Ovarian age-based stimulation of young women with diminished ovarian reserve results in excellent pregnancy rates with in vitro fertilization

Transcription

1 Ovarian age-based stimulation of young women with diminished ovarian reserve results in excellent pregnancy rates with in vitro fertilization Norbert Gleicher, M.D., and David Barad, M.D. Center for Human Reproduction, New York, New York, and The Foundation for Reproductive Medicine, Chicago, Illinois Objective: To determine whether pregnancy outcome can be positively affected if the ovarian stimulation in women with diminished ovarian reserve (DOR) is changed. Design: Case control study. Setting: Medical school-affiliated private infertility center. Patient(s): Sixty-two women, aged years, with the diagnosis of DOR (study group) and 62 age-matched controls with apparently normal ovarian function (control group). Intervention(s): Study group patients received a modified (ovarian age-based) stimulation with microdose GnRH agonist (GnRH-a) and IU of gonadotropin daily. Women in the control group received a standard chronological age-based stimulation with long luteal phase GnRH-a and up to 300 IU of gonadotropins (two patients in this group received an antagonist in place of an agonist). Main Outcome Measure(s): The IVF cycle outcome parameters, including pregnancy rates (PR). Result(s): Women with DOR and controls did not differ significantly in patient profiles or underlying infertility conditions and received identical embryo numbers at transfer. The DOR patients demonstrated a strong trend toward lower gravidity. Peak E 2 levels were similar between groups but controls produced significantly more oocytes, a strong trend toward more embryos and significantly more cycles of cryopreservation. Both groups achieved a 47% rate of first positive pregnancy test (hcg), with controls demonstrating a 39% and DOR patients a 32% ongoing PR. Conclusion(s): Women with DOR, if treated with an ovarian age-based rather than chronological age-based ovarian stimulation protocols, will demonstrate surprisingly good PR with IVF in comparison to women with normal ovarian function. (Fertil Steril 2006;86: by American Society for Reproductive Medicine.) Key Words: Ovarian resistance, premature ovarian failure, ovarian stimulation, premature ovarian aging, IVF The treatment of subfertility in women with diminished ovarian reserve (DOR) has remained one of the largest challenges in infertility care. Although DOR is expected in older women, the challenge is especially pronounced when it occurs in young women because they, in addition, appear at risk for early menopause (1, 2) In such women, a number of investigators have recently advocated a hypothesis of early ovarian aging (3), which is based on the observation that the time interval between accelerated decline in fertility (aged years), and menopause appears fixed (4). The normal, physiological aging curve appears, thus, shifted toward younger age, with younger women experiencing the same measured decline in ovarian reserve that, in a large majority of women, is only experienced at older age (5). In young women with DOR, ovarian physiology, under this concept is, therefore, suggested to have aged prematurely. Received January 24, 2006; revised and accepted April 28, Dr. N. Gleicher is a Visiting Professor, Department of Obstetrics of Gynecology, Yale University School of Medicine, New Haven, Connecticut. Reprint requests: Norbert Gleicher, M.D., Center for Human Reproduction, New York, 21 East 69th Street, New York, NY (FAX: ; ngleicher@thechr.com). As is known from the age-appropriate, physiologically aging ovary, DOR leads to a number of rather predictable clinical results. For example, DOR is characterized by ovarian resistance to stimulation with ovulation-inducing medications (6). Although an increase in medication dosage is generally not considered to improve stimulation results in such patients (7), in milder cases an increase in medication dosage or change in stimulation protocol may improve oocyte yield after all (8). This is also the reason why many fertility experts increase the dosages of gonadotropins with advancing female age. Ovarian stimulation protocols are usually chosen based on the woman s age. There is, however, no consensus on how stimulation characteristics should vary. At our center, older women with DOR receive radically different stimulation protocols from younger women with normal ovarian reserve. We reported previously that, in our hands, a microdose agonist protocol, with maximal gonadotropin dosage (of IU) achieved superior IVF outcomes to other stimulation protocols in women with DOR (8). Therefore we have been treating the older ovary different from the younger ovary, as women with normal ovarian function practically /06/$32.00 Fertility and Sterility Vol. 86, No. 6, December 2006 doi: /j.fertnstert Copyright 2006 American Society for Reproductive Medicine, Published by Elsevier Inc. 1621

2 routinely receive a long agonist protocol with a maximum of 300 IU of gonadotropins. Because physiologically aging ovaries, in our hands, appear to benefit from a stimulation change to an agonist microdose protocol, it appeared tempting to speculate that ovaries in younger women with DOR may also show improved responses with such a protocol. Therefore, women with proven prematurely aging ovaries and DOR might benefit from an ovarian stimulation protocol, which was no longer based on the chronological age of the patient but, instead, was based on her ovarian age. This study addresses this question in the format of a case control study. MATERIALS AND METHODS We retrospectively selected by chart review 62 consecutive IVF cycles, conducted at two medical practices, in New York City and Chicago, during a 12-month period, between 2002 and 2003, in women aged 35 years, with a proven diagnosis of DOR. All of these cycles received ovarian stimulation with a microdose leuprolide acetate (Lupron, Tap Pharmaceutical Products Inc., Lake Forest, IL) protocol, involving ovarian stimulation with IU of gonadotropins. Microdose agonist treatment involved the SC injection of 50 g of agonist, twice daily, starting on day 2 of cycle and through ovulation induction with hcg. Gonadotropins were started 2 days after initiation of the agonist. In confirmation of ovarian resistance in this study group of patients, none developed ovarian hyperstimulation, despite obviously very high gonadotropin dosages, considering patient ages. Among these study patients, 38 had been diagnosed with DOR based on abnormally elevated baseline FSH or E 2 levels and 24 based on observed severe ovarian resistance in a prior stimulation cycle. The FSH levels were considered elevated if 9.9 IU/L and E 2 levels if 74.9 pg/ml (5). A diagnosis of DOR was reached, based on a prior cycle performance if oocyte or embryo yields were considered inadequate despite agespecific ovarian stimulation with at least 225 IU of gonadotropins. Such subpar oocyte/embryo yield was defined, based on the young age of the patients, at less than 7 oocytes or less than 4 embryos, respectively. Patients in this study group were then matched with the next consecutive patients of identical age who was presumed to have normal ovarian function, based on normal baseline hormone levels and no prior history of ovarian resistance. The absence of a history of ovarian resistance in control patients, however, did not preclude the first-time diagnosis of previously undiagnosed DOR in a first IVF cycle (as this study will demonstrate). Because the study design specifically precluded the exclusion of selected control patients, the control group contained patients with statistically identical infertility causes as the study group. This meant that women with, for example, advanced stages of endometriosis were included. There were no cases of patients who had undergone chemotherapy in either group. Study outcomes may, therefore, be mildly biased against the study group, as, quite expectedly, controls ended up containing a small number of previously unrecognized DOR patients. Study and control groups were also compared for associated medical conditions. Because, in view of a DOR diagnosis, associated immune diagnoses appeared of special relevance, there was special emphasis placed on evaluating patient and family histories for endometriosis, autoimmune diseases, allergies, and repeated Candida infections. Both groups demonstrated no significant differences. Control patients, with presumed normal ovarian function, received an ovarian stimulation protocol that was based on the patient s chronological age. It involved down-regulation with a long, luteal phase agonist (leupleuride acetate, Lupron, Tap Pharmaceutical Products Inc.), followed by, at most, 300 IU of gonadotropins daily. (Two patients received an antagonist in place of the agonist. In all cycle parameters they were in approximately mid-range for the group, and therefore were not excluded from analysis.) These stimulation protocols have been described in detail and routinely included a combination of recombinant FSH and hmg (9, 10). Study and control groups were then compared for IVF cycle outcome parameters, such as peak E 2 levels, number of oocytes retrieved, number of high quality embryos, and pregnancy rates (PR). All data were extracted from patient chart by either embryology or nursing staffs and transcribed into a questionnaire. The investigators then entered the data onto a spreadsheet that was used for statistical evaluations, using a standard statistical computer program package (SPSS for Windows, standard version , SPSS Co., Chicago, IL). Statistical tests used for this study included the two-tailed t-test, 2, and Fisher s exact test, multiple logistic regression analysis and univariant comparison between two groups (ANOVA), depending on statistical indication. This study did not require review by an Institutional Review Board as it involved the retroactive, and anonymous, review of medical records to which all patients had consented in writing. RESULTS Based on the age-based matching process, age of study and control populations were identical (Table 1). The table also demonstrates that the study group showed a strong trend toward lower mean gravidity (P.08). As noted earlier, the underlying diagnoses, leading patients into infertility treatment, did not vary between study and control groups. The IVF cycle outcomes are summarized in Table 2. Peak E 2 levels were similar between both groups, although the 1622 Gleicher and Barad Ovarian age-based stimulation Vol. 86, No. 6, December 2006

3 TABLE 1 Demographics of study (diminished ovarian reserve, DOR) and control patients. DOR group Controls P value Age Mean SD NS Range Gravidity Mean SD Range Parity Mean SD NS Range Note: SD standard deviation of the mean; NS not statistically significant. Gleicher. Ovarian age-based stimulation. Fertil Steril 2006 total number as well as the number of mature oocytes were significantly larger in control patients (P.01). High quality embryos trended toward higher numbers in controls but missed statistical significance (P.06). Controls, however, cryopreserved embryos at a statistically higher rate than study patients in 47% versus 24% of cycles (P.05). Both groups achieved positive first pregnancy tests in 47% of IVF cycles. Controls demonstrated a trend toward higher clinical PR (39% vs. 32%) but this difference did not reach statistical significance. These PR were achieved in both patient groups with the transfer of mainly only two embryos unless, in rare instances, poor embryo quality mandated otherwise (Table 2). Assuming a 25% clinical PR in frozen thawed IVF cycles (this assumes PR is below historic PR achieved in our program; N. Gleicher, unpublished data), a hypothetical cumulative PR calculated was significantly higher in controls than in study patients (P.05). Univariate analyses of PR, based on E 2 quartiles, three groups of oocyte numbers, two age groups, three groups of high quality embryo numbers, and two groups, each, of embryos transferred fresh and cryopreserved revealed that in study patients older age (range years) was significantly correlated to lower PR (P.05) in comparison to younger female age (range years). No other statistically significant correlations were noted, although trends toward lower pregnancy in study patients were also observed with very high E 2 levels (P.07) and to a lesser degree with large oocyte numbers and small number of embryos. Multiple logistical regression analysis revealed, however, a surprising finding in that the chance of study patients to conceive was actually statistically higher if the data were adjusted for female age and number of oocytes retrieved (P.03; 95% confidence interval [CI] 1, ). DISCUSSION This study attempted to address whether a switch from chronological age-specific to ovarian age-specific ovarian stimulation improves the clinical outcome of IVF in young women with DOR. Such patients have been widely reported, under standard stimulations, to show disappointing PR with IVF (11, 12). Because initial pregnancy outcomes between study patients with DOR and women with apparently overwhelmingly normal ovarian function were practically identically high, it appears reasonable to conclude that pregnancy outcomes were improved in women with DOR by placing them on an ovarian age-based stimulation protocol. The premise of this study was that ovarian age-based, rather than chronological age-based, ovarian stimulation would improve pregnancy outcomes in young women. Such stimulation, in women with normal, age-appropriate ovarian function undoubtedly would have lead to a high prevalence of ovarian hyperstimulation. This study, where not a single case of clinically significant hyperstimulation occurred, confirms that this patient population, in fact, suffered from DOR. Control patients, with presumed normal ovarian function, in contrast, received chronological age-based ovarian stimulation, never exceeding a dosage of 300 IU of gonadotropins daily. Such a dosage represents adequate stimulation for the young and normally functioning ovary but, as this study also showed, is inadequate for young women with unrecognized DOR. A few women with DOR were inadvertently included in the control group, witnessed by a small number of women who demonstrated only relatively low peak E 2 levels (the lowest level was 270 pg/ml) and very small oocyte numbers (Table 2). Because some patients in the study group responded to stimulation with very high peak E 2 levels and large oocyte numbers, this does not contradict their diagnosis of DOR, as milder cases of DOR can still be expected to respond well to an increase in medication dosages (7). Although study patients appear quite rigidly selected, controls appear contaminated by a small number of women with DOR, potentially biasing the results of this study unfavorably for the DOR study group. To address such concerns, we broke up both patient groups into group quartiles and investigated, by univariant analysis, whether the elimination of outliers would affect conclusions of the study. Because no such effects were apparent, any potential biases in the control group should not affect the conclusions of this study. The validity of the study group is further supported by the observation that women with DOR demonstrated a strong trend toward lower fecundity, even at younger ages, when they did not yet perceive themselves as infertile. These findings appear to support the concept of prematurely aging ovaries, as recently proposed (3). They also Fertility and Sterility 1623

4 TABLE 2 IVF cycle characteristics. DOR group (n 62) Controls (n 62) P value E 2 (pg/ml) a Mean SD 2,042 1,331 3, NS Range 20 5, Oocytes total Mean SD Range Mature Mean SD Range Embryos b Mean SD Range Cycles with cryopreservation (%) 15 (24) 29 (47) c.05 Embryo transfer (n) Mean SD NS Range Pregnancies (n%) Positive hcg 29 (47) 29 (47) NS Chemical 5 (8) 2 (3) NS Ongoing 20 (32) 24 (39) NS Spontaneous abortion 2 (3) 3 (5) NS Ectopic 2 (3) 0 (0) NS a Levels were available in only 57/62 DOR and 60/62 control patients. b Only good quality embryos were considered (i.e., embryos considered of good enough quality for embryo transfer or cryopreservation). c Five couples who had embryos for cryopreservation chose not to do so. They are included in these statistics. Gleicher. Ovarian age-based stimulation. Fertil Steril 2006 suggested that correct, quick treatment of such young women is of considerable urgency as, even among young women, age matters: the younger they are at time of treatment, the better their pregnancy chances. Timely treatment with ovarian age-based protocols will, however, result in surprisingly excellent pregnancy outcomes. The study confirms a number of recent reports in the literature that have suggested that young women with elevated FSH levels (if taken as evidence for DOR) still have reasonably good pregnancy chances with IVF. Elevated baseline FSH levels at younger ages do not appear to carry the same negative predictive value as in older patients (13). Ovarian age-based treatment means a number of things. Equally aggressive infertility treatment should be given as for older women. We consider this to require the aggressive use of IVF, as in women with DOR it is impossible to predict how quickly ovarian function will deteriorate. Clinical advice, based on a worst case scenario, has to assume a rapid decline in ovarian function. Ovarian age-specific treatment also means ovarian stimulation based on the presumed ovarian age and not the patient s chronological age. We have for many years stimulated the older ovary exclusively with a microdose agonist protocol because alternative stimulations, inclusive of antagonist protocols, have been in our hands less successful (8). Such an approach has also been suggested by other investigators (14). This study confirms that such an ovarian stimulation protocol results in surprisingly good clinical PR in women with inherently poor prognosis. These women were young but their prematurely aging ovaries were old. The ovarian stimulation used in these women did not only improve oocyte yield, it appeared to produce excellent embryo quality, as witnessed by the superior PR after statistical correction for age and oocyte numbers. This finding correlates with recent reports that elevated baseline FSH levels in younger women reflect more a quantitative than a qualitative decline in ovarian function (13). Oocyte yields can, however, not be fully normalized, as this study also, once more demonstrated, because controls produced significantly more oocytes, despite much milder ovarian stimulation Gleicher and Barad Ovarian age-based stimulation Vol. 86, No. 6, December 2006

5 In summary, DOR, if diagnosed early and if treated aggressively with an ovarian age-based IVF protocol, will yield surprisingly good PR. The proper treatment of DOR will assume increasing importance for infertility practices, as improving IVF outcomes in women with normally functioning ovaries lead to their quick departure from infertility treatment programs, and DOR patients can be expected to accumulate disproportionally. A new approach to such women may, therefore, significantly affect program statistics to the better. Acknowledgments: We acknowledge the help of our Chicago embryology and New York clinical coordinator staff in extracting chart information. REFERENCES 1. de Boer EJ, den Tonkelaar I, te Velde ER, Burger CW, Klip H, van Leeuwen FE. A low number of retrieved oocytes at in vitro fertilization treatment is predictive of early menopause. Fertil Steril 2002;77: de Boer EJ, den Tonkelaar I, te Velde ER, Burger CW, Klip H, van Leeuwen FE. Increased risk of early menopause transition and natural menopause after poor response at first IVF treatment. Hum Reprod 2003;18: Nikolaou D, Templeton A. Early ovarian aging: a hypothesis. Detection and clinical relevance. Hum Reprod 2003;18: Faddy MJ, Gosden RG, Gougeon A, Richardson SJ, Nelson JF. Accelerated disappearance of ovarian follicules in mid-life; implications for forecasting menopause. Hum Reprod 1992;7: Gleicher N. Ovarian aging: is there a norm? Contemp Ob Gyn 2005;50: Beckers NG, Macklon NS, Eijkemans MJ, Fauser BC. Women with regular menstrual cycles and poor response to ovarian hyperstimulation for in vitro fertilization exhibit follicular phase characteristics suggestive of ovarian aging. Fertil Steril 2002;78: Van Hooff MH, Alberda AT, Huisman GJ, Zeilmaker GH, Leerentveld RA. Doubling the human menopausal gonadotropin dose in the course of in vitro fertilization treatment cycles in low responders: a randomized study. Hum Reprod 1993;8: Karande V, Gleicher N. A rational approach to management of low responders in in vitro fertilization. Hum Reprod 1999;14: Karande V, Rao R, Pratt D, Balin M, Levrant S, Morris R, et al. A randomized prospective comparison between intrauterine inseminations and fallopian tube perfusion for the treatment of infertility. Fertil Steril 1995;64: The European and Middle East Orgalutron Study Group. Comparable clinical outcome using GnRH antagonist ganirelix or a long protocol of the GnRH agonists triptorelin for the prevention of premature LH surges in women undergoing ovarian stimulation. Hum Reprod 2001; 16: Bancsi LF, Broermans FJ, Mol BW, Habbema JD, Tevelde ER. Performance of basal follicle-stimulating hormone in the prediction of poor ovarian response and failure to become pregnant after in vitro fertilization: a meta analysis. Fertil Steril 2003;79: Sallam HN, Ezzeldin F, Agameya AF, Rahman AF, El-Garem Y. Defining poor responders in assisted reproduction. Int J Fertil Womens Med 2005;50: Abdalla H, Trum MY. An elevated basal FSH reflects a quantitative rather than qualitative decline of the ovarian reserve. Hum Reprod 2004;19: Fasouliotis SJ, Simon A, Laufer N. Evaluation and treatment of low responders in assisted reproductive technology. A challenge to meet. J Assist Reprod Genet 2000;17: Fertility and Sterility 1625