| Infertility has become one of the serious problems affecting social and marital harmony.Assisted reproductive technology(ART)is the main treatment to solve the problem of infertility.At present,the clinical pregnancy rate following ART(including In vitro fertilization,IVF/Intracytoplasmic sperm injection,ICSI)is about40%~50%and the live birth rate is about 30%.Therefore,10%to 20%of pregnancy loss following ART treatment is considered to be related with the basic physiological characteristics of the infertility couples,basic diseases of the women,and medical interventions during ART treatment,such as excessive estrogen levels,multiple pregnancies,and abortion,stillbirth related to perinatal complications in women after ART pregnancy.The maternal and neonatal safety during the perinatal period has been the focus of ART.This research aimed at evaluating the differences of the perinatal complications and neonatal outcomes between the ART pregnancy and the spontaneous pregnancy,and investigating the effects of ART on perinatal outcomes.A healthy live baby is the ultimate goal of ART,and it is of great clinical significance to explore the influencing factors of live birth outcome.Combining the basic physiological characteristics of infertile couples,such as age,duration of infertility,body mass index(BMI)and so on with the key protocols in controlled ovarian stimulation(COS)and embryo transfer(ET)to screen the influencing factors of the live birth outcome and establish a risk assessment model is a new area of infertility treatment.On the one hand,this model can help doctors tailor treatment protocols,and provide guidance in the choice of the ART treatment;On the other hand,the risk assessment model for live birth outcome is developed to improve counseling of the infertile couples,and make the basic understanding of the final live birth outcome following ART,so as to make the final decisions to avoid blind treatment.It can also reduce the pain of the patient during the treatment physically and psychologically,reduce the cost of treatment,and avoid or reduce the short-term and long-term risks of ART.Therefore,this study was divided into two parts:the first part concerned about the effects of ART on perinatal outcomes,and the second part concerned about the screening of the influencing factors of the live birth outcome following ART and establishment of the risk assessment model of live birth outcome.Part oneEffects of ART on perinatal outcomesObjective:To assess the differences of the perinatal complications and neonatal outcomes between the ART pregnancy and the spontaneous pregnancy,and to investigate the effects of ART on perinatal outcomes.Methods:From January 2012 to December 2015,345 ART primiparas received ART protocol in Tianjin Medical University General Hospital and delivered in the obstetrics department of the Tianjin Medical University General Hospital,Tianjin Hongqiao Hospital,Tianjin Beichen District Chinese Medicine Hospital and Tianjin Binhai Area Tanggu Maternity Hospital were selected as the ART group.21159spontaneous pregnancy primiparas delivered in the four hospitals during the same period were selected as the control group.Perinatal complications such as hypertensive disorders complicating pregnancy(HDCP),gestational diabetes mellitus(GDM),placenta previa,placenta increta,placental abruption,premature rupture of membranes(PROM),premature delivery,fetal distress,oligohydramnios,postpartum hemorrhage(PPH),hysterectomy,mode of delivery and neonatal outcomes such as neonatal sex ratio,neonatal average birth weight,and the incidence of the neonatal asphyxia,low birth weight infants,macrosomia,and neonatal transfer to NICU were compared between the two groups.Results:(1)Among the single birth delivery women,there were no differences with regard to the HDCP,placental increta,placental abruption,fetal distress,oligohydramnios and postpartum hemorrhage between the ART group and the spontaneous pregnancy group(P>0.05);The risk of GDM,placenta previa,premature delivery,and cesarean delivery rate in ART group were significantly higher than in spontaneous pregnancy group(P<0.01).(2)Among the single birth delivery women,there were no differences with regard to the neonatal sex ratio,neonatal average birth weight and the incidence of neonatal asphyxia when compared between the two groups(P>0.05);the risk of low birth weight infants,macrosomia and neonatal transfer to NICU rate in ART group were significantly higher than in spontaneous pregnancy group(P<0.05).(3)Among the twin birth delivery women,there were no differences with regard to the incidence HDCP,placental increta,placental abruption,PROM,premature delivery,fetal distress,oligohydramnios,postpartum hemorrhage and the cesarean delivery rate between the two groups(P>0.05);The risk of GDM,placenta previa and the hysterectomy due to postpartum hemorrhage rate in ART group were significantly higher than in spontaneous pregnancy group(P<0.05).(4)Among the twin birth delivery women,the rate of neonatal asphyxia in ART group was significantly lower than in spontaneous pregnancy group(P<0.05);the neonatal average birth weight and the rate of low birth weight infants in ART group was significantly higher than in spontaneous pregnancy group(P<0.05);However,there were no differences with regard to the neonatal sex ratio and neonatal transfer to NICU when compared between the two groups(P>0.05).(5)In ART pregnancy,the risk of HDCP,premature delivery,and postpartum hemorrhage were significantly increased in the twin birth delivery group than in the single birth delivery group(P<0.01).Conclusions:(1)Compared with spontaneous pregnancy group,the adverse perinatal outcomes of GDM and placenta previa were significantly increased in ART group.(2)Pregnancy following ART had significantly increased the adverse neonatal outcomes when compared with the spontaneous pregnancy.Part two Establishment of risk assessment model of live birth outcome in ARTObjective:To assess the risk factors of live birth outcome in ART,and establish the risk assessment model of live birth with fresh embryo embryo transfer(Fresh-ET)cycles and frozen-thawed embryo transfer(FET)cycles independently.Methods:(1)A retrospective study was conducted in 1005 Fresh-ET cycles in the reproductive medicine center,Tianjin Medical University General Hospital from January 2012 to December 2015.The basic characteristics of infertile couples(including female age,male age,type of the infertility,delivery history,causes of infertility,the number of pregnancy,duration of infertility,BMI,basal FSH level,basal LH level,FSH/LH ratio and antral follicles count),the factors in COS procedures(including COS protocol,fertilization methods,the number of oocytes retrieved cycle,the time of using Gn days and Gn dosages,the number of retrieved oocytes,serum estradiol level of HCG trigger day)and the factors during the embryo transfer procedures(the number of ET times,the number of embryo transferred,the stage of embryos transferred,endometrial thickness,endometrial types and endometrial blood flow measured by color Doppler)correlating with the outcome of live birth were analyzed.(2)A retrospective study was conducted in 1031 FET cycles.The basic characteristics of infertile couples(including female age,male age,type of the infertility,delivery history,causes of infertility,the number of pregnancy,duration of infertility,BMI,basal FSH level,basal LH level,FSH/LH ratio and antral follicles count)and the factors during the ET procedures(the number of ET times,the number of embryo transferred,the stage of embryos transferred,endometrial thickness,endometrial types and endometrial blood flow measured by color Doppler)correlating with the outcome of live birth were analyzed.(3)Data were analyzed by SPSS22.0 software.Enumeration data were analyzed by Independent Samples T Tests.All P-values were two-sided,and P<0.05 was considered to be statistically significant in all tests.Comparison of the rates between groups was done byc~2 test.Logistic multivariate regression analysis was used for screening the influencing factors of live birth outcomes,and the cut-off value of the independent influencing factor was determined by the receiver operating characteristic curve(ROC)and Youden index.The factors proved to affect the live birth outcome were used to establish the risk assessment model.From January 2016to June 2016,the risk assessment model was validated internally in the ET cycles from this reproductive medical center to verify the fitness of the model.Results:(1)There were 963 clinical pregnancy cycles in 2036 ET cycles,the total clinical pregnancy rate was 47.3%and the total live birth rate was 37.6%.In 1005Fresh-ET cycles,478 clinical pregnancy cycles and 385 live birth cycles were obtained,the live birth rate was 38.3%;In 1031 FET cycles,485 clinical pregnancy cycles and 380 live birth cycles were obtained,the live birth rate was 36.9%.There were no significant differences in clinical pregnancy rate,live birth rate,single live birth rate,multiple live birth rate,abortion rate,ectopic pregnancy rate,stillbirth rate and induced labor rate between the Fresh-ET cycles and FET cycles(P>0.05).(2)Univariate logistic regression analysis showed that female age,male age,type of the infertility,delivery history,causes of infertility,the number of pregnancies,duration of infertility,basal FSH level,antral follicles count,COS protocol,the number of oocytes retrieved cycle,the use of Gn dosages,the number of retrieved oocytes,the stage of embryos transferred,endometrial thickness and endometrial types would affect the outcome of live birth in Fresh-ET cycles.The influential variables in univariate logistic regression analysis were introduced into logistic multivariate analysis,and the results showed that female age(P=0.000,OR=1.613,95%CI:1.266-2.057),duration of infertility(P=0.022,OR=1.064,95%CI:1.009-1.123)and the number of oocytes retrieved cycle(P=0.030,OR=1.501,95%CI:1.04-2.168)were the independent risk factors affecting the live birth outcome in Fresh-ET cycles;the stage of embryos transferred(P=0.013,OR=0.732,95%CI:0.573-0.935)and endometrial thickness(P=0.000,OR=0.909,95%CI:0.867-0.953)were independent protection factors affecting the live birth outcome in Fresh-ET cycles.According to the cut-off values of the independent influencing factors,the results of live birth in Fresh-ET cycles decreased significantly when the woman was over 40 years old,the number of infertility years was over 3.5 years,and the number of the number of oocytes retrieved cycle was over 2 times;The results of live birth in Fresh-ET cycles increased significantly when the embryo was blastocyst and the thickness of endometrium was more than 11.75 mm on the ET day.Variables in logistic multivariate regression analysis with diagnostic significance for live birth outcome were introduced to the risk assessment model in Fresh-ET cycles.The risk assessment model was as follows:P=1/(1+exp(-y));logit(y)=0.478*female age group+0.062*duration of infertility+0.406*the number of oocytes retrieved cycle-0.312*the stage of embryos transferred-0.095*endometrial thickness+1.042.The risk assessment model was tested by the likelihood ratio test,the result showed that regression equation was statistically significant(P=0.000,c~2=357.313).The ROC curve was used to evaluate the discriminant ability of the model.The area under ROC was 0.655,the standard error=0.008,P=0.000,95%CI:0.620-0.689.Choosing P=0.597 as the cut-off level,the sensitivity of the model was 64.0%,and the specificity of the model was 60.3%.225 Fresh-ET cycles were introduced into the risk assessment model for the internal verification.According to the cut-off level,the theoretical predictive probability was divided into theoretical live births and theoretical non-live births,and it was correlated with the actual live birth.Spearman correlation was made to verify the prediction of the equation model.The result showed that r=0.209,P=0.002,and the correlation was positive which indicated that the model was well fitted.(3)Univariate logistic regression analysis showed that female age,male age,delivery history,causes of infertility,the number of pregnancy,duration of infertility,basal LH level,FSH/LH ratio,antral follicles count,the number of oocytes retrieved cycles,the number of ET times,the number of embryo transferred,the stage of embryos transferred and endometrial thickness would affect the live birth outcome in FET cycles.The influential variables in univariate logistic regression analysis were introduced into logistic multivariate analysis,and the results showed that female age(P=0.000,OR=1.648,95%CI:1.274-2.132)and the number of ET times(P=0.000,OR=1.447,95%CI:1.220-1.717)were the independent risk factors affecting the live birth outcome in FET cycles;the stage of embryos transferred(P=0.000,OR=0.599,95%CI:0.474-0.758),the number of embryo transferred(P=0.005,OR=0.652,95%CI:0.482-0.882)and the endometrial thickness(P=0.000,OR=0.882,95%CI:0.830-0.938)were independent protection factors affecting the live birth outcome in FET cycles.According to the cut-off values of the independent influencing factors,the results of live birth in FET cycles decreased significantly when the woman was over 40 years old and the number of ET times were over 2 times;The results of live birth in FET cycles increased significantly when the embryo was blastocyst,the number of embryo transferred were more than 2 and the thickness of endometrium was more than 9.15 mm on the ET day.Variables in logistic multivariate regression analysis which would affect the live birth were introduced to the risk assessment model in FET cycles.Risk assessment model was as follows:P=1/(1+exp(-y));logit(y)=0.5*female age group-0.512*the numbers of embryo transferred+0.37*the number of ET times-0.428*the numbers of embryo transferred-0.125*endometrial thickness+2.756.The risk assessment model was tested by means of Hosmer-Lemeshow test,and the result showed that regression equation was statistically significant(P=0.000,c~2=98.336).ROC curve was used to evaluate the discriminant ability of the model.The area under ROC was 0.681,the standard error was 0.017,P=0.000,95%CI:0.648-0.714.Choosing P=0.622 as the cut-off level,the sensitivity of the model was 61.2%,and the specificity of the model was 66.8%.231 FET cycles were introduced into the risk assessment model for the internal validation.The result indicated that the model was well fitted.Conclusions:(1)Female age,duration of infertility and the number of oocytes retrieved cycles were the independent risk factors affecting the live birth outcome in Fresh-ET cycles;and the stage of embryos transferred and endometrial thickness were the independent protection factors affecting the live birth outcome in Fresh-ET cycles.The results of live birth in Fresh-ET cycles decreased significantly when the woman was over 40 years old,the number of infertility years was over 3.5 years,and the number of oocytes retrieved cycle was over 2 times;The results of live birth in Fresh-ET cycles increased significantly when the embryo was blastocyst and the thickness of endometrium was more than 11.75 mm on the ET day.(2)Female age and the number of ET times were the independent risk factors affecting the live birth outcome in FET cycles;the stage of embryos transferred,the numbers of embryo transferred and endometrial thickness were the independent protection factors affecting the live birth outcome in FET cycles.The results of live birth in FET cycles decreased significantly when the woman was over 40 years old and the number of ET times was over 2 times;the results of live birth in FET cycles increased significantly when the embryo was blastocyst,the numbers of embryo transferred was more than 2 and the endometrial thickness was more than 9.15 mm on the ET day.(3)This study established the risk assessment model of live birth with fresh embryo embryo transfer(Fresh-ET)cycles and frozen-thawed embryo transfer(FET)cycles independently,and the model fitted well after the internal validation.This model could effectively evaluate the outcome of live birth in ART,and it can provide clinical counseling and further optimize the diagnosis and treatment strategies for the infertile couples. |
PDF Full Text Request