Pregnancy Results Of 401 Women With Poor Ovarian Response With Consecutive In-vitro Fertilization Treatment Cycles

Background and objectives:Since the first report of a woman who manifested lower number of follicles and retrieved oocytes, a peak value of estrogen<300 pg/ml during ovarian stimulation in 1983, poor ovarian response (POR) has become a major challenge for clinicians. The prevalence of POR ranges from 9% up to 24% in in-vitro fertilization (IVF) treatment. Although different protocols and adjuvants for ovarian stimulation have been proposed to improve ovarian response, it seemed that patients with POR benefit little from the interventions. Furthermore, disagreement on the definition of POR limited the comparison and generalization of results from different clinical trials until Bologna consensus was published in 2011. Poor pregnancy results, such as live birth rate (LBR) per cycle being less than 10% and varying from 2.6% to 8.3%, similarly, and LBR per woman varying from 7.4% to 20.5%, have been reported. However, few comparative data about LBR and cumulative live birth rate (CLBR) in continuous cycles have been published until now. CLBR instead of LBR per-cycle does provide a reference for continued treatment for individual patients.It has been established that both LBR and CLBR decline with increasing maternal age for an infertile population. According to previously published reviews applying different diagnostic criteria for POR, young patients established more pregnancies compared with older patients with POR. However, in most studies using Bologna criteria, no association between age and LBR has been revealed.The aim of the present study was to observe the LBR and CLBR in 401 patients with POR diagnosed according to Bologna criteria, undergoing consecutive cycles of in-vitro fertilization/introcytoplasmic sperm injection-embryo transfer (IVF/ICSI-ET), and compare the pregnancy results among different subgroup and different age after multiple treatment cycles. Moreover, Multivariable logistic regression analysis was performed to identify potential factors contributing to live birth.Materials and methodsFour hundred and one women with POR diagnosed according to the Bologna criteria were enrolled, who underwent their first IVF/ICSI-ET treatment cycles from January to December of 2013 in the Reproductive Hospital Affiliated to Shandong University. All consecutive treatment cycles were followed. The Bologna criteria were applied as follows:At least two of the three features should be present:Ⅰ. Advanced maternal age (≥40 years) or any other risk factors for POR (previous ovarian surgery, ovarian endometrioma, genetic abnormalities for POR, previous chemotherapy);Ⅱ. A previous POR (< 3 oocytes retrieved with a conventional stimulation protocol using at least 150IU gonadotropin for initiation);Ⅲ. An abnormal ovarian reserve test (i.e. AFC<5 follicles or AMH<1.1 ng/ml).Ⅳ. Two episodes of ≤3 oocytes retrieved with a conventional stimulation protocol in the absence of advanced maternal age (≥40 years) and abnormal ovarian reserve tests were also included.Comparisons of LBR and CLBR among different groups were performed using chi-squared test and Fisher’s exact test. Multivariable logistic regression analysis was performed to identify potential factors contributing to live birth. SPSS version 19.0 was used for analysis.Results1、The overall LBR per cycle was 18.3% in the 401 women undergoing 700 cycles of IVF/ICSI-ET. The CLBR per woman was 31.9%.2、No significant was found both in LBR per cycle and CLBR of different POR subgroups, with LBR per cycle ranging from 8.7% to 21.5%, and CLBR varying from 20.0% to 31.0%, respectively.3% From cycle 1 to cycle 3, the LBR per cycle decreased significantly (P<0.01), from 22.2% in the first cycle to 11.1% in the third cycle. The CLBR increased from 22.2% to 31.7% accordingly. Extremely low LBR (2.4%) was obtained in patients with 4 consecutive cycles or more.4、Age was divided into 3 groups and different LBR were observed; i.e.30.0% for women younger than 35 years of age,17.0% for those aged 35-40 years, and 9.0% for women exceeding 40 years of age (P<0.01). The CLBR was significantly higher for young patients aged<35 years (48.0%), compared with 30.1% for those aged 35-39 years, and 16.9% for the women exceeding 40 years of age (P<0.01). In the first cycle, younger patients (i.e.<35 years old) showed higher LBR (36.0%) compared with older women (i.e.19.4% for 35-40 years old and 10.1% for ≥40 years old) (P<0.01). However, these differences between young and old patients disappeared among the third and subsequent treatment cycles.5、Multivariable logistic regression analysis of first treatment cycle showed that age, number of embryos and endometrial thickness on HCG day are associated with the LBR. The ORs for live birth were 0.86 (95% CI:0.82-0.92),2.01 (95%CI: 1.62-2.50), OR=1.21, (95%CI:1.06-1.39),respectively.Conclusions:1. Bologna criterion for POR women can identify homologous population with declined ovarian reserve and poor pregnancy results.2. From cycle 1 to cycle 3, the LBR per cycle decreased significantly, extremely low LBR was obtained in patients with 4 consecutive cycles or more. Given this low rate of live birth (2.4%), a treatment consisting of more than three cycles of IVF in women with POR should not be recommended.3. Younger poor ovarian response women achieved better pregnancy results in the first three IVF cycles. Our findings suggest that patients with POR or high risk for POR should seek infertility treatment as early as possible.4. Maternal age, number of embryos and endometrial thickness on HCG day may be potential predictors for LBR.