Background

Ectopic pregnancy has been reported to occur in approximately 2–5% of all clinical pregnancies after IVF-ET [1–4]. Although the direct injection of transfer media with embryos into the fallopian tubes may account for the development of tubal pregnancies after IVF, migration of embryos to the tubes by reflux expulsion from uterine contractions has been proposed as another possible explanation. [3, 5]

Uterine junctional zone activity has been shown to decrease with increasing time after oocyte retrieval. [6]. When comparing day 2 to day 3 transfers, Lesny et al. [3] showed a trend for a lower ectopic pregnancy rate in the day 3 transfer group which they attributed to the decreased uterine contractility further along in the luteal phase. Fanchin et al. [7] reported a significant reduction in retrograde uterine contractility, from the cervix to the fundus, 7 days after hCG administration compared to both 4 days after and the day of hCG injection. These findings suggest that blastocyst transfer should be associated with a lower incidence of ectopic pregnancy compared to cleavage stage transfer. The larger diameter of the blastocyst was proposed as an additional factor in reducing the rate of tubal pregnancies after day 5 transfer. [8]

Despite these theoretical considerations, large series that specifically compare the incidence of ectopic pregnancy with blastocyst versus cleavage stage transfers are lacking in the literature. The purpose of our study is to shed light on this issue by examining the ectopic pregnancy rates after day 3 transfer compared to day 5 transfer in our program over a 5 year period.

Methods

We reviewed all clinical pregnancies conceived in our IVF program since 1998 when blastocyst transfer was introduced to our center. The incidence of ectopic pregnancy was compared between day 3 and day 5 transfers in the same time period.

The controlled ovarian hyperstimulation protocol consisted of pretreatment with oral contraceptive pills with overlapping GnRH agonist down-regulation followed by FSH/hHMG and hCG, microdose flare or antagonist protocols. Oocytes were inseminated conventionally or by ICSI 3–4 hours after retrieval. Embryos were cultured in groups under mineral oil in 150 μL droplets of P1 medium (Irvine Scientific, Santa Ana, CA, USA) with 10% Serum Substitute Supplement (SSS) at 37 degrees Celsius in a 5% O2, 5% CO2 and 90% N2 environment for 72 hours. For the blastocyst transfer group, the embryos were moved on day 3 into Blastocyst medium (Irvine Scientific) with 10% SSS and cultured for 48 hours before transfer. Additional blastocysts were cryopreserved on day 5 or day 6.

All transfers were performed using a Tefcat catheter (Cook Ob/Gyn, Spencer, IN, USA) 1 to 1.5 cm short of the fundus under transabdominal ultrasound guidance. The transfer volume was 20–30 μL.

Clinical pregnancies were defined by seeing a gestational sac on transvaginal ultrasound or by diagnosing an ectopic pregnancy. Ectopic pregnancies were diagnosed by ultrasound or by laparoscopic visualization of a gestational sac in the fallopian tube or by the absence of an intrauterine gestational sac and rising βhCG levels following the failure of suction D&C to reveal products of conception.

The rate of ectopic pregnancies for day 3 and day 5 transfers was compared. Chi-square testing was used for statistical analysis. Significance was set at P < 0.05. Institutional review board approval was obtained for chart review.

Results

There were 623 clinical pregnancies resulting from day 3 transfer of which 22 were ectopic (3.5%). In day 5 transfers, there were 13 ectopic pregnancies out of 333 clinical pregnancies (3.9%). The difference between these rates is not statistically significant (P = 0.8). Of the 22 ectopic pregnancies with day 3 transfer, 9 were in patients with tubal disease compared to 5 out of the 13 ectopic pregnancies with day 5 transfer (P = 0.9). More importantly, the incidence of tubal disease was similar in the day 3 transfer and the day 5 transfer groups, 22 and 24%, respectively (P = 0.4). The mean ages were 37.7(± 4.9) years in the day 3 group and 35.3 (± 4.7) years in the day 5 group (P < 0.01). The mean BMI was similar in both groups. (Table 1)

Table 1 Comparison of ectopic pregnancy rates in day 3 versus day 5 embryo transfers
Full size table

In our program, we primarily perform cryopreservation at the blastocyst stage. [9] Accordingly, the day 5 transfer group includes the vast majority of the thaw embryo transfers. When these pregnancies are excluded and only fresh transfers are considered, the ectopic pregnancy rate remains similar for day 3 and day 5 at 3.5% (22/615) and 3.3% (9/271), respectively. (P = 0.8)

Discussion

Studies that showed decreased uterine contractility further along in the luteal phase [6, 7] would imply that the ectopic pregnancy rates should be reduced after a day 5 transfer compared to a cleavage stage transfer. It has also been postulated that the larger size of the blastocyst may decrease the chances of the day 5 embryo from migrating to the fallopian tube. [8] Despite these theoretical mechanisms which suggest that day 5 transfer is associated with a lowered ectopic pregnancy risk, our study, which examined close to a thousand pregnancies, failed to show such a trend. It is possible that when a blastocyst is transferred, it does indeed have a lower probability of entering the fallopian tube. However, the blastocyst that does reach the tube may have a higher tendency to implant there while the day 3 embryo has 2 additional days, compared to the day 5 embryo, to migrate back into the uterine cavity.

A potential source of bias in our study is the fact that blastocyst transfer was offered to patients with more than 3 eight cell embryos on day 3 which is likely to occur in patients with a higher number of oocytes and higher estrogen levels. Specific data on oocyte number is not available for our study, and we do not routinely measure estradiol levels in our program. However, when specifically analyzed in previous studies in the literature [1, 2], these parameters were not found to affect the incidence of tubal pregnancy. Although the patients in the blastocyst transfer group were on the average 2 years younger than those in the day 3 group, it is unlikely that this small difference could have had an impact on increasing the ectopic pregnancy rate in the day 5 group. If anything, the rate of ectopic pregnancy has been reported to increase with age [10]. Another confounding factor could be the prevalence of tubal disease in the two patient populations studied, as tubal pathology has been shown to be a major risk factor for the development of an ectopic pregnancy with IVF [11, 12] The incidence of tubal disease is unlikely to be a source of bias in our study since it was similar in our day 3 and day 5 transfer groups.

Although a comparison of the incidence of ectopic pregnancy between cleavage stage transfer and blastocyst transfer has not been the specific subject of any prior study in the literature, information on this issue can be found in a report by Marek et al. [13] In their study, the authors compared the pregnancy rates in their program when they switched from day 3 to day 5 transfer for all patients. The ectopic pregnancy rates can be extrapolated from their tabulated data as being 1% (2/199) with day 3 and 1.3% (2/159) with day 5 transfers. The findings of this smaller series confirm the absence of a decrease in ectopic rates after blastocyst transfer.

The literature contains 2 additional studies that incidentally report data allowing the computation of the ectopic pregnancy rate with blastocyst transfer without any information on day 3 transfers. In one study, Pantos et al. [14] examined the influence of age on the pregnancy rate after blastocyst transfer and mentioned 4 ectopic pregnancies out of a total of 99 pregnancies (4%). In the other study, Tarlatzis et al. [15] looked at monozygotic twinning with blastocyst transfer after ICSI and conventional IVF and noted an ectopic pregnancy rate of 2 out of 48 pregnancies (4.2%). Although the purpose of these studies was not related to the issue of ectopic pregnancy and they lacked day 3 controls, the rates of about 4% are in line with what has been reported with day 3 transfers and suggest that blastocyst transfer does not reduce the likelihood of tubal pregnancy.

Conclusion

We believe that blastocyst transfer is a valuable tool that has enabled IVF programs to more accurately select the embryos with the highest potential for implantation [16–18] allowing for a good pregnancy rate while avoiding high order multiple gestations [19, 20]. In our program, we offer blastocyst transfer to patients of any age [21] if they have more than 3 eight cell embryos. Although some authors have advocated routine blastocyst transfer in all patients [13, 22], offering extended culture when there are no eight cell embryos on day 3 has been reported to be detrimental [23]. We suggest that programs establish the criteria that work for them for offering blastocyst culture and transfer. However, based on the results of this study, the presence of risk factors favoring ectopic pregnancy should not be taken into account in the decision making for choosing to transfer on day 3 or day 5.

Author's Contributions

AAM, faculty attending physician, treated the patients, conceived of the study and co-wrote the manuscript. SHJ, resident, collected the data and co-wrote the manuscript.