Luteal phase support

Transcription

1 FERTILITY AND STERILITY VOL. 77, NO. 2, FEBRUARY 2002 Copyright 2002 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A. Luteal phase support Alan S. Penzias, M.D. Boston IVF, Waltham, Massachusetts; Division of Reproductive Endocrinology and Infertility, Beth Israel Deaconess Medical Center, Boston, Massachusetts; and Harvard Medical School, Boston, Massachusetts Objective: To develop a consensus regarding the need for luteal phase support during assisted reproductive technology (ART), and to establish the optimal compound and route of administration for this purpose. Design: Review of the published literature on luteal phase support. Patient(s): Women undergoing assisted reproductive technologies. Intervention(s): Progesterone was administered orally, vaginally, or by intramuscular (i.m.) injection to supplement the luteal phase after assisted reproductive technology (ART). Main Outcome Measure(s): Pregnancy following ART. Result(s): Gonadotropin releasing hormone (GnRH)-agonist protocols necessitate the use of luteal phase support. Progesterone and human chorionic gonadotrophin (hcg) have both been used for this purpose, with comparable outcomes. Progesterone is the product of choice, however, as it is associated with a lower incidence of ovarian hyperstimulation syndrome (OHSS). Its use is indicated up to the serum pregnancy test. Oral, i.m., and vaginal progesterone preparations are available. Intramuscular and vaginal preparations lead to comparable rates of implantation and clinical pregnancy, despite higher serum progesterone levels after i.m. injection. Oral formulations are inferior products for luteal support. Although widely used, i.m. progesterone is uncomfortable and inconvenient for patients. By contrast, the vaginal progesterone gel (Crinone 8%) is more convenient and easier to use. Conclusion(s): Progesterone support of the luteal phase in in vitro fertilization (IVF) cycles is indicated, though support beyond the serum pregnancy test may not be needed. The pregnancy rates after vaginal and i.m. progesterone support are comparable, despite higher serum levels after i.m. injection. Patients prefer the vaginal progesterone gel. (Fertil Steril 2002;77: by American Society for Reproductive Medicine.) Key Words: Assisted reproduction technologies, human chorionic gonadotropin, intramuscular, in vitro fertilization, luteal phase support, outcomes, progesterone, vaginal gel Received May 5, 2001; revised and accepted July 31, Presented at Optimizing Ovulation Induction in ART: A Consensus Meeting, Santa Monica, California, August 18 20, Reprint requests: Alan S. Penzias, M.D., Boston IVF, 40 Second Avenue, Suite 300, Waltham, Massachusetts (FAX: ; E- mail: alan.penzias@ bostonivf.com) /02/$22.00 PII S (01) Following ovulation, the luteal phase of a natural cycle is characterized by the formation of a corpus luteum, which secretes steroid hormones, including progesterone. Following conception and implantation, the developing blastocyst secretes human chorionic gonadotrophin (hcg). The role of hcg is to maintain the corpus luteum and its secretions. Normal luteal function is essential for maintaining pregnancy several studies have shown that removal of the corpus luteum during early pregnancy results in complete abortion (1, 2). These data suggest that progesterone is necessary for the survival of the early pregnancy. THE RATIONALE FOR LUTEAL SUPPORT DURING ART In the mid-1980s, the incorporation of gonadotropin releasing hormone (GnRH) agonists into ovarian stimulation regimens became associated with improved outcomes after in vitro fertilization (IVF) and other assisted reproduction technologies (ART). Pituitary function does not resume completely until 2 3 weeks after the end of GnRH-agonist therapy; however, luteal phase support was considered essential to counter any luteal insufficiency that may have a negative impact on an early pregnancy (3, 4). GnRH-Antagonists have recently appeared with increasing frequency in ART treatment. In contrast to the agonist, antagonists are administered for a significantly shorter period of time and have a shorter duration of effect. Their impact on the corpus luteum is unknown and merits further study. Initially, using hcg or progesterone provided luteal phase support. Over the years, it became clear that luteal phase support had a positive effect on outcome a 1994 meta-anal- 318

2 ysis of randomized trials indicated that the use of hcg or progesterone led to significantly higher pregnancy rates than placebo (5). Progesterone has become the agent of choice for luteal supplementation, because hcg is associated with a higher risk of the ovarian hyperstimulation syndrome (OHSS) (6 9). Most treatment protocols advocate the use of progesterone throughout the first trimester of pregnancy, based on the findings of Shamma et al., who used 17- hydroxyprogesterone as a marker to demonstrate ongoing corpus luteum activity up to week 10 of pregnancy (10). This finding confirmed earlier estimates of 8 10 weeks (11, 12). Progesterone is given with the aim of assisting a corpus luteum that may have been compromised during ovulation induction or oocyte retrieval. REEVALUATING THE NEED FOR LUTEAL PHASE SUPPORT Interestingly, two recent studies suggest that progesterone supplementation may not, in fact, be necessary in all women following the detection of pregnancy. Stovall et al. (13) studied 188 women who conceived with IVF or zygote intrafallopian transfer (ZIFT) between January 1994 and June Progesterone supplementation was discontinued at 4 weeks gestation in patients whose initial progesterone levels were 60 ng/ml. In these women, progesterone levels were maintained at 30 ng/ml without support. The delivery rate in women whose progesterone supplementation was discontinued was comparable with that in women who continued to receive progesterone. In a later study, Stelling et al. (14) evaluated the outcome of 1,322 positive pregnancy tests in women who conceived with IVF or gamete intrafallopian transfer (GIFT) between 1996 and In 1,056 women, progesterone was continued following the positive pregnancy test but was discontinued in 266 others. There was no difference in the rates of biochemical pregnancy, clinical/ongoing pregnancy, spontaneous abortion, or ectopic pregnancy between the two groups. The ability to stop progesterone in some patients following detection of pregnancy after ART raises the question of whether luteal phase support is needed at all between oocyte retrieval and the onset of menses or detection of pregnancy. Hutchinson-Williams et al. (15) studied the interval of time between oocyte retrieval and menses or detection of pregnancy. Twenty-eight women (group I) underwent 28 IVF cycles without subsequent luteal phase support. Estradiol and progesterone levels in that cohort of women were compared with 40 women (group II) who underwent 42 IVF cycles with hcg supplementation in the luteal phase. The groups were matched for age and cause of infertility. Estradiol and progesterone were measured on the day of embryo transfer and every 3 to 4 days thereafter. Luteal phase hcg support significantly augmented estradiol and progesterone levels in the conception cycles of group II compared with group I and progesterone, but not estradiol levels in the nonconception cycles of group II compared with group I. The midluteal decline in estradiol and progesterone, which was observed in group I, was minimized or prevented in group II. The authors concluded that luteal phase hormone profiles of IVF cycles were improved by supplementation with hcg. They further hypothesized that this type of intervention may serve to rescue potentially failing corpora lutea and thereby optimize the periimplantation hormonal milieu. If progesterone supplementation were not needed between oocyte retrieval and pregnancy detection, then a direct comparison of two different methods of progesterone administration would exhibit no clinical differences; however, a prospective randomized trial did reveal a significant difference in implantation rates between women who received intramuscular (i.m.) progesterone and those who received an oral micronized progesterone preparation (16). Although the two formulations of progesterone produced similar progesterone levels 7 days after oocyte retrieval, the implantation rate per embryo among women treated with i.m. progesterone was more than twice that of those treated with the oral preparation (40.9% vs. 18.1%: P.004). Although the clinical pregnancy rate was also higher among women who received i.m. progesterone (57.9% vs. 45.8%), the investigators had to terminate the study prematurely for ethical reasons because of the marked difference in implantation rates. This meant that the sample size was not large enough to determine a statistically significant difference between clinical pregnancy rates. In another randomized study, investigators compared a vaginal progesterone gel (Crinone 8%, Wyeth Laboratories, St. Davids, PA), 90 mg once daily, with an oral progesterone preparation (Utrogestan), 400 mg once daily, and i.m. progesterone in oil, 50 mg once daily. The clinical and ongoing pregnancy rates were comparable between the vaginal gel and i.m. groups, but significantly (P.05) lower with the oral formulation (17). These studies strongly suggest the inferiority of oral progesterone for luteal phase support during ART cycles. PHARMACOKINETICS OF PROGESTERONE SUPPLEMENTATION Various formulations of progesterone are now available, including oral, vaginal, and i.m. Parenteral administration of progesterone, vaginally or intramuscularly, does not subject the compound to the significant metabolic consequences of oral administration. Progesterone administered orally is subjected to first-pass prehepatic and hepatic metabolism. This metabolic activity results in progesterone degradation to its 5 - and 5 -reduced metabolites. Levine and Watson (18) compared the pharmacokinetics of an oral, micronized progesterone preparation (Prome- FERTILITY & STERILITY 319

3 TABLE 1 Pharmacokinetic parameters of progesterone after administration of Crinone 8% vaginal gel, 90 mg, and oral progesterone capsules, 100 mg. Results are presented as means SD. Reproduced with permission (18). Crinone 8%, 90 mg (N 6) Progesterone capsules, 100 mg (N 6) C Max (ng/ml) a Dose-normalized C Max (ng/ml/mg) a C Avg(0 24) (ng/ml) a T Max (h) a AUC 0 24 (ng.h/ml) a Dose-normalized (ng.h/ml/mg) a SD standard deviation; C Max maximum concentration; C Avg(0 24) Mean drug concentration over 24 hours after administration; T Max time to maximum concentration; AUC 0 24 area under concentration-time curve to 24 hours after administration. a P.05, Student s t-test. Penzias. Luteal phase support. Fertil Steril trium, 100 mg, Solvay Pharmaceuticals Inc., Marietta, GA) with that of a vaginal progesterone gel (Crinone 8%, 90 mg). Results showed that the vaginal gel was associated with a higher maximum serum concentration of progesterone (Table 1). Furthermore, the 24-hour area under the curve for drug concentration vs. time (AUC 0 24 ) was higher in the group that had received the vaginal preparation. This signifies greater total progesterone exposure over 24 hours for a single dose of progesterone administered in a vaginal gel compared with a similar dose administered orally. Levine and Watson concluded that the vaginal administration of progesterone results in a greater bioavailability with less relative variability than oral progesterone. Oral administration of progesterone yields poorly sustained plasma progesterone concentrations (19, 20), yet plasma progesterone concentration is a poor marker of bioavailability. Cicinelli et al. (21) recently compared i.m. progesterone administration with vaginal progesterone gel administration. They found that serum progesterone levels were higher after i.m. administration than after vaginal gel administration, although endometrial progesterone levels were significantly higher after vaginal progesterone gel administration than after i.m. injection. The ratio of endometrial to serum progesterone concentrations was markedly higher in women who had received vaginal progesterone gel (14.1 median, range; 95% confidence interval 9.89, 38.79) compared with i.m. injections (1.2 median, range; 95% confidence interval 0.48, 7.39) (P.005). CLINICAL PERFORMANCE OF VAGINAL PROGESTERONE Donor Egg Cycles The efficacy of luteal support with vaginal progesterone has been investigated in donor egg cycles in women without inherent ovarian function. This is an excellent population in which to study luteal supplementation, because these women require complete hormone replacement, in contrast with women undergoing IVF in which luteal support is used to supplement endogenous progesterone production. A recent study compared the effects of vaginal progesterone gel (Crinone 8%), 90 mg, once daily, with i.m. injections of progesterone in oil, 100 mg, once daily, in 86 women (22). The authors reported a higher mean serum progesterone level in women receiving i.m. progesterone compared with the vaginal route; however, there was no difference between groups in the percentage of patients with an in-phase endometrial biopsy, the mean endometrial thickness, the clinical pregnancy rate, the miscarriage rate, or the ongoing pregnancy rate (Table 2). IVF Cycles Various studies have evaluated the efficacy of vaginal progesterone gel compared with other progesterone formulations in women undergoing ovarian stimulation and IVF. An open prospective study, performed by Chantilis et al. (23) compared data from 100 patients who used vaginal progesterone gel (Crinone, 8%), 90 mg once daily, with data from a historic group of 106 patients who had received i.m. progesterone in oil, 50 mg once daily. The results showed that the vaginal progesterone group had lower serum progesterone concentrations and higher rates of biochemical pregnancy loss than the i.m. group; however, the rates for total positive pregnancy tests and ongoing pregnancy were comparable between groups. In a second study, 44 patients using vaginal progesterone gel (Crinone 8%), 90 mg once daily, were compared with 203 historic controls that had used i.m. progesterone, 50 mg once daily (24). The authors of this study found a higher biochemical pregnancy rate in the vaginal progesterone group though lower implantation and clinical pregnancy rates. The patient populations studied did have one important and statistically significant difference between them, which may have limited the investigators ability to draw conclu- 320 Penzias Luteal phase support Vol. 77, No. 2, February 2002

4 TABLE 2 Clinical outcomes in donor egg cycles with a vaginal progesterone gel (Crinone 8%), 90 mg once daily, or intramuscular progesterone, 100 mg once daily. Values shown are means SD (where relevant). Reproduced with permission (22). Variable Crinone 8% (N 42) Intramuscular progesterone (N 44) Percentage of endometrial biopsies showing in-phase development Day 26 serum progesterone level (ng/ml) Day 26 endometrial thickness (mm) Percentage of embryos that implanted (no (34/159) a 19.0 (30/158) a gestational sacs/no. of embryos transferred) Percentage of women with a clinical 45.7 (21/46) a 52.3 (23/44) a pregnancy (no. of women with gestational sacs/no. of women undergoing ET) Percentage of women who spontaneously 14.3 (3/21) a 21.7 (5/23) a miscarried (no. of women miscarrying/ no. of clinical pregnancies) Percentage of women with ongoing pregnancies (no. of women with pregnancies past 20 weeks/no. of women who underwent ET) 39.1 (18/46) a 40.9 (18/44) a SD standard deviation; ET embryo transfer. a Rates not statistically different between Crinone 8% and IM progesterone. Penzias. Luteal phase support. Fertil Steril sions from the study. In this study, the women treated with vaginal progesterone required 29% more ampoules of gonadotropin per cycle (42.6 vs. 32.9, P.01) than patients treated with i.m. progesterone. This suggests that patients who used vaginal progesterone may have been inherently less fertile than those who used i.m. progesterone. Another analysis by Alper and Penzias involved women who were given one of three different progesterone formulations (25). The authors reviewed 1,675 IVF cycles performed between January 1 and December 31, 1998, at Boston IVF. The analysis excluded patients who underwent oocyte donation or freezing of all embryos. Following oocyte retrieval, patients received progesterone for luteal support in one of three forms: vaginal suppository, vaginal gel, or i.m. injection. The choice of formulation was based upon doctor and patient preference. Vaginal progesterone gel was used with increasing frequency during the study period from 21.7% of cases during the first quarter rising to 50.1% in the fourth quarter. Most of the difference was attributable to a shift away from progesterone suppositories. No statistically significant differences in implantation or clinical pregnancy rate were found as a function of progesterone formulation (Figure 1). Williams et al. (26) compared luteal support with a vaginal progesterone gel (Crinone 8%), 90 mg twice daily, vaginal tablets (Prometrium), 200 mg three times daily, or i.m. progesterone, 50 mg once daily. The choice of treatment was based on patient preference. Although serum progesterone concentrations were significantly higher during administration of the i.m. preparation, pregnancy rates were not statistically significantly different as a function of formulation: vaginal gel (39%), vaginal tablets (30%), and i.m. injections (23%) (P.05). A cohort study compared all good prognosis in vitro fertilization/embryo transfer (IVF/ET) patients in a single IVF program in two consecutive years. In 1999, all good prognosis patients (under age 40 with at least 7 embryos on day 1 and remaining cryopreserved embryos after embryo transfer) received vaginal progesterone gel (Crinone 8%), 90 mg once daily. Their outcomes were compared with all good prognosis patients who, in 1998, had received i.m. progesterone, 50 mg once daily (27). The investigators found a trend toward higher implantation and clinical pregnancy rates in the vaginal progesterone gel group. Well-controlled, randomized studies still are needed to refine and confirm the findings of observational, retrospective, and nonrandomized cohort studies. PATIENT PREFERENCES Although i.m. progesterone in oil is a widely used preparation in ART protocols, from a practical perspective, it is uncomfortable and inconvenient for patients. It also has the potential to produce injection-site reactions and serious ad- FERTILITY & STERILITY 321

5 FIGURE 1 (A), Implantation rate, assessed over 1,675 IVF cycles, in relation to the type of progesterone preparation used for luteal phase support; vaginal suppositories, intramuscular injection, Crinone o. (B), Clinical pregnancy rate in patients who received either Crinone 8% vaginal progesterone gel or another type of progesterone preparation for luteal phase support; non-crinone, Crinone. Reproduced with permission (25). Penzias. Luteal phase support. Fertil Steril verse effects, including allergic reactions to the oil vehicle. In these respects, vaginal progesterone offers a clear advantage over i.m. administration. Recently, in a multicenter U.S. study involving almost 2,000 women, Levine (28) found that, not only were pregnancy rates comparable between women who had used i.m. progesterone and those who had used Crinone 8% vaginal gel, but patients preferred the gel and found it easier to administer. CONCLUSION Progesterone is required for the maintenance of an intrauterine pregnancy in the first trimester. Historically, luteal adequacy was judged by the measurement of serum progesterone concentration. The assumption that high serum progesterone levels are required to achieve biological efficacy in the endometrium appears to be incorrect. Adequate tissue levels of progesterone are obtained after the vaginal administration of progesterone gel via a previously unrecognized uterine first-pass effect. In fact, when compared to i.m. progesterone injection, vaginal administration results in higher endometrial progesterone levels. Observation of patients using vaginal or i.m. progesterone in the luteal phase suggests that both are effective, whereas orally administered progesterone appears to be inferior. Consensus Points from the Meeting Progesterone support of the luteal phase (up to the serum pregnancy test) in IVF cycles is supported by the literature, though support beyond the pregnancy test may not be indicated. Although serum progesterone levels are higher after i.m. administration than after vaginal administration, the pregnancy rates after these two types of support are comparable. References 1. Csapo AI, Pulkkinen MO, Kaihola HL. The relationship between the timing of luteectomy and the incidence of complete abortions. Am J Obstet Gynecol 1974;118: Csapo AI, Pulkkinen MO, Wiest WG. Effects of luteectomy and progesterone replacement therapy in early pregnant patients. Am J Obstet Gynecol 1973;115: Smitz J, Erard P, Camus M, et al. Pituitary gonadotrophin secretory capacity during the luteal phase in superovulation using GnRH-agonists and HMG in a desensitization or flare-up protocol. Hum Reprod 1992; 7: Smitz J, Bourgain C, Van Waesberghe L, Camus M, Devroey P, Van Steirteghem AC. A prospective randomized study on estradiol valerate supplementation in addition to intravaginal micronized progesterone in buserelin and HMG-induced superovulation. Hum Reprod 1993;8: Soliman S, Daya S, Collins J, Hughes EG. The role of luteal phase support in infertility treatment: a meta-analysis of randomized trials. Fertil Steril 1994;61: Herman A, Ron-El R, Golan A, Raziel A, Soffer Y, Caspi E. Pregnancy rate and ovarian hyperstimulation after luteal human chorionic gonadotropin in in vitro fertilization stimulated with gonadotropin-releasing hormone analog and menotropins. Fertil Steril 1990;53: Mochter MH, Hogerzeil HV, Mol BW. Progesterone alone versus progesterone combined with HCG as luteal support in GnRHa/HMG induced IVF cycles: a randomized trial. Hum Reprod 1996;11: McClure N, Leya J, Radwanska E, Rawlins R, Haning RV Jr. Luteal phase support and severe ovarian hyperstimulation syndrome. Hum Reprod 1992;7: MacDougall MJ, Tan SL, Jacobs HS. In vitro fertilization and the ovarian hyperstimulation syndrome. Hum Reprod 1992;7: Shamma FN, Penzias AS, Thatcher S, DeCherney AH, Lavy G. Corpus luteum function in successful in vitro fertilization cycles. Fertil Steril 1992;57: Yoshimi T, Strott CA, Marshall JR, Lipsett MB. Corpus luteum function in early pregnancy. J Clin Endocrinol Metab 1969;29: Tulchinsky D, Hobel CJ. Plasma human chorionic gonadotropin, estrone, estradiol, estriol, progesterone, and 17 alpha-hydroxyprogester- 322 Penzias Luteal phase support Vol. 77, No. 2, February 2002

6 one in human pregnancy. 3. Early normal pregnancy. Am J Obstet Gynecol 1973;117: Stovall DW, Van Voorhis BJ, Sparks AE, Adams LM, Syrop CH. Selective early elimination of luteal support in assisted reproduction cycles using a gonadotropin-releasing hormone agonist during ovarian stimulation. Fertil Steril 1998;70: Stelling JR, Barrett CB, Penzias AS, Alper MM, Berger MJ, Oskowitz SP, et al. Progesterone support in early IVF/GIFT pregnancies may not be necessary (abstract O-095). Presented at the American Society for Reproductive Medicine and Canadian Fertility and Andrology Society Conjoint Annual Meeting September , Toronto, Canada. Copyright 1999 American Society for Reproductive Medicine, Birmingham, Alabama. 15. Hutchinson-Williams KA, DeCherney AH, Lavy G, Diamond MP, Naftolin F, Lunenfeld B. Luteal rescue in in vitro fertilization-embryo transfer. Fertil Steril 1990;53: Licciardi FL, Kwiatkowski A, Noyes NL, Berkeley AS, Krey LL, Grifo JA. Oral versus intramuscular progesterone for in vitro fertilization: a prospective randomized study. Fertil Steril 1999;71: Saucedo LLE, Galache VP, Hernandez AS, Santos HR, Arenas ML, Patrizio P. Randomized trial of three different forms of progesterone supplementation (abstract P-175). Fertil Steril 2000;74(3S):S Levine H, Watson N. Comparison of the pharmacokinetics of Crinone 8% administered vaginally versus Prometrium administered orally in postmenopausal women. Fertil Steril 2000;73: Whitehead MI, Townsend PT, Gill DK, Collins WP, Campbell S. Absorption and metabolism of oral progesterone. Br Med J 1980;280: Nahoul K, Dehennin L, Jondet M, Roger M. Profiles of plasma estrogens, progesterone and their metabolites after oral or vaginal administration of estradiol or progesterone. Maturitas 1993;16: Cicinelli E, DeZiegler D, Bulletti C, Matteo MG, Schonauer LM, Galantino P. Direct transport of progesterone from vagina to uterus. Obstet Gynecol 2000;95: Jobanputra K, Toner JP, Denoncourt R, Gibbons WE. Crinone 8% (90 mg) given once daily for progesterone replacement therapy in donor egg cycles. Fertil Steril 1999;72: Chantilis SJ, Zeitoun KM, Patel SI, Johns DA, Madziar VA, McIntire DD. Use of Crinone vaginal progesterone gel for luteal support in in vitro fertilization cycles. Fertil Steril 1999;72: Damario MA, Goudas VT, Session DR, Hammitt DG, Dumesic DA. Crinone 8% vaginal progesterone gel results in lower embryonic implantation efficiency after in vitro fertilization-embryo transfer. Fertil Steril 1999;72: Alper MM, Penzias AS. Crinone offers excellent implantation rates in patients undergoing in vitro fertilization (abstract P-059). Presented at the 16th Annual Meeting of the European Society of Human Reproduction and Embryology, Bologna, Italy, June Williams SC, Donahue J, Muasher SJ. Vaginal progesterone therapy during programmed cycles for frozen embryo transfer: an analysis of serum progesterone levels and pregnancy rates (abstract P-363). Fertil Steril 2000;74(3S):S Coutifaris C, Patrizio P, Schafer D, Bunso S, Bucci J, Barnhart K. Is the use of Crinone for the support of the luteal phase detrimental to pregnancy outcome after transfer of non-cryopreserved embryos in good prognosis patients? A preliminary report (abstract P-350). Fertil Steril 2000;74(3S):S Levine HL. Luteal support with Crinone 8% in 1827 women undergoing assisted reproduction technology (ART) procedures (abstract P-182). Fertil Steril 2000;74(3S):S152. FERTILITY & STERILITY 323