P-557 CLOMIPHENE CITRATE VS. LETROZOLE: COMPARISON OF CYCLE CHARACTERISTICS AND PREGNANCY RATES. C. M. DeUgarte, K. Hayter, C. M. Blacker, G. Wegienka, R. C. Strickler. Reproductive Endocrinology and Infertility, Pacific Fertility Center, Los Angeles, CA; Obstetrics & Gynecology, Henry Ford Hospital, Detroit, MI; Epidemiology and Biostatistics, Henry Ford Hospital, Detroit, MI. OBJECTIVE: To evaluate the pregnancy rates of patients who underwent ovulation induction with Clomiphene Citrate(CC) or Letrozole. DESIGN: Prospective Observational. MATERIALS AND METHODS: 167 women underwent 338 cycles of ovulation induction using Clomiphene Citrate (144 cycles) or Letrozole (194 cycles) combined with hCG and IUI or coitus. The patients, all treated in one clinic, had dysfunctional ovulation, male factor or a combination of the two. Medications were administered on cycle days 3–7 at dose ranges of CC 50–150 mg and Letrozole 2.5–5 mg per day. Since patients may have had more than one treatment cycle with different medications, we lim- ited our analysis of pregnancy rates to the first treatment cycle for each woman (77 CC cycles and 88 Letrozole cycles). We compared the pregnancy rate estimates between the two groups using a Chi-square test. We also deter- mined 95% confidence intervals for the estimates. RESULTS: The average age for patients using CC was 32.4 years (SD ¼ 4.4) and 33.4 years (SD ¼ 4.9) for Letrozole users. The proportions of women who became pregnant on the first cycles were: 18.2% (95% CI 9.6%, 26.8%) for the CC group and 10.2% (95% CI 3.9%, 16.6%) for the Letrozole group (Chi-square test of rates between the groups P¼0.14). Of the 14 CC pregnancies, 7 (50%) were with IUI and 6 (42.9%) were with coitus and 1 (7.1%) had missing data. Of the 9 Letrozle pregnancies, 1 (11.1%) was with IUI and 8 followed coitus. CONCLUSIONS: Although there is no statistical difference in pregnancy rates between the Clomiphene vs. Letrozole groups, the trend favored preg- nancy in the Clomiphene group. Supported by: None. P-558 OVARIAN STIMULATION CYCLES WITH A FIXEDVS. A FLEXI- BLE GONADOTROPIN RELEASING HORMONE (GNRH) ANTAG- ONIST PROTOCOL COMPARED TO A LONG GNRH ANALOG PROTOCOL IN AN OOCYTE DONATION PROGRAM. J. C. Lopez, F. Cano, A. Jaller, J. L. Giraldo, M. N. Posada, G. C. Raigosa. Obstetrics and Gynecology, Instituto Antioqueno de Reproduccion INSER, Medellin, Antioquia, Colombia. OBJECTIVE: GnRH agonists(a) and antagonists prevent premature luteinization during follicular stimulation. Although GnRH antagonist cycles are shorter and require less gonadotropins they are often associated with lower embryo implantation and pregnancy rates compared to GnRHa cycles. The objective of this study is to compare the effect of GnRHa and antagonists on oocyte donation cycles. The main outcome measures are: number of oocytes retrieved, units of gonadotropins required, clinical and implantation rates per cycle. DESIGN: Retrospective study evaluating all oocyte donation cycles between January 2002 and December 2006 at the Instituto Antioqueno de Reproduccion. MATERIALS AND METHODS: All donors allocated to the long protocol received leuprolide acetate 0.1 mg SC starting on cycle day 21 until the day of hCG (human chorionic gonadotropin). Donors in the GnRH antagonist group received 0.25 mg of ganirelix starting on stimulation day 6 (fixed protocol) or when the leading follicle reached an average diameter of 14 mm (flexible protocol) until the day of HCG. All donors received recombinant follicular stimulating hormone with or without human menopausal gonadotropin start- ing on cycle day 2 or 3. HCG was administered when two follicles reached an average diameter of 18mm. Good quality oocytes were classified as meta- phase II. Statistical analysis was done with epi-info 2004, ANOVA, and Krushal-Wallis. A P value of < 0.05 was considered clinically significant. RESULTS: The total units of gonadotropins administered per cycle were significantly lower in the fixed protocol with GnRH antagonists (1469 IU) compared to the flexible protocol and the long protocol with GnRHa (1842 IU, P¼0.01 and 1892, P¼0.004, respectively). There was a non significant (NS) difference in the other outcomes. TABLE. Agonists And Antagonists In An Oocyte Donation Program GnRH Agonists (n ¼ 21) Antagonists Fixed (n ¼ 21) Antagonists Flexible (n ¼ 43) P value Age of donor 25.3  3 25.76  3.3 26.5  3 NS Gonadotropins requirement (IU) 1892  379 a 1469  512 b 1842  585 c <0.05 b Oocytes retrieved 16.5  6 17.7  7 15.7  6 NS % Metaphase II 60 64 60 NS Embryos transferred 3.2  0.9 2.8  1.1 3.2  0.8 NS Clinical pregnancy/transfer 57 57 63 NS % Fertilization 65 50 72 NS % Implantation 22 39 30 NS Miscarriages 1 0 2 NS Technique Hyper- stimulation Pregnancy Rate(PR)/ Cycle PR/embryo transfer(ET) Miscarriage Rate Ongoing PR/ET Patients with cryo Avg # cryo Cut-Back (n ¼ 8) 0/8 (0%) 8/8 (100%)* 8/8 (100%) 0/8 (0%) 8/8 (100%) 7/8 (88%) 6 Static Dose (n ¼ 14) 5/14 (36%) 8/14 (57%)* 8/9 (89%) 2/8 (25%) 6/9 (67%) 11/14 (79%) 8 Mean  SD. CONCLUSIONS: The fixed protocol with GnRH antagonists is an effi- cient and effective stimulation protocol for IVF in an oocyte donation pro- gram. Future larger prospective studies are needed to confirm this result. Supported by: None. P-559 EFFICACY AND SAFETY OF COMMERCIALLY AVAILABLE HIGHLY PURIFIED FSH VS. RECOMBINANT FOLLICLE STIMU- LATING HORMONE: A META-ANALYSIS. H. G. Al-Inany, A. M. A- bou-Setta, R. T. Mansour, G. I. Serour, M. A. Aboulghar. The Egyptian IVF-ET Center, El-Maadi, Cairo, Egypt; Private Clinic, Pyramids, Giza, Egypt. OBJECTIVE: Previous systematic reviews comparing highly purified (HP) follicle-stimulating hormone (FSH) with recombinant (r) FSH in IVF/ICSI cycles have demonstrated conflicting results. This may be related to different patient populations, study quality and individual drugs compared. DESIGN: Meta-analysis of randomized trials comparing only currently available HP-FSH vs. rFSH. MATERIALS AND METHODS: In order to assess the clinical profile and efficacy of these drugs meticulous computerized searches (last performed April 2007) were conducted. Furthermore, the reference lists of all known primary studies, review articles, citation lists of relevant publications, and included studies were examined to identify additional relevant citations. Finally, ongoing and unpublished trials were sought by contacting experts in the field and com- mercial entities. Primary outcome measures were the ongoing pregnancy/live birth rate and the rate of ovarian hyperstimulation syndrome. Secondary out- comes were the clinical pregnancy, multiple pregnancy and miscarriage rates, in addition to cycle characteristics (e.g. treatment duration, number of am- poules, E2 on day of HCG, number of oocytes retrieved). RESULTS: The primary outcomes, ongoing pregnancy/live-birth (O.R ¼ 1.24, 95% CI ¼ 0.76 to 2.01) and OHSS rates (O.R ¼ 6.70, 95%CI ¼ 0.4 to 7.17) were not significantly different between the two groups. As for the second- ary outcomes, there was significantly less treatment days (WMD ¼0.43, 95% CI ¼0.72 to 0.14) and total dose of FSH (IU) (WMD ¼1172.29, 95% CI ¼1445.13 to 899.45) in the HP-FSH group compared with the rFSH group. Even so, the number of oocytes retrieved (WMD ¼0.46, 95% CI ¼1.05 to 0.13) was not significantly different between the two groups. In addition, there were no significant differences with regards the clinical pregnancy rate (O.R ¼ 1.34, 95% CI ¼ 0.96 to 1.88) or the other secondary outcomes. Figure 1. S292 Abstracts Vol. 88, Suppl 1, September 2007