Ultrasound-guided embryo transfer: a prospective randomized controlled trial

Oi Shan Tang,1, Ernest Hung Yu Ng, William Wai Kei So and Pak Chung Ho

Department of Obstetrics and Gynaecology, The University of Hong Kong, Queen Mary Hospital, Hong Kong SAR, China


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
BACKGROUND: Recent reports suggested that ultrasound guidance during embryo transfer might improve the pregnancy rate. METHODS: A prospective randomized controlled trial was performed to compare embryo transfer under ultrasound guidance versus the clinical touch method. A total of 800 embryo transfers was studied; 400 were randomized to ultrasound-guided transfers and 400 were randomized to the clinical touch group. Of these, 441 were fresh cycles and 359 were frozen–thawed cycles. RESULTS: The clinical pregnancy rate was 26.0% in the ultrasound-guided group and 22.5% in the clinical touch group; the difference was not statistically significant. The ongoing pregnancy rate was 23.5% in the ultrasound-guided group compared with 19.0% in the clinical touch group and the difference was again not statistically significant. The implantation rate was slightly higher in the ultrasound-guided group (15.3%) than the clinical touch group (12.0%) (P = 0.048). There were no differences in the incidences of ectopic pregnancy, miscarriage and multiple pregnancy between the two groups. CONCLUSIONS: A significant improvement in implantation rate was observed following the use of ultrasound guidance during embryo transfer. The extent of improvement in the pregnancy rate may depend on the specific techniques and methods of embryo transfer used in individual centres.

Key words: embryo transfer/IVF/transabdominal ultrasound


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
The methods of ovarian stimulation and oocyte recovery have undergone major revision since the advent of IVF and embryo transfer. However, the method of embryo transfer has remained very much the same. In most centres, embryo transfer is a blind procedure. The transfer catheter is placed in the uterine cavity guided only by the clinician's tactile senses. It is an imprecise procedure and it is difficult to ascertain that the embryos have been placed in the desired position in the uterine cavity.

The use of ultrasound guidance during the transfer procedure has previously been advocated to improve the pregnancy rate, as tactile assessment of embryo transfer catheter placement might be unreliable. Woolcott and Stanger found by using transvaginal ultrasound examination that in 17.4% of transfers the outer guiding catheter abutted the fundal endometrium, the outer guiding cannula indented the endometrium in 24.8% and the transfer catheter embedded in the endometrium in 33.1% (Woolcott and Stanger, 1997Go). The use of ultrasound-guided transfer avoided tubal transfer in 7.4% of transfers. It was difficult to judge from the study how these factors could affect the pregnancy rate since this was not a randomized study. In a randomized trial, the use of transabdominal ultrasound guidance during embryo transfer improved the pregnancy rate from 33.7 to 50% (Coroleu et al., 2000Go). In another retrospective study (Wood et al., 2000Go) the pregnancy rate improved from 25 to 38% by using ultrasound during embryo transfer. However, three other prospective trials failed to show such a significant improvement (Hurley et al., 1991Go; Al-Shawaf et al., 1993Go; Kan et al., 1999Go).

In view of the controversies in the literature, it was the objective of this prospective randomized controlled study to find out whether the use of transabdominal ultrasound guidance during embryo transfer could improve the pregnancy rate.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Infertile patients undergoing treatment with IVF, intracytoplasmic sperm injection (ICSI) and replacement of frozen–thawed embryos in the Assisted Reproduction Unit at the Department of Obstetrics and Gynaecology, Queen Mary Hospital were recruited into the study between September 1999 and October 2000. Written consent was obtained from patients prior to participating in the study. The study protocol was approved by the Ethics Committee, Faculty of Medicine, University of Hong Kong. Eight hundred patients were recruited. In all, 441 patients had fresh embryo transfer and 359 women had frozen–thawed embryo transfer. Patients were excluded from the study if they required general anaesthesia for embryo transfer or if they refused to participate. They were randomized prospectively by computer-generated random table on the day of embryo transfer into two groups: (i) transbdominal ultrasound-guided embryo transfer; (ii) clinical touch embryo transfer. The randomization table was made and kept by a nurse who was not involved in the recruitment of subjects. All the embryo transfers were performed either by the two clinicians or the specialist nurse in the unit and this was the usual practice at our centre. The ultrasound examination was performed by one of the two clinicians (O.S.T. or E.H.Y.N.).

Ovarian stimulation protocol
The details of the long protocol of ovarian stimulation regimen used in the IVF and ICSI cycles at our centre have been described previously (Ng et al., 2000Go). Patients were down-regulated with gonadotrophin-releasing hormone (GnRH) analogue, buserelin (Suprecur; Hoechst, Frankfurt, Germany) nasal spray 150 µg four times daily from the mid-luteal phase of the pre-treatment cycle. Human menopausal gonadotrophin (HMG: 75 IU FSH and LH; Pergonal, Serono, Switzerland; or Humegon; Organon, Netherlands) injection was started on the second day of the treatment cycle. The ovarian response was monitored by serial transvaginal ultrasound scanning and serum oestradiol measurement. Human chorionic gonadotrophin (HCG, Profasi; Serono, Switzerland) 10 000 IU was given i.m. when the leading follicle reached 18 mm in mean diameter and there were at least three follicles >15 mm in mean diameter. Transvaginal ultrasound-guided oocyte retrieval (Aloka SSD-620, 5 MHz; Aloka Co. Ltd, Japan) was scheduled 36 h after the HCG injection. Embryo transfer was performed 48 h after the retrieval. The luteal phase was supported by a 1500 IU HCG i.m. injection on the day of embryo transfer and again 6 days later. Progesterone vaginal pessaries (cyclogest, Cox Pharmaceuticals, Barnstaple, UK) 400 mg twice daily for 10 days were used if the serum oestradiol on the day of the ovulatory dose of HCG was >18 000 pmol/l.

Frozen–thawed cycles
The protocol for frozen–thawed embryo replacement at our centre has been described previously (Ng et al., 2000Go). Frozen–thawed embryos are replaced at our centre in natural cycles, clomiphene citrate (CC) (Clomid, Merrell, Staines, UK)-induced cycles or controlled cycles using pituitary down-regulation followed by hormone replacement therapy, depending on the patient's ovulatory history. If the patient ovulated either spontaneously or after CC (50–150 mg daily from day 3 to day 7 of the menstrual cycle), her serum oestradiol and LH concentrations were monitored starting from 18 days before the expected date of the next period. Embryo transfer was performed on the third day after the LH surge.

For patients who failed to ovulate on CC, hormone replacement was required. Buserelin nasal spray (150 µg four times a day) was used for pituitary down-regulation from the mid-luteal phase of the pretreatment cycle. Oestradiol (Estrofem; Novo Nordisk, Surrey, UK) was started on day 2 of the thaw cycle (2 mg daily for 5 days, 4 mg daily for 4 days and then 6 mg daily for 4 days). After 13 days of Estrofem therapy, transvaginal ultrasound scanning was performed. Estrofem was reduced to 4 mg and progesterone (Cyclogest) vaginal pessaries 400 mg twice daily were added if the endometrial thickness was >=8 mm by transvaginal ultrasound scanning. Embryo transfer was performed on the fourth day after the administration of Cyclogest.

Technique of embryo transfer
Patients were admitted on the morning of embryo transfer and were randomized into two groups: ultrasound-guided transfer and clinical touch transfer. Those patients randomized to the ultrasound-guided group were asked to keep a full bladder before the transfer. The patients in the clinical touch were allowed to pass urine according to their need and this was the usual practice of our centre. The patients were put in a lithotomy position. In the ultrasound-guided transfer group, the pelvis was examined by transabdominal ultrasound (Aloka SSD-620, 3.5 MHz). The axis, degree of flexion and the configuration of the uterine cavity were noted. A bivalve speculum was used to expose the cervix. The cervix was cleansed with sterile saline and then culture medium. The embryo transfer was done by using the Jansen–Anderson catheter (Cooke, Queensland, Australia) which is a coaxial catheter with an outer bulb tip cannula (2 French) and an inner silicon transfer catheter with a lumen measuring 0.5 mm in diameter. The outer cannula was inserted into the cervical canal 4 cm from the external os and just beyond the internal os. A malleable obturator was used if difficulty was encountered. To ensure smooth entry to the uterine cavity, a dummy inner transfer catheter was inserted through the cervical canal so that the tip of it was 6 cm from the external os and was within the uterine cavity. In the ultrasound-guided group, the whole transfer procedure was done under ultrasound guidance. The aim was to put the tip of the catheter inside the uterine cavity 6 cm from the external os and care was taken not to advance the tip beyond 1 cm from the uterine fundus. Embryo(s) in 5 µl of medium (IVFTM –20/–100; IVF Science Scandinavia, Sweden) were then aspirated into the real transfer catheter by the embryologist. The transfer catheter with the loaded embryo(s) was then inserted into the uterine cavity via the outer cannula and the transfer volume was gently expelled by the embryologist. In the ultrasound group, intrauterine transfer was identified by the bright echoes, which represented the air bubbles in the transfer volume. The degree of difficulty of the transfer procedure was rated by the attending clinician or the specialist nurse. The procedure was considered difficult if an obturator, uterine sound or dilatation of the cervix was/were required.

A urine pregnancy test was performed 16 days after embryo transfer and ultrasound examination was done 10 days later to confirm intrauterine pregnancy and the number of gestational sacs.

The primary outcome measures were clinical pregnancy, on-going pregnancy and implantation rates. Clinical pregnancy was defined as a positive urinary pregnancy test and the presence of at least one gestational sac, irrespective of whether it was intra- or extrauterine, by ultrasound examination. An ongoing pregnancy was defined as the presence of at least one fetus with positive heart pulsations seen with ultrasound examination at or beyond 10 weeks gestation. Implantation rate was defined as the total number of gestational sacs seen on ultrasound examination in relation to the total number of embryos replaced. The incidences of ectopic pregnancy and missed abortion were also noted.

The average pregnancy rate in our centre was 17% per transfer and therefore a sample size of 400 in each arm would have 80% power to detect an 8% increase in pregnancy rate at 5% level of significance.

SPSS 7.5 for Windows Statistical Package was used for statistical analysis. Continuous variables were compared by Student's t-test if they were normally distributed. The Mann–Whitney U-test was used where the data were skewed. Differences in proportions were analysed with the {chi}2-test or Fisher's exact test as appropriate. P < 0.05 was considered statistically significant.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
A total of 800 women who underwent embryo transfer in our units were recruited into the study. There were 441 fresh embryo transfers and 359 frozen–thawed embryo transfers. One hundred and sixty one of the fresh embryo transfer cycles used embryos from ICSI. The characteristics of the women are summarized in Table IGo. There were no differences in the age, duration and aetiology of infertility and the number of embryos transferred between the ultrasound-guided and clinical touch transfer groups. The percentages of IVF, ICSI and frozen–thawed cycles were similar in both groups. The treatment characteristics of fresh IVF and ICSI cycles were also similar in both groups.


View this table:
[in this window]
[in a new window]
 
Table I. The clinical and treatment characteristics of the 800 embryo transfer cycles
 
Table IIGo shows the pregnancy outcomes of the 800 cycles. The clinical pregnancy rate was 26.0% in the ultrasound-guided group and 22.5% in the clinical touch group. The difference was not statistically significant. The ongoing pregnancy rate was 23.5% in the ultrasound group and 19.0% in the clinical group and again the difference was not statistically significant. The implantation rate in the ultrasound-guided group (15.3%) was found to be significantly higher than the clinical touch group (12.0%) (P = 0.048). There were no differences in the incidences of ectopic pregnancy, missed abortion and multiple pregnancy between the two groups. For the frozen–thawed cycles, there was no difference in the clinical pregnancy rates between the ultrasound-guided and clinical touch groups in spontaneous, CC-induced and hormone replacement cycles (Table IIIGo).


View this table:
[in this window]
[in a new window]
 
Table II. The pregnancy outcomes of the 800 embryo transfer cycles
 

View this table:
[in this window]
[in a new window]
 
Table III. Clinical pregnancy rates for different types of frozen–thawed cycles
 
The clinical pregnancy rate was similar in both groups when one, two or three embryos were transferred (Table IVGo). The use of ultrasound guidance during embryo transfer did not reduce the number of embyro transfers classified as difficult. The percentages of difficult transfers were the same in both groups, 16.3% in the ultrasound-guided group and 18% in the clinical touch group (data not shown). The pregnancy rates were the same in both groups independent of whether the transfers were considered as ‘easy’ or ‘difficult’ by the operators (Table VGo).


View this table:
[in this window]
[in a new window]
 
Table IV. Effects of number of embryos transferred on the clinical pregnancy rate
 

View this table:
[in this window]
[in a new window]
 
Table V. The effect of difficult embryo transfer on the clinical pregnancy rate
 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
The technique of embryo transfer remained more or less unchanged despite significant advances in the ovarian stimulation protocols and oocyte retrieval technique. Implantation of the embryos is said to involve an interaction between the embryo and endometrium. The quality of the embryo and the uterine receptivity are certainly important factors in determining the success of implantation. The embryo transfer technique itself may also be important.

Strickler et al. were the first to describe the use of transabdominal ultrasound in embryo transfer (Strickler et al., 1985Go). It was advocated that the use of ultrasound could help to position the catheter tip in the fundus of the uterus, and ejection of the transfer bubble could be documented. This was reassuring to both the clinician and the patients. This study was too small (16 embryo transfers) to show any improvement in pregnancy rate. Hurley et al. found that pregnancy rate was improved by using transvaginal guidance, although statistical significance was only reached in the subgroup of single embryo transfer (Hurley et al., 1991Go). This was not a true randomized trial as the patients in the ultrasound-guided group were only selected if the transfers were carried out at a time when two of the authors were available. The statistical calculation was also questionable (Coetsier and Dhont, 1991Go). Woolcott and Stanger performed transvaginal ultrasound in 121 embryo transfers to observe the position of the transfer catheter. It was found that clinical touch was unreliable in that the outer guiding catheter abutted the fundus in 17.4%, the outer guiding cannula indented the endometrium in 24.8%, the transfer catheter embedded in the endometrium in 33.1% and accidental tubal transfer occurred in 7.4% of transfers. The significance of these observations on the pregnancy rate was unknown (Woolcott and Stanger, 1997Go).

Two other retrospective studies showed that the use of ultrasound guidance improved the pregnancy rate in oocyte donation cycles and fresh IVF cycles (Lindheim et al., 1999Go; Wood et al., 2000Go). In the later study, the embryo transfer catheter was changed from a hard to a soft catheter during the study period and it was shown in the same study that the use of a soft catheter resulted in a better pregnancy rate. Therefore, it was very difficult to determine whether ultrasound guidance or the use of a soft catheter improved pregnancy rate in this particular study. Other studies also showed that the choice of transfer catheter had a significant impact on the pregnancy rate (Wisanto et al., 1989Go; Gonen et al., 1991Go).

Two prospective trials have addressed this problem more recently (Kan et al., 1999Go; Coroleu et al., 2000Go). The authors in both studies used transabdominal ultrasound and similar transfer catheters (Edwards–Wallace catheter). One study (Kan et al., 1999Go) found a higher pregnancy rate (37.8 versus 28.9%) and implantation rate (20.4 versus 16.2%) in the ultrasound-guided group compared with clinical touch group respectively. The difference was not statistically significant. This was not a randomized trial as the patients were enrolled in the ultrasound group depending on the availability of the ultrasonographer. Another study (Coroleu et al., 2000Go) found a higher pregnancy rate (50 versus 33.7%) and implantation rate (25.3 versus 18.1%) in the ultrasound group compared with the clinical touch group and the differences, this time, were statistically significant. It was difficult to explain the difference in the degree of improvement in the pregnancy rate in these two studies.

The current study was, to the best of our knowledge, the largest prospective randomized trial to examine the use of ultrasound guidance in embryo transfer. Although there was no significant improvement in the pregnancy rate, a consistent trend towards a better pregnancy rate in the ultrasound group in both fresh and frozen–thawed cycles was observed. The overall implantation rate of fresh and frozen–thawed embryos improved significantly from 12 to 15.3% if ultrasound guidance was used (P = 0.048).

Several factors may explain the discrepancy in the extent of improvement in the pregnancy rate between this and another study (Coroleu et al., 2000Go). Transabdominal ultrasound was used in this study as in most other recent randomized trials but a different transfer catheter (Jansen–Anderson) was used. It was shown that the choice of transfer catheter could influence the pregnancy rate following embryo transfer (Wisanto et al., 1989Go; Gonen et al., 1991Go). It remains to be proven that ultrasound can improve pregnancy rate in embryo transfer using an Edwards–Wallace catheter instead of a Jansen–Anderson catheter.

In our study, a dummy transfer was performed immediately before the real transfer and the embryos were transferred to the low uterine cavity 6 cm from the external os. A mock dummy transfer was shown in a randomized trial to improve pregnancy rate (Mansour et al., 1990Go). It has been performed as an interval procedure in most centres and the use of mock embryo transfer immediately before the real transfer was reported in a non-randomized study to be an acceptable practice and resulted in an implantation rate of 20.6% (Sharif et al., 1995Go). It has been our policy to perform a dummy transfer immediately before the real transfer for the last 10 years. Dummy transfers were not done in the other studies that showed improvement in pregnancy rates with ultrasound guidance. Perhaps the benefit from ultrasound guidance was less if a dummy transfer was performed immediately before the real transfer.

It was suggested that embryo transfer to the low uterine cavity resulted in a higher pregnancy rate than transfer to near the fundus (Waterstone et al., 1991Go). The embryos were transferred to the uterine cavity 6 cm from the external os and the chance of the transfer catheter touching the fundus was low, even in the clinical touch group. In the previous studies that showed an improvement in pregnancy rate using ultrasound guidance, the embryos were placed 0.5–1.5 cm from the fundus (Coroleu et al., 2000Go; Wood et al., 2000Go). The chance of touching the fundus was higher and it was possible that ultrasound guidance could prevent it. It was suggested that contact between the catheter and the uterine fundus could stimulate strong fundo-cervical contractions and random uterine waves and this might affect the pregnancy rate (Lesny et al., 1999Go).

One of the possible confounding factors in this study was that the subjects in the ultrasound-guided group had a full bladder during embryo transfer whereas the subject in the clinical touch group might not. This confounding factor also existed in the other studies using transabdominal ultrasound scan (Kan et al., 1999Go; Coroleu et al., 2000Go; Wood et al., 2000Go). Many centres asked the patients to attend for the embryo transfer with a full bladder. It was believed that filling up the bladder could straighten the angle between the internal os and the uterus, as ~80% of the uterus is anteverted (Sharif et al., 1995Go). This would make the embryo transfer easier and might improve the pregnancy rate. However, the evidence on the effect of a full bladder without the use of ultrasound guidance on the pregnancy rate was conflicting (Mitchell et al., 1989Go; Lewin et al., 1997Go). Moverover, it has been shown in a retrospective study that the use of transvaginal ultrasound without a full bladder could also improve the pregnancy rate (Lindheim et al., 1999Go). In the current study, the use of transabdominal ultrasound guidance could not decrease the chance of having a difficult transfer as the percentages of difficult transfer were the same in both groups.

No difference was found in the incidence of ectopic pregnancy between the clinical touch group (4.0%) and the ultrasound group (1.9%). It would be difficult to prove in a randomized trial whether ultrasound guidance could reduce the chance of ectopic pregnancy as the incidence of ectopic pregnancy is so low.

In conclusion, there was a significant improvement in implantation rate by using ultrasound guidance during embryo transfer. The extent of improvement in the pregnancy rate would depend on the specific techniques and methods of embryo transfer used in individual centres.


    Notes
 
1 To whom correspondence should be addressed at: 6/F, Professorial Block, Department of Obstetrics and Gynaecology, The University of Hong Kong, Queen Mary Hospital, 102, Pokfulam Road, Hong Kong SAR, China. E-mail: ostang{at}graduate.hku.hk Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Al-Shawaf, T., Dave, R., Harper, J. et al. (1993) Transfer of embryos into the uterus: how much do technical factors affect pregnancy rates? J. Assist. Reprod. Genet., 10, 31–36.[ISI][Medline]

Coetsier, T. and Dhont, M. (1991) Value of ultrasound-guided embryo transfer. Fertil. Steril., 56, 1005–1006.

Coroleu, B., Carreras, O., Veiga, A. et al. (2000) Embryo transfer under ultrasound guidance improves pregnancy rates after in-vitro fertilization. Hum. Reprod., 15, 616–620.[Abstract/Free Full Text]

Gonen, Y., Dirnfeld, M., Goldman, S. et al. (1991) Does the choice of catheter for embryo transfer influence the success rate of in-vitro fertilization? Hum. Reprod., 6, 1092–1094.[Abstract]

Hurley, V.A., Osborn, J.C., Leoni, M.A. and Leeton, J. (1991) Ultrasound-guided embryo transfer: a controlled trial. Fertil. Steril., 55, 559–562.[ISI][Medline]

Kan, A.K.S., Abdalla, H.I., Gafar, A.H. et al. (1999) Embryo transfer: ultrasound-guided versus clinical touch. Hum. Reprod., 14, 1259–1261.[Abstract/Free Full Text]

Lesny, P., Killick, S.R. and Tetlow, R.L. (1999) Embryo transfer and uterine junctional zone contraction. Hum. Reprod. Update, 5, 87–88.[Free Full Text]

Lewin, A., Schenker, J.G., Avrech, O. et al. (1997) The role of uterine straightening by passive bladder distension before embryo transfer in IVF cycles. J. Assist. Reprod. Genet., 14, 32–34.[ISI][Medline]

Lindheim, S.R., Cohen, M.A. and Sauer, M.V. (1999) Ultrasound guided embryo transfer significantly improves pregnancy rates in women undergoing oocyte donation. Int. J. Obstet. Gynaecol., 66, 281–284.

Mansour, R., Aboulghar, M. and Serour, G. (1990) Dummy embryo transfer: a technique that minimizes the problems of embryo transfer and improves the pregnancy rate in human in vitro fertilization. Fertil. Steril., 54, 678–681.[ISI][Medline]

Mitchell, J.D., Wardle, P.G., Foster, P.A. and Hull, M.G. (1989) Effect of bladder filling on embryo transfer. J. In Vitro Fertil. Embryo Transfer, 6, 263–265.[ISI][Medline]

Ng, E.H.Y., Yeung, W.S.B., Lau, E.Y.L. et al. (2000) High serum oestradiol levels in fresh IVF cycles do not impair implantation and pregnancy rates in subsequent FET cycles. Hum. Reprod., 15, 250–255.[Abstract/Free Full Text]

Sharif, K., Afnan, A. and Lenton, W. (1995) Mock embryo transfer with a full bladder immediately before the real transfer for in-vitro fertilization treatment: the Birmingham experience of 113 cases. Hum. Reprod., 10, 1715–1718.[Abstract]

Strickler, R.C., Christianson, C., Crane, J.P. et al. (1985) Ultrasound guidance for human embryo transfer. Fertil. Steril., 43, 54–61.[ISI][Medline]

Waterstone, J., Curson, R. and Parsons, J. (1991) Embryo transfer to low uterine cavity. Lancet, 337, 1413[ISI][Medline]

Wisanto, A., Camus, M., Janssens, R. et al. (1989) Performance of different embryo transfer catheters in a human in vitro fertilization program. Fertil. Steril., 52, 79–84.[ISI][Medline]

Wood, E.G., Batzer, F.R., Go, K.J. et al. (2000) Ultrasound-guided soft catheter embryo transfers will improve pregnancy rates in in-vitro fertilization. Hum. Reprod., 15, 107–112.[Abstract/Free Full Text]

Woolcott, R. and Stanger, J. (1997) Potentially important variables identified by transvaginal ultrasound-guided embryo transfer. Hum. Reprod., 12, 963–966.[ISI][Medline]

Submitted on March 5, 2001; accepted on July 23, 2001.