Background: This study aimed to determine if the gonadotropin releasing hormone (GnRH) antagonist protocol is optimal for expected poor ovarian responders with tubal factor undergoing in vitro fertilization-embryo transfer (IVF-ET). Methods: A total of 341 IVF-ET cycles were retrospectively identified. The following inclusion criteria were applied: age ≥ 40 years and patients with tubal factors. The cycles were divided into two groups: a GnRH antagonist group (157 cycles) and a GnRH agonist group (184 cycles). Results: The duration of stimulation and the total doses of gonadotropin in the GnRH agonist group were significantly more than those in the GnRH antagonist group (P < 0.05). There were significant differences in LH and P values on the hCG measurement days between the two groups (0.91 ± 1.17 vs. 4.82 ± 4.69 U/L and 0.69 ± 0.42 vs. 1.03 ± 0.50 ng/mL, P < 0.05). The implantation rate of the GnRH antagonist group was 12.24%, which was slightly higher than that of the GnRH agonist group (10.10%, P = 0.437). The clinical pregnancy rate of the two groups showed no statistical differences (23.36% vs. 23.03%, P = 1.000). Conclusion: For expected poor ovarian responders, the GnRH antagonist protocol was, to some extent, superior to the GnRH agonist protocol in terms of the implantation and clinical pregnancy rates.