| ObjectiveTo retrospectively analyze factors affecting the clinical pregnancy the pregnancy outcome in the fresh embryo transfer cycle after treatment with the assisted reproductive technology(ART) and to establish forecast model for clinical pregnancy, so as to guide the preparation of individualized treatment protocol and enhance the rate of clinical pregnancy and pregnancy outcome. MethodThe investigators selected patients who received in-vitro fertilization(IVF) or intracytoplasmic sperm injection(ICSI) for embryo transfer with support of ART at Reproductive & Genetic Center of Tangshan Maternal and Child Health Hospital of Hebei Province between Jan. 2010 and June 2015, and altogether 2510 cycles were involved. They were divided into successful clinical pregnancy group(996 cases) and unsuccessful clinical pregnancy group(1514 cases) according to the results of color doppler ultrasound 35 days after the transplantation. According to the pregnancy outcome,the patients were divided into two groups, successful conception group(712 cases) and unsuccessful conception group(284 cases). The single factor analysis was employed to compare the two groups’ general conditions(ages, weight indices, infertility types and years of infertility), complications(hysteromyoma, adenomyosis, and uterine deformity etc), basal endocrine levels, different controlled super-ovulation protocols(long protocol, ultra-long protocol, short protocol, and minimal ovarian stimulation), the use of gonadotropin(Gn)(days and dosage of use) during the treatment, the condition on the day of administration of HCG(endometrial thickness, ovarian size and number of extracted eggs), and transplantation conditions(number of transplanted embryos, and mode of transplantation), so as to find out the factors of statistical significance and relevance. In addition, investigators also made a multiple-factor analysis and established a risk forecast model for global variables of clinical pregnancy and pregnancy outcome to validate the effectiveness of the equation.The SPSS 19.0 statistical package was used for data analysis. The 2 Independent Samples T Tests were employed to analyze the single factors. The comparison between rates of samples was done by R×C contingency table χ2 test. The Logistic regression analysis was adopted for the multiple-factor analysis, and if P is less than 0.05, it means that the difference is of statistical significance. The global variable forecast model was established for factors affecting clinical pregnancy and pregnancy outcome, which used whether clinical pregnancy or pregnancy outcome was successful as dependent variable. The investigators made a non-conditional Logistic regression analysis of the data, and imported factors of diagnostic significance to the step-by-step multiple-factor analysis into the models. The Enter method was adopted to work out the forecast equation of clinical pregnancy and pregnancy outcome. The forecast probability and the data on whether clinical pregnancy and pregnancy outcome was successful were used to draw the ROC curve and to obtain the area under the curve to verify the efficiency of the equation. Results1. Single factor analysis: with respect to the unsuccessful clinical pregnancy group, the female patients’ ages, years of infertility,FSH, FSH/LH and Gn dosage were higher than those in the successful pregnancy group(p<0.05), while the endometrial thickness, the volume of both ovaries, total number of extracted eggs, and the number of extracted eggs from both ovaries on the day of administration of HCG were smaller than those of the successful clinical pregnancy group(p < 0.05). In addition, as for the unsuccessful clinical pregnancy group, the rate of uterine deformity, adenomyosis, hydrosalpinx and polycystic ovarian syndrome were higher than those of the successful pregnancy group, and p<0.05 means that the difference was of statistical significance. A multiple-factor regression analysis was made over the patients’ general conditions, complications, basal endocrine and ovulation promotion, and transplantation-related indices that were derived from the above single factor analysis. the factors of statistical significance to the clinical pregnancy included:age, years of infertility,combined uterine malformations, PCOS, hydrosalpinx,basal FSH,basal FSH/LH,COH,HMG dosage,endometrial thickness on the day of administration of HCG..2.Whether clinical pregnancy was successful was used as the dependent variable and the factors in the above multiple-factor regression analysis which affected the pregnancy were used as the independent variables. The Enter method was adopted to obtain the forecast model(P) for clinical pregnancy probability: P=1/(1+y); y=exp{-(3.538-0.025×age of female patient-0.054×years of infertility-2.479×uterie malformation-0.323×PCOS-0.707×hydrosalpinx-0.020×basal FSH-0.236×FSH/LH-0.055×therapeuticprotocol+0.000×HMGdosage+0.069×endometrial thickness on the day of administration of HCG)}. The forecast probability and the data concerning whether clinical pregnancy was successful were used as basis to draw the ROC curve and to work out the area under the curve. The area was 0.668, with the 95% confidence interval being 0.638-0.698, which is of statistical significance(p=0.000). It implies that the multiple-factor regression equation related to pregnancy is of certain diagnostic value.3.Single factor analysis: after a comparison was made between the successful conception group and unsuccessful conception group in terms of the E2 and LH of basal endocrine indices, the investigators found that the difference was of statistical significance; in the unsuccessful conception group, the rate of the complications, including hysteromyoma, adenomyosis, and uterine surgery, was higher than that of the successful conception group; in the unsuccessful conception group, the FSH and HMG dosage during COH process were higher than those in the successful conception group; after a comparison was made between the two groups in terms of the number of extracted eggs and the number of transplanted embryos at the left side during the transplantation process, the numbers in the successful conception group was higher than that in the unsuccessful pregnancy group; after the comparison was made between the two groups in terms of the blood HCG detected value 28 days after the transplantation, the value of successful conception group was higher than that of the unsuccessful pregnancy group, and the difference was of statistical significance.A multiple-factor regression analysis was carried out on the patients’ general conditions, complications, basal endocrine and ovulation promotion, and transplantation related indices according to the results of the above single factor analysis. The obtained factors which were of statistical significance to the success of pregnancy included: combined adenomyosis.history of uterine surgery, basal E2, FSH dosage during COH, HMG dosage, HCG values 28 days after transplantation. The above factors independently affect the outcome of pregnancy outcome.4. The forecast probability and the data on whether conception was successful were used as basis to draw the ROC curve and to obtain the area under the curve. The ROC area under the curve was 0.668, with the 95% confidence interval 0.634-0.742, being of statistical significance(p=0.000). The result indicated that the multiple-factor regression equation related to successful conception was of certain diagnostic value.Conclusion1.Through single factor analysis and multiple-factor analysis, the effect factors to the clinical pregnancy included:age, years of infertility,combined uterine malformations, PCOS, hydrosalpinx,basal FSH,basal FSH/LH,COH,HMG dosage,endometrial thickness on the day of administration of HCG. The forecast model(P) for clinical pregnancy probability draw the ROC curve. It implies that the multiple-factor regression equation related to pregnancy is of certain diagnostic value.2. Through single factor analysis and multiple-factor analysis, the effect factors to the pregnancy outcome included: combined adenomyosis.history of uterine surgery, basal E2, FSH dosage during COH, HMG dosage, blood HCG values 28 days after transplantation. The forecast model(P) for pregnancy outcome probability draw the ROC curve. It implies that the multiple-factor regression equation related to pregnancy outcome is of certain diagnostic value. |
PDF Full Text Request