Ovarian response prediction in controlled ovarian stimulation for IVF using anti-Müllerian hormone in Chinese women: A retrospective cohort study

Abstract

The predictive value of anti-Müllerian hormone (AMH) in Chinese women undergoing in vitro fertilization (IVF) treatment is data deficient. To determine the attributes of AMH in IVF, oocyte yield, cycle cancellation, and pregnancy outcomes were analyzed. All patients initiating their first IVF cycle with gonadotropin-releasing hormone agonist treatment in our center from October 2013 through December 2014 were included, except patients diagnosed with polycystic ovarian syndrome. Serum samples collected prior to IVF treatment were used to determine serum AMH levels. A total of 4017 continuous cycles were analyzed. The AMH level was positively correlated with the number of oocytes retrieved. Overall, AMH was significantly correlated with risk of cycle cancellation, poor ovarian response (POR, 3, or fewer oocytes retrieved) and high response (>15 oocytes), with an area under the curve (AUC) of 0.83, 0.89, and 0.82 respectively. An AMH cutoff of 0.6 ng/mL had a sensitivity of 54.0% and a specificity of 90.0% for the prediction of cycle cancellation, and cutoff of 0.8 ng/mL with a sensitivity of 55.0% and a specificity of 94.0% for the prediction of POR. Compared with AMH >2.0 ng/mL, patients with AMH <0.6 ng/mL had a 53.6-fold increased risk of cancellation (P<0.001), and AMH <0.80 ng/mL were 17.5 times more likely to experience POR (P<0.001). However, AMH was less predictive of pregnancy and live birth, with AUCs of 0.55 and 0.53, respectively. Clinical pregnancy rate, ongoing pregnancy rate, and live birth rate per retrieval according to the AMH level (≤0.40, 0.41-0.60, 0.61-0.80, 0.81-1.00, 1.01-1.50, 1.51-2.00, and >2.00 ng/mL) showed no significant differences. Even with AMH ≤0.4 ng/mL, 50.0% of all the patients achieved pregnancy and 34.8% of patients achieved live birth after transfer. Our results suggested that AMH is a fairly robust metric for the prediction of cycle cancellation and oocyte yield for Chinese women, but it is a relatively poor test for prediction of pregnancy outcomes. Patients with low levels of AMH still can achieve reasonable treatment outcomes and low AMH levels in isolation do not represent an appropriate marker for withholding fertility treatment.