Long GnRH-agonist protocol in an IVF program: Is it appropriate for women with normal FSH levels and high FSH/LH ratios?

OBJECTIVE: To determine whether subjects with an elevated ratio of follicle-stimulating hormone (FSH) to luteinizing hormone (LH) but normal basal FSH levels should be regarded as poor responders to controlled ovarian hyperstimulation. STUDY DESIGN: One hundred twenty-five women undergoing in vitro fertilization (IVF) for the first time were recruited in this retrospective cohort study. Women over 40 years old or having serum basal FSH > 10 mIU/mL were excluded. RESULTS: Various cutoff values for the FSH/LH ratio were chosen, and the ratio demonstrated that pregnancy rates were apparently higher in patients with the long protocol than with the short one if they had an FSH/LH ratio < 3.0 (48.5% vs. 25.8%, P = .034), < 2.5 (53.3% vs. 28.6%, P = .030) or < 2.0 (57.8% vs. 21.7%, P = .005). Pregnancy rates were similar with the long and short protocols in patients with FSH/LH ≤ 3.0 (57.1% vs. 70%, P = .521), FSH/LH ≤ 2.5 (40% vs. 53.8%, P = .435) or FSH/LH ≤ 2.0 (40% vs. 55.6%, P = .281). CONCLUSION: This study failed to demonstrate that FSH/LH was a useful parameter for predicting reproductive outcome in IVF programs and for patient selection for the long or short gonadotropin-releasing hormone agonist protocol.