The effect of growth hormone on low-prognosis patients: a retrospective study based on POSEIDON criteria

Abstract

Abstract Objective: To investigate the effect of growth hormone on low-prognosis patients who were diagnosed by POSEIDON criteria. Methods: Poor ovarian reserve patients were included if their AMH <1.2ng/ml, and according to POSEIDON criteria they were further divided into Group 3 and Group 4. Mild stimulation protocol was adopted in all groups. Growth hormone was administered in GH group. Further study on non-first cycles was conducted. Linear regression and logistic regression analysis were carried out to control the confounding factors. Results: A total of 574 cycles (155 in GH group, 419 in Control group) were analyzed. For all the poor ovarian reserve patients, there were no significant differences between GH and control group, in aspects of HMG dosage, duration of HMG, the number of oocytes retrieved, the number of 2PN, the number of embryos, transferrable embryos ,good-quality embryos, clinical pregnancy rate, miscarriage rate and clinical live birth rate. In stratified analysis, the variables of G4 were not significantly different, and in G3, duration of HMG (8.11±1.86vs. 8.80±1.98, P﹤0.05) was significantly different. Further study on non-first cycle patients, a total of 274 cycles (122 in GH group, 152 in Control group) were analyzed, there were no differences between GH group and control group, in aspects of HMG dosage, duration of HMG, the number of oocytes retrieved, the number of 2PN, the number of embryos, transferrable embryos ,good-quality embryos, clinical pregnancy rate, miscarriage rate and clinical live birth rate. In stratified analysis, the variables of G3’ were not significantly different. In G4’, duration of HMG (8.74±2.31 vs. 7.90±2.56, P﹤0.05) was significantly different, the number of oocytes retrieved(8.74±2.31 vs. 7.90±2.56, P﹤0.05), clinical pregnancy rate(22.97% vs. 8.45%, P﹤0.05), and clinical live birth rate(14.86% vs. 4.23%, P﹤0.05) were significantly different between GH and control groups. The outcomes of linear regression and logistic regression analysis were approximately consistent with that of Chi square test. Conclusions: GH co-treatment with the mild stimulation protocol in poor ovarian reserve patients who reached or were older than 35 years old, and failed in at least one previous cycle, could significantly increase the number of oocytes retrieved, clinical pregnancy rate and live birth rate.