Division of Reproductive Endocrinology, Department of Obstetrics and Gynecology, Taipei Veterans General Hospital; 2

RBMOnline - Vol 16. No 5. 2008 632-639 Reproductive BioMedicine Online; www.rbmonline.com/article/3209 on web 27 March 2008 Article Prospective comparison of short and long GnRH agonist protocols using recombinant gonadotrophins for IVF/ICSI treatments Dr Chi-Hong Ho received his MD degree in 2001 from the National Taiwan University, Taipei, Taiwan. After residency training at the National Taiwan University Hospital, he joined the hospital board in obstetrics and gynecology (2005) and completed a two-year infertility subspecialty training there. Currently, he is visiting staff in the Division of Reproductive Endocrinology, Taipei Veteran General Hospital. His main research interests are assisted reproduction techniques and folliculogenesis. Dr Chi-Hong Ho Chi-Hong Ho 1, Shee-Uan Chen 2,4, Fu-Shiang Peng 3, Chih-Yuan Chang 2, Yih-Ron Lien 2, Yu-Shih Yang 2 1 Division of Reproductive Endocrinology, Department of Obstetrics and Gynecology, Taipei Veterans General Hospital; 2 Department of Obstetrics and Gynecology, National Taiwan University Hospital and College of Medicine; 3 Department of Obstetrics and Gynecology, Far Eastern Memorial Hospital, Taipei, Taiwan 4 Correspondence: Tel: +886 2 23123456 ext. 5168; Fax: +886 2 23418557; e-mail: csu@ha.mc.ntu.edu.tw Abstract This is a prospective comparative study investigating cost and effectiveness of IVF/ intracytoplasmic sperm injection (ICSI) treatments after stimulation with recombinant gonadotrophins following either the short or long gonadotrophin-releasing hormone (GnRH) agonist protocol. Patients in the short protocol (n = 120) were administered buserelin nasal sprays from day 2 of the menstrual cycle and recombinant FSH from day 5. Patients in the long protocol (n = 120) were administered buserelin from the previous mid-luteal phase and recombinant FSH after achieving down-regulation. The average age and basal FSH concentrations of both groups were similar. The serum LH concentrations during ovarian stimulation were significantly higher with the short protocol. The total cost of recombinant gonadotrophins (US$527 ± 184 versus US$795 ± 244, P < 0.001) was significantly lower in the short protocol, but there was no significant difference in delivery rates (47.5 versus 36.7%) between the short and long protocols. LH flare-up during the short protocol does not seem to impair the treatment outcome. Using recombinant gonadotrophins, the short GnRH agonist protocol is an effective and cheaper choice for IVF/ICSI treatments. Keywords: IVF, long GnRH agonist protocol, short GnRH agonist protocol Introduction 632 The use of gonadotrophin-releasing hormone (GnRH) agonists during IVF programmes helps prevent spontaneous LH surge, thus reducing cycle cancellation rate and improving treatment outcome (Hughes et al., 1992). Strategies for ovarian stimulation that combine gonadotrophins and GnRH agonists fall mainly into two categories, namely the short and the long protocols. Both protocols can effectively prevent spontaneous LH surges (Acharya et al., 1992), yet they are significantly different in terms of cycle dynamics. In the short protocol, the GnRH agonist is administered from day 2 of the cycle until the use of human chorionic gonadotrophin (HCG), then gonadotrophins are given 2 3 days after GnRH agonist initiation. The short protocol takes advantage of the initial stimulatory effect of GnRH agonist on pituitary gonadotrophin release (also known as the flareup effect) to promote follicular development. In the long protocol, the use of GnRH agonist starts in the mid-luteal phase of the pretreatment cycle to attain pituitary downregulation with suppression of endogenous gonadotrophin secretion. After suppression of the pituitary ovarian axis, ovarian stimulation with exogenous gonadotrophins is commenced while GnRH agonist administration is continued concomitantly until the day of HCG administration. The question of which is superior between the long or the short 2008 Published by Reproductive Healthcare Ltd, Duck End Farm, Dry Drayton, Cambridge CB3 8DB, UK

protocol has been an issue of debate, and previous studies have involved retrospective or prospective approaches (Tan et al., 1992; Tarlatzis et al., 1993; Loutradis et al., 2005). In 2000, a Cochrane Review indicated that the long protocol was superior to the short protocol for IVF cycles with respect to the clinical pregnancy rate, with an odds ratio of 1.27 (Daya, 2000). The increased LH activity from the flare-up effect in the short protocol was proposed by some investigators to affect oocyte quality (San Roman et al., 1992; Cramer et al., 1999). However, in these studies, human menopausal gonadotrophin (HMG) consisting of equivalent activities of FSH and LH was primarily used. Recent development and application of recombinant FSH (rfsh) has resulted in the gradual replacement of urinary HMG by rfsh in most assisted reproductive units. Several studies have been performed on the effects of the GnRH agonist long protocol in combination with rfsh (Cheon et al., 2004; Al-Inany et al., 2005). These indicated that downregulation with GnRH agonist in some normogonadotrophic women led to a profound lowering of the LH concentrations and this may, in turn, impair adequate oestradiol synthesis, increase the duration of stimulation and possibly reduce the rate of clinical pregnancies (Liu et al., 2005). Studying the rate of DNA fragmentation and caspase-3 activity for cumulus cells in women receiving the long protocol, Ruvolo et al. (2007) found that the apoptosis rate was higher for those treated with rfsh alone than for those receiving rfsh and recombinant LH (rlh). Patients with profound LH suppression during the long GnRH agonist treatment might benefit from supplementation with rlh. However, the effects of rfsh in the short GnRH agonist protocol and its superiority or inferiority compared with the long protocol have not been investigated. To address these questions, a prospective, comparative study was designed for a group of women undergoing IVF or intracytoplasmic sperm injection (ICSI) treatment with a normal ovarian reserve and who were <40 years of age. The endocrine changes in serum LH, oestradiol and progesterone concentrations during the follicular phase were explored, together with oocyte number, pregnancy outcome, and the drug cost for both the short and long GnRH agonist protocols. As far as is known, this is the first report that prospectively compares the short and long GnRH agonist protocols when using recombinant gonadotrophins. Materials and methods Study population This study was approved by the Institutional Ethics Review Board of National Taiwan University Hospital. Infertile patients were enrolled after standard medical work-up procedures in order to determine whether they met the inclusion criteria and whether treatment with IVF or ICSI was necessary. Informed consent was obtained from each participant. Between May 2004 and June 2006, a total of 240 patients were enrolled for IVF/ICSI treatment. Eligible patients were randomized 1:1 to the long or the short GnRH agonist protocol, based on a computer-generated randomization list. Randomization was performed by a centralized telephone procedure as soon as informed consent was available. The inclusion criteria were: (i) women aged between 20 and 39 years in good physical and mental health; (ii) a history of regular menstrual cycles, ranging from 25 to 35 days; (iii) tubal factor, male factor, or unexplained infertility; (iv) a body mass index between 18 and 29 kg/m 2 at the time of randomization; (v) hysterosalpingography or transvaginal ultrasound documenting a uterus with expected normal function within 6 months before randomization; (vi) transvaginal ultrasound documenting the presence of bilateral ovaries without evidence of abnormality; and (vii) an early follicular phase serum FSH concentration of 1 12 IU/l. The exclusion criteria included: (i) endometriosis stage III/IV or polycystic ovary syndrome; (ii) more than three previously consecutive unsuccessful IVF cycles; (iii) a previous poor response to an IVF cycle, defined as cycle cancellation due to limited response or fewer than four oocytes retrieved; (iv) a history of recurrent miscarriage; (v) severe ovarian hyperstimulation syndrome (OHSS) in a previous IVF cycle; (vi) any significant systemic disease, endocrine or metabolic abnormalities; and (vii) hypersensitivity to any trial product. Study design The GnRH agonist used in this study was a buserelin nasal spray (Supremon; Hoechst, Frankfurt am Main, Germany). Patients in the short protocol were administered buserelin 200 μg q.i.d. from day 2 of their menstrual cycle and rfsh (Puregon; Organon, Oss, Netherlands) 200 IU/day subcutaneously from day 5. In the long protocol, patients were administered buserelin from their previous midluteal phase, which is 7 days before estimated start of next menses. After down-regulation was achieved (menstrual bleeding occurred and serum oestradiol < 50 pg/ml), the dose of buserelin was halved (100 μg q.i.d.) and ovarian stimulation was commenced on the next day with a daily dose of 200 IU rfsh subcutaneously. Blood sampling (FSH, LH and oestradiol) was performed on menstrual cycle day 2 (short protocol) or the day just before the commencement of gonadotrophin stimulation (long protocol). From day 7, transvaginal ultrasonography and serum hormone analysis (LH, oestradiol, and progesterone) were performed every other day. The dose of rfsh was adjusted from day 7 according to the ovarian response monitored by transvaginal ultrasonography and serum oestradiol concentration. If the serum LH concentration was below 1.0 IU/l, rlh (Luveris; Serono, Geneva, Switzerland), 75 IU/day subcutaneously was added to the treatment until the day of HCG administration. When at least two follicles had reached a diameter of 18 mm or more, GnRH agonist and gonadotrophins were stopped and a single bolus of 10,000 IU HCG (Profasi; Serono) was administered intramuscularly. The cycle was cancelled if fewer than two follicles were observed during the ultrasound scan. In fact, all the patients had at least three follicles on the day of HCG administration and therefore no cycle was cancelled. All follicles with a diameter 10 mm were retrieved transvaginally 34 36 h after HCG administration. Subsequently, IVF or ICSI was performed. Embryo transfer 633

634 was performed on day 3 after oocyte retrieval. According to the guidelines of the Taiwan Society for Reproductive Medicine, no more than three embryos could be transferred to women younger than 35 years and no more than four embryos to women between 35 and 39 years. Luteal support commenced 2 days after oocyte retrieval with intramuscular progesterone (Tai Yu, Hsinchu, Taiwan) 50 mg/day and oral oestradiol valerate (Estrade; Synmosa, Taipei, Taiwan) 6 mg b.i.d. Serum HCG was checked on day 16 after oocyte retrieval and a concentration above 50 IU/l was considered positive. Ultrasound examination was performed 1 week later to confirm clinical pregnancy and determine the number of intrauterine gestational sacs. The implantation rate was defined as the ratio of the number of gestational sacs to the number of embryos transferred. The presence of at least one viable fetus at 12 weeks gestation was classified as ongoing pregnancy, and all pregnancies were followed up to delivery. A delivery cycle was defined as a cycle that resulted in at least one liveborn neonate. Serum FSH, LH, oestradiol and progesterone concentrations were measured by means of immunoassay (Immulite 2000; DPC, Flanders, NJ, USA). Patient serum samples were assayed immediately upon sample acquisition. Medical staff assessing biochemical results and performing ultrasound investigation for follicle and pregnancy status were blinded to the specific group assignment of the participating patients. Study end-points and statistical analysis The study end-points were cost and effectiveness per IVF/ ICSI treatment cycle, including delivery rate, clinical pregnancy rate, ongoing pregnancy rate, number of oocytes retrieved, fertilization rate, and the cost of recombinant gonadotrophins. Patients underwent only one treatment cycle in the study. All cases of ectopic pregnancy and moderate/ severe OHSS were recorded. Comparisons of outcomes between the two groups were performed by independent t-test for continuous data and by chi-squared or Fisher s Exact test for binary variables. A P- value <0.05 was considered to be statistically significance. Data were analysed using the commercially available software package Statistics Package for Social Sciences (SPSS) 11.0 (SPSS Inc., Chicago, IL, USA) and presented as mean ± SD unless otherwise specified. Results Baseline A total of 240 infertile couples who fulfilled the inclusion/ exclusion criteria were included in this study, and these couples were equally divided into the short and long protocol groups. Table 1 shows the demographic features and baseline characteristics of the two groups. No significant difference was found between the two groups with regard to the distribution of age (33.8 ± 3.5 versus 33.7 ± 3.2 years) and baseline serum FSH concentration (6.8 ± 1.9 versus 6.4 ± 1.6 IU/l). There was also no difference in the distribution of infertility causes or the percentage undergoing ICSI. For the couples using IVF, there was no difference in the sperm concentration but the sperm motility was higher in the long protocol group (54.5 versus 59.9%, P = 0.016). Treatment outcomes Clinical outcome parameters are shown in Table 2. The total amount of rfsh (1315 ± 455 versus 1866 ± 512 IU, P < 0.001), the need for rlh supplementation (1 versus 49 patients, P < 0.001), and the duration of rfsh stimulation (7.3 ± 1.7 versus 9.9 ± 1.6 days, P < 0.001) were significantly lower in the short protocol. The serum LH concentrations of the short protocol on day 2, day 7 and the day of HCG administration were all higher (Figure 1), as were the serum oestradiol concentrations (Figure 2). However, the serum progesterone concentrations on day 7 (1.2 ± 0.8 versus 1.1 ± 0.7 ng/ml) and the day of HCG administration (1.8 ± 1.1 versus 1.8 ± 1.0 ng/ml) were comparable. The thickness of the endometrium on the day of HCG administration was significantly greater in the long protocol (11.6 ± 2.1 versus 12.3 ± 2.7 mm, P = 0.023). The number of follicles (12.3 ± 6.5 versus 16.3 ± 8.1, P < 0.001), retrieved oocytes (11.2 ± 5.8 versus 14.5 ± 6.6, P < 0.001) and fertilized oocytes (8.0 ± 4.3 versus 10.3 ± 5.2, P < 0.001) were also significantly higher in the long protocol. Thus, the serum oestradiol per oocyte on day of HCG administration was significantly lower in this group (P < 0.001). The fertilization rate and the number of transferred embryos were comparable. The short protocol had significantly higher implantation rate (26.8 versus 19.8%, P = 0.024) and a trend toward better clinical pregnancy rate (50.8 versus 43.3%), ongoing pregnancy rate (49.2 versus 40%, P = 0.153) and delivery rate (47.5 versus 36.7%), although the differences were not significant at the 5% level. Nineteen patients (33.3%) in the short protocol group and 13 patients (29.5%) in the long protocol group had twin live births. The risk of ectopic pregnancy and moderate/severe OHSS was comparable across both groups. To further understand the effect of serum LH concentration on treatment outcomes, the delivery rates at different LH concentrations for each group were compared. In the short protocol, regardless of whether it was on day 2, day 7 or day of HCG, the patients with an LH concentration higher than the mean did not show a poorer delivery rate. In the long protocol, profound LH suppression (LH <1 IU/l) on day 2, day 7 or day of HCG did not impair delivery rate either (Table 3). However, among those with serum LH concentrations 1 IU/l on the day of HCG administration, 26 patients received rlh supplementation and 11 had live births, but all others failed to become pregnant (11/26 versus 0/19, P = 0.001). Discussion Because of the use of HMG, the results of previous studies may favour the long GnRH agonist protocol in normogonadotrophic women over the short protocol (Tan et al., 1992; Tarlatzis et al., 1993; Cramer et al., 1999; Daya 2000). Nevertheless, in this prospective comparison, it has been demonstrated that using rfsh as the primary gonadotrophin, the short protocol achieved a comparable delivery rate to the long protocol in normogonadotrophic women. The cause for this discrepancy was not clear, but

Table 1. Demographics and baseline characteristics of patients in the short and long protocol groups. Parameter Short protocol Long protocol P-value (n = 120) (n = 120) Mean age (years) ± SD 33.8 ± 3.5 33.7 ± 3.2 NS Mean baseline FSH (IU/l) ± SD 6.8 ± 1.9 6.4 ± 1.6 NS No. with tubal factor infertility (%) 30 (25.0) 36 (30.0) NS No. with male factor infertility (%) 83 (69.2) 80 (66.7) NS No. with unexplained infertility (%) 16 (13.3) 17 (14.2) NS No. of patients treated with ICSI (%) 81 (67.5) 80 (66.7) NS Sperm concentration of IVF couples ( 10 6 /ml) 79.9 ± 46.5 62.1 ± 36.1 NS Sperm motility of IVF couples (mean % ± SD) 54.5 ± 9.7 59.9 ± 9.7 0.016 ICSI = intracytoplasmic sperm injection; NS = not statistically significant. Table 2. Comparison of clinical parameters and treatment outcomes for patients in the short and long protocol groups. Parameter Short protocol Long protocol P-value (n = 120) (n = 120) Dosage of rfsh (IU) 1315 ± 455 1866 ± 512 <0.001 Duration of rfsh stimulation (days) 7.3 ± 1.7 9.9 ± 1.6 <0.001 No. of patients needing rlh 1 (0.8) 49 (40.8) <0.001 Total cost of gonadotrophins (US$) 527 ± 184 795 ± 244 <0.001 Endometrial thickness on day of HCG 11.6 ± 2.1 12.3 ± 2.7 0.023 administration (mm) No. of follicles 10 mm 12.3 ± 6.5 16.3 ± 8.1 <0.001 No. of oocytes retrieved 11.2 ± 5.8 14.5 ± 6.6 <0.001 Serum oestradiol per oocyte (pg/ml) 288 ± 134 181 ± 96 <0.001 No. of embryos fertilized 8.0 ± 4.3 10.3 ± 5.2 <0.001 Fertilization rate (%) 73.7 ± 17.5 72.1 ± 17.8 NS No. of embryos transferred 3.1 ± 0.6 3.2 ± 0.5 NS No. of ectopic pregnancies (%) 5 (4.2) 2 (1.7) NS No. with moderate/severe OHSS 5 (4.2) 4 (3.3) NS Implantation (%) 98/366 (26.8) 75/379 (19.8) 0.024 No. of clinical pregnancies (%) 61 (50.8) 52 (43.3) NS No. of ongoing pregnancies (%) 59 (49.2) 48 (40.0) NS No. of deliveries (%) 57 (47.5) 44 (36.7) NS Values are mean ± SD, unless otherwise stated. rfsh = recombinant FSH; rlh = recombinant LH; HCG = human chorionic gonadotrophin; NS = not statistically significant; OHSS = ovarian hyperstimulation syndrome. 635

5000 Short protocol Long protocol 4000 Oestradiol concentration (pg/ml) 3000 2000 1000 0 Day 2 Day 7 Day of HCG Figure 1. Changes in serum luteinizing hormone (LH) concentrations during the treatment cycles. The short protocol group exhibited a higher serum LH concentration on day 2 (4.0 ± 2.0 versus 2.7 ± 1.8 IU/l, P < 0.001), day 7 (7.1 ± 3.5 versus 1.7 ± 1.7 IU/l, P < 0.001), and day of human chorionic gonadotrophin administration (day of HCG, 5.2 ± 3.2 versus 1.6 ± 1.2 IU/l, P < 0.001). 12 Short protocol Long protocol 10 8 LH concentration (IU/l) 6 4 636 2 0 Day 2 Day 7 Day of HCG Figure 2. Changes in serum oestradiol concentrations during treatment cycles. Serum oestradiol concentrations were significantly higher with the short protocol on day 2 (33 ± 12 versus 24 ± 10 pg/ml, P < 0.001), day 7 (541 ± 421 versus 378 ± 361 pg/ml, P = 0.001), and day of human chorionic gonadotrophin administration (day of HCG, 3030 ± 1685 versus 2580 ± 1846 pg/ml, P = 0.050).

Table 3. Delivery rates for various LH concentrations (IU/l) on day 2, day 7 and day of human chorionic gonadotrophin (HCG) administration. Day No. of livebirths (%) Short protocol Long protocol Day 2 LH < 4.0 (n = 69) LH 4.0 (n = 51) LH < 1.0 (n = 16) LH 1.0 (n = 104) 35 (50.7) 22 (43.1) 5 (31.3) 39 (37.5) Day 7 LH < 7.1 (n = 69) LH 7.1 (n = 51) LH < 1.0 (n = 48) LH 1.0 (n = 72) 35 (50.7) 22 (43.1) 20 (41.7) 24 (33.3) Day of HCG LH < 5.2 (n = 70) LH 5.2 (n = 50) LH < 1.0 (n = 45) LH 1.0 (n = 75) administration 33 (47.1) 24 (48.0) 11 (24.4) 33 (44.0) There were no statistically significant differences between the delivery rates at different LH concentrations on any of the days. HMG, which consists of equivalent activities of FSH and LH, was primarily used in those previous studies. In the short protocol when using HMG, excessive endogenous LH flareup and exogenous LH from the HMG may be a cause of a poorer pregnancy outcome (San Roman et al., 1992). LH stimulates androgen production in the theca cells, which is aromatized into oestradiol by granulosa cells under the action of FSH (Drakakis et al., 2005). During the mid-follicular phase, granulosa cells acquire LH receptors besides the FSH receptors; both are important for pre-ovulatory oestradiol production and possibly for the maturation of competent oocytes (Hillier, 2001). Furthermore, LH may also play an important role in implantation, and may influence uterine receptivity via ovarian secretion of oestradiol or through direct effects on endometrium, myometrium and uterine vessels (Rao, 2001; Shemesh, 2001). Circulating endogenous concentrations of LH are reduced in women undergoing down-regulation with the long GnRH agonist protocol (Humaidan et al., 2004). When using rfsh for the long protocol, the necessity of LH supplementation has been an issue of considerable debate (Westergaard et al., 2001; Tesarik and Mendoza 2002; Humaidan et al., 2004; Cabrera et al., 2005; De Placido et al., 2005; Andersen et al., 2006; Tarlatzis et al., 2006). A greater than 50% reduction of endogenous LH concentration during the early to mid-follicular phase is associated with a significant reduction in the live birth rate (Lahoud et al., 2006). The threshold concentration of LH equates to a serum concentration of 0.5 1.35 IU/l (Fischer, 2007). In a study analysing an oocyte donation programme (Tesarik and Mendoza 2002), the addition of exogenous LH could increase the oocyte number and implantation rate when LH was suppressed to 1 IU/l. In this study, rlh was added when endogenous LH was significantly suppressed (<1 IU/l) during ovarian stimulation. It was found that all 19 patients with LH concentrations 1 IU/l on the day of HCG administration, but without rlh supplementation did not achieve a live birth. During the second half of the follicular phase, preovulatory follicular development progresses normally if LH is at a concentration within the window (Palermo, 2007). Marrs et al. (2004) suggested that rlh supplementation is beneficial for women aged 35 years, but the mean age of the 19 patients was 33.2 ± 3.4 years, with only four patients 35 years. Routine rlh supplementation seems to be beneficial to these patients. The long protocol necessitates a rather long treatment period in order to achieve pituitary down-regulation, and requires the substantial administration of exogenous gonadotrophins to attain adequate follicular growth (Ben-Rafael et al., 1991; Smitz et al., 1992). Using GnRH antagonist instead of long GnRH agonist can reduce the number of days and ampoules of gonadotrophins, but the clinical pregnancy rate is significantly lower (Al-Inany et al., 2007). In this study, the average cost of recombinant gonadotrophins in the short protocol (US$527) was significantly lower (P < 0.001) than that in the long protocol (US$795). It is estimated that recombinant gonadotrophins and GnRH agonist account for approximately 50% of IVF costs (Hovatta and Cooke, 2006). Therefore, when recombinant gonadotrophins replace urinary HMG, the short protocol has significant advantages in terms of lower expense, fewer injections and greater convenience. The initial GnRH agonistic action of the short protocol may result in an inappropriately high concentration of LH, which may increase intrafollicular androgen and thus affect oocyte quality (Tan et al., 1992). Liu et al. (2005) indicated that using the short protocol, higher serum LH (i.e. higher than the mean concentration) on day 8 significantly reduced implantation and clinical pregnancy rates. However, in this study, patients with serum LH concentrations on day 7 higher than the mean (7.1 IU/l) did not show a poorer delivery rate. The effect of endogenous LH concentration during the short protocol on the pregnancy outcome may need further clarification. Significantly fewer oocytes were retrieved (11.2 ± 5.8 versus 14.5 ± 6.6, P < 0.001) with the short protocol, which is consistent with previous studies (Hughes et al., 1992; San Roman et al., 1992; Tan et al., 1992; Filicori et al., 1996; Cramer et al., 1999). Reduced folliculogenesis during the flare-up protocol could be due to asynchronized development 637

638 of the follicles or a shorter duration of stimulation. However, no detrimental effect on implantation, pregnancy, and delivery rates was found with the short protocol. The influence of high serum oestradiol concentrations on the outcome of IVF cycle is controversial (Kosmas et al., 2004). In the present study, serum concentrations of oestradiol and the oestradiol:oocyte ratios on day of HCG were found to be higher in the short protocol, but the IVF outcomes for the short protocol were not impaired. Previous research has indicated that no pregnancy occurs if endometrial thickness measures < 7 mm (Schild et al., 2001). Above the minimum thickness, however, there is no relationship between endometrial thickness and IVF outcome (Dietterich et al., 2002; Rashidi et al., 2005). In this study, the difference in mean endometrial thickness of the two protocols was less than 1 mm (11.6 versus 12.4 mm), with both thicker than 7 mm. The reduced thickness of the endometrium found with the short protocol may be caused by the shorter duration of stimulation (Yoeli et al., 2004). Oocyte quality from long GnRH agonist and GnRH antagonist protocol seems to be comparable within cryopreserved embryo transfer cycles (Eldar-Geva et al., 2007) and within oocyte donation cycles (Prapas et al., 2005). Bodri et al. (2006) compared the quality of oocytes from donors using the GnRH antagonist protocol and the short GnRH agonist protocol. Ovarian response, embryo development, pregnancy rate and implantation rate were comparable between the two protocols, with the exception of significantly higher serum oestradiol concentrations for the short GnRH agonist group. These results again provide evidence that the LH flare-up effect in the short protocol using rfsh does not seem to affect oocyte quality. In conclusion, the short and long GnRH agonist protocols have similar delivery rates in normogonadotrophic women when recombinant gonadotrophins are used during IVF/ICSI treatment. The short protocol is a cheaper method and is worth wider adoption when using recombinant gonadotrophins. References Acharya U, Small J, Randall J et al. 1992 Prospective study of short and long regimens of gonadotropin-releasing hormone agonist in in vitro fertilization program. Fertility and Sterility 57, 815 818. Al-Inany H, Abou-Setta AM, Aboulghar M 2007 Gonadotrophinreleasing hormone antagonists for assisted conception: a Cochrane Review. Reproductive BioMedicine Online 14, 640 649. Al-Inany H, Aboulghar MA, Mansour RT et al. 2005 Ovulation induction in the new millennium: recombinant folliclestimulating hormone versus human menopausal gonadotropin. Gynecological Endocrinology 20, 161 169. Andersen AN, Devroey P, Arce JC 2006 Clinical outcome following stimulation with highly purified hmg or recombinant FSH in patients undergoing IVF: a randomized assessor-blind controlled trial. Human Reproduction 21, 3217 3227. Ben-Rafael Z, Lipitz S, Bider D et al. 1991 Ovarian hyporesponsiveness in combined gonadotropin-releasing hormone agonist and menotropin therapy is associated with low serum follicle-stimulating hormone levels. Fertility and Sterility 55, 272 275. Bodri D, Vernaeve V, Guillen JJ et al. 2006 Comparison between a GnRH antagonist and a GnRH agonist flare-up protocol in oocyte donors: a randomized clinical trial. Human Reproduction 21, 2246 2251. Cabrera RA, Stadtmauer L, Mayer JF et al. 2005 Follicular phase serum levels of luteinizing hormone do not influence delivery rates in in vitro fertilization cycles down-regulated with a gonadotropin-releasing hormone agonist and stimulated with recombinant follicle-stimulating hormone. Fertility and Sterility 83, 42 48. Cheon KW, Byun HK, Yang KM et al. 2004 Efficacy of recombinant human follicle-stimulating hormone in improving oocyte quality in assisted reproductive techniques. Journal of Reproductive Medicine 49, 733 738. Cramer DW, Powers DR, Oskowitz SP et al. 1999 Gonadotropinreleasing hormone agonist use in assisted reproduction cycles: the influence of long and short regimens on pregnancy rates. Fertility and Sterility 72, 83 89. Daya S 2000 Gonadotropin releasing hormone agonist protocols for pituitary desensitization in in vitro fertilization and gamete intrafallopian transfer cycles. Cochrane Database of Systemic Reviews, CD001299- De Placido G, Alviggi C, Perino A et al. 2005 Recombinant human LH supplementation versus recombinant human FSH (rfsh) step-up protocol during controlled ovarian stimulation in normogonadotrophic women with initial inadequate ovarian response to rfsh. A multicentre, prospective, randomized controlled trial. Human Reproduction 20, 390 396. Dietterich C, Check JH, Choe JK et al. 2002 Increased endometrial thickness on the day of human chorionic gonadotropin injection does not adversely affect pregnancy or implantation rates following in vitro fertilization-embryo transfer. Fertility and Sterility 77, 781 786. Drakakis P, Loutradis D, Kallianidis K et al. 2005 Small doses of LH activity are needed early in ovarian stimulation for better quality oocytes in IVF-ET. European Journal of Obstetrics, Gynecology, and Reproductive Biology 121, 77 80. Eldar-Geva T, Zylber-Haran E, Babayof R et al. 2007 Similar outcome for cryopreserved embryo transfer following GnRHantagonist/GnRH-agonist, GnRH-antagonist/HCG or long protocol ovarian stimulation. Reproductive BioMedicine Online 14, 148 154. Filicori M, Flamigni C, Cognigni GE et al. 1996 Different gonadotropin and leuprorelin ovulation induction regimens markedly affect follicular fluid hormone levels and folliculogenesis. Fertility and Sterility 65, 387 393. Fischer R 2007 Understanding the role of LH: myths and facts. Reproductive BioMedicine Online 15, 468 477. Hillier SG 2001 Gonadotropic control of ovarian follicular growth and development. Molecular and Cellular Endocrinology 179, 39 46. Hovatta O, Cooke I 2006 Cost-effective approaches to in vitro fertilization: Means to improve access. International Journal of Gynaecology and Obstetrics 94, 287 291. Hughes EG, Fedorkow DM, Daya S et al. 1992 The routine use of gonadotropin-releasing hormone agonists prior to in vitro fertilization and gamete intrafallopian transfer: a meta-analysis of randomized controlled trials. Fertility and Sterility 58, 888 896. Humaidan P, Bungum M, Bungum L et al. 2004 Effects of recombinant LH supplementation in women undergoing assisted reproduction with GnRH agonist down-regulation and stimulation with recombinant FSH: an opening study. Reproductive BioMedicine Online 8, 635 643. Kosmas IP, Kolibianakis EM, Devroey P 2004 Association of estradiol levels on the day of hcg administration and pregnancy achievement in IVF: a systematic review. Human Reproduction 19, 2446 2453. Lahoud R, Al-Jefout M, Tyler J et al. 2006 A relative reduction in mid-follicular LH concentrations during GnRH agonist IVF/ ICSI cycles leads to lower live birth rates. Human Reproduction

21, 2645 2649. Liu SY, Han JL, Peng XD et al. 2005 Effect of midfollicular luteinizing hormone levels on ovarian response and pregnancy outcome in patients undergoing in vitro fertilization in a shortterm protocol. Fertility and Sterility 83, 1043 1046. Loutradis D, Stefanidis K, Drakakis P et al. 2005 Comparison between short and long protocols in an ICSI programme. European Journal of Obstetrics, Gynecology, and Reproductive Biology 120, 69 72. Marrs R, Meldrum D, Muasher S et al. 2004 Randomized trial to compare the effect of recombinant human FSH (follitropin alfa) with or without recombinant human LH in women undergoing assisted reproduction treatment. Reproductive BioMedicine Online 8, 175 182. Palermo R 2007 Differential actions of FSH and LH during folliculogenesis. Reproductive BioMedicine Online 15, 326 337. Prapas N, Prapas Y, Panagiotidis Y et al. 2005 GnRH agonist versus GnRH antagonist in oocyte donation cycles: a prospective randomized study. Human Reproduction 20, 1516 1520. Rao CV 2001 Multiple novel roles of luteinizing hormone. Fertility and Sterility 76, 1097 1100. Rashidi BH, Sadeghi M, Jafarabadi M et al. 2005 Relationships between pregnancy rates following in vitro fertilization or intracytoplasmic sperm injection and endometrial thickness and pattern. European Journal of Obstetrics, Gynecology, and Reproductive Biology 120, 179 184. Ruvolo G, Bosco L, Pane A et al. 2007 Lower apoptosis rate in human cumulus cells after administration of recombinant luteinizing hormone to women undergoing ovarian stimulation for in vitro fertilization procedures. Fertility and Sterility 87, 542 546. San Roman GA, Surrey ES, Judd HL et al. 1992 A prospective randomized comparison of luteal phase versus concurrent follicular phase initiation of gonadotropin-releasing hormone agonist for in vitro fertilization. Fertility and Sterility 58, 744 749. Schild RL, Knobloch C, Dorn C et al. 2001 Endometrial receptivity in an in vitro fertilization program as assessed by spiral artery blood flow, endometrial thickness, endometrial volume, and uterine artery blood flow. Fertility and Sterility 75, 361 366. Shemesh M 2001 Actions of gonadotrophins on the uterus. Reproduction 121, 835 842. Smitz J, Ron-El R, Tarlatzis BC 1992 The use of gonadotrophin releasing hormone agonists for in vitro fertilization and other assisted procreation techniques: experience from three centres. Human Reproduction 7 (Suppl. 1), 49 66. Tan SL, Kingsland C, Campbell S et al. 1992 The long protocol of administration of gonadotropin-releasing hormone agonist is superior to the short protocol for ovarian stimulation for in vitro fertilization. Fertility and Sterility 57, 810 814. Tarlatzis B, Tavmergen E, Szamatowicz M et al. 2006 The use of recombinant human LH (lutropin alfa) in the late stimulation phase of assisted reproduction cycles: a double-blind, randomized, prospective study. Human Reproduction 21, 90 94. Tarlatzis BC, Pados G, Bontis J et al. 1993 Ovarian stimulation with buserelin/hmg/hcg: prospective randomized study of short versus long protocol. Human Reproduction 8, 807 812. Tesarik J, Mendoza C 2002 Effects of exogenous LH administration during ovarian stimulation of pituitary down-regulated young oocyte donors on oocyte yield and developmental competence. Human Reproduction 17, 3129 3137. Westergaard LG, Erb K, Laursen SB et al. 2001 Human menopausal gonadotropin versus recombinant follicle-stimulating hormone in normogonadotropic women down-regulated with a gonadotropin-releasing hormone agonist who were undergoing in vitro fertilization and intracytoplasmic sperm injection: a prospective randomized study. Fertility and Sterility 76, 543 549. Yoeli R, Ashkenazi J, Orvieto R et al. 2004 Significance of increased endometrial thickness in assisted reproduction technology treatments. Journal of Assisted Reproduction and Genetics 21, 285 289. Declaration: The authors report no financial or commercial conflicts of interest. Received 28 September 2007; refereed 8 October 2007; accepted 2 January 2007. 639