Objective : The aim of this study was to evaluate the outcome the GnRH antagonist (Cetrotide) short protocol in controlled ovarian hyperstimulation comparing with GnRH agonist long protocol. Materials and Method: From July 2000 to November 2001, 26 patients, 28 cycles were performed in controlled ovarian hyperstimulation by GnRH antagonist and GnRH agonist. GnRH antagonist (Cetrotide) was administered in 12 patients (14 cycles, Group 1) and GnRH agonist (Lucrin, Sub Q, Group 2) in 14 patients (14 cycles). Ovulation induction was performed by hMG (Pergonal) in group 1, and by Combo (Metrodine HP + Pergonal) in group 2. We compared the fertilization rate, good quality embryo, and clinical pregnancy rate between the two groups. Student-t test and Chi-square were used to determine statistical significance. Statistical significance was defined as p<0.05. Results: Ovarian hyperstimulation syndrome did not occurred in which estradiol (E2) level was $3874{pm}809;pg/ml$ and the number of retrieved oocytes was $18.4{pm}2.4$. The number of used gonadotropin ampules was significantly decreased in Group 1 (26.0 vs. 33.1, p<0.04). There were no significant difference in the number of preovulatory oocyte ($10.6{pm}6.9$ vs. $10.0{pm}6.1$), fertilization rate ($74.8{pm}23.4$ vs. $72.2{pm}21.8$), good quality embryo ($58.7{pm}23.6$ vs. $38.7{pm}36.6$), and embryo transfer ($4.3{pm}1.6$ vs. $4.4{pm}1.6$). Although the age of the group 1 was older than the group 2 (34.4 vs. 30.8), there was no significant difference in clinical pregnancy rate (50.0% vs. 57.1%). Conclusions: We suggest that GnRH antagonist was a safe, effective, and alternative method in the controlled ovarian hyperstimulation, especially in PCOD patients who will be develop the ovarian hyperstimulation syndrome.