Doing Something vs. Doing Nothing: Luteal Phase Deficiency and Progesterone Supplementation

Most women who have undergone an infertility workup, and almost all women who have gone through infertility treatment, have probably been tested for their progesterone level at some point in the process. A recent statement published by the American Society for Reproductive Medicine (ASRM) provides a good opportunity to revisit what we know (and what we don’t) about the role of progesterone in infertility treatment.

Broadly speaking, progesterone is the hormone that supports pregnancy. Discovered in the 1930’s at the University of Rochester, pro-gest-er-one was named for its role as the “pro-gestational (i.e. supporting gestation, or pregnancy) steroidal ketone.” During a natural menstrual cycle, as part of the highly synchronized natural interplay between the ovaries and the uterus that occurs each month in order to facilitate the possibility of pregnancy, progesterone is produced starting just after ovulation; the source of the progesterone, conveniently enough, is actually the very follicle left behind by the egg that has been ovulated. This structure, called the “corpus luteum,” can usually be seen on transvaginal ultrasound within a couple of days after ovulation. The main job of the progesterone at this point is to transform the cells in the uterine lining (aka endometrium) into a surface that will be receptive to the implantation of a fertilized egg (aka embryo).

The corpus luteum typically continues to produce progesterone for about 12-14 days; at that point, if no embryo has implanted, the corpus luteum disintegrates, and the resulting drop in progesterone causes a menstrual bleed.  If, however, the egg that was ovulated is fertilized and the resulting embryo implants in the lining of the uterus, the embryo itself produces the hormone hCG (human chorionic gonadotropin) and “rescues” the corpus luteum, allowing it to continue producing progesterone. Progesterone, in turn, supports the continued growth of the pregnancy. Quite poetically therefore, it turns out that the pregnancy needs progesterone as much as the source of the progesterone (the corpus luteum) needs the pregnancy (for hCG). If the progesterone production is somehow interrupted -- for example, if the ovary containing the corpus luteum is surgically disturbed or removed during this period of time (up to about 8 weeks gestation, at which the point the placenta takes over progesterone production), the pregnancy will most likely fail, resulting in miscarriage. 

Given the critical role that progesterone plays in “supporting” pregnancy, it has long been speculated that progesterone deficiency, referred to as “luteal phase deficiency (LPD),” might be an important cause of infertility in couples who have no other obvious issue. As the recent ASRM statement confirms, however, whether or not LPD is a real cause of infertility in otherwise healthy women is doubtful at best. Furthermore, even if we assume LPD is real, there is little convincing evidence that treating LPD (typically with supplemental progesterone) will improve the situation. An important exception to this statement is in IVF/ART cycles (as opposed to non-medicated cycles or IUI cycles), when luteal function is clearly interrupted, and supplementing progesterone has been shown to be important. (Tangentially, new data actually questions whether we need to continue the progesterone supplementation as long as we do)

ASRM summarizes our current knowledge about LPD as follows:

1. Abnormal luteal function may occur as the result of a medical condition (e.g., elevated prolactin, abnormal thyroid function), and infertile women should be investigated for these disorders with appropriate treatment of identified conditions.
2. No diagnostic test for luteal phase insufficiency has been proven reliable in a clinical setting. The roles of basal body temperature, luteal progesterone levels, endometrial biopsy, and other diagnostic studies have not been established, and performance of these tests cannot be recommended.
3. No treatment for luteal phase insufficiency has been shown to improve pregnancy outcomes in natural, unstimulated cycles.
4. Luteal support after ART procedures with progesterone or hCG improves pregnancy outcomes, but hCG increases the risk of OHSS.
5. There is no proven role in adding progesterone or hCG for luteal support once a pregnancy has been established. Use of supplemental progesterone, in a non-ART cycle beyond the time of expected menses (i.e., 2 weeks after ovulation), is not proven beneficial.

References:

1. The Clinical Relevance of Luteal Phase Deficiency. Practice Committee of ASRM. Fertility and Sterility 2012.

2. Early progesterone cessation after in vitro fertilization/intracytoplasmic sperm injection: a randomized, controlled trial. Kohls et al. Fertility and Sterility 2012.

Measures of Success

Fertility medicine is quite remarkable for its almost obsessive focus on quantitative outcomes. This is probably partly due to the binary nature of what we do (pregnant vs. not pregnant, baby vs. no baby, etc.) and partly to the elective (and therefore competitive) nature of the “fertility business,” a trend that was undoubtedly intensified by the 1992 federal law requiring all IVF clinics to report their outcomes to the CDC.

While “pregnancy rate” (percent of women getting pregnant per IVF attempt) has historically been the mainstay of how outcomes are reported, many fertility specialists advocate alternate measures of success.   One approach emphasizes the importance of a baby (“live birth”) over a mere pregnancy (with the difference being miscarriages and stillbirths). In a different vein, others have advocated adjusting the time frame from “per cycle” success to “cumulative” success, arguing that what is really most important to a patient is what the chances are that at the conclusion of a course of treatment, she will achieve success. This last point is particularly germane in an infertile population, as there is usually some degree of “natural selection” that occurs over the course of multiple treatment attempts: the “least infertile” couples (i.e. those with the most mild problems) tend to get pregnant sooner, leaving behind an increasingly infertile group to try again the next time. Statistically, this often leads to worse outcomes in subsequent treatment cycles (one can think of this as “diminishing returns” on the treatment investment after multiple failed attempts).

Two large studies using newer measures of treatment success have been published recently (links below). The first, published in the New England Journal of Medicine, analyzed cumulative live birth data from the national (U.S.) database compiled by the Society for Assisted Reproductive Technology. The authors “linked” data belonging to individual women who underwent multiple IVF cycles and calculated cumulative success rates after multiple attempts. They found several trends:

a)       Amongst women using their own eggs, cumulative success increases dramatically in the first 3 attempts and increases more slowly during IVF attempts #4-6 (very few women pursued 7 or more cycles)

b)      Whereas age is a big factor in cumulative success amongst women using their own eggs (i.e. younger women have more success), “older” women using donor eggs do not have significantly decreased cumulative success despite their age, as compared with younger women using donor eggs

c)       Blastocyst (day 5 or 6) embryo transfers are associated with higher cumulative success than cleavage stage (day 2 or 3) embryo transfer (Although this may not be a true cause-and-effect relationship, since the women having blastocyst transfers tend to be younger and/or have more available embryos from which to choose)

The second study, published in Fertility and Sterility, looked specifically at cumulative success in egg donation cycles, and used a new measure they termed “cumulative newborn rate per number of embryos transferred” – a rate using number of newborns (as opposed to pregnancy) as the numerator and number of embryos transferred -- instead of number of treatment cycles performed --  as the denominator. This outcome measure will thus take into account the efficiency of the treatment, recognizing that transferring 3 embryos in one cycle is not the same as transferring one embryo in one cycle.

This group of researchers showed that most of the babies (65%) are achieved with the first 5 embryos transferred, 85% with the first 15 embryos, and cumulative success ultimately levels off between 90-95% after 16-25 embryos have been transferred. Interestingly, and in accordance with many previous studies (such as the one mentioned above), neither the age of the recipient nor the diagnosis (i.e. reason for pursuing oocyte donation) had any significant influence on cumulative outcomes.

1. Cumulative birth rates with linked assisted reproductive technology cycles.
2. Cumulative newborn rates increase with the total number of transferred embryos according to an analysis of 15,792 ovum donation cycles.

Jumping the Gun: Progesterone Levels Before Oocyte Retrieval and IVF Success

While every IVF clinic does things a bit differently, there is broad consensus that the two core components of “monitoring” response to medication during an IVF cycle include:

1. Blood estradiol
2. Follicle growth as measured by transvaginal ultrasound

That said, there are multiple reasonable approaches to IVF; some clinics also like to check other hormones, such as FSH, LH, and progesterone at certain times during the IVF stimulation.

Three recent studies (links below) highlight growing evidence supporting the importance of checking progesterone during an IVF cycle, particularly in the latter part of the cycle (after 6 or 7 days of medication). Progesterone is the hormone that causes the uterine lining to become receptive to pregnancy (“pro-” i.e. supporting, “gest” i.e. gestation/pregnancy). In normal physiology, progesterone is produced mainly by the corpus luteum, the structure resulting from the follicle after the egg has been released. By definition, therefore, progesterone action is timed by nature to kick in after ovulation, when a conceptus may be looking for a place to implant.

In an IVF cycle, however, due to the growth of multiple follicles (rather than just one, as nature would have it), progesterone levels sometimes start to rise prematurely – before ovulation, and even before the oocyte retrieval is performed. Early exposure to elevated progesterone levels is theorized by some to have detrimental effects on an embryo’s ability to implant. In other words, even the slight increase in progesterone levels caused by the IVF stimulation may throw off the synchronicity of the embryo and the lining enough to decrease chances of pregnancy.

There have been multiple studies aiming to address this subject in the past 10-20 years, and the results have been inconsistent. Many of these have been small studies using different cutoffs for what is considered an “elevated” progesterone level. Two of the studies cited below are amongst the largest ones that demonstrate a link between elevated progesterone and decreased chances of IVF success. The third, and perhaps most intriguing, takes this idea of “progesterone timing” even further and looks at the duration of progesterone elevation in addition to the amount. Taken together, I believe these studies warrant serious consideration when there is evidence of a premature progesterone rise prior to egg retrieval in an IVF cycle. While no studies address the question of what is the best plan of action in this situation, one logical option would be to forgo a fresh embryo transfer in order to allow the lining to “recover,” and plan for a frozen-thaw cycle with a healthy, well-timed lining 1-2 months later. 

1. Subtle progesterone rise on the day of human chorionic gonadotropin administration is associated with lower live birth rates in women undergoing assisted reproductive technology: a retrospective study with 2,555 fresh embryo transfers.
2. Premature progesterone rise negatively correlated with live birth rate in IVF cycles with GnRH agonist: an analysis of 2,566 cycles.
3. The duration of pre-ovulatory serum progesterone elevation before hCG administration affects the outcome of IVF/ICSI cycles.