| BackgroudEndometriosis(EMS)is a disease known to be detrimental to fertility.Women with endometriosis may require in vitro fertilization/intracytoplasmic sperm injection(IVF/ICSI)to achieve a pregnancy.The specific impact of endometriosis on the reproductive outcome of women undergoing IVF/ICSI are areas that require further clarification.Once IVF/ICSI decided,the choice of the controlled ovarian stimulation(COS)protocol remains debatable.Thus,we intended to explore the impact of endometriosis on the outcome of women undergoing IVF/ICSI,and how to choose the COS protocol for endometriosis patients.PartⅠ Analysis of clinical outcome and related factors of IVF/ICSI-ET in patients with endometriosisObjectiveTo investigate the effects of endometriosis on the pregnancy outcome of IVF/ICSI-ET and the associated factors in infertile patients with endometriosis.Methods1 Research objectA retrospective study of infertility patients who diagnosed as endometriosis and underwent IVF/ICSI-ET at our center from January 2013 to December 2016 was performed Inclusion criteria are:1)infertility factor is endometriosis,which confirmed by laparoscopy or laparotomy combined with pathological examination,and no recurrence after operation;2)woman’s actual age is≤40 years old.The control group consisted of patients with laparoscopically confirmed tubal factor unrelated with endometriosisExclusion criteria are:1)Operative history of ovarian lesions other than ovarian endometriosis;2)Uterine malformation(double uteri,mediastinal uterus,etc.),endometrial polyps and other uterine cavity abnormalities confirmed by ultrasonography,hysterosalpingography or hysteroscopy;3)Adenomyosis,uterine fibroids or related surgical history;4)PCOS;5)hydrosalpinx;6)Chromosomal abnormalities.2 Research methodsThe basal conditions,cycle characteristics and pregnancy outcome were collected.SPSS(version 20.0)was used for analysis.To further clarify the factors associated with IVF/ICSI-ET outcomes in endometriosis group,fresh embryo transfer cycles were divided into clinical pregnancy group and non-pregnancy group.Univariate analysis of these two groups was performed.And then analyze with multivariate non-conditional logistic regression.Results1 Comparison of basal conditions between EMS group and control groupThe age was not statistically significant different(P>0.05).BMI in EMS group was significantly smaller than control group(P<0.05),while infertility years,odds of primary infertility in EMS group was significantly greater than those of control group(P<0.05).2 Comparison of COS and laboratory conditions between EMS group and control groupComparing COS conditions of the two groups,Gn days and dose in EMS group were significantly higher than those in control group(P<0.05);level of E2 and LH on HCG day,and number of mature follicles in EMS group were significantly lower than those in control group(P<0.05).There were not statistically significances in level of P,level of E2 per mature follicle and endometrial thickness on HCG day(P>0.05).Comparing laboratory conditions of the two groups,there were not statistically significances in number of oocytes retrieved,rate of normal fertilization oocytes and cleavage rate(P>0.05),while number of metaphase Ⅱ(MⅡ)oocytes,number of normal fertilization oocytes,number of cleavage,number of good-quality embryos,rate of MⅡoocytes,rate of good-quality embryos in EMS group were all significantly smaller than control group(P<0.05).3 Comparison of clinical outcome of fresh embryo transfer cycles between EMS group and control groupThere were 185 fresh embryo transfer cycles in EMS group,while 184 cycles in control group.Embryo implantation rate,clinical pregnancy rate,live birth rate in EMS group were smaller than control group,while early abortion rate in EMS group was larger than control group.However,the differences of clinical parameters above were not statistically significant.4 Comparison of clinical parameters between clinical pregnancy group and non-pregnancy groupComparing basal conditions of the two groups,age in clinical pregnancy group is significantly smaller than non-pregnancy group,while AFC is significantly larger than non-pregnancy group(P<0.05).No difference was found in BMI,infertility duration,type of infertility,basal FSH,stage and type of EMS,duration from endometriosis surgery to oocytes retrieval,male factor between these tow groups(P>0.05).Comparing COS and laboratory conditions of the two groups,Gn dose in pregnancy group is significantly smaller than non-pregnancy group,while endometrial thickness on HCG day is significantly larger than non-pregnancy group(P<0.05).No difference was found in Gn days,COS ptotocals,insemination method,number of embryo transferred between these tow groups(P>0.05).5 Multivariate logistic regression analysis on pregnancy outcomes of fresh embryos transferred cycles in EMS GroupSelect clinical pregnancy as the dependent variable,P<0.05 factors as independent variables(age,AFC,Gn dose,endometrial thickness on HCG day),and analyze with multivariate non-conditional logistic regression.One factor,age,was selected into the final regression equation(OR=0.876,P=0.001).However,AFC,Gn dose and endometrial thickness on HCG day were not selected into the final regression equation.ConclusionThe adverse effects of endometriosis on the outcome of IVF/ICSI-ET were mainly due to diminished ovarian reserve and decreased quality of oocytes and embryos,which result in greater Gn dose,fewer oocytes retrieved,fewer mature oocytes,and fewer high-quality embryos.Pregnancy outcomes in endometriosis patients were similar to those of infertility patients due to tubal factors.Part Ⅱ Comparison of GnRH-ant protocol and GnRH-a long protocol for IVF/ICSI-ET in patients with endometriosisObjectiveComparing the influence of GnRH-ant protocol and GnRH-a long protocol on clinical outcomes of IVF/ICSI-ET in endometriosis patients after propensity score matching.Methods1 Research objectWe retrospectively analyzed data of infertility patients with endometriosis who underwent IVF/ICSI-ET treatment from January 2013 to December 2016 in our center.Inclusion criteria were as follows:1)infertility factors of the woman are endometriosis,which confirmed by laparoscopy or laparotomy combined with pathological examination,and no recurrence after operation;2)woman’s actual age is<40 years old.3)controlled ovarian stimulation protocol was GnRH-ant protocol or GnRH-a long protocol.Exclusion criteria were the same as Part Ⅰ.2 Research methodsThe basal conditions,cycle characteristics and pregnancy outcome were collected.SPSS(version 20.0)was used for analysis.We use propensity score matching to make groups comparable.We choose basal variables that were thought to be related to the outcomes for propensity score matching including female age,BMI,infertility duration,AFC,basal FSH,stage and type of EMS,and male factor.Subsequent analyses were comparisons of groups after propensity score matching.Results1 Results of propensity score matchingIn order to compare clinical outcomes of endometriosis patients using different protocols,we choose basal variables that were thought to be related to the outcomes for propensity score matching,so as to reduce bias due to differences in baseline data.After propensity score matching,these variables were all comparable between the two groups(P>0.05).2 Comparison of COS and laboratory conditions between GnRH antagonist protocol group and GnRH-a long protocol groupComparing COS conditions of the two groups,level of LH on HCG day in antagonist group was significantly higher than those in long protocol group,while endometrial thickness was significantly smaller than those in long protocol group,both of the differences were statistically significant(P<0.05);There were not statistically significances in Gn days,Gn doses,level of P,E2 and number of mature follicles on HCG day(P>0.05).Comparing laboratory conditions of the two groups,number of oocytes retrieved,mature oocytes,normal fertilization oocytes,cleavage,good-quality embryos,rate of mature oocytes and normal fertilization oocytes,cleavage rate,and rate of good-quality embryos were not statistically significant(P>0.05).3 Analysis on pregnancy outcomes of fresh embryos transferred cycles between GnRH-ant protocol group and GnRH-a long protocol groupThere were 53 fresh embryo transfer cycles in GnRH-ant protocol group,while 59 cycles in GnRH-a long protocol group.Embryo implantation rate,clinical pregnancy rate,early abortion rate and live birth rate in GnRH-ant protocol group were all lower than GnRH-a long protocol group,but not statistically significant(P>0.05).4 Analysis on pregnancy outcomes of FET cycles between GnRH-ant protocol group and GnRH-a long protocol groupThere were 40 FET cycles in GnRH-ant protocol group,while 44 cycles in GnRH-a long protocol group.Endometrial thickness on embryo transferred day,number of transferred embryos and live birth rate in GnRH-ant were smaller than GnRH-a long protocol group,while embryo implantation rate,clinical pregnancy rate and early abortion rate in GnRH-ant protocol group were higher than GnRH-a long protocol group.However,the differences of clinical parameters above were not statistically significant(P>0.05).ConclusionGnRH-ant protocol is an effective,convenient,and friendly COS protocol for patients with endometriosis.However,taking into account the potential adverse effects of antagonist protocol on endometrial receptivity,it is suggested that performing embryo transfer in freeze-all cycles after antagonists protocol may achieve clinical outcomes similar to those of GnRH-a long protocol.Part Ⅲ Effect of GnRH-ant protocols,GnRH-a long protocols and GnRH-a ultra long protocols on clinical outcomes of IVF/ICSI-ET in patients with endometriosis:systematic review and meta-analysisObjectiveMeta-analysis was used to compare the effects of three COS protocols for endometriosis patients,in order to provide guidance for the selection of COS protocols by evidence-based medicine.Methods1 Inclusion and exclusion criteria1.1 Type of studiesClinical studies focusing on effect of different COS protocols on clinical outcomes of IVF/ICSI-ET in patients with endometriosis from January 2000 to January 2018 were included.Study design included cohort study,case-control study(retrospective or prospective),and randomized control trial(RCT).The literature language was limited to Chinese and English.1.2 Type of participantsInfertility patients who diagnosed as endometriosis and underwent IVF/ICSI-ET Endometriosis was confirmed by imaging or laparoscopy or laparotomy combined with pathological examination.1.3 Type of interventionsInfertility patients with endometriosis underwent IVF/ICSI-ET treatment and used the following protocol for COS:1)GnRH-ant protocol(fixed or flexible protocol),with or without oral contraceptive(OC)pretreatment;2)GnRH-a ultra long protocol;3)Short-acting/long-acting GnRH-a long protocol.1.4 Exclusion criteriaStudies were excluded if:l)Comparation with other COS protocols;2)Comparation with non-endometriosis patients;3)COS protocols in the studies were unclear;4)The studies were reviews,comments and systematic reviews;5)The studies with republished data and selected only the most recently published or largest sample;6)animal studies and non-clinical studies.2 Literature searching and quality evaluationSearch for literature is mainly based on electronic searches.Manual search,reference search,and grey literature search are supplemented.RCT were scored using the Cochrane Collaboration’s biased risk assessment tool and non-RCT were scored using the Newcastle-Ottawa Scale(NOS)scale.3 Meta-analysisMeta-analysis was performed by Revman 5.3 software.I2 statistic test was performed to examine the heterogeneity.When ten studies were included in the meta-analysis or more,the publication bias was tested using a funnel plot and Begg’s test.All statistics were two-tailed test and P<0.05 was considered statistically significant.Results1 Literature searching resultsAccording to our inclusion and exclusion criteria,1227 articles related to the study were initially retrieved.By browsing the titles and abstracts,1203 articles that obviously did not meet the inclusion and exclusion criteria were eliminated.After reading the full text carefully,11 studies were excluded.Finally,13 studies met our inclusion criteria and were included in meta-analysis.2 Description of studies includedOf the 13 studies included,4 were RCT and the remaining 9 were non-RCT.1 of the 9 studies were prospective designs,while 8 studies were retrospective designs.4 studies compared the GnRH-ant protocol and the GnRH-a long protocol,2 of which GnRH-ant protocol were pretreated with oral contraceptives.9 studies compared GnRH-a ultral long protocol and long protocol.GnRH-a dosage in ultral long protocol was ranging from 1.25 to 3.75 mg,administrating 2 to 6 cycles continuously.3 Quality assessment of studiesAll four RCTs were highly biased due to lack of intention to treat analysis,and all other biases were low risk.All non-randomized studies were awarded with 8-9 star.Therefore,the included studies were high-quality overall.4 Meta-analysis result4.1 Analysis on clinical outcomes of IVF/ICSI-ET between GnRH-ant protocol and GnRH-a long protocolThis meta-analysis included five studies comparing outcomes of IVF/ICSI-ET using GnRH-ant protocol versus GnRH-a long protocol in patients with endometriosis.The results showed that comparing with GnRH-a long protocol group,Gn days,number of oocytes retrieved,number of mature oocytes in antagonist protocol group were significantly smaller(P<0.05),while no differences were found in Gn doses,clinical pregnancy rate,early abortion rate and live birth rate(P>0.05).4.2 Analysis on clinical outcomes of IVF/ICSI-ET between GnRH-a ultral long protocol group and GnRH-a long protocol groupThis meta-analysis included nine studies comparing outcomes of IVF/ICSI-ET using GnRH-a ultral long protocol versus long protocol in patients with different stages of endometriosis.The results showed that comparing with GnRH-a long protocol,Gn doses,Gn days in ultral-long protocol were significantly larger,but clinical pregnancy rate and embryo implantation rate were significantly higher(P<0.05).No differences were found in number of oocytes retrieved,early abortion rate and live birth rate(P>0.05).According to the meta-analysis of subgroups of different ASRM stage,in stage Ⅰ-Ⅱ endometriosis patients,Gn doses and implantation rate were significantly higher,number of oocytes retrieved was significantly fewer in ultral long protocol than in long protocol(P<0.05),no difference was found in Gn days and clinical pregnancy rate(P>0.05).In stage Ⅲ-Ⅳ endometriosis patients,clinical pregnancy rate was significantly higher in ultral long protocol(P<0.05),while no differences were found in Gn doses,Gn days,number of oocytes retrieved and implantation rate(P>0.05).ConclusionThe effect of GnRH-ant protocol is similar to GnRH-a long protocol in endometriosis patients.GnRH-a ultra long protocol could enhance the pregnancy rate,although more Gn doses and days are needed.For stage Ⅰ-Ⅱ endometriosis patients,ultra long protocol could not improve the pregnancy rate,and cost more Gn. |
PDF Full Text Request