Assisted Reproduction Unit, American Hospital of Istanbul, Istanbul, Turkey

Transcription

1 FERTILITY AND STERILITY VOL. 72, NO. 6, DECEMBER 1999 Copyright 1999 American Society for Reproductive Medicine ublished by Elsevier Science Inc. rinted on acid-free paper in U.S.A. ROGESTERONE Elevated serum progesterone level on the day of human chorionic gonadotropin administration does not adversely affect implantation rates after intracytoplasmic sperm injection and embryo transfer Bulent Urman, M.D., Cengiz Alatas, M.D., Senai Aksoy, M.D., Ramazan Mercan, M.D., Aycan Isiklar, MSc., and Basak Balaban, BSc. Assisted Reproduction Unit, American Hospital of Istanbul, Istanbul, Turkey Objective: To evaluate the association between serum levels on the day of hcg administration and the outcome of intracytoplasmic sperm injection (ICSI). Design: Retrospective case study. Setting: Assisted reproduction unit of a tertiary care private hospital. atient(s): Nine hundred eleven ICSI cycles that proceeded to ET were studied. Intervention(s): The decision to administer hcg was based on serum E 2 levels and follicle size. Serum was measured from frozen sera obtained on the day of hcg administration. Cycles were stratified according to serum levels of 0.9 ng/ml (n 298) or 0.9 ng/ml (n 613). This cutoff level was selected because it yielded the highest sensitivity and specificity according to a receiver operator characteristic curve. Main Outcome Measure(s): Implantation and clinical pregnancy rates. Result(s): In cycles with high serum levels, more oocytes were retrieved and more embryos were available for transfer. Clinical pregnancy rates per ET in the low and high groups were 36.9% and 45.4%, respectively (.05). The implantation rate per embryo was similar in the two groups (14.9% and 16.4%, respectively, in cycles with levels 0.9 vs 0.9. Abortion rates were 22.7 and 25.8%, respectively (.05). Conclusion(s): Our data showed no adverse effect of high serum levels on the day of hcg administration on implantation rates after ICSI and ET. (Fertil Steril 1999;72: by American Society for Reproductive Medicine.) Key Words: rogesterone, fertilization in vitro, intracytoplasmic sperm injection, implantation Received March 5, 1999; revised and accepted July 8, Reprint requests: Bulent Urman, M.D., VKV Amerikan Hastanesi, Guzelbahce sok no: 20, Nisantasi 80200, Istanbul, Turkey (FAX: ; burman@superonline.com) /99/$20.00 II S (99) Elevated serum levels before administration of hcg have been reported to adversely affect the outcome of IVF-ET in some studies (1 12). ossible explanations for the lack of concordance among studies include the use of retrospective study designs, the use of different protocols for controlled ovarian hyperstimulation, and the use of different cutoff levels at the time of data analysis. The decision to administer the ovulatory dose of hcg is commonly based on serum E 2 levels and follicle size. However, a high level on the anticipated day of hcg is often considered to be an indication to cancel oocyte retrieval. This practice is based on the assumption that high levels may have a negative impact on endometrial receptivity. In this study, levels were assessed after completion of the treatment cycle and therefore had no impact on the timing of hcg or cyclecancellation policy. This aspect of the study is important because it eliminates the bias introduced when decision making is based on levels; that is, women with high levels may proceed to oocyte recovery less frequently than women with low levels. In this study, we analyzed the outcome of 911 IVF-ET cycles in which intracytoplasmic sperm injection (ICSI) was performed and in 975

2 which hcg was administered without previous knowledge of the serum level. MATERIALS AND METHODS The study group comprised 911 couples undergoing their first treatment cycle for male infertility. Serum levels were assessed from frozen sera previously withdrawn on the day of hcg administration. Approval of the American Hospital Ethics Committee was obtained before initiation of the study. The decision to administer hcg had been based on the serum E 2 level and follicle size and was reached without knowledge of the serum level. rocedure Ovarian stimulation was undertaken using SC buserelin acetate (Suprefact proinjection; Hoechst AG, Frankfurt am Main, Germany) in a luteal long protocol combined with pure FSH (Metrodin 75; I.F. Serono, Rome, Italy). Cycles initiated without GnRH analogues or with other GnRH analogue protocols were excluded. Gonadotropins were initiated once down-regulation was confirmed (serum E 2 50 pg/ml, serum 1 ng/ml, and absence of ovarian cystic structures measuring 10 mm). The initial starting dose of FSH was IU, depending on the patient s previous or anticipated ovarian response. The dose was subsequently individualized in a step-down manner. Ovarian response was monitored with ultrasonographic measurements of follicle size and serum E 2 levels. We administered hcg when the leading follicle reached or exceeded 20 mm in size with a serum E 2 level of pg/ml per mature follicle. Oocyte retrieval was performed 36 hours after the injection of 10,000 U of hcg (rofasi H 5000; I.F. Serono). After ICSI, the oocytes were cultured in IVF-50 medium (Scandinavian IVF Science AB, Gotenborg, Sweden). Four embryos, selected according to their morphology, were transferred transcervically on the second or third day after insemination. The number of transferred embryos was increased to five in women 38 years and in women with only grade-4 embryos available for transfer. Selective assisted hatching was performed in women 35 years of age and/or in women with at least one previous implantation failure. Tetracycline 200 mg bid (Monodoks; DEVA, Istanbul, Turkey) and methylprednisolone 16 mg/d (rednol 16 mg; Mustafa Nevzat Ilac Sanayi, Istanbul, Turkey) were administered for 5 days starting from the day of oocyte retrieval. The luteal phase was supplemented with intravaginal natural at a dose of 200 mg three times a day (Utrogestan; Laboratories Besins Iscovesco, aris, France) starting on the day of oocyte retrieval. regnancy was defined as two -hcg titers showing appropriate doubling assessed 12 and 14 days after ET. Clinical pregnancy was defined as the presence of a gestational sac or sacs with a viable embryo on transvaginal ultrasonography performed approximately 24 days after ET. FIGURE 1 Receiver operating characteristic curve analysis of different levels on the day of hcg administration and the occurrence of clinical pregnancy after ICSI and ET ng/ml yields a sensitivity of 71.6% and a specificity of 35.9%). The hypothetical line shows 50% sensitivity and 50% specificity. rogesterone Assay Serum levels were analyzed using a solid-phase, ligandlabeled, competitive chemiluminescent immunoassay (Immulite rogesterone; Diagnostic roducts Corporation, Los Angeles, CA). Samples were stored at 20 C. Before assay, the samples were brought to room temperature and then were mixed gently. The assay was performed according to the manufacturer s instructions. Low- and high-quality controls were used to analyze assay performance and to monitor the continued applicability of the stored master curve. Controls were processed at the beginning of every run containing patient samples to be tested for and also when readjusting. The assay has a calibration range of ng/ml. Withinrun and run-to-run coefficients of variation were 5.8 and 7.2%, respectively. The detection limit of the assay, defined as the concentration 2 SDs below the response at zero dose, is approximately 0.2 g/ml. Statistics A high level was defined as a value of 0.9 ng/ml. This arbitrary cutoff level was chosen to favor further comparison of our results with those of other investigators. Furthermore, according to our data, a serum cutoff level of 0.9 ng/ml yielded the highest sensitivity and specificity for predicting clinical pregnancy according to a receiver operating characteristic curve analysis (Fig. 1, Table 1). The groups were compared with use of Student s t-test, 2 analysis, and Fisher s exact test where applicable..05 was considered statistically significant. Because this was a 976 Urman et al. Serum and ICSI outcome Vol. 72, No. 6, December 1999

3 TABLE 1 Accuracy of measurements on the day of hcg for predicting the occurrence of pregnancy after ICSI and ET. cutoff Sensitivity Specificity retrospective study, a power analysis was not performed. If we had done a prospective study, we would have postulated a lower pregnancy rate (R) in the high group. To detect a clinically significant 25% decrease in the implantation rate (currently 14% in our second- or third-day ETs) with 0.05 and power 0.8, 1,091 case subjects and 2,182 control patients should have been included in the study. RESULTS 1 Specificity * * cutoff level that yielded the highest sensitivity and specificity. Table 2 shows the cycle characteristics and the response to controlled ovarian hyperstimulation in the high and low groups. Although the mean age was similar in the two groups, cycles with high levels were associated with a lower FSH level in the early follicular phase, reflective of a higher ovarian reserve in these women. Women with high levels had significantly more follicles of 14 mm in mean diameter, reached higher E 2 levels on the day of hcg administration, and yielded significantly more oocytes to follicle aspiration (.05). The fertilization rate was similar in TABLE 2 Cycle characteristics in women with low and high serum levels on the day of hcg. Cycle characteristic Low High value No. of ET cycles Mean age (y) NS Day-3 E 2 (pg/ml) NS Day-3 FSH (IU/L) Duration of stimulation (d) NS No. of FSH ampules NS eak E 2 (pg/ml) 1,858 2, No. of follicles of 14 mm Diameter of lead follicle (mm) NS No. of oocytes No. of MII oocytes (ICSI cycles) N fertilization (%) NS Cleaved embryos (%) NS No. of transferred embryos Note: NS not significant; N pronuclei. the two groups. More embryos were transferred in the high group, reflecting the increased number of embryos available for transfer. Clinical R per transfer, implantation rate per embryo, and abortion rates are shown in Table 3. The clinical R was higher in the high group (.05). However, the implantation rate per embryo was similar in the two groups. The increased clinical R was attributed to the increased number of embryos that were transferred. Abortion rates of 22.7% and 25.8% in the low and high groups were similar. There were 23 transfer cycles in which the serum level on the day of hcg administration was 3 ng/ml, the level considered as frank luteinization. Of these cycles, 5 (21.7%) resulted in a clinical pregnancy, of which 2 ended in firsttrimester spontaneous abortions. The numbers are too small TABLE 3 Outcome of ET cycles in women with low and high levels on the day of hcg. Outcome Low High value No. of ET cycles No. of embryos transferred Clinical pregnancies (%) 110 (36.9) 278 (45.4).05 Implantation rate per embryo (%) NS Abortions (%) 25 (22.7) 72 (25.8) NS Note: NS not significant. FERTILITY & STERILITY 977

4 to reach a conclusion regarding the impact of very high levels on cycle outcome. One patient conceived with a level of 5.8 ng/ml and an intrauterine gestational sac was documented; however, no embryo was visible and curettage was required. DISCUSSION Several reports have suggested an association between elevated serum levels on the day of hcg administration and unfavorable cycle outcomes in patients undergoing assisted reproduction. This has been thought to be mainly due to impaired endometrial receptivity (2, 13, 14). In this context, Fanchin et al. (13) showed that the rate of embryos reaching the blastocyst stage was similar in women with low and high levels; however, the R was significantly different and the difference was attributed to a low implantation rate. The above notwithstanding, the contention that a high serum level on the day of hcg administration is detrimental to cycle outcome has been questioned by the results of other studies (4, 6, 9, 10, 12, 15). Most of the data in the literature are derived from IVF-ET cycles; however, recently Ubaldi et al. (16) showed that embryos derived from ICSI are of similar quality and possess the same potential for implantation regardless of the serum level on the day of hcg administration. Our data suggest that in women undergoing controlled ovarian hyperstimulation with a long GnRH analogue protocol and pure FSH, the serum level on the day of hcg correlates with the degree of ovarian hyperstimulation. In this scenario, elevated levels do not seem to have an adverse effect on Rs and implantation rates. This also has been confirmed in previous studies (1, 4, 6, 8, 12, 15 17). Furthermore, in our study clinical Rs were higher in the high group because of an increased number of embryos available for transfer. An elevated serum level before the administration of hcg appears to reflect the total output of from multiple developing follicles and is closely linked to the degree of ovarian hyperstimulation. It has been shown previously that in accidentally overstimulated gonadotropin cycles, E 2 levels decline when gonadotropin administration is withheld before injection of hcg (18). Recently, Benadiva et al. (19) showed a similar decline in as well as E 2 when IVF cycles at risk for ovarian hyperstimulation syndrome were managed by so-called coasting. This finding further supports the notion that the serum level is closely related to the degree of ovarian hyperstimulation in response to gonadotropin treatment. In previous studies, with the exception of the study by Givens et al. (4), measurements of influenced the decision of the clinician regarding cycle cancellation. It is likely that some cycles with higher levels were canceled more often than cycles with lower levels. In our study, assays for were performed after the patients had completed their treatment and therefore had no impact on decision making or patient management. A power analysis performed at the beginning of data analysis showed that with an of 0.05 and a power of 0.8, 1,049 case subjects and 2,098 controls were necessary to show an approximately 25% decrease (14% to 10.5%) in the implantation rate of embryos in women with high levels. In our study population, retrospective power analysis yielded a power of for the detection of a similar decline in implantation rates. However, the initial assumption that a high level is detrimental for implantation was not supported in this study. In fact, we observed a higher implantation rate, although nonsignificant, in this group. The conclusion that levels are not indicative of treatment outcome may not hold for cycles with very high levels ( 3, which suggest frank luteinization. Only the occasional patient will present with a very high level on the day of hcg administration (23/911; 2.5% in our patient population). In this latter group, there were 5 (21.7%) clinical pregnancies, of which 2 (40%) miscarried. In conclusion, levels on the day of hcg appear to reflect the degree of ovarian hyperstimulation and have no impact on implantation rates. Furthermore, cycles with elevated levels are associated with an increased number of retrievable oocytes and transferable embryos. References 1. Edelstein M, Seltman H, Cox B, Robinson S, Shaw R, Muasher S. rogesterone levels on the day of human chorionic gonadotropin administration in cycles with gonadotropin-releasing hormone agonist suppression are not predictive of pregnancy outcome. Fertil Steril 1990;54: Silverberg K, Burns W, Olive D, Riehl R, Schenken R. Serum progesterone levels predict success of in vitro fertilization embryo transfer in patients stimulated with leuprolide acetate and human menopausal gonadotropins. J Clin Endocrinol Metab 1991;73: rien S, Canez M, Messer R. Increases in progesterone following human chorionic gonadotropin administration may predict cycle outcome in patients undergoing in vitro fertilization and embryo transfer. Fertil Steril 1994;62: Givens C, Schirock E, Dandekar, Martin M. Elevated serum progesterone levels on the day of human chorionic gonadotropin administration do not predict outcome in assisted reproduction cycles. Fertil Steril 1994;62: Dumesic D. eriovulatory serum progesterone levels as a predictor of pregnancy outcome during ovarian hyperstimulation for assisted reproductive technology. Fertil Steril 1994;62: Bustillo M, Stern J, Coulam C. Serum progesterone at the time of human chorionic gonadotropin does not predict pregnancy in in-vitro fertilization and embryo transfer. Hum Reprod 1995;11: Levy M, Smotrich D, Widra E, Sagoskin A, Murray D, Hall J. The predictive value of serum progesterone levels on in vitro fertilization outcome. J Assist Reprod Genet 1995;12: Abuzeid M, Sasy M. Elevated progesterone levels in the late follicular phase do not predict success of in vitro fertilization-embryo transfer. Fertil Steril 1996;65: Hofmann G, Khoury J, Johnson J, Thie J, Scott R. remature luteinization during controlled ovarian hyperstimulation for in vitro fertilization-embryo transfer has no impact on pregnancy outcome. Fertil Steril 1996;66: Moffitt D, Queenan J, Shaw R, Muasher S. rogesterone levels on the day of human chorionic gonadotropin do not predict pregnancy outcome from the transfer of fresh or cryopreserved embryos from the same cohort. Fertil Steril 1997;67: Ubaldi F, Camus M, Smitz J, Bennink H, Van Steirteghem A, Devroey. remature luteinization in in vitro fertilization cycles using gonadotropin releasing hormone agonist (GnRH-a) and recombinant follicle stimulating hormone (FSH) and GnRH-a and urinary FSH. Fertil Steril 1996;66: Huang JC, Jackson KV, Hornstein MD, Ginsburg ES. The effect of 978 Urman et al. Serum and ICSI outcome Vol. 72, No. 6, December 1999

5 elevated serum progesterone during ovulation induction in in vitro fertilization and embryo transfer. J Assist Reprod Genet 1996;13: Fanchin R, Righini C, Olivennes F, de Ziegler D, Selva J, Frydman R. remature progesterone elevation does not alter oocyte quality in in vitro fertilization. Fertil Steril 1996;65: Ezra Y, Simon A, Sherman Y, Benshushan A, Younis J, Laufer N. The effect of progesterone administration in the follicular phase of an artificial cycle on endometrial morphology: a model of premature luteinization. Fertil Steril 1994;62: Hofmann GE, Bentzien F, Bergh A, Garissi GJ, Williams MC, Guzman I, et al. remature luteinization in controlled ovarian hyperstimulation has no adverse effect on oocyte and embryo quality. Fertil Steril 1993;60: Ubaldi F, Smitz J, Wisanto A, Joris H, Schiettecatta J, Derde M, et al. Oocyte and embryo quality as well as pregnancy rate in intracytoplasmic sperm injection are not affected by high follicular phase serum progesterone. Hum Reprod 1995;10: Burns W, Witz C, Klein N, Silverberg K, Schenken R. Serum progesterone concentrations on the day after human chorionic administration and progesterone/oocyte ratios predict in vitro fertilization/embryo transfer outcome. J Assist Reprod Genet 1994;11: Urman B, ride S, Ho Yuen B. Management of overstimulated gonadotrophin cycles with a controlled drift period. Hum Reprod 1992;7: Benadiva C, Moomjy M, Kligman I, Liu H, Davis O, Rosenwaks Z. Withholding gonadotropin administration is an effective alternative to cryopreservation of all embryos for the prevention of ovarian hyperstimulation syndrome (OHSS). [Abstract no. -245]. In: Fifty-First Annual Meeting of the American Society for Reproductive Medicine, Seattle, WA. 1995; FERTILITY & STERILITY 979