Feng Xiong1, Guangui Li1, Qing Sun1, Peilin Chen1, Zhuran Wang1, Caiyun Wan1, Zhihong Yao1, Huixian Zhong1, Yong Zeng2. 1. Shenzhen Key Laboratory of Reproductive Immunology for Peri-implantation, Shenzhen Zhongshan Institute for Reproduction and Genetics, Fertility Center, Shenzhen Zhongshan Urology Hospital, Shenzhen 518045, Guangdong, People's Republic of China. 2. Shenzhen Key Laboratory of Reproductive Immunology for Peri-implantation, Shenzhen Zhongshan Institute for Reproduction and Genetics, Fertility Center, Shenzhen Zhongshan Urology Hospital, Shenzhen 518045, Guangdong, People's Republic of China. Electronic address: zengyong1966@sina.com.
Abstract
RESEARCH QUESTION: Do pregnancy, obstetric and perinatal outcomes differ according to initial maternal serum human chorionic gonadotrophin (HCG) level measured on day 11 after single blastocyst transfer? DESIGN: Vitrified-warmed single blastocyst transfer cycles (n = 640) were collected between 1 January 2013 and 30 April 2017 with positive HCG values and retrospectively analysed by receiver operating characteristic curves to predict clinical pregnancy, ongoing pregnancy and delivery. Cycles were divided into a low HCG group (n = 155) and high HCG group (n = 485) based on cut-off value of live birth prediction. Cycles in the HCG group were subdivided into a low-high subgroup (n = 162), medium-high subgroup (n = 162) and high-high subgroup (n = 161) based on tertile points. Pregnancy rates and obstetric and perinatal outcomes were compared. RESULTS: The area under curves for clinical pregnancy, ongoing pregnancy and live birth prediction were 0.95, 0.81 and 0.79, respectively; corresponding cut-off values were 152.2 IU/l, 211.9 IU/l and 211.9 IU/l; HCG less than 211.9 IU/l indicated an extremely low clinical pregnancy rate (34.84%), a high early miscarriage rate (61.11%) and a low live birth rate (12.26%). Rates of gestational diabetes mellitus (GDM) (P = 0.007) and female neonates (P = 0.001) were significantly higher in the LHG group compared with the HHG group; no significant differences were observed in the low versus high HCG group overall. CONCLUSIONS: Lower initial maternal serum HCG levels indicated poorer clinical outcomes. Within the high HCG group, a lower initial maternal HCG level was found to be associated with GDM occurrence and proportion of female neonates.
RESEARCH QUESTION: Do pregnancy, obstetric and perinatal outcomes differ according to initial maternal serum human chorionic gonadotrophin (HCG) level measured on day 11 after single blastocyst transfer? DESIGN: Vitrified-warmed single blastocyst transfer cycles (n = 640) were collected between 1 January 2013 and 30 April 2017 with positive HCG values and retrospectively analysed by receiver operating characteristic curves to predict clinical pregnancy, ongoing pregnancy and delivery. Cycles were divided into a low HCG group (n = 155) and high HCG group (n = 485) based on cut-off value of live birth prediction. Cycles in the HCG group were subdivided into a low-high subgroup (n = 162), medium-high subgroup (n = 162) and high-high subgroup (n = 161) based on tertile points. Pregnancy rates and obstetric and perinatal outcomes were compared. RESULTS: The area under curves for clinical pregnancy, ongoing pregnancy and live birth prediction were 0.95, 0.81 and 0.79, respectively; corresponding cut-off values were 152.2 IU/l, 211.9 IU/l and 211.9 IU/l; HCG less than 211.9 IU/l indicated an extremely low clinical pregnancy rate (34.84%), a high early miscarriage rate (61.11%) and a low live birth rate (12.26%). Rates of gestational diabetes mellitus (GDM) (P = 0.007) and female neonates (P = 0.001) were significantly higher in the LHG group compared with the HHG group; no significant differences were observed in the low versus high HCG group overall. CONCLUSIONS: Lower initial maternal serum HCG levels indicated poorer clinical outcomes. Within the high HCG group, a lower initial maternal HCG level was found to be associated with GDM occurrence and proportion of female neonates.