcome, although it often comes at great expense. In many cases, the cost is not only financial but also emotional. It is quite common for reproductive endocrinologists to have patients express a desire for pregnancy as soon as possible without complete understanding on the patient's part of what that might exactly entail.
When deciding on a particular therapy, patients and their doctors need to consider many factors including: success of therapy, cost of treatment, inconvenience, side effects, and risks of treatment.
Certain diagnoses leave little choice. For example, a patient who has had her fallopian tubes removed has no choice other than IVF if she desires pregnancy. Similarly, patients with certain anatomic abnormalities require surgery. For most patients, however, there are several options for treatment. Lengthy discussion between the patient, her partner and their doctor is necessary to properly tailor treatment.
After an initial work-up for infertility has been performed, one of three conditions will exist: 1) a diagnosis with a specific treatment will be evident or 2) a diagnosis with treatment choices will be found or 3) no specific diagnosis will be identified and the doctor and patient will have to decide which treatment from the appropriate list will be attempted.
In real terms patients will be found to have tubal disease, ovulatory dysfunction, male factor, uterine abnormalities, endometriosis-associated infertility, a combination of the above or no diagnosis at all. Though other infertility diagnoses exist (such as immune dysfunction or luteal phase dysfunction), they are uncommon and poorly defined. Consequently, there is no consensus for treatment of these potential problems.
Also in very real terms, all fertility care can be reduced to one of four categories: expectant management (which means do nothing special), surgery, ovulation induction with or without intra-uterine insemination (IUI) or lastly, In-vitro fertilization (IVF).
For patients with early stage endometriosis, mild male factor, ovulation errors or unexplained infertility, many doctors recommend, and many patients request, therapies based on intra-uterine insemination. IUI can be timed by ovulation induction protocols where the doctor monitors the growth of a patient's ovarian follicles (egg sacs) while the patient takes either oral or injectible medication. IUI can also be timed by ovulation predictor kits in which a patient dips her urine to detect the most fertile time in a cycle by detecting the presence of LH, the ovulation hormone, in urine
Insemination has been around quite a long time and predates IVF treatment by 25+ years. The first recorded inseminations were performed with donor sperm in the 1950s and were surrounded by much controversy regarding the ethics of the procedure. These inseminations were of the 'turkey baster' variety, where donor sperm was placed in the vagina or on the cervix but not injected into the uterine cavity.
Clomiphene Citrate (Clomid) was approved for use by the FDA for ovulation induction in anovulatory women in 1967 and within a few years, insemination was added to the protocol in many centers. Initial experience with intrauterine insemination yielded the observation that direct injection of raw semen into the uterus could cause intense cramping and pain. These symptoms are caused by the hormones in human semen called prostaglandins. By the mid-1970s methods of washing sperm out of semen and re-suspending live sperm in nutrient media became available and allowed for IUI to be performed without side effects.
Unfortunately, reliable clinical analysis of the efficacy of IUI wasn't available until much later. It was (understandably) assumed by many that simply increasing the number of sperm in the fallopian tube (where fertilization takes place) would increase the chance of pregnancy. At the time that ovulation induction and insemination therapy was coming into common practice, doctors performing these procedures required no special training and scientific evaluation of new therapies was often slipshod. Since there really wasn't anybody tracking outcomes adequately in the early years of IUI, it took a long time to realize how limited the therapy really is.
To understand the limits of IUI it helps to know the natural rate of pregnancy in couples attempting conception at different ages. The maximal chance of pregnancy per month occurs at age 27 and is about 22% per month. This means that in any given month of trying one in four or five 27 year olds will conceive. At age 35 the month to month chance of pregnancy is 16 % and at age 40 it is 7% per month. We now know that IUI alone DOES NOT CHANGE the month to month chance of conceiving AT ALL because the number of sperm is, sadly, irrelevant. The limit to fertility at any given age is the egg. In cases where there is only one egg, it matters not whether there are millions of sperm around (as is the case with IUI) or tens to hundreds as seen with natural intercourse. The inherent potency of the single available egg will set the pregnancy rate.
That being the case, several fertility centers began offering therapies to women to increase the available egg number per cycle. The rationale here was obvious: increase the number of targets for the arrows and one becomes more likely to hit a bullseye. The first reports of combining ovulation induction drugs and IUI in women who already ovulated normally began appearing in the early 1980s. This too was met with great controversy, but the combination of drugs plus IUI was more successful than IUI alone or drugs alone and IVF was still in its infancy, so the therapy stuck.
Reports on the efficacy of drug therapy plus insemination began to pop up frequently in the literature but again, there were no truly reliable reports available to quantify outcomes. It was not until 1991 that the first paper considering the cumulative benefit of IUI was published in Fertility and Sterility. The paper, by Linda Chaffkin, MD, showed that with injectible drugs and insemination, pregnancy rate varied by diagnosis AND very few pregnancies occurred after a 3rd attempt for some diagnoses or a 4th attempt for others. As expected, patients with ovulation dysfunction had the highest cumulative pregnancy rates while patients with severe male factor did quite poorly.
Irrespective of diagnosis though, the paper showed quite clearly that with IUI it is basically 'three strikes and yer out!' Plainly, and less bluntly put, this means that for ANY IUI treatment regimen in ANY patient population, it makes little or no sense to keep trying if no pregnancy occurs by the third attempt. Though similar data does not exist for Clomiphene plus IUI, other authors have considered the existing publications and derived similar conclusions. These same is also true for IUI based on LH timing alone (no drugs).
With regard to the use of oral ovulation induction agents, clomiphene in particular, all the reliable data on cumulative pregnancy rates is derived from women with ovulation dysfunction. In work dating back to the 1960s and validated by Guzick, et al. in the 1980s, women with PCOS type ovulation dysfunction could expect 30-37% cumulative pregnancy rates over the course of 6 attempts. Most of the pregnancies occurred within the first three attempts. These studies were conducted without IUI as part of treatment but they illustrate that, generally speaking, effective therapies work early on in their course or something more potent will be needed to cause pregnancy.
In 1998, Guzick et. al also published a comprehensive review that details what is outlined above. He examined papers on the treatment of couples with unexplained infertility and compiled the data to find average outcomes by treatment choice. Once a patient has had a year or more of infertility, the month to month chance of pregnancy in the future is 1-4%. The table below details the expected outcomes by treatment protocol.
Note that Clomiphene is 'Clomid' and Gonadotropins are injectible drugs containing FSH (follicle stimulating hormone).
In his study, Guzick found that on average, IVF worked no better than 21% of the time. The data he examined was from the late 1980s and early 1990's. IVF outcomes have improved markedly since then and in the most recent analysis from CDC (2007 data) the overall chance of having a baby from IVF is about 35% per try for all patients entering treatment. When broken down by age, patients under 35 have close to a 50/50 chance while 40 year olds have about an 18% chance per IVF cycle.
Which brings us to the main the point of this article...IVF may in fact be the better first choice treatment based on outcome. Given the findings from the Chaffkin paper, patients who decline IVF as primary therapy should do no more than 3 cycles of IUI with gonadotropins before giving IVF a try. Classically, patients would start with Clomid alone and then progress to clomid plus IUI followed by injectibles plus IUI before considering IVF. For a patient who did everything mentioned above 3 times before trying IVF, the total amount spent before IVF would be about 18,000.00. This is more than a single IVF cycle. Though I did not mention it before, the Chaffkin paper showed CUMULATIVE pregnancy rates for all diagnoses other than ovulation dysfunction to be less than 38%. This means that in 2009, using current technology, patients at any age engaging in three IUI cycles with injectible drugs will spend more for treatment than one IVF cycle and wind up with a lower overall chance of pregnancy to boot.
Cost efficiency analyses have been performed by several authors and they all show the same thing: for advanced stages of endometriosis, male factor, severe PCOS (ovulation dysfunction) or tubal disease, IVF should be the first choice of treatment because overall, patients will spend less per live birth than if they chose IUI-based treatment. Even for diagnoses where IUI proved (slightly) more cost effective (unexplained infertility or early stage endometriosis), other considerations should move many couples to IVF quickly.
IUI based therapies with injectibles are notorious for causing higher order multiple births because doctors have poor control of what happens once the eggs drop and the sperm flies north. Though twins are not generally considered a poor outcome, triplets and above are. Almost all of the morbidity and mortality from fertility treatment occurs as a result of higher order multiple birth. Drugs plus IUI therapies are too imprecise to prevent this. In IVF, embryo number can be controlled and the higher-order multiple birth rate kept to a minimum.
IUI-based therapies also fail to preserve the 'extras'. A patient may make 3-5 ripe eggs in a typical IUI cycle. Pregnant or not, the eggs are lost once the cycle is over. In IVF, additional embryos can be cryopreserved for many years. With new freezing technology now available, eggs can be frozen too so that patients can both get the advantage of the 'left-overs' and avoid abandoning embryos in the future. Egg freezing allows patients with religious or ethical concerns about discarding embryos to avoid this quandary altogether.
In sum, IUI is a reasonable therapy for some infertility patients but should be abandoned for IVF by the end of the third IUI cycle at the latest. This is clearly true where the drug therapy is injectible FSH but is also probably true for Clomiphene as well simply because the original data on clomiphene for anovulatory patients showed very few pregnancies past a third to fourth attempt. Unfortunately, reliable data for oral ovulation induction agents like clomiphene in OVULATORY women is not as clear as it is for injectible medication, though Guzick's data shows abysmally poor pregnancy rates for clomiphene with or without IUI in ovulatory patients IUI without any medication has almost no role except in single women or couples with coital failure as IUI for proven infertile couples offers no benefit at all. For patients with concerns about multiple birth who will not accept selective reduction to avoid delivery of higher order multiples, advanced IUI treatment should be avoided altogether.
Consequently, IVF should be considered much earlier in a treatment course than has been the case in classical management. This statement applies regardless of diagnosis or (within reason) the age of the female patient. IVF offers higher pregnancy rates per attempt that the cumulative rates for any IUI therapy. IVF allows cryopreservation of extra eggs and/or embryos and IVF allows the patient and the doctor to better control multiple birth by controlling embryo number for transfer to the host uterus.
Daniel Shapiro, M.D. is a reproductive endocrinologist based in Atlanta, GA