Embryo transfer (ET) has been noted as a crucial step that has an influence on pregnancy rates in patients undergoing the in vitro fertilization (IVF) process [1,2]. The factors that affect the success of ET include embryo quality, endometrial receptivity, and the ET technique itself [2]. Although ET techniques have improved immensely, the incidence of embryo retention (ER) in the transfer catheter following attempted ET has remained consistent; the reported incidence rates of ER were between 1% and 7% in ET with cleavage-stage embryos [3,4,5,6,7,8]. There have been several retrospective studies about the effect of retransferring retained embryos on outcomes in IVF-ET with cleavage-stage embryos [3,4,5,6,7,8]. Among them, several studies [3,4,5,6,8] have reported that retransferring retained embryos in the transfer catheter did not affect the IVF outcomes. With great advances in embryo culture and transfer techniques, blastocyst transfer has taken center stage in IVF as an efficient strategy for avoiding multiple pregnancies while improving the pregnancy rate [9,10]. An error during the ET procedure resulting in ER cannot be ignored under the pressure to ensure that only one or two blastocysts are transferred and to achieve a high probability of pregnancy. Such an error may lead to impaired IVF outcomes. However, there is a paucity of data about the incidence and reproductive outcomes of retained blastocysts in the transfer catheter. Moreover, some studies have speculated that retained blastocysts had a larger diameter and a more fragile zona pellucida, in which case the ET procedure may cause some trauma to the blastocyst and epigenetic effects on the embryo [11,12,13,14]. Keeping these studies in consideration, blastocyst transfer itself could be a factor that may affect the incidence of ER and ER in blastocyst transfer may affect the blastocysts and their subsequent growth in the uterus, which would have a negative effect on IVF and perinatal outcomes.

In this study, we analyzed 10 years of data from Cheil General Hospital to determine the incidence of blastocysts retained in the transfer catheter during Day 5 blastocyst transfer cycles where one or two blastocysts were transferred. We also investigated whether retransferring retained blastocysts had an impact on IVF and perinatal outcomes.

A total of 1,131 patients, comprising 776 fresh ET cycles and 355 frozen-thawed ET cycles, were recruited retrospectively at Cheil General Hospital from January 2004 to December 2014. Cycles for preimplantation genetic diagnosis and donor oocyte cycles were excluded. Among the patients included, 223 single blastocyst transfers (SBTs) and 908 double blastocyst transfers (DBTs) were performed. In fresh cycles, every patient had undergone controlled ovarian stimulation. Ovulation triggering was performed with the administration of 250 µg of human chorionic gonadotropin (hCG) (Ovidrel, Merck Serono, Darmstadt, Germany) as soon as two to three follicles 17 mm in diameter were observed by transvaginal ultrasound. Transvaginal oocyte retrieval was performed approximately 36 to 38 hours after the administration of hCG. In frozen-thawed cycles, we used 6 to 8 mg/day of oral estradiol valerate (Progynova, Bayer Schering Pharma, Berlin, Germany) for endometrial preparation and we used intramuscular progesterone (50 mg in oil, daily) for luteal support. One or two blastocysts that survived vitrification were transferred into the patient's uterus 5 days after the initiation of progesterone treatment. The embryos obtained were scored according to the modified Gardner and Schoolcraft grading system [15].

ET was performed with a full bladder under transabdominal ultrasound guidance. A Cook Sydney IVF catheter (Cook Medical, Bloomington, IN, USA) was used. The patient was placed in the lithotomy position without any sedation or anesthesia. After the removal of cervical mucus, an empty outer transfer catheter was passed through the external cervical os to the level of the internal cervical os. Then, the embryos were loaded into an inner catheter with approximately 20 µL of media. The inner catheter contained a 5-µg air bubble at the tip to aid visualization during ultrasound guidance. The embryologist checked the catheter microscopically for the presence of retained embryos. If any embryo was found to have been retained in the catheter, the retained embryos were immediately reloaded by the embryologist and a second transfer was performed. Then, the catheter was checked again.

A pregnancy test was performed by determining the serum β-hCG level 12 days after the oocyte retrieval day in fresh ET cycles or 12 days after the initiation of progesterone treatment in frozen-thawed cycles. If the pregnancy test was positive, patients underwent an ultrasound scan 3 weeks later to establish the number of gestational sacs and embryo viability, and to exclude ectopic pregnancy.

The implantation rate was calculated as the number of gestational sacs per the number of transferred embryos. Clinical pregnancy was defined as the presence of an intrauterine gestational sac with a yolk sac, a fetal pole, and fetal heart pulsations. Live birth was defined as the birth of a live infant at ≥24 weeks of gestation.

We used a chi-square test and Fisher's exact test, as appropriate, to compare pregnancy and live birth and multiple gestation rates. A student's t-test was used to evaluate continuous parameters. A p-value <0.05 was considered statistically significant. The statistical analysis was performed using SPSS ver. 22.0 (IBM Co., Armonk, NY, USA). In comparing pregnancy rates, we did not distinguish between patients with one or more than one retained embryo.

A total of 1,131 ET procedures with single or double blastocysts, with Day 5 blastocysts, were performed by eight experienced physicians from January 2004 to December 2014. The cycle characteristics of single and double groups (SBT and DBT, respectively) are presented in Table 1. SBT and DBT groups were similar with regard to body mass index, basal estradiaol and FSH, endometrial thickness at day of hCG injection, and number of retrieved oocytes. Patients in the DBT group were significantly younger and had experienced fewer previous ET cycles (p<0.001 and p<0.001, respectively).

The overall incidence of finding retained embryos during the study period was 2.8% (32/1,131). Only two ET cycles required a third transfer attempt. There was no case of ER during transfer in the SBT group. Therefore, to compare IVF outcomes, only the DBT group was classified by whether the embryos were successfully transferred at the first attempt or not. The cycle characteristics of the without-ER group and the ER group are presented in Table 2. The without-ER group and ER group were similar with regard to age, body mass index, the number of previous ET cycles, basal E2 and FSH, endometrial thickness at day of hCG injection, and number of retrieved oocytes. The pregnancy outcomes of the two groups are described in Table 3. Implantation rates (30.0% vs. 26.6%), positive β-hCG rates (57.2% vs. 56.2%), clinical pregnancy rates (45.3% vs. 46.9%), and live birth rates (38.9% vs. 43.8%) were not significantly different between the two groups (Table 3).

Table 4 shows the perinatal outcomes of children born in this study. Out of 341 deliveries in the without-ER group, 440 babies were born. The mean gestational age was 269.3±17.2 days and birth weight was 2,928.4±631.8 g. There were 92 SBTs resulting in twins. Among them, seven were monozygotic twins. Triplets resulted in two cases. In the ER group, out of 14 deliveries, 17 babies were born. The mean gestational age was 264.2±25.7 days and the mean birth weight was 2,948.7±497.8 g. There were three sets of twins. Among them, there was one set of monozygotic twins and there were no triplets. No statistically significant differences in the mean gestational age at birth and the mean birth weight were noted between the without-ER and ER groups (p=0.316 and p=0.903, respectively).

A classification of all observed birth defects is listed in Table 5. Birth defects were classified according to the criteria of the National Center on Birth Defects and Developmental Disabilities, Centers for Disease Control and Prevention, United States Department of Health and Human Services [16]. A total of nine cases of congenital birth defects were found in this study population. Eight were observed in the without-ER group, and one in the ER group.

Finding retained embryos in the transfer catheter may be one of the most stressful and inconvenient situations for both the clinician performing the transfer and the patient. Furthermore, during an ET procedure, it is important that the patient be relaxed, because psychological factors such as stress and anxiety could affect rates of IVF outcomes [17,18,19].

Over the decades, it has long been debated whether retransferring embryos retained during the transfer could have a negative effect on pregnancy rates in IVF patients. Recent studies [3,4,5,6,8] have reported that retransferring retained embryos in the transfer catheter did not have an adverse effect on IVF outcomes. However, all of these studies concerned cleavage-stage embryos. Nowadays, blastocyst transfer with one or two embryos has become common in order to reduce multiple pregnancy rates and to enhance IVF outcomes. However, there has been a lack of consensus regarding the incidence of ER and its influence on blastocyst transfer. Moreover, Lee et al. [3] stated that the difference in the morphological and hydrodynamic features according to the developing embryo stage might have an influence on the incidence of retained embryos. In other words, the larger, less dense, expanded blastocyst compared to the dense, compact character of cleavage-stage embryos might result in an increased risk of blastocyst retention in the transfer catheter. In addition, some studies have speculated that the zona pellucida of the blastocyst is more fragile near the time of hatching, and ET may cause some trauma to the blastocyst [11,12]. Considering these blastocyst features, it seems reasonable to postulate that blastocyst transfer itself might be a risk factor for ER or additional procedures like embryo retransfer might result in compromised IVF and prenatal outcomes. Two prior studies have evaluated ER in blastocyst transfer. Lee et al. [3] compared initial pregnancy rates for Day 3 embryo transfers with ER to those of blastocyst transfer without ER (40% vs. 36%, respectively) and found no statistically significant difference between them. However, they did not present concrete information about blastocyst transfer cycles, such as the number of study cases or the incidence of ER. On the other hand, although Silberstein et al. [6] reported that the embryonic stage at ET did not demonstrate a statistically significant difference, they simply compared the incidence of ER in cleavage-stage transfer cycles to that of blastocyst transfer cycles (2.3% vs. 4.4%). Hence, the primary goal of this study was to determine the incidence of retransferring retained embryos and its influence on IVF outcomes in Day 5 blastocyst transfer.

Our data showed that the overall incidence of ER in Day 5 blastocyst transfer (2.8%) was similar to the incidence of ER in ET with cleavage-stage embryos. This result was identical with our previous unpublished data regarding the incidence of ER in cleavage-stage ET (2.8%). No ER was observed in SBT, while the incidence of ER in DBT was 3.5%. As expected, the number of transferred blastocysts might be a factor that leads to an increased incidence of ER. From this viewpoint, our results suggest that a lower number of blastocysts transferred has the advantage of decreasing the incidence of retained embryos in catheters.

For the purpose of comparing IVF and prenatal outcomes in a homogenous population between the without-ER and ER groups, we divided patients according to the number of transferred embryos. As there was no case of ER in the SBT group, only the DBT group was split into two groups depending on whether at least one embryo was retained in the transfer catheter after ET. Our data show that IVF outcomes such as the implantation rate, positive β-hCG rate, clinical pregnancy rate, live birth rate, and ectopic pregnancy rate were not significantly affected by retransferring retained embryos during ET.

Some reports have noted a concern about the perinatal outcomes of blastocyst transfer, such as an increased risk for monozygotic twins (1.6%–5.6%) and for congenital malformations (2.5%–6.8%) [13,20,21,22,23]. Although the exact mechanism leading to this increased incidence is not yet fully understood, one suggested mechanism has been that micromanipulation of the fragile zona pellucida of the blastocyst may interfere with the natural process of hatching and the normal development of the embryo [13,20]. Considering this mechanism, it seems reasonable to postulate that ER requiring additional procedures might result in compromised IVF and prenatal outcomes. Comparing the perinatal outcomes of blastocyst transfer without ER to those of blastocyst transfer with ER, there were no statistically significant differences in the mean birth weight and the mean gestational age at birth between the without-ER and the ER groups, except for the mean gestational age. There was one case of monozygotic twins in the ER group. Although the incidence of congenital birth defects was higher in the ER group (2.1% vs. 5.8%), the number of cases with ER was too small to perform a proper statistical analysis. Accordingly, our data could not prove that the possibility existed for a lack of increased risk of congenital malformation in the ER group. However, considering the result of this presentation, the apprehensions might prove to be unfounded that blastocyst transfer could ininfluence the incidence of retained embryos or blastocyst transfer and retransfer could be more traumatic to the embryo.

Our results demonstrate that Day 5 blastocyst transfer does not increase the incidence of retained embryos and proper retransfer of retained embryos has no adverse impact on IVF and perinatal outcomes in Day 5 blastocyst transfer cycles. To our knowledge, this study is the first to postulate the prenatal outcomes of blastocyst transfer with ER in addition to evaluating the incidence and IVF outcomes of blastocyst transfer with ER. Thus, we hope that the results of this study could help in resolving some of the debates found in the literature on blastocyst transfer. Furthermore, our results support the conclusion that a lower number of blastocysts transferred, especially with SBT, has the advantage of decreasing the incidence of retained embryos in catheters, in addition to preventing the multiple pregnancies associated with a transfer of multiple embryos.