Initial β-hCG levels and 2-day-later increase rates effectively predict pregnancy outcomes in single blastocyst transfer in frozen-thawed or fresh cycles: A retrospective cohort study

To investigate whether the initial beta-human chorionic gonadotropin (beta-hCG) levels and their rate of increase differ after single fresh and frozen blastocyst transfers, and whether these values effectively predict pregnancy outcomes. This retrospective cohort study was conducted at the Sisli Memorial Hospital, assisted reproductive technology, and Reproductive Genetics Center in Istanbul, Turkey, between January 2016 and January 2022. Three thousand two hundred thirty-eight single blastocyst transfers with positive pregnancy test results were evaluated. Of these, 738 were fresh transfer cycles and 2500 were frozen-thawed embryo transfer (FET) cycles. beta-hCG test results from 9 days after fresh and FET cycles were compared between the groups with biochemical pregnancy, early pregnancy loss, and live birth outcomes. The threshold values were determined for each pregnancy outcome. The rate of increase between the first and second beta-hCG tests performed 2 days apart was determined for each pregnancy outcome. Finally, the listed values were compared between the FET and fresh cycle. Mean baseline beta-hCG levels were significantly higher in FET cycles than in fresh cycles, regardless of pregnancy outcomes (P<.005). Baseline beta-hCG levels were higher in fresh cycles with live births (171.76 +/- 109.64 IU/L) compared to biochemical and clinical pregnancy losses (50.37 +/- 24.31 and 114.86 +/- 72.42, respectively) (P<.001). Live births in FET cycles resulted in higher baseline beta-hCG levels (193.57 +/- 100.38 IU/L) compared to biochemical and clinical pregnancy loss groups (68.41 +/- 51.85 and 149.29 +/- 96.99 IU/L, respectively) (P<.001). The beta-hCG threshold for live birth for fresh cycles was 116.5 IU/L (sensitivity 80%, specificity 70%, positive predictive value 90%, negative predictive value 54%) and 131.5 IU/L for FET cycles (sensitivity 71%, specificity 68%, positive predictive value 87%, negative predictive value 50%). The percentage of the area under the curve for single fresh blastocyst transfers was 0.81 and 0.76 for frozen transfers. The rate of increase in beta-hCG was similar in fresh and FET cycles. Initial beta-hCG levels and 2-day increases are effective parameters for diagnosing pregnancy in fresh and FET cycles. The initial beta-hCG level was significantly higher in the FET cycles than in the fresh cycles. Predicting outcomes earlier helps clinicians to manage and follow high-risk pregnancies.