Oocyte Donation

Infertility secondary to absent or irreversibly abnormal oocyte development was once considered to present an absolute barrier to fertility. Women with intractable oocyte disorders were told that they were functionally sterile and that their only option for building a family was adoption. The development and refinement of the assisted reproductive technologies in the 1980s provided clinicians and scientists with another means of helping these women. In vitro access to human oocytes and embryos allowed embryos that result from a healthy oocyte produced by one woman to be transferred to the uterus of another, where they may implant, develop, and ultimately lead to the delivery of a healthy child.

The first report of a successful pregnancy established in one woman (the recipient) with an oocyte from another (the donor) was in 1983. The original technique described by Buster involved intracervical artificial insemination of a normal female volunteer with the infertile couple's husband's spermatozoa, uterine lavage during the perinidatory period, and subsequent intrauterine embryo transfer to a synchronized recipient.¹ Although revolutionary, the technique had several shortcomings. Concerns regarding infectious diseases; technical difficulties, which meant that many of the cycles did not lead to recovery of a viable embryo; relatively low implantation rates, even when an embryo was available; and the possibility of pregnancy in the donor limited the application of this technique.

Oocyte donation is now achieved most often through in vitro fertilization (IVF). In this circumstance, controlled ovarian hyperstimulation of the donor is used to increase the number of mature oocytes available at the time of retrieval. IVF is then completed in a routine fashion with the obvious exception that the transfer is to the recipients' endometrial cavity. Many of the various assisted reproductive technologies have been used successfully to achieve donor oocyte pregnancies. These include GIFT, ZIFT, TEST, and many others.

Special management issues in oocyte donation cycles include the need for embryo-endometrial synchronization; maintenance of hormonal support of the endometrium until after the physiologic luteal-placental shift; and the myriad of issues surrounding recruitment, selection, and screening of oocyte donors. Successful oocyte donation requires an established assisted reproduction technology program and a specially trained team of physicians, nurses, social support personnel, and embryologists. Once in place, these programs may enjoy exceptional clinical success. Several thousand babies who developed from donor oocytes have been born throughout the world.

The development of oocyte donation in humans began in the early 1980s, but development of these techniques may be found in the animal literature and dates back to the late 19th century. Heape authored the first published description of a successful embryo transfer in 1890 using the rabbit model.¹ High-volume animal experimentation began in the 1950s after the description of a successful bovine embryo transfer in 1951. Commercial application of these technologies is now routinely applied in cattle and sheep. In these animals, synchronization was initially achieved via hormonal manipulation of the cycles of both the donor and the recipient animals. More recently, cryopreservation has allowed thawing and transfer in subsequent recipient cycles, permitting maximal synchrony while significantly simplifying the logistics of these cycles.

The first description of successful embryo donation to primates was by Meyer¹ in 1968. Subsequently, other reports of pregnancies in castrate monkeys receiving sequential estrogen and progesterone replacement followed. These studies showed that exogenous steroids permitted the development of an endometrial environment that was receptive to implantation.

The human oocyte donation literature contains a number of “firsts.” The first pregnancy was reported by Buster in 1983 and used the uterine lavage technique without IVF. The first IVF-related pregnancy was reported the same year by Trounson in a woman with functional ovaries who was inaccessible secondary to severe pelvic adhesive disease. Regrettably, that pregnancy miscarried in the second trimester. By 1984, a pregnancy was reported in a woman whose endometrium had been prepared by exogenous hormone replacement therapy.

The indications for oocyte donation are all related to the lack of availability of high-quality oocytes capable of producing healthy offspring. Functionally, four broad categories of patients may be identified. These are ovarian failure, genetic abnormalities in the recipient that she would like to avoid passing on to her progeny, the depletion of high-quality oocytes (diminished ovarian reserve), and inaccessible ovaries.

Ovarian Failure

Women may have ovarian failure develop for a variety of reasons. Among these are X chromosome deletions or mosaicism; iatrogenic interventions such as surgery, radiation, and chemotherapy; autoimmune disease; idiopathic gonadal dysgenesis; Savage syndrome (resistant ovary syndrome); and idiopathic premature ovarian failure (POF). Once the diagnosis has been made, most of these patients generally require oocyte donation to be able to conceive. Some patients may have intermittent remissions, and conception with their own oocytes may occur. However, the substantial majority of the time, these patients require oocyte donation to conceive.

Genetically Transmitted Disease

Women who carry specific heritable disorders may be candidates for oocyte donation. These include a variety of different classes of disorders. Straightforward indications such as single gene mutations (e.g., sickle cell, cystic fibrosis) and balanced translocations remain a major indication. This is despite the fact that preimplantation genetic diagnosis is becoming available for larger number of disorders and acceptable pregnancy rates have been shown. As preimplantation genetic diagnosis technologies improve and probes become reliably available for a larger number of disorders, this indication may become less common. In those cases in which specific probes are not available, oocyte donation remains an excellent treatment option.

Diminished Ovarian Reserve

The physiologic decline in reproductive potential correlates temporally with the processes of follicular depletion and the development of diminished oocyte quality. This is why women ovulate approximately the same number of times per year through their reproductive life but have dramatic differences in natural fecundity (with or without treatment). At some point, all women deplete their ovarian reserve to the point that the reproductive potential of their oocytes is so low that they are very unlikely to conceive. These women are said to have diminished ovarian reserve.

Diminished ovarian reserve may be identified in several ways. Women with elevated basal follicle-stimulating hormone levels, repetitive poor responses to gonadotropin stimulation, recurrent pregnancy loss (especially those with advanced age or with concepti that have been shown repetitively to be aneuploid), and those who have not responded to multiple treatment cycles with impaired embryo development may all be considered to have diminished ovarian reserve and are superb candidates for oocyte donation.

Inaccessible Ovaries

A common indication for oocyte donation in the early 1980s was the presence of severe pelvic adhesive disease or other problems that precluded access to the ovaries for oocyte retrieval. The development of high-quality sonographic imaging equipment has all but eliminated this as an indication for oocyte donation. There is still the occasional individual whose ovaries are not safely accessible to either ultrasound-directed or laparoscopic retrieval who is a good candidate for oocyte donation.

Pretreatment evaluation generally begins by counseling the patient that she is a candidate for oocyte donation. More specifically, the patient is generally told that she is not a good candidate for other treatment methods that would allow preservation of her own genetic contribution. The specific nature of the counseling is highly dependent on the clinical setting. This often is the most difficult portion of the entire process for many women, especially those with advanced age but otherwise normal gynecologic and menstrual histories who are far into the physiologic process of depleting their ovarian reserve. In contrast, it may be simple in women who have ovarian failure or who are seeking treatment alternatives because they are carriers of a genetic disease.

The procedural aspects of donor oocyte IVF should be reviewed with the couple and should cover the timeline for treatment, risks, and cost. Donor oocyte success rates for the individual center should also be presented.

The next step in the process is to determine that the recipient is a good candidate to become pregnant if this was not already accomplished during the couple's initial evaluation. It is intuitive that both the oocyte recipient and her partner should be relatively healthy and there should be no physical contradictions to pregnancy. What exactly constitutes “sufficient” health is quite difficult to define with rigor. Any patient with a medical problem for which he or she would require regular medical care should be counseled in detail about the additional risks that occur with pregnancy. The clinician should quickly consider perinatal or other appropriate consultation in those settings in which he or she is not completely comfortable counseling the patient (and providing documentation of that counseling) regarding the potential risks in pregnancy.

Physical examination of the female should include a pelvic examination, including a trial embryo transfer or “catheter check” to determine uterine depth and, most important, the ease with which the endocervical canal may be negotiated. A Papanicolaou smear and cervical cultures also are obtained at this time. Screening for infectious diseases in the patient and her partner should include human immunodeficiency virus (HIV; I, II, and human T-cell leukemia [HTLV-1]), hepatitis, and syphilis.

Some form of evaluation of the uterine cavity is important. Methods include hysterosalpingography, office hysteroscopy, and saline hysterosonography. Evaluation and treatment of suspected abnormalities are equivalent to those used for IVF patients. Any intrauterine pathology that could compromise the chances for implementation or increase the risk of pregnancy wastage should be treated aggressively.

The male partner should have a semen analysis to rule out a coexisting male factor. If previous IVF data are available, fertilization should be assessed, and a decision should be made as to whether there is an indication for intracytoplasmic sperm injection using routine clinical criteria.

In the natural cycle, endometrial function is directly linked to the sex steroid milieu created by the dominant follicle and corpus luteum. This provides a natural synchrony between oocyte/embryo development and the endometrial changes that allow implantation, although many recipients continue to have natural cycles. The requirement for synchrony between these two processes mandates that control be maintained over endometrial development and maturation.

A series of elegant studies showed that the endometrial receptivity is present for a limited time each cycle. This “window” of implantation lasts for approximately 3 days, although some authors have described that it may be as wide as 5 days. More important, this window of receptivity is related principally to the duration of progesterone exposure. An elegant study by Navot and coworkers² showed great flexibility in the length of proliferative development of the endometrium with very short and very long durations of estrogen priming, resulting in similar pregnancy rates when appropriate doses and durations of progesterone were provided before embryo transfer. This is quite similar to the model that nature provides in oligo-ovulatory women. They may have prolonged enestrogenic intervals before ovulating and then conceiving. The proliferative phase varies markedly, but the consistency in the duration of progesterone exposure and the window of receptivity remains fixed.

Women with ovarian function are downregulated with a gonadotropin-releasing hormone (GnRH)-agonist. Obviously, women with ovarian failure require no GnRH-agonist treatment. Exogenous hormonal replacement is then used to simulate the natural cycle and to effect perinidatory uterine receptivity.

A variety of steroid replacement regimens have been developed, all designed to approximate the pattern of hormone secretion seen in the natural cycle. These include oral and intramuscular estradiol valerate, micronized estradiol-17, estradiol-impregnated polysiloxane vaginal rings, transdermal estradiol patches, and gels. There are no studies showing higher pregnancy rates with the varying routes of administration. This is despite the fact that there may be marked differences in circulating estrogen levels. Estrone levels may be 10-fold higher in oral estrogen-treated cycles than with transdermal or transvaginal replacement. Replacement protocols result in late follicular and midluteal estradiol concentration of approximately 200 and 100 pg/mL, respectively.

Progesterone usually is administered as intramuscular progesterone-in-oil or micronized progesterone vaginal suppositories. Midluteal progesterone levels are typically 20 ng/mL or greater.³

A progesterone vaginal gel that adheres to the vaginal mucosa and provides very high local levels of progesterone at the endometrial level recently has been introduced and has shown promising results. Larger clinical experiences are required to determine whether it is clinically equivalent to other forms of progesterone replacement.

At most programs, the recipient undergoes a preparatory cycle before the actual treatment cycle to ensure the adequacy of her response to the replacement regimen. Serial serum estradiol and progesterone determinations are performed, and ultrasonographic assessment of adequate endometrial proliferation is accomplished. If normal, no further evaluation is required. In those cases in which the sonographic assessment is abnormal or unclear, a midluteal endometrial biopsy specimen is obtained on cycle days 21 through 25 and dated according to histologic criteria of Noyes and coworkers.⁴ Replacement protocols frequently result in midluteal endometrium displaying glandular-stromal dysynchrony, with glands lagging behind appropriately advanced stroma; when the stroma is in phase, pregnancy is normal.

Perhaps the greatest difficulty in establishing and maintaining a donor oocyte program is the limited availability of oocyte donors. Donors come from a variety of sources but are most commonly anonymous. Some patients are fortunate enough to have a close family member or friend who is a good candidate to be a donor. There are even episodes in which identical twin sisters have donated to each other.

Anonymous donors are generally recruited by the program and are young, healthy volunteers who are undergoing treatment specifically to be an oocyte donor. Other sources include IVF patients who are willing to donate a portion of their oocytes (typically in exchange for having their cycle partially or fully paid for by the recipient couple) and women scheduled for laparoscopic tubal sterilization who volunteer to undergo preoperative superovulation and monitoring, with oocyte harvest at the time of their surgery.

The availability of embryo cryopreservation has diminished the pool of IVF patients willing to donate oocytes, as most elect to freeze any “excess” embryos for possible future replacement. Anonymous volunteers, family members, and close friends compromise the majority of donors at most centers.

The relative scarcity of donors has led to a search for other sources of oocytes. Cha and associates⁵ recently described the recovery of immature oocytes from unstimulated ovaries and ovarian biopsy specimens removed at the time of surgery. These immature oocytes were matured in vitro and subsequently inseminated. The transfer of five embryos to a woman with POF resulted in a viable triplet pregnancy. These data should be considered preliminary and have yet to be reproduced at other centers. Pregnancy rates, cost, availability, and many other clinical and ethical issues remain to be explored before proceeding with this type of care on a routine basis.

Donor Screening Requirements

The American Society of Reproductive Medicine provides guidelines regarding appropriate screening of donors. These recommendations provide the minimum screening, which is done at many centers. Additional screening is based on the needs of the recipients relative to their medical histories (e.g., carrier states for various genetic diseases) or any specific problems that may be in question in the donor's medical, personal, or family histories. All donors require a complete history and physical, detailed review of genetic and infectious disease history, a psychologic evaluation, and extensive counseling regarding the risks of the procedure.

A detailed menstrual and pregnancy history is important since potential donors who have histories of infertility, recurrent miscarriages, or very irregular cycles may be suboptimal candidates. The protocol of the Cornell group is as follows and is comprehensive. The laboratory evaluation should include a complete blood count and sedimentation rate to identify occult infection. In selected cases, hemoglobinopathies such as sickle cell trait should be ruled out. The screen for infectious disease should, at minimum, include serology for HIV, HTLV, hepatitis B and C, and syphilis and cervical cultures for chlamydia, gonorrhea, and mycoplasma. A urine toxicology screen can provide an additional screen for high-risk lifestyles such as substance abuse. Genetic screening should include a detailed family history. We include cystic fibrosis carrier testing in all donors. Additional specific genetic or medical testing is done if indicated by the donor's personal or ethnic history. Examples would be sickle cell screening in donors of African heritage and Tay-Sachs screening in donors who are Ashkenazi Jews.

A psychological evaluation is done by some type of written personality assessment or inventory to identify donors with any psychological instability. There are obvious limits to these tests, so that donors should also be interviewed by a mental health professional to ensure that they have no obvious psychopathology and that they understand all the implications of being an oocyte donor and are comfortable in proceeding. Donors with any positive findings on medical or psychological testing should be excluded. The goal is to have a healthy donor before and subsequent to donation who perceives the donation experience as a positive event with minimal risk to herself.

Pregnancy rates are discussed below but closely parallel those of women undergoing IVF with normal endometrial development who are the age of the donors. IVF rates are equivalent at or younger than 32 years of age, and this is also the best age group for donor candidates. Older donors are sometimes used, especially when they are family members of the recipient. This is perfectly acceptable as long as the recipient is sufficiently counseled about the impact of donor age on pregnancy rates and pregnancy outcome. It is important to inform the donor regarding the probability of the cycle leading to a successful pregnancy. They need to understand that although it is likely that a child will result from the cycle, some cycles are not successful. Issues of disclosure to the donor as to the outcome of the treatment cycle vary from cycle to cycle. We have elected not to tell our donors the outcome of their cycles, not because we had legal concerns regarding parenting but rather because many felt very guilty if the recipient failed to conceive. Most donors feel very good about providing another couple with an opportunity to build their family. Disclosing a negative cycle may deprive them of the personal satisfaction and ego strengthening that may accompany being a donor.

Finally, informed consent is an important part of this process and mandates that donors be very well-informed about the process and any real or putative short- or long-term risks with the process. The short history of oocyte donation mandates that no long-term data regarding the psychologic implications of oocyte donation are available. Good data are available from sperm donation programs and indicate that donating has not produced long-term problems. The uncertain long-term effect of controlled ovarian hyperstimulation on ovarian cancer risks should prompt all centers to strictly limit the amount of cycles for any single donor (typically less than 4). Many centers are now trying to establish a national registry for oocyte donors to provide a central registry regarding the number of cycles that these women have completed and their outcomes.

Stimulation and Retrieval of the Donor

Stimulation of an oocyte donor is equivalent to stimulation of an IVF patient of the same age. Each center has its own practice pattern and is familiar with what is most successful in its own practice. Particular attention should be paid to avoiding an exaggerated response that may increase the risk of ovarian hyperstimulation syndrome.

Oocyte retrieval is performed routinely with ultrasound-guided vaginal aspiration with the patient under intravenous sedation in a fashion similar to all IVF patients. A follow-up ultrasound is required of the donor at her next menses to ensure that she is doing well and that she returned to normal.

Embryonic-Endometrial Synchronization

The very nature of oocyte donation mandates that ovarian and endometrial events are independent and require synchronization. Once a source of major anxiety, the process is now straightforward.

A significant experience published in the animal literature indicated that a maximally efficient window of transfer exists. Elegant work by Navot and coworkers⁵ showed that the transfer of embryos (approximately 48 hours after retrieval) into endometrium histologically developed to days 17 to 19 is most likely to result in implantation and pregnancy. In this setting, day 15 is equated to the day of ovulation and is the same as the first day of progesterone exposure. Thus, “day 2” embryos are transferred on the third day of progesterone treatment, “day 3” embryos are transferred on the fourth day of progesterone treatment, and blastocysts or “day 5” embryos are transferred on the sixth day of progesterone replacement. The window of transfer is wider than a single day, but synchronous transfer provides a margin of safety in the process in the event that endometrial development is accelerated or retarded by the exogenous replacement that has been prescribed.

The flexibility that is possible with the duration of estrogen treatment during the proliferative phase of the recipient's cycle is an important practical matter during actual treatment cycle. Because it is impossible to state precisely when the donor's oocytes will be ready for transfer before initiating her stimulation, the recipient is generally starts taking her estrogen replacement at the same time or in advance of the donor initiating her stimulation.

This allows ample time to achieve stimulation of normal proliferative development of the endometrium before the donor going to retrieval. Progesterone replacement is initiated on the day of the donor's retrieval to allow synchronization. Some programs initiate progesterone the day before the donor retrieval and some as late as the day after. This likely reflects that the window of receptivity is at least 3 days wide.

Clinical Results of Oocyte Donation

Success rates in ART are proportional to the age of the woman producing the oocytes, and oocyte donation is no exception. The oocytes are derived from young (younger than 32 years old) women, and thus the clinical pregnancy and delivery rates are extremely high. Pregnancy rates in our oocyte donation program are equivalent to those seen in young IVF patients who have normal endometrial development.

An analysis of the 1996 statistics of the United States IVF Registry (most recent that have been published) showed a 45.1% clinical pregnancy and 39.1% delivery rate per transfer, resulting from 3345 donor oocyte embryo transfers.⁶ These statistics are clearly favorable when viewed in the context of overall per-transfer IVF clinical and viable pregnancy and delivery rates of 33.3% and 27.9%, respectively, reported in the same survey.

Pregnancy rates in better programs now exceed 65%. Regrettably, multiple pregnancy rates are quite high and exceed 50% in many programs with the substantial majority of these being twins. The development of blastocyst culture systems that allow routine culture for the first 5 days of development may allow better embryo selection and, thus, enhanced pregnancy rates with fewer embryos transferred with lower multiple pregnancy rates. Further experience is required to document clinical outcomes. Recent experience in some centers has shown pregnancy rates that exceed 80% in those cases in which two high-quality blastocysts are available for transfer.

Oocyte donation provides an elegant model for studying the temporal window of transfer and, by inference, the window of implantation in humans, as embryos of known developmental stage can be transferred to histologically defined endometrium (albeit not physiologic) following differing days of progesterone exposure. Analysis of such data has shown that the transfer of embryos with two or more blastomeres to endometrium that is histologically developed to days 17 to 19 of the idealized 28-day cycle permits implantation and viable pregnancies.⁷ Assuming that a six- to 16-cell embryo develops to the blastocyst stage within 2 to 3 days after transfer, these results suggest that the window of endometrial receptivity in humans does not extend beyond day 22 or 23. Bergh and Navot timed implantation by an ultrasensitive human chorionic gonadotropin assay and confirmed this “window of receptivity.” These and other studies indicate an implantation window of approximately 3.5 days.⁸

At least one study has suggested a window of up to 7 days; Formigli and coworkers⁹ obtained donor embryos by uterine lavage 5 days after ovulation and suggested that blastula may implant into day 22 endometrium. This report has many shortcomings, including a failure to precisely determine the timing of ovulation and the inability to exclude the possibility of spontaneous pregnancy in patients with ovarian function.

It must be stressed that conclusions about uterine receptivity that are derived from these data rely on assumptions regarding in vivo cleavage rates of embryos resulting from in vitro fertilized oocytes (i.e., the time required for the attainment of embryonic implantation competence). This problem may be obviated if advances in embryo culturing techniques evolve to permit the transfer of fresh blastocysts. Further, although hormonal replacement protocols permit histologic definition of the endometrium, the inadequacy of currently available systems for grading embryos limits the ability to accurately define embryo quality. Acknowledging these limitations, oocyte donation has afforded a unique opportunity to investigate the temporal window of uterine receptivity.

Luteal-Placental Shift

Donor oocyte IVF provides a unique human in vivo model for studying the luteal-placental shift, as recipient subjects with POF have no endogenous ovarian function. In the early 1970s, Csapo and colleagues¹⁰ estimated that placental “takeover” of steroidogenesis occurs during the eighth week of gestation based on a study of lutectomized pregnant women. More recently, we have studied a group of nine patients with ovarian failure who conceived after oocyte donation. Hormone replacement consisted of oral or transdermal estradiol coadministered with intramuscular progesterone-in-oil, resulting in both constant and physiologic serum levels. Serum estradiol and progesterone concentrations were assayed serially, and statistically significant mean increases in these levels were noted by the fourth and fifth weeks after embryo transfer, respectively (sixth and seventh weeks of gestation by last menstrual period). Linear regression analysis of steroid concentrations versus gestational age was performed, and the intersection with basal levels was determined, suggesting onset of placental steroidogenesis during the third week after transfer. Although this model is imperfect, given the clinical necessity for exogenous sex steroid administration for early pregnancy maintenance, it nonetheless provides valuable insight into the time course of development of placental steroidogenic competence.

As illustrated by the above-mentioned study, exogenous hormonal support is required during the interval preceding the luteal-placental shift. Monitoring of peripheral estradiol and progesterone concentrations is continued once pregnancy is documented. The onset in adequate steroidogenesis is marked by a significant rise in these steroid levels, at which point the dosage of the replacement hormone is halved. Estradiol and progesterone levels are reassessed in another week and if a continued rise is observed, hormonal therapy is discontinued. Generally, hormonal replacement is unnecessary beyond the seventh developmental week. Endovaginal sonographic assessment is also done to exclude an extrauterine gestation and to evaluate the number of implantation sites that are developing appropriately.

Understanding Clinical Success and Failures in Recipients

The most important variable in oocyte donation is the number and quality of the oocytes available for the IVF portion of the cycle. Thus, a failure with one cycle does not preclude success in a subsequent cycle. A recent review has shown that there is no diminution in pregnancy rates with subsequent cycle up through and including at least four attempts.¹¹ Similar data are available from our program.

Interesting insights may be gained by close scrutiny of failed oocyte donation cycles. The most common cause of failure may be suboptimal embryo development. Failed cycles have a higher prevalence of morphologic abnormalities among the oocytes and embryos. This is true even when comparing successful and unsuccessful cycles from individual donors. Although not controlled for any male contribution to embryo quality, these data suggest that even young, healthy donors who undergo similar stimulations are not equivalent for each cycle and that cycles may possess intrinsically greater or lesser reproductive potential than others. Recipients should be counseled that donating oocytes by no means ensures that a large number of highquality oocytes will be available each and every cycle.

Ovum donation has also proved a useful model in addressing a major controversy in relation to the age-related decline in female fecundity (i.e., the relative importance of uterine versus oocyte senescence). Success rates following conventional IVF decline significantly after the age of 40 years, and viable pregnancies are infrequent beyond the age of 42 years. Oocyte donation permits dissociation of uterine and oocyte age. Sauer and coworkers¹² published a series of seven patients 40 to 44 years of age (mean age, 41.4) who underwent appropriate hormone replacement and transfer of embryos derived from oocytes from donors younger than 35 years of age. They reported a per-embryo implantation rate of 35.7%, with an overall clinical pregnancy rate of 75%, which was not significantly different from success rates seen in donor egg recipients younger than 40 years of age (mean age, 31.8 years).

There have since been numerous reports of viable pregnancies conceived by premenopausal and menopausal women via oocyte donation.^10,13,14,15 Pregnancy and delivery rates were superior to women older than 40 years who attempted IVF without oocyte donation. However, unanswered by these studies was whether there is a phenomenon of uterine senescence.

One study examined 24 recipients with a mean age of 40 years versus 24 IVF patients with a mean age of 32 years who provided the oocytes for donation. The implantation rate for the younger patients (42.5%) was significantly higher than the implantation rate in the older recipients (25.6%) who received embryos derived from the same cohort of oocytes.¹⁶ Cario and coworkers¹⁷ prospectively studied recipient age via a shared matching oocyte donation program in which a cohort of eggs were distributed to a recipient younger than 40 years and a recipient older than 40 years. They found no difference of implantation rates but a significantly higher miscarriage rate in the older recipients. Borini and coworkers¹⁸ also reported on the implantation and pregnancy rates for younger and older recipients receiving the same cohort of eggs at donation. They found a significantly lower pregnancy and implantation rate for the recipients 40 to 49 years old compared with the younger recipients of this shared matching program.

A review of our clinical data for all recipients showed a significantly lower implantation rate for recipients older than age 43 years compared with those younger than 43 years. Although a lowered implantation rate may limit the multiple pregnancy rate, we have maintained uniform pregnancy and delivery rates across the age groups with the transfer of three embryos.¹⁹ Analysis of outcome variables according to etiology of ovarian failure showed no impact of clinical diagnosis on pregnancy, miscarriage, or implantation rates. Our recipients have been divided into five diagnostic subgroups: POF, incipient ovarian failure, physiologic menopause, poor responder in IVF, and cancer survivor. Given the great potential for pregnancy success for any promising recipient following oocyte donation, it appears that psychosocial and obstetric consequences, rather than medical indications, serve to limit the provision of oocyte donation.

Oocyte donation is an important clinical method that is now widely available. It is a highly successful approach to the management of infertility caused by ovarian failure. A better understanding of reproductive senescence has led to the extension of indications for this technology and now includes poor oocyte quality, inaccessible ovaries, advanced age, and carriage of potentially transmissible genetic disorders. Indeed, the limited availability of oocyte donors appears to be the major limiting factor in the utilization of this type of health care.

It is likely that oocyte donation will eventually become an obsolete technology for most patients with the obvious exception of those with absolute ovarian failure. Advances in our understanding of reproductive senescence and pathologic changes in oocytes will likely allow directed treatment with patients' own oocytes. Until that time, this allows patients to experience pregnancies, maintain the paternal genetic contribution, and achieve their ultimate goal of building a family.