Risk factors of ectopic pregnancy after in vitro fertilization-embryo transfer in Chinese population: A meta-analysis.

In recent years, there has been a notable increase in the annual incidence of infertility. According to statistics, one out of every seven couples of childbearing age experience infertility [ 1 ]. The rapid development and widespread application of in vitro fertilization-embryo transfer (IVF-ET) have become important methods for infertile couples to achieve fertility. However, ectopic pregnancy (EP), a high-risk complication of IVF-ET, occasionally occurs. EP is not only a pregnancy failure but also a direct threat to the patient’s life. The incidence of EP after IVF-ET in China is between 3.2% and 8.6% [ 2 , 3 ], which is significantly higher than that observed after natural conception [ 4 ]. There are many related original studies; however, their results differ [ 5 , 6 ]. Furthermore, only a few studies have used meta-analyses to quantitatively and systematically evaluate these findings. Therefore, this study aimed to analyze the risk factors for EP after IVF-ET in the Chinese population using evidence-based medicine and provide a reference for identifying high-risk groups and implementing targeted prevention. We have successfully achieved this aim.

Initially, a total of 1,786 articles were retrieved, and after screening based on the inclusion and exclusion criteria, 34 articles were finally included in the study. The selection process is shown in S1 Fig . All data included in the literature can be seen in S1 Data . Among the 34 articles, 25 were case-control studies, and nine were cohort studies. The sample sources involved 16 provinces and municipalities across the country. In this study, the research data from Zhu BY [ 7 ], Wang [ 8 ], and Zhang CC [ 9 ] were divided into fresh cycle and freeze-thaw cycle categories, resulting in two separate data extractions. The basic characteristics of the included studies are presented in Table 1 . a = Fresh, b = Frozen; 1 = EMT on HCG administration day, 2 = History of EP, 3 = Infertility type, 4 = EMT at transplantation, 5 = History of induced abortion, 6 = PCOS, polycystic ovary syndrome, 7 = Male factor infertility, 8 = DOR, diminished ovarian reserve, 9 = Tubal factor infertility, 10 = Embryo transfer stage, 11 = Type of transfer, 12 = Endometrial preparation, 13 = E2 level on HCG day, 14 = Previous tubal surgery, 15 = Maternal age, 16 = No. of transferred embryos, 17 = Fertilization method, 18 = Previous of cesarean section, 19 = Ovulation Protocol, 20 = No. of oocytes retrieved, 21 = Maternal BMI, 22 = Previous pregnancy, 23 = History of pelvic surgery, 24 = Dose of gonadotrophin (IU). Heterogeneity tests showed that infertility type, history of induced abortion, polycystic ovarian syndrome (PCOS), decreased ovarian reserve, thawed endometrial preparation plan, estradiol (E2) level on the day of HCG administration, history of tubal surgery, history of cesarean section, number of oocytes aspirated, maternal body mass index (BMI), previous pregnancy history, and total dose of gonadotropin were less heterogeneous among the literature; thus, a fixed effect model was used for consolidation. A heterogeneity test of other factors included in the literature showed that I 2 > 50%, indicating significant heterogeneity. Therefore, a random effects model was used for consolidation. The results of the meta-analysis showed that a thin endometrium on the days of HCG administration and embryo transfer (OR 1.951, 95% CI [1.598–2.381]), (OR 1.511, 95% CI [1.197–1.908]), history of EP (OR 1.541, 95%CI [1.213–1.957]), secondary infertility (OR 1.326, 95% CI [1.171–1.502]), history of induced abortion (OR 2.054, 95% CI [1.310–3.222]), PCOS (OR 2.164, 95% CI [1.386–3.381]), decreased ovarian reserve (OR 1.751, 95% CI [1.346–2.279]), tubal factor infertility (OR 1.851, 95% CI [1.609–2.130]), cleavage stage embryo transfer (OR 1.870, 95%CI [1.417–2.466]), fresh embryo transfer (OR 1.463, 95% CI [1.062–2.016]), artificial cycle (OR 2.067, 95% CI [1.718–2.487]), higher E2 level on HCG day (OR 1.001, 95% CI [1.001–1.001]), history of fallopian tube surgery (OR 2.692, 95% CI [2.075–3.494]), two or more number of embryo transfer (OR 1.517, 95% CI [1.226–1.878]), history of cesarean section (OR 1.632, 95% CI [1.005–2.652]), past pregnancy history (OR 1.227, 95% CI [1.057–1.423]) and history of pelvic surgery (OR 1.909, 95% CI [1.349–2.701]) were identified as risk factors for EP after IVF-ET. Additionally, maternal age and male factor infertility were found to be associated with EPs. The fertilization method, ovulation induction protocol, number of aspirated oocytes, maternal BMI, and total dose of gonadotropin were unrelated to EP after IVF-ET. The detailed results of the meta-analysis are shown in Table 2 . A forest plot example based on the type of infertility is shown in S2 Fig . EMT:endometrial thickness; HCG:human chorionic gonadotropin; EP:ectopic pregnancy; PCOS:polycystic ovarian syndrome; DOR:diminished ovarian reserve; E2:estradiol; BMI:body mass index. Statistically significant influencing factors were analyzed using both random and fixed effects models. The results showed that a history of cesarean section was a significant risk factor for EP in the fixed effects model. However, under the random-effects model, with P = 0.062, it was no longer statistically significant, suggesting that the results were unstable; further investigation is needed to determine whether a history of cesarean section is indeed a risk factor for EP after IVF-ET. The other risk factors did not change significantly between the two effect models, and the combined results were stable. The specific values are listed in Table 3 . EMT:endometrial thickness; HCG:human chorionic gonadotropin; EP:ectopic pregnancy; PCOS:polycystic ovarian syndrome; DOR:diminished ovarian reserve; E2:estradiol. More than nine articles included endometrial thickness on HCG administration day, history of EP, tubal factor infertility, cleavage stage embryo transfer, and the number of embryos transferred. Publication bias was assessed using Egger’s test. The results indicated the possibility of publication bias concerning endometrial thickness on the day of HCG administration. In contrast, the possibility of publication bias for other risk factors was small, the specific values are shown in Table 3 . Publication bias regarding endometrial thickness on the day of HCG administration was corrected using the clipping method. The corrected results showed that after the data from eight virtual studies were included, the combined result was OR = 1.390, with a 95% CI of 1.156–1.672, which was not significantly different from the value before correction, indicating that this result was minimally affected by publication bias.

The risk factors for EP after IVF-ET are a thin endometrium on the days of HCG administration and embryo transfer, history of EP, secondary infertility, history of induced abortion, PCOS, decreased ovarian reserve, tubal factor infertility, cleavage-stage embryo transfer, fresh embryo transfer, artificial cycle, high E2 level on the day of HCG administration, history of tubal surgery, two or more embryo transfers, previous pregnancy history, and pelvic surgery history.

The meat-analysis, was performed following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines; the PRISMA Checklist is presented in S1 Checklist . We systematically searched the China National Knowledge Infrastructure, Wang fang Database, China Science Technology Journal Database, Chinese Biomedical Literature Database, PubMed, Web of Science and Embase databases from their establishment to April 2023. We sought relevant literature on the risk factors associated with EP after IVF-ET in Chinese women. Additionally, the references within the selected studies were reviewed. A combination of subject-specific terms and free-text keywords was used for retrieval and adjusted according to different databases. The search terms included “assisted reproductive technology”, “in vitro fertilization and embryo transfer”, “ectopic pregnancy”, “risk factors”, “influencing factors”, “related factors”. Studies were included in the meta-analysis if: (1) the study participants were Chinese women who had undergone IVF-ET; (2) the fertilization technique used was IVF or intracytoplasmic sperm injection (ICSI); (3) the original literature used multivariate logistic regression analysis to identify relevant risk factors; (4) the study type was case-control or cohort study; (5) they were published in Chinese or English; (6) literature quality score was ≥ 7 points; (7) there were clear definitions of cases and risk factors in the literature; (8) the number of studies on the same risk factor was ≥ 2; and (9) if the same study population was reported in different articles, articles with more risk factors were included. The exclusion criteria employed were as follows: (1) literature aimed at studying the risk factors for heterotopic pregnancy or recurrent EP; (2) literature with the same data published repeatedly or in different articles; (3) inaccessibility to full text, invalid data, data self-contradictory literature; (4) studies involving unconventional techniques, such as preimplantation embryo genetic diagnosis and intratubal transplantation. The retrieved literature was imported into Endnote, a literature management software. After removing duplicate studies, the remaining literature was screened according to the inclusion and exclusion criteria to determine whether they were included. The Newcastle-Ottawa Scale was used to evaluate the quality of the literature that might be included. Finally, the included literature was determined, and data were extracted, including the author’s name, year of publication, country of publication, study type, sample size, and identified risk factors. Two researchers independently executed the entire process after unified training, and the results were cross-checked. The part with differences in the results was decided by the third person. We used Stata17 for statistical analysis. Odds ratio (OR) and their 95% confidence interval (CI) were used to represent the effects of the statistical analysis. Heterogeneity between the studies was evaluated using I 2 values. An I 2 value of ≤ 50% indicated low heterogeneity, prompting the selection of a fixed effect model for statistical aggregation. Otherwise, indicated high heterogeneity, prompting the selection of a random effect model for statistical aggregation. The sensitivities of the results were evaluated by transforming the two analysis models. The Egger’s test was used to analyze publication bias for risk factors with more than nine articles. The stability of the results for risk factors with publication bias was evaluated using the clipping method. Statistical significance was set at P < 0.05.

(TIF) Click here for additional data file. (TIF) Click here for additional data file. (DOCX) Click here for additional data file. (XLS) Click here for additional data file.