Objective: To evaluate and compare the efficacy of a microdose GnRH agonist flare (MF) and a GnRH antagonist/letrozole (A/L) protocols in poor responders undergoing in vitro fertilization (IVF). Materials and Methods: Ninety-four poor responder patients were randomized to an ovarian stimulation protocol with either a MF or a letrozole and high dose FSH/hMG and flexible GnRH antagonist protocol. Results: There were no significant difference in mean age, body mass index (BMI), basal serum FSH and estradiol (E2) levels, duration of infertility, distribution of etiology of infertility and the number of previous failed IVF cycles. The days of stimulation, mean gonadotropin dose, the number of mature follicles and oocytes retrieved and metaphase ΙΙ oocytes retrieved, serum E2 level on the day of hCG administration and the percentage of top and good quality embryos were significantly higher in the MF group. The endometrial thickness, the fertilization rate and the number of embryos transferred were similar in both groups. The implantation and clinical pregnancy rates were higher in the MF group and the total cancellation rate were higher in the A/L group, but these findings were not statistically significant. Conclusion: The addition of letrozole to GnRH antagonist for poor responders does not improve outcome of assisted reproductive technology (ART) cycles. The MF protocol remains the most appropriate protocol in poor responders.