A luteal estradiol protocol for anticipated poor-responder patients may improve delivery rates

Transcription

1 A luteal estradiol protocol for anticipated poor-responder patients may improve delivery rates Micah J. Hill, D.O., a Grant D. E. McWilliams, D.O., b Kathleen A. Miller, B.S., c Richard T. Scott, Jr, M.D., H.C.L.D., c and John L. Frattarelli, M.D. c a Blanchfield Army Community Hospital, Fort Campbell, Kentucky; b Tripler Army Medical Center, Honolulu, Hawaii; and c Reproductive Medicine Associates of New Jersey, Morristown, New Jersey Objective: To compare IVF data and outcomes between a standard protocol and a luteal phase E 2 protocol. Design: Retrospective cohort analysis. Setting(s): Large academic assisted reproduction technologies center. Patient(s): Fifty-seven infertile patients with a history of poor response to IVF stimulation and 228 matched control patients. Intervention(s): IVF with a standard protocol or a luteal phase E 2 protocol. Main Outcome Measure(s): Live-birth rates. Result(s): Patients in the luteal E 2 protocol required more days of stimulation and total gonadotropins and had higher peak E 2 levels when compared with the control group. The luteal E 2 protocol showed a greater percentage of embryos with R7 cells on day 3. A trend toward improved delivery rates was seen in the luteal E 2 protocol (28.1% vs. 22.4%; relative risk, 1.25, ). Conclusion(s): A luteal E 2 protocol results in improved day 3 embryo development as demonstrated by the percent of embryos at the R7-cell stage. Likewise, the luteal E 2 protocol may ultimately improve pregnancy outcomes for patients with poor response to IVF stimulation. (Fertil Steril Ò 2009;91: Ó2009 by American Society for Reproductive Medicine.) Key Words: Poor responders, IVF outcome, luteal phase, estradiol, embryo morphology, oocytes, pregnancy, microdose flare, GnRH antagonist It is estimated that 5% 18% of all IVF cycles are complicated by poor response to ovarian hyperstimulation. Subsequently, these patients have poor IVF outcomes with successful pregnancy rates as low as 2% 4% (1 5). A number of criteria have been proposed and evaluated that may be used to prognosticate ovarian responsiveness to exogenous gonadotropin stimulation, the quality of the oocytes, and the subsequent implantation and pregnancy rates. These parameters and criteria include few oocytes or follicles, low peak E 2, the requirement of excessive gonadotropins for follicular development, elevated baseline FSH, patient age, and basal antral follicle count (1, 3, 6 16). The challenge the poor responder presents lies in how best to stimulate the ovaries. A novel strategy for treating poor responders is to give E 2 in the luteal phase before IVF hyperstimulation (17 19). Endogenous FSH in the preceding luteal phase may selectively stimulate larger follicles and subsequently lead to a size discrepancy in the developing follicular Received October 2, 2007; revised and accepted December 24, M.J.H. has nothing to disclose. G.D.E.M. has nothing to disclose. K.A.M. has nothing to disclose. R.T.S. has nothing to disclose. J.L.F. has nothing to disclose. The views expressed in this manuscript are those of the authors and do not reflect the official policy or position of the Department of the Army, Department of Defense, or the U.S. Government. Reprint requests: John L. Frattarelli, M.D., Associate Professor, Robert Wood Johnson Medical School, Reproductive Medicine Associates of New Jersey, 100 Franklin Square Drive, Suite 200, Somerset, NJ (FAX: ; jfrattarelli@rmanj.com). cohort. This size discrepancy may cause fewer follicles to be responsive to gonadotropin stimulation. Poor responders to hyperstimulation show an improvement in fertilization rates and greater number of embryos when given luteal phase E 2 (17). Fanchin et al. showed that luteal phase E 2 resulted in a greater number of follicles R16 mm, more mature oocytes, and more available embryos when compared with a control population (19). In a prior paired-analysis study, a luteal E 2 protocol was associated with a greater number of embryos with R7 cells, oocytes retrieved, mature oocytes, and total embryos when compared with a standard protocol (20). The purpose of this study was to investigate the effects of a luteal E 2 protocol, where exogenous E 2 is administered in the preceding luteal phase, on IVF parameters and pregnancy outcomes in patients proven or suspected to be poor responders. We therefore undertook this retrospective cohort analysis to evaluate pregnancy outcomes in patients at our center who have been treated with a luteal E 2 protocol and compared them with matched patients treated with a standard poor-responder protocol. MATERIALS AND METHODS This retrospective cohort study was designed to evaluate the ability of a fairly new and rarely used protocol (the luteal E 2 protocol) to enhance the stimulation parameters in patients thought to be poor responders to IVF. In a review of our electronic database, we found 57 patients who were treated with /09/$36.00 Fertility and Sterility â Vol. 91, No. 3, March doi: /j.fertnstert Copyright ª2009 American Society for Reproductive Medicine, Published by Elsevier Inc.

2 the luteal E 2 protocol. Patients over the age of 42 were excluded (n ¼ 3), as these patients typically have poorer IVF outcomes. For each subject in the luteal E 2 group, four control patients were chosen. Controls were matched for date of IVF procedure (6 months, two controls before and two after the subject cycle date), age, basal antral follicle count (BAFC; 2), infertility diagnosis, intracytoplasmic sperm injection (ICSI) or conventional insemination, and stimulation protocol (either microflare agonist or GnRH antagonist). The first four controls who underwent IVF and met the matching criteria immediately before and after the case s IVF cycle were chosen. Control criteria were confirmed by an author (MJH) who was blinded to all data except cycle date, patient age, diagnosis, insemination method, embryos transferred, and protocol used. The main outcome measure was live-birth rate. Other outcome measures included the number of follicles R14 mm on day of hcg administration, peak E 2 level (defined as the level of E 2 on the day of hcg administration), ampules of gonadotropins administered, days of stimulation, endometrial response to stimulation, number of oocytes retrieved, number of embryos, chemical and clinical pregnancy rates, and pregnancy loss rates. Patients The records of all patients from January 1, 2004, to August 31, 2006, undergoing a fresh IVF cycle, with or without ICSI, who had a luteal E 2 protocol stimulation at Reproductive Medicine Associates of New Jersey were evaluated. Fifty-seven patients who had a luteal E 2 protocol with either a microdose flare protocol or a GnRH antagonist protocol and a diagnosis of poor responder were identified for analysis. A diagnosis of poor responder was made if patients had a history of poor response in a prior cycle (%5 oocytes retrieved, poor-quality oocytes, poor-quality embryos, or cycle cancellation due to poor response) or were anticipated to be a poor responder based on initial testing (basal FSH over 12 miu/ml or BAFC %5). Two-hundred twenty-eight control patients were chosen who underwent the same protocol (microdose flare or GnRH antagonist) but did not receive the luteal phase E 2 protocol. The patients in the control group were matched (4:1) to the subject group as previously described. All patients were treated with these protocols secondary to being diagnosed as an IVF poor responder based on patient evaluation, prior response, and physician preference. No egg recipient or cryopreserved/thawed embryo cycles were included. Study Design This is a retrospective cohort analysis of 57 cases and 228 controls undergoing 285 fresh IVF cycles at Reproductive Medicine Associates of New Jersey. Transvaginal ultrasound to assess basal antral follicle number was performed on menstrual cycle day 3 (the start of gonadotropins). Institutional Review Board approval was obtained from the Western Institutional Review Board, Olympia, Washington. Patient Treatment Protocols Patients underwent controlled ovarian hyperstimulation generally using a step-down protocol, with a typical initiating dose of IU/day of recombinant FSH or a combination of recombinant FSH and low-dose hcg administered at IU/day. The two methods of controlled ovarian hyperstimulation used either a microdose flare GnRH agonist or intracycle GnRH antagonist administration. If microdose flare GnRH agonist was used, it was initiated on menstrual cycle day 3 using leuprolide acetate (Lupron; TAP Pharmaceuticals, Deerfield, IL) at a daily dose of 0.05 mg given SC and continued until the day of hcg administration. For antagonist cycles, a GnRH antagonist was administered when the lead follicle reached mm in greatest diameter at a dose of 250 mg in 0.5 ml/day until the day of hcg injection. When not using the luteal E 2 protocol, a short course of oral contraceptives was used in the previous cycle with either GnRH agonist or gonadotropins (for the microdose flare or the GnRH antagonist protocol) beginning on menstrual cycle day 3 after discontinuation of the oral contraceptives. For luteal E 2 protocols, oral micronized 17b-estradiol (Estrace; Mead Johnson, Evansville, IN) 2 mg twice a day orally was started on luteal day 21 and continued through the first 3 days of gonadotropin stimulation (gonadotropins were started on menstrual cycle day 3). No oral contraceptive therapy was used in combination with the luteal E 2 protocol. When the largest two or three follicles reached the 18-mm range, a single 10,000 IU IM dose of hcg (Pregnyl, Organon, West Orange, NJ; Novarel, Ferring Pharmaceuticals Inc., Tarrytown, NY) or its recombinant equivalent (Ovidrel 500 mg, Serono Laboratories, Rockland, MD) was administered. Transvaginal follicular aspiration took place hours later. ET Technique ETs were all performed 3 5 days after oocyte retrieval. Patients were instructed to have a full bladder, which would provide an acoustic window for visualization of the uterus, in preparation for the ultrasound-guided ET. Each patient was placed in the dorsal lithotomy position without anesthesia or sedation. Each ET was performed with an Embryon Genesis Catheter System (Rocket Medical PLC, Hingham, MA), while the ultrasonographer performed the abdominal ultrasound using a 5-MHz probe (GE Logiq 400 Pro Series, General Electric Company, Pewaukee, WI). Statistical Analysis For normally distributed data, a t-test was used to compare the mean values between two different stimulation protocols. For data that were not normally distributed, a Mann-Whitney rank sum test was used to compare the mean values between two stimulation protocols. Differences in outcome rates were analyzed using a c 2 or two-tailed Fisher s exact test. An alpha error of 0.05 was considered significant for all comparisons. Relative risk and 95% confidence intervals are displayed 740 Hill et al. Estradiol protocol improves IVF outcomes Vol. 91, No. 3, March 2009

3 where appropriate. All data are reported as means with their associated SDs. RESULTS Fifty-seven patients deemed to be poor responders underwent IVF hyperstimulation with a luteal phase E 2 protocol. Fourteen patients in the subject group underwent stimulation with a microdose flare protocol, and 43 underwent a GnRH antagonist protocol. Two-hundred twenty-eight control patients were matched to the luteal E 2 subjects by date of IVF procedure, age, BAFC, infertility diagnosis, ICSI or conventional insemination, and stimulation protocol (either microflare agonist or GnRH antagonist). Fifty-six patients in the control group were treated with a microdose flare protocol, and 172 were treated with a GnRH antagonist protocol. Table 1 displays the demographics and stimulation variables for the patient population. No difference was noted between the subject group and the control group with regard to patient age, partner age, or patient weight. The patients in the E 2 group were taller (P<.05), but the body mass index was not different between the two groups. There was no difference between the two protocols with respect to BAFC, number of follicles R14 mm on day of hcg surge, or endometrial thickness on day of hcg surge. However, the luteal phase E 2 protocol required more total gonadotropins (P<.001) and more days of stimulation (P<.001) and had higher peak E 2 levels (P<.05). Ooctye and embryo outcome data were compared between the luteal E 2 protocol and the standard protocol (Table 2). There was no difference between the two populations with respect to the number of oocytes retrieved, mature oocytes, and fertilized embryos. There was no difference between the two groups in regards to chemical pregnancy, clinical pregnancy, pregnancy loss, or live-birth rates (Table 3). However, the percent of embryos at the R7-cell stage on day 3 was significantly higher in the luteal phase E 2 group (P<.05; Table 3). A trend toward a higher delivery rate in the luteal E 2 group was noted (28.1% vs. 22.4%), although statistical significance was not achieved (P¼.36). A power analysis was performed, determining a need for 946 subjects to confirm a statistical difference in the live-birth rates between the two protocols. A subanalysis of only the GnRH antagonist protocol showed similar trends in favor of the luteal E 2 protocol (data not shown). DISCUSSION The luteal E 2 protocol may represent a novel and more successful way to treat poor responders during IVF cycles. The theory is that by suppressing FSH in the preceding luteal phase, asynchronous follicular stimulation may be avoided. This ultimately may result in a larger and more coordinated cohort of follicles responding to the stimulation process, leading to improved outcomes (20). The primary outcome of this paper was the live-birth rate. We did not find a statistically significant improvement in livebirth rates with the luteal E 2 protocol. However, there was a trend toward an improved live-birth rate in patients receiving the luteal E 2 protocol, with these patients having a delivery rate that increased by 25% (22.4% vs. 28.1%). A power analysis of these data revealed a sample size of 946 subjects to statistically show improvement in the live-birth rate. While a randomized controlled trial is certainly needed to demonstrate efficacy, it will be difficult to recruit and randomize such a large number of poor-responder patients. Likewise, few centers have the patient volume to recruit such a large number of patients deemed to have poor response either retrospectively or prospectively. We noted an increase in the total gonadotropins used and days of stimulation in the luteal E 2 group. A possible reason TABLE 1 Population demographic and IVF stimulation data for luteal phase E 2 and standard protocol IVF cycles. Variable protocol (n [ 57) protocol (n [ 228) Patient age, years Partner age, years Body mass index BAFC No. of previous IVF cycles Days of stimulation <.001 a Total gonadotropins, ampules <.001 a Peak E 2, pg/ml <.05 a Follicles R14 mm on day of hcg Endometrial thickness on day of hcg, mm a Statistically significant findings (P<.05). P Fertility and Sterility â 741

4 TABLE 2 Oocyte and embryo data for luteal phase E 2 and standard protocol IVF cycles. Outcome variable protocol (n [ 57) protocol (n [ 228) P Retrieved oocytes Mature oocytes PN a R7 Cells on day Embryos transferred a 2PN ¼ number of fertilized oocytes on day 1. for the increased use of gonadotropins is that the inhibitory effect of E 2 on FSH in the luteal phase may result in slower and more coordinated growth of the follicles once stimulation has started. Our findings are consistent with the published literature in many regards. While Fanchin et al. studied normal responding patients, our data are consistent with theirs with respect to more days of gonadotropin stimulation and higher peak E 2 levels (17 19). Prior studies have also shown that variations of the luteal E 2 protocol are associated with a higher number of oocytes retrieved, mature oocytes, and embryos available (19 21). These findings should translate into improved pregnancy outcomes. The increase in gonadotropin dose is thought to be a result of a slower and more coordinated stimulation process secondary to a more homogenous antral follicle cohort (18), as an increased gonadotropin dose itself is not associated with improved outcomes in poor responders. In contrast to Fanchin et al., we did not find an increase in the number of follicles with this protocol (17). However, this may be due to the fact that we used R14 mm as the cutoff value, whereas Fanchin et al. used R16 mm. Both our study and that of Fanchin et al. used oral 17-b-estradiol, while the study by Dragisic et al. used a transdermal E 2 patch and a GnRH antagonist in the preceding luteal phase (21). The obvious weakness of the paper is its retrospective design, a common limitation in studies for poor responders. It is difficult to achieve the appropriate power for such studies in a randomized prospective fashion. A source of potential bias in this design is the question of what factors in a patient s history led the clinician to choose who did or did not receive the luteal E 2 protocol. We matched our control population to the subjects in five critical demographics to attempt to minimize this bias. Additionally, poor response was a clinical definition required for entry into the study. This was based on historic response to hyperstimulation or diagnostic evaluation made by the clinician; however, entry into the study was not defined with strict criteria. The definition of a poor responder varies from investigator to investigator, making comparisons between studies more difficult. Another weakness is the underpowered sample size. We attempted to correct for this by using a 4:1 ratio of controls to subjects. In summary, giving E 2 in the luteal phase before gonadotropin stimulation is associated with an increase in the amount of stimulation required, the peak E 2 levels, and the percent of embryos reaching the R7-cell stage in patients who are poor responders during IVF. The slower and more co-coordinated follicular growth may lead to an improvement in embryo and oocytes quantity, possibly leading to an increase in delivery rates. It is the possibility of improved IVF outcomes with a luteal E 2 protocol for poor responders that is most intriguing. Such a protocol could represent an important method for treating poor-prognosis patients to ultimately optimize outcome success. Further research in this area is warranted to confirm and advance these findings, specifically with studies powered to evaluate for delivery rates. TABLE 3 Clinical outcome rates for luteal E 2 protocol and standard protocol in patients with poor response to IVF cycles. Outcome variable protocol, % (n [ 57) protocol, % (n [ 228) RR (95% CI) P % of embryos R7 cell 46.4 (154/332) 40.6 (548/1349) 1.14 ( ).05 a stage on day 3 b Implantation rate 19.7 (31/157) 21.8 (130/596) 0.91 ( ).57 Chemical pregnancy rate 45.6 (26/57) 44.3 (101/228) 1.03 ( ).86 Clinical pregnancy rate 38.6 (22/57) 36.4 (83/228) 1.06 ( ).76 Total pregnancy loss rate 30.8 (8/26) 32.7 (33/101) 0.94 ( ).85 Delivery rate 28.1 (16/57) 22.4 (51/228) 1.25 ( ).36 a Statistically significant value (P¼.05). b Represents the percentage of fertilized embryos (2PN) that progressed to the R7 cell stage. 742 Hill et al. Estradiol protocol improves IVF outcomes Vol. 91, No. 3, March 2009

5 REFERENCES 1. Mohamed KA, Davies WA, Allsopp J, Lashen H. Agonist flare-up versus antagonist in the management of poor responders undergoing in vitro fertilization treatment. Fertil Steril 2005;83: Schmidt DW, Bremner T, Orris JJ, Maier DB, Benadiva CA, Nulsen JC. A randomized prospective study of microdose leuprolide versus ganirelix in in vitro fertilization cycles for poor responders. Fertil Steril 2005;83: Schoolcraft W, Schlenker T, Gee M, Stevens J, Wagley L. Improved controlled ovarian hyperstimulation in poor responder in vitro fertilization patients with a microdose follicle-stimulating hormone flare, growth hormone protocol. Fertil Steril 1997;67: Ben-Rafael Z, Orvieto R, Feldberg D. The poor-responder patient in an in vitro fertilization-embryo transfer (IVF-ET) program. Gynecol Endocrinol 1994;8: Ulug U, Ben-Shlomo I, Turan E, Erden HF, Akman MA, Bahceci M. Conception rates following assisted reproduction in poor responder patients: a retrospective study in 300 consecutive cycles. Reprod Biomed Online 2003;6: Tarlatzis BC, Zepiridis L, Grimbizis G, Bontis J. Clinical management of low ovarian response to stimulation for IVF: a systematic review. Hum Reprod Update 2003;9: Cooperman AB. Antagonists in poor-responder patients. Fertil Steril 2003;80: Akman MA, Erden HF, Tosun SB, Bayazit N, Aksoy E, Bahceci M. Comparison of agonistic flare-up-protocol and antagonistic multiple dose protocol in ovarian stimulation of poor responders: results of a prospective randomized trial. Hum Reprod 2001; Hellberg D, Waldenstrom U, Nilsson S. Defining a poor responder in in vitro fertilization. Fertil Steril 2004;82: Detti L, Williams DB, Robins JC, Maxwell RA, Thomas MA. A comparison of three downregulation approaches for poor responders undergoing in vitro fertilization. Fertil Steril 2005;84: Weissman A, Farhi J, Royburt M, Nahum H, Glezerman M, Levran D. Prospective evaluation of two stimulation protocols for low responders who were undergoing in vitro fertilization embryo transfer. Fertil Steril 2003;79: Sallam HN, Ezzeldin F, Agameya AF, Rahman AF, El-Garem Y. Defining poor responders in assisted reproduction. Int J Fertil Women Med 2005;50: Frattarelli JL, Levi AJ, Miller BT, Segars JH. A prospective assessment of the predictive value of basal antral follicles in in vitro fertilization cycles. Fertil Steril 2003;80: Scott RT Jr, Hofmann GE. Prognostic assessment of ovarian reserve. Fertil Steril 1995;63: Toner JP, Philput CB, Jones GS, Muasher SJ. Basal follicle-stimulating hormone level is a better predictor of in vitro fertilization performance than age. Fertil Steril 1991;55: Frattarelli JL, Lauria-Costab D, Miller BT, Bergh PA, Scott RT. Basal antral follicle number and mean ovarian diameter predict cycle cancellation and ovarian responsiveness in assisted reproductive technology cycles. Fertil Steril 2000;73: Fanchin R, Cunha-Filho JS, SchonauerLM. Coordination of early antral follicles by luteal estradiol administration provides a basis for alternative controlled ovarian hyperstimulation regimens. Fertil Steril 2003;79: Fanchin R, Salomon L, Castelo-Branco A, Olivennes F, Frydman N, Frydman R. Luteal estradiol pre-treatment coordinates follicular growth during controlled ovarian hyperstimulation with GnRH antagonists. Hum Reprod 2003;18: Fanchin R, Mendez Lozano DH, Schonauer LM, Cunha-Filho JS, Frydman R. Hormonal manipulations in the luteal phase to coordinate subsequent antral follicle growth during ovarian stimulation. Reprod Biomed Online 2005;10: Frattarelli JL, Hill MJ, McWilliams GD, Miller KA, Bergh PA, Scott RT Jr. A luteal estradiol protocol for expected poor-responders improves embryo number and quality. Fertil Steril In press. 21. Dragisic KG, Davis OK, Fasouliotis SJ, Rosenwaks Z. The use of a luteal estradiol patch and a gonadotropin-releasing hormone antagonist suppression protocol before gonadotropin stimulation for in virto fertilization in poor responders. Fertil Steril 2005;84: Fertility and Sterility â 743