LAUSR

RESEARCH QUESTION Do live birth rates (LBR) following modified natural IVF (mnIVF) differ according to serum anti-Müllerian hormone (AMH) concentration? DESIGN Retrospective cohort study including 638 women aged ≤39 years starting their first mnIVF cycle at a university-affiliated private IVF centre. Patients were divided into three groups, by concentration of AMH: ≤0.5 ng/ml (25th percentile), 0.51-2.03 ng/ml (25-75th percentile, reference) and 2.04-6.56 ng/ml (75th percentile). Analyses were stratified by AMH percentile and the age of patients (<35, 35-39 years). Logistic regression assessed the impact of age and AMH percentile on outcomes. LBR was the primary outcome measure. RESULTS LBR per started cycle were comparable across AMH percentiles (11.6%, 12.4% and 17.0% for the 25th, 25-75th and 75th percentile, respectively). No statistically significant difference was found between the three AMH groups with respect to cancellation, successful egg retrieval, embryo transfer, or biochemical and clinical pregnancy rates. Logistic regression analysis did not identify AMH percentile as a significant predictor of live birth. Compared with the reference group, the odds ratios (OR [95% confidence interval, CI]) for live birth in the <25th and >75th AMH percentile groups were 0.97 (0.54-1.76) and 1.41 (0.82-2.41), respectively. The results were the same regardless of age group (<35 years, 35-39 years). CONCLUSIONS Serum AMH cannot be used to predict mnIVF outcomes. Patients in lower/upper AMH percentiles showed pregnancy and LBR comparable to patients with normal AMH.