Selecting the Optimal Ovulation Induction Protocol for the Patient Undergoing In-Vitro Fertilization: An Exchange of Ideas

At a symposium held in conjunction with the American Society for Reproductive Medicine’s 58th Annual Meeting, four leading fertility experts gathered to discuss how best to induce ovulation in women undergoing IVF.

This program was supported by an educational grant from Ferring Pharmaceuticals.

Redefining the Roles of FSH and LH in Follicular Development and Maturation

Marco Filicori, MD, Director of the Reproductive Endocrinology Center, University of Bologna, Italy, discussed the growing realization of the importance of luteinizing hormone (LH) in ovulation induction. “Only in the last three to four years,” he noted, “has the role of LH in folliculogenesis been understood.”

He described the two-cell, two-gonadotropin concept: FSH works on granulosa cells, stimulating their proliferation and growth, while LH acts on theca cells, causing them to produce androgens. FSH is active in the early follicular phase, simulating small follicles, while LH takes over in the mid- to late-follicular phase, stimulating large-follicle growth and causing atresia of small follicles.

Therefore, he explained, LH appears to help prevent complications associated with giving steady or rising amounts of FSH, such as multiple gestations and ovarian hyperstimulation syndrome (OHSS).

Dr. Filicori and his colleagues recently completed a series of studies investigating the role of LH in folliculogenesis. They used either human-derived LH (traditional human menopausal gonadotropin, hMG) or low-dose hCG. Giving LH along with FSH accelerated the growth of large ovarian follicles, cut the amount of time needed for controlled ovarian stimulation, and made it possible to give a lower dose of FSH to women with suppressed gonadotropin-releasing hormone agonist activity or secondary amenorrhea and hypogonadotropic hypogonadism.

Also, they found, the amount of LH given was inversely related to the development of many small preovulatory follicles, which is a risk factor for OHSS. And once the ovary had been primed with FSH, it was possible to achieve full oocyte development and pregnancy with LH alone.

Dr. Filicori and his team also looked at the role of LH in premature luteinization, which can affect the outcome of
assisted reproduction by causing endometrial changes. They found no link with LH levels, but rather that premature luteinization actually seemed to be caused by increased activity of the granulosa cells stimulated by FSH.

To prove the importance of LH, Dr. Filicori and his colleagues compared four different regimens in patients undergoing IVF (JCE&M. 2002;87:1156). One group received FSH throughout stimulation. The second group, at day seven, had their FSH reduced from 150 to 50 units per day, and received 50 units per day of hCG. The third group had their FSH dose cut further on day eight to 25 units daily, and received 100 units of hCG daily. In the fourth group FSH was halted altogether, and patients received 200 units of hCG during the last week of stimulation.

“This was sort of bold and we were expecting a high failure rate,” Dr. Filicori noted. “To our surprise, the study turned out to be less traumatic for the patients than we thought.”

There were no differences between the four groups in terms of estrogen and progesterone levels or the development of large follicles. And in patients with any LH activity, small follicles began to disappear before ovulation, thus cutting the risk of OHSS.

“Co-administration of LH and FSH will reduce the frequency of gonadotropin treatment,” Dr. Filicori said. “It provides clinical results in assisted reproduction that are comparable to recombinant FSH; it makes ovulation induction less expensive by lowering the cost of the drug and the monitoring procedure.”

LH and FSH in Ovulation Induction for IVF: Too Little, Too Much, Just Right

William R. Keye, Jr., MD, Director of the Division of Reproductive Endocrinology and Infertility, William Beaumont Hospital, Royal Oak, Michigan, discussed how best to determine the adequate dosage of FSH and LH.

“These requirements vary from woman to woman,” he said, “and it’s impossible to determine the individual threshold and response before treatment.”

“It’s fairly clear,” Dr. Keye added, “that there is a threshold for the amount of FSH given, below which follicular response will be inadequate. But too much FSH can lead to ovarian hyperstimulation syndrome.”

And some women have thresholds for LH as well: those with hypogonadotropic hypogonadism and those whose own LH levels have been profoundly suppressed by administration of a gonadotropin-releasing hormone agonist, which has produced a drug-induced hypogonadotropic hypogonadism.

Studies of women with naturally occurring hypogonadotropic hypogonadism have shown that without LH follicular responses are inadequate. The threshold appears to be somewhere below 75 IU per day.

“There was a clear positive relationship between the dose of LH and estradiol levels and endometrial thickness,” Dr. Keye said.

In women with drug-induced hypogonadotropic hypogonadism, FSH and estradiol levels are profoundly reduced. For some of these women LH levels are also suppressed below the level needed for follicular development, while others retain adequate LH. Women whose LH levels remain adequate only require the administration of FSH for normal follicular development, while those with inadequate levels require LH as well.

The difficulty, Dr. Keye explained, is that it is not yet clear how to determine the threshold below which LH supplementation is needed. “There are two problems in trying to define the threshold. One is the variation in kits used to measure LH, so that the LH measurement from your laboratory may not be the same as the LH measurement in my laboratory,” he said. “And the second problem is whether or not immunoreactive LH is indeed a valid and the most valuable way of measuring biologically active andrelevant LH.”

Most clinicians have found that as much as 300 IU of LH per day for seven days or more does not adversely affect implantation or pregnancy rates. But he noted that the question of whether there is a ceiling of LH activity, and what that ceiling might be, remains unanswered.

“While there are considerable data supporting the concept of thresholds and ceilings for FSH and LH, we cannot describe the exact value nor do we know exactly how much these values vary from woman to woman nor from cycle to cycle,” Dr. Keye concluded.

Use of GnRH Antagonists and Their Impact on Ovulation Induction in
Patients Undergoing IVF

Richard T. Scott, Jr., MD, Director of the Assisted Reproductive Program, Reproductive Medicine Associates of New Jersey, Morristown, spoke about how the use of GnRH antagonists has changed the process of follicular stimulation.

GnRH antagonists, when used appropriately, lead to lower cancellation rates than GnRH agonists, he noted, especially for women over 38 and those with low basal antral follicle counts or a history of decreased gonadotropin responsiveness.

“The GnRH antagonists have changed the way ovulation is induced, but not only in terms of how the prevention of premature LH surges are prevented,” Dr. Scott said. They also cause a more abrupt decrease in both LH and FSH levels, making it possible to delay treatment into the follicular phase. The drugs also produce a faster rise in estradiol, and a faster increase in the diameter of the largest follicle. Patients, however, tend to end up with slightly lower levels of estradiol.

The antagonists tend to shorten treatment by one to two days, thus also reducing the number of injections a patient needs. On average, patients will end up with about 17 fewer injections. “In some sense I think antagonists can be a little more patient-friendly, even though I think outcomes are our most important parameters,” Dr. Scott said.

The risk of ovarian hyperstimulation syndrome and histamine-type releasing problems are about the same with antagonists and agonists, he noted.

While a number of studies have shown a 5% lower pregnancy rate for antagonist-treated versus agonist-treated cycles, Dr. Scott said, at his own center he and his colleagues have seen no such difference.

“The take-home message is that I think implantation rates are equivalent and I don’t see any impairment in our program in terms of outcomes for patients,” he said (Figure 1). “The use of antagonists need not interfere with patient synchronization,” he added.

There have also been concerns about the fact that antagonists can bring down estradiol levels in patients in mid cycle. In his own research, these declining levels appeared not to affect pregnancy in patients younger than 35, but did appear to reduce success among older patients. “And so I think if you can have a stimulation approach that minimizes the prevalence of declining estradiols, I think that is more physiologic and ultimately you’ll see a little bit better outcome for your patients,” Dr. Scott said. “A decline in estradiol, however, certainly does not preclude pregnancy in that cycle and I would not want to imply that at any point.”

Another caveat with GnRH use, Dr. Scott added, is that starting them too early can decrease pregnancy rates; it’s most effective to wait for the sixth day. “Similarly, if your patients respond very quickly and they’re done certainly by cycle day nine or the sixth day of medication, be very hesitant about giving them hCG.”

Also, Dr. Scott said, antagonists are not a wonder drug for patients who are low responders. “The reality is that nothing is a magic cure for those people and this is certainly not either.”

He and his colleagues now tend to prefer using antagonists in their program for low responders, rather than microdose flares, due to a lower rate of rejections.

Choosing the Right Ovulation Induction Protocol for the Difficult Patient Undergoing IVF

Richard P. Marrs, MD, Director of the Center for Assisted Reproductive Medicine, Santa Monica-UCLA Medical Center, spoke on how best to treat patients who have responded poorly to past attempts at ovulation induction.

Not all poor responders are difficult to treat for the same reasons, Dr. Marrs noted. “Poor responders are those who have normal FSH and estradiol levels who fail to produce multiple follicles with stimulation approaches,” he explained. “Some patients are poor responders because they do have adequate or appear to have adequate follicle
development but they have very poor estradiol profiles. And then there’s the poor responder who has fluctuating
FSH levels.”

While clomiphene challenge can be used to identify poor responders, the response in that cycle may not predict future cycles. And while age is a factor, it doesn’t always identify patients who will be poor responders, some of whom are in their 30s.

The first step with such patients is to review their stimulation history. “Every one of those cycles is important to review, even going back to look at patients who are on clomiphene citrate at the very beginning of their stimulation history,” Dr. Marrs said.

“Menstrual cycle length may be important when choosing a stimulation protocol.” Patients with long follicular phases and low estradiol levels at the outset of their ovulatory cycle, for example, may do better with a later stimulation start. “Probably the most important action that you can take outside of reviewing their past stimulations is to look at their ovaries on cycle day two,” said Dr. Marrs. “Looking at the number of basal follicles, ovarian volume, and the appearance of their ovaries on ultrasound will give you some indication of the best choice for these difficult patients.”

And of course no one protocol will work for all difficult patients, he noted. But, certain “trigger points” in a patient’s past stimulation history can help fertility specialists to make the choice.

For example, patients who were on oral contraceptive suppression before starting Lupron® or agonist down-regulation may become oversuppressed.

And for patients who have been on agonists for a long period of time, it’s important to determine whether they might have had a functional cyst. It’s essential to check such patients with an LH-timed ultrasound to follow follicle rupture, and make sure the follicle rupture happens before agonist down-regulation begins. “And then you can shorten their time on agonist and end up hopefully getting a better response,” Dr. Marrs said.

Some patients may need LH but may have only received FSH, while others need an adjustment in their agonist dosage or timing. “I think a lot of these patients are made poor responders even though they didn’t start out in that category,” he said.

Investigating a woman’s ovarian appearance is also essential, especially if a woman tends always to have a single dominant follicle. “That’s an individual that really needs to use a down-regulation scheme,” Dr. Marrs said. Another possibility is oral contraceptive synchronization on day two, he added. “Let her cohort synchronize itself, stop your oral contraceptives, and then go ahead and use your gonadotropin, antagonist approach.”

Some women may also need adjustment in the FSH-LH ratio for their stimulation protocol, making one or the other predominant.

For women with low estradiol and FSH only, converting them to HMG or a combination of FSH-LH can be
effective.

Again, prolonged suppression can make patients more difficult to treat, particularly those who may be extra-sensitive to suppression. “It’s going to take longer to stimulate, it’s going to take more drug to stimulate,” Dr. Marrs said. “So look at these patients in light of what they’ve been prepared with, not just what they’ve been stimulated with, but what happened before they ever started their stimulation drugs.”

Patients with low estradiol can benefit from the “old method” of clomiphene citrate and gonadotropins, he added, and antagonists make this method easier to use.

In his practice, Dr. Marrs said, patients who are “poor responders” are generally asked to come in on their regular cycle, and they’re not given preliminary treatment. On day two, if the patient has minimal follicular asynchrony and evidence of primary follicles, gonadotropins can begin, and antagonists can be added later.

He and his colleagues have stopped using GnRH protocols and microdose flares in poor responders.

“In summary, with this poor stimulating population you need to use every bit of information you can get from their prior stimulations. It will help you make decisions.”

Such patients won’t produce a dozen eggs, he added; “you’ll be lucky if you get three, four, five good quality oocytes in these patients. So we have to readjust what we expect.”

Early on, he added, it’s important to talk to such patients about their options and the possibility of alternatives such as egg donation. “We cannot go on and stimulate over and over and over again in these individuals,” he added. “I think it’s very important that these patients have an individualized approach based on their ovarian function and their ovarian pattern. We need to look at them as individuals and look at their ovarian pattern and make our decisions of how to stimulate.”

All contents Copyright © 1999 - 2003 Medical Association Communications