| Objective:The purpose of this study was to explore the factors that influence pregnancy in patients with AEH and EEC and to establish a Nomogram prediction model.Methods:A total of 175 women who attempted to conceive and had achieved complete remission(CR)after fertility sparing therapy were retrospectively analyzed.They were divided into pregnancy group and non-pregnancy group,patient pregnancy rate distributions were calculated using Kaplan-Meier analysis combined with log-rank test,and univariate and multivariate analysis were performed using Cox proportional hazards regression models to determine pregnancy-related factors,the prediction model of Nomogram is established by using R language based on pregnancy-related factors.Results:A total of 73 of the 175 patients had 85 pregnancies,including 59 live births,with a mean gestation time of 33.18±4.58 months;The proportion of patients with BMI>=28kg/m~2,irregular menstrual cycle,complete remission time>6 months,relapse and natural conception in non-pregnancy group was higher than that in pregnancy group[55 cases(75.3%)vs 18 cases(24.7%),79 cases(77.5%)VS 38cases(52.1%),46 cases(46.0%)vs 22 cases(30.0%),39 cases(38.2%)vs 17 cases(23.3%),66 cases(64.7%)vs 24 cases(32.9%)]and the difference was statistically significant(P<0.05),there was no significant difference in age,pathological type,treatment and abnormal glucose and lipid metabolism between pregnancy group and non-pregnancy group(P>0.0 5).Multivariate Cox regression analysis showed that the independent risk factors of pregnancy were recurrence[HR=0.492,95%CI(0.284,0.850),P=0.011],type of pregnancy[HR=2.493,95%CI(1.524,4.078),P=0.001],BMI[HR=0.541,95%CI(0.314.0.931),P=0.002],regularity of menstrual cycle HR=0.512,95%CI(0.322,0.816);Although in univariate analysis,there was not a significant difference in AEH and maintenance therapy after CR(P>0.05),their relative risks were 1.319[95%CI(0.710,2.450),P=0.396],1.578[95%(0.932,2.703),P=0.089],in other words,the pregnancy rate was increased by31.9%and 57.8%;furthermore,Age>=32 years HR was 0.608[95%CI(0.623,1.670),P=0.04],in univariate analysis,but HR was 0.646[95%(0.399,1.004),P=0.07]in multivariate analysis,the pregnancy rate decreased by 33.4%when age>=32 years.Therefore,in addition to taking into account independent risk factors:whether recurrence、type of pregnancy,menstrual regularity and BMI,whether the patient was older than 32 years old,the type of pathological tissue and whether the patient received maintenance therapy after CR were also included,the Nomogram was drawn by R language environment according to the relative risk of each factor of Cox proportional hazard model,and the C-index of the model was 0.712(95%CI:0.699-0.726).Conclusion:The center’s pregnancy rate was 41.7%,the median pregnancy time was33 months(interquartile range:33 months),which proved that EEC and AEH were feasible to preserve fertility;Patients who use assisted reproductive technology have higher pregnancy rates;Keeping a BMI<28kg/m~2during the course of treatment is associated with a higher likelihood of pregnancy.Relapse is an independent risk factor of pregnancy,so in addition to pregnancy as soon as possible after complete remission,maintenance treatment such as oral progesterone,compound oral contraceptives or wearing LNG-IUD placement should be taken;A prediction model with good accuracy and discrimination degree is established,by means of pregnancy mode,age,recurrence status,medication after complete remission,BMI,regularity of menstrual cycle,and histopathologic type;the nomograms were drawn,it can directly and accurately predict the probability of pregnancy in the second and third year after CR,but it is limited in predicting the probability of pregnancy in the first,fourth year after CR,the prediction model has been verified internally with a C-index of0.71 2(95%CI:0.699-0.726),which has a good prediction ability,but has not been verified externally,whether the prediction model based on patients in our center can be extended to a large scale needs to be validated by a multicenter study with a large sample. |
PDF Full Text Request