Clinical or biochemical pregnancies after ICSI cycles: threshold values of beta-human chorionic gonadotropin (β-HCG) helping in better prediction and counselling

Abstract

Human chorionic gonadotropin is secreted by syncytial trophoblast and appears in maternal circulation approximately 6-8 days after fertilization. The maternal plasma levels of hCG continue to double about every 2 days and peaks at 6-8 weeks after conception. Thus, the β-hCG levels can be used to assist and guide the clinician in better management and monitoring of post-IVF pregnancies in conjunction with a transvaginal sonography. (1,2) Low values after transfer require to repeat the 48 hours later to confirm pregnancy. Clinical pregnancy (defined by sonography when a gestational sac with a heartbeat is observed) is a good predictor for evolutionary pregnancy. Determining threshold values of β-hCG that are predictive of clinical pregnancy (CP) is important for the future management of the patient. Also, to identify values below which pregnancy is not achieved (staying as biochemical pregnancy (BP)), could also help to manage levels of anxiety caused by the assisted reproduction treatment (3,4,5,6,7). The aim of this study was to determine β-hCG thresholds predictive of a clinical pregnancy (CP) after the transfer of embryos in cleavage (D2/D3) or blastocyst (D5) stage. Also, to determine if there is any minimum value below which it is not possible to observe CP. single center retrospective cohort study. 387 women that underwent fresh ICSI cycles were included. The determination of β-hCG was performed by ECLIA (ELECSYS-ROCHE) fourteen days after an embryo transfer (ET). The sensitivity of the assay was 2 IU/L. All patients underwent transvaginal ultrasound at 6 weeks after ET. Considering the day in which the embryos was transferred to the uterus of the mother, the groups were classified in D+2 , in D+3 and D+5. In order to determine the threshold of β-hCG that could predict a CP, different COR (Operative Feature of the Receiver) curves were made for each group (D+2, D+3 and D+5). In addition, sensitivity and specificity levels (S/S) for each of these values was determined. Similar ROC curve analyses were performed to determine β-hCG threshold predictive of CP. The proposed optimal thresholds predictive for CP were: D2: 183 IU/L [ROC: D2=0,89 (J= 0,68), S/S 80,6%, 87,5%]; D3: 256 IU/L [ROC D3= 0,89 (J= 0,68) S/S 83,2,6%, 85,2%] and D5: 379 IU/L [ROC= 0,93 (J=0,85) S/S 92,6%, 92,9%]. Under 60 IU/L none of the pregnancies evolved to CP. The data from this study provide a set of values for initial β-hCG levels that are dependent on the day of ET and are a reliable and highly predictive tool for CP outcomes. Above those values, repeating the β-hCG at 48 hrs is unjustified.. We propose each institution must find a cut-off value below which BP is defined, helping in better patient counseling. Further studies are needed to confirm our findings, in order to establish valid and clinically informative β-hCG thresholds.