Background: Several markers were studied previously in order to predict the pregnancy outcome of assisted reproductive techniques; however, serum beta human chorionic gonadotropin was found to be the most predictive marker.
Aims: To evaluate the value of serum beta human chorionic gonadotropin levels in discriminating biochemical and clinical pregnancies 12 days after embryo transfer, while determining the factors predicting ongoing pregnancy was established as the secondary aim.
Study Design: Retrospective cross-sectional study.
Methods: A total of 445 pregnant cycles were retrospectively analysed in 2359 embryo transfer cycles. Patients were divided into two groups according to the outcome of pregnancy: biochemical and clinical.
Results: The cut-off value of beta human chorionic gonadotropin levels on day 12 in predicting clinical pregnancies was 86.8 IU/mL with 65.1% sensitivity and 74.7% specificity [CI: 0.76 (0.71-0.81). Receiver operating characteristic curve analysis revealed different cut-off values for embryo transfer days (57 mIU/mL for day 3 embryo transfer CI: 0.59-0.79 and 87 mIU/mL for day 5 embryo transfer, CI: 0.74-0.86). Subgroup analysis of clinical pregnancies revealed a significant difference between ongoing pregnancies and early fetal losses regarding duration of infertility (81.3±54.4 vs. 100.2±62.2 months), serum oestradiol on hCG day (2667.4±1276.4 vs. 2094.6±1260.5 pg/mL), number of transferred embryos (1.9±0.8 vs. 1.5±0.7) and the prevalence of diminished ovarian reserve as an indication (2.3% vs 12.2%).
Conclusion: Beta human chorionic gonadotropin levels on day 12 following embryo transfer provide an important parameter for the prediction of clinical pregnancy; however, other stimulation parameters are indicated in the prediction of ongoing pregnancies.