UvA-DARE (Digital Academic Repository) Optimizing the embryo transfer technique Abou-Setta, A.M. Link to publication

Transcription

1 UvA-DARE (Digital Academic Repository) Optimizing the embryo transfer technique Abou-Setta, A.M. Link to publication Citation for published version (APA): Abou-Setta, A. M. (2008). Optimizing the embryo transfer technique General rights It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons). Disclaimer/Complaints regulations If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible. UvA-DARE is a service provided by the library of the University of Amsterdam ( Download date: 21 Dec 2017

2

3 Optimizing the Embryo Transfer Technique Ahmed Mohamed Abou-Setta

4 Ahmed Abou-Setta, Amsterdam, 2008 Optimizing the Embryo Transfer Technique PhD Thesis, University of Amsterdam- with references- with summary in Dutch & Arabic Key words: embryo transfer, technique, catheter, ultrasound, evidencebased medicine, meta-analysis, pregnancy rate Cover: Picture of the Nile River in Luxor. In Egypt, the Nile represents life, hope, fertility and prosperity for the Egyptian people since the days of the Pharaohs. Printed by: Gildeprint drukkerijen BV

5 Optimizing the Embryo Transfer Technique ACADEMISCH PROEFSCHRIFT ter verkrijging van de graad van doctor aan de Universiteit van Amsterdam op gezag van de Rector Magnificus Prof.dr. D.C. van den Boom ten overstaan van een door het college voor promoties ingestelde commissie, in het openbaar te verdedigen in de Agnietenkapel op dinsdag 1 april 2008, te uur door Ahmed Mohamed Abou-Setta Geboren te Giza, Egypte

6 Promotiecommissie Promotor: Co-promotor Overige leden: Prof.dr. F. van der Veen Dr. H.G. Al-Inany Prof.dr. M.A. Aboulghar Prof.dr. P. Braude Prof.dr. M.J. Heineman Prof.dr. J.A.M. Kremer Prof.dr. N.S. Macklon Dr. P.G.A. Hompes Faculteit der Geneeskunde

7 Contents Page Chapter 1 Introduction 1 Timing of the embryo transfer Chapter 2 Aboulghar MM, Aboulghar MA, Mansour RT, Serour GI, Amin YM, Abou-Setta AM. Pregnancy rate is not improved by delaying embryo transfer from days 2 to 3. Eur J Obstet Gynecol Reprod Biol. 2003;107: The embryo transfer procedure Chapter 3 Eskandar M, Abou-Setta AM, El-Amin M, Almushait MA, Sobande AA. Removal of cervical mucus prior to embryo transfer improves pregnancy rates in women undergoing assisted reproduction. Reprod Biomed Online. 2007;14: Chapter 4 Abou-Setta AM. Effect of passive uterine straightening during embryo transfer: a systematic review and meta-analysis. Acta Obstet Gynaecol Scand. 2007;86: Chapter 5 Saldeen P, Abou-Setta AM, Bergh T, Sundström P, Holte J. A prospective randomized controlled trial comparing two embryo transfer catheters in an ART program. Fertil Steril. 2007;Nov 28 [Epub ahead of print]. 59 Chapter 6 Aboulfotouh I, Abou-Setta AM, Khattab S, Mohsen IA, Askalani A. Firm versus Soft Embryo Transfer Catheters under Ultrasound Guidance: Does Catheter Choice Really Influence the Pregnancy Rates. Fertil Steril. 2007;Jul 17 [Epub ahead of print]. 73

8 Page Chapter 7 Abou-Setta AM, Al-Inany HG, Mansour RT, Serour GI, Aboulghar MA. Soft versus firm embryo transfer catheters for assisted reproduction: a systematic review and meta-analysis. Hum Reprod. 2005;11: Chapter 8 Abou-Setta AM. Air-fluid versus fluid-only models of embryo catheter loading: a systematic review and meta-analysis. Reprod Biomed Online. 2007;14: Chapter 9 Abou-Setta AM. What is the best site for embryo deposition? A systemic review and meta-analysis. Reprod Biomed Online. 2007;14: Chapter 10 Eskandar MA, Abou-Setta AM, Almushait MA, El-Amin M, Mohmad SE. Ultrasound-guidance during embryo transfer: a prospective, single operator, randomized, controlled trial. Fertil Steril. (In-Press). 149 Chapter 11 Abou-Setta AM, Mansour RT, Al-Inany HG, Aboulghar M, Serour GI, Aboulghar MA. Among women undergoing embryo transfer, is the probability of pregnancy and live birth improved with ultrasound-guidance than with clinical touch alone? a systemic review and meta-analysis of prospective randomized trials. Fertil Steril. 2007;88: Chapter 12 Abou-Setta AM, Sharkawy S, Shahed M, Amer M. Difficult embryo transfers and the presence of blood on the embryo transfer catheter negatively affect clinical pregnancy rates. (Submitted) 193 Chapter 13 Abou-Setta AM, Mansour RT, Al-Inany HG, Serour GI, Aboulghar MA, El-Wassify M. Difficult embryo transfer: the impact of propofol anesthesia. Middle East Fertil Soc J. 2007;12:

9 Page Chapter 14 General Discussion 219 Chapter 15 Summary 229 Appendices 240 Samenvatting (الملخص العربى) Arabic summary Dankwoord (Acknowledgements) 260 Curriculum Vitae and other publications 261

10

11 Chapter I Introduction

12 Assisted reproductive technology (ART) is a fast-paced and everchanging field in medicine. All parts of the treatment cycle are evolving towards more evidence-based procedures, resulting in increasing success rates. Almost all aspects of the IVF procedure (e.g. ovarian stimulation protocols, laboratory techniques, techniques for embryo transfer and luteal phase support) have been evaluated and optimized through increased effectiveness and efficiency of each individual step. Embryo transfer is the last phase of a complementing process involving ovarian stimulation, in vitro fertilization and finally intrauterine placement of the generated embryos. This step is considered to be an important one among all those involved in assisted fertilization procedures since any errors at this stage or poor technique negate all previous work. Historically, little attention has been paid to the embryo transfer procedure. This fact is reflected by the dearth of scientific publications regarding embryo transfer compared with other aspects of IVF (e.g. ovarian stimulation), and also the reluctance of physicians to modify their own personal habits to encompass a more evidence-based approach. Physicians, too, often underestimate the importance of the embryo transfer technique, being an apparently simple manoeuvre. Most inexperienced clinicians do not consider inserting a catheter through the uterine cervix and ejecting embryo-containing fluid to be a difficult task. Be that as it may, multiple variables have been identified that might influence the success of the procedure (e.g. individual technical training and performance, catheter type and model, ultrasound-guidance and minimization of cervical/ endometrial trauma during transfer). Over the years several modifications to the embryo transfer method have been introducted since it was first described by Steptoe and Edwards (1). Each has built on the available technical developments and previous experience in order to achieve an effective transfer. In general, the most used approach is through the natural vaginal-cervical-uterine orifice, the transcervical intrauterine transfer, with or without ultrasound-guidance. Other techniques, such as ultrasound-guided transmyometrial transfer, gamete intra-fallopian transfer (GIFT), zygote intra-fallopian transfer (ZIFT), and embryo intra-fallopian transfer (EIFT) are more surgically invase and have been described, but are nowadays hardly used. 2

13 However, until not long ago, most clinicians relied heavily on prior experience and personal preference when performing embryo transfer, even though the success rate after embryo transfer has been demonstrated to show marked variability among different IVF programmes, and also among physicians within the same programme (2-5). Recently it was shown that pregnancy rates are strongly affected by the embryo transfer technique (6). This has fortunately been reflected in an increased interest and awareness among clinicians and some evidence-based guidelines have been provided by medical organizations (5). Today, most patients who undergo IVF/ ICSI will reach the embryo transfer stage with good quality embryos available for replacement. It is this development that emphasizes the clinical importance of an optimal embryo transfer. There are several issues regarding the embryo technique that are still debated in the literature. They can easily be distinguished into the timing of the embryo transfer and aspects related to the procedure itself. Re timing of the embryo transfer: In the early days of IVF, Edwards et al. performed embryo transfer (ET) in the 8 16-cell stage (i.e. 3 4 days after egg collection) (1). Recently, variations to this timing have ranged from as little as 2, to as many as 7 days post-oocyte pick-up. In view of these discrepancies, we conducted a prospective, controlled trial to compare the IVF outcomes between days 2 and 3 transfer to determine if a change is policy is warranted. Re the embryo transfer procedure: In general, the transfer procedure may be arbitrarily divided into four distinct sections: (i) preparation prior to embryo transfer (e.g. dummy embryo transfer, cervical preparation, uterine position); (ii) technical aspects related to the embryo transfer catheter (e.g. catheter type and catheter loading); (iii) the embryo transfer procedure (e.g. the site of embryo deposition within the uterus and techniques to assist with the accurate placement of the embryo within the uterus); and (iv) posttransfer aspects (e.g. expulsion of fluid/embryos from the cervix after embryo transfer and bed rest following embryo transfer) (7). 3

14 Ad (i) preparation prior to embryo transfer. Data on preparation prior to the actual embryo transfer are conflicting. Techniques involved include: (i) performing a trial (e.g. dummy) embryo transfer prior to the actual transfer to determine the length and direction of the cervico-uterine axis, the cervico-uterine angle and identifying possible difficulty in traversing the cervix (8); (ii) aspiration of the cervical mucus prior to transfer to remove the natural cervical mucus, with or without irrigating the endometrial cavity to remove possible contaminants; (iii) passively widening the utero-cervical angle by filling the urinary bladder. We conducted a prospective, controlled trial to determine whether removal of cervical mucus was beneficial. In addition, we systematically reviewed the available evidence on the use of passive uterine straightening during embryo transfer. Ad (ii) technical aspects related to the embryo transfer catheter. Several embryo transfer catheters are commercially available. All are mainly composed of non-toxic plastics and/or metal, but vary in length, calibre, location of the distal port (end- or side-loading), and degree of stiffness and malleability. They can be subdivided by the material they are made off (i.e. metal, hard or soft plastics) and whether they are equipped with, or without, an introducing cannula that facilitates the transfer procedure. In general, soft catheters have been associated with better pregnancy rates, but also with a higher incidence of failed and difficult transfers. To determine the most suitable catheter for embryo transfer in a clinical setting, we conducted a multicenter, randomized controlled trial comparing two commercially available soft transfer catheters. In addition, we performed a retrospective comparison of soft versus firm catheters under ultrasound guidance. Finally we systematically reviewed the current evidence comparing clinical pregnancy rates achieved with soft versus firm embryo transfer catheters. Catheter loading, ie the use of air brackets around the embryocontaining medium has been suggested to be beneficial to the success of the embryo transfer. By these air brackets the embryos would be protected from the cervical mucus and from accidental discharge before entering the endometrial cavity. On the other hand, it has also been suggested that even a small amount of air in the uterus could be a non 4

15 physiological factor with a deleterious effect on the embryos and implantation (9). Accordingly, we systematically reviewed the literature to summarize the evidence on the use of air brackets during embryo transfer. (iii) the embryo transfer procedure. With regards the embryo transfer procedure itself, the influence of the depth of embryo replacement into the uterine cavity has been postulated as being one of the most important factors to the success of an IVF treatment cycle (10). Traditionally, most IVF programmes have relied on the clinician s clinical touch for the placing of the transfer catheter within the uterine cavity at a point near the fundus (3, 11). At present, the best site of embryo deposition is still not clear and remains highly debated. Therefore, we performed a systematic review of the literature to determine the best site of embryo deposition. Another important aspect of the embryo transfer technique is whether ultrasound guidance is able to assist with accurate placement of the embryo within the uterus. More than twenty-years after the first reports on ultrasound guidance during the blind embryo transfer procedure, the routine use of ultrasonography to guide the intrauterine embryo transfer catheter placement are still highly debated (12, 13). This has been fuelled by the conflicting results of published clinical trials, with some concluding that ultrasound guidance improves clinical pregnancy and implantation rates, while others reporting no such improvement in their results. We therefore compared the use of ultrasound guided embryo transfer to the standard clinical touch embryo transfer performed by a single operator, and at the same time performed a systematic review of the literature to summarize the evidence on the use of ultrasound guidance during embryo transfer. Lastly, during the embryo transfer, the aim is to manipulate the catheter atraumatically through the cervix into the uterine cavity; without touching the fundus and minimizing trauma to the endometrium. Factors related to tissue trauma such as the presence of blood and/or mucus on the transfer catheter has been shown to decrease implantation and pregnancy rate (14, 15). 5

16 This led us to investigate the relationship between difficulties during transfer and presence of blood on the catheter tips following embryo transfer and the outcomes of IVF. In addition any clinical manoeuvre that increases the ease of transfer is highly welcomed. Therefore we investigated whether the use of Propofol anaesthesia improves the clinical outcomes of difficult embryo transfers. Propofol acts as both a hypnotic agent and exerts smooth muscle relaxation. It has been shown to be safe and useful in alleviating anxiety in women undergoing embryo transfer (16). 6

17 Aim of this thesis The aim of this thesis was to answer the following questions:- 1- Is the pregnancy rate improved by delaying embryo transfer from day two to day three? 2- Does removal of cervical mucus prior to embryo transfer improve the results? 3- What is the effect of passive uterine straightening during embryo transfer on pregnancy rate? 4- Is there a difference in clinical outcomes when two soft embryo transfer catheters are compared for embryo transfer? 5- Is there a difference in clinical outcomes between firm and soft embryo transfer catheters when embryo transfer is performed under ultrasound guidance? 6- Are soft embryo transfer catheters more favourable than firm catheters during embryo transfer? 7- What is the effect of using air to bracket embryos in the transfer catheter? 8- What is the best site in the uterine cavity for embryo replacement? 9- Does ultrasound-guidance during embryo transfer improve the outcomes when performed by a single operator? 10- Is the probability of pregnancy and live birth improved with ultrasound-guidance than with clinical touch alone? 11- What is the effect of blood found on the tips of the post-transfers on clinical outcomes? 12- Does the use of Propofol anaesthesia improve the clinical outcomes of difficult embryo transfers? 7

18 Outline of this thesis Chapter 2 provides results of a prospective, controlled, clinical trial comparing the clinical outcomes following embryo transfer on day two and day three. The primary outcome was the clinical pregnancy rate per woman. Chapter 3 provides results of a prospective, controlled, clinical trial comparing the clinical outcomes following cervical mucus aspiration compared with no aspiration prior to embryo transfer. The primary outcome was the clinical pregnancy rate per woman. Chapter 4 presents a systematic review and meta-analysis of randomized controlled trials that compare between the uses of passive uterine straightening during embryo transfer compared with no intervention. The studies were included and assessed according to the principles of the Cochrane Collaboration. Primary outcomes were livebirth, ongoing and clinical pregnancy rates. Chapter 5 presents the results of a randomized controlled trial that investigated the clinical outcomes following the use of two soft embryo transfer catheters during embryo transfer. Primary outcomes were the live-birth and clinical pregnancy rates per woman randomized. Chapter 6 presents the results of a retrospective analysis of using firm embryo transfer catheters compared with soft catheters under ultrasound-guidance. Primary outcome was the clinical pregnancy rate. Chapter 7 reports the results of a systematic review and meta-analysis of randomized controlled trials that compared soft embryo transfer catheters with firm catheters during embryo transfer. The studies were included and assessed according to the principles of the Cochrane Collaboration. The primary outcome measures used for this systematic review were implantation rate, clinical pregnancy rate and ongoing/takehome baby rate. Chapter 8 presents a systematic review and a meta-analysis of randomized controlled trials that compare between the air-fluid and the fluid-only models of embryo catheter loading. The studies were included 8

19 and assessed according to the principles of the Cochrane Collaboration. Primary outcomes were live-birth, ongoing and clinical pregnancy rates. Chapter 9 presents a systematic review and meta-analysis of randomized controlled trials that compare different sites of embryo placement in the uterine cavity. The studies were included and assessed according to the principles of the Cochrane Collaboration. Primary outcomes were live-birth, ongoing and clinical pregnancy rates. Chapter 10 presents the results of a randomized controlled trial that compared ultrasound with clinical touch methods of embryo catheter guidance performed by a single operator. Primary outcomes were the live-birth/ongoing pregnancy and clinical pregnancy rates per randomized woman. Chapter 11 reports on results of a systematic review and meta-analysis of randomized controlled trials that compared ultrasound with clinical touch methods of embryo catheter guidance. The studies were included and assessed according to the principles of the Cochrane Collaboration. Primary outcome measures were the live-birth, ongoing pregnancy, and clinical pregnancy rates. Chapter 12 reports on results of a retrospective analysis on a large cohort of women who underwent embryo transfer to determine whether the presence of difficulty during embryo transfer, the presence of blood and mucus on the post-transfer catheter and retained embryos affected the clinical outcomes. Primary outcome was clinical pregnancy rate. Chapter 13 presents a retrospective analysis to determine whether the use of Propofol anaesthesia improves the clinical outcomes when used in women presenting with difficulty during embryo transfer. The primary outcome measures for this study were the odds of a clinical pregnancy and embryo implantation following Propofol anesthesia compared to no anesthesia. Chapters 14 and 15 present a general discussion and the summary and conclusions of the preceding chapters. 9

20 References 1. Steptoe PC, Edwards RG. Birth after the reimplantation of a human embryo. Lancet Aug 12;2(8085): Karande VC, Morris R, Chapman C, Rinehart J, Gleicher N. Impact of the "physician factor" on pregnancy rates in a large assisted reproductive technology program: do too many cooks spoil the broth? Fertil Steril Jun;71(6): Hearns-Stokes RM, Miller BT, Scott L, Creuss D, Chakraborty PK, Segars JH. Pregnancy rates after embryo transfer depend on the provider at embryo transfer. Fertil Steril Jul;74(1): Salha OH, Lamb VK, Balen AH. A postal survey of embryo transfer practice in the UK. Hum Reprod Apr;16(4): National Institute for Clinical Excellence. Fertility assessment and treatment for people with fertility problems. February 2004, p RCOG Press, London UK. 6. Mansour RT, Aboulghar MA. Optimizing the embryo transfer technique. Hum Reprod May;17(5): Al-Inany HG, Abou-Setta AM, Garzo G. ET catheters for assisted reproduction. Cochrane Database Syst Rev. 2006; Issue 1:Art. No: CD DOI: / CD Mansour R, Aboulghar M, Serour G. Dummy embryo transfer: a technique that minimizes the problems of embryo transfer and improves the pregnancy rate in human in vitro fertilization. Fertil Steril Oct;54(4): Krampl E, Zegermacher G, Eichler C, Obruca A, Strohmer H, Feichtinger W. Air in the uterine cavity after embryo transfer. Fertil Steril Feb;63(2): Naaktgeboren, N., Dieben, S., Heijnsbroek, I. et al Embryo transfer, easier said than done. Abstracts of the 16th World Congress on Fertility and Sterility and 54th Annual Meeting of the American Society for Reproductive Medicine, San Francisco, CA, USA. S Schoolcraft WB, Surrey ES, Gardner DK. Embryo transfer: techniques and variables affecting success. Fertil Steril Nov;76(5): Strickler RC, Christianson C, Crane JP, Curato A, Knight AB, Yang V. Ultrasound guidance for human embryo transfer. Fertil Steril Jan;43(1): Leong M, Leung C, Tucker M, Wong C, Chan H. Ultrasound-assisted embryo transfer. J In Vitro Fert Embryo Transf Dec;3(6): Goudas VT, Hammitt DG, Damario MA, Session DR, Singh AP, Dumesic DA. Blood on the embryo transfer catheter is associated with decreased rates of embryo implantation and clinical pregnancy with the use of in vitro fertilization-embryo transfer. Fertil Steril Nov;70(5): Alvero R, Hearns-Stokes RM, Catherino WH, Leondires MP, Segars JH. The presence of blood in the transfer catheter negatively influences outcome at embryo transfer. Hum Reprod Sep;18(9): Al-Inany HG, Wasseef M, Aboulghar MA, Mansour RT, Serour GI, Abou-Setta AM. Embryo Transfer under Propofol Anaesthesia: The Impact on Implantation and Pregnancy Rate. MEFS J. 2003; 8(3):

21 Chapter 2 Pregnancy rate is not improved by delaying embryo transfer from days 2 to 3. Mona Aboulghar, Mohamed Aboulghar, Ragaa Mansour, Gamal Serour, Yehia Amin, Ahmed M. Abou-Setta European Journal of Obstetrics & Gynecology and Reproductive Biology 2003;107:176-9.

22 Abstract Objective: To compare the outcome of assisted reproduction in day 2 versus day three embryo transfer. Design: Prospective study. Participants: A total of 927 consecutive embryo transfers for IVF and ICSI cycles including 626 embryo transfers on day 2 and 301 on day 3. Intervention: IVF and ICSI. Outcome measure: Clinical pregnancy rate. Results: There is no significant difference in the pregnancy rate between ET on day 2 (50.9%) and ET on day 3 (50.5%). Conclusion: Embryo transfer could be done on days 2 or 3 according to the convenience of the patient and the medical team. Condensation: Embryo transfer could be done on days 2 or 3 according to the convenience of the medical team with similar results. 12

23 Introduction Since human IVF was established as an efficient mode of treatment for infertility, many steps of the IVF procedure became standardized. However the optimum timing of embryo transfer (ET) is still debatable. In the early days of IVF, Edwards et al. (1) performed embryo transfer (ET) in the 8 16-cell stage, i.e. 3 or 4 days after egg collection. Later, most of the centers worldwide performed ET on day 2 when embryos are in the 4-cell stage. Lately, ET at the blastocyst stage was introduced (2). It was reported that the first and the second cleavage divisions do not accurately predict the potential of future development and a large proportion of human embryos which arrest in vitro do so between the 4- and 8-cell stage (3). Based on this data, the idea of transferring embryos on day 3 should eliminate embryos which arrest at this sensitive stage and therefore allow selection of more viable embryos for transfer hoping that this might improve implantation and pregnancy rates (4). The objective of this study is to compare the IVF outcome between days 2 and 3 transfer. 13

24 Materials and Methods Patients This is a prospective study performed at the Egyptian IVF-ET center, in Cairo, Egypt during the period from January 1998 to June A total of 927 patients who reached the stage of embryo transfer are included in the study. This represents consecutive embryo transfers including only patients during their first trial, 39 years old or less using the long GnRH agonist protocol and using only ejaculate sperm. Patients with previous failed trials, patients above 39 years, patients who received short agonist protocol, antagonist protocols, clomiphene citrate or hmg only protocols were excluded from the study. Patients with very high or low body mass index, polycystic ovarian disease and general diseases not contradicting with performing IVF were not excluded from the study. All patients were treated by our standard GnRH long protocol, which was published before (5). In our IVF center, we work 6 days a week with Friday as a holiday. Patients who have oocyte retrieval on Saturday, Sunday, Monday and Tuesday had their ET after 48h. Patients who have retrieval on Wednesday or Thursday had ET after 72h to avoid egg retrieval and embryo transfer on Friday. The protocol was approved by IRB and our ethical committee. The power of the study was examined and it revealed that a sample size of 814 women provides 80% power and a two-sided significance level of 0.05 to test whether day 3 is equivalent or even superior to day 2 in an IVF/ET program. This sample size has adequate power to detect a difference of 5% in pregnancy rate/treated. Oocyte and embryo culture After retrieval, the oocytes were incubated in G1.2 media (Vitrolife, Motndalswaga, Goteborg) under mineral oil in tissue culture dishes (Falcon 3001) at 37 C and 5% CO 2 in air. ICSI procedure was performed about 2 3h after retrieval as described previously (6) after about 16 20h, the oocytes were checked for PN formation and the normally fertilized ones were transferred to fresh G media that was incubated overnight and no further changes of media was done till the day of embryo transfer. The embryos were cultured in groups of 3 4 in 50ml droplets of tissue culture media under mineral oil. On the day of ET, the best morphologically looking embryos were chosen. Selection of 14

25 the embryos for transfer was based on the number of blastomeres, absence of fragmentation and the most advanced stage of development. The best three embryos were transferred except in patients above the age of 37 with bad quality embryos where four embryos could sometimes be transferred. Two weeks after embryo transfer serum BhCG was measured and patients who had a positive pregnancy test were scheduled for ultrasound scan three weeks later. Clinical pregnancy was diagnosed by the presence of positive pregnancy test, ultrasound sac, echoes and pulsations. 15

26 Results The mean age of patients was 31.7 ± 5.6 years for day 2 transfer as compared to 31.9 ± 5.2 years for day 3 transfer. There was no significant difference in the age between the two groups. A total of 927 embryo transfers were included in the study on either day 2 (N = 626) or day 3 (N = 301). Causes of infertility in both groups are shown in Table 1. Number of oocytes retrieved, fertilization rate, mean number of embryos transferred, number of clinical pregnancies, pregnancy rate, implantation rate and multiple pregnancies are shown in Table 2. There was no significant difference in all parameters between days 2 and 3. There were four sets of triplets in day 2 ET and two sets of triplets in day 3 ET. The proportion of grades 1 three embryos was 27, 61 and 9 on day 2 ET versus 15, 71 and 12 on day 3 ET. The cleavage stage of embryos transferred on days 2 or 3 is shown in Fig. 1. According to the number of embryos available for transfer, the patients were further divided into two subgroups. Subgroup A, patients with a maximum of three embryos available for transfer. Subgroup B, patients with more than three embryos available for transfer (Table 3). There was no significant difference between the pregnancy rate in both subgroups between days 2 and 3 transfer. However, there was a significant higher PR in the subgroups with more than three embryos available for ET as compared with those who had less than three embryos available. Statistical analysis Statistical analysis of results was performed using the Student s t-test and the percentage data were compared using chi-square test. 16

27 Discussion Embryo transfer was performed either on day 2 or 3, until recently when blastocyst transfer was introduced as a possible option (2). From the available literature, it is not exactly clear if there is an advantage in delaying transfer to day 3 over day 2. Transfer of embryos to the uterus on day 3 after oocyte retrieval may be closer to the physiological time of arrival of embryo to the uterine cavity than transfer on day 2. Moreover, delaying embryo transfer would allow the selection of the most vital embryos for transfer (7). The present study has shown that there is no significant difference in the PR between ET on days 2 or 3. Although a randomized study would have been ideal, yet there is no possibility of selection bias in selecting the day of embryo transfer as patients who had egg retrieval on Wednesdays and Thursdays had their ET on day 3, and those who had egg retrieval on other days of the weeks had their ET done on day 2. Criteria for giving hcg was fixed all through the study when the lead follicle was 19mm in diameter in the presence of three or more follicles. This meant that the day of ET was allocated at random. Although the percentage of 4-cell stage embryos on day 2 ET was 83% as compared to the 55% 8-cell stage on day 3, this shows that embryos were arrested at 4- and 6-cell stage after extending the culture in vitro for 24h more (Fig. 1), yet this difference did not improve the pregnancy rate in day 3 over day 2 after we had the opportunity to exclude arrested embryos at 4- and 6-cell stage. Even in patients with few embryos (three or less) there was no difference in the pregnancy rate between days 2 and 3. Several retrospective studies have investigated this issue. Dawson et al. (4) in a retrospective study analyzed pregnancy rates in a series of patients comparing days 2 and 3 ET. There was no significant difference in the clinical PR between the two groups. The implantation rate as measured by the percentage of embryos developing to the fetal heart stage was significantly higher following transfer on day 3 (23%) than after day 2 (19%) suggesting that selection of viable embryos is improved on day 3. Furthermore, the embryos which gave rise to fetal sac significantly fewer miscarried before fetal heart stage 6% on day 3 as compared to 12% on day 2. In a retrospective study, Carrillo et al. (8) compared day 2 with day 3 ET and found out that the implantation and pregnancy rates were 17

28 significantly higher on day 3; 24 and 44% as compared to 13 and 26% on day 2. In this study only patients with large numbers of oocytes were included. In a large retrospective study, Huisman et al. (9) compared IVF results after day 2, 3 and 4 of ET. In general, the implantation and PR were similar in the three groups; however the implantation rate of 73 cavitating morula on day 4 was surprisingly higher (41%). The authors suggested that transfer on day 4 may give the ability to recognize embryos with very high implantation potential. Some prospective studies compared the pregnancy rates on days 2 and 3 ET. Laverge et al. (7) in a prospective randomized study found that delaying embryo transfer by 1 day did not improve the pregnancy rate. They found a statistically lower percentage of excellent and good quality embryos on day 3 compared to day 2. However, this did not affect the pregnancy rate possibly because they were able to select good quality embryos for transfer because only patients with al least seven normally fertilized oocytes were included in the study. In a prospective study of unselected group, patients were randomized for day 2 versus day 3 ET (10). Implantation rate was 15.8% for day 2 and 14.3 in day 3, and birth rate per ET was 18.5% for day 2 and 22.6% for day 3. There was no significant difference between both groups. Van Os et al. (11) in a prospective randomized study of day 2 versus day 3 ET, the mean number of embryos transferred was equal in the two groups. The PR per ET was not significantly different between both groups. However, they found a significantly higher percentage of clinical abortions after ET on day 3. This is contrary to what Edwards has reported earlier (12). Extending the culture period to beyond the time of expected activation of the embryonic genome might therefore optimize the selection of viable embryos for transfer (13). However, Extended culture may be compromised by the sub optimal embryo development in vitro compared to that in vivo. It has also been observed that a large proportion of human embryos which arrest in vitro do so between the 4- and 8-cell stage (3). Daunson et al. (4) reported that there was no change in embryo quality between days 2 and 3 as measured by the distribution of embryo grades. However, 16% of embryos had either arrested or became 18

29 developmentally retarded between days 2 and 3 and it was therefore possible to avoid these embryos at the time of ET. The probability of selecting embryos on day 2 which will subsequently arrest in culture was calculated to be between 2 and 3%. Selection on the basis of morphology and the rate of development will result in probably selecting embryos which have not arrested if ET was delayed to day 3. Ertzeid et al. (10), randomized all patients with at least one egg fertilized to days 2 or 3 ET. There was no difference in the implantation and live birth rate. The outcome might have been different if only patients with several embryos on day 2 were included rendering better selection of embryos on day 3. They also mentioned that the rate of embryo development appears to be a more reliable indication of embryo viability than morphology. They observed a higher frequency of clinical pregnancy resulting from early cleaved two-cell embryos. In conclusion, our data showed that although delay of in vitro culture of the embryo by 24h may eliminate arrested and poor quality embryos, yet ET on day 3 did not improve the pregnancy rate after IVF. This applied also to patients with three embryos or less, as well as patients with higher number of embryos. It is recommended that embryo transfer could be done on either days 2 or 3 depending upon the convenience of the patients and the medical team. Our center achieved similar pregnancy rate in days 2 and 3 ET without performing ET during the weekend. 19

30 References 1. Edwards RG, Steptoe PC, Purdy JM. Establishing full-term human pregnancies using cleaving embryos grown in vitro. Br. J. Obstet. Gynecol. 1980;87: Gardner DK, Vella P, Lane M, Wagley T, Schlenker T, Schoolcraft WB. Culture and transfer of human blastocyst increases implantation rates and reduces the need for multiple embryo transfer. Fertil. Steril. 1998;69: Bolton NV, Hawes SM, Taylor CT, Parsons JH. Development of spare human preimplantation embryos in vitro: an analysis of the correlations among gross morphology. J. In Vitro Fertil. Embryo Transfer. 1989;6: Dawson KJ, Cnaghan J, Ostera GR, Winston RML, Hardy K. Delaying transfer to the third day post-insemination, to select non-arrested embryos, increases development to the fetal heart stage. Hum. Reprod. 1995;10: Aboulghar MA, Mansour RT, Serour GI, Amin MY. The prognostic value of successful in vitro fertilization on subsequent trials. Hum. Reprod. 1994;9: Mansour RT, Aboulghar MA, Serour GI, Amin YM, Ramzi AM. The effect of sperm parameters on the outcome of intracytoplasmic sperm injection. Fertil. Steril. 1995;64: Laverge H, De Sutter P, Van Der Elst J, Dhont M. A prospective. Hum. Reprod. 2001;16: Carrillo AJ, Lane B, Pridham DD, Risch PP, Pool TB, Silverman IH, et al. Improved clinical outcomes for in vitro fertilization with delay of embryo transfer from 48 to 72 h after oocyte retrieval: use of glucose- and phospate-free media. Fertil. Steril. 1998;69: Huisman GJ, Alberda ATh, Leerentveld RA, Verhoeff A, Zeilmaker GH. A comparison of in vitro fertilization results after embryo transfer after 2, 3, and 4 days of embryo culture. 10. Ertzeid G, Dale PO, Tanbo T, Storeng R, Kjekshus E, Abyholm T. Clinical outcome of day 2 versus day three embryo transfer using serum-free culture media: a prospective randomized study. J. Assist. Reprod. Genet. 1999;16: Van Os HC, Janssen-Caspers HAB, Leerentveld RA, Scholtes MCW, Zeilmaker GH. Alberda Ath. Fertil. Steril. 1989;51: Edwards RG, Fishel SB, Cohen J, Fehilly CB, Purdy JM, Slater JM, et al. Factors influencing the success of in vitro fertilization for alleviating human infertility. J. In Vitro Fertil. Embryo Transfer. 1984;1:3 13. Braude P, Bolton V, Moore S. Human gene expression first occurs between the four- and eight-cell of preimplantation development. Nature. 1988;332:

31 Table 1. Characteristics of patients in days 2 and three embryo transfer Day 2 Day 3 Number of cycles Age 31.7 ± ± 5.2 Male factor 470 (75%) 220 (73%) Tubal 94 15% 51 17% Unexplained infertility 50 (8%) 21 (7%) Others 12 (2%) 9 (3%) No significant difference between all patient characteristics. Table 2. Outcome of IVF and ICSI on days 2 and three embryo transfer Day 2 Day 3 Number of patients Number of oocytes retrieved Mean number of oocytes 11.5 ± ± 3.4 Fertilization rate 62% 61% Mean number of embryos transferred 3.2 ± ± 1.2 Number of clinical pregnancies Clinical pregnancy rate per ET 50.90% 50.50% Multiple pregnancy rate 121 (39.8%), four triplets 46 (31.9%), two triplets Implantation rate 21.40% 20.60% No significant difference in all parameters between the two groups. 21

32 Table 3. Pregnancy outcome in subgroup A with a maximum of three embryos available and subgroup B with more than three embryos available No. of pregnancies (%) Day 2 (%) Day 3 (%) Subgroup A 39 (31) 19 (30.65) Subgroup B 265 (56.1) 97 (59.9) No significant difference in pregnancy in subgroups A and B between days 2 and 3 transfer. 22

33 Fig. 1. The number of blastomeres in days 2 and three embryo transfer. 90 Percentage Day 2 Day No. of Blastomeres

34 24

35 Chapter 3 Removal of cervical mucus prior to embryo transfer improves pregnancy rates in women undergoing assisted reproduction. Mamdoh A. Eskandar, Ahmed M. Abou-Setta, Mohamed El-Amin, Mona A. Almushait, Adekunle A. Sobande Reproductive Biomedicine Online. 2007;14:

36 Abstract The removal of cervical mucus during embryo transfer has been postulated to increase the pregnancy and implantation rates by not interfering with embryo implantation. Even so, this is a time-consuming procedure that may increase the incidence of difficult transfers by removing the naturally lubricant mucus. In addition, any cervical manipulations at the time of embryo transfer may cause unwarranted uterine contractions. In this prospective, controlled study, 286 women undergoing embryo transfer between January and May 2006 were divided into two groups according to whether the cervical mucus was scheduled to be aspirated (group A) or not (group B). The two groups were similar with regards to the demographics, cause of infertility, characteristics of ovarian stimulation and embryos transferred. Even so, the clinical pregnancy rate was significantly higher in group (A) than group (B) (OR = 2.18, 95% CI = ), although there were easier transfers in group (B) than group (A) (OR = 3.00, 95% CI = ). This demonstrates that even though embryo transfers were easier to perform when the cervical mucus was left in place, aspiration resulted in an increased chance of clinical pregnancy. Key words: embryo transfer, cervical mucus, aspiration, catheter, clinical trial 26

37 Introduction The majority of patients undergoing assisted reproduction through IVF/intracytoplasmic sperm injection (ICSI) will reach the transfer stage, but a small proportion of them will achieve a clinical pregnancy, an ongoing pregnancy or a live-birth (1, 2). The pregnancy rate following embryo transfer is generally dependent upon multiple factors, including embryo quality, endometrial receptivity and the technique of embryo transfer itself (3). Traditionally, unlike other aspects of assisted procreation which have been more thoroughly addressed by clinicians and researchers alike, the steps involved in the final stage of IVF, the transfer of embryos into the receptive uterus has mainly been left up to personal preferences. This fact is reflected by both the scarce volume of scientific publications regarding the embryo transfer technique as a whole, or its subunits in particular, compared with other aspects of IVF (e.g. ovulation induction). This is even more evident in the reluctance of physicians to modify their own personal habits to a more evidence-based approach. Even so, in recent trends among clinicians, more stress is being placed on optimizing and standardizing the embryo transfer protocol. Factors such as the use of a dummy embryo transfer (e.g. trial transfer) (4), ease of the procedure (5), catheter choice (6, 7), and ultrasound guidance (8, 9) have proven to improve the clinical outcomes. The influence of removing the cervical mucus prior to embryo transfer has been highly debated due to conflicting results in the medical literature. Some authors have demonstrated improved pregnancy rates, while others have shown no improvement (10-12). In addition, other studies supported the use of cervical mucus aspiration, but did not report the clinical pregnancy rates (13, 14). In light of this controversy, and the need to clearly identify the potential value of removing the cervical mucus, it was decided to perform a prospective, controlled trial to investigate this individual step in the embryo transfer procedure. 27

38 Materials and methods This prospective controlled trial was approved by the institutional review board. Two-hundred and eighty-six patients undergoing embryo transfer in the assisted reproduction unit between January and May 2006 were prospectively included. Patients were divided into two groups: group (A), which consisted of patients having the cervical mucus aspirated before the embryo transfer, and group (B), which consisted of patients not having the cervical mucus aspirated. At the time of embryo transfer, patients in group (A) had the cervical mucus removed by a commercially available catheter designed for the removal of cervical mucus (9.3 Fr/25 cm) (Aspiracath, J-ASP ; Cook Women s Health, USA). The cervix was not flushed, and the mucus was aspirated using the catheter introduced up to about 2 cm from the external cervical os. Patients in group (B) did not have the cervical mucus removed prior to embryo transfer. All other aspects of the ovarian stimulation, oocyte retrieval, embryo transfer and luteal phase support protocols were similar between the two groups. Moreover, all embryo transfers were performed on day 3 by a single physician using a standardized technique. In brief, ovarian stimulation, oocyte retrieval and luteal phase support were performed in accordance with the standard protocol of the department. Women were down-regulated using a gonadotrophinreleasing hormone agonist (GnRH agonist) (Decapeptyl; Ferring NV, Belgium) protocol, followed by ovarian stimulation using recombinant FSH (rfsh, Puregon; NV Organon, Oss, The Netherlands) and/or human menopausal gonadotrophin (Menogon; Ferring NV, Belgium) till the day of human chorionic gonadotrophin (HCG) administration. When the leading follicle reached ~18 mm in diameter, 10,000 IU of HCG (Pergonyl; NV Organon) was given intramuscularly, and oocyte retrieval was performed h later. Embryo transfer was performed using a soft Edward-Wallace or Cook catheter connected to a tuberculin syringe. A complete column of fluid was used (e.g. no air bubbles) to avoid the possibility of artificially introducing air into the uterine cavity. In addition, the catheter was held with its tip slightly downwards to prevent embryos from travelling through the liquid column to the end connected to the syringe. It is important to note that although the vast majority of patients in each group were less than 35 years old, an average of about three 28

39 embryos were transferred since, as a referral centre, most of the patients had previous failed trials in different centres. Therefore, it was decided to transfer an average of three embryos for such patients according to the centre s protocol. It is also important to note that a trial transfer was not performed prior to the embryo transfer. This was because the use of a trial transfer is considered to be empirical, and there are limited clinical trials on the beneficial effect of using a trial, or dummy, transfer. As a standard protocol in the unit, ultrasound-guided transcervical intrauterine embryo transfer is used exclusively. The embryos are deposited ~1 cm from the uterine fundus. Then the catheter is extracted and examined for retained embryos, blood and/or mucus by the embryologist under a stereomicroscope. Luteal phase support is provided in the form of daily progesterone vaginal suppositories t.i.d. (Cycologest 400 mg; Hoechst Roussel Limited, UK). The primary outcome measure for this trial was the clinical pregnancy rate per woman. Clinical pregnancy was defined by the presence of a positive β-hcg subunit measurement 2 weeks post-transfer and a clinically viable gestational sac with fetal heart pulsation on ultrasound 3 weeks later. In addition, embryo implantation rates, the incidences of difficult transfers and the presence of retained embryos, blood and/or mucus on the catheter tip was evaluated. Difficult transfers were defined as difficulties in placing the catheter inside the uterine cavity due to position of the uterus in relation to the cervical canal, cervical stenosis, or if embryo transfer took more than 5 min. Statistical analysis was performed according to the intention to treat principle. All analyses of significance were two-sided and tested at the 5% level; values of P < 0.05 were considered to indicate significant differences. Continuous variables were tested if they presented normal distribution using the F-test. The results of the two groups were compared using the t-test or Mann-Whitney U-test for parametric and non-parametric data respectively. Qualitative variables were compared with the use of the chi-squared test with Yates correction or Fisher s exact test, when necessary, and the 95% confidence intervals (95% CI) using the Woolf (logit) approximation. Odds ratios (OR) and 95% confidence intervals (95% CI) were calculated to examine the odds of improving clinical outcomes. Clinical and demographic data are also presented as mean (± SD) or as frequency distribution for simplicity. 29

40 Statistical analysis was performed using the computer statistical package Stats Direct (Stats Direct Ltd, UK). 30

41 Results The cause of infertility was similar in both groups, mainly being male factor (Table 1). In addition, the two groups were similar with regard to other demographics and cycle characteristics. There was no significant difference with regard to patient age, period of infertility, and day 3 FSH concentrations (Table 2). In addition, there were no significant differences in the numbers of gonadotrophin ampoules, days of stimulation, and numbers of oocytes retrieved, MII oocytes, injected oocytes, fertilized oocytes, embryos produced, embryo quality, and numbers of embryos transferred (Table 2). Finally, there were no significant differences in the catheter choice or the volume of injected culture media in the two groups (data not presented). With regard to the primary outcome measure, clinical pregnancy rates, there was a significantly higher number of clinical pregnancies in group (A) (63/143) than group (B) (38/143) (P = 0.003; odds ratio = 2.18, 95% CI = ). In contrast, there was a significantly higher incidence of easy transfers in favour of group (B) (Table 3, OR = 3.00, 95% CI = ). The implantation rates were similar in both groups (13.80 versus 13.38%). With regard to the post-transfer examination of the catheter tips, there were no significant differences in the presence of blood (in and/or on the catheter tip), mucus or blood and mucus on the tip of the embryo transfer catheter (Table 3), even though, ironically, there were more retained embryos in the mucus aspiration group than in the noaspiration group (Table 3). The results of this prospective clinical trial with 143 patients in each treatment arm demonstrated a statistically significant (P = 0.003) absolute difference of 17.48% between the two groups with regard to the clinical pregnancy rate. With this absolute difference, 129 women would be needed in each arm (with an alpha = 0.05) to have a power of 80%, therefore supporting the results. In addition, this difference relates to a number needed to treat of six (95% CI = 4-16). 31

42 Discussion The embryo transfer is the final stage of the IVF cycle. It is also the area in which clinical manipulations can directly alter the outcomes of the IVF cycle, and has shown marked variability both among different IVF programmes and physicians in the same programme (15, 16). During the embryo transfer, the aim is to manipulate the catheter atraumatically through the cervix into the uterine cavity, without touching the fundus and minimizing trauma to the endometrium (17). Physicians too often underestimate the importance of the embryo transfer technique, regarding it as an apparently simple manoeuvre. Most inexperienced clinicians do not consider inserting a catheter through the uterine cervix and ejecting embryo-contained fluid to be a difficult task, especially since many gynaecologists today perform intrauterine insemination (IUI) in a private clinic setting. Be that as it may, it has been shown recently that physician attitudes toward the embryo transfer technique are positively changing (18, 19). Recently, the techniques and variables affecting the success of embryo transfer have attracted more attention. Today, in light of global trends such as single embryo transfer (SET), more stress has been placed on optimizing and standardizing the embryo transfer protocol than ever before. Although most patients who undergo assisted procreation, via IVF or ICSI, will reach the embryo transfer stage with good quality embryos available for replacement, embryo implantation remains the rate-limiting step in the success of this form of therapy. The aim should be to meticulously and accurately place embryos within the uterus, in order to allow for proper implantation and fetal development (20). In order to ascertain the importance of each step involved in the embryo transfer procedure, individual factors must be evaluated independently. The removal of cervical mucus prior to embryo transfer has been suggested to directly influence the embryo implantation rates. Nevertheless, this clinical query is not clearly answered in the literature. Cervical mucus has been postulated to interfere with proper embryo replacement. It has also been suggested that the presence of cervical mucous can prevent the embryos from leaving the catheter by acting as a 'plug' at the catheter tip (3). In addition, transferred embryos may stick to the cervical mucus around the catheter and be dragged from their original site of deposition during the withdrawal of the catheter. 32

43 Moreover, the mucus may also interfere with implantation if pushed or injected into the uterine cavity. This has even led some authors to test the efficacy of endometrial flushing to remove any excessive mucus that may prevent implantation (21). Although the theory of the cervical mucus acting as a condom around the transfer catheter is possible, leading to retained embryos, this is unlikely to decrease the chances of treatment success since studies have shown that when the embryos are immediately transferred back into the endometrial cavity in a second attempt, the implantation and pregnancy rates are not reduced (14, 22). In addition, if the cervical mucus is dragged into the endometrial cavity with the transfer catheter, cervical canal mucous may entangle the embryos, interfere with implantation and increase the risk of cervical expulsion of the embryos post-transfer (13, 23). Another benefit of cervical cleaning and mucous removal is to reduce the risk of bacterial contamination of the catheter and endometrial cavity. Contamination of the catheter tip with micro-organisms such as streptococci (groups B and D), E. coli, staphylococci, mycoplasma and ureaplasma has been shown to reduce implantation and pregnancy rates by 40-60% (24-28), even when prophylactic antibiotics have been used. The hypothetical, but ineffective, role of probiotics in this situation would be to positively support the presence of the natural vaginal flora (e.g. lactobacilli) (29). Cleaning the cervical canal can therefore play a beneficial role by modifying the cervical micro-environment and reducing the risk of introducing pathogens into the endometrial cavity. As with any procedure, there are also negative aspects, and cervical mucus aspiration is no different. It is true that aspiration of the cervical mucus prior to embryo transfer only takes a few minutes, but in a busy IVF clinic it could accumulate to be time-consuming. This is due to the time taken to perform the procedure plus an additional period of time to allow the uterus to become quiescent again. It is generally preferable to have at least 5-10 min between the aspiration and the actual embryo transfer. Of course, this is an empirical approach, but since the extra wait does not do any harm to the parties involved, but could prove a beneficial factor, it has been decided to use it. Now in a busy IVF unit, when you add up the extra 5-10 min per embryo transfer, this could amount to a considerable amount of time invested in this procedure alone each day. 33

44 Mansour et al. (13) demonstrated that removing the cervical mucus before a methylene blue dummy transfer significantly reduced the extrusion of the dye. In addition, Nabi et al. (14) demonstrated that embryos were significantly more likely to be retained when the embryo catheter was contaminated with mucus (3.3 versus 17.8%, P = ). Consequently, the removal of the cervical mucus prior to embryo transfer has been claimed to improve the pregnancy and implantation rates, but so far as is known, there have been only a limited number of randomized, controlled trials on the routine aspiration of the mucus prior to embryo transfer (11, 12). It is important to note also that they were all published as conference abstracts, and not in peer-reviewed journals. McNamee et al. (10) performed a retrospective study to determine the effect of vigorous cervical irrigation prior to embryo transfer. The pregnancy rate following embryo transfer by five physicians (two who irrigated the cervix and three who did not) were compared. They demonstrated a significantly higher pregnancy (60 versus 32%) and ongoing pregnancy/delivery rates (44 versus 24%) in patients who had the cervical mucus removed. Glass et al. (11) randomized 253 patients to mucus aspiration during a mock embryo transfer or a control group. Randomization was performed using sealed envelopes. Before the aspiration, the cervical canal was flushed intermittently with moderate force. If mucus or blood remained in the canal, a second irrigation with the trial catheter was performed. With regard to the clinical pregnancy rate, they noted that there was a trend towards significance in the control group. In a double-blind randomized, controlled trial, performed on 424 patients, Visschers et al. (12) suggested that removal of cervical mucus prior to embryo transfer does not improve pregnancy rates. Even so, it is important to note that the patients in the cervical removal group underwent meticulous removal of cervical mucus prior to embryo transfer by means of a cervical brush. It is unclear if this brush increases endometrial contractility, as has been proven to occur with cervical and endometrial manipulations. In this prospective, controlled trial, it was possible to demonstrate that the gentle removal of the cervical mucus has a beneficial effect on embryo implantation and clinical pregnancy rates. A significantly higher percentage of women undergoing cervical mucus removal prior to 34

45 embryo transfer have a clinical pregnancy (P = 0.003). However, there was also a greater number of patients in the cervical removal group with difficult transfers. This relationship points to the natural lubricant effect of the cervical mucus. It is important to note that no natural or synthetic lubricant is currently advised for use on the embryo transfer catheter in difficult cases, and this may be an issue for further research in the future. Moreover, logistic regression showed a trend towards significance (P = ) for the association between a difficult transfer and the presence of retained embryos. The actual reasons for the association between difficult embryo transfers and the presence of retained embryos is not clear, but may be due to more cervical bleeding and contamination of the catheter tip with blood in cases of difficult transfer. This may form a barrier to proper delivery of the embryos in the uterine cavity. In conclusion, cervical aspiration prior to embryo transfer should be performed routinely in all patients undergoing embryo transfer through the cervical route. Even so, randomized controlled studies with adequate sample sizes should be performed to confirm these findings, and to determine if there is a true relationship between the presence of a difficult embryo transfer and the presence of retained embryos. 35

46 References 1. Edwards RG 1995 Clinical approaches to increasing uterine receptivity during human implantation. Human Reproduction 10, Adamson GD, de Mouzon J, Lancaster P, Nygren KG, Sullivan E, Zegers- Hochschild F, International Committee for Monitoring Assisted Reproductive Technology 2000 World collaborative report on in vitro fertilization. Fertility and Sterility 85, Mansour RT, Aboulghar MA 2002 Optimizing the embryo transfer technique. Human Reproduction 17, Mansour R, Aboulghar M, 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. Fertility and Sterility 54, Schoolcraft WB, Surrey ES, Gardner DK 2001 Embryo transfer: techniques and variables affecting success. Fertility and Sterility 76, Abou-Setta AM, Al-Inany HG, Mansour RT et al Soft versus firm embryo transfer catheters for assisted reproduction: a systematic review and metaanalysis. Human Reproduction 20, Abou-Setta AM 2006 Firm embryo transfer catheters for assisted reproduction: a systematic review and meta-analysis using direct and adjusted indirect comparisons. Reproductive BioMedicine Online 12, Sallam HN, Sadek SS 2003 Ultrasound-guided embryo transfer: a meta-analysis of randomized controlled trials. Fertility and Sterility 80, Abou-Setta AM, Mansour RT, Al-Inany HG et al Among women undergoing embryo transfer, is the probability of pregnancy and live birth improved with ultrasound-guidance than with clinical touch alone? A systemic review and metaanalysis of prospective randomized trials. Fertility and Sterility 88: McNamee M, et al Signifi cant increase in pregnancy rate achieved by vigorous irrigation of endocervical mucus prior to embryo transfer with the Wallace Catheter in an IVF ET programme. Fertility and Sterility; 67 (Suppl 1), Glass KB, Green CA, Fluker MR 2000 Multicenter randomized controlled trial of cervical irrigation at the time of embryo transfer [abstract no. O-085). Fertility and Sterility 74 (suppl. 1), S Visschers BAJT, Bots RSGM, Mol BW, Van Dessel HJHM 2006 Removal of cervical mucus prior to embryo transfer does not improve pregnancy rates in IVF/ICSI. Human Reproduction 21 (Suppl. 1), i Mansour RT, Aboulghar MA, Serour GI, Amin YM 1994 Dummy embryo transfer using methylene blue dye. Human Reproduction 9, Nabi A, Awonuga A, Birch H et al Multiple attempts at embryo transfer: does this affect in-vitro fertilisation treatment outcome? Human Reproduction 12, Karande VC, Morris R, Chapman C et al Impact of the 'physician factor' on pregnancy rates in a large assisted reproductive technology program: do too many cooks spoil the broth? Fertility and Sterility 71,

47 16. Hearns-Stokes RM, Miller BT, Scott L et al Pregnancy rates after embryo transfer depend on the provider at embryo transfer. Fertility and Sterility 74, Sallam HN 2005 Embryo transfer: factors involved in optimizing the success. Current Opinion in Obstetrics and Gynecology 17, Kovacs GT 1999 What factors are important for successful embryo transfer after in-vitro fertilization? Human Reproduction 14, Salha OH, Lamb VK, Balen AH 2001 A postal survey of embryo transfer practice in the UK. Human Reproduction 16, Coroleu B, Barri PN, Carreras O et al The influence of the depth of embryo replacement into the uterine cavity on implantation rates after IVF: a controlled, ultrasound-guided study. Human Reproduction 17, Berkkanoglu M, Isikoglu M, Seleker M, Ozgur K 2006 Flushing the endometrium prior to the embryo transfer does not affect the pregnancy rate. Reproductive BioMedicine Online 13, Lee HC, Seifer DB, Shelden RM 2004 Impact of retained embryos on the outcome of assisted reproductive technologies. Fertility and Sterility 82, Awonuga A, Nabi A, Govindbhai J et al Contamination of embryo transfer catheter and treatment outcome in IVF. Journal of Assisted Reproduction and Genetics 13, Egbase PE, Udo EE, Al-Sharhan M, Grudzinskas JG 1999 Prophylactic antibiotics and endocervical microbial inoculation of the endometrium at embryo transfer. Lancet Egbase PE, Al-Sharhan M, A-Othman S et al Incidence of microbial growth from the tip of the embryo transfer catheter after embryo transfer in relation to clinical pregnancy rate following IVF and embryo transfer. Human Reproduction 11, Fanchin R, Harmas A, Benaoudia F et al Microbial flora of the cervix assessed at the time of embryo transfer adversely affects IVF outcome. Fertility and Sterility 70, Moore DE, Soules MR, Klein NA et al Bacteria in the transfer catheter tip influence the live-birth rate after IVF. Fertility and Sterility 74, Salim R, Ben-Shlomo I, Olodner R et al Bacterial colonisation of the uterine cervix and success rate in assisted reproduction: results of a prospective study. Human Reproduction 17, Gilboa Y, Bar-Hava I, Fisch B et al Does intravaginal probiotic supplementation increase the pregnancy rate in IVF-embryo transfer cycles? Reproductive BioMedicine Online 11,

48 Table 1. Frequencies of indications leading to assisted reproduction. Mucus aspiration No aspiration Male factor 65 (45.5) 61 (42.7) Tubal factor 38 (26.6) 50 (35.0) Anovulation 28 (19.6) 16 (11.2) Unexplained infertility 9 (6.3) 15 (10.5) Mixed male and female factors 3 (2.1) 1 (0.7) Values in parentheses are percentages. There were no significant differences between groups. Table 2. Patient demographics and cycle characteristics between the two groups. Characteristic Mucus aspiration No aspiration Age ( 30 years)a 66 (46.15) 71 (49.65) Age (range years) a 41 (28.67) 40 (27.97) Age (range years) a 26 (18.18) 22 (15.38) Age (range years) a 10 (6.99) 10 (6.99) Period of infertility (years) 6.51 ± ± 3.25 Day 3 FSH concentration 5.19 ± ± 2.21 No. of ampoules ± ± 8.93 Days of stimulation ± ± 2.34 No. of oocytes retrieved ± ± 7.36 No. at metaphase II 9.59 ± ±

49 Characteristic Mucus aspiration No aspiration Injected 9.76 ± ± 6.25 Fertilized 6.87 ± ± 4.90 Good embryo g ± ± 3.86 Fair embryo g ± ± 1.75 Poor embryo g3, g ± ± 2.19 No. of embryos/transfer 3.50 ± ± 0.93 Values are means ± SD unless otherwise indicated. a Mean (%). Table 3. Primary and secondary outcome measures in the two groups. Mucus aspiration a No aspiration a OR 95% CI Clinical pregnancy rate b 63 (44.1) 38 (26.6) Implantation rate 69/500 (13.8) 63/471 (13.4) Rate of difficult embryo transfer b 14 (9.8) 5 (3.5) Embryo transfer catheter tip No blood or mucus present 119 (83.2) 109 (76.2) Presence of blood only 5 (3.5) 13 (9.1) Presence of mucus only 14 (9.8) 19 (13.3) Presence of both blood and mucus 5 (3.5) 2 (1.4) Retained embryos b 15 (10.5) 2 (1.4) a Values are numbers with percentages in parentheses. b Differences between groups are statistically significant. 39

50 40

51 Chapter 4 Effect of passive uterine straightening during embryo transfer: a systematic review and meta-analysis. Ahmed M. Abou-Setta Acta Obstetricia et Gynecologica Scandinavica. 2007;86:

52 Abstract Background: Part of the success of ultrasound-guided embryo transfer has been associated with the beneficial effect of uterine straightening by passive bladder distention. Even so, this has not been properly analysed in the literature. Methods: This is a systematic review and meta-analysis of prospective, randomised, controlled trials, comparing embryo transfer with a full versus empty bladder. Electronic (e.g. PubMed, EMBASE, Cochrane Library) and hand searches were performed to locate trials. Primary outcomes were live-birth, ongoing and clinical pregnancy rates. Secondary outcomes were rates of implantation, miscarriage, multiple and ectopic pregnancies, and retained embryos. Also, the ease of transfer, need for instrumental assistance, and presence of blood on the catheter tip were evaluated. Four studies were identified, of which 1 study was excluded. Meta-analysis was conducted with the Mantel- Haenszel method, utilising the fixed-effect model. Results: For the primary outcome measures, no data was available for the LBR rate. There was a significantly higher chance of an ongoing pregnancy [OR = 1.44 (95% CI = )] and clinical pregnancy [OR = 1.55 (95% CI = )] with a full bladder. For the secondary outcomes, there was a significantly greater incidence of difficulty, or need for instrumental assistance, with an empty bladder. Other outcome measures were not significantly different. Conclusion: There is evidence in the literature advising to fill the bladder prior to embryo transfer. Key words: Embryo transfer, bladder distention, meta-analysis, in vitro fertilisation 42

53 Background The embryo transfer (ET) procedure is considered to be the final and most crucial step in an in vitro fertilisation (IVF) cycle. When compared with other aspects of the IVF procedure, the ET is poorly efficient, with up to 85% of transferred embryos not implanting (1). In addition, it is estimated that about 80% of women undergoing IVF will reach the ET stage, but only a small portion will achieve pregnancy. The pregnancy rate after ET is dependent upon multiple factors, including the quality of the embryos, proper endometrial receptivity and development, and the technique by which embryos are transferred (2). Traditionally, little attention has been focused on the technique of ET. Moreover, no standard clinical protocol is accepted worldwide as an evidence-based protocol for ET. Most clinicians have been left to exercise personal preferences when performing the ET, unlike other aspect of IVF which have been more thoroughly addressed in the literature. This publication bias in the IVF literature has been mainly attributed to the lack of importance of the ET technique as a defining step in the success of the IVF procedure. Only recently have the techniques and variables affecting the success of ET attracted more interest. Today, in light of global trends, such as single ET, more emphasis has been placed on optimising and standardising the ET protocol. Numerous studies have shown that the success rates in different IVF clinics may be directly associated with the clinical techniques used during ET. Factors, such as ease of procedure (3), catheter choice (4, 5), and dummy ET (6) have proven to improve the clinical outcomes. During the ET, the aim is to manipulate the catheter atraumatically through the cervix into the uterine cavity, without touching the fundus and minimising trauma to the endometrium. Factors related to tissue trauma, such as the presence of blood and/or mucus on the transfer catheter, debatably, has been shown to decrease implantation and pregnancy rate (7, 8). Therefore, any clinical manoeuver that increases the ease of transfer is highly welcomed. To date, the possible beneficial role of passive uterine straightening by bladder distension is still a subject of debate. This has been fueled by conflicting results from published clinical trials, with some concluding that bladder distension improves the pregnancy rates following ET, while others reporting no such improvement in their results. Some authors 43

54 have even concluded that the success of ultrasound-guided ET is partly due to the use of bladder distention. Even so, many clinicians prefer to perform ultrasound-guided ET with a partially filled bladder, or even an empty bladder, in order to avoid immediate post-transfer micturation. In addition, vaginal ultrasound-guided ET is performed with an empty bladder. In light of this controversy, and the need to clearly identify the relative efficacy of this simple procedure, we decided to systematically locate, analyse, and review the current best available evidence for the use of bladder distension to passively straighten the utero-cervical angle during ET. 44

55 Materials and methods Criteria for considering studies for this review All published, unpublished, and ongoing randomised trials reporting data which compared outcomes for women undergoing ET through the cervical route following IVF, or ICSI, and rando-mised to either having a full bladder or an empty bladder during the time of transfer were sought in all languages. Types of outcome measures The primary outcome measures were the live-birth (LBR), ongoing pregnancy (OPR) and clinical pregnancy (CPR) rates. The secondary outcomes were the implantation, multiple pregnancies, ectopic pregnancy, and miscarriage rates. In addition, the incidences of difficult transfers or need for instrumental assistance during the transfer (e.g. stylette, tenaculum, dilatation, sounding) were evaluated. Lastly, the tips of the post-transfer catheters were evaluated for signs of cervical or endometrial trauma (e.g. presence of blood, mucus, or both), in addition to retained embryos. Search strategy for identification of studies Meticulous computerised searches (last performed July 2006) were conducted using MEDLINE (1966 to present), EMBASE (1980 to present), the Cochrane Central Register of Controlled Trials (CENTRAL) on the Cochrane Library Issue 3, 2006, the National Research Register (NRR), and the trial register of controlled trials ( Furthermore, the reference lists of all known primary studies, review articles, citation lists of relevant publications, abstracts of major scientific meetings (e.g. ESHRE and ASRM) and included studies were examined to identify additional relevant citations. Finally, ongoing and unpublished trials were sought by contacting experts in the field and commercial entities. Methods of the review A standardised data extraction form was developed and piloted for consistency and completeness. Trials were considered for inclusion, and trial data extracted. Data management and statistical analyses were conducted using the Review Manager (Rev-Man) 4.2 and Power and Sample Size Calculations (PS) statistical software packages. 45

56 Individual outcome data were included in the analysis if they met the pre-stated criteria. Where possible, data were extracted to allow for an intention-to-treat analysis - defined as including in the denominator all randomised cycles. If data from the trial reports were insufficient or missing, the inves-tigators of individual trials were contacted via for additional information, in order to perform analyses on an intentionto-treat basis. For the meta-analysis, the number of participants experiencing the event in each group of the trial was recorded. Heterogeneity of the included studies was determined using the χ 2 -test for heterogeneity. In addition, the quantity I 2 test was used to attempt quantifying any apparent inconsistency. The I 2 test describes the percentage of the variability in effect estimates that is due to heterogeneity rather than sampling error (chance) (9). An I 2 value >50% may be considered to represent substantial heterogeneity. For the meta-analysis, the number of participants experiencing the event in each group of the trial was recorded. Meta-analysis was undertaken using the Mantel-Haenszel method, utilising the fixed effect model, and the odds ratio (and 95% confidence intervals (CI)) evaluated. Search results A total of 4 prospective, randomised, controlled trials were identified (3 full-text manuscripts and 1 conference abstract). Subsequently, the conference abstract (10) was excluded for duplicate publication (e.g. publication as a conference abstract and full-text manuscript). The remaining 3 trials were included, the methodological quality of each trial assessed, and data extracted to allow for an intention-to-treat analysis. Description of included studies Mitchell et al. (11) conducted a prospective, randomised, controlled trial including 142 women undergoing 142 ET cycles. Patients were allocated into 1 of 2 groups by a random number table to have a partially full or empty bladder at the time of ET (Table I). Patients in the partially full bladder group were asked to empty their bladders, then drink 250 ml of water 1 h prior to the actual ET. Patients in the empty bladder group were asked to empty their bladders immediately before the transfer. The number of cycles in each arm was as follows: full bladder group 46

57 (66 ET cycles), and empty bladder group (76 cycles). None of the reported cycles were frozen embryo replacement (FER) or used oocyte donation (OD). Lewin et al. (12) conducted a prospective, randomised, controlled trial, including 796 women undergoing 796 ET cycles. Using alternate randomisation of days, the cycles were randomised into 2 groups: full bladder and empty bladder. Patients in the full bladder group were asked to urinate, and then drink 1,000 ml of water 1 h prior to the actual ET. At the time of ET, ultrasound was used to demonstrate bladder fullness. The number of cycles in each arm was as follows: full bladder group (411 ET cycles), and empty bladder group (385 cycles). None of the reported cycles were FER or used OD. Lorusso et al. (13) conducted a prospective, randomised, controlled trial including 171 women undergoing 171 ET cycles. Patients were allocated into 1 of 2 groups by a computer-generated randomisation table to have a full bladder or an empty bladder at the time of ET. The directions given to patients on the day of the ET were not available for review, but since it was mentioned that the transfer was carried out under ultrasound guidance, bladder distention must have been appraised at the time of the ET. The number of cycles in each arm was as follows: full bladder group (67 ET cycles), and empty bladder group (64 cycles). None of the reported cycles were FER or used OD. In addition, the authors mentioned that 40 patients undergoing ET using the clinical touch method was used a control group. This group was not included in the analysis due to the confounding factor of using clinical touch versus ultrasound-guided during the ET. 47

58 Results Primary outcome measures For the primary outcome measures, no data was available for the LBR rate. Even so, there was a significantly higher chance of an OPR and CPR with a full bladder than an empty bladder [103/478 versus 72/449 (OR = 1.44; 95% CI = )] and [148/544 versus 102/525 (OR = 1.55; 95% CI = )], respectively. Furthermore, when only the properly randomised trials were compared, there was no apparent differ-ence in the chance of an OPR [22/67 versus 21/64 (OR = 1.00; 95% CI = )], but a signifi-cantly increased chance of a CPR [136/478 versus 89/449 (OR = 1.61; 95% CI = )] with a full bladder than an empty bladder. However, it is important to mention that with regard to the OPR rate, there was only 1 study (13) included in the analysis. Secondary outcome measures For the secondary outcome measures, no data were retrievable for the multiple pregnancy and miscarriage rates. In addition, there was no significant difference between the implantation rates between the 2 groups [12/137 versus 18/158 (OR = 0.75; 95% CI = )]. Even so, there was a significantly greater incidence of difficulty [72/140 versus 36/133 (OR = 2.85; 95% CI = )], and need for instrumental assistance [47/64 versus 16/67 (OR = 8.81; 95% CI = )] during ET with an empty bladder than with a full bladder, respectively. Moreover, with regard to the ease of transfer, there was marked heterogeneity between the included studies (p = , χ 2 = 14.40, I 2 = 93.1%). The other outcome measures (e.g. implantation rate, presence of blood or embryos on the post-transfer catheter) were not significantly different (p > 0.05). 48

59 Discussion Although most patients who undergo assisted procreation, via IVF or ICSI, will reach the ET stage with good quality embryos available for replacement, embryo implantation remains the rate-limiting step in the success of this form of therapy. The aim of the ET procedure is to atraumatically and accurately place embryos within the uterus; in order to allow for proper implantation and fetal development. Studies have shown that different factors may be involved in a successful transfer. These include the experience of the physician (14), ET catheter choice (4, 5), the use of ultrasound guidance (15), the ease of the procedure (3), the presence or absence of blood on the catheter (7), and bacterial contamination of the catheter (16). In addition, other factors concerning ET that might affect the chance for an OPR have been identified, such as the use of cervical introducers or obturators (17), the value of resting after transfer (18), the position of embryo insertion in the uterus (19, 20), flushing of the cervical canal to remove mucus (21), microbiological factors in terms of the local flora (22), and retention of embryos in the catheter (23). In order to ascertain the importance of each step involved in the ET procedure, individual factors must be evaluated independently. Therefore, the ET procedure may be arbitrarily divided into four distinct sections: (1) preparation prior to ET (e.g. patient position, cervical preparation, uterine position and the dummy ET); (2) technical aspects related to the ET catheter (e.g. catheter type and catheter loading); (3) the ET procedure (e.g. the site of embryo deposition within the uterus and techniques to assist with the accurate placement of the embryo within the uterus); and (4) post-transfer aspects (e.g. expulsion of fluid/embryos from the cervix after ET and bed rest following ET) (24). Since it would be difficult to accurately compare several factors at the same time, it was decided to concentrate on 1 factor, the beneficial value of bladder distention during the ET procedure. Sundstrom et al. (25) were the first to identify the beneficial effects of uterine straightening by bladder distension. However, their study had a relatively small sample size (n = 14), and used a historical control. Since then, only a handful of clinical trials have tackled the same issue, therefore emphasising the need for a systematic review of the best available evidence in the literature. Systematic reviews and meta-analysis of randomised, controlled, trials 49

60 have proven to be the highest level of evidence in the hierarchy of medical knowledge. Even so, publication and search biases may confound the results of any systemic review, as studies showing positive results are more likely to be published (26, 27). Therefore, every effort has been made to avoid bias by searching a wide variety of databases, including Medline (PubMed), EMBASE, the Cochrane Library, with no language barriers, in addition to hand-searching the abstract books of major conferences (e.g. ASRM, ESHRE), reference lists of review articles and included trials. In addition, even though all the included studies in this systematic review were published trials in peer-reviewed journals, every attempt was made to locate abstracts from conference proceedings, unpublished trials, and currently ongoing trials. The objective was to minimise the chance of publication or selection bias in order to strengthen the validity of the results of the systematic review. Failure to identify trials reported in conference proceedings might affect the results or threaten the validity of a systematic review (28). Another important issue in clinical trials and systematic reviews are sample sizes. It is theorised that smaller studies might not have sufficient sample sizes to detect minor differences between study groups. This current meta-analysis included 1,069 ET cycles, and could detect an absolute difference of 7.5% with 80% power in a two-tailed analysis (assuming a CPR rate of 30% with a full bladder and a significance level of 0.05). The absolute difference between full and empty bladder was 7.8%, therefore validating our results. The results of this systematic review demonstrate that passive uterine straightening with the use of bladder distention during ET catheter placement may be a beneficial tool in optimising the outcome of the ET procedure, regardless of the method used for embryo catheter placement (e.g. ultrasound-guided or clinical touch). However, patient counseling is important, since most patients will need to micturate shortly after the transfer procedure. This action may be presumed to negatively affect the outcome of the IVF procedure. Therefore, proper counseling must be undertaken early in the cycle in order to decrease any anxiety over early mobilisation or micturation following the transfer. Lastly, it is also important to note that until today, no study has reported any direct adverse effects of bladder distention during the ET. Even so, one may argue that the main disadvantages are patient 50

61 discomfort, time for bladder distension to take place, and the possible psychological distress caused to patients who are forced to micturate after the transfer procedure. 51

62 Conclusion The results of this systematic review demonstrate that passive uterine straightening by the use of bladder distention during ET may be a beneficial tool in optimising the outcome of the ET procedure. Even so, more randomised, controlled, trials are needed to support the results of this systematic review and to address other issues, such as partial versus complete filling of the bladder, and the amount of fluid intake needed to achieve bladder fullness. 52

63 References 1. Edwards RG. Clinical approaches to increasing uterine receptivity during human implantation. Hum Reprod. 1995;10: Mansour RT, Aboulghar MA. Optimizing the embryo transfer technique. Hum Reprod. 2002;17: Lesny P, Killick SR, Tetlow RL, Robinson J, Maguiness SD. Embryo transfer - can we learn anything new from the observation of junctional zone contractions? Hum Reprod. 1998;13: Abou-Setta AM, Al-Inany HG, Mansour RT, Serour GI, Aboulghar MA. Soft versus firm embryo transfer catheters for assisted reproduction: a systematic review and meta-analysis. Hum Reprod. 2005;20: Epub 22 July Abou-Setta AM. Firm embryo transfer catheters for assisted reproduction: a systematic review and meta-analysis using direct and adjusted indirect comparisons. Reprod Biomed Online. 2006;12: Mansour R, Aboulghar M, Serour G. Dummy embryo transfer: a technique that minimizes the problems of embryo transfer and improves the pregnancy rate in human in vitro fertilization. Fertil Steril. 1990;54: Goudas VT, Hammitt DG, Damario MA, Session DR, Singh AP, Dumesic DA. Blood on the embryo transfer catheter is associated with decreased rates of embryo implantation and clinical pregnancy with the use of in vitro fertilization-embryo transfer. Fertil Steril. 1998;70: Alvero R, Hearns-Stokes RM, Catherino WH, Leondires MP, Segars JH. The presence of blood in the transfer catheter negatively in.uences outcome at embryo transfer. Hum Reprod. 2003;18: Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ. 2003;327: Lorusso F, Depalo R, Bettocchi S, Vacca M, Vimereati A, Selvaggi L. Outcome of in vitro fertilization after transabdominal ultrasound-assisted embryo transfer with a full or empty bladder. Research Papers in Fertility and Reproductive Medicine: Proceedings of the 18th World Congress on Fertility and Sterility (IFFS 2004); Mitchell JD, Wardle PG, Foster PA, Hull MG. Effect of bladder.lling on embryo transfer. J In Vitro Fert Embryo Transf. 1989;6: Lewin A, Schenker JG, Avrech O, Shapira S, Safran A, Friedler S. The role of uterine straightening by passive bladder distension before embryo transfer in IVF cycles. J Assist Reprod Genet. 1997;14: Lorusso F, Depalo R, Bettocchi S, Vacca M, Vimercati A, Selvaggi L. Outcome of in vitro fertilization after transab-dominal ultrasound-assisted embryo transfer with a full or empty bladder. Fertil Steril. 2005;84: Lu MC. Impact of non-physician factors on the physician factor of in vitro fertilization success: is it the broth, the cooks, or the statistics? Fertil Steril. 1999;71: Abou-Setta AM, Mansour RT, Al-Inany HG, Aboulghar MM, Serour GI, Aboulghar MA. Among women undergoing embryo transfer, is the probability of pregnancy 53

64 and live birth improved with ultrasound-guidance than with clinical touch alone? A systemic review and meta-analysis of prospective randomized trials. Fertil Steril. 2007;88: Egbase PE, al-sharhan M, al-othman S, al-mutawa M, Udo EE, Grudzinskas JG. Incidence of microbial growth from the tip of the embryo transfer catheter after embryo transfer in relation to clinical pregnancy rate following in-vitro fertilization and embryo transfer. Hum Reprod. 1996;11: Ghazzawi IM, Al-Hasani S, Karaki R, Souso S. Transfer technique and catheter choice in.uence the incidence of transcervical embryo expulsion and the outcome of IVF. Hum Reprod. 1999;14: Woolcott R, Stanger J. Ultrasound tracking of the movement of embryoassociated air bubbles on standing after transfer. Hum Reprod. 1998;13: Yovich JL, Turner SR, Murphy AJ. Embryo transfer techni-que as a cause of ectopic pregnancies in in vitro fertilization. Fertil Steril. 1985;44: Waterstone J, Curson R, Parsons J. Embryo transfer to low uterine cavity. Lancet. 1991;337: Sallam HN, Sadek SS. Ultrasound-guided embryo transfer: a meta-analysis of randomized controlled trials. Fertil Steril. 2003;80: Ralph SG, Rutherford AJ, Wilson JD. In.uence of bacterial vaginosis on conception and miscarriage in the first trimester: cohort study. BMJ. 1999;319: Friedler S, Lewin A, Schenker JG. Methodology of human embryo transfer following assisted reproduction. J Assist Reprod Genet. 1993;10: Al-Inany HG, Abou-Setta AM, Garzo G. ET catheters for assisted reproduction. Cochrane Database Syst Rev, Issue 1 Art. No: CD DOI: / CD Sundstrom P, Wramsby H, Persson PH, Liedholm P. Filled bladder simpli.es human embryo transfer. Br J Obstet Gynaecol. 1984;91: Easterbrook PJ, Berlin JA, Gopalan R, Matthews DR. Publication bias in clinical research. Lancet. 1991;337: Dickerson K, Min YL, Meinert CL. Factors in.uencing publication of research results. JAMA. 1992;267: Higgins JPT, Green S, editors. Cochrane handbook for systematic reviews of interventions (updated May 2005). The Cochrane Library 3. Chichester, UK: John Wiley & Sons Ltd.;

65 Table 1. Review table of the included studies, comparing full and empty bladder during embryo transfer. Included Patients Studies ET cycles A-priori Sample size Calculation Lewin et al., Not performed Mitchell et al., 1989 Lorusso et al., Not performed Not performed Fixed alternate days Random numbers method Method of Randomization Method of Randomization Concealment Intentionto-treat Followup Confounders Computergenerated randomization table Inadequate Unclear Unclear OPR ITT only CPR ITT only OPR ITT only ITT = Intention to treat analysis performed; CPR = Clinical pregnancy rate; OPR = Ongoing pregnancy rate. Multiple physicians - Instrumental assistance not recorded None evident Multiple studies performed on same patient population 55

66 Figure 1: Quorum flow diagram. Potentially relevant RCTs identified and screened for retrieval (n=4) RCTs excluded, Duplicate publication (n=1) RCTs retrieved for more detailed evaluation (n=3) RCTs excluded, (n=0) Potentially appropriate RCTs to be included in the meta-analysis (n=3) RCTs excluded (n=0) RCTs included in metaanalysis (n=3) RCTs withdrawn (n=0) RCTs with usable information, by outcome (n=3) 56

67 Figure 2. Meta-analysis forest plot showing ongoing pregnancy rate. Figure 3. Meta-analysis forest plot showing clinical pregnancy rate. 57

68 58

69 Chapter 5 A prospective randomized controlled trial comparing two embryo transfer catheters in an ART program. Pia Saldeen, Ahmed M. Abou-Setta, Torbjörn Bergh, Per Sundström, Jan Holte. Fertiltiy & Sterility. Published online November 28, 2007 [Epub ahead of print].

70 Abstract Objective: To compare the performance of the Cook Sydney IVF (SIVF) embryo transfer (ET) catheter and the Edwards-Wallace (EW) ET catheter. Design: Prospective randomized controlled clinical trial with an intention-to-treat analysis. Setting: Two-center study. Patient(s): Four hundred consecutive women <40 years undergoing ET with two fresh embryos. Intervention(s): Women were randomly allocated to undergo ET with either the EW or the SIVF catheter, with possible catheter change in case of insertion difficulties. Main Outcome Measure(s): Live birth and clinical pregnancy rates. Result(s): Two hundred two women were allocated to the SIVF catheter and 198 to the EW catheter. No significant differences in the clinical pregnancy rates (odds ratio [OR] 0.99, 95% confidence interval [CI] ) and live-birth rates (OR 1.09, 95% CI ) were found. The EW catheter had to be changed more often than the SIVF catheter (OR 9.5, 95% CI ) because of catheter insertion problems. Conclusion(s): The pregnancy and live birth rates were not significantly different with the two catheters, but catheter insertion failure was significantly more common with the EW catheter than with the SIVF catheter. 60

71 Introduction Assisted reproductive technology (ART) is a fast-paced and everchanging field in medicine. All parts of IVF/ICSI treatment are evolving toward more evidence-based procedures, resulting in increasing success rates. Although older reports suggested that many clinicians were little aware of the great importance of transfer technique skills, numerous published trials now document that the embryo transfer (ET) procedure has a huge impact on pregnancy and delivery rates (1 3). Clearly, the aim of the ET technique must be to minimize the risk of misplacing the embryo, and to minimize the risk of cervical and endometrial trauma. Attempts at optimizing the ET procedure include the use of dummy ET (4), softer ET catheters (5 6), and ultrasound guidance (7 10). The literature now includes numerous randomized trials and systematic reviews assessing different ET catheters (5 6, 11). In the majority of these studies, the Wallace Classic catheter was used as the standard. Its popularity has only been shadowed by its relatively higher rate of difficult transfers, and in some cases failure to navigate the unrelenting cervix, compared with its firmer rivals. Today, a new generation of competitor catheters have entered the market. They share the softness of the Wallace catheter and incorporate newer technologic advances in their design. Even so, there is limited evidence in the literature on the success of these new rivals compared with the time-tested Wallace catheter. Therefore, in the present study we wished to compare the standard Edwards-Wallace (EW) catheter to the Cook Sydney IVF (SIVF) catheter in a clinical setting in Sweden to determine if a change in policy should be implemented. Before the study, we had routinely used the EW catheter for ET. Our aim was to compare the pregnancy rates, live birth rates, and ease of use (defined as the rate of successful insertion) of the two catheters in IVF treatments. 61

72 Materials and methods This two-center prospective randomized controlled clinical trial was performed to determine if the Cook SIVF catheter performed as well as the standard EW catheter in our IVF programs. The protocol was approved by the Medical Faculty Research Ethics Committees at Lund University and Uppsala University, and all of the women gave their oral and written consent to participate. Participants Four hundred consecutive women undergoing IVF/ICSI were recruited from IVF-Kliniken CURA, Malmo, and Carl von Linne Kliniken, Uppsala, Sweden. The study was performed from August 2001 to February Inclusion criteria were age <40 years and at least two embryos generated from a fresh IVF cycle on the day of ET. Randomization was performed on the day of ET using a computer-generated randomization table with a 1:1 ratio into two groups: Cook SIVF and EW. Operator blinding was deemed unnecessary, because patients were randomized immediately before the transfer, and there is no feasible way of blinding the clinicians to the type of catheter allocation. Even so, clinicians were discouraged from discussing the group allocation with the patients. All patients signing the consent form participated in the study. Cook Sydney IVF Catheter The SIVF catheter (K-Jets-7019-SIVF; Cook IVF, Eight Miles Plains, Queensland, Australia) consists of an outer firm and an inner ultrasoft catheter. The outer guiding catheter (17 cm long) is slightly stiff, with a preshaped curve and a rounded bulb tip to help negotiate the cervical canal. It has a depth marker at 4 cm from the tip, which can be pulled back to a second marker at 5 cm. The inner catheter (23 cm long) is made of a soft material with a rounded bullet tip. In general, the inner catheter does not negotiate the cervical canal directly but rather is introduced into the uterine cavity through the outer catheter. Edwards-Wallace Embryo Replacement Catheter The Edwards-Wallace catheter (Classic Embryo Replacement Catheter; Smiths Medical, Hythe, Kent, U.K.) design also consists of an outer firm and inner soft catheter. The outer guiding catheter (18 cm long) is straight and made of rigid Teflon. The inner catheter (23 cm long) is 62

73 made of a soft polyethylene. In contrast to the Sydney IVF catheter, generally the inner catheter of the Edwards-Wallace is introduced directly through the cervix. Ovarian Stimulation Protocol Our standard ovarian stimulation protocol has previously been described in detail (12). In brief, the majority of patients were down-regulated with the long GnRH agonist protocol, although a minority of the women were treated with a GnRH antagonist. For ovarian stimulation, recombinant FSH (Gonal F; Laboratories Serono, Geneva, Switzerland; or Puregon; Organon, The Netherlands) was used in daily doses ranging from 75 IU to 450 IU SC depending on individual patient response. Final oocyte maturation was induced with urinary hcg (10,000 IU Profasi SC; Laboratories Serono) when the leading follicles were mm in diameter, and ovum pick-up was scheduled 36 hours later. Embryo Transfer and Luteal Phase Support Embryo transfer was performed hours after ovum pick-up by one of four experienced clinicians. A bivalve speculum was inserted into the vagina to expose the cervix. Endovaginal secretions and excess mucus was wiped off with a sterile cotton swab. The embryo transfers with SIVF were performed in two steps. First, the guiding SIVF catheter was passed through the cervical canal until the bulb tip of the catheter was located just past the internal os. Then the transfer (inner) catheter was loaded with the two best quality embryos available in a continuous column of ml culture medium (i.e., no air) and passed through the outer catheter and gently advanced into the uterine cavity. The embryos were expelled at a level of cm from the external os. The SIVF catheter was withdrawn directly after deposition of embryos and without moving the inner catheter back inside the outer catheter. In ETs with the EW catheter, the inner catheter of the EW was introduced directly through the cervix and retracted into the outer one for negotiation of the cervical canal only if difficulties were encountered. The outer catheter was used only for gentle manipulation and for stabilization of the inner catheter, when needed, and was advanced just beyond the external os. The embryos were expelled at cm from the external os. 63

74 If the study catheter could not be inserted through the internal os, the cervical canal was straightened with a forceps or tenaculum. If it still was not possible to insert the catheter, a change of catheter was made. Dilatation of the cervical canal was not performed in any case. Obturators for the EW catheter was not used. Ultrasound guidance was not used. In case of insertion failure despite manipulations as stated above, primarily the other study catheter was used, and only if neither of the study catheters could be inserted a complete switch of catheter, to the Frydman Tight-Difficult-Transfer (TDT) catheter, was made. After embryo injection, the catheter was flushed and checked under a stereomicroscope to ascertain that there were no retained embryos. Patients were allowed to ambulate immediately following the transfer procedure. For luteal phase support, natural micronized progesterone pessaries (Apoteksbolaget, Stockholm, Sweden), 400 mg tid, were prescribed for 18 days. Outcome Measures The primary outcomes were the rates of clinical pregnancies and live births. Immediately after the transfer, all events that had occurred were recorded, including catheter insertion difficulties. Clinical pregnancy was defined as the presence of a gestational sac on ultrasound at 7 weeks of gestation. Live birth was defined as delivery of a living baby after at least 28 weeks of gestation. Statistical Analysis Statistical analysis was performed according to the intention-to-treat principle. All analyses were two sided, and values of P<.05 were considered to be significant. Continuous variables were compared using the t test or Mann-Whitney U test for parametric and nonparametric data, respectively. Qualitative variables were compared with the chi-squared test or Fisher exact test, accordingly, and the 95% confidence intervals (CI). Odds ratios (ORs) and 95% CIs were calculated to examine the odds of improving clinical outcomes. Clinical and demographic data are also presented as mean ± SD or as frequency distribution for simplicity. Statistical analyses were performed with the aid of StatView (SAS Institute, Cary, NC) and MedCalc (Mariakerke, Belgium) software. 64

75 Results Of all basic variables, only maternal parity differed between the two groups, with more nulliparas in the SIVF group (Table 1). There was no significant difference in number of oocytes retrieved and preembryo quality between the EW and SIVF groups. There was no significant interclinician difference in pregnancy rate (P=.5). For the primary outcomes there were no significant differences between the clinical pregnancy rate (OR 0.99, 95% CI ), or live birth rates (OR 1.09, 95% CI ) between the two catheters (Table 2). The implantation rate (number of gestational sacs divided by number of embryos transfered) was 103/404 (25.5%) with the SIVF catheter and 101/396 (25.5%) with the EW catheter. Failure to negotiate the cervix, requiring changing the catheter to a either the other study catheter or a firmer alternative (TDT catheter) was more frequently seen with the EW catheter (32/198) than with the Cook catheter (4/202; OR 9.5, 95% CI ). There was no difference in pregnancy and delivery rate for the actual catheter used, as shown in Table 3. When comparing failure rates for the EW catheter in relationship to physician, one physician had a statistically significantly higher failure rate (P=.002). However, when excluding this physician from the analysis of insertion failure, the failure rate for the EW catheter was still significantly higher 25/184 (13.6%) compared with the SIVF catheter 4/202 (2%; Fisher exact test P<.0001; OR 7.8, 95% CI ). Because the pregnancy rate was lower with the rigid TDT catheter, a sensitivity analysis was performed in easy transfers not requiring the use of the TDT catheter. The clinical pregnancy rates in the groups randomized to SIVF and EW (excluding TDT) were then 40% and 41%, respectively (P=.8). According to the study protocol, a switch to the other study catheter was performed in the cases of failed catheter insertion, followed by an attempt with the TDT catheter if the second catheter also failed. Among the 32 catheter changes made in women randomized to the EW catheter, the switch to the SIVF catheter was successful in 26 cases, and in the remaining six cases a second switch to the TDT catheter had to be done. The clinical pregnancy rate in women randomized to the EW catheter but successfully changed to SIVF was 11/26 (42%), and in the further switch to the TDT catheter it was 1/6 (17%). In the four cases of 65

76 failed transfer in patients randomized to the SIVF catheter, a switch to the EW catheter was successful in one case, whereas a second switch to the TDT catheter had to be performed in the remaining three patients. The pregnancy rate was 1/3 (33%) in patients switched to TDT. 66

77 Discussion This prospective randomized study showed no difference in pregnancy or delivery rate between the EW and SIVF catheters according to an intention-to-treat analysis. The pregnancy rates were 40.4% and 40.1%, respectively, and the delivery rates 31.8% and 33.7%. As a secondary outcome, the rate of insertion failure was studied. The SIVF catheter was successfully inserted through the cervix at first attempt in 98% of cases, whereas the first EW catheter insertion failed in 32 women (16%). All of these 32 women had a second attempt with a SIVF catheter, which was successful in 26 cases, and in the remaining six cases a stiff TDT catheter was successfully inserted. Thus, in the EW group an SIVF catheter was successfully used as a rescue catheter in 26/ 198 (13%) of the women, and this switch did not affect the outcome negatively according to the intention-to-treat analysis. Although the pregnancy and delivery rates with the actual catheter used (randomized catheter results added with the secondary alternative) were similar in the EW and SIVF groups, it should be emphasized that this result in the SIVF group was accomplished in spite of a higher proportion of difficult transfers. Transfers with the TDT catheter, i.e., the third catheter choice in both groups, resulted in a delivery rate of 2/9 (22%). Because it is well known that, compared with stiff catheters, a soft catheter performs better in terms of pregnancy rate (5, 6), we thought it was unethical to change directly from a soft to a stiff catheter (e.g., a TDT catheter) in case of insertion failure with the EW or SIVF catheters, when an alternative soft catheter was available. We believe the higher percentage of successful transfers with the SIVF catheter is related to its semifirm design with a bulb-shaped tip with maintained stability and tactile transmission. The EW catheter works very well in easy transfers but had a higher insertion failure rate in the present study. If a stiff obturator had been used for the EW catheter, the insertion failure rate might have been lower. However, for the SIVF catheter no obturator was available on the Scandinavian market at the time of the study, and using an obturator for only one of the catheters would have biased the results. All clinicians participating in the present trial each had experience with several thousand embryo transfers. Moreover, before the trial, the EW catheter was used routinely, and all of the four clinicians had very 67

78 limited experience with the SIVF catheter. Our previous insertion failure rate in the two clinics with the EW catheter was 9% and 10%, respectively, lower than in the current study. An enthusiasm for a new soft catheter might have made us more liberal to replace the EW with SIVF in case of difficulties. After the study period when the SIVF catheter was accepted, the overall insertion failure rate with SIVF has stabilized at approximately 1%, confirming the results from the study. The present study is the second ever randomized trial comparing the EW and SIVF catheters. McIlveen et al. (13) randomized 150 women undergoing a fresh ET to the EW or SIVF catheters and found no significant difference in pregnancy rates (risk ratio 0.96, 95% CI ; Fig. 1), but the SIVF was, as in our study, associated with a significantly lower frequency of catheter change or change to an obturator. This is remarkable, because women with a history of difficult embryo transfers were excluded from the study by McIlveen et al. A meta-analysis of the two studies showed similar homogenous results (Fig. 1). The SIVF catheter was compared with the Tom Cat catheter in a randomized trial by McDonald and Norman (14). In contrast to the two studies comparing SIVF and EW catheters, the SIVF catheter was superior to the Tom Cat catheter regarding pregnancy rate: 29.6% versus 20.5% (OR 1.63, 95% CI ). The authors speculate that the reasons are the protection of the tip of the inner transfer catheter inside the guiding catheter when inserted through the cervical canal, and the softer tip of the inner catheter causing less trauma to the endometrium. Despite differences in operator experience in favor of the EW catheter in the present study, we accomplished more successful transfers with the SIVF catheter, indicating that the SIVF catheter is user friendly with a short learning curve. Despite the soft tip of the SIVF catheter, McIlveen et al. (13) noticed that the SIVF catheter tip was blood stained in 37% of cases. We also noticed a high incidence of bloodstained tips in the SIVF catheters, but the rate was not consistently registered. However, blood on the catheter tip did not endanger the chance of pregnancy, according to both studies, and it is possible that contamination of blood on the SIVF catheter is related more to cervical than endometrial trauma. 68

79 Conclusion According to the intention-to-treat protocol, there is no significant difference in clinical pregnancy or live birth rates between the EW catheter and the Cook SIVF catheter. Even so, catheter insertion failure was significantly more common with the EW catheter than with the SIVF catheter. 69

80 References 1. Schoolcraft WB, Surrey ES, Gardner DP. Embryo transfer techniques and variables affecting success. Fertil Steril. 2001; 76: Mansour RT and Aboulghar MAA. Optimizing the embryo transfer technique. Hum Reprod 2002; 17: Sallam HN. Embryo transfer: factors involved in optimizing the success, Curr Opin Obstet Gynecol. 2005; 17: Mansour RT, Aboulghar MA, Serour G. Dummy embryo transfer; a technique that minimized the problems of embryo transfer and improves the pregnancy rate in human in vitro fertilization. Fertil Steril 1990; 54: Abou-Setta AM A-I Inany HG, Mansour RT, Serour GI, Aboulghar MA. Soft versus firm embryo transfer catheters for assisted reproduction: a systematic review and meta-analysis. Hum Reprod 2005; 20: Buckett WM. A review and meta-analysis of prospective trials comparing different catheters used for embryo transfer. Fertil Steril 2006; 85: Sallam HN and Sadek SS. Ultrasound-guided embryo transfer: a meta-analysis of randomized controlled trials. Fertil Steril Oct;80(4): Buckett WM. A meta-analysis of ultrasound-guided versus clinical touch embryo transfer. Fertil Steril 2003; Flisser E and Grifo JA. Is what we clearly see really so obvious? Ultrasonography and transcervical embryo transfer- a review. Fertil Steril 2006; 87: Abou-Setta AM, Mansour RT, Al-Inany HG, Aboulghar MM, Serour GI, Aboulghar MA. Among women undergoing embryo transfer, is the probability of pregnancy and live birth improved with ultrasound-guidance than with clinical touch alone? a systemic review and meta-analysis of prospective randomized trials. Fertil Steril. 2007;88: Abou-Setta AM. Firm embryo transfer catheters for assisted reproduction: a systematic review and meta-analysis using direct and adjusted indirect comparisons. Reprod Biomed Online Feb;12(2): Saldeen P and Sundström P. Nuclear status of four-cell preembryos predicts implantation potential in in vitro fertilization treatment cycles. Fertil Steril 2005; 84, McIlween M, Lok FD, Pritchard J, Lashen H. Modern embryo transfer catheters and pregnancy outcome; a prospective trial. Fertil Steril 2005; 84: McDonald JA and Norman RJ. A randomized controlled trial of a soft double lumen embryo transfer catheter versus a firm single lumen catheter: significant improvements in pregnancy rates. Hum Reprod 2002;

81 Table 1: Patient characteristics Sydney IVF Edwards Wallace Significance No. of patients Mean age 32.7 ± ± 3.5 P = 0.06 Primigravida P = 0.01 Rank of cycle P = 0.9 BMI (mean) P = 0.9 Smoking P = 0.8 Table 2: Outcome measures- intention to treat analysis Sydney IVF Edwards Wallace Significance No of patients Clinical pregnancy rate 81 (40.1%) 80 (40.4%) O.R = 0.99; 95% CI = 0.66 to 1.47 Live-birth rate 68 (33.7%) 63 (31.8%) O.R = 1.09; 95% CI = 0.72 to 1.65 Table 3: Outcome measures- actual catheter used Sydney IVF Edwards Wallace TDT No of patients Clinical pregnancy rate 91 (40.0%) 68 (41.2%) 2 (22.2%) Live-birth rate 76 (33.3%) 53 (32.5 %) 2 (22.2%) 71

82 Figure 1: Meta-analysis forest plot comparing the clinical pregnancy and catheter failure rates between the Edwards-Wallace and Cook Sydney IVF catheters 72

83 Chapter 6 Firm versus soft embryo transfer catheters under ultrasound guidance: Does catheter choice really influence the pregnancy rates? Ismail Aboulfotouh, Ahmed M. Abou-Setta, Sherif Khattab, Iman Abdel Mohsen, Ahmed Askalani, Rasha E. El-Din Fertiltiy & Sterility. Published online July 17, 2007 [Epub ahead of print].

84 In recent years, the relationship between the ET catheter choice and the outcomes of assisted reproduction has become evident. The majority of the published trials comparing the outcomes for soft compared with firm catheters revealed conflicting results, necessitating a systematic review of the evidence to answer this clinical query. This gap in the literature, and ensuing argument, resulted in the production of two recent metaanalyses comparing soft versus firm ET catheters (1, 2). They demonstrated that softer catheters are associated with higher clinical pregnancy rates when compared with firmer catheters. In contrast, softer catheters also were demonstrated to be more likely to fail passage through the cervix, requiring instrumental assistance or catheter replacement (1). In addition, the majority of the studies were not performed under ultrasound guidance and did not consider the physician factor as a possible cause of the difference in results. Ultrasound guidance during the ET procedure has been shown to increase the clinical pregnancy (3 5) and live birth rates (5) when compared with the traditional clinical touch method. The aim of the present study is to investigate whether under ultrasound guidance the type of ET catheter could influence clinical outcomes. A detailed chart review of all IVF cycles (n = 666) performed in our center from 2004 to 2006 was performed. Details on patient demographics, cycle characteristics, and outcomes were extracted and compared for a semifirm catheter compared with soft catheters. Local ethical review committee approval was provided to perform this study. In our center, ultrasound-guided ET is performed by two highly trained and experienced physicians (I.A.F., S.K.) with similar pregnancy rates. Ultrasonography is performed by a trained sonographer with extensive experience in ultrasound during ET. The primary outcome of this study was the clinical pregnancy rate per woman. Secondary outcomes were the incidence of detecting indirect signs of cervical and/or uterine injury (e.g., blood, mucus) on the catheter tip and the incidence of difficult transfers with the respective catheters. Statistical analysis was performed with use of the computer statistical package StatsDirect (StatsDirect Ltd, Cheshire, United Kingdom). Odds ratios and 95% confidence intervals were calculated to examine the odds of improving clinical outcomes. Clinical and demographic data are also presented as mean (± SD) or as frequency distribution for 74

85 simplicity. Patient demographics and cycle characteristics were similar between the two groups (Table 1). For the primary outcome, there was a trend toward significance with the softer catheters than with the semifirm catheter, but this did not reach statistical significance. In addition, no significant difference between the clinical pregnancy rates between the individual soft catheters was found. Finally, no significant differences were found with regard to the secondary outcomes. It is theorized that through the use of ultrasound visualization of the cervical and endometrial canals, the incidence of difficult transfers can be minimized. In turn the easier the transfer, the lesser the incidence of trauma to the endocervical and endometrial canal. The results of this study also are in line with those of a recent prospective randomized controlled trial comparing the clinical performance of the Labotect (Labotect Embryo Transfer Cather, set, Ref. No ; Labotect Labor-Technik-Göttingen GmbH, Göttingen, Germany) and Wallace (Wallace Embryo Replacement Catheter, Ref. 1816N; Smiths Medical International Limited, Kent, UK) ET catheter sets in 260 women undergoing ET in a single assisted reproduction center (6). There was no statistically significant difference between the Wallace and Labotect catheters with regard to the clinical pregnancy (44.6% vs. 34.6%), implantation (23.2% vs. 18.9%), and ongoing pregnancy (38.5% vs. 27.7%) rates. Even so, catheter change because of unsuccessful negotiation of the internal cervical os was statistically significantly more frequently necessary with the Wallace than the Labotect catheter (33% and 2%, respectively). Therefore in conclusion, and in contrast to the previously available evidence, under ultrasound guidance individual catheter choice does not seem to statistically significantly affect the clinical pregnancy rate in a modern clinical IVF practice. This may be as a result of decreasing the incidence of difficult transfers and endometrial injury. 75

86 References 1. Abou-Setta AM, Al-Inany HG, Mansour RT, Serour GI, Aboulghar MA. Soft versus firm embryo transfer catheters for assisted reproduction: a systematic review and meta-analysis. Hum Reprod. 2005;20: Buckett WM. A review and meta-analysis of prospective trials comparing different catheters used for embryo transfer. Fertil Steril. 2006;85: Buckett WM. A meta-analysis of ultrasound-guided versus clinical touch embryo transfer. Fertil Steril. 2003;80: Sallam HN, Sadek SS. Ultrasound-guided embryo transfer: a meta-analysis of randomized controlled trials. Fertil Steril. 2003;80: Abou-Setta AM, Mansour RT, Al-Inany HG, Aboulghar MM, Serour GI, Aboulghar MA. Among women undergoing embryo transfer, is the probability of pregnancy and live birth improved with ultrasound-guidance than with clinical touch alone? A systemic review and meta-analysis of prospective randomized trials. Fertil Steril. Published online June 7, 2007 [Epub ahead of print]. 6. Ata B, Isiklar A, Balaban B, Urman B. Prospective randomized comparison of Wallace and Labotect embryo transfer catheters. Reprod BioMed Online 2007;14:

87 Table 1. Patient demographics, cycle characteristics, and clinical outcomes. Firm catheter (mean ± SD) Soft catheters (mean ± SD) Patient demographics and cycle characteristics No. of patients Age (y) ± ± 5.60 Cumulus complex ± ± 9.44 Metaphase II 9.96 ± ± 8.13 Metaphase I 1.39 ± ± 2.20 Germinal vesicle 0.96 ± ± 2.02 Zona pellucida 0.38 ± ± 0.63 Total embryos replaced 3.91 ± ± 1.52 Outcome measures Clinical pregnancy rate (%) Mucus on catheter tip (%) Blood on catheter tip (%) Difficult transfer rate (%)

88 78

89 Chapter 7 Soft versus firm embryo transfer catheters for assisted reproduction: a systematic review and meta-analysis. Ahmed M. Abou-Setta, Hesham G. Al-Inany, Ragaa T. Mansour, Gamal I. Serour, Mohamed A. Aboulghar. Human Reproduction. 2005;11:

90 Abstract Background: The true impact of the embryo transfer catheter choice on an IVF programme has not been fully examined. We therefore decided to systematically review the evidence provided in the literature so that we may evaluate a single variable in relation to a successful transfer, the firmness of the embryo transfer catheter. Methods: An extensive computerized search was conducted for all relevant articles published as full text, or abstracts, and critically appraised. In addition, a hand search was undertaken to locate any further trials. Results: A total of 23 randomized controlled trials (RCT) evaluating the types of embryo transfer catheters were identified. Only ten of these trials, including 4141 embryo transfers, compared soft versus firm embryo catheters. Pooling of the results demonstrated a statistically significantly increased chance of clinical pregnancy following embryo transfer using the soft (643/2109) versus firm (488/2032) catheters [P = 0.01; odds ratio (OR) = 1.39, 95% confidence interval (CI) = ]. When only the truly RCT were analysed, the results were again still in favour of using the soft embryo transfer catheters [soft (432/1403) versus firm (330/1402)], but with a greater significance (P < ; OR = 1.49, 95% CI = ). Conclusion: Using soft embryo transfer catheters for embryo transfer results in a significantly higher pregnancy rate as compared to firm catheters. Key words: catheter/ embryo transfer/ ICSI/ IVF/ meta-analysis/ randomized controlled trial 80

91 Background Embryo transfer is the final and most crucial step in IVF. About 80% of patients undergoing IVF reach the embryo transfer stage, but only a small proportion of them achieve pregnancy. The pregnancy rate after embryo transfer is dependent upon multiple factors including embryo quality, endometrial receptivity and the technique of the embryo transfer itself (1). Recently, several surveys have shown that the embryo transfer catheter ranks high as an important, independent factor in the success of an IVF programme. A survey of Australian clinicians rated the type of catheter used as the third most important variable in embryo transfer (2). In addition, a postal survey in the UK found that the type of catheter used was believed to be the fourth most important variable (3). The ideal embryo transfer catheter should avoid any trauma to the endocervix and/or endometrium as it finds its way into the uterine cavity. Several studies have compared different kinds of catheters for embryo transfer but most of these studies are either observational, retrospective, or are prospective but non-randomized. Even in the few prospective, randomized trials published, the majority had small sample sizes; sizes too small to reach a definite conclusion with statistical soundness. Therefore, the impact of the embryo transfer catheter choice on an IVF programme has been investigated in relatively small samples, albeit the examination of a single factor in reproductive medicine is more reliable when large groups are involved (4, 5). We therefore decided to systematically review the evidence provided in the literature so that we may evaluate a single variable in relation to a successful transfer, the firmness of the embryo transfer catheter. 81

92 Materials and methods Criteria for considering studies for this review All published, unpublished and ongoing randomized trials reporting data that compares outcomes for women undergoing embryo transfer through the cervical route following IVF, or ICSI using soft compared with firm embryo transfer catheters, were sought in all languages. Types of outcome measures The primary outcome measures used for this systematic review were implantation rate (IR), clinical pregnancy rate (CPR) and ongoing/takehome baby rate. The secondary outcomes were ease of transfer (catheter failure rate) and simultaneous occurrence of traumatic events (e.g. use of a tenaculum, stylette, sounding, and/or dilatation). In addition, the presence of blood, mucus and/or retained embryos on the tip of the catheter was evaluated. Search strategy for identification of studies A computerized search was conducted using MEDLINE (1978 to present), EMBASE (1980 to present), the Cochrane Central Register of Controlled Trials (CENTRAL) on the Cochrane Library Issue 2, 2005, and the National Research Register [a register of ongoing and recently completed research projects funded by, or of interest to, the UK s National Health Service (NHS)] as well as entries from the Medical Research Council s Clinical Trials Register, and details on reviews in progress collected by the NHS Centre for Reviews and Dissemination. The following Medical Subject Headings and text words were used: embryo transfer, embryo transfer technique, embryo transfer catheter, Cook, Erlangen, Frydman DT, Frydman, Gynetics, Rocket, TDT, Tom Cat, Wallace, and randomised controlled trial(s), randomized controlled trial(s) (RCTs). Furthermore, the reference lists of all known primary studies and review articles were also examined to identify additional relevant citations. In addition, a hand search of the citation lists of relevant publications, review articles, abstracts of major scientific meetings and included studies were searched for trials. Moreover, the reviewers sought ongoing and unpublished trials by contacting experts in the field, and commercial entities. 82

93 Methods of the review A standardized data extraction form was developed and piloted for consistency and completeness. Two reviewers (A.M.A.S. and H.G.A.I.) considered trials for inclusion, evaluated methodological quality and extracted trial data independently. Differences in interpretation were resolved by discussion and mutual agreement and refereeing by a third reviewer (R.T.M.). Data management and analysis was then conducted using the Review Manager (RevMan) 4.2 statistical software package. Individual outcome data were included in the analysis if they met the pre-stated criteria. Where possible, data were extracted to allow an intention-to-treat analysis. If data from the trial reports were insufficient or missing, the authors contacted the investigators of individual trials for additional information, in order to perform analyses on an intention-totreat basis. For the meta-analysis, the number of participants experiencing the event in each group of the trial was recorded. Heterogeneity by visual inspection of the outcome tables and by using the X 2 -test for heterogeneity with a 10% level of statistical significance was utilized. Where statistical heterogeneity was found, the reviewers looked for an explanation. If studies with heterogeneous results were thought to be comparable, statistical synthesis of the results using a random effects model was undertaken. Furthermore, a meta-regression analysis (subgroup analyses) was undertaken to determine, if possible, the source behind the heterogeneity. In addition, the I 2 test was used to attempt at quantifying any apparent inconsistency. An I 2 value greater than 50% may be considered substantial hererogeneity. In the absence of heterogeneity, results were pooled using a fixed effect model, the relative risk and risk difference [and 95% confidence intervals (CI)]. Description of studies During the course of this review, we came across several commercially available embryo transfer catheters. They were divided into two groups: soft or firm, according to the available literature and the experience of the authors. 83

94 Soft embryo transfer catheters The Frydman embryo transfer catheter has a soft 23 cm long inner polyurethane catheter with an external diameter of 1.53 mm with an open end. The Edwards Wallace embryo transfer catheter system set is openended and made of polyethylene, and has a firm outer Teflon introducer. It has an 18 or 23 cm long inner silicon catheter with an external diameter of 1.6 mm and an open end. The Cook Soft-Pass embryo transfer catheter system consisted of two parts fitted coaxially. The outer sheath of the catheter was 6.8 French size (FR) with an overall length of 17 cm and an inner catheter of 4.4 FR, measuring 23.5 cm. The tip of the inner sheath incorporates an echogenic stainless-steel band embedded circumferentially within a polyethylene sheath to enable its imaging at the time of transabdominal ultrasonogram. The Cook Soft-Trans embryo transfer catheter system set consists of a single lumen cannula with a 12.5 cm firm proximal part and a 4.0 cm soft distal part. The transfer catheter is made of an undisclosed soft polyurethane material. The Cook Sydney IVF catheter system set consists of a double lumen catheter set. The guiding (outer) catheter is 19 cm long, has a polycarbonate hub, a bulb tip and the distal end is angled. The transfer (inner) catheter is 23 cm long and the tip is 2.8 French size. The Gynetics Delphin embryo transfer catheter is single lumen catheter set, 21 cm in length. It uses a combination of a soft, flexible intrauterine catheter and a solid cervix catheter, but is softer than Gynetics Emtrac-A embryo transfer catheter. Firm embryo transfer catheters The Erlangen embryo transfer catheter consists of an introducing metal cannula (fitted with an obturator) and an insertion catheter. The cannula has an external diameter of 2 mm, and its tip is olive-shaped with a diameter of 3 mm. The silicon movable collar is usually placed 2 3 cm from the tip. The instrument has a length of 25 cm. To facilitate handling, the proximal end of the instrument is provided with a ring to accommodate the operator s finger. The quality of the steel used for the instrument permits the cannula to be bent to match the individual angle of kink between the uterine corpus and the cervix. 84

95 The Tom Cat embryo transfer catheter was initially used for draining the bladder of male cats; hence its name. It is 11.5 mm long and is made of polyethylene. The external and internal diameters of the tip are 1 mm and 0.3 mm respectively. The base is 6 mm in diameter and fits onto a 1 ml disposable syringe. The TDT (Tight Difficult Transfer) embryo transfer catheter consists of a single lumen 18 cm long polyethylene/polyprene cannula (Frydman 4.5) and a partly polyethylene, partly metal transfer catheter. The cannula is standard equipped with a malleable metal obturator, allowing bending it into the required curve necessary for passage through the cervical canal. The Rocket Embryon embryo transfer catheter is 18 cm in length. The inner transfer catheter is made of polyurethane and the outer sheath is made of white polythene. The Gynetics Emtrac-A embryo transfer catheter is a single lumen catheter set 21 cm in length. It uses a combination of a soft, flexible intrauterine catheter and a solid cervix catheter. A total of 23 prospective RCT evaluating the types of embryo transfer catheters were identified (10 full-text papers, 12 conference abstracts and one unpublished trial comparing different types of embryo transfer catheters). Of these studies, one was excluded because it compared a soft embryo transfer catheter to surgical placement of the embryos in the uterine cavity using a hysteroscope (6). In the remaining studies, only ten trials compared soft versus firm embryo catheters including 4141 embryo transfers (7-16) (Table I). The remaining studies either compared soft versus soft transfer catheters (17-25) or firm versus firm transfer catheters (26-28) (See Figure 1a). Methodological quality of included studies The methodological quality of each trial was assessed in terms of randomization, blinding of the patients, sample size, the absence of confounders and the extent of follow-up. Each trial was judged, and given a quality rating as adequate or inadequate: A = adequate, B = unclear, C = inadequate, D = not used. Furthermore, validity scores were given to each item: A = 4, B = 3, C = 2, D = 1 and the total was tabulated (Table II). High quality trials were defined as those receiving >15 points. Moderate quality trials were defined as receiving points. Poor quality trials were defined as receiving <10 points. 85

96 Furthermore, a funnel plot assessed publication bias, quality and heterogeneity (Figures 2a, b). Randomization was considered to be proper when computer generated number tables or sealed envelopes were used. Quasi-randomization was considered to be an inadequate form of randomization. As one study used alternate randomization (10) and the randomization was not clear from the manuscript in five studies (8, 9, 12, 14, 16), only four studies described a proper method of randomization (7, 11, 13, 15). Furthermore, blinding was examined with regards to who was blinded in the trials. All levels were sought and categorized as follows: (i) single blind (the investigator only knew of the allocation), (ii) no blinding (both investigator and participant knew the allocated treatment), (iii) unclear. It is important to note that double blind was not sought since it would be impossible to blind the operator from knowing the type of catheter being used. In all the studies, the exact level of blindness could not be extracted, therefore they were stated as unclear. Sample size calculations were considered to be proper when the authors of the studies pre-calculated the number needed in each arm prior to starting the trial. This prevents the occurrence of Type II errors. Only two studies (13, 15) undertook sample size calculations. As for the presence of confounders, for the purpose of this systematic review, confounders included any factors that might have helped to alter the results. These included more than one operator (as pregnancy rates are often operator dependent), testing several factors at the same time (e.g. soft versus firm catheters combined with ultrasound-guided versus clinical touch), unequal characteristics in the two groups (e.g. statistically significant differences in patient age), the transfer of fresh IVF, cryo embryos and/or donor oocytes in the same trial. Four studies (7, 8, 10, 13) showed obvious presence of confounders. Furthermore, the remainder of the studies could not be considered free from confounders since either they were only published as abstracts in conference proceedings and/or the authors did not provide enough information in the text of the published manuscripts. Finally, quality scores were assigned to each trial for completeness of follow-up. Only three studies (8, 11, 15) completed follow-up until delivery, while the remaining studies were discontinued before follow-up could occur. 86

97 Results Primary outcome measures Implantation rate For the implantation rate, data were only available from two studies (8, 14). Using the fixed effect model, there was no statistically significant difference in the chance of embryo implantation following embryo transfer using the soft (103/573) versus firm (60/360) catheters [P = 0.34, odds ratio (OR) = 1.18, 95% CI = ]. However, there was statistical heterogeneity between the studies (P = , I 2 = 92.5%). Therefore, the random effects model was used for the metaanalysis, but this did not alter the results: soft (103/573) versus firm (60/360) catheters (P = 0.71, OR = 1.28, 95% CI = ). In addition, meta-regression analyses were undertaken to determine the source of the heterogeneity. Subgroup analyses were undertaken by excluding the moderate quality study (8) to determine if the heterogeneity was caused by this factor. When Grunert et al. (8) was removed and the results were re-analysed using the fixed effect model, this did alter the overall statistical outcome of the results in favour of the soft catheters: soft (53/170) versus firm (23/154) catheters (P = 0.001, OR = 2.51, 95% CI = ). Clinical pregnancy rate For the clinical pregnancy rate, data were available from all ten studies. Using the fixed effect model, pooling of the results demonstrated a statistically increased chance of clinical pregnancy following embryo transfer using the soft (643/2109) versus firm (488/2032) catheters (P < , OR = 1.39, 95% CI = ). Nevertheless, there was significant statistical heterogeneity between the studies (P = 0.003, I 2 = 63.4%). In order to nullify this heterogeneity, the random effects model was utilized. Pooling of the results still demonstrated a significantly increased chance of clinical pregnancy following embryo transfer using the soft versus firm catheters (P = 0.01, OR = 1.39, 95% CI = ). In addition, meta-regression (subgroup analyses) was undertaken to determine the source of the heterogeneity. Subgroup analyses were undertaken by excluding one study at a time to determine if the heterogeneity was caused by one factor, or if multiple factors were involved. When Ghazzawi et al. (10) was removed and the results were 87

98 re-analysed using the fixed effect model, the heterogeneity was nullified (P = 0.15, I 2 = 32.9%). However, this did not alter the overall statistical outcome of the results: soft (612/1949) versus firm (440/1872) catheters, but on the contrary it increased the statistical gap between the two groups (P < , OR = 1.49, 95% CI = ). Furthermore, when only the fresh IVF procedure (i.e. excluding frozen replacement and donor cycles) were analysed, using the random effect model, the results were still in favour of using the soft embryo transfer catheters [soft (613/1964) versus firm (466/1882) catheters] (P = 0.02, OR = 1.38, 95% CI = ). In addition, when only the true RCT were analysed, using the fixed effect model, the results were again still in favour of using the soft embryo transfer catheters [soft (432/1403) versus firm (330/1402) catheters] (P < , OR = 1.49, 95% CI = ). Ongoing pregnancy/take-home baby rate For the ongoing pregnancy/take-home baby rate, data were available from three studies (8, 11, 15). Using the fixed effect model, pooling of the results demonstrated a significantly increased ongoing pregnancy/ take-home baby rate following embryo transfer using the soft (290/978) versus firm (240/948) catheters (P = 0.03, OR = 1.25, 95% CI = ). Secondary outcome measures Catheter failure For the failure rate using the assigned catheter, data were available from five studies (7, 11, 12, 13, 15). Using the random effect model, pooling of the results demonstrated a trend towards statistical significance, but this increased chance of failure following embryo transfer using the soft (100/1563) versus firm (11/1571) catheters did not reach statistical significance (P = 0.06, OR = 7.51, 95% CI = ). Traumatic events Data pertaining to traumatic events during the embryo transfer, use of a tenaculum, stylette, sounding and/or dilatation, were recorded in three studies (7, 10, 13). Using the random effects model, pooling of the overall results demonstrated a significantly increased chance of 88

99 traumatic events during embryo transfer using the soft (229/684) versus firm (104/686) catheters (P < , OR = 5.40, 95% CI = ). For the rate of using a tenaculum, data were available from two studies (7, 13). Using the fixed effect model, pooling of the results demonstrated a statistically increased chance of tenaculum use following embryo transfer using the soft (84/524) versus firm (62/526) catheters (P = 0.02, OR = 1.61, 95% CI = ). For the rate of using a stylette in corporation with the embryo transfer, data were available from only one study (13), which did not use a stylette in any of the cases. For the rate of using sounding, data were available from two studies (10, 13). Using the fixed effect model, pooling of the results demonstrated a significantly increased chance of need for sounding during embryo transfer using the soft (36/484) versus firm catheters (6/486) (P < , OR = 7.45, 95% CI = ). For the rate of using dilatation, data were available from three studies (7, 10, 13). Using the fixed effect model, pooling of the results demonstrated a significantly increased chance of need for dilatation during embryo transfer using the soft (45/684) versus firm catheters (16/686) (P = , OR = 3.07, 95% CI = ). Catheter tip Data pertaining to another important aspect of embryo transfer is the catheter tip. Blood, mucus, and the retention of embryos at the tip of the embryo transfer catheters were described in four studies (7, 10, 12, 13). Using the random effects model, pooling of the overall results demonstrated a significantly increased chance of these events during embryo transfer using the soft (163/1328) versus firm (55/1320) catheters (P < 0.02, OR = 5.63, 95% CI = ). Blood on the tip of the catheter was described in two studies (7, 13). Using the fixed effect model, there was no significant difference between the two groups: soft (42/524) versus firm (39/526) (P = 0.37, OR = 1.10, 95% CI = ). Mucus on the tip of the catheter was described in one study (10). There was a significantly increased chance of finding mucus on tip of the catheter in the soft catheter group (65/160) when compared with the firm catheter group (0/160) (P = ). 89

100 Retained embryos were described in three studies (10, 12, 13). Using the random effects model, there was a trend towards increased likelihood of retained embryos using the soft catheters (56/644) versus firm catheters (16/634), but this did not reach statistical significance (P = 0.05, OR = 4.52, 95% CI = ). 90

101 Discussion Although most patients who undergo assisted procreation, via IVF or ICSI, reach the embryo transfer stage and have embryos of good quality available for transfer, embryo implantation remains the rate-limiting step in the success of this form of therapy. The main factors that affect embryo implantation are uterine receptivity, embryo quality, and efficiency of the embryo transfer procedure. The aim must be to transfer the embryos with a high degree of reliability atraumatically. There have been many publications over the years discussing ways of improving embryo transfer and hopefully pregnancy rates. Multiple factors may affect the success of uterine embryo transfer including the experience of the physician (29), the use of ultrasound guidance (30, 31), the ease of the procedure (32), the presence or absence of blood on the catheter (33) and bacterial contamination of the catheter (34). In addition, other factors concerning embryo transfer that might affect the chance for an ongoing pregnancy have been identified, such as the use of cervical introducers or obturators (10), the value of resting after transfer (35), the position of embryo insertion in the uterus (36, 37), flushing of the cervical canal to remove mucus (38), microbiological factors in terms of the local flora (39) and retention of embryos in the catheter (40, 41). Since it would be difficult to compare several factors at the same time, we decided to concentrate on one factor, the firmness of the embryo transfer catheter, as a possible cause of limiting the success of the embryo transfer. One crucial factor that has not gained enough attention and scrutiny as a deciding factor is catheter technology. There is no conclusive evidence for the preferred use of any particular catheter and previous randomized trials have been too small to show significant differences in pregnancy rates. Moreover, some authors have concluded that the success rate of embryo transfers is not even influenced by the choice of the embryo transfer catheter used (42). Therefore the catheter choice has been mainly left to personal choice, availability and implied costeffectiveness. Several embryo transfer catheters are commercially available. All are mainly composed of non-toxic plastics and/or metal, but vary in length, calibre, location of the distal port (end- or side-loading), and degree of stiffness and malleability. These catheters can be subdivided by the material they are made of (i.e. metal, hard or soft plastics) and whether 91

102 they are equipped with, or without, an introducing cannula that facilitates the transfer procedure. In this systematic review, soft embryo transfer catheters overall performed better compared with the firm embryo transfer catheters. Even though the implantation rate seemed to be questionable in favour of the soft catheters, there is definitely a strong statistical trend for clinical pregnancies using the soft catheters. This was apparent in the clinical pregnancy and ongoing pregnancy/take-home baby rate. One theory why the softer catheters produce better results is built on decreasing the trauma to the endometrium. The softer the materials used, the lesser the chance for damage to the endometrium and the lesser the chance for uterine contractions. The soft transfer catheters follow the natural curvature of the uterine cavity better than the firmer catheters, possibly reducing the risk of burrowing into the posterior endometrium in the anteflexed uterus, or stimulating uterine junctional zone contractions. This is supported by the ultrasound-detected endometrial changes following intrauterine insemination, which differ between firm and soft catheters (43). The Tom Cat catheter was shown to cause significantly more trauma to the endometrium than did the Edwards Wallace catheter. In addition, a retrospective analysis of 518 embryo transfers, comparing five catheters [firm (Tefcat, Tom Cat, Norfolk) and soft (Frydman, Wallace)] found that a soft catheter was associated with higher pregnancy rates than a firm catheter (44). The results of this meta-analysis confirm that this increased pregnancy rate is both statistically and clinically significant. In contrast, the soft catheters were also associated with a higher degree of failure to negotiate the cervix and therefore the simultaneous occurrence of traumatic events (use of tenaculum, stylette, sounding and/or dilatation). In addition, they had a higher rate of blood, mucus and retained embryos at the tip of the embryo transfer catheter, but overall these events did not seem to alter the pregnancy rates. Passing soft catheters through the cervical canal is often difficult and even sometimes impossible. In a series of 876 embryo transfer procedures by Wood et al. (45), 1.3% were impossible, 3.2% were very difficult (requiring manipulation for >5 min or cervical dilatation) and 5.6% were difficult (requiring manipulation) to perform. In another study by Mansour et al. (46), soft catheters resulted in the highest rate (37.6%) of difficult embryo transfer with the consequences of lowering 92

103 the pregnancy rate. Furthermore, difficult transfers have been associated with lower pregnancy rates (46, 47). Our review supports the theory that softer catheters are associated with a higher incidence of difficult transfers, but not negatively affecting the pregnancy rates. Since difficult transfers have been associated with a poorer outcome than easy transfer, it would be useful to directly examine the uterine cavity for any lesions post-transfer. Unfortunately, this would not be possible without ultimately affecting the pregnancy rate. Therefore indirect measures of the degree of difficulty are utilized. These include patient discomfort during the procedure, the need for use of a tenaculum, stylette, sounding and/or cervical dilatation, and the presence of blood on the catheter post-transfer. Different approaches have been described in cases of difficult embryo transfers with varying success rates (46, 48-50). A commonly used initial approach is to negotiate the cervix using the outer sheath of the catheter, with its inner noodle withdrawn (51). Once the uterine cavity is entered, the inner noodle is used to deposit the embryos, taking care to avoid the fundus. Even though this technique works efficiently in certain situations, in others it is not sufficient. Therefore more invasive and potentially traumatic events are sometimes undertaken by clinicians to overcome the problematic cervix. These include the use of a tenaculum, stylette, sounding and/or cervical dilatation. Overall these events have been associated with increased uterine junctional zone contractions and a decreased pregnancy rate (32, 47, 49, 52). Alternatively, the cervical route may be bypassed and the embryos may be transferred transmyometrially into the uterine cavity using the Towako method (48). Another tell-tale sign of a difficult transfer is the post-transfer presence of blood on the transfer catheter. Amongst clinicians, the absence of blood on the catheter or cannula is ranked high as an important factor towards success (2). This opinion is supported by literature reports in which the presence of blood on the transfer catheter has been associated with lower pregnancy rates (52, 33). In addition, Perin et al. (26) found that contamination of the catheter with blood and mucus accounted for significantly lower implantation and clinical pregnancy rates. In our review, even though these events were present more 93

104 frequently with the softer catheters, it did not seem to drastically alter the outcome. Finally, the incidence of retained embryos was shown in this review to be higher with the softer embryo catheters. The role of retained embryos in decreasing the pregnancy rate is controversial with some studies claiming a negative effect (52) and other claiming no such effect (53, 33). Again, we could not confirm that the increased incidence of retained embryos with the soft catheters had any great influence on the overall outcome. In conclusion, the results of this study clearly indicate that the type of embryo transfer catheter contributes significantly to the success rate of an IVF programme. Soft catheters rather than firm catheters are associated with better pregnancy rate, even though a soft catheter is also associated with more traumatic events. More adequately powered, high quality RCT are needed to support the development of an ideal soft catheter that finds its way to the cavity with minimal failure rate. 94

105 Acknowledgements The authors would like to thank all the corresponding authors that were contacted for more information and provided us with assistance. Special thanks to Dr Janelle McDonald, Dr Robert Norman and Dr Raja Karaki, whose direct assistance has helped to increase the accuracy of this systematic review by providing us with missing information. 95

106 References 1. Mansour R and Aboulghar M (2002) Optimizing the embryo transfer technique. Hum Reprod 17, Kovacs GT (1999) What factors are important for successful embryo transfer after in-vitro fertilization? Hum Reprod 14, Salha OH, Lamb VK and Balen AH (2001) A postal survey of embryo transfer practice in the UK. Hum Reprod 16, Templeton A, Morris JK and Parslow W (1996) Factors that affect outcome of invitro fertilisation treatment. Lancet 348, Ramsay S (1999) Experts give practical advice in reproductive medicine. Lancet 354, Sweet CR, Hamidi J, El Mansari El, Guttieri T and Harrington T (1998) Hysteroscopic vs. catheter embryo transfer: a randomized prospective pilot study. American Society Reproductive Medicine (San Francisco, USA). Fertil Steril 70,(Suppl 1)S Wisanto A, Janssens R, Deschacht J, Camus M, Devroey P and Van Steirteghem AC (1989) Performance of different embryo transfer catheters in a human in vitro fertilization program. Fertil Steril 52, Grunert GM, Dunn RC, Valdes CT, Wun CC and Wun WSA (1998) Comparison of Wallace, Frydman DT and Cook embryo transfer catheter for IVF: a prospective randomised study [abstract]. Annual Meeting of the American Society of Reproductive Medicine, Fertil Steril 70, S Amorcho B, Gomez E, Pontes L, Campos I and Landeras J (1999) Does the selection of catheter for embryo transfer affect the success rate of an ART unit? [abstract] 15th Annual Meeting of the ESHRE (Tours, France), Hum Reprod 14(Suppl 1) Ghazzawi IM, Al-Hasani S, Karaki R and Souso S (1999) Transfer technique and catheter choice influence the incidence of transcervical embryo expulsion and the outcome of IVF. Hum Reprod 14, Curfs MHJM, Cleine JH, van Kamp AA, Kruse-Blankestijn ME, Hondelink MN and Leerentveld RA (2001) Comparison of the Wallace versus TDT embryo-transfer catheter: a prospective, randomized study. Third Biennial Alpha Conference, September 8 11, 2001, New York, New York, USA. Reprod Biomed Online : 3(Suppl 1). 12. Lavery S, Ravhon A, Skull J, Ellenbogen A, Taylor J, Dawson K, Trew G, Margara R and Winston R (2001) A prospective randomized controlled trial of Wallace and Rocket embryo transfer catheters in an IVF embryo transfer programme [abstract]. Hum Reprod 16(Suppl 1), McDonald JA and Norman RJ (2002) A randomized controlled trial of a soft double lumen embryo transfer catheter versus a firm single lumen catheter: significant improvements in pregnancy rates. Hum Reprod 17, Mortimer S, Fluker M and Yuzpe A (2002) Effect of embryo transfer catheter on implantation rates [abstract]. 58th Annual Meeting of the American Society for Reproductive Medicine. Fertil Steril 78(3), S

107 15. van Weering HG, Schats R, McDonnell J, Vink JM, Vermeiden JP and Hompes PG (2002) The impact of the embryo transfer catheter on the pregnancy rate in IVF. Hum Reprod 17, Foutouh IA, Youssef M, Tolba M, Rushdi M, Nakieb A and Meguid WA (2003) Does embryo transfer catheter type affect pregnancy rate? Middle East Fertil Soc J 8, al-shawaf T, Dave R, Harper J, Linehan D, Riley P and Craft I (1993) Transfer of embryos into the uterus: how much do technical factors affect pregnancy rates? J Assist Reprod Genet 10, Mayer JF, Nechiri F, Jones EL, Weedon VM, Kalin HL, Lanzendorf SE, Oehinger SC, Toner JP and Muasher SJ (1999) Prospective randomized analysis of the impact of two different transfer catheters on clinical pregnancy rates [abstract]. ASRM/CFAS Conjoint Annual Meeting September 25 30, 1999 (Toronto, Canada). Fertil Steril (Suppl 1), S Boone WR, Johnson JE, Blackhurst DM and Crane MM 4th (2001) Cook versus Edwards Wallace: are there differences in flexible catheters? J Assist Reprod Genet 18, Karande V, Hazlett D, Vietzke M and Gleicher N (2002) A prospective randomized comparison of the Wallace catheter and the Cook Echo-Tip catheter for ultrasound-guided embryo transfer. Fertil Steril 77, Levi-Setti PE, Albani E, Baggiani AM, Zannoni E, Colombo G, Liprandi V (2002) Prospective randomized study comparing two soft catheters for embryo transfer. Fertil Steril 78;3(Suppl 1)S Saldeen P, Bergh T, Sundstrom P and Holte J (2003) A prospective randomized controlled trial comparing two embryo transfer catheters in an ART programme [abstract]. 19th Annual Meeting of the European Society of Human Reproduction and Embryology. Hum Reprod 18(Suppl 1), Mcllveen M, Lok D, Pritchard J and Lashen H (2004) A randomised controlled trial comparing two embryo transfer catheters. Hum Reprod 19,127S. 24. Taylor TH, Colturato LF, Jones AE, Wright G, Elsner CW, Kort HI and Nagy ZP (2005) Prospective comparison of three soft ET catheters. Hum Reprod 20(Suppl 1), i Lashen H (unpublished data) A prospective randomised trial comparing two embryo transfer catheters. National Research Register (NRR): Perin PM (1999) The influence of two different transfer catheters on the pregnancy rate in a human in vitro fertilization program. Reprod Clim 14, Meriano J, Weissman A, Greenblatt EM, Ward S and Casper RF (2000) The choice of embryo transfer catheter affects embryo implantation after IVF. Fertil Steril 74, Schiewe M (2001) IVF-ICSI pregnancy outcomes in a randomized embryo transfer catheter trail comparing the cook soft-pass catheter to the Edward Wallace catheter [abstract] (2001) 51st Annual Meeting of the American Society of Reproductive Medicine, Fertil Steril 76;3(Suppl 1),

108 29. Lu MC (1999) Impact of non-physician factors on the physician factor of in vitro fertilization success: is it the broth, the cooks, or the statistics? Fertil Steril 71, Buckett WM (2003) A meta-analysis of ultrasound-guided versus clinical touch embryo transfer. Fertil Steril 80, Sallam HN and Sadek SS (2003) Ultrasound-guided embryo transfer: a metaanalysis of randomized controlled trials. Fertil Steril 80, Lesny P, Killick SR, Tetlow RL, Robinson J and Maguiness SD (1998) Embryo transfer: can we learn anything new from the observation of junctional zone contractions? Hum Reprod 13, Goudas VT, Hammitt DG and Damario MA (1998) Blood on the embryo transfer catheter is associated with decreased rates of embryo implantation and clinical pregnancy with the use of in vitro fertilization embryo transfer. Fertil Steril 70, Egbase PE, al-sharhan M, al-othman S, al-mutawa M, Udo EE and Grudzinskas JG (1996) Incidence of microbial growth from the tip of the embryo transfer catheter after embryo transfer in relation to clinical pregnancy rate following invitro fertilization and embryo transfer. Hum Reprod 11, Woolcott R and Stanger J (1997) Potentially important variables identified by transvaginal ultrasound-guided embryo transfer. Hum Reprod 12, Yovich JL, Turner SR and Murphy AJ (1985) Embryo transfer technique as a cause of ectopic pregnancies in in vitro fertilization. Fertil Steril 44, Waterstone J, Curson R and Parsons J (1991) Embryo transfer to low uterine cavity. Lancet 337, Sallam HN, Farrag F, Ezzeldin A, Agameya A, Sallam AN (2000) Vigorous flushing of the cervical canal prior to embryo transfer, a prospective randomised study. Fertil Steril 74;3(Suppl 1)S Ralph SG, Rutherford AJ and Wilson JD (1999) Influence of bacterial vaginosis on conception and miscarriage in the first trimester: cohort study. Br Med J 319, Friedler S, Lewin A and Schenker JG (1993) Methodology of human embryo transfer following assisted reproduction. J Assist Reprod Genet 10, Moore DE, Soules MR, Klein NA, Fujimoto VY, Agnew KJ and Eschenbach DA (2000) Bacteria in the transfer catheter tip influence the live-birth rate after in vitro fertilization. Fertil Steril 74, Diedrich K, Van der ven H, Al-Hasani S and Krebs D (1989) Establishment of pregnancy related to embryo transfer techniques after in-vitro fertilization. Hum Reprod 4(Suppl 1), Lavie O, Margalioth EJ, Geva-Eldar T and Ben-Chetrit A (1997) Ultrasonographic endometrial changes after intrauterine insemination: a comparison of two catheters. Fertil Steril 68, Wood EG, Batzer FR, Go KJ, Gutmann JN and Corson SL (2000) Ultrasoundguided soft catheter embryo transfers will improve pregnancy rates in in-vitro fertilization. Hum Reprod 15,

109 45. Wood C, McMaster R, Rennie G, Trounson A and Leeton J (1985) Factors influencing pregnancy rates following in vitro fertilization and embryo transfer. Fertil Steril 43, 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, Lesny P, Killick SR, Tetlow RL, Robinson J and Maguiness SD (1999) Embryo transfer and uterine junctional zone contractions. Hum Reprod Update 5, Kato O, Takatsuka R and Asch R (1993) Transvaginal transmyometrial embryo transfer: the Towako method; experience of 104 cases. Fertil Steril 59, Groutz A, Lessing JB, Wolf Y, Yovel I, Azem F and Amit A (1997) Cervical dilation during ovum pick-up in patients with cervical stenosis: effect on pregnancy outcome in an in vitro fertilization embryo transfer program. Fertil Steril 67, Tur-Kaspa I, Yuval Y, Bider D, Levron J, Shulman A and Dor J (1998) Difficult or repeated sequential embryo transfers do not adversely affect in-vitro fertilization pregnancy rates or outcome. Hum Reprod 13, Glass KB, Green CA, Fluker MR, Schoolcraft WB, McNamee PI, Meldrum DR (2000) Multicenter randomized controlled trial of cervical irrigation at the time of embryo transfer. (abstract no. O-085) Fertil Steril 74 (Suppl 1) S Visser DS, Fourie FL and Kruger HF (1993) Multiple attempts at embryo transfer: effect on pregnancy outcome in an in vitro fertilization and embryo transfer program. J Assist Reprod Genet 10, Nabi A, Awonuga A, Birch H, Barlow S, Stewart B (1997) Multiple attempts at embryo transfer: does this affect in-vitro fertilization treatment outcome? Hum Reprod 12,

110 Table I. Review table of the prospective, randomized, controlled studies analysed, comparing soft (SC) versus firm (FC) embryo transfer catheters Sample Included studies Catheter type and firmness Randomization size Edwards Wallace catheter (SC) 100 Frydman catheter (SC) 100 Wisanto et al., 1989 Tight Difficult Transfer (TDT) catheter (FC) 100 Random number table Tight Difficult Transfer (TDT) catheter (FC) + ultrasound-guided 100 Edwards Wallace catheter (SC) 49 Grunert et al., 1998 Cook catheter (SC) 50 Not determined Frydman DT catheter (FC) 51 Gynetics Delphin catheter (SC) 113 Amorocho et al., 1999 Not determined Gynetics Emtrac-A catheter (FC) 101 Edwards Wallace catheter (SC) 160 Ghazzawi et al., 1999 Erlangen metal catheter (FC) 160 Alternative randomization Edwards Wallace catheter (SC) 240 Curfs et al., 2001 Tight Difficult Transfer (TDT) catheter (FC) 240 Dark sealed envelopes 100

111 Included studies Catheter type and firmness Sample size Lavery et al., 2001 Edwards Wallace catheter (SC) 160 Rocket embryo transfer catheter (FC) 148 McDonald and Norman, Cook catheter (SC) Tom Cat catheter (FC) 324 Mortimer et al., 2002 Cook SIVF 6019 catheter (SC) 58 Tom Cat catheter (FC) 60 Van Weering et al., 2002 Cook K-soft 5000 soft trans universal 639 set (SC) Tight Difficult Transfer (TDT) catheter 657 (FC) Foutouh et al., 2003 Edwards Wallace catheter (SC) 114 Rocket embryo transfer catheter (FC) 91 Randomization Not determined Computergenerated Not determined Random number table Not determined 101

112 Table II. Review table of the validity scores for the included studies Randomization Study (4) Sample size (4) Blinding (4) Confounders (4) Followup (4) Total score (20) Wisanto et al., Grunert et al., Amorocho et al., Ghazzawi et al., Curfs et al., Lavery et al., McDonald and Norman,

113 Randomization Study (4) Sample size (4) Blinding (4) Confounders (4) Followup (4) Total score (20) Mortimer et al., Van Weering et al., Foutouh et al.,

114 Figure 1. Quorum flow diagram. Potentially relevant RCTs identified and screened for retrieval (n=21) RCTs excluded, with easons (n=1) RCTs retrieved for more detailed evaluation (n=20) RCTs excluded, with reasons (n=10) Potentially appropriate RCTs to be included in the metaanalysis (n=10) RCTs excluded from metaanalysis, with reasons (n=0) RCTs included in meta-analysis (n=10) RCTs withdrawn, by outcome, with reasons (n=0) RCTs with usable information, by outcome (n=10) 104

115 Figure 2. Funnel plots comparing soft vs firm ET catheters for all RCTs (a) and for the truly RCTs (b). (a) (b) 105

116 Figure 3. Meta-analysis of clinical pregnancy rates for all RCTs (random effects model), Truly RCTs (fixed effect model) and fresh IVF cycles only (random effects model.) 106

117 Chapter 8 Air-fluid versus fluid-only models of embryo catheter loading: a systematic review and meta-analysis. Ahmed M. Abou-Setta Reproductive BioMedicine Online. 2007; 14:79 83.

118 Abstract The objective of this systematic review was to determine the beneficial or detrimental effect of using air bubbles to bracket the embryocontaining medium during embryo transfer. To test this theory, a metaanalysis of randomized trials comparing air-fluid versus fluid-only methods was performed. The primary outcome measures were live birth, ongoing and clinical pregnancy rates. The secondary outcome measures were the rates of implantation, miscarriage, multiple and ectopic pregnancies and retained embryos. Electronic (e.g. PubMed, EMBASE, Cochrane Library) and hand searches of the literature revealed two included studies (298 women). Meta-analysis was conducted using the Mantel Haenszel method (fixed-effect model). For the primary outcome measures, there were no significant differences between the two methods with regards to live birth (OR = 1.34; 95% CI = ), ongoing pregnancy (OR = 1.34; 95% CI = ) and clinical pregnancy (OR = 1.13; 95% CI = ) rates. For the secondary outcomes, there were no significant differences between the two groups. In conclusion there is insufficient evidence to suggest that the fluid-only method is superior to the use of air brackets during embryo loading. There is a need for well-designed and powered randomized trials to determine any possible benefit to either method. Key words: embryo transfer, IVF, loading of transfer catheters, metaanalysis 108

119 Introduction Embryo transfer is the last stage of a series of complementary processes leading to success, or failure, of an IVF cycle. Over the years, it has been accepted as a non-decisive phase, as compared with other aspects of the IVF process. Even so, recently it has been shown that the pregnancy rates are strongly affected by the embryo transfer technique, and this has reflected in an increased interest and awareness among clinicians. The pregnancy rate after embryo transfer is dependent upon multiple factors including embryo quality, endometrial receptivity and the technique of the embryo transfer itself (1). In recent years, more stress has been placed on optimizing and standardizing the embryo transfer protocol. Factors such as catheter choice (2, 3), ease of the procedure (4), ultrasound-guidance (5), and dummy embryo transfer (6) have proved to improve the clinical outcomes. Consequently, any modification in the standard protocol that will improve the outcomes has great value. The use of air brackets around the embryo-containing medium has, in theory, been debated to be beneficial to the success of the embryo transfer, by protecting the embryos from the cervical mucus and from accidental discharge before entering the endometrial cavity. On the other hand, supporters of the fluid-only method of catheter loading believe that the introduction of even a small amount of air in the uterus could be a non-physiological factor with a deleterious effect on the embryos and implantation (7). In light of this controversy, and the need to clearly identify the relative efficacy of different embryo catheter-loading techniques, it was decided that there is a need for a systematic review to locate and analyse the best available evidence today in the medical literature. 109

120 Materials and methods Criteria for considering studies for this review All published, unpublished and ongoing randomized trials that had reported data comparing outcomes for women undergoing embryo transfer through the cervical route using air brackets (e.g. air fluid) versus fluid-only methods of embryo catheter loading, following IVF or intracytoplasmic sperm injection (ICSI), were sought in all languages. Types of outcome measures The primary outcome measures were the live birth (LBR), ongoing pregnancy (OPR) and clinical pregnancy (CPR) rates. The secondary outcomes were the rates of embryo implantation, multiple and ectopic pregnancies and miscarriage rates. Lastly, the tips of the post-transfer catheters were evaluated for retained embryos. Search strategy for identification of studies Meticulous computerized searches (last performed July 2006) were conducted using Medline (PubMed) (1966 to present), EMBASE (1980 to present), the Cochrane Central Register of Controlled Trials (CENTRAL) on the Cochrane Library Issue 3, 2006, the National Research Register (NRR), and the trial register of controlled trials ( Furthermore, the reference lists of all known primary studies, review articles, citation lists of relevant publications, abstracts of major scientific meetings, e.g. European Society for Human Reproduction & Embryology (ESHRE) and American Society of Reproductive Medicine (ASRM) and included studies were examined to identify additional relevant citations. Finally, ongoing and unpublished trials were sought by contacting experts in the field and commercial entities. Methods of the review A standardized data extraction form was developed and piloted for consistency and completeness. Trials were considered for inclusion, and trial data extracted. Data management, statistical analyses and power calculations were conducted using the Review Manager (RevMan) 4.2 and Power and Sample Size Calculations (PS) statistical software packages. 110

121 Individual outcome data were included in the analysis if they met the pre-stated criteria. Where possible, data was extracted to allow for an intention-to-treat analysis, defined as including all randomized cycles in the denominator. If data from the trial reports was insufficient or missing, the investigators of individual trials were contacted for additional information, in order to perform analyses on an intention-totreat basis. For the meta-analysis, the number of participants experiencing the event in each group of the trial was recorded. Heterogeneity of the included studies was determined by visual inspection of the outcome tables and by using the chi-squared test for heterogeneity. Meta-analysis of binary data was performed using the Mantel Haenszel method utilizing a fixed effect model and the odds ratio (OR), and 95% confidence intervals (CI), evaluated. 111

122 Results Search results A total of three prospective randomized controlled trials were identified (7-9) (Figure 1). Of these studies, two were published as full-text articles in peer-reviewed journals and one was published as an abstract in a conference proceeding. One study was excluded because it compared the standard air-fluid method (e.g. two air bubbles used to bracket the embryo containing media) to a modified air-fluid method, using only one air bubble at the tip of the embryo transfer catheter (9). The methodological quality of the remaining trials was assessed and data extracted to allow for an intention-to-treat analysis. Description of included studies Krampl et al. (7) conducted a prospective, randomized controlled trial including 196 women undergoing 196 embryo transfer cycles. Patients were allocated at the time of embryo transfer into one of two groups by an unclear form of randomization, with an equal number in each group. None of the reported cycles were frozen embryo replacement or oocyte donation cycles. In the air-fluid group, embryos were loaded as follows: 10 ml of air in the proximal part of the catheter, followed by 5 10 ml of medium containing the embryos to be transferred, and 10 ml of air at the tip of the catheter. In the fluid-only group, the syringe and the entire catheter were filled with medium and the embryo-containing medium (5 10 ml) was aspirated without being bracketed by air spaces. Moreno et al. (8) conducted a prospective, randomized controlled trial including 102 women undergoing 102 embryo transfer cycles. Patients were allocated at the time of embryo transfer into one of two groups by a computer-generated randomization table, the results of which were placed in consecutively numbered and sealed opaque envelopes. The number of cycles in each arm was as follows: air-fluid group (52 women); fluid-only group (50 women). None of the reported cycles were frozen embryo replacement or oocyte donation cycles. In the air-fluid group, embryos were loaded as follows: 200 ml of air in the syringe, ml of air in the proximal part of the catheter, ml of medium containing the embryos to be transferred, and 10 ml of air at the tip of the catheter. In the fluid-only group, the syringe and the entire catheter were filled with medium and the embryo-containing 112

123 medium (20 25 ml) was aspirated without being bracketed by air spaces. In both groups, the embryos were positioned about one-fourth of the way along the column from the catheter tip. Outcome measures With regards to the primary outcome measures, there was no significant difference between the two methods in live birth rate (19/52 versus 15/50; OR = 1.34; 95% CI = ) (Figure 2) and clinical pregnancy rate (54/150 versus 49/148; OR = 1.13; 95% CI = ) (Figure 3). For the secondary outcomes, there were also no significant differences between the two methods in rate of embryo implantation (31/127 versus 24/128; OR = 1.40; 95% CI = ), multiple pregnancy (1/22 versus 0/17; OR = 2.44; 95% CI = ), ectopic pregnancy (0/150 versus 3/148; OR = 0.14; 95% CI = ), miscarriage (3/22 versus 2/17; OR = 1.18, 95% CI = ), and retained embryos (1/52 versus 1/50; OR = 0.96; 95% CI = ). 113

124 Discussion Embryo transfer is the final stage in the IVF cycle with manual clinical manipulation. Although most patients who undergo assisted reproduction with IVF or ICSI will reach the embryo transfer stage with good quality embryos available for replacement, embryo implantation remains the rate-limiting step in the success of this form of therapy. The aim of the embryo transfer procedure is to atraumatically and accurately place embryos within the uterus in order to allow for proper implantation and fetal development. The embryo transfer procedure can be broken down into several parts, all of which are under the control of the clinician. In the past, the choice of catheters, ultrasound guidance, and other aspects (e.g. catheterloading techniques) were left up to personal preferences. Today, in the era of evidence-based medicine, the performance of each step should be scrutinized in order to standardize and perfect this obviously inefficient technique. In order to ascertain the importance of each step involved in the embryo transfer procedure, individual factors must be evaluated independently. In addition, since it would be difficult to accurately compare several factors at the same time, it was decided to concentrate on only one factor, the possible beneficial or detrimental role of using air brackets around the embryo-containing medium. Embryo transfer protocols have shown that different combinations of air and fluid volumes may be introduced in the catheter during embryo loading. In general, there are two accepted catheter-loading techniques, the air-fluid and the fluid-only models. In the air-fluid model, the loading of the syringe catheter complex with the transferred volume consists of the transfer media (which contains the embryos) separated by air spaces on both sides. In the fluid-only model, the embryos are placed in a complete column of fluid, without any air brackets or bubbles (10). In addition, a new modification of the standard air-fluid loading technique has been suggested which uses only one air bubble at the tip of the embryo transfer catheter instead of two air bubbles to bracket the embryo containing media (9), therefore introducing a lesser amount of air into the uterine cavity. To this day, the use of air brackets is controversial, and the evidence to support its use or avoidance is lacking. Some clinicians prefer the use of air brackets so that the embryo-containing media is easily identified in 114

125 the catheter, and on ultrasound. In addition, it has a psychological benefit for both the clinician and patient. The physician is guaranteed that the embryos will not be released from the catheter prematurely before proper placement in the uterus (11), or move up towards the syringe, therefore increasing the risk of being retained. Also the added visibility on ultrasound helps to detect the embryo catheter tip, in order to allow proper distancing from the uterine fundus. As for the patients, they are given the added comfort of visualizing the embryo-containing droplet on ultrasound through the detection of the two air bubbles surrounding the media. Even though some clinicians support the use of air columns to identify the positioning of embryos in the uterus, others suggest that the presence of air could increase the likelihood of embryo entrapment and increase reactive oxygen species, movement of embryos to other areas away from the uterus, or the occurrence of retained embryos within the catheter (7). It is important to note that these theories have yet to be proven, and the results of this systematic review provide evidence of clinical equivalence with regards to the two loading methods. In general, systematic reviews and meta-analyses of randomized controlled trials have proven to be the highest level of evidence in the hierarchy of medical knowledge. Even so, publication and search biases may confound the results of any systemic review, as studies showing positive results are more likely to be published (12, 13). Therefore, every effort has been made to avoid bias by searching a wide variety of databases, including Medline, EMBASE, the Cochrane Library, with no language barriers, in addition to hand searching the abstract books of major conferences (e.g. ASRM, ESHRE), reference lists of review articles and included trials. Even with all these precautions, only a limited number of trials were retrieved from the literature. This is evidence for the need for more trials to be properly planned, implemented and published regarding the different steps of the embryo transfer technique. Another important issue in clinical trials and systematic reviews is sample size. Since none of the included studies performed power calculations a priori, and due to their small sample sizes, it is postulated that they did not have sufficient power to detect minor differences between the study groups. Moreno et al. (8) determined from the 115

126 results of their implantation rates that a sample size of more than 800 patients in each arm would be necessary. Moreover, since clinical pregnancy rates are a more accepted outcome measure than implantation rates, it was decided to determine the properly necessary sample size. In order to detect a 5% difference between the air-fluid and fluid-only methods, it was determined that 1,307 women would be needed in each arm to provide an 80% statistical power of avoiding a type II error and a 5% chance of making a type I error in a two-tailed analysis (assuming a clinical pregnancy rate of 30% with the air-fluid method and a significance level of 0.05). The final outcome of this systematic review demonstrates that until today, even though there is no proven beneficial role to the use of air brackets when compared with the fluid-only method of embryo loading, there is also no evident detrimental role. In addition, it once again indirectly supports the notion of both clinical and publication bias against the all-so-important embryo transfer procedure in the literature. This is, of course, owing to the small number of studies planned, executed and published available to date. It is hoped that this trend will change in the near future, and the importance of each step in the embryo transfer technique will be further realized and scrutinized in order to create a truly evidence-based protocol for the transfer of human embryos following assisted reproduction. In conclusion, there is insufficient evidence to suggest the superiority of the air-fluid or fluid-only methods during embryo loading. There is also a strong need for more well-designed and powered randomized controlled trials in order to truly determine any possible beneficial, or detrimental, effect to either method. 116

127 Acknowledgements The author would like to thank all the corresponding authors that were contacted for more information and provided us with assistance. A special thanks is in order to Dr. Juan Balasch, whose direct assistance has helped to increase the accuracy of this systematic review by providing missing information. 117

128 References 1. Mansour R, Aboulghar M 2002 Optimizing the embryo transfer technique. Human Reproduction 17, Abou-Setta AM, Al-Inany HG, Mansour RT et al. 2005a Soft versus firm embryo transfer catheters for assisted reproduction: a systematic review and metaanalysis. Human Reproduction 20, Abou-Setta AM 2006 Firm embryo transfer catheters for assisted reproduction: a systematic review and meta-analysis. Reproductive BioMedicine Online 12, Lesny P, Killick SR, Tetlow RL et al Embryo transfer can we learn anything new from the observation of junctional zone contractions? Human Reproduction 13, Abou-Setta AM, Al-Inany HG, Mansour RT et al. 2005b Ultrasound-guided versus clinical touch embryo transfer: a systematic review and meta-analysis. Fertility and Sterility 84 (suppl. 1), S51-S Mansour R, Aboulghar M, Serour G 2000 Dummy embryo transfer: a technique that minimizes the problems of embryo transfer and improves the pregnancy rate in human in vitro fertilization. Fertility and Sterility 54, Krampl E, Zegermacher G, Eichler C et al Air in the uterine cavity after embryo transfer. Fertility and Sterility 63, Moreno V, Balasch J, Vidal E et al Air in the transfer catheter does not affect the success of embryo transfer. Fertility and Sterility 81, Allahbadia GN, Gandhi GN, Kadam KS et al A prospective randomized comparison of two different embryo transfer catheter loading techniques. Fertility and Sterility 84 (suppl. 1), S Lee HC, Seifer DB, Shelden RM 2004 Impact of retained embryos on the outcome of assisted reproductive technologies. Fertility and Sterility 82, Marek DE, Langley MT, Pultorak MJ et al Incidence of retained embryos following embryo transfer when a full column of media is utilized to expel the embryos compared with a column of air. Fertility and Sterility 82 (suppl. 2), S Easterbrook PJ, Berlin JA, Gopalan R, Matthews DR 1991 Publication bias in clinical research. Lancet 337, Dickerson K, Min YL, Meinert CL 1992 Factors influencing publication of research results Journal of the American Medical Association 267,

129 Figure 1. QUOROM statement flow diagram. Potentially relevant RCTs identified and screened for retrieval (n = 3) RCTs excluded, (n = 0) RCTs retrieved for more detailed evaluation (n = 3) Potentially appropriate RCTs to be included in the meta-analysis (n = 2) RCTs excluded, (compared standard airfluid method to modified method using only one air bracket at tip of catheter) (n = 1) RCTs withdrawn, (n = 0) RCTs included in meta-analysis (n = 2) RCTs excluded, (n = 0) RCTs with usable information, by outcome (n = 2) RCT = randomized controlled trial. 119

130 Figure 2. Meta-analysis forest plot showing live birth rates. Figure 3. Meta-analysis forest plot showing clinical pregnancy rates. 120

131 Chapter 9 What is the best site for embryo deposition? A systemic review and meta-analysis. Ahmed M. Abou-Setta Reproductive Biomedicine Online. 2007;14:611-9.

132 Abstract The site of embryo replacement has been postulated as being important to the success of IVF/ICSI. In order to determine the best site for embryo deposition during embryo transfer, a meta-analysis of randomized trials comparing different uterine deposition sites was undertaken. Electronic (e.g. PubMed, EMBASE, Cochrane Library, LILACS) and hand searches were performed to locate trials. Outcomes measures were the live-birth (LBR), ongoing pregnancy (OPR), and clinical pregnancy rates (CPR). Assesments of the endometrial cavity length (ECL) and the distance from the fundus to the tip of the catheter (DTC) were utilized. Six studies were identified, of which three were excluded. Meta-analysis was conducted with the Mantel-Haenszel method, utilizing the fixed-effects model. The LBR and OPR showed an increasing trend when transfers were performed to the lower half of the uterine cavity. For the DTC, all rates were significantly higher for the ~20 mm versus ~10 mm distance from the uterine fundus, supporting the results of the ECL analysis. The results of this systematic review show that there is limited evidence of the superiority of lower cavity transfers (e.g. ~20 mm) compared with the traditional high cavity (e.g. ~10 mm) transfers. More well-designed and powered randomized trials are needed to confirm this conclusion. Key words: embryo transfer, endometrial cavity length, IVF, metaanalysis, pregnancy rates 122

133 Introduction Treatment of infertile couples through IVF has shown remarkable improvements since it was first described by Steptoe and Edwards (1). Since then, almost all aspects of the IVF procedure have been optimized in order to increase pregnancy rates through increased efficiency and effectiveness of the different aspects involved. Thus far, this is only partially true for the embryo transfer technique. Ever since it was first described, few modifications have been made to this inefficient procedure. At present there is no standard, worldwide-accepted and evidencebased clinical protocol for the intrauterine transfer of embryos. In contrast to other aspects of the IVF procedure which have been more thoroughly addressed in the literature, most clinicians have relied on prior experience and personal preference when performing embryo transfer. In addition, historically, little attention has been paid to the embryo transfer procedure. This fact is reflected by the dearth of scientific publications regarding embryo transfer compared with other aspects of IVF (e.g. ovulation induction), and also the reluctance of physicians to modify their own personal habits to encompass a more evidence-based approach. Physicians, too, often underestimate the importance of the embryo transfer technique, being an apparently simple manoeuvre. Most inexperienced clinicians do not consider inserting a catheter through the uterine cervix and ejecting embryo-containing fluid to be a difficult task. This task may even be compared with the simple hystersalpingography procedure done by most first year gynaecology residents. Nevertheless, it has been shown that the attitude of physicians towards the embryo transfer technique are changing for the better (2, 3). Recently, the techniques and variables affecting the success of embryo transfer have attracted more attention. Today, in light of global trends such as single embryo transfer (SET), more emphasis has been placed on optimizing and standardizing the embryo transfer protocol than ever before (2, 3). The pregnancy rate after embryo transfer is dependent upon multiple factors, including embryo quality, endometrial receptivity and the technique of the embryo transfer itself (4). In recent years, more emphasis has been placed on optimizing and standardizing the embryo 123

134 transfer protocol. Factors such as catheter choice (5, 6), ease of the procedure in order to prevent endocervical and endometrial damage (7, 8), ultrasound-guidance (9-11), flushing the endometrium (12) and dummy embryo transfer (13) have proven to affect clinical outcomes. In order to ascertain the importance of each step involved in the embryo transfer procedure, individual factors must be evaluated independently. Consequently, any modification in the standard protocol that will improve the outcomes is of great value. The influence of the depth of embryo replacement into the uterine cavity has been postulated as being one of the most important factors to the success of an IVF treatment cycle (14). Traditionally, most IVF programmes have relied on the clinician s clinical touch for the placing of the transfer catheter within the uterine cavity at a point near the fundus (3, 15). Today this procedure has been modified to utilize ultrasound guidance to direct the placement of the catheter tip, allowing for more accurate placement (11). At present, the best site of embryo deposition is still not clear and remains highly debated. This debate has been fuelled by conflicting results from published clinical trials. Therefore, in light of this controversy and the need to clearly identify the best site for catheter placement and embryo deposition during embryo transfer, it was decided to systematically locate, analyse and review the best available current evidence on the site of embryo deposition. 124

135 Materials and methods Criteria for considering studies for this review All published, unpublished and ongoing randomized trials reporting data that compared outcomes for women undergoing embryo transfer through the cervical route following IVF, or intracytoplasmic sperm injection (ICSI), and randomized according to distance to the tip of the catheter (DTC) and the uterine fundus or endometrial cavity length (ECL) were sought, in all languages. Types of outcome measures The outcome measures for this systematic review were the live-birth (LBR), ongoing pregnancy (OPR), and clinical pregnancy (CPR) rates. Search strategy for identification of studies Meticulous computerized searches were conducted using MEDLINE ( July 2006), EMBASE (1980 July 2006), the Cochrane Central Register of Controlled Trials (CENTRAL) on the Cochrane Library Issue 3, 2006, the National Research Register (NRR), and the Trial Register of Controlled Trials ( and the Latin American and Caribbean Health Sciences Literature database (LILACS). The following Medical Subject Headings (MeSH) and text words were used: embryo transfer, embryo transfer technique, ultrasound, ultrasoundguided embryo transfer, endometrial cavity length, embryo implantation, and randomized controlled trial(s). Furthermore, the reference lists of all known primary studies, review articles, citation lists of relevant publications, abstracts of major scientific meetings (e.g. ESHRE and ASRM) and included studies were examined to identify additional relevant citations. Finally, ongoing and unpublished trials were sought by contacting experts in the field and commercial organisations. Methods of review A standardized data extraction form was developed and piloted for consistency and completeness. Trials were considered for inclusion, and trial data extracted. Data management and statistical analyses were conducted using the Review Manager (RevMan) 4.3 and Power and Sample Size Calculations (PS) statistical software packages. 125

136 Individual outcome data were included in the analysis if they met the pre-stated criteria. Where possible, data was extracted to allow for an intention-to-treat analysis; defined as including in the denominator all randomized cycles. If data from the trial reports was insufficient or missing, the investigators of individual trials were contacted via for additional information, in order to perform analyses on an intentionto-treat basis. For the meta-analysis, the number of participants experiencing the event in each group of the trial was recorded. Heterogeneity of the included studies was determined by visual inspection of the outcome tables and by using the I 2 -test for heterogeneity. In addition, the I 2 test was used to try to quantify any apparent inconsistency. The I 2 test is a statistical measure used to quantify heterogeneity. It describes the percentage of the variability within effect estimates that is due to heterogeneity rather than sampling error (chance) (16). An I 2 value greater than 50% may be considered to represent substantial heterogeneity. Comparison methods For the meta-analysis, the number of participants experiencing the event in each group of the trial was recorded. Two comparative methods were used for evaluation: the direct (head-to-head) and the adjusted indirect comparison methods. For the direct comparisons, comparison of the result of group B with the result of group C within a randomized controlled trial gave an estimate of the efficacy of intervention B versus C. Direct comparison was undertaken using the Mantel-Haenszel method utilizing the fixed-effects model, and the odds ratio and 95% confidence interval (CI) evaluated. If direct comparison was not possible due to the lack of available trials comparing group B with group C, then adjusted indirect comparison was performed using the method described by Bucher et al. (17, 18). The indirect comparison of intervention B and C was adjusted by the results of their direct comparisons with a common intervention A. In brief, given two estimated effects θ AB and θ AC for comparisons of group A versus group B (AvB) and group A versus group C (AvC), respectively, then the effect for the comparison of group B versus C (BvC) is estimated as follows: θ BC = θ AB θ AC, and its variance is var(θ BC ) = var(θ AB ) + var(θ AC ). A 95% confidence interval for θ BC is 126

137 obtained as θ BC ± 1.96 [var(θ BC )]. The estimates of effect, denotes θ, relate to scale on which the data would be analyzed; in this case being the log odds ratio. Search results A total of six prospective randomized controlled trials were identified (four full-text manuscripts and two conference abstracts) (Figure 1). Subsequently, three trials were excluded due to duplicate publication [e.g. publication as a conference abstract and full-text manuscript (19, 20), or double publication as full-text manuscripts (21)]. The remaining three trials were included, and the methodological quality of each trial assessed. Finally, data was extracted to allow for an intention-to-treat analysis. Description of included studies Nazari et al. (22) conducted a randomized controlled trial including 1008 women undergoing 1590 embryo transfer cycles. All cycles were performed using the clinical touch method of catheter placement. In order to determine the best site for embryo deposition, patients were randomly allocated into one of two groups: embryo placement <5 mm (group I), and embryo placement >15 mm (group II) from the uterine fundus. The length of the uterine cavity was determined during a mock embryo transfer. For the purpose of this systematic review, groups I and II will be denoted as N-I and N-II respectively. The cycles were randomly allocated using a 2:1 ratio: N-I (n = 660 cycles) and N-II (n = 930 cycles). The transfer procedures were performed by five physicians (two allocated to N-I and three allocated to N-II), and both fresh and frozen embryo replacement (FER) cycles were included in the analysis. Only one catheter type was used for all transfers (Short Frydman Embryo Transfer Set; Laboratoire CCD, Paris, France). Physicians were allocated to the two groups by randomly picking numbered envelopes. Clinical pregnancy rates were defined as the presence of a gestational intrauterine sac on ultrasound. Coroleu et al. (23) conducted a randomized controlled trial including 180 women undergoing 180 embryo transfer cycles. All cycles were performed using ultrasound guidance to guide catheter placement. In order to determine the best site for embryo deposition, patients were randomly allocated into one of three groups, using a computer- 127

138 generated randomization table: embryo placement 10 ± 1.5 mm (group I), embryo placement 15 ± 1.5 mm (group II), and embryo placement 20 ± 1.5 mm (group III) from the uterine fundus. The site of deposition was determined during the actual embryo transfer by the use of ultrasonography. For the purpose of this systematic review, groups I, II and III will be denoted as C-I, C-II and C-III respectively. The cycles were randomly allocated using a 1:1:1 ratio: C-I (n = 61 cycles), C-II (n = 59 cycles), and C-III (n = 60 cycles). The transfer procedures were done by one physician, with the same catheter type used in all transfers (Edwards-Wallace Embryo Replacement Catheter; Smiths Medical International Ltd., Kent, UK), and only fresh, non-donor cycles were included in the analysis. Clinical pregnancy rates were defined as the presence of a gestational intrauterine sac on ultrasound. Franco et al. (24) conducted a prospective, randomized controlled trial including 360 women undergoing 400 embryo transfer cycles. All cycles were performed using ultrasound guidance to guide catheter placement. In order to determine the best site for embryo deposition, patients were randomly allocated into one of two groups, by drawing lots from a random-number table, depending on the site of embryo deposition. In all cases the endometrial cavity length (ECL) was measured immediately before transfer. In group I, embryos were deposited in the upper half of the endometrial cavity (e.g. <50% ECL) and in group II embryos were deposited in the lower half of the endometrial cavity (e.g. 50% ECL). The site of deposition was determined during the actual embryo transfer by the use of ultrasonography. For the purpose of this systematic review, groups I and II will be denoted as F-I and F-II respectively. The cycles were randomly allocated using a 1:1 ratio: F-I (n = 200 cycles) and C-II (n = 200 cycles). The transfer procedures were carried out by the same physician, with the same catheter type used in all transfers (Frydman Embryo Classic Catheter 4.5), and only fresh, non-donor cycles were included in the analysis. Clinical pregnancy rates were defined as the presence of a gestational intrauterine sac with cardiac activity on ultrasound. Ongoing pregnancy rates were defined as a viable fetus on ultrasound at 12 weeks. 128

139 Results Table 1 provides an assessment of the quality of the included trials. An a priori sample size calculation, to determine the adequate sample size needed, was performed by only one study (22). The method of randomization and randomization concealment was clear and adequate in two studies (23, 24). In addition, all studies used an intention-totreat analysis when analysing their data. Finally, possible confounders were obvious in the study by Nazari et al. (22). They included the use of both fresh and frozen embryo transfer cycles and multiple (n = 5) physicians performing the embryo transfer. In the original publications, none of the included studies presented the live-birth rate, but two studies (23, 24) presented the ongoing pregnancy rate. The missing ongoing pregnancy and live-birth rates for all the studies were provided following personal contact with the respective authors. Table 2 shows the intended distance of embryo disposition from the uterine fundus and the actual average distance of embryo deposition for the included studies. It is noticeable that even though physicians intended to deposit the embryos at an exact distance from the uterine fundus, factors including endometrial wave contractions and hydraulic force, served to move the embryos up or down a few millimetres from the intended site. Table 3 shows the numbers of clinical pregnancies, ongoing pregnancies and live births in the original studies. Two analyses were performed in an attempt to locate the best site of embryo deposition. The first was performed by dividing the averagesized uterus into two parts (i.e. upper half and lower half) according to the endometrial cavity length (ECL). In a second series of analyses, the position of the catheter tip was examined with regard to the distance from the fundus to the tip of the catheter (DTC). Finally, a subgroup analysis was performed for studies that used ultrasound-guidance during embryo transfer, as this is considered the proper method of determining the exact site of embryo deposition. Assessment of endometrial cavity length The average uterine endometrial cavity length (ECL) is considered to be ~30 mm (21). Therefore, the individual study groups were allocated to transfer to the upper half (i.e. ECL <50% or <15 mm) or the lower half (i.e. ECL >50% or >15 mm) of the uterus. This analysis was performed twice using two different algorithms (Table 4). The site of embryo 129

140 deposition was not significantly different with regard to the live-birth, ongoing pregnancy, and clinical pregnancy rates between the two groups in either analysis. Even so, it should be noted that there was an apparent trend with regard to transfers to the lower half of the uterine cavity. Assessment of distance from the fundus to the tip of the catheter (DTC) For the purpose of these analyses, the uterine cavity was arbitrarily divided into five regions as follows: DTC-I = <7.25 mm, DTC-II = 10 ± 2.5 mm, DTC-III = 15 ± 2.5 mm, DTC-IV = 20 ± 2.5 mm and DTC-V = >22.5 mm (Table 5). It is important to note that an additional section (i.e. DTC-III/DTC-IV) was added to allow for a more appropriate analysis of the available data. Furthermore, each section was compared with the other sections and the results are presented in Table 6. In this series, only the pooled analysis for live-birth, ongoing pregnancy and clinical pregnancy rates were significantly different for the analyses DTC-II (e.g. 10 ± 2.5 mm) versus DTC-IV (e.g. 20 ± 2.5 mm) (for ongoing pregnancy rates OR = 0.43, 95% CI = and for clinical pregnancy rates OR = 0.43, 95% CI = ). In addition, the analysis DTC-I versus DTC-III/IV showed significance with regard to the live-birth (OR = 0.70, 95% CI = ) and ongoing pregnancy rates (OR = 0.71, 95% CI = ), but this analysis was built on data from only one trial. The other analyses were not significantly different. Subgroup analysis In the analyses of the subgroups of studies that used ultrasound-guided embryo transfer, there were no differences in the clinical pregnancy, ongoing pregnancy or live births in the two ECL assessments, therefore denoting no significance between embryo deposition in the upper or lower halves of the uterine cavity. As for the DTC assessments, only the DTC-II versus DTC-IV analyses showed statistical significance with respect to the clinical pregnancy (OR = 0.43; 95% CI = ), ongoing pregnancy (OR = 0.43; 95% CI = ) and live-birth (OR = 0.42; 95% CI = ) rates. It is important to note that this result was generated from only one included study (22). 130

141 Discussion Embryo transfer is the final stage of the IVF cycle. It is also the instance where clinical manipulations can directly alter the outcome of the IVF cycle, and has shown marked variability among different IVF programmes, and also among physicians within in the same programme (25, 26). Although most patients who undergo assisted reproduction, via IVF or ICSI, will reach the embryo transfer stage with good quality embryos available for replacement, embryo implantation remains the ratelimiting step in the success of this form of therapy. The aim should be to place embryos meticulously and accurately within the uterus, in order to allow for proper implantation and fetal development (14, 27). The site of embryo placement in the uterine cavity has been suggested to directly influence the embryo implantation rates. Nevertheless, this clinical issue is not clearly addressed in the literature, with some authors recommending the tip of the embryo transfer catheter be placed ~10 mm below the fundal endometrial surface (28, 29) or close to the uterine fundus (27, 30), and others suggesting that improved results may be obtained when the embryos are placed at lower levels in the uterine cavity (7, 14, 22, 31-36). Finally, some authors postulate that the site of embryo transfer is of no importance, since it does not influence implantation, as long as embryos are placed in the upper half of the cavity (22, 37, 38). Ironically, the site of transfer has also been related to poor pregnancy outcomes. Transferring embryos within 1 cm of the uterine endometrial fundus has been criticized as being associated with a higher than normal tubal ectopic pregnancy rate (22, 39). At the same time, low implantations have been shown to have a higher rate of spontaneous abortion, and cervical ectopic pregnancies (27, 37). Another important factor is the use of ultrasound guidance in order to accurately place the tip of the embryo transfer catheter at the required site of embryo deposition. In addition to the accuracy provided by ultrasound visualization of the uterus, ultrasound-guided embryo transfer has been shown to improve the clinical pregnancy rate (9, 10) and live-birth rates (11) when compared with the traditional clinical touch method. Even so, this has been challenged by a recent large randomized trial that demonstrated similar clinical pregnancy and livebirth rates between the two methods (40). Nonetheless, there is no 131

142 disagreement of the fact that only with ultrasound guidance can the clinician confidently determine the site of deposition. Two of the included studies (23, 24) utilized ultrasound guidance to determine the exact site of embryo deposition, while one study (22) used the traditional clinical touch method. The results of the subgroup analyses of only the studies using ultrasound guided embryo transfer did not differ from the results of the original analyses. More importantly, the importance of the site of embryo deposition revolves around locating the best site for optimum embryo implantation in the human uterus. In a prospective study, it was documented that following embryo transfer, approximately 80% of embryos implant in areas to which they are transferred initially and approximately 20% implant in other areas (41). Furthermore, using three-dimensional transvaginal ultrasound, Minami et al. (42) determined that the upper region of the uterine cavity contained the majority of early gestational sacs in an unselected population of pregnant women. Moreover, the miscarriage rate was significantly lower when the early sac was found in the upper region than in the middle and lower regions. In another recent study, Cavagna et al. (43) studied 63 pregnancies following embryo transfer to the middle point of the endometrial cavity. This study documented that in singleton pregnancies, 66.0% of the gestational sacs were detected in the upper region, 29.8% in the middle region and 4.2% in the lower region. In multiple pregnancies, the rates were 45.5, 51.5 and 3.0% respectively. These results demonstrate that even when embryos are transferred to the central area of the uterine cavity, there is still a high chance of embryo implantation in the upper region. Since systematic reviews and meta-analysis of randomized controlled trials have been proven to provide the highest level of evidence in the hierarchy of medical knowledge, it was decided to test the soundness of these theories. During this systematic review, both direct and adjusted indirect analyses were utilized. Since it is well accepted that well designed prospective, randomized, controlled trials provide the most valid evidence of relative efficacy of competing interventions, they are utilized first and foremost in evaluations. Nevertheless, many competing interventions have not been compared directly (head-to-head) in randomized trials. Because of the lack of direct evidence, indirect 132

143 comparisons have been recommended and used for evaluating the efficacy of alternative interventions (44). Still, it may be argued that this review might not have sufficient sample sizes to detect minor differences between study groups. Study power and adequate sample sizes help to prevent the occurrence of type II errors. This current meta-analysis included 2570 embryo transfer cycles. With this large sample size, the ECL assessments could detect a 6% difference with 80% power in a two-tailed analysis (assuming a clinical pregnancy rate of 30% and a significance level of 0.05). Even so, since the DTC analyses were divided into more groups, ranging from 59 to 1189 cycles, there was evident heterogeneity in the examined sample sizes. The results of this systematic review and meta-analysis show that (i) pregnancy rates are similar when the upper and lower halves of the endometrial cavity are compared, and (ii) mid-cavity transfer (e.g. ~20 mm) is superior to the traditional high transfer (e.g. ~10 mm). Even so, it is important to note that due to the sample size, firm conclusions cannot be made. Moreover, it is of the utmost importance to note that there was clinical heterogeneity between the included studies. Therefore a series of metaregression analyses were utilized to compare all aspects of embryo site deposition. Finally, even though every effort was used to locate trials in the medical literature, the final sample sizes in some of the calculations (e.g. DTC) tended to suggest a lack of statistical power, which brings in the possibility of type II errors. This will only be offset by the publication of more well designed and powered randomized controlled trials. In conclusion, the results of this systematic review demonstrate that live-birth, ongoing pregnancy and clinical pregnancy rates may be influenced by the site of the embryo deposition. With the available evidence at hand, it may be recommended to position the tip of the catheter in the middle area of the endometrial cavity rather than at the currently recommended ~10 mm from the uterine fundus. Even so, due to the heterogeneity of the included studies and in some cases, the relatively small sample sizes, there is only limited evidence to support the proper site of embryo deposition. More well designed, and powered randomized controlled trials comparing the standard ~10 mm distance and the new suggested ~20 mm distance are needed to support the results of this systematic review. In addition, studies are needed to 133

144 address surrogate outcomes such as the effect of embryo deposition site on the rate of ectopic and spontaneous miscarriage. 134

145 Acknowledgements The author would like to thank all the corresponding authors and their respective staff for providing additional unpublished data. A special acknowledgment is warranted for Juan Balasch MD, Jerome H Check MD, Buenaventura Coroleu MD, Jose Franco Jr MD, Joao Batista A Oliveira MD and Mrs Laurie Aurand, whose direct assistance is highly appreciated. 135

146 References 1. Steptoe PC, Edwards RG 1978 Birth after the reimplantation of a human embryo. Lancet 12, Kovacs GT 1999 What factors are important for successful embryo transfer after in-vitro fertilization? Human Reproduction 14, Salha OH, Lamb VK, Balen AH 2001 A postal survey of embryo transfer practice in the UK. Human Reproduction 16, Mansour R, Aboulghar M 2002 Optimizing the embryo transfer technique. Human Reproduction 17, Abou-Setta AM, Al-Inany HG, Mansour RT et al Soft versus firm embryo transfer catheters for assisted reproduction: a systematic review and metaanalysis. Human Reproduction 20, Abou-Setta AM 2006 Firm embryo transfer catheters for assisted reproduction: a systematic review and meta-analysis. Reproductive BioMedicine Online 12, Lesny P, Killick SR, Tetlow RL et al Embryo transfer can we learn anything new from the observation of junctional zone contractions? Human Reproduction 13, Cevrioglu AS, Esinler I, Bozdag G et al Assessment of endocervical and endometrial damage inflicted by embryo transfer trial: a hysteroscopic evaluation. Reproductive BioMedicine Online 13, Buckett WM 2003 A meta-analysis of ultrasound-guided versus clinical touch embryo transfer. Fertility and Sterility 80, Sallam HN, Sadek SS 2003 Ultrasound-guided embryo transfer: a meta-analysis of randomized controlled trials. Fertility and Sterility 80, Abou-Setta AM, Mansour RT, Al-Inany HG et al Among women undergoing embryo transfer, is the probability of pregnancy and live birth improved with ultrasound-guidance than with clinical touch alone? A systemic review and metaanalysis of prospective randomized trials. Fertility and Sterility, in press. 12. Berkkanoglu M, Isikoglu M, Seleker M et al Flushing the endometrium prior to the embryo transfer does not affect the pregnancy rate. Reproductive BioMedicine Online 13, Mansour R, Aboulghar M, Serour G 2000 Dummy embryo transfer: a technique that minimizes the problems of embryo transfer and improves the pregnancy rate in human in vitro fertilization. Fertility and Sterility 54, Naaktgeboren N, Broers FC, Heijnsbroek I et al Hard to believe hardly discussed, nevertheless very important for the IVF/ICSI results: embryo transfer technique can double or halve the pregnancy rate. Human Reproduction 12, i Schoolcraft WB, Surrey ES, Gardner DK 2001 Embryo transfer: techniques and variables affecting success. Fertility and Sterility 76, Higgins JPT, Thompson SG, Deeks JJ et al Measuring inconsistency in meta-analysis. British Medical Journal 327,

147 17. Bucher HC, Guyatt GH, Griffith LE et al The results of direct and indirect treatment comparisons in meta-analysis of randomized controlled trials. Journal of Clinical Epidemiology 50, Song F, Altman DG, Glenny AM et al Validity of indirect comparison for estimating efficacy of competing interventions: empirical evidence from published meta-analyses. British Medical Journal 326, Nazari A, Askari HA, Check JH et al. 1993a Embryo transfer technique as a cause of ectopic pregnancy in in-vitro fertilization. Annual Pacific Coast Fertility Society Meeting Conference Booklet. 20. Franco JG Jr, Martins AM, Baruffi RL et al. 2004a Increase of implantation and pregnancy rates with the embryo transfer in the middle area of endometrial cavity. Human Reproduction 19 (Suppl. 1), i Oliveira JB, Martins AM, Baruffi RL et al Increased implantation and pregnancy rates obtained by placing the tip of the transfer catheter in the central area of the endometrial cavity. Reproductive BioMedicine Online 9, Nazari A, Askari HA, Check JH et al. 1993b Embryo transfer technique as a cause of ectopic pregnancy in in-vitro fertilization. Fertility and Sterility 60, Coroleu B, Barri PN, Carreras O et al The influence of the depth of embryo replacement into the uterine cavity on implantation rates after IVF: a controlled, ultrasound-guided study. Human Reproduction 17, Franco JG Jr, Martins AM, Baruffi RL et al. 2004b Best site for embryo transfer: the upper or lower half of endometrial cavity? Human Reproduction 19, Karande V, Morris R, Chapman C et al Impact of the `physician factor on pregnancy rates in a large assisted reproductive technology program: do too many cooks spoil the broth? Fertility and Sterility 71, Hearns-Stokes RM, Miller BT, Scott L et al Pregnancy rates after embryo transfer depend on the provided at embryo transfer. Fertility and Sterility 74, Meldrum DR, Chetkowski R, Steingold KA et al Evolution of a highly successful in vitro fertilization embryo transfer program. Fertility and Sterility 48, Webster J 1986 Embryo replacement. In: Fishel, S, Symonds, EM (eds) In Vitro Fertilisation: Past, Present and Future. IRL Press, Oxford, pp Jones HW 1988 In vitro fertilization. In: Behrman SJ, Kistner RW, Patton GW (eds) Progress in Infertility, 3rd edn. Little, Brown and Co., Boston, pp Krampl E, Zegermacher G, Eichler C et al Air in the uterine cavity after embryo transfer. Fertility and Sterility 63, Waterstone J, Curson R, Parsons J 1991 Embryo transfer to low uterine cavity. Lancet 337, Woolcott R, Stanger J 1998 Ultrasound tracking of the movement of embryoassociated air bubbles on standing after transfer. Human Reproduction 13,

148 33. Naaktgeboren N, Dieben S, Heijnsbroek I et al Embryo transfer, easier said than done. Fertility and Sterility 70 (Suppl. 1), S Brinsden PR 1999 Oocyte recovery and embryo transfer. In: Brinsden PR (ed.) A Textbook of In Vitro Fertilization and Assisted Reproduction, 2nd edn. Parthenon, London pp Frankfurter D, Silva CP, Mota F et al The transfer point is a novel measure of embryo placement. Fertility and Sterility 79, van de Pas MMC, Weima S, Looman CWN et al The use of fixed distance embryo transfer after IVF/ICSI equalizes the success rates among physicians. Human Reproduction 18, Al-Shawaf T, Dave R, Harper J et al Transfer of embryos into the uterus: how much do technical factors affect pregnancy rates? Journal of Assisted Reproduction and Genetics 10, Roselund B, Sjöblom P, Hillensjö T et al Pregnancy outcome related to the site of embryo deposition in the uterus. Journal of Assisted Reproduction and Genetics 13, Yovich JL, Turner SR, Murphy AJ 1985 Embryo transfer technique as a cause of ectopic pregnancies in in-vitro fertilization. Fertility and Sterility 44, Drakeley AJ, Lunt R, Aust T et al A randomised trial of 2250 women having ultrasound guided embryo transfer. Human Reproduction 21 (Suppl. 1), i Baba K, Ishihara O, Hayashi N, et al Where does the embryo implant after embryo transfer in humans? Fertility and Sterility 73, Minami S, Ishihara K, Araki T 2003 Determination of blastocyst implantation site in spontaneous pregnancies using three-dimensional transvaginal ultrasound. Journal of Nippon Medical School 70, Cavagna M, Contart P, Petersen CG, et al Implantation sites after embryo transfer into the central area of the uterine cavity. Reproductive BioMedicine Online 13, McAlister F, Laupacis A, Wells G, Sackett D 1999 Users guides to the medical literature: XIX. Applying clinical trial results B. Guidelines for determining whether a drug is exerting (more than) a class effect. Journal of the American Medical Association 282,

149 Figure 1. QUOROM statement flow diagram. Potentially relevant RCTs Identified and screened for retrieval (n=6) RCTs excluded, duplicate publication (n=3) RCTs retrieved for more detailed evaluation (n=3) RCTs withdrawn (n=0) Potentially appropriate RCTs to be included in the meta-analysis (n=3) RCTs withdrawn (n=0) RCTs included in metaanalysis (n=3) RCTs withdrawn (n=0) RCTs with usable information, by outcome (n=3) RCT = randomized controlled trials. 139

150 TABLE 1. Review table of the study quality of the included trials. Quality Assessment A-priori Study sample size calculation Method of Randomization Method of Randomization Concealment Possible confounders Nazari et al. (22) Not performed Physicians randomly picking numbered Not performed ITT LBR envelopes Multiple physicians, FET included Coroleu et al. (23) Not performed Intention-totreat Followup Computergenerated randomization table Telerandomization OPR None ITT evident Franco et Performed al. (24) Random-number table Drawing lots at time of embryo ITT OPR None evident transfer ITT = Intention to treat analysis performed; FET = Frozen-thawed embryo replacement cycles. 140

151 TABLE 2. Review table of the site of expected and actual distance of embryo deposition from the uterine fundus as reported in the included studies. Study Group A priori transfer distance from Fundus Site of Embryo Deposition Actual transfer distance from Fundus Nazari et al. (22) N-I <5 mm NA N-II >15 mm NA Coroleu et al. (23) C-I 10 ± 1.5 mm 10.2 ± 0.9 mm C-II 15 ± 1.5 mm 14.6 ± 0.7 mm C-III 20 ± 1.5 mm 19.3 ± 0.8 mm Franco et al. (24) F-I <50 ECL 13.3 ± 1.6 mm F-II >50% ECL 18.3 ± 3.2 mm N-I, II = Group I, II in Nazari et al. (22), respectively. C-I, II, III = Group I, II and III in Coroleu et al. (23), respectively. F-I, II = Group I, II in Franco et al. (24), respectively. ECL = endometrial cavity length. NA = Not available. 141

152 TABLE 3. Review table of the clinical pregnancy and live-birth rates as reported in the included studies. Grou Study p Number of embryo transfers Number of Clinical Pregnancies (%) Number of Ongoing Pregnancies (%) Number of Live Births (%) Nazari et al. N-I (12.42%) 63 (9.55%) 61 (9.24%) (22) N-II (14.19%) 120 (12.90%) 118 (12.69%) Coroleu et al. (23) C-I (39.34%) 20 (32.79%) 19 (31.15%) C-II (49.15%) 26 (44.07%) 25 (42.37%) C-III (60.00%) 32 (53.33%) 31 (51.67%) Franco et al. F-I (35.00%) 58 (29.00%) 58 (29.00%) (24) F-II (29.50%) 50 (25.00%) 50 (25.00%) N-I, II = Group I, II in Nazari et al. (22), respectively. C-I, II, III = Group I, II and III in Coroleu et al. (23), respectively. F-I, II = Group I, II in Franco et al. (24), respectively. 142

153 TABLE 4. Distribution of groups according to the endometrial cavity length (ECL). ECL-I Study (<15 mm) ECL-II (>15 LBR OPR CPR mm) Nazari et al. N-I (22) N-II Coroleu et al. C-I (23) C-II/ C-III Analysis I Franco et al. F-I (24) F-II OR = %CI = 0.62 to 1.01 OR = %CI = 0.63 to 1.01 OR = %CI = 0.72 to 1.14 Nazari et al. N-I (22) N-II Coroleu et al. (23) Franco et al. (24) C-I/ C-III C-II F-I F-II OR = %CI = 0.63 to 1.01 OR = %CI = 0.63 to 1.01 OR = %CI = 0.72 to 1.13 Analysis II N-I, II = Group I, II in Nazari et al. (22), respectively. C-I, II, III = Group I, II and III in Coroleu et al. (23), respectively. F-I, II = Group I, II in Franco et al. (24), respectively. O.R = odds ratio; CI = confidence interval. CPR = clinical pregnancy rate; LBR = live birth rate; OPR = ongpoing pregnancy rate. 143

154 TABLE 5. Distribution of groups according to the distance from the fundus to the tip of the catheter (DTC). DTC-I Study (<7.25 mm) DTC-II (10 ± 2.5 mm) DTC-III (15 ± 2.5 mm) DTC-IV (20 ± 2.5 mm) DTC-V (>22.5 mm) Nazari et al. (22) N-I N-II Coroleu et al. (23) C-I C-II C-III Franco et al. (24) F-I F-II N-I, II = Group I, II in Nazari et al. (22), respectively. C-I, II, III = Group I, II and III in Coroleu et al. (23), respectively. F-I, II = Group I, II in Franco et al. (24), respectively. 144

155 TABLE 6. Review table of the results of direct and adjusted indirect meta-analysis according to the distance from the fundus to the catheter tip (DTC). (A) Live-birth rate DTC-II DTC-III DTC-IV DTC-III/ DTC-IV DTC-I OR = 0.76* 95%CI = 0.47 to 1.23* NA NA OR = %CI = 0.51 to 0.97 OR = 0.62 DTC-II 95%CI = 0.29 to 1.30 OR = %CI = 0.20 to 0.89 OR = %CI = 0.64 to 1.32 OR = 0.69 DTC-III NA 95%CI = 0.33 to 1.42 DTC-IV NA DTC-I = <7.25 mm; DTC-II = 10 ± 2.5 mm; DTC-III = 15 ± 2.5 mm; DTC-IV = 20 ± 2.5 mm; O.R = odds ratio; CI = confidence interval; * adjusted indirect analysis; NA = not applicable. 145

156 (B) Ongoing pregnancy rate DTC-II DTC-III DTC-IV DTC-III/ DTC-IV OR = 0.77* DTC-I NA NA 95%CI = 0.48 to 1.25* OR = %CI = 0.52 to 0.98 OR = 0.62 DTC-II 95%CI = 0.29 to 1.30 OR = %CI = 0.20 to 0.89 OR = %CI = 0.64 to 1.32 OR = 0.69 DTC-III NA 95%CI = 0.33 to 1.42 DTC-IV NA DTC-I = <7.25 mm; DTC-II = 10 ± 2.5 mm; DTC-III = 15 ± 2.5 mm; DTC-IV = 20 ± 2.5 mm; O.R = odds ratio; CI = confidence interval; * adjusted indirect analysis; NA = not applicable. 146

157 (C) Clinical pregnancy rate DTC-II DTC-III DTC-IV DTC-III/ DTC-IV OR = 0.88* DTC-I NA NA 95%CI = 0.57 to 1.36* OR = %CI = 0.64 to 1.15 OR = 0.67 DTC-II 95%CI = 0.33 to 1.38 OR = %CI = 0.21 to 0.90 OR = %CI = 0.69 to 1.38 OR = 0.64 DTC-III NA 95%CI = 0.31 to 1.33 DTC-IV NA DTC-I = <7.25 mm; DTC-II = 10 ± 2.5 mm; DTC-III = 15 ± 2.5 mm; DTC-IV = 20 ± 2.5 mm; O.R = odds ratio; CI = confidence interval; * adjusted indirect analysis; NA = not applicable. 147

158 148

159 Chapter 10 Ultrasound-guidance during embryo transfer: a prospective, single operator, randomized, controlled trial. Mamdoh Eskandar, Ahmed M. Abou-Setta, Mona A. Almushait, Mohamed El-Amin, Saria E. Y. Mohmad Fertiltiy & Sterility. (In-Press).

160 Abstract Objective: Ultrasound (US) assisted embryo transfer (ET) has been proposed to be more effective than the standard clinical touch (CT) method of catheter placement. Even so, heterogeneity between the published studies may be in part due to multiple operators. Design: Prospective, single-operator, randomized, controlled trial comparing ultrasound to clinical touch methods of embryo catheter guidance. Setting: Saudi Center for Assisted Reproduction. Patient(s): 378 women. Intervention(s): Transcervical, intrauterine embryo transfer with or without ultrasound guidance. Main outcome measure(s): Primary outcomes were the livebirth/ongoing pregnancy and clinical pregnancy rates per randomized woman. Secondary outcomes were the incidences of difficult transfers, blood and/or mucus on the catheter tip, spontaneous miscarriages and ectopic pregnancies. Result(s): Demographics and cycle characteristics were not different between the two groups. Live-birth/ongoing pregnancy rate was significantly higher in the US-ET group [68/183 (40.98%)] than the CT- ET group [50/190 (28.42%)] (O.R= 1.66, 95%CI= ). In addition, there was a significantly higher number of clinical pregnancies in the US-ET group [75/183 (40.98%)] than the CT-ET group [54/190 (28.42%)] (O.R= 1.75, 95%CI= ). Secondary outcomes were not significantly different between the two groups. Conclusion(s): Ultrasound-guided embryo transfer significantly increases the chance of ongoing/live-birth and clinical pregnancy rates compared to the clinical touch method. Key words: Ultrasound, clinical touch, embryo transfer, randomized controlled trial, in vitro fertilization 150

161 Background The majority of patients undergoing assisted reproduction through in vitro fertilization (IVF)/ intracytoplasmic sperm injection (ICSI) will reach the transfer stage, but a small proportion of them will achieve a clinical pregnancy, an ongoing pregnancy or a live-birth (1, 2). Although there have been miraculous improvements in respect to pituitary downregulation and ovulation stimulation protocols, embryo culture medias, and laboratory techniques, embryo implantation remains the ratelimiting step in the success of IVF. Embryo implantation is considered by many to be the most important limiting factor to success in assisted reproduction. The main factors that are believed to affect embryo implantation are embryo quality, uterine receptivity, and more importantly the proper technique of embryo transfer (3). Ever since the birth of the first in-vitro fertilization (IVF) baby in 1978 (4), the technique of uterine embryo transfer has remained largely unchanged since it was first described. Even so, recently a renewed interest in the embryo transfer technique has evolved among clinician and researchers, alike. Factors such as the use of soft embryo transfer catheters (5) and ease of the technique have been shown to affect the clinical outcomes. A new innovation is the use of ultrasound-guidance during embryo transfer in order to improve the ease of transfer, and precision in determining the site of transfer when compared to the standard clinical touch method. Recent systematic reviews have reported that embryo transfer under ultrasound guidance improves clinical pregnancy, ongoing pregnancy and live birth rates (6-8). Even so, there was heterogeneity between the included studies. This heterogeneity may be a result of individual transfer techniques, role of multiple practitioners performing embryo transfer, or study quality. As a result of the importance of the embryo transfer stage in determining the clinical outcomes for patients undergoing IVF treatment, it is vital for each program to properly evaluate their protocols and to adjust to new innovations in assisted reproduction treatment. Therefore we decided to perform a prospective, randomized, single operator, trial in order to determine if the implementation of ultrasound-guidance will improve the clinical outcomes in our patient population. 151

162 Materials and Methods This prospective, randomized trial was approved by our institutional review board. Three-hundred and seventy-three patients undergoing embryo transfer in our assisted reproduction unit between January 2005 and November 2006 were prospectively included. Patient population The study objectives were explained thoroughly to all prospective patients and their partners entering our assisted reproduction program. Couples who agreed to enter the clinical trial provided both verbal and written consent. Inclusion criteria were broad so as to truly represent our center s patient population. Women undergoing embryo transfer in a fresh cycle, and with good quality embryos on the day of transfer, were randomized to one of the two groups. Randomization was performed by the use of dark, sealed envelopes. Ovulation induction and IVF protocols All aspects of the IVF procedure were similar between the two groups, with the exception of the transfer technique. In brief, ovarian stimulation, oocyte retrieval and luteal phase support were performed in accordance with the standard protocol of our department. Women were down-regulated using a gonadotropin-releasing hormone agonist (GnRH-agonist) (Decapeptyl, Ferring NV, Belgium) protocol, followed by ovarian stimulation using recombinant FSH (rfsh, Puregon, NV Organon, Oss, The Netherlands) and/ or human menopausal gonadotrophin (Menogon, Ferring NV, Belgium) till the day of human chorionic gonadotrophin (HCG) administration. When the leading follicle reached ~18 mm in diameter, 10,000 IU of hcg (Pergnyl, NV Organon, The Netherlands) was given intra muscularly, and oocyte retrieval was performed hours later. Luteal phase support is provided in the form of daily progesterone vaginal suppositories tid (Cycologest 400 mg; Hoechst Roussel Limited, UK). Embryo transfer technique All embryo transfers were performed on day 3 by a single physician (M.A.E.) using a standardized technique. In all cases patients were asked to maintain a full bladder prior to transfer. Embryo transfer was performed using a Sydney IVF Embryo Transfer Set (K-JETS

163 SIVF) connected to a tuberculin syringe. A single air bubble was used to prevent the accidental expulsion of the embryo containing media. In addition, the catheter was held with its tip slightly downwards to prevent embryos from traveling through the liquid column to the end connected to the syringe. In both groups, the target was to deposit the embryos ~2 cm from the uterine fundus. In the ultrasound group, this was assisted with the use of transabdominal ultrasound guidance (2101 Falcon B-K Medical, 3.5 MHz probe) by a trained ultrasonographer. In the clinical touch group, catheter placement was accomplished using clinical sense and judgment. Following the transfer the catheter was extracted slowly and examined for retained embryos, blood and/ or mucus by the embryologist under a stereomicroscope. Outcome measures The primary outcome measure for this trial was the live birth/ ongoing pregnancy and clinical pregnancy rates per randomized woman. Live birth was defined as a living fetus born >28 weeks gestation. Ongoing pregnancy was defined as an ongoing pregnancy >14 weeks gestation, and clinical pregnancy was defined by the presence of a positive β-hcg subunit measurement two weeks post-transfer and a clinically viable gestational sac with fetal heart pulsation on ultrasound three weeks later. In addition, the incidences of difficult transfers, blood and/ or mucus on the catheter tip, spontaneous miscarriages and ectopic pregnancies were evaluated. Difficult transfers were defined as difficulties in placing the catheter inside the uterine cavity due to position of the uterus in relation to the cervical canal, cervical stenosis or if embryo transfer took more than 5 minutes. Statistical analysis Statistical analysis was performed according to the intention to treat principle. All analyses of significance were two-sided and tested at the 5% level; values of P < 0.05 were considered to indicate significant differences. Continuous variables were tested if they presented normal distribution using the f-test. The results of the two groups were compared using the t-test or Mann-Whitney U test for parametric and nonparametric data, respectively. Qualitative variables were compared with the use of the chi-squared test with Yates correction or Fisher s 153

164 exact test, when necessary, and the 95% confidence intervals (95% CI) using the Woolf (logit) approximation. Odds ratios (OR) and 95% confidence intervals (95% CI) were calculated to examine the odds of improving clinical outcomes. Clinical and demographic data are also presented as mean (± SD) or as frequency distribution for simplicity. Statistical analysis was performed using the computer statistical package Stats Direct (Stats Direct, Ltd, UK). 154

165 Results The cause of infertility was similar in both groups, mainly being male factor (Table 1). Also, the two groups were similar with regards other demographics, cycle characteristics. There was no significant difference with regards to patient age, period of infertility, and day-3 FSH levels (Table 1). In addition, there were no significant differences in the days of stimulation, and numbers of oocytes retrieved, MII oocytes, injected oocytes, fertilized oocytes, embryos produced, embryo quality, and numbers of embryos transferred (Table 1). Finally, there were no significant differences in the volume of injected culture media in the two groups (data not presented). With regards to the primary outcome measure, live-birth/ ongoing pregnancy rate there was a significantly higher number of live births/ ongoing pregnancies in the US-ET group [68/ 183 (40.98%)] than the CT-ET group [50/ 190 (28.42%)] (P = ; O.R = 1.66, 95% CI = 1.07 to 2.57). In addition, there was a significantly higher number of clinical pregnancies in the US-ET group [75/ 183 (40.98%)] than the CT-ET group [54/ 190 (28.42%)] (P = ; O.R = 1.75, 95% CI = 1.14 to 2.69). With regards to the post-transfer examination of the catheter tips, there were no significant differences in the presence of blood (in and/or on the catheter tip), mucus or blood and mucus on the tip of the embryo transfer catheter (Table 2). Also, there was no significant differences in the incidence of easy transfers in the two groups. Finally, there were also no significant differences in the incidence of spontaneous miscarriages and ectopic pregnancies in the two groups. Finally, a subgroup analysis was performed to determine if the presence of blood and/ or mucus on the catheter tip was a negative predictor of IVF outcome. In both the USG and CT groups, cycles with post-transfer evidence of blood or mucus had a less likely chance of a clinical pregnancy, but this did not reach statistical significance [(33.33% vs %, Chi² = 0.35, P = 0.55) and (7.69% vs %, Chi² = 1.96, P = 0.16), respectively]. 155

166 Discussion The embryo transfer is the final stage of the in vitro fertilization cycle. It is also the instance where clinical manipulations can directly alter the outcomes, and has shown marked variability among different IVF programs and physicians in the same program (9, 10). During embryo transfer, the aim is to manipulate the catheter atraumatically through the cervix into the uterine cavity; without touching the fundus and minimizing trauma to the endometrium (11). Recently, the techniques and variables affecting the success of embryo transfer have attracted more attention, and physician attitudes have been changing accordingly (12, 13). Today, in light of global trends such as single embryo transfer (SET), more stress has been placed on optimizing and standardizing the embryo transfer protocol than ever before. Although most patients who undergo assisted procreation, via IVF or ICSI, will reach the embryo transfer stage with good quality embryos available for replacement, embryo implantation remains the ratelimiting step in the success of this form of therapy. The aim should be to meticulously and accurately place embryos within the uterus; in order to allow for proper implantation and fetal development (14). In order to ascertain the importance of each step involved in the embryo transfer procedure, individual factors must be evaluated independently. This can only be properly performed in prospective, randomized trials in which all possible confounders are virtually the same in both the control and intervention groups. In addition, the two groups of patients should be similar with respect to the known clinical factors that influence pregnancy rate in fresh embryo transfer cycles (e.g. patient demographics, cause of infertility). Moreover, all transfers should be carried out by the same physician using the same transfer catheter, patient position, bladder preparation, and catheter loading techniques. In the current study, these factors were all similar with regards the two randomized groups. Today, more than twenty-years since the first reports of the beneficial effect of ultrasound guidance during the blind embryo transfer procedure were published (15, 16), the routine use of ultrasonography to guide the intrauterine embryo transfer catheter placement are still highly debated. This has been fueled by the conflicting results of published clinical trials, with some concluding that ultrasound guidance 156

167 improves the clinical pregnancy, and implantation rates, while others reporting no such improvement in their results. In the literature there are numerous clinical trials that have tested the theory of improved success with ultrasound-guidance when compared to the standard clinical touch. Even so, the majority of published trials are flawed by study design (e.g. retrospective, quasi-randomization), including both fresh and frozen embryo replacement cycles and donor oocytes, and/or the inclusion of multiple cycles for the same patient in the trial. These issues may be a factor in the obvious heterogeneity in the presented results. Recent systematic reviews have shown that ultrasound-guidance does in factor improve the clinical pregnancy, ongoing pregnancy and live-birth rates in women when compared to the standard clinical touch method (6 8, 17). Even so, the majority of studies did not provide the live-birth or ongoing pregnancy rates for the randomized groups. The exact mechanism whereby ultrasound-guidance improves the outcomes following embryo transfer remains unclear. Several theories have been proposed to identify the mechanisms including confirming the position of the catheter tip within the uterine cavity, properly determining the site of embryo deposition, and decreasing the frequency of difficult embryo transfers (7). Nevertheless, some clinicians argue that the real benefit of ultrasound guidance lies is the ability of increasing the clinical appreciation of the pelvic anatomy, and to notice early signs of ovarian hyperstimulation prior to transfer. Whatever the underlying mechanism, the overall conclusion from this randomized controlled trial is that ultrasound-guided embryo transfer using 2Dtransabdominal ultrasound is significantly more effective than embryo transfer by clinical touch alone. 157

168 References 1. Edwards RG. Clinical approaches to increasing uterine receptivity during human implantation. Hum Reprod 1995; 10(Suppl 2): International Committee for Monitoring Assisted Reproductive Technology; Adamson GD, de Mouzon J, Lancaster P, Nygren KG, Sullivan E, Zegers- Hochschild F. World collaborative report on in vitro fertilization, Fertil Steril 2006; 85: Mansour RT, Aboulghar MA. Optimizing the embryo transfer technique. Hum Reprod 2002; 17: Steptoe PC, Edwards RG. Birth after the reimplantation of a human embryo. Lancet 1978; 12: Abou-Setta AM, Al-Inany HG, Mansour RT, Serour GI, Aboulghar MA. Soft versus firm embryo transfer catheters for assisted reproduction: a systematic review and meta-analysis. Hum Reprod 2005; 20: Buckett WM. A meta-analysis of ultrasound-guided versus clinical touch embryo transfer. Fertil Steril 2003; 80: Sallam HN and Sadek SS. Ultrasound-guided embryo transfer: a meta-analysis of randomized controlled trials. Fertil Steril 2003; 80: Abou-Setta AM, Mansour RT, Al-Inany HG et al Among women undergoing embryo transfer, is the probability of pregnancy and live birth improved with ultrasound-guidance than with clinical touch alone? A systemic review and metaanalysis of prospective randomized trials. Fertil Steril (in press). 9. Karande VC, Morris R, Chapman C, Rinehart J, Gleicher N. Impact of the "physician factor" on pregnancy rates in a large assisted reproductive technology program: do too many cooks spoil the broth? Fertil Steril 1999; 71: Hearns-Stokes RM, Miller BT, Scott L, Creuss D, Chakraborty PK, Segars JH. Pregnancy rates after embryo transfer depend on the provider at embryo transfer. Fertil Steril 2000; 74: Sallam HN. Embryo transfer: factors involved in optimizing the success. Curr Opin Obstet Gynecol 2005; 17: Kovacs GT. What factors are important for successful embryo transfer after invitro fertilization? Hum Reprod. 1999; 14: Salha OH, Lamb VK, Balen AH. A postal survey of embryo transfer practice in the UK. Hum Reprod 2001; 16: Coroleu B, Barri PN, Carreras O, Martinez F, Parriego M, Hereter L, Parera N, Veiga A, Balasch J. The influence of the depth of embryo replacement into the uterine cavity on implantation rates after IVF: a controlled, ultrasound-guided study. Hum Reprod 2002; 17: Leong M, Leung C, Tucker M, Wong C, Chan H. Ultrasound-assisted embryo transfer. J In Vitro Fert Embryo Transf 1986; 3: Strickler RC, Christianson C, Crane JP, Curato A, Knight AB, Yang V. Ultrasound guidance for human embryo transfer. Fertil Steril 1985; 43:

169 17. Brown JA, Buckingham K, Abou-Setta A, Buckett W. Ultrasound versus 'clinical touch' for catheter guidance during embryo transfer in women. Cochrane Database Syst Rev 2007; 24:CD

170 Table 1: Review table of demographic and cycle outcomes: US-ET Mean (SD) CT-ET Significance Mean (SD) Age (5.47) (5.46) NS FSH 5.19 (2.09) 5.17 (2.32) NS Days of stimulation 9.81 (1.87) 9.78 (1.98) NS No. of Oocytes Retrieved (7.37) (7.90) NS No. of MII 8.79 (6.12) 9.07 (6.41) NS Injected 8.78 (6.10) 9.05 (6.38) NS Fertilized 6.16 (4.28) 6.48 (4.66) NS Good Embryo G (3.10) 2.98 (2.94) NS Fair Embryo G (2.01) 2.16 (2.01) NS Poor Embryo G3, G (1.63) 1.36 (1.90) NS No. of ET 3.36 (1.07) 3.32 (1.04) NS Clinical Pregnancy Rate (%) 75/ 183 (40.98) 54/ 190 (28.42) P = ; O.R = 1.75, 95% CI = 1.14 to 2.69 Multiple Pregnancy Rate 3 2 NS Ectopic Pregnancy 1 0 NS 160

171 US-ET Mean (SD) CT-ET Significance Mean (SD) Miscarriages 5 4 NS Live-birth/ Ongoing Pregnancy Rate (%) 68/ 183 (37.16) 50/ 190 (26.32) P = ; O.R = 1.66, 95% CI = 1.07 to 2.57 Table 2: ET Catheter tip: US-ET Frequency (%) CT-ET Significance Frequency (%) No blood or mucus 159 (86.89) 177 (93.16) NS Blood 5 (2.73) 3 (1.58) NS Mucus 15 (8.20) 7 (3.68) NS Blood + Mucus 4 (2.19) 3 (1.58) NS Retained embryos 3 (1.64) 5 (2.63) NS 161

172 162

173 Chapter 11 Among women undergoing embryo transfer, is the probability of pregnancy and live birth improved with ultrasound-guidance than with clinical touch alone? a systemic review and meta-analysis of prospective randomized trials. Ahmed M. Abou-Setta, Ragaa T. Mansour, Hesham G. Al-Inany, Mona M. Aboulghar, Gamal I. Serour, Mohamed M. Aboulghar. Fertiltiy & Sterility. 2007;88:

174 Abstract Objective: To investigate the theory that ultrasound guidance during ET improves clinical outcomes. Design: Systematic review of prospective, randomized, controlled trials comparing ultrasound with clinical touch methods of embryo catheter guidance. Setting: Infertility centers. Patient(s): 5,968 ET cycles in women. Intervention(s): Embryo transfer with or without ultrasound guidance. Main Outcome Measure(s): Meticulous electronic (e.g., PubMed, EMBASE, CENTRAL) and hand searches were performed to locate trials. Primary outcome measures were the live-birth, ongoing pregnancy, and clinical pregnancy rates. Secondary outcome measures were the implantation, multiple pregnancies, and miscarriage rates. In addition, the incidences of ectopic pregnancies and difficult transfers were evaluated. Result(s): Twenty-five studies were retrieved, of which five were excluded. Meta-analysis of the remaining studies (5,968 ET cycles in women) was conducted by using the Mantel-Haenszel method (fixedeffect model). There was a significantly increased chance of a live birth (odds ratio [OR] = 1.78, 95% confidence interval [CI] = 1.19 to 2.67), ongoing pregnancy (OR = 1.51, 95% CI = 1.31 to 1.74), clinical pregnancy (OR = 1.50, 95% CI = 1.34 to 1.67), embryo implantation (OR = 1.35, 95% CI = 1.22 to 1.50), and easy transfer rates after ultrasound guidance (OR = 0.68, 95% CI = 0.58 to 0.81). There was no difference in multiple pregnancy, ectopic pregnancy, or miscarriage rates. Conclusion(s): Ultrasound-guided ET significantly increases the chance of live birth and ongoing and clinical pregnancy rates compared with the clinical touch method. Key words: Ultrasound-guided ET, clinical touch ET, meta-analysis, in vitro fertilization 164

175 Background Despite the major advancements made in ovarian stimulation protocols and in vitro embryo development, the pregnancy and embryo implantation rates after ET have remained relatively low. The majority of patients undergoing IVF will reach the transfer stage, with goodquality embryos available for transfer, but only a small proportion of them will ever achieve a clinical pregnancy, an ongoing pregnancy, or a live birth. It is estimated that up to 85% of the embryos replaced into the uterine cavity will fail to implant (1). This makes the ET procedure a highly vital and, at the same time, highly inefficient step in IVF. Consequently, any modification to the standard protocol that will improve the outcomes is of great value. The pregnancy rate after ET is dependent on multiple factors including embryo quality, endometrial receptivity, and the technique of the ET itself (2). In recent years, more stress has been placed on optimizing and standardizing the ET protocol. Factors such as ease of the procedure (3), catheter choice 4, 5, and dummy ET (6) have proved to improve the clinical outcomes. Today, more than 25 years after the first reports of the beneficial effect of ultrasound guidance during the blind ET procedure were published 7, 8, the routine use of ultrasonography to guide the intrauterine ET catheter placement still is debated highly. This has been fueled by the conflicting results of published clinical trials, with some concluding that ultrasound guidance improves the clinical pregnancy and implantation rates and others reporting no such improvement in their results. In light of this controversy, and the need to identify clearly the relative efficacy of ultrasound guidance during ET, we decided to locate, analyze, and review systematically the best available evidence today for the use of ultrasonography during ET. 165

176 Materials and methods Criteria for Considering Studies for This Review All published, unpublished, and ongoing randomized trials reporting data that compare outcomes for women undergoing ET through the cervical route after IVF, or ICSI, using ultrasound-guided versus the clinical touch method were sought in all languages. Types of Outcome Measures The primary outcome measures were the live-birth, ongoing pregnancy, and clinical pregnancy rates. The secondary outcomes were the implantation, multiple pregnancies, and miscarriage rates. In addition, the incidences of ectopic pregnancies, difficult or failed transfers, and need for instrumental assistance during the transfer (e.g., stylette, tenaculum, dilatation, sounding), were evaluated. Last, the tips of the posttransfer catheters were evaluated for signs of cervical or endometrial trauma (e.g., presence of blood, mucus, or both), in addition to retained embryos. Search Strategy for Identification of Studies Meticulous computerized searches (last performed July 2006) were conducted by using MEDLINE (1966 to present); EMBASE (1980 to present); the Cochrane Central Register of Controlled Trials (CENTRAL) on the Cochrane Library issue 3, 2006; the National Research Register; the trial register of controlled trials ( the Latin American and Caribbean Health Sciences Literature database (LILACS); and details on reviews in progress collected by the National Health Service Centre for Reviews and Dissemination. Furthermore, the reference lists of all known primary studies, review articles, citation lists of relevant publications, abstracts of major scientific meetings (e.g., European Society of Human Reproduction and Embryology and American Society for Reproductive Medicine), and included studies were examined to identify additional relevant citations. Finally, ongoing and unpublished trials were sought by contacting experts in the field and commercial entities. 166

177 Methods of the Review A standardized data extraction form was developed and pilot tested for consistency and completeness. Trials were considered for inclusion, and trial data extracted. Data management and statistical analyses were conducted by using the Review Manager 4.2, Meta-analysis With Interactive Explanations 1.2, and Power and Sample Size Calculations statistical software packages. Individual outcome data were included in the analysis if they met the prestated criteria. Where possible, data were extracted to allow for an intention-to-treat analysis, defined as including in the denominator all randomized cycles. If data from the trial reports were insufficient or missing, the investigators of individual trials were contacted via for additional information, to perform analyses on an intention-to-treat basis. A response was received from the correspondence authors of nine trials (9-17), with two providing the unpublished live-birth rates (11, 14). For the meta-analysis, the number of participants experiencing the event in each group of the trial was recorded. Statistical heterogeneity of the included studies was determined by visual inspection of the outcome tables and more formally by using the χ 2 and I 2 tests for heterogeneity. The I 2 test is a statistical measure used to identify and quantify heterogeneity. It describes the percentage of the variability in effect estimates that is due to heterogeneity rather than sampling error (chance) (18). An I 2 value greater than 50% may be considered to represent substantial heterogeneity. Meta-analysis of binary data was performed with the Mantel-Haenszel method and a fixed-effect model, and the odds ratio (OR) and 95% confidence intervals (95% CI) evaluated. Subsequently a sensitivity analysis was conducted by using the random-effects model. Furthermore, where appropriate, subgroup analyses were performed. Search Results A total of 25 prospective randomized controlled trials were identified (13 full-text papers, 11 conference abstracts, and one unpublished study) (Fig. 1). Of these, three studies (19-21) were excluded initially for duplicate publication (i.e., published as abstract in a conference proceeding and as full text in a peer-reviewed journal). In addition, two 167

178 trials (10, 12) were excluded because no data were available for review. Finally, the methodologic quality of the remaining 20 trials was assessed and data extracted to allow for an intention-to-treat analysis (Table 1). 168

179 Results Primary Outcome Measures With regard to the primary outcome measures, there was a significantly increased chance of a live birth (ultrasonography: 76/305 vs. clinical touch: 49/310, OR = 1.78, 95% CI = 1.19 to 2.67), an ongoing pregnancy (ultrasonography: 656/1,902 vs. clinical touch: 557/2,065, OR = 1.51, 95% CI = 1.31 to 1.74), and clinical pregnancy (ultrasonography: 1086/2,941 vs. clinical touch: 884/3,027, OR = 1.50, 95% CI = 1.34 to 1.68) with ultrasound guidance than with the standard clinical touch method (Figure 2, Figure 3, Figure 4). Moreover, subgroup analyses of only the truly randomized trials, fresh nondonor cycles, and the frozen embryo replacement cycles revealed similar results (Table 2). Even so, patients undergoing oocyte donor cycles did not demonstrate any significant findings between the ultrasound-guided and the clinical touch groups. Nevertheless, it is important to note that only one trial used oocyte donor cycles. Secondary Outcome Measures As for the secondary outcome measures, there was a significant increased chance of embryo implantation after ultrasound guidance compared with the standard clinical touch method (ultrasonography: 974/5,662 vs. clinical touch: 810/5,841, OR = 1.35, 95% CI = 1.22 to 1.50). Even so, there were similar rates of multiple pregnancies, ectopic pregnancies, and spontaneous miscarriages in the two groups (P>.05). As for the ease of transfer and the need for instrumental assistance, there was a statistically significant increased chance of having a difficult transfer, need for instrumental assistance, and failure to transfer with the assigned catheter during clinical touch alone compared with ultrasound guidance (Table 2). Even so, there was significant heterogeneity between the study trials: ease of transfer (χ 2 = 53.06, I 2 = 83.0%, P<.00001); need for instrumental assistance (χ 2 = 57.83, I 2 = 89.6%, P<.00001); failure to transfer with the assigned catheter (χ 2 = 2.92, I 2 = 65.8%, P<.09). Therefore, a series of sensitivity analyses were undertaken. The results of these analyses could only confirm the increased usage of instrumental assistance with the clinical touch group compared with the ultrasound-guided group (P=.04). The remaining comparisons could not be verified. 169

180 Finally, the catheter tips were evaluated for the presence of blood, mucus, and/or retained embryos. There were no significant differences for these rates between the two groups with the exception of increased incidence of finding blood on the catheter tips after clinical touch ET. 170

181 Discussion Although most patients who undergo assisted procreation, via IVF or ICSI, will reach the ET stage with good-quality embryos available for replacement, embryo implantation remains the rate-limiting step in the success of this form of therapy. The aim of the ET procedure is to place embryos atraumatically and accurately within the uterus to allow for proper implantation and fetal development. Studies have shown that different factors may be involved in a successful transfer. These include ET catheter choice (4, 5), the use of ultrasound guidance (35, 36), the experience of the physician (37), the ease of the procedure (3), the presence or absence of blood on the catheter (38), the use of cervical introducers or obturators (39), the value of resting after transfer (40), the position of embryo insertion in the uterus (41, 42), and retention of embryos in the catheter (43). To ascertain the importance of each step involved in the ET procedure, individual factors must be evaluated independently. Therefore, the ET procedure may be divided arbitrarily into four distinct sections: [1] preparation before ET (e.g., patient position, cervical preparation, uterine position, and the dummy ET); [2] technical aspects related to the ET catheter (e.g., catheter type and catheter loading); [3] the ET procedure (e.g., the site of embryo deposition within the uterus and techniques to assist with the accurate placement of the embryo within the uterus; and [4] posttransfer aspects (e.g., expulsion of fluid/embryos from the cervix after ET and bed rest after ET) (44). Because it would be difficult to compare several factors accurately at the same time, we decided to concentrate on one factor, the beneficial value of ultrasound guidance to guide catheter placement during the ET procedure. All the studies retrieved from our search have examined the role of twodimensional (2-D) transabdominal ultrasound guidance during ET. Even so, vaginal ultrasound-guided ET (40, 45), as well as the use of threedimensional (46) and fourth-dimensional ultrasound-guided ET have been reported. Nonetheless, it is important to mention that to the best of our knowledge there are no randomized controlled trials comparing any of these new modalities either with the standard clinical touch method or with the 2-D ultrasound guided method. Systematic reviews and meta-analysis of randomized controlled trials have proved to be the highest level of evidence in the hierarchy of 171

182 medical knowledge. Even so, publication and search biases may confound the results of any systemic review, because studies showing positive results are more likely to be published (47, 48). Therefore, every effort has been made to avoid bias by searching a wide variety of online databases (with no language barriers), hand searching of the gray literature, and contacting commercial entities in search of unpublished or ongoing trials. The objective was to minimize the chance of publication or selection bias to strengthen the validity of the results of the systematic review. Failure to identify trials reported in conference proceedings might affect the results or threaten the validity of a systematic review (49). In addition, multiple methods were used to detect publication, location, or selection biases including funnel plot, trim and fill funnel plot, and the Egger regression analyses (50). The funnel plot is a graphic display of study precision plotted against the effect size. It is used to investigate whether there is a link between study size and treatment effect. The funnel plot of the included studies shows asymmetry that may be due to publication bias. Even so, it is important to note that there may be other etiologies for the asymmetry found in funnel plot, including selection biases (e.g., location biases, language bias, citation bias, multiple publication bias), poor methodologic quality of smaller studies (e.g., poor methodologic design, inadequate analysis, fraud), true heterogeneity (i.e., size of effect differs according to study size), artifactual, and just by chance (51). Duval and Tweedie (52) have proposed the trim and fill method for defining the extent of publication bias. It is based on adding studies to a funnel plot so that it becomes symmetrical. Then the smaller studies are omitted until the funnel plot is symmetrical (trimming). The trimmed funnel plot is used to estimate the true center of the funnel, and then the omitted studies and their missing counterparts around the center are replaced (filling). This provides an estimate of the number of missing studies and an adjusted treatment effect, including the filled studies. Funnel plots, plots of the trials' effect estimates against sample size, are skewed and asymmetric in the presence of publication bias and other biases. Funnel plot asymmetry, measured by regression analysis, predicts discordance of results when meta-analyses are compared with single large trials. One of the commonly used regression methods is the 172

183 Egger method. This method uses a linear regression approach to measure funnel plot asymmetry on the natural logarithm scale of the OR (50). In this systematic review, we formally investigated the presence of publication bias using a series of publication bias regression and sensitivity plots (e.g., the Egger regression method and the trim and fill method). For the ongoing pregnancy rate, publication bias was found to be unlikely (Egger score = 0.24; 95% CI = , P=.83). Even so, using the trim and fill method to determine bias sensitivity revealed that one study (empty circle in Fig. 1B) may be imputed to bring symmetry to the funnel plot. The resulting meta-analysis outcome would still be significant: OR = 1.50 (95% CI = ). For the clinical pregnancy rate, publication bias was found to be unlikely (Egger score = 0.21; 95% CI = , P=.74). In addition, using the trim and fill method to determine bias sensitivity revealed that no studies need to be imputed to bring symmetry to the funnel plot. Therefore the resulting meta-analysis remains significant: OR = 1.50 (95% CI = ). The possibility of publication bias opens up the door for questioning the results of any meta-analysis; therefore, further inspection and interpretation are needed to formulate definite conclusions. With regard to the 20 included trials, there was a general trend toward better results with ultrasound guidance, with only a limited number of studies demonstrating equivocal or better results with the clinical touch method. In addition, there was no statistical heterogeneity shown by either the scatter plot or the χ 2 / I 2 tests, with regard to the primary outcomes. It may be assumed that studies with equivocal results demonstrate that some clinicians have an excellent clinical sense, therefore nullifying any extra advantage provided by sonographic visualization. Another important issue is the sample size of clinical trials and systematic reviews. It is theorized that some of the smaller trials did not have sufficient sample sizes to detect minor differences between the study groups. The largest included study (N = 800) had enough power to detect an 8% difference in clinical pregnancy rates (16). Our current meta-analysis included 5,968 ET cycles and could detect an absolute difference of 3.3% with 80% power in a two-tailed analysis (assuming a clinical pregnancy rate of 30% with clinical touch alone and a significance level of.05). The absolute difference between ultrasound- 173

184 guided and clinical touch ET was 7.7%, therefore validating our results. The number needed to treat to obtain one additional clinical pregnancy with use of ultrasound guidance is 13 (95% CI = 10 19). The exact mechanism whereby ultrasound-guided ET improves pregnancy rates and embryo implantation remains unclear. Several theories have been proposed to identify the mechanisms whereby the transfer technique is optimized. These include confirming the position of the tip of the ET catheter within the uterine cavity, the site of embryo deposition, increasing the frequency of easy ETs, and avoiding endometrial indentation (36). Nevertheless, some clinicians argue that the real benefit of ultrasound guidance lies in the ability to increase the clinical appreciation of the pelvic anatomy during transfer. They infer that ultrasound guidance, compared with the standard clinical touch alone, will not significantly increase the pregnancy rates when ET is performed by experienced professionals. Even so, this simple modification will allow for standardization of the transfer technique and therefore decrease any unexpected variation in pregnancy rates among different clinicians in the same center. Whatever the underlying mechanism, the overall conclusion from this meticulous systematic review is that ultrasound-guided ET with use of 2- D transabdominal ultrasound is significantly more effective than ET by clinical touch alone. It is believed that our overall analysis, and subgroup analyses, represents both a robust evaluation of the available evidence in the medical literature and also the best available evidencebased conclusions to support ultrasound guidance during ET. The results of this systematic review demonstrate that ultrasound guidance during ET placement is a beneficial tool in optimizing the outcome of the ET procedure. This is in line with results of previously published clinical trials, systematic reviews (35, 36), and published guidelines (53). In addition, we have demonstrated that the beneficial effect is not limited only to patients with fresh but also those with frozen embryo replacement. It is also hoped that this evidence will be easily and quickly translated from the medical literature to everyday clinical practice. Last, it is also important to note that until today, no study has reported any direct adverse effects of ultrasound-guided ET. Even so, one may argue that the main disadvantages of ultrasound-guided ETs are the 174

185 need for more time, space, equipment, and trained ultrasound personnel. In addition, possible psychological stress may be caused in some patients who are forced to micturate after the transfer procedure, therefore disrupting the period of absolute bed rest. This issue can be managed easily through proper counseling before the procedure on the amount of time needed for bed rest after ET, which is currently recommended as 20 minutes (53). Also an additional cost to the transfer cycle may be associated with the preferred usage of echogenic catheters by some clinicians. The manufacturers of ET catheters have began to market new echogenic versions of their standard catheters, which are sold for a 25% to 40% higher price tag than the original product, adding additional costs to the treatment cycle. Even so, it is of the utmost importance to note that even though echogenic catheters promise to be cost-effective by providing clinicians with more ultrasonographic visibility, easier transfers, and higher pregnancy rates than their competitor nonechogenic catheters, these points have not been proved in the literature, apart from more visibility. The pregnancy rates have been equivocal with the echogenic catheters. Therefore echogenicity currently should be left up to individual clinical preference and not standardized. Finally, it would be helpful for clinicians to perform in-house costeffective analyses to determine the best combination of needed equipment, personnel, and catheter choice to provide the most efficient and cost-effective treatment options for patients. In conclusion, according to the best available evidence in the medical literature today, the use of transabdominal ultrasound guidance to guide catheter placement during ET both is beneficial and should become the standard of care for all patients. 175

186 Acknowledgments The authors thank all the corresponding authors who were contacted for additional information. In addition, we would like especially to thank Jose Franco, Jr., M.D., Juan Garcia-Velasco, M.D., Roberto Matorras, M.D., Yannis Prapas, M.D., Oi Shan Tang, M.D., and Ariel Weissman, M.D., for providing us with missing data. 176

187 References 1. Edwards RG. Clinical approaches to increasing uterine receptivity during human implantation. Hum Reprod. 1995;10(Suppl 2): Mansour RT, Aboulghar MA. Optimizing the embryo transfer technique. Hum Reprod. 2002;17: Lesny P, Killick SR, Tetlow RL, Robinson J, Maguiness SD. Embryo transfer can we learn anything new from the observation of junctional zone contractions?. Hum Reprod. 1998;13: Abou-Setta AM, Al-Inany HG, Mansour RT, Serour GI, Aboulghar MA. Soft versus firm embryo transfer catheters for assisted reproduction: a systematic review and meta-analysis. Hum Reprod. 2005;20: Epub 2005 Jul Abou-Setta AM. Firm embryo transfer catheters for assisted reproduction: a systematic review and meta-analysis using direct and adjusted indirect comparisons. Reprod Biomed Online. 2006;12: Mansour R, Aboulghar M, Serour G. Dummy embryo transfer: a technique that minimizes the problems of embryo transfer and improves the pregnancy rate in human in vitro fertilization. Fertil Steril. 1990;54: Strickler RC, Christianson C, Crane JP, Curato A, Knight AB, Yang V. Ultrasound guidance for human embryo transfer. Fertil Steril. 1985;43: Leong M, Leung C, Tucker M, Wong C, Chan H. Ultrasound-assisted embryo transfer. J In Vitro Fert Embryo Transf. 1986;3: al-shawaf T, Dave R, Harper J, Linehan D, Riley P, Craft I. Transfer of embryos into the uterus: how much do technical factors affect pregnancy rates?. J Assist Reprod Genet. 1993;10: Blake K. Prospective randomised trial to evaluate the role of ultrasound during transcervical embryo transfer. (Unpublished.) National Research Register. Available from: URL: de Camargo Martins AM, Baruffi RL, Mauri AL, Petersen C, Oliveira JB, Contart P, et al. Ultrasound guidance is not necessary during easy embryo transfers. J Assist Reprod Genet. 2004;21: Drakeley AJ, Lunt R, Aust T, Williamson P, Gazvani R, Sklavounos J, et al. A randomised trial of 2250 women having ultrasound guided embryo transfer. Hum Reprod. 2006;21(Suppl 1):i Garcia-Velasco JA, Isaza V, Martinez-Salazar J, Landazabal A, Requena A, Remohi J, et al. Transabdominal ultrasound-guided embryo transfer does not increase pregnancy rates in oocyte recipients. Fertil Steril. 2002;78: Matorras R, Urquijo E, Mendoza R, Corcostegui B, Exposito A, Rodriguez- Escudero FJ. Ultrasound-guided embryo transfer improves pregnancy rates and increases the frequency of easy transfers. Hum Reprod. 2002;17: Prapas Y, Prapas N, Hatziparasidou A, Vanderzwalmen P, Nijs M, Prapa S, et al. Ultrasound-guided embryo transfer maximizes the IVF results on day 3 and day 4 embryo transfer but has no impact on day 5. Hum Reprod. 2001;16:

188 16. Tang OS, Ng EH, So WW, Ho PC. Ultrasound-guided embryo transfer: a prospective randomized controlled trial. Hum Reprod. 2001;16: Weissman A, Farhi J, Steinfeld Z, Mutsafi R, Nahum H, Levran D. A prospective, randomized study of ultrasound-guided embryo transfer. Fertil Steril. 2003;80(Suppl 3):S Higgins JPT, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analysis. BMJ. 2003;327: García-Velasco JA, Martinez-Salazar J, Isaza V, Landazabal A, Requena A, Simón C. Does ultrasound guidance at embryo transfers improve pregnancy rates in oocyte recipients?. Fertil Steril. 2001;76(Suppl 1):S75 S Li R, Zhuang GL, Cai ZM, Wang H, Zhong K, Zhou WY. [Clinical analysis of ultrasound-guided embryo transfer after in-vitro fertilization]. Zhonghua Fu Chan Ke Za Zhi. 2004;39: Martins AMVC, Baruffi RLR, Mauri AL, Petersen C, Oliveira JBA, Contart P, et al. Ultrasound guidance for embryo transfer in patients previously identified by mock transfer as likely to have an easy transfer: a controlled trial. The 20th Annual Meeting of the European Society of Human Reproduction and Embryology. Hum Reprod. 2004;19(Suppl 1):i Abdelmassih VG, Abdelmassih ST, Nagy ZP, Abdelmassih R, Balmaceda J. The effect of ultrasound (US) guided embryo transfer (ET) and the choice of catheter on the outcome of IVF. Fertil Steril. 2001;76(Suppl 1):S88 S Bar Hava I, Meltzer S, Rabinson J, Ayash I, Sega S, Tur Kaspa I. Ultrasound guided versus blind tactile embryo transfer: a prospective randomized study. Hum Reprod. 2003;18(Suppl 1):xviii Coroleu B, Barri PN, Carreras O, Martinez F, Veiga A, Balasch J. The usefulness of ultrasound guidance in frozen-thawed embryo transfer: a prospective randomized clinical trial. Hum Reprod. 2002;17: Coroleu B, Carreras O, Veiga A, Martell A, Martinez F, Belil I, et al. Embryo transfer under ultrasound guidance improves pregnancy rates after in-vitro fertilization. Hum Reprod. 2000;15: Kan AK, Abdalla HI, Gafar AH, Nappi L, Ogunyemi BO, Thomas A, et al. Embryo transfer: ultrasound-guided versus clinical touch. Hum Reprod. 1999;14: Kosmas IP, Janssens R, De Munck L, Al Turki HF, Tournaye H, Van Steirteghem AC, et al. Ultrasound guidance during embryo transfer does not offer any benefit in clinical outcome: a randomized controlled trial. Hum Reprod. 2006;21(Suppl 1):i Li R, Lu L, Hao G, Zhong K, Cai Z, Wang W. Abdominal ultrasound-guided embryo transfer improves clinical pregnancy rates after in vitro fertilization: experiences from 330 clinical investigations. J Assist Reprod Genet. 2005;22: Maldonado LG, Ajzen SA, Busato WC, Iaconelli A, Bibancos M, Borges E. Impact of previous hysterossonometry on embryo transfer. Fertil Steril. 2005;84(Suppl 1):S

189 30. Marconi G, Young E, Vilela M, Bello A, Young E, Sueldo C. Prospective randomized comparison of an ultrasound-guided embryo transfer versus a blind catheter placement. Fertil Steril. 2003;80(Suppl 3):S Moraga-Sanchez MR, Saucedo-de la Llata E, Batiza-Resendiz V, Santos-Haliscak R, Galache-Vega P, Hernández-Ayup S, et al. Abstracts of the 20th Annual Meeting of the European Society of Human Reproduction and Embryology, Berlin, Germany. Hum Reprod. 2004;19(Suppl 1):i Prapas Y, Prapas N, Hatziparasidou A, Prapa S, Nijs M, Vanderzwalmen P, et al. The echoguide embryo transfer maximizes the IVF results. Acta Eur Fertil. 1995;26: Sallam HN, Agameya AF, Rahman AF, Ezzeldin F, Sallam AN. Ultrasound measurement of the uterocervical angle before embryo transfer: a prospective controlled study. Hum Reprod. 2002;17: Wisanto A, Janssens R, Deschacht J, Camus M, Devroey P, Van Steirteghem AC. Performance of different embryo transfer catheters in a human in vitro fertilization program. Fertil Steril. 1989;52: Buckett WM. A meta-analysis of ultrasound-guided versus clinical touch embryo transfer. Fertil Steril. 2003;80: Sallam HN, Sadek SS. Ultrasound-guided embryo transfer: a meta-analysis of randomized controlled trials. Fertil Steril. 2003;80: Lu MC. Impact of non-physician factors on the physician factor of in vitro fertilization success: is it the broth, the cooks, or the statistics?. Fertil Steril. 1999;71: Goudas VT, Hammitt DG, Damario MA. Blood on the embryo transfer catheter is associated with decreased rates of embryo implantation and clinical pregnancy with the use of in vitro fertilization embryo transfer. Fertil Steril. 1998;70: Ghazzawi IM, Al-Hasani S, Karaki R, Souso S. Transfer technique and catheter choice influence the incidence of transcervical embryo expulsion and the outcome of IVF. Hum Reprod. 1999;14: Woolcott R, Stanger J. Ultrasound tracking of the movement of embryoassociated air bubbles on standing after transfer. Hum Reprod. 1998;13: Waterstone J, Curson R, Parsons J. Embryo transfer to low uterine cavity. Lancet. 1991;337: Yovich JL, Turner SR, Murphy AJ. Embryo transfer technique as a cause of ectopic pregnancies in in vitro fertilization. Fertil Steril. 1985;44: Friedler S, Lewin A, Schenker JG. Methodology of human embryo transfer following assisted reproduction. J Assist Reprod Genet. 1993;10: Al-Inany HG, Abou-Setta AM, Garzo G. ET catheters for assisted reproduction. Cochrane Database Syst Rev. 2006;1:CD DOI: / CD

190 45. Kojima K, Nomiyama M, Kumamoto T, Matsumoto Y, Iwasaka T. Transvaginal ultrasound-guided embryo transfer improves pregnancy and implantation rates after IVF. Hum Reprod. 2001;16: Baba K, Ishihara O, Hayashi N, Saitoh M, Taya J, Kinoshita K. Three-dimensional ultrasound in embryo transfer. Ultrasound Obstet Gynecol. 2000;16: Dickerson K, Min YL, Meinert CL. Factors influencing publication of research results. JAMA. 1992;267: Easterbrook PJ, Berlin JA, Gopalan R, Matthews DR. Publication bias in clinical research. Lancet. 1991;337: Higgins JPT, Green S, editors. Cochrane handbook for systematic reviews of interventions [updated May 2005]. In: The Cochrane Library, Issue 3, Chichester, United Kingdom: John Wiley & Sons. 50. Egger M, Davy-Smith G, Sneider M, Minder CE. Bias in meta-analysis detected by a simple, graphical test. BMJ. 1997;315: Deeks JJ, Higgins, JPT, Altman DG, editors. Analysing and presenting results. In: Higgins JPT, Green S, editors. Cochrane handbook for systematic reviews of interventions [updated May 2005]; Section 8. Available from: URL: Accessed May 25, Duval S, Tweedie R. Trim and fill: a simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis. Biometrics. 2000;56: National Institute for Clinical Excellence. Fertility assessment and treatment for people with fertility problems. London: RCOG Press; 2004;

191 Table 1: Review table of the included studies, comparing ultrasound-guided ET with clinical touch ET. Included Studies ET cyc -les Follow -up Abdelmassih et al. (22) NA 39 Not performed Unclear Unclear ITT CPR Present - multiple catheters Al-Shawaf et al. (9) NA 241 Not performed Availability of Ultrasonog rapher Inadequate ITT CPR Multiple studies Bar Hava et al. (23) NA 131 Not performed Unclear Unclear ITT CPR Not evident Coroleu et al. (24) Not performed Patients A-priori Sample Size Calculation Method of Randomization Method of Randomization Concealm ent Intentionto-treat Confounders Computergenerated randomizat ion table Unclear ITT OPR Not evident 181

192 Included Studies Coroleu et al. (25) de Camargo Martins et al. (11) Garcia- Velasco et al. (13) Kan et al. (26) ET cyc -les Patients A-priori Sample Size Calculation Method of Randomization Performed Not performed Computergenerated randomizat ion table Computergenerated randomizat ion table Random Performed permuted blocks Not performed Availability of Ultrasonog rapher Follow -up Method of Randomization Concealm ent Intentionto-treat Confounders Unclear ITT OPR Not evident Unclear ITT LBR Not evident Adequate Multiple ITT OPR physicians Inadequate ITT CPR Multiple physicians 182

193 Included Studies ET cyc -les Patients A-priori Sample Size Calculation Kosmas et al. (27) Not performed Li et al. (28) Not performed Maldonado et al. (29) Not performed Marconi et al. (30) Not performed Matorras et al. (14) Not performed Morago- Sanchez et al. (31) Not performed Follow -up Present - Unclear Not ITT OPR multiple catheters Unclear Unclear ITT CPR Not evident Unclear Unclear ITT CPR Not evident Unclear Unclear ITT OPR Multiple catheters Method of Randomization Method of Randomization Concealm ent Intentionto-treat Confounders Computergenerated randomizat ion table Computergenerated Unclear randomizat ITT LBR Not evident ion table Unclear Unclear ITT CPR Multiple studies 183

194 Included Studies Prapas et al. (32) Prapas et al. (15) Sallam et al. (33) Tang et al.(16) Weissman et al. (33) ET cyc -les Patients A-priori Sample Size Calculation NA 132 Not performed Not performed Not performed Performed Not performed Follow -up Alternate Randomiza Inadequate ITT CPR Not evident tion Alternate randomizat Inadequate ITT CPR Not evident ion Alternate randomizat Inadequate ITT CPR Not evident ion Method of Randomization Method of Randomization Concealm ent Intentionto-treat Confounders Computergenerated randomizat ion table Adequate Multiple ITT OPR physicians Unclear Unclear ITT CPR Not evident 184

195 Included Studies ET cyc -les Follow -up Wisanto et al. (34) Not performed Patients A-priori Sample Size Calculation Method of Randomization Method of Randomization Concealm ent Intentionto-treat Confounders Computergenerated randomizat ion table Unclear not ITT CPR Multiple studies, physicians ITT = Intention to treat analysis performed; CPR = Clinical pregnancy rate; OPR = Ongoing pregnancy rate; LBR = Live-birth rate 185

196 Table 2: Review table of the primary outcome measures of the included studies, comparing ultrasound-guided ET with clinical touch ET. Outcome Comparison No of No of Effect Measure Studies Cycles size* Live- birth Rate All trials [1.19 to 2.67] Properly randomized trials only Fresh, non-donor cycles only Oocyte donation cycles only Frozen embryo replacement cycles only [1.19 to 2.67] [1.19 to 2.67] 0 0 Not estimable 0 0 Not estimable Ongoing Pregnancy Rate All trials [1.31 to 1.74] Properly randomized trials only [1.25 to 1.78] Fresh, non-donor cycles only Oocyte donation cycles only Frozen embryo replacement cycles only [1.33 to 1.82] [0.84 to 1.90] [1.04 to 1.93] Clinical Pregnancy Rate All trials [1.34 to 1.67] Properly randomized trials only [1.23 to 1.72] Fresh, non-donor cycles only [1.36 to 1.74] 186

197 Outcome Comparison No of No of Effect Measure Studies Cycles size* Oocyte donation cycles only Frozen embryo replacement cycles only [0.81 to 1.84] [0.97 to 1.62] Implantation Rate All trials [1.22 to 1.50] Secondary Outcomes Properly randomized trials only Fresh, non-donor cycles only Oocyte donation cycles only Frozen embryo replacement cycles only Multiple Pregnancy Rate Ectopic Pregnancy Rate [1.25 to 1.72] [1.21 to 1.51] 0 0 Not estimable [1.11 to 1.81] [0.72 to 1.27] [0.30 to 1.45] Miscarriage Rate [0.61 to 1.26] Ease of Transfer Difficult embryo transfer [0.58 to 0.81] Failure using the assigned catheter [0.09 to 0.45] Instrumental Assistance during Transfer All aspects [0.50 to 0.74] 187

198 Outcome Comparison No of No of Effect Measure Studies Cycles size* Use of Tenaculum [0.39 to 0.64] Using Stylette [0.70 to 1.26] Using Sounding [0.02 to 0.32] Using Dilatation [0.10 to 1.24] Catheter tip Blood [0.40 to 0.70] Mucus [0.01 to 8.21] Retained embryos [0.16 to 1.01] * Statistically significant 188

199 Figure 1: (A) Funnel and (B) trim and fill plots for ongoing pregnancy rates. RD = risk difference. 189

200 Figure 2: Meta-analysis forest plots showing live-birth rate following ultrasound-guided versus clinical touch ET. US = ultrasound. 190

201 Figure 3: Meta-analysis forest plots showing ongoing pregnancy rate following ultrasound-guided versus clinical touch ET. US = ultrasound. 191

202 Figure 4: Meta-analysis forest plots showing clinical pregnancy rate following ultrasound-guided versus clinical touch ET. US = ultrasound. 192

203 Chapter 12 Difficult embryo transfers and the presence of blood on the embryo transfer catheter negatively affect clinical pregnancy rates. Ahmed M. Abou-Setta, Sherif Sharkawy, Maha Shahed, Medhat Amer. (Submitted)

204 Abstract Introduction: The embryo transfer (ET) technique has become recognized as one of the most complex and significant variables in the success of assisted reproduction, with multiple factors associated with its success or failure. Design: Retrospective cohort study. Setting: Infertility centers. Patient(s): 943 women, aged years old, who underwent 1,122 ETs between January 1, 2005, and December 31, Intervention(s): NA. Main outcome measure(s): A semiquantitative system for grading and recording the difficulty of transfer, presence of retained embryos, mucus and blood found inside and outside the transfer catheter after ET was used to determine the correlation of each factor with the clinical pregnancy rate. Result(s): Difficult transfer and presence of blood on the outside or inside of the transfer catheter after ET was associated with decreased clinical pregnancy rates. In addition, there was a correlation between difficult transfers and the presence of blood. In contrast, presence of retained embryos or mucus did not significantly affect the outcomes. Conclusion(s): The relationship between difficult embryo transfer and the presence of blood on the outside or inside of the transfer catheter has been demonstrated to decrease the clinical pregnancy rates. Every effort should be made to minimalize difficulty, and hence bleeding, during embryo transfer. Key words: Difficult embryo transfer, blood, embryo catheter, retained embryos, IVF 194

205 Background Although most patients who undergo in vitro fertilization (IVF)/ intracytoplasmic sperm injection (ICSI) will reach the embryo transfer stage and have good quality embryos available for transfer, embryo implantation remains the rate-limiting step in the success of this form of therapy. The main factors that affect embryo implantation can be related to uterine receptivity, embryo quality, and efficiency of the embryo transfer procedure (1). The aim must be to transfer the embryos with a high degree of reliability atraumatically. In the early days of assisted reproduction, the embryo transfer technique was considered to be an unimportant factor in determining the outcome of the treatment cycle. This is mainly reflected by the relative lack of published studies or modification of the procedures constituting embryo transfer since it was first described. Even so, the importances of the transfer technique and each step within have recently become more appreciated by clinicians. In a series of communications with clinicians, factors such as lack of blood on the transfer catheter, mucus aspiration prior to transfer and lack of instrumental assistance during the transfer ranked high in importance (2, 3). In addition, multiple factors have recently been shown to significantly affect the success of transcervical intra-uterine embryo transfer; including the softness of the catheter (4, 5), the use of ultrasound guidance (6 8), the ease of the procedure (9), the absence of blood on the catheter (10, 11), the use of cervical introducers or obturators (12), flushing of the cervical canal (13) or gentle aspiration (14) to remove cervical mucus, and retention of embryos in the catheter (15, 16). Most importantly, difficulty in traversing the cervix has been considered to be one of the most important factors having a negative impact on the pregnancy rate. Furthermore, the presence of blood following embryo transfer has been associated with difficult embryo transfers and poor outcomes. The latter is believed to be as a result of endocervical and possibly endometrial damage, possibly resulting in uterine or junctional zone contractions (9, 17). In this study, we aimed to determine whether the ease of the procedure, the presence or absence of blood, mucus or retained embryos on the tips of the post-transfer catheter were correlated with the probability of a clinical pregnancy in a large cohort of women 195

206 undergoing ultrasound-guided embryo transfer with high quality embryos. 196

207 Materials and Methods Patient Population A detailed chart review of all ultrasound-guided embryo transfers performed in our centre from January 1, 2005 December 31, 2006 was performed. Details on patient demographics, cycle characteristics and outcomes were extracted, with special emphasis on the presence of blood, mucus, retained embryos and occurrence of difficulty during embryo transfer. Inclusion criteria consisted of female partner age at the time of transfer between years old, availability of high quality embryos (grade I or grade II) or blastocysts on the day of transfer, and undergoing a fresh embryo transfer. Patients aged >40 years, undergoing cryo-embryo transfer, or lacking high quality embryos on the day of transfer were excluded. The final cohort consisted of 943 women, undergoing 1,122 cycles. Of these, 787 women were having their first IVF-ET. This study was reviewed and approved by our local institutional review board (IRB). Ovarian stimulation In general, the majority of women were down-regulated using a long agonist protocol of pituitary down-regulation beginning on day 21 of the previous cycle. When down-regulation was considered adequate (e.g. serum E2 level of <35 pg/ml), ovarian hyperstimulation was begun with a standard initial daily dose of purified FSH, hmg and/or recombinant FSH. This was followed by individualization of the regimens according to the patient demographics (e.g. patient s age, ovarian reserve, response to previous stimulation), follicular development and physician preference. When at least two follicles 18 mm in diameter were seen on transvaginal ultrasound (US), human chorionic gonadotropin (10,000 IU IM) was administered, and transvaginal US-guided oocyte retrieval was performed ~36 hours later. Embryo transfer The majority of cases received an embryo transfer (ET) hours post-oocyte retrieval. In a minority of patients, a blastocyst transfer was performed. ET was performed with the patient in the dorso-lithotomy position. All transfers were performed by two experienced clinicians with similar pregnancy rates under ultrasound guidance. The Cook Sydney 197

208 IVF embryo catheter system was loaded with culture media till the tip (e.g. no air bubbles). The ultrasonography was performed by an experienced ultrasonographer, and patients were asked to maintain a full bladder during the transfer. Following embryo transfer, the embryologist microscopically inspected the transfer catheter for retained embryos, and any found were immediately reloaded and transferred. This was followed by recording of all events that occurred during the transfer. A semiquantitative scale was used to record the amount of mucus or blood (e.g., 0 = none, 1 = minimal, and 2 = significant). Scores were obtained separately for blood located inside and outside the catheter. In addition, the physician provided a score on the difficulty of the transfer (e.g. 0 = no resistance, 1 = minimal resistance, 2 = some resistance, 3 = marked resistance, 4 = dilator used, 5 = uterine sound used). Statistical analysis The primary outcome of this study was the clinical pregnancy rate per woman; defined as the visualization of a fetal sac 4 6 weeks posttransfer associated with an increasing maternal β-hcg titer. All analyses were performed on the overall cohort and on two selected populations: (1) patients undergoing their first cycle of IVF-ET, and (2) patients with only easy transfers. Univariate analyses using the χ 2 test were performed to determine whether the occurrence of each variable was associated with the occurrence of a clinical pregnancy. Moreover, multivariate logistic regression analyses were performed to determine the influence of the independent variables on the occurrence of a clinical pregnancy. For the purpose of the univariate analyses, in the blood or mucus categories, the cases with presence of minimal amounts of the respective variable were grouped with the absent group to form two distinctive groups: (1) absent or (2) present. In addition, difficulty during the embryo transfer was regrouped into two categories: (1) no to minimal resistance and (2) moderate to marked resistance, the latter including cases in which the use of instrumental assistance during transfer was required. All statistical analyses were performed using Statistical Package for the Social Sciences (SPSS). A P value <0.05 was considered to be statistically significant. 198

209 Results Difficulty during the embryo transfer procedure was recorded during 45 embryo transfers (4.02%) in the overall cohort and in 29 transfers (3.68%) in the 1 st cycle subgroup. Women with an easy transfer were significantly more likely to have a clinical pregnancy than women with a difficult transfer (45.40% vs %; O.R = 4.52, 95% CI = 2.00 to 10.20; P < ). In addition, in the subgroup of patients with only their first transfer, the trend was also similar (46.31% vs %; O.R = 2.71, 95% CI = 1.14 to 6.42; P = 0.03). Mucus was noted post-transfer in 271 cases (24.15%) in the overall cohort and in 197 cases (25.03%) in the 1st cycle only subgroup. In the overall cohort, the presence of mucus did not seem to alter the chances for a positive outcome (44.65% vs %; O.R = 1.08, 95% CI = 0.82 to 1.42; P = 0.64). In the 1 st cycle subgroup, the results were also similar (45.42% vs %; O.R = 0.99, 95% CI = 0.72 to 1.37; P = 1). Blood was noted on the outside of the transfer catheter in 94 instances (8.38%) and on the inside of the catheter in 60 (5.35%) in the overall cohort. In the 1 st cycle subgroup, the respective figures were 59 (7.50%) and 35 (4.45%). The presence of blood demonstrated a negative impact on the clinical pregnancy rates with pregnancy rates higher in cases in which there was no blood on the outside of the catheter (45.72% vs %; O.R = 2.20, 95% CI = 1.38 to 3.52; P = 0.001), or on the inside of the catheter (44.82% vs %; O.R = 1.63, 95% CI = 0.94 to 2.82; P = 0.11), but the latter did not reach statistical significance. In addition, the 1 st cycle subgroup revealed similar results for the presence of blood on the outside of the catheter (46.57% vs %; O.R = 1.84, 95% CI = 1.04 to 3.23; P = 0.046) and on the inside of the catheter (45.74% vs %; O.R = 1.27, 95% CI = 0.63 to 2.5; P = 0.62). Retained embryos were noted following 69 transfers (6.15%) and 44 transfers (5.59%) in the overall cohort and 1 st cycle subgroup, respectively. The retention of embryos and subsequent retransfer did not seem to significantly affect the pregnancy rates in either group (44.44% vs %; O.R = 1.17, 95% CI = 0.71 to 1.92; P = 0.62) and (45.49% vs %; O.R = 1.00, 95% CI = 0.54 to 1.85), respectively. 199

210 Moreover, using a series of regression analyses to determine the association between the variable and the chance for a clinical pregnancy, in the original cohort, multiple linear regression analysis demonstrated that only the difficulty in embryo transfer significantly affected the clinical pregnancy rate (P = ). The remaining variables (presence of mucus, retained embryos and blood) were considered to be non-significant. Even so, the influence of each individual variable on the occurrence of a clinical pregnancy using simple linear regression revealed differing results. The analyses not only confirmed that difficult transfers were correlated with the probability of a clinical pregnancy, but so were the presence of blood either on the outside or inside of the transfer catheter (Table 1). In addition, the presence of mucus and the occurrence of retained embryos were not associated with the primary outcome. Subgroup analyses of only patients undergoing their first embryo transfer (e.g. 1 st cycle subgroup) revealed similar results as the multiple linear regression analysis of the original cohort. Moreover, there was more homogeneity with the results of the individual simple linear regression analyses, with the difficulty of the embryo transfer being the only significant variable [(r) = -0.08, (r²= 0.007); 95% CI for r = to -0.02; P = 0.02]. Finally, in a subgroup analysis of only patients with an easy transfer revealed an association with the presence of blood, especially on the outside of the catheter with the occurrence of a clinical pregnancy. Other variables were considered to be non-significant. 200

211 Discussion The current study emphasizes the correlation between difficulty in performing the embryo transfer, presence of blood on the catheter and poor outcomes. Since difficult transfers have been associated with a poorer outcome than easy transfer, it would be useful to directly examine the uterine cavity for any lesions post-transfer. Since this is only possible in mock transfers, indirect measures of the degree of difficulty were utilized. These include physician assessment of difficulty, the need for instrumental assistance (e.g. tenaculum), and the presence of blood on the catheter post-transfer. Different approaches have been described in cases of difficult embryo transfers with varying success rates (18 21). A commonly used initial approach is to negotiate the cervix using the outer sheath of the catheter, with its inner noodle withdrawn (22). Once the uterine cavity is entered, the inner noodle is used to deposit the embryos, taking care to avoid the fundus. Even so, the pregnancy rates with this method have not been acceptable when the catheter guide enters the endometrial cavity (23). Moreover, more invasive and potentially traumatic events are sometimes undertaken by clinicians to overcome the problematic cervix. These include the use of a tenaculum, stylette, sounding and/or cervical dilatation. Overall these events have been associated with increased uterine junctional zone contractions and a decreased pregnancy rate (9, 20, 24, 25). Alternatively, the cervical route may be bypassed and the embryos may be transferred transmyometrially into the uterine cavity using the Towako method (19). Another tell-tale sign of a difficult transfer is the post-transfer presence of blood on the transfer catheter. Amongst clinicians, the absence of blood on the catheter or cannula is ranked high as an important factor towards success (2, 3). This opinion is supported by literature reports in which the presence of blood on the transfer catheter has been associated with lower pregnancy rates (10, 24). In addition, Perin et al. (26) found that contamination of the catheter with blood and mucus accounted for significantly lower implantation and clinical pregnancy rates. In the current study blood was negatively correlated with the clinical pregnancy rate. This was more evident for blood on the outside of the catheter than on the inside of the catheter. Moreover, even in easy 201

212 transfers, blood on the outside of the catheter was found to be a negative predictor of success. Finally, the role of retained embryos and mucus in decreasing the pregnancy rate is controversial with some studies claiming a negative effect (24) and other claiming no such effect (10, 27). In our study, we could not find an association between the retransfer of retained embryos and a drop in the pregnancy rate. In conclusion, the relationship between difficult embryo transfer, the presence of blood on the transfer catheter has been demonstrated to decrease the clinical pregnancy rates. Therefore, every effort should be made to minimalize difficulty, and hence bleeding, during embryo transfer. 202

213 References 1. Mansour RT, Aboulghar MA. Optimizing the embryo transfer technique. Hum Reprod May;17(5): Kovacs GT. Which factors are important for successful embryo transfer after in-vitro fertilization? Hum Reprod Oct;14(10): Salha OH, Lamb VK, Balen AH. A postal survey of embryo transfer practice in the UK. Hum Reprod Apr;16(4): Abou-Setta AM, Al-Inany HG, Mansour RT, Serour GI, Aboulghar MA. Soft versus firm embryo transfer catheters for assisted reproduction: a systematic review and meta-analysis. Hum Reprod Nov;20(11): Buckett WM. A review and meta-analysis of prospective trials comparing different catheters used for embryo transfer. Fertil Steril Mar;85(3): Buckett WM. A meta-analysis of ultrasound-guided versus clinical touch embryo transfer. Fertil Steril Oct;80(4): Sallam HN, Sadek SS. Ultrasound-guided embryo transfer: a meta-analysis of randomized controlled trials. Fertil Steril Oct;80(4): Abou-Setta AM, Mansour RT, Al-Inany HG, Aboulghar MM, Aboulghar MA, Serour GI. Among women undergoing embryo transfer, is the probability of pregnancy and live birth improved with ultrasound guidance over clinical touch alone? A systemic review and meta-analysis of prospective randomized trials. Fertil Steril Jun 6; [Epub ahead of print]. 9. Lesny P, Killick SR, Tetlow RL, Robinson J, Maguiness SD. Embryo transfer-- can we learn anything new from the observation of junctional zone contractions? Hum Reprod Jun;13(6): Goudas VT, Hammitt DG, Damario MA, Session DR, Singh AP, Dumesic DA. Blood on the embryo transfer catheter is associated with decreased rates of embryo implantation and clinical pregnancy with the use of in vitro fertilization-embryo transfer. Fertil Steril Nov;70(5): Alvero R, Hearns-Stokes RM, Catherino WH, Leondires MP, Segars JH. The presence of blood in the transfer catheter negatively influences outcome at embryo transfer. Hum Reprod Sep;18(9): Ghazzawi IM, Al-Hasani S, Karaki R, Souso S. Transfer technique and catheter choice influence the incidence of transcervical embryo expulsion and the outcome of IVF. Hum Reprod Mar;14(3): Sallam HN, Farrag F, Ezzeldin A, Agameya A, Sallam AN. Vigorous flushing of the cervical canal prior to embryo transfer, a prospective randomized study. Fertil Steril Oct;74(3, Suppl 1):S Eskandar MA, Abou-Setta AM, El-Amin M, Almushait MA, Sobande AA. Removal of cervical mucus prior to embryo transfer improves pregnancy rates in women undergoing assisted reproduction. Reprod Biomed Online Mar;14(3):

214 15. Friedler S, Lewin A, Schenker JG. Methodology of human embryo transfer following assisted reproduction. J Assist Reprod Genet Aug;10(6): Moore DE, Soules MR, Klein NA, Fujimoto VY, Agnew KJ, Eschenbach DA. Bacteria in the transfer catheter tip influence the live-birth rate after in vitro fertilization. Fertil Steril Dec;74(6): Fanchin R, Righini C, Olivennes F, Taylor S, de Ziegler D, Frydman R. Uterine contractions at the time of embryo transfer alter pregnancy rates after in-vitro fertilization. Hum Reprod Jul;13(7): Mansour R. Intracytoplasmic sperm injection: a state of the art technique. Hum Reprod Update Jan-Feb;4(1): Kato O, Takatsuka R, Asch RH. Transvaginal-transmyometrial embryo transfer: the Towako method; experiences of 104 cases. Fertil Steril Jan;59(1): Groutz A, Lessing JB, Wolf Y, Yovel I, Azem F, Amit A. Cervical dilatation during ovum pick-up in patients with cervical stenosis: effect on pregnancy outcome in an in vitro fertilization-embryo transfer program. Fertil Steril May;67(5): Tur-Kaspa I, Yuval Y, Bider D, Levron J, Shulman A, Dor J. Difficult or repeated sequential embryo transfers do not adversely affect in-vitro fertilization pregnancy rates or outcome. Hum Reprod Sep;13(9): Glass KB, Green CA, Fluker MR, Schoolcraft WB, McNamee PI, Meldrum DR. Multicenter randomized controlled trial of cervical irrigation at the time of embryo transfer. (abstract no. O-085). Fertil Steril. 2000;74 (Suppl 1) S Abdelmassih VG, Neme RM, Dozortsev D, Abdelmassih S, Diamond MP, Abdelmassih R. Location of the embryo-transfer catheter guide before the internal uterine os improves the outcome of in vitro fertilization. Fertil Steril May 15; [Epub ahead of print]. 24. Visser DS, Fourie FL and Kruger HF (1993) Multiple attempts at embryo transfer: effect on pregnancy outcome in an in vitro fertilization and embryo transfer program. J Assist Reprod Genet 10, Lesny P, Killick SR, Tetlow RL, Robinson J, Maguiness SD. Embryo transfer and uterine junctional zone contractions. Hum Reprod Update Jan- Feb;5(1): Perin PM. The influence of two different transfer catheters on the pregnancy rate in a human in vitro fertilization program. Reprod Clim. 1999;14: Nabi A, Awonuga A, Birch H, Barlow S, Stewart B. Multiple attempts at embryo transfer: does this affect in-vitro fertilization treatment outcome? Hum Reprod Jun;12(6):

215 Table 1: Simple linear regression of independent variable and clinical pregnancy rate in the overall cohort. Correlation coefficient 95% CI for r P value r = Difficulty in transfer to P = r² = 0.01 Blood on the outside of the catheter r = to P = r² = Blood on the inside of the catheter r = to P = r² = r = Mucus to 0.03 P = 0.31 r² = r = Retained embryos to 0.04 P = 0.53 r² =

216 206

217 Chapter 13 Difficult Embryo Transfer: The Impact of Propofol Anesthesia. Ahmed M. Abou-Setta, Ragaa T. Mansour, Hesham G. Al-Inany, Gamal I. Serour, Mohamed A. Aboulghar, Mohamed El-Wassify Middle East Fertility Society Journal. 2007;12:

218 Abstract Background: Difficult embryo transfers (ET) requiring general anesthesia are occasionally encountered in clinical practice. Little evidence is present in the literature as to the success rates when compared with difficult transfers not requiring anesthesia. Objective: To evaluate the impact of using Propofol anesthesia during difficult embryo transfers on the implantation and clinical pregnancy rates. Design: Retrospective patient chart review. Materials and methods: Women undergoing ICSI cycles in the Egyptian IVF-ET center, from January 2000 December 2002, and having difficult ET requiring general anesthesia (Group I = 99 women) were included. A matching group of women with difficult ET, without anesthesia (Group II = 99 women) were used as a control. Results: There were no significant differences in the patient demographics (e.g. age, period of infertility, number of oocytes retrieved, fertilization rate, embryo quality, number of embryos transferred. Moreover, there was no significant differences in implantation (Group I = 19.15%, Group II = 20.86%) or clinical pregnancy rates (Group I = 36.36%, Group II = 33.33%). Conclusion: The use of propofol general anesthesia during difficult embryo transfer does not seem to improve the implantation and pregnancy rates. Even though, prospective randomized trials are needed to confirm these findings. Key words: Propofol, Embryo transfer, Assisted Reproduction, ICSI, IVF 208

219 Introduction Embryo transfer (ET) is the final step in the IVF/ ICSI treatment process. It is also the last step in which clinical manipulation may directly affect the outcome of the assisted reproduction treatment cycle. The majority of patients undergoing IVF/ ICSI will reach the transfer stage, with good quality embryos available for transfer, but only a small proportion of them will ever achieve a clinical pregnancy. It is estimated that up to 85% of the embryos replaced into the uterine cavity will fail to implant (1). The pregnancy rate following embryo transfer is generally dependent upon multiple factors including embryo quality, endometrial receptivity and the technique of the embryo transfer itself (2). One important aspect of the embryo transfer technique that has received limited attention is the use of propofol anesthesia during difficult embryo transfers. Whether Propofol anesthesia during embryo transfer could have a potential impact on conception is not clear from the medical literature. This issue was not even mentioned as an important variable that might affect the outcome of embryo transfer in two recent reviews of the literature (2, 3). Even so, it is clear from a recent systematic review and meta-analysis that difficult embryo transfer is associated with a significant drop in implantation rate when compared to easy transfers (O.R = 0.64, 95% CI = ) (4). In the medical literature, there is conflicting evidence on the impact of different anesthetic agents used during oocyte retrieval on pregnancy rates following embryo transfer (5 8). Moreover, only a few studies have discussed the impact of general anesthesia during the embryo transfer procedure on the implantation and clinical pregnancy rates (9 11). In a previous work by our group, there was no adverse or beneficial effect of propofol anaesthesia on both implantation and pregnancy rate in women with easy embryo transfer (12). Therefore in the present study, we wished to evaluate the possible beneficial or detrimental effect of using propofol anesthesia on IVF-ET outcome in patients undergoing a difficult embryo transfer. 209

220 Materials and Methods Data obtained from clinical records of subfertile couples undergoing embryo transfer under general anesthesia over a two-year period were obtained and analyzed. The inclusion criteria were: female age < 39 years old, with normal hormonal profile and no pelvic pathology, and no surgically retrieved sperms. Inclusion criteria were difficultly in cannulating the cervix encountered during embryo transfer with clinician preference of using general Propofol anesthesia. Patients undergoing ICSI in the same period of time with the same inclusion criteria and which did not have general anesthesia for ET were selected as a control group. All participants received the GnRHa long protocol, 0.1mg/day (Decapeptyl, Ferring Pharmaceuticals) starting on day 20 of the cycle till the day of hcg injection. After down regulation was confirmed, I.U of hmg/day was started for 7 days; then the dose was adjusted according to the ovarian response as measured by serum estradiol levels and ultrasound monitoring. Ten thousand international units of hcg (Pregnyl; Nile Co., Cairo, Egypt) were given I.M. when two or more follicles reached ~18 mm in mean diameter. Ovum retrieval using transvaginal ultrasound was scheduled 36 hours after hcg injection. All participants were enrolled in our ICSI program, which is described elsewhere (13). None of the patients in either group received premedication prior to the embryo transfer. In the anaesthesia group, a 22-gauge catheter was inserted in one arm, ECG (lead II) was connected and blood pressure and pulse oximetry were instituted. Induction of general anaesthesia consisted of pre-oxygenation by face mask, followed by intravenous bolus of 2 mg/kg propofol (Diprivan ; Zeneca, Manchester, UK) as an induction dose and anaesthesia was maintained by inhalation of isoflurane 1.5% and oxygen 100% through a face mask. Embryo transfer was performed according to the standard procedure used in our center (14). In general, embryo transfer was performed h after oocyte pick-up using the Wallace catheter (H.G.Wallace Ltd, West Sussex, UK). In both groups, the embryo transfer was done at the lithotomy position. The previously taken ultrasound picture of the uterus and dummy embryo transfer (Mansour et al., 1990) were revised to get an idea of the length and direction of the uterine cavity. After performing the dummy embryo transfer, the embryos were loaded into 210

221 a new catheter, either Wallace or Cook according to the dummy embryo transfer as follows. The embryo transfer catheter was rinsed then filled with tissue culture medium. About ml tissue culture medium was aspirated, then the embryos (up to three) were aspirated in ml tissue culture medium so that the embryos would be away from the tip of the catheter. If the Wallace catheter was used, the soft internal catheter, protruding from the external rigid sheath, was introduced gently through the internal cervical os and stopped 1 cm below the fundus. The outer rigid sheath was stopped just at the internal cervical os and not pushed beyond it. If the Cook catheter was used, the tip of the inner catheter was positioned flush with the external sheath until it passed the internal cervical os, then the internal sheath one was advanced 2 cm into the uterine cavity. Luteal phase support was given routinely in the form of a daily progesterone injection (100 mg, progesterone; Steris, Phoenix, AZ, USA). A serum β-hcg test was done to confirm pregnancy two weeks after the embryo transfer. Clinical pregnancy was diagnosed 3 weeks after a positive test by the presence of a gestational sac with fetal echoes and pulsations on ultrasound. The primary outcome measures for this study was the odds of a clinical pregnancy and embryo implantation following Propofol anesthesia compared to no anesthesia. Statistical evaluation Data are presented as mean ± SD. Different outcome measures were compared using Student's t-test, X 2 or Fisher's exact test where appropriate. Odds ratios (using the Woolf (Logit) method), 95% confidence intervals and P-values are presented. A P-value of <0.05 was considered to be significant. Statistic analysis was performed using Arcus Quickstat BioMedical (Version 1.0). 211

222 Results The present study enrolled 198 women who had completed an IVF/ET cycle, and divided into two groups: Group I (99 women who had difficult embryo transfer under general anesthesia), and Group II (99 women who had difficult embryo transfer without general anesthesia). There was no statistically significant difference in the age, infertility duration, number of oocytes retrieved, fertilization rate or embryos obtained between both groups (Table I). In addition, there was no statistically significant difference between both groups regarding clinical pregnancy (Group I = 36/ 99 versus Group II = 33/ 99; O.R = 1.14, 95% CI = 0.64 to 2.05; P = 0.77) or implantation rates (Group I = 54/ 282 versus Group II = 58/ 278; O.R = 0.89, 95% CI = 0.59 to 1.36; P = 0.69). 212

223 Discussion Embryo transfer is the final and most crucial step in IVF. Patients experiencing difficult embryo transfer are not uncommon in daily practice, especially in large infertility centers. In general, difficultly in threading the cervix occurs in about ~5% of all embryo transfers (15). Several techniques have been discussed as ways of bypassing the unrelenting cervix. These include the use of cervical dilatation, using dilators (16) or hygroscopic rods (17), cervical shaving to widen the cervical canal in cases of cervical stenosis (18) and the use of instrumental assistance during the embryo transfer such as tenaculums, sounding or dilatation. As a last resort, some authors prefer to use transmyometrial embryo transfer to deposit the embryos in the uterine cavity (19). Even though these techniques may assist the clinician in reaching the endometrial cavity, they also carry associated risks, including stimulating uterine contractions, cervical and endometrial injuries and lacerations, increase in the presence of blood on the catheter tip and cervix, and most importantly, cumulatively a decrease in clinical pregnancy and implantation rates (20). Propofol (Diprivan ) is an intravenous sedative-hypnotic drug widely used as a sole anesthetic agent in minor interventions, such as ovum pick up, cervical dilatation and minor intrauterine procedures. Propofol (2,6-diisopropylphenol) is an alkylphenol that has a rapid onset of action and recovery. In addition, Propofol exerts smooth muscle relaxation, a mechanism that may involve IP 3 -induced SR Ca 2+ release in calcium signaling pathways. Propofol acts as a hypnotic and can be used as both an induction agent and/ or as maintenance anesthetic. A three compartment linear model, characterizes Propofol s pharmacokinetics, with fast distribution into the tissues, rapid metabolic clearance, and a slow return to the peripheral compartment. Following the initial bolus dose, propofol equilibrates between the plasma and the brain. Plasma levels, however, decline quickly as a result of high metabolic clearance and prompt distribution to the tissues. These properties account for Propofol's rapid onset and short duration of action. Distribution time decreases as tissues equilibrate with plasma and become saturated. Elimination is triphasic; with the distribution half-life being 2 10 minutes; the second phase half-life being minutes; and the terminal elimination half-life

224 to almost 30 hours. The last phase is not thought to be clinically significant. Mean induction time is 30 to 40 seconds after a 2.0 to 2.5 mg/kg bolus. Discontinuation of propofol anesthesia usually results in a rapid decrease in plasma concentrations and prompt awakening. Propofol has a clearance rate of ml/kg/minute. The presence of hepatic cirrhosis or renal insufficiency does not appear to significantly alter its pharmacokinetics (21). To overcome the limitations of retrospective analysis, both groups (with and without anesthesia) were matching regarding age, duration of infertility, number of oocytes retrieved, fertilization rates and number of embryos transferred. The participants of the control group were only recruited during the same period of time in order to guarantee that the lab conditions were the same.. Despite controversial reports with regard to the influence of propofol anesthesia on implantation rates and clinical pregnancy rates in humans (22 25). The previous studies investigated the impact of propofol used during ovum pickup, or Gamete Intra-fallopian Transfer (GIFT), or during easy embryo transfer on pregnancy rate and implantation rates. To the best of our knowledge this is the first study to determine the beneficial or detrimental effect of Propofol anesthesia during difficult embryo transfers and it confirms our previous work (12). In the present study, there was no statistically significant difference in implantation, or clinical pregnancy rates. This concludes that there is no evidence from our data that the administration of propofol during the procedure of embryo transfer had a negative impact on the embryos as measured by probability of a clinical pregnancy or implantation rates. Therefore, Propofol anesthesia may offer clinicians a complementary measure while dealing with a difficult embryo transfer. Even so adequately powered randomized controlled trials are needed to confirm our findings. 214

225 References 1. Edwards RG. Clinical approaches to increasing uterine receptivity during human implantation. Hum Reprod. 1995;10(Suppl 2): Mansour RT, Aboulghar MA. Optimizing the embryo transfer technique. Hum Reprod May;17(5): Schoolcraft WB, Surrey ES, Gardner DK. Embryo transfer: techniques and variables affecting success. Fertil Steril Nov;76(5): Sallam H, Sameh S, Sadek S, and Agameya A. Does a difficult embryo transfer affect the results of IVF and ICSI? A meta-analysis of controlled studies. Fertil Steril. 2003; 80 (3 Suppl): Gonen O, Shulman A, Ghetler Y, Shapiro A, Judeiken R, Beyth Y, Ben-Nun I. The impact of different types of anesthesia on in vitro fertilization-embryo transfer treatment outcome. J Assist Reprod Genet Nov;12(10): Coetsier T, Dhont M, De Sutter P, Merchiers E, Versichelen L, Rosseel MT. Propofol anaesthesia for ultrasound guided oocyte retrieval: accumulation of the anaesthetic agent in follicular fluid. Hum Reprod Nov;7(10): Ben-Shlomo I, Moskovich R, Golan J, Eyali V, Tabak A, Shalev E. The effect of propofol anaesthesia on oocyte fertilization and early embryo quality. Hum Reprod Oct;15(10): Christiaens F, Janssenswillen C, Verborgh C, Moerman I, Devroey P, Van Steirteghem A, Camu F. Propofol concentrations in follicular fluid during general anaesthesia for transvaginal oocyte retrieval. Hum Reprod Feb;14(2): Fishel S, Webster J, Faratian B, Jackson P. General anesthesia for intrauterine placement of human conceptuses after in vitro fertilization. J In Vitro Fert Embryo Transf Oct;4(5): van der Ven H, Diedrich K, Al-Hasani S, Pless V, Krebs D.The effect of general anaesthesia on the success of embryo transfer following human in-vitro fertilization. Hum Reprod Oct;3 Suppl 2: Diedrich K, van der Ven H, al-hasani S, Krebs D. Establishment of pregnancy related to embryo transfer techniques after in-vitro fertilization. Hum Reprod Nov;4(8 Suppl): Al-Inany HG, Wasseef M, Aboulghar MA, Mansour RT, Serour GI, Abou-Setta AM. Embryo Transfer under Propofol Anaesthesia: The Impact on Implantation and Pregnancy Rate. Middle East Fertil Soc J. 2003;8(3): Mansour RT, Aboulghar MA, Serour GI, Tawab NA, Amin YM, Sattar MA. Successful intracytoplasmic sperm injection without performing cytoplasmic aspiration. Fertil Steril 1996;66: Mansour R. Minimizing embryo expulsion after embryo transfer: a randomized controlled study. Hum Reprod Jan;20(1): Epub 2004 Nov Mansour RT, Aboulghar MA. Optimizing the embryo transfer technique. Hum Reprod May;17(5): Abusheikha N, Lass A, Akagbosu F, Brinsden P. How useful is cervical dilatation in patients with cervical stenosis who are participating in an in vitro fertilization- 215

226 embryo transfer program? The Bourn Hall experience. Fertil Steril Oct;72(4): Serhal P, Ranieri DM, Khadum I, Wakim RA. Cervical dilatation with hygroscopic rods prior to ovarian stimulation facilitates embryo transfer. Hum Reprod Dec;18(12): Noyes N. Hysteroscopic cervical canal shaving: a new therapy for cervical stenosis before embryo transfer in patients undergoing in vitro fertilization. Fertil Steril May;71(5): Sharif K, Afnan M, Lenton W, Bilalis D, Hunjan M, Khalaf Y. Transmyometrial embryo transfer after difficult immediate mock transcervical transfer. Fertil Steril May;65(5): Lesny P, Killick SR, Tetlow RL, Robinson J, Maguiness SD. Embryo transfer--can we learn anything new from the observation of junctional zone contractions? Hum Reprod Jun;13(6): Fulton B, Sorkin EM. Propofol: an overview of its pharmacology and a review of its clinical efficacy in intensive care sedation. Drugs 1995;50: Pierce ET, Smalky M, Alper MM, Hunter JA, Amrhein RL, Pierce EC Jr. Comparison of pregnancy rates following gamete intrafallopian transfer (GIFT) under general anesthesia with thiopental sodium or propofol. J Clin Anesth Sep- Oct;4(5): Vincent RD Jr, Syrop CH, Van Voorhis BJ, Chestnut DH, Sparks AE, McGrath JM, Choi WW, Bates JN, Penning DH. An evaluation of the effect of anesthetic technique on reproductive success after laparoscopic pronuclear stage transfer. Propofol/nitrous oxide versus isoflurane/nitrous oxide. Anesthesiology Feb;82(2): Rosenblatt MA, Bradford CN, Bodian CA, Grunfeld L. The effect of a propofolbased sedation technique on cumulative embryo scores, clinical pregnancy rates, and implantation rates in patients undergoing embryo transfers with donor oocytes. J Clin Anesth Dec;9(8): Beilin Y, Bodian CA, Mukherjee T, Andres LA, Vincent RD Jr, Hock DL, Sparks AE, Munson AK, Minnich ME, Steinkampf MP, Christman GM, McKay RS, Eisenkraft JB. The use of propofol, nitrous oxide, or isoflurane does not affect the reproductive success rate following gamete intrafallopian transfer (GIFT): a multicenter pilot trial/survey. Anesthesiology Jan;90(1):

227 Table 1. Demographic and cycle characteristics of the Propofol anesthesia and no anesthesia groups. Anesthesia (Mean ± SD) No Anesthesia (Mean ± SD) Significance Age ± ± 4.68 P = 0.78 Infertility 8.99 ± ± 3.72 P = 0.14 Oocytes ± ± 6.95 P = 0.32 PB ± ± 5.26 P = PN 7.01 ± ± 3.53 P = 0.91 Cryopreserved Embryos Embryos Transferred High Quality Embryos (G1) Good Quality Embryos (G2) Fair Quality Embryos (G3) Fertilization Rate Pregnancy Rate Implantation Rate 2.51 ± ± 3.16 P = ± ± 0.79 P = ± ± 0.88 P = ± ± 0.86 P = ± ± 0.62 P = % 64.42% P = % 33.33% P = % 20.86% P =

228 218

229 Chapter 14 General Discussion

230 Since the early days of in vitro fertilization, emphasis has mainly been on methods of ovarian stimulation and laboratory methods for embryo manipulation. The results of countless hours of laboratory and clinical research are evident in the marked advancement of these refined techniques, and the excellent results available today. As a consequence, most patients who enter an assisted reproduction program today will reach the stage of embryo transfer with adequate numbers of welldeveloped embryos. Even so, the majority will fail to become clinically pregnant. This phenomenon makes the understanding of the elements related to embryo implantation, and the techniques by which human embryos are transferred into the recipient uterus, vital for the advancement of assisted reproduction. Traditionally, the steps involved in the transfer of embryos into the receptive uterus have mainly been left up to personal preferences. This is reflected by both the scarce volume of scientific publications regarding the embryo transfer technique as a whole, or its subunits in particular, compared with other aspects of IVF. The only way a program can expect stability in clinical outcomes and a decrease in the often marked fluctuations between different clinicians and time periods is to implement a standardized protocol based upon evidence for all phases of the IVF cycle, including the embryo transfer technique. However, many of the steps involved in the transfer technique are enforced empirically and are still highly debated in the literature. Many centres have opted to using a retrospective or observational approach to improving their pregnancy rates. This mainly utilizes a method of identifying the aspects leading to success by different operators and forcing all clinicians to mimic the procedures of the most successful operators. Even though this may beneficial is some circumstances, it is both low on the evidence pyramid, may be highly biased, and is often incorrect. In the early days of IVF, the embryo transfer technique was mainly built on the descriptions provided by Edwards et al. (1). Over the years, many aspects of the embryo transfer technique have been put to the test, modified and perfected in order to provide the best pregnancy rates, and lowest chances of complications (e.g. ectopic pregnancies) possible. Unfortunately, this is not true for all aspects of the procedure, some of which have been neglected, while others have been accepted as 220

231 they are, without much questioning of the clinical effectiveness or efficiency. The individual technical performance of the clinician performing the embryo transfer has been demonstrated to be one of the most critical factors affecting success (2 3). Inter-operator variation in subtle issues as catheter type or gentleness in inserting the catheter has been shown to dramatically affect the results in the same assisted fertilization programme (4 5). One study comparing the pregnancy rates following embryo transfer by different physicians in the same assisted reproduction programme revealed several key factors affecting the outcomes (6). The authors examined 393 clinical pregnancies resulting from 854 embryo transfers, in which the number and quality of embryos transferred did not differ significantly between the different operators. Even so, pregnancy rate fluctuations more than three-fold (range: 17.0% to 54.3%) were observed in different operators. Some authors have noted that success rates are correlated to operator experience (7), denoting that acquired experience helps to fine-tune a physician s clinical judgement when performing an embryo transfer. These data have been interpreted as meaning that training and experience are the two most important issues in embryo transfer. Accordingly, it has been argued that well trained nurses may perform the embryo transfer as well as physicians (8), which now has become standard practice in many centres. Regardless of the academic qualifications of the individual (qualified clinician, in-training physician or trained nurse) performing this highly sensitive and meticulous step in IVF, it is clear that technical skills and experience are important factors. But this should not lead to a programme founded on personal experience complemented with a method of trial and error. In the same way that there is no reason to re-invent the wheel, there is also no reason to re-investigate each and every step in the ET technique by every operator. An excellent example of how clinical intuition alone has failed in the past is the discovery of the negative effects of uterine contractions at the time of embryo transfer. These contractions have been adequately demonstrated to lead to embryo expulsion following embryo transfer (9). Before this important discovery was made, it was customary to determine the site of embryo transfer by introducing the embryo 221

232 transfer catheter, containing the embryos, until the tip touches the uterine fundus, then pulling back ~ 1 cm and depositing the embryos. What is needed is an adequate assessment of what is known and how the evidence supports these aspects of the transfer technique. Fortunately, during the last years more and more publications discussing ways of improving embryo transfer and hopefully pregnancy rates have been published. In short, among the multiple factors shown to affect the success of embryo transfer are the experience of the physician (7, 10), the use of ultrasound guidance (11 13), the ease of the procedure (14 15), the presence or absence of blood on the catheter (16) and bacterial contamination of the catheter (18). In addition, other factors that might affect the chance for an ongoing pregnancy have been identified, such as the use of cervical introducers or obturators (19), resting after transfer (20), the position of embryo insertion in the uterus (21 22), flushing of the cervical canal to remove mucus (23), microbiological factors in terms of the local flora (24) and retention of embryos in the catheter (25-26). Since it would be difficult to determine the effect that each aspect of the transfer technique may have on limiting the success of the embryo transfer, we have divided the embryo transfer technique into four distinct sections: (i) preparation prior to embryo transfer; (ii) technical aspects related to the embryo transfer catheter; (iii) the embryo transfer procedure; and (iv) post-transfer aspects. The aim of this thesis was to address several of the individual points that make up each step of the procedure in order to determine what is beneficial and what is not when performing embryo transfer. Regarding the first phase of embryo transfer, the preparation prior to embryo transfer, we compared the outcome of cervical mucus removal prior to embryo transfer to no aspiration, and whether passive uterine straightening during embryo transfer affects pregnancy rates. We determined that both cervical mucus aspiration and passive uterine straightening are beneficial prior to embryo transfer. Therefore it was our recommendation that both should be performed in women prior to the actual embryo transfer procedure. Even so, there are other aspects in this phase that were not covered in our research, and which might affect the clinical outcomes. These include the patient position (e.g. lithotomy, knee-chest, or supine position), mock (or dummy) embryo transfers prior to the actual 222

233 procedure, cervical dilation (in cases of expected difficulty in transfer) and medications to decrease uterine irritability (e.g. oxytocin antagonists). Future research into these clinical queries is sure to add more efficiency and effectiveness to the transfer procedure by decreasing the rates of difficult transfers and improving the outcomes. Regarding the second phase of embryo transfer, the technical aspects related to the embryo transfer catheter, we tested the theories that both catheter type and catheter loading techniques could affect the clinical outcomes. Our research demonstrated that in general soft catheters performed better than firmer catheters, except under ultrasoundguidance. In addition, we determined that two of the commonly used catheters have similar outcomes. Finally, we determined that the use of air bubbles to bracket the embryo containing media inside the embryo transfer catheter does not affect the clinical outcomes. Other areas for future consideration include the use of different commercially available catheters, the use of echogenic catheters compared to non-echogenic catheters during ultrasound-guided embryo transfer and the use of different transfer medias (e.g. enriched with hyaluronic acid). Regarding the third phase of embryo transfer, the embryo transfer procedure, we investigated the use of ultrasound-guidance to assist in the accurate placement of the embryo within the uterus, and attempted to determine the best site of embryo deposition within the human uterus. Our results showed that ultrasound-guidance during embryo transfer was highly beneficial, and that a lower (e.g. ~20 mm) site of transfer seems to be preferable over the commonly used 10 mm distance from the uterine fundus. With regards the latter, more trials are needed to clarify the exact site for transfer. Last but not least, regarding the fourth phase of embryo transfer, the post-transfer aspects, we did not investigate these factors in this thesis. Issues still under debate include bed rest (and what period of length) following embryo transfer, and methods of decreasing embryo expulsion following transfer. In addition to the former points that were discussed, we also investigated whether the day of transfer, the presence of blood on the post-transfer catheter and the use Propofol anaesthesia during difficult transfers could affect the clinical outcomes. The results of these investigations revealed that the day of transfer (day 2 vs. day 3) and 223

234 the use of Propofol anaesthesia did not negatively affect the outcomes negating these decisions to a case by case scenario and the preference of the clinical staff. Even so, we did notice that the presence of blood on the post-transfer catheter was related negatively to the pregnancy rates; this denoting that an easy transfer results in more preferable outcomes. Finally, it is important to reflect back on the cumulative achievements of the research presented in this thesis and how it should be implemented into everday clinical work. It is important to note that the most important measure of success in IVF is neither the results of ovarian stimulation nor the quantity/quality of embryos produced in the laboratory. The most important criterium of success is patient satisfaction, and this will not be accomplished without a pregnancy. This is what patients undergoing IVF desire, and it is our duty as clinicians to perform everything in our power to attain this goal. What is evident from this research presented here, is that as part of the standard protocol of embryo transfer in every assisted reproduction program, passive uterine straightening and cervical mucus aspiration should be performed prior to embryo transfer. In addition, soft catheters under abdominal ultrasound-guidance should be used to assist in the accurate placement of the embryos within the uterus at an average of ~20 mm from the uterine fundus. The site of deposition may also be more toward the center of the uterine cavity. Finally, it is of paramount importance to decrease the rates of difficult transfers by any method possible. Our experience with light anaesthesia (e.g. Propofol) has been shown to be effective and should be implemented in patients with anticipated difficulty in traversing the unrelenting cervix. Finally, embryo transfers may be scheduled on day 2 3 following oocyte pick-up according to clinical judgement, patient scheduling, and center flow on a case-by-case scenario without fearing any change in clinical outcomes. In conclusion, it is important to note that the embryo transfer technique, as other areas of assisted reproduction, is becoming more evidence-based, relying on good, solid evidence to help modify the available protocols of practice. Even so it is important to note, that just as any other clinical procedure, in the end a human factor is involved. This factor needs to be trained and fine-tuned according to the best available evidence today so that the efficiency and effectiveness of this procedure can reach the great success as other aspects of assisted 224

235 reproduction. This will not take place until clinicians realize that personal preferences and experience do not out way the evidence, but on the contrary, complement and modify the evidence. Once we reach this point, we will truly have reached the optimum level of performance of this highly sensitive procedure. 225

236 References: 1. Edwards RG, Fishel SB, Cohen J, Fehilly CB, Purdy JM, Slater JM, et al. Factors influencing the success of in vitro fertilization for alleviating human infertility. J. In Vitro Fertil. Embryo Transfer. 1984;1:3. 2. Mansour RT, Aboulghar MA. Optimizing the embryo transfer technique. Hum Reprod May;17(5): Pasqualini RS, Quintans CJ. Clinical practice of embryo transfer. Reprod Biomed Online Jan-Feb;4(1): Nazari, A., Askari, H.A., Check, J.H. et al. Embryo transfer technique as a cause of ectopic pregnancy in in-vitro fertilization. Fertil. Steril. 1993;(60) Boomsma CM, Macklon NS. What can the clinician do to improve implantation? Reprod Biomed Online Dec;13(6): Hearns-Stokes RM, Miller BT, Scott L, Creuss D, Chakraborty PK, Segars JH. Pregnancy rates after embryo transfer depend on the provider at embryo transfer. Fertil Steril Jul;74(1): Papageorgiou TC, Hearns-Stokes RM, Leondires MP, Miller BT, Chakraborty P, Cruess D, Segars J. Training of providers in embryo transfer: what is the minimum number of transfers required for proficiency? Hum Reprod Jul;16(7): Cheung WM, Ng EH, Lau EY, Yeung WS, So WW, Ho PC. Is there any difference in pregnancy and implantation rates when nurses perform embryo transfer in an IVF-ET program? Gynecol Obstet Invest. 2003;56(1): Fanchin R, Righini C, Olivennes F, Taylor S, de Ziegler D, Frydman R. Uterine contractions at the time of embryo transfer alter pregnancy rates after in-vitro fertilization. Hum Reprod Jul;13(7): Lu MC (1999) Impact of non-physician factors on the physician factor of in vitro fertilization success: is it the broth, the cooks, or the statistics? Fertil Steril 71, Buckett WM (2003) A meta-analysis of ultrasound-guided versus clinical touch embryo transfer. Fertil Steril 80, Sallam HN and Sadek SS (2003) Ultrasound-guided embryo transfer: a metaanalysis of randomized controlled trials. Fertil Steril 80, Abou-Setta AM, Mansour RT, Al-Inany HG, Aboulghar M, Serour GI, Aboulghar MA. Among women undergoing embryo transfer, is the probability of pregnancy and live birth improved with ultrasound-guidance than with clinical touch alone? a systemic review and meta-analysis of prospective randomized trials. Fertil Steril. 2007;88: Lesny P, Killick SR, Tetlow RL, Robinson J and Maguiness SD (1998) Embryo transfer: can we learn anything new from the observation of junctional zone contractions? Hum Reprod 13, Lesny P, Killick SR, Tetlow RL, Robinson J and Maguiness SD (1999) Embryo transfer and uterine junctional zone contractions. Hum Reprod Update 5,

237 16. Goudas VT, Hammitt DG and Damario MA (1998) Blood on the embryo transfer catheter is associated with decreased rates of embryo implantation and clinical pregnancy with the use of in vitro fertilization embryo transfer. Fertil Steril 70, Abou-Setta AM, Sharkawy S, Shahed M, Amer M. Difficult embryo transfers and the presence of blood on the embryo transfer catheter negatively affect clinical pregnancy rates. (Submitted). 18. Egbase PE, al-sharhan M, al-othman S, al-mutawa M, Udo EE and Grudzinskas JG (1996) Incidence of microbial growth from the tip of the embryo transfer catheter after embryo transfer in relation to clinical pregnancy rate following invitro fertilization and embryo transfer. Hum Reprod 11, Ghazzawi IM, Al-Hasani S, Karaki R and Souso S (1999) Transfer technique and catheter choice influence the incidence of transcervical embryo expulsion and the outcome of IVF. Hum Reprod 14, Woolcott R and Stanger J (1997) Potentially important variables identified by transvaginal ultrasound-guided embryo transfer. Hum Reprod 12, Yovich JL, Turner SR and Murphy AJ (1985) Embryo transfer technique as a cause of ectopic pregnancies in in vitro fertilization. Fertil Steril 44, Waterstone J, Curson R and Parsons J (1991) Embryo transfer to low uterine cavity. Lancet 337, Sallam HN, Farrag F, Ezzeldin A, Agameya A, Sallam AN (2000) Vigorous flushing of the cervical canal prior to embryo transfer, a prospective randomised study. Fertil Steril 74;3(Suppl 1)S Ralph SG, Rutherford AJ and Wilson JD (1999) Influence of bacterial vaginosis on conception and miscarriage in the first trimester: cohort study. Br Med J 319, Friedler S, Lewin A and Schenker JG (1993) Methodology of human embryo transfer following assisted reproduction. J Assist Reprod Genet 10, Moore DE, Soules MR, Klein NA, Fujimoto VY, Agnew KJ and Eschenbach DA (2000) Bacteria in the transfer catheter tip influence the live-birth rate after in vitro fertilization. Fertil Steril 74,

238 228

239 Chapter 15 Summary

240 The aim of this thesis was to determine the best available evidence on procedures used to optimize the embryo transfer technique. Intrauterine transfer of human embryos into the recipient uterus may, at present, well be the rate limiting procedure and many factors have been shown to affect clinical outcome. In order to determine which embryo transfer techniques are truly associated with better outcomes, we performed some clinical trials and systematic reviews of the medical literature. Accordingly, the thesis aimed as outlined in chapter I to answer the following twelve specific clinical queries. In addition, two appendices have been provided. The first details the author s contributions to each publication in accordance with the Guidelines on Good Publication Practice published by the Committee on Publication Ethics (COPE). The second provides a broad search strategy that can be modified to allow readers to perform a comprehensive search on the topic of embryo transfer. Chapter 2 addresses the first question Is the pregnancy rate improved by delaying embryo transfer from day two to day three? To answer this question, we conducted a prospective quasi-randomized controlled trial comparing the outcome of embryo transfer on day two versus day three post-oocyte pick-up. We included a total of 927 consecutive embryo transfers (626 transfers on day 2 and 301 on day 3) following IVF/ICSI. The results demonstrated that there was no significant difference in the pregnancy rate between ET on day 2 (319/626) and ET on day 3 (152/ 301) (O.R = 1.02, 95% CI = 0.77 to Although our study was not randomized, selection bias is probably minimal, as patients who had egg retrieval on Wednesdays and Thursdays had their ET on day 3, and those who had egg retrieval on other days of the week had their ET done on day 2. Criteria for giving hcg was fixed all through the study when the lead follicle was 19mm in diameter in the presence of three or more follicles. We therefore concluded that embryo transfer could be done on days 2 or 3 according to the convenience of the patient and the medical team, without fear of comprising the clinical pregnancy rate. 230

241 The results of this study are reassuring to both medical staff and patients since it allows for more flexibility in scheduling the day of transfer. Even so, it should be noted that that the optimium number of embryos transferred per patient during the study period was somewhat different than what is current practice today. The effects of day 2 versus 3 might be substantial if only 1 2 embryos were replaced, as opposed to the higher numbers transferred during this trial. Even so, it should be noted that even though single embryo transfer has become a common practice in Europe, this policy is only beginning to become accepted in countries where IVF expenses are not covered by government or medical insurance. In these regions, the primary target is a clinical pregnancy and subsequent live birth, regardless the number of embryos transferred to accomplish this goal. Chapter 3 addresses the second question: Does removal of cervical mucus prior to embryo transfer improve the results? To answer this question, we conducted a prospective controlled trial comparing the outcome of cervical mucus removal prior to embryo transfer to no aspiration. We included a total of 286 women undergoing embryo transfers that were divided into two groups according to whether the cervical mucus was scheduled to be aspirated (group A) or not (group B). In addition, all transfers were performed in a similar manner by one operator (Dr. Mamdouh Eskandar). The clinical pregnancy rate was found to be significantly higher in group (A) (63/143) than group (B) (38/143) (P =0.003; OR = 2.18, 95% CI = ), although there was a higher frequency of easy transfers in group (B) than group (A) (OR = 3.00, 95% CI = ). This demonstrated that even though embryo transfers were easier to perform when the cervical mucus was left in place, aspiration resulted in an increased chance of clinical pregnancy, which is of more clinical importance. Even so, randomized controlled studies with adequate sample sizes should be performed to confirm these findings, and to determine if there is a true relationship between the presence of a difficult embryo transfer and the presence of retained embryos. 231

242 Chapter 4 addresses the third question: What is the effect of passive uterine straightening during embryo transfer on pregnancy rate? The first group to advocate the full bladder in order to assist straightening the cervico-uterine angle were Sundstrom and colleagues (1984). They described the use of a full bladder using both ultrasoundguided and standard clinical touch methods for embryo catheter placement. To answer the clinical question of whether passive uterine straightening during embryo transfer affects pregnancy rates, a systematic review and meta-analysis of prospective, randomised, controlled trials, comparing embryo transfer with a full versus empty bladder was conducted. Electronic (e.g. PubMed, EMBASE, Cochrane Library) and hand searches were performed to locate trials. Primary outcomes were live-birth, ongoing and clinical pregnancy rates. Secondary outcomes were rates of implantation, miscarriage, multiple and ectopic pregnancies, and retained embryos. Also, the ease of transfer, need for instrumental assistance, and presence of blood on the catheter tip were evaluated. Three studies (1109 ET cycles in women) were included. Meta-analysis was conducted with the Mantel-Haenszel method, utilizing the fixedeffect model. For live birth rate no data were available. Pooling of the results revealed a significantly higher chance of an ongoing pregnancy (OR = 1.44, 95% CI = 1.04 to 2.04) and clinical pregnancy (OR = 1.55, 95% CI = 1.16 to 2.08) with a full bladder. In addition, there was a significantly greater incidence of difficulty, or need for instrumental assistance, with an empty bladder. For the other secondary outcomes no differences were found. This current meta-analysis could detect an absolute difference of 7.5% with 80% power in a two-tailed analysis (assuming a CPR rate of 30% with a full bladder and a significance level of 0.05). The absolute difference between full and empty bladder was 7.8%, therefore validating our results. However, patient counseling is important, since most patients will need to micturate shortly after the transfer procedure. This action may be presumed to negatively affect the outcome of the IVF procedure. Therefore, proper counseling must be undertaken early in the cycle in order to decrease any anxiety over early mobilisation or micturation following the transfer. It was concluded that the evidence in the literature supports the use of bladder filling prior to 232

243 embryo transfer. Whether this is also applicable to patients with a retroverted uterus, or for patients undergoing vaginal ultrasound-guided embryo transfer, we do not know since there are no prospective clinical trials comparing empty to full bladders in patients undergoing embryo transfer with these two pre-requisites Chapter 5 addresses the fourth question: Is there is a difference in clinical outcomes when two soft embryo transfer catheters are compared for embryo transfer? To answer this question, we performed a two-center prospective randomized controlled clinical trial, in order to compare the performance of two soft embryo transfer catheters. Four hundred consecutive women aged less than 40 years of age, and undergoing ET with two fresh embryos were included. The women were randomly allocated to undergo ET with one of the two catheters, with possible catheter change in case of insertion difficulties. Main outcome measures were clinical pregnancy rate and live birth rate. No significant difference in the clinical pregnancy rate (O.R = 0.99; 95%CI = ) and live-birth rates (O.R = 1.09; 95%CI = ) was found between the two catheters. It was therefore concluded from this trial that the pregnancy and live-birth rates were not significantly different with the two tested soft catheters. Chapter 6 addresses the fifth question: Is there a difference in clinical outcomes between firm and soft embryo transfer catheters when embryo transfer is performed under ultrasound guidance? To answer this question, we performed a detailed chart review of all IVF cycles performed in our center from 2004 to 2006 (n = 666). The primary outcome of this study was the clinical pregnancy rate per woman. Secondary outcomes were the incidence of detecting indirect signs of cervical and/or uterine injury (e.g., blood, mucus) on the catheter tip and the incidence of difficult transfers with the respective catheters. 233

244 Details on patient demographics, cycle characteristics, and outcomes were extracted and compared for a semifirm catheter compared with soft catheters. The results of the study showed no significant difference between the clinical pregnancy rates between the individual catheters compared. Also, no significant differences were found with regard to the secondary outcomes. We concluded that under ultrasound guidance, the degree of firmness of the embryo transfer catheter may not be a critical variable. Chapter 7 addresses the sixth question Are soft embryo transfer catheters more favourable than firm catheters during embryo transfer? Following up on the previous question, we decided to systematically review the evidence provided in the literature so that we may evaluate a single variable in relation to a successful transfer, the firmness of the embryo transfer catheter. Electronic (e.g. PubMed, EMBASE, Cochrane Library) and hand searches of the literature revealed a total of ten trials, (4141 ET in women) comparing soft versus firm embryo catheters and were included. The primary outcome measures used for this systematic review were implantation rate (IR), clinical pregnancy rate (CPR) and ongoing/take-home baby rate. Of these outcomes, the most important primary outcome was considered to be the ongoing/take-home baby rate. The secondary outcomes were ease of transfer (catheter failure rate) and simultaneous occurrence of traumatic events (e.g. use of a tenaculum, stylette, sounding, and/or dilatation). In addition, the presence of blood, mucus and/or retained embryos on the tip of the catheter was evaluated. Meta-analysis was conducted using the Mantel Haenszel method (fixedeffect model). Pooling of the results demonstrated a statistically significantly increased chance of clinical pregnancy following embryo transfer using the soft (643/2109) versus firm (488/2032) catheters [P=0.01; odds ratio (OR) = 1.39, 95% confidence interval (CI) = ]. When only the truly RCT were analysed, the results were again still in favour of using the soft embryo transfer catheters [soft (432/1403) versus firm (330/1402)], but with a greater significance (P < ; OR = 1.49, 95% CI = ). 234

245 Softer catheters may produce better results by decreasing the trauma to the endometrium. The softer the materials used, the lesser the chance for damage to the endometrium and the lesser the chance for uterine contractions. The soft transfer catheters follow the natural curvature of the uterine cavity better than the firmer catheters, possibly reducing the risk of burrowing into the posterior endometrium in the anteflexed uterus, or stimulating uterine junctional zone contractions. We concluded from this study that using soft embryo transfer catheters for embryo transfer results in a significantly higher clinical pregnancy rate as compared to firm catheters. Chapter 8 addresses the seventh question: What is the effect of using air to bracket embryos in the transfer catheter? To answer this question, a systematic review and meta-analysis of randomized trials was performed to determine the beneficial or detrimental effect of using air bubbles to bracket the embryo-containing medium during embryo transfer. Electronic (e.g. PubMed, EMBASE, Cochrane Library) and hand searches of the literature revealed three studies of which two were included (298 women). The primary outcome measures were live birth, ongoing and clinical pregnancy rates. Of these outcomes, the most important primary outcome was considered to be the live-birth rate. The secondary outcome measures were the rates of implantation, miscarriage, multiple and ectopic pregnancies and retained embryos. Meta-analysis was conducted using the Mantel Haenszel method (fixed-effect model). The results demonstrated that there were no significant differences between the two methods with regards to live birth (OR = 1.34; 95% CI = ), ongoing pregnancy (OR = 1.34; 95% CI = ) and clinical pregnancy (OR = 1.13; 95% CI = ) rates. For the secondary outcomes, there were no significant differences between the two groups. It was therefore concluded that there is insufficient evidence to suggest that the fluid-only method is superior to the use of air brackets during embryo loading. More well-designed and powered randomized trials are needed to determine any possible benefit to either method. 235

246 Chapter 9 addresses the eighth question: What is the best site in the uterine cavity for embryo replacement? To answer this question, and in order to determine the best site for embryo deposition during embryo transfer, a meta-analysis of randomized trials comparing different uterine deposition sites was undertaken. Electronic (e.g. PubMed, EMBASE, Cochrane Library, LILACS) and hand searches were performed to locate trials. Outcomes measures were the live-birth, ongoing pregnancy and clinical pregnancy rates. Assessments of the endometrial cavity length (ECL) and the distance from the fundus to the tip of the catheter (DTC) were utilized. Out of the identified six studies, three trials (2170 ET in women) were included. Meta-analysis was conducted with the Mantel-Haenszel method, utilizing the fixed-effects model. The live-birth rate and ongoing pregnancy rates showed an increasing trend when transfers were performed to the lower half of the uterine cavity. For the DTC, all rates were significantly higher for the ~20 mm versus ~10 mm distance from the uterine fundus, supporting the results of the ECL analysis. The results of this systematic review showed that there is limited evidence of the superiority of lower cavity transfers (e.g. ~20 mm) compared with the traditional high cavity (e.g. ~10 mm) transfers. More well-designed and powered randomized trials are needed to confirm this conclusion. Chapter 10 addresses the ninth question: Does ultrasound-guidance during embryo transfer improve the outcomes when performed by a single operator? To answer this question, we performed a prospective, single-operator, randomized, controlled trial comparing ultrasound to clinical touch methods of embryo catheter guidance. The study randomized 378 women to transcervical, intrauterine embryo transfer with (183 women) or without ultrasound guidance (190 women). In addition, all the transfers were performed in a similar manner by one operator (Dr. Mamdouh Eskandar). Primary outcomes were the live-birth/ongoing pregnancy and clinical pregnancy rates per randomized woman. Of these outcomes, the most important primary outcome was considered to be the live-birth/ongoing pregnancy rate. Secondary outcomes were the 236

247 incidences of difficult transfers, blood and/or mucus on the catheter tip, spontaneous miscarriages and ectopic pregnancies. The results of this trial demonstrated that the live-birth/ongoing pregnancy rate was significantly higher in the US-ET group [68/183 (40.98%)] than the CT-ET group [50/190 (28.42%)] (O.R= 1.66, 95%CI= ). In addition, there was a significantly higher number of clinical pregnancies in the US-ET group [75/183 (40.98%)] than the CT-ET group [54/190 (28.42%)] (O.R= 1.75, 95%CI= ). Secondary outcomes were not significantly different between the two groups. The exact mechanism whereby ultrasound-guidance improves the outcomes following embryo transfer remains unclear. Confirming the position of the catheter tip within the uterine cavity, properly determining the site of embryo deposition, and decreasing the frequency of difficult embryo transfers or increasing the clinical appreciation of the pelvic anatomy have all been proposed as possible factors. Whatever the mechanism, from this trial it seems that ultrasoundguided embryo transfer significantly increases the chance of ongoing/live-birth and clinical pregnancy rates compared to the clinical touch method when performed by a single operator. Chapter 11 addresses the tenth question: Is the probability of pregnancy and live birth improved with ultrasoundguidance than with clinical touch alone? To answer this question, we performed a systematic review of prospective, randomized, controlled trials comparing ultrasound with clinical touch methods of embryo catheter guidance. Meticulous electronic (e.g., PubMed, EMBASE, CENTRAL) and hand searches were performed to locate trials. Primary outcome measures were the livebirth, ongoing pregnancy, and clinical pregnancy rates. Of these outcomes, the most important primary outcome was considered to be the live-birth rate. Secondary outcome measures were the implantation, multiple pregnancies, and miscarriage rates. In addition, the incidences of ectopic pregnancies and difficult transfers were evaluated. Twentyfive studies were retrieved, of which five were excluded, leaving twenty studies (5,968 ET cycles in women) for analysis. Pooling of the data was 237

248 conducted by using the Mantel-Haenszel method (fixed-effect model). The results revealed that there was a significantly increased chance of a live birth (odds ratio [OR] = 1.78, 95% confidence interval [CI] = 1.19 to 2.67), ongoing pregnancy (OR = 1.51, 95% CI = 1.31 to 1.74), clinical pregnancy (OR = 1.50, 95% CI = 1.34 to 1.67), embryo implantation (OR = 1.35, 95% CI = 1.22 to 1.50), and easy transfer rates after ultrasound guidance (OR = 0.68, 95% CI = 0.58 to 0.81). There was no difference in multiple pregnancy, ectopic pregnancy, or miscarriage rates. We therefore concluded that ultrasound-guided ET significantly increases the chance of live birth and ongoing and clinical pregnancy rates compared with the clinical touch methodand we hope that this evidence will be easily and quickly translated from the medical literature to everyday clinical practice. However, clinicians should perform in-house cost-effective analyses to determine the best combination of needed equipment, personnel, and catheter choice to provide the most efficient and cost-effective treatment options for their patients. Chapter 12 addresses the eleventh question: What is the effect of blood found on the tips of the post-transfers on clinical outcomes? To answer this question, we performed a detailed chart review of 943 women, aged years old, who underwent 1,122 ETs between January 1, 2005 and December 31, A semi quantitative system for grading and recording the difficulty of transfer, presence of retained embryos, mucus and blood found inside and outside the transfer catheter after ET was used to determine the correlation of each factor with the clinical pregnancy rate. The results demonstrated that an easy transfer (O.R = 4.52, 95% CI = 2.00 to 10.20) and absence of blood on the outside (O.R = 2.20, 95% CI = 1.38 to 3.52; P = 0.001) or inside O.R = 1.63, 95% CI = 0.94 to 2.82) of the transfer catheter after ET was associated with increased clinical pregnancy rates. In addition, there was a correlation between difficult transfers and the presence of blood. In contrast, presence of retained embryos or mucus did not significantly affect the outcomes. It was therefore concluded that the 238

249 relationship between difficult embryo transfer and the presence of blood on the outside or inside of the transfer catheter decreases the chance for a clinical pregnancy rate. Therefore every effort should be made to minimalize difficulty, and hence bleeding, during embryo transfer. Chapter 13 addresses the twelfth question: Does the use of Propofol anaesthesia improve the clinical outcomes of difficult embryo transfers? To answer this question, we performed a detailed patient chart review of 198 women undergoing embryo transfer with difficultly during the procedure from January December We included 99 women requiring general anesthesia (Group I), and a matching group of women with difficult ET, without anesthesia (Group II) as a control group. Primary outcome measures were embryo implantation rate and clinical pregnancy rate. The results showed that there were no significant differences in implantation (Group I = 19.15%, Group II = 20.86%) or clinical pregnancy rates (Group I = 36.36%, Group II = 33.33%) (O.R = 1.14, 95% CI = 0.64 to 2.05; P = 0.77). Therefore it was concluded that the use of Propofol general anesthesia during difficult embryo transfer does not seem to affect the implantation and pregnancy rates. Therefore, Propofol anesthesia may offer clinicians a complementary measure while dealing with a difficult embryo transfer. Even so, prospective randomized trials are needed to confirm these findings. Conclusions 1. Passive uterine straightening and cervical mucus aspiration should be performed prior to embryo transfer as part of the standard protocol of embryo transfer. 2. Soft catheters under ultrasound-guidance should be used to assist in the accurate placement of the embryos within the uterus at an average of ~20 mm from the uterine fundus. 3. It is of paramount importance to decrease the rates of difficult transfers by any method possible. 239

250 Appendix 1: Summary of the intellectual contributions of the PhD candidate to the conception, design, statistical analysis and writing of the individual studies in accordance with the Guidelines on Good Publication Practice published by the Committee on Publication Ethics (COPE). Chapter Contribution Timing of the embryo transfer Chapter 2 Aboulghar MM, Aboulghar MA, Mansour RT, Serour GI, Amin YM, Abou-Setta AM. Pregnancy rate is not improved by delaying embryo transfer from days 2 to 3. Eur J Obstet Gynecol Reprod Biol. 2003;107: Responsible for data entry, statistically analyzing and interpreting the outcomes. The embryo transfer procedure Chapter 3 Eskandar M, Abou- Setta AM, El-Amin M, Almushait MA, Sobande AA. Removal of cervical mucus prior to embryo transfer improves pregnancy rates in women undergoing assisted reproduction. Reprod Biomed Online. 2007;14: Responsible for the first draft of the text of final manuscript, statistically analyzing and interpreting the outcomes. 240

251 Chapter Contribution Chapter 4 Abou-Setta AM. Effect of passive uterine straightening during embryo transfer: a systematic review and metaanalysis. Acta Obstet Gynaecol Scand. 2007;86: Responsible for the draft of the text of both the protocol and review, for designing the search strategy, searching for studies, evaluating the methodological quality and extracting data from the included studies, and finally performing the meta-analysis. Chapter 5 Saldeen P, Abou-Setta AM, Bergh T, Sundström P, Holte J. A prospective randomized controlled trial comparing two embryo transfer catheters in an ART program. Fertil Steril. 2007; November 28 [Epub ahead of print]. Responsible for the first draft of the text of final manuscript, assisted in statistically analyzing and interpreting the outcomes and performed a meta-analysis of similar trials. Chapter 6 Aboulfotouh I, Abou- Setta AM, Khattab S, Mohsen IA, Askalani A. Firm versus Soft Embryo Transfer Catheters under Ultrasound Guidance: Does Catheter Choice Really Influence the Pregnancy Rates. Fertil Steril. 2007; Jul 17 [Epub ahead of print]. Responsible for the first draft of the protocol and text of final manuscript, contributing to the study concept and design, statistically analyzing and interpreting the outcomes. 241

252 Chapter Contribution Chapter 7 Abou-Setta AM, Al- Inany HG, Mansour RT, Serour GI, Aboulghar MA. Soft versus firm embryo transfer catheters for assisted reproduction: a systematic review and metaanalysis. Hum Reprod : Responsible for the first draft of the text of both the protocol and review, assisted in designing the search strategy, searching for studies, evaluating the methodological quality and extracting data from the included studies, and finally performing the meta-analysis. Chapter 8 Abou-Setta AM. Airfluid versus fluid-only models of embryo catheter loading: a systematic review and metaanalysis. Reprod Biomed Online. 2007; 14: Responsible for the draft of the text of both the protocol and review, for designing the search strategy, searching for studies, evaluating the methodological quality and extracting data from the included studies, and finally performing the meta-analysis. Chapter 9 Abou-Setta AM. What is the best site for embryo deposition? A systemic review and meta-analysis. Reprod Biomed Online. 2007;14: Responsible for the draft of the text of both the protocol and review, for designing the search strategy, searching for studies, evaluating the methodological quality and extracting data from the included studies, and finally performing the meta-analysis. 242

253 Chapter Contribution Chapter 10 Eskandar MA, Abou- Setta AM, Almushait MA, El-Amin M, Mohmad SE. Ultrasoundguidance during embryo transfer: a prospective, single operator, randomized, controlled trial. Fertil Steril. (In-Press). Responsible for the first draft of the text of final manuscript, statistically analyzing and interpreting the outcomes. Chapter 11 Abou-Setta AM, Mansour RT, Al-Inany HG, Aboulghar M, Serour GI, Aboulghar MA. Among women undergoing embryo transfer, is the probability of pregnancy and live birth improved with ultrasound-guidance than with clinical touch alone? a systemic review and meta-analysis of prospective randomized trials. Fertil Steril Jun 6; [Epub ahead of print]. Responsible for the first draft of the text of both the protocol and review, assisted in designing the search strategy, searching for studies, evaluating the methodological quality and extracting data from the included studies, and finally performing the meta-analysis. Chapter 12 Abou-Setta AM, Sharkawy S, Shahed M, Amer M. Difficult embryo transfers and the presence of blood on the embryo transfer catheter negatively affect clinical pregnancy rates. (Submitted) Responsible for the first draft of the text of both the protocol and final manuscript, assisted in designing the study strategy, extracting data and finally statistically analyzing and interpreting the outcomes. 243

254 Chapter Contribution Chapter 13 Abou-Setta AM, Mansour RT, Al-Inany HG, Serour GI, Aboulghar MA. Difficult Embryo Transfer: The Impact of Propofol Anesthesia. Middle East Fertility Society Journal. 2007;(In-press). Responsible for the first draft of the text of both the protocol and final manuscript, assisted in designing the study strategy, extracting data and finally statistically analyzing and interpreting the outcomes. 244

255 Appendix 2: Electronic search strategy for locating randomized controlled trials on Medline and Embase: MEDLINE may be searched using the following subject headings and keywords: 1. randomised controlled trial.pt. 2. controlled clinical trial.pt. 3. randomised controlled trials/ 4. random allocation/ 5. double-blind method/ 6. single-blind method/ 7. or/ clinical trial.pt. 9. exp clinical trials/ 10. (clin$ adj25 trial$).tw. 11. ((singl$ or doubl$ or treb$ or tripl$) adj25 (blind$ or mask$)).tw. 12. placebos/ 13. placebo$.tw. 14. random$.tw. 15. research design/ 16. or/ animal/ not (human/ and animal/) or not Embryo transfer/ 21. (embryo$ adj5 transfer).tw. 22. or/ and

256 The EMBASE database may be searched using the following subject headings and keywords: 1. Controlled study/ or Randomised Controlled Trial/ 2. Double Blind Procedure/ 3. Single Blind Procedure/ 4. Crossover Procedure/ 5. Drug Comparison/ 6. Placebo/ 7. Random$.tw. 8. latin square.tw. 9. crossover.tw. 10. cross-over.tw. 11. placebo$.tw. 12. ((doubl$ or singl$ or tripl$ or trebl$) adj5 (blind$ or mask$)).tw. 13. (comparativ$ adj5 trial$).tw. 14. (clinical adj5 trial$).tw. 15. animal/ not (human/ and animal) 16. or/ not Embryo transfer/ 19. (embryo$ adj5 transfer).tw. 20. or/ and

257 Samenvatting Het doel van dit proeschrift was het beste bewijs te vinden dat beschikbaar is over de procedures die worden gebruikt om de embryo transfer techniek te optimaliseren. Van vele aspecten van de embryo transfer is aangetoond dat zij de klinische uitkomst beinvloeden en op dit moment zou de embryo transfer wel eens de bepalende factor kunnen zijn bij de succeskansen van IVF/ICSI. Om vast te stellen welke technieken werkelijk samenhangen met betere uitkomsten, hebben wij enkele klinische trials uitgevoerd en systematische reviews van de literatuur verricht. In concreto was het doel van dit proefschrift zoals aangegeven in hoofdstuk I de volgende twaalf specifieke klinische vragen te beantwoorden: Bovendien zijn twee bijlagen toegevoegd. In de eerste bijlage worden de bijdragen die de auteur heeft geleverd aan ieder hoofdstuk vermeld overeenkomstig de Guidelines on Good Publication Practice zoals gepubliceerd door the Committee on Publication Ethics (COPE). In de tweede bijlage wordt een brede zoek strategie beschreven die door de lezer kan worden aangepast om het mogelijk te maken zelf een gedegen zoekactie te verrichten op het gebied van embryo transfer. Hoofdstuk 2 behandelt de eerste vraag Verbetert uitstel van de embryo transfer van dag twee naar dag drie de kans op zwangerschap? Om deze vraag te beantwoorden, verrichtten wij een prospectieve quasi-gerandomiseerde klinische trial waarin de uitkomst van embryo transfer op dag twee versus dag drie na de eicelpunctie werd vergeleken. Wij includeerden 927 consecutieve embryo transfers (626 transfers op dag 2 en 301 op dag 3) na IVF/ICSI. De resultaten lieten zien dat er geen significant verschil was in de kans op zwangerschap tussen een embryo transfer op dag 2 (319/626) en dag 3 (152/ 301) (O.R = 1.02, 95% BI = 0.77 tot 1.34). Hoewel onze studie niet gerandomiseerd was, is selectie bias waarschijnlijk minimaal omdat patienten die hun eicelpunctie op woensdag of donderdag hadden een embryo transfer op dag 3 kregen en degenen die hun eicelpunctie op de andere dagen van de week hadden een embryotransfer op dag 2 247

258 kregen. Criteria voor de hcg injectie waren hetzelfde gedurende de hele studie, nl bij een dominante follikel van 19mm in diameter in aanwezigheid van drie of meer follikels. De conclusie is dan ook dat embryo transfer zowel op dag 2 als 3 kan worden verricht afhankelijk van de omstandigheden en het gemak voor zowel de patient als het medisch team, zonder de kans op zwangerschap in gevaar te brengen. De resultaten van deze studie zijn geruststellend voor zowel de medische staf als patienten omdat ze flexibiliteit bij het afspreken van de dag van terugplaatsing mogelijk maken. Het aantal embryos dat per patient werd teruggeplaatst tijdens de onderzoeksperiode verschilde echter van wat tegenwoordig gangbaar is. Het verschil tussen dag 2 versus 3 zou substantieel kunnen zijn als slechts 1 2 embryos worden teruggeplaatst, in tegenstelling tot de hogere aantallen tijdens deze trial. Desalniettemin moet worden vastgesteld dat, hoewel terugplaatsing van 1 embryo gebruikelijk is geworden in Europa, dit beleid beslist nog niet standaard is in landen waar de kosten van IVF niet worden gedekt door de staat of verzekering. In deze landen is het primaire doel een zwangerschap met daarna geboorte van een levend kind, ongeacht het aantal embryos dat wordt teruggeplaatst om dit doel te bereiken. Hoofdstuk 3 heeft betrekking op de tweede vraag: Leidt het verwijderen van het cervix slijm voor de embryo transfer tot betere resultaten? Om deze vraag te beantwoorden, verrichtten wij een prospectieve klinische trial waarin de uitkomst van het verwijderen van het cervix slijm voorafgaand aan de embryo transfer werd vergeleken met het in situ laten van het slijm. Wij includeerden 286 vrouwen die een embryo transfer ondergingen. Zij werden verdeeld in twee groepen, namelijk de groep waarin het cervix slijm werd verwijderd (groep A) en de groep waarin dit niet gebeurde (groep B). De kans op een klinische zwangerschap was significant hoger in groep (A) (63/143) dan in groep (B) (38/143) (P =0.003; OR = 2.18, 95% BI = 1.32 tot 3.58), hoewel er meer gemakkelijke embryo transfers waren in groep (B) dan in groep (A) (OR = 3.00, 95% BI = 1.05 tot 8.55). Dit toont aan dat er na verwijdering van het cervix slijm een verhoogde kans op een klinische 248

259 zwangerschap was, hoewel embryo transfers gemakkelijker uit te voeren waren als het cervix slijm niet werd verwijderd. Desalniettemin zouden gerandomiseerde studies van voldoende omvang moeten worden verricht om deze bevindingen te bevestigen en te bepalen of er een echte relatie is tussen een moeilijke embryo transfer en het achterblijven van embryos in de catheter. Hoofdstuk 4 is gericht op de derde vraag: Wat is het effect van het strekken van de uterus door blaasvulling gedurende embryo transfer op de zwangerschapskans? De eerste groep die een volle blaas adviseerden om te helpen bij het strekken van de hoek tussen cervix en uterus waren Sundstrom et al (1984). Zij beschreven het gebruik van een volle blaas gebruikmakend van zowel echo geleide als standaard clinical touch methodes voor het plaatsen van de embryo catheter. Om deze vraag te beantwoorden, werd een systematische review en meta-analyse van gerandomiseerde, gecontroleerde trials verricht, waarbij embryo transfer met een volle versus een lege blaas werd vergeleken. Om de trials te vinden werd zowel een electronische ( PubMed, EMBASE, Cochrane Library) als een handmatige zoekactie uitgevoerd. Primaire uitkomsten waren levend geborenen, doorgaande zwangerschap en klinische zwangerschap. Secundaire uitkomsten waren implantatie graad, miskraam, meerling- en extra uteriene zwangerschappen en achtergebleven embryos. Ook werden het gemak van de transfer, de behoefte aan aanvullende instrumentatie en de aanwezigheid van bloed aan de catheter tip geevalueerd. Drie studies (1109 ET cycli) werden geincludeerd. Meta-analysis werd uitgevoerd met de Mantel-Haenszel methode, gebruikmakend van het fixed-effect model. Voor levend geborenen waren geen gegevens beschikbaar. Samenvoegen van de resultaten toonde een significant hogere kans op een doorgaande zwangerschap (OR = 1.44, 95% BI = 1.04 tot 2.04) en klinische zwangerschap (OR = 1.55, 95% BI = 1.16 tot 2.08) met een volle blaas. Bovendien was er een significant hogere incidentie van moeilijke terugplaatsingen of behoefte aan instrumentele assistentie bij een lege blaas. Bij de andere secundaire uitkomsten werden geen verschillen gevonden. In deze meta-analyse kon met een tweezijdige 249

260 analyse een absoluut verschil van 7.5% worden aangetoond bij een power van 80% (bij de aanname van een kans op een klinische zwangerschap van 30% bij een volle blaas en een significantie niveau van 0.05). Het absolute verschil tussen volle en lege blaas was 7.8%, waarmee onze resultaten betrouwbaar zijn. Voorlichting aan de patient is echter belangrijk, omdat de meeste patienten kort na de transfer procedure zullen moeten plassen. Hiervan zou de patient kunnen denken dat het de uitkomst van de IVF procedure negatief zou kunnen beinvloeden. Daarom moet goede voorlichting vroeg in de cyclus worden gegeven om eventuele angst over vroege mobilisatie of plassen na de transfer weg te nemen. De conclusie van deze meta-analyse was dat de gegevens in de literatuur het nut van blaasvulling voorafgaand aan de embryo transfer aantonen. Of dit ook geldt voor patienten met een uterus in retroversie, of voor patienten die een embryo terugplaatsing ondergaan onder controle van een transvaginale echoscopie, weten wij niet omdat er geen prospectieve klinische trials zijn die lege met volle blaas vergelijken in deze twee situaties. Hoofdstuk 5 gaat over de vierde vraag: Is er verschil in klinische uitkomsten bij het gebruik van twee verschillende zachte embryo transfer catheters? Om deze vraag te beantwoorden, verrichtten wij een gerandomiseerde klinische trial in twee centra. Vierhonderd vrouwen jonger dan 40 jaar en die een embryotransfer met twee verse embryos ondergingen, werden achtereenvolgens geincludeerd. De vrouwen werden na randomisatie toegewezen om een embryotransfer te ondergaan met 1 van de 2 catheters, met eventuele wisseling van de catheter in geval de catheter moeilijk in te brengen was. De belangrijkste uitkomstmaten waren klinische zwangerschap en levend geborenen. Er werden geen significante verschillen in de kans op klinische zwangerschap (O.R = 0.99; 95%BI = 0.66 tot 1.47) en levend geborenen (O.R = 1.09; 95%BI = 0.72 tot 1.65) gevonden tussen de twee geteste catheters. 250

261 Hoofdstuk 6 concentreert zich op de vijfde vraag: Is er een verschil in klinische uitkomsten tussen harde en zachte embryo transfer catheters bij terugplaatsing onder echoscopische controle? Om deze vraag te beantwoorden, onderzochten wij in detail de gegevens van alle IVF cycli die waren verricht in ons centrum van 2004 tot 2006 (n = 666). De primaire uitkomst van deze studie was klinische zwangerschap per vrouw. Secundaire uitkomsten waren indirecte tekenen van cervicale en/of uteriene beschadiging, zoals bloed en slijm op de catheter tip en het aantal moelijke transfers met beide catheters. Details over patient kenmerken, cyclus karakteristieken en uitkomsten werden vastgelegd en vergeleken voor beide types catheter. De resultaten van de studie toonden geen significant verschil aan tussen de catheters in kans op een klinische zwangerschap. Ook werden geen significante verschillen gevonden met betrekking tot de secundaire uitkomsten. Wij concludeerden dat onder echogeleiding, de buigzaamheid van de embryo transfer catheter mogelijk niet van doorslaggevende betekenis is bij de embryo transfer. Hoofdstuk 7 behandelt de zesde vraag: Zijn zachte embryo transfer catheters gunstiger dan stijve catheters bij embryo transfer? In aansluiting aan onze vorige vraag besloten wij de literatuur gegevens over de buigzaamheid van embryo transfer catheters systematisch te analyseren om zo de buigzaamheid als individuele variabele voor een successvolle transfer te evalueren. Een electronische (PubMed, EMBASE, Cochrane Library) en handmatige zoekactie van de literatuur leverde tien trials op met in het totaal 4141 embryo transfers waarin zachte versus stijve embryo catheters werden vergeleken. De primaire uitkomstmaten gebruikt in deze systematische review waren implantatie graad, klinische zwangerschap en doorgaande zwangerschap. De secundaire uitkomsten waren het gemak waamee de catheter kon worden ingebracht en het gebruik van traumatische ingrepen zoals een kogeltang, stylette of dilatatie. Bovendien werd de aanwezigheid van 251

262 bloed, slijm en/of achtergebleven embryos op de tip van de catheter geevalueerd. Meta-analysis werd verricht met de Mantel Haenszel methode (fixedeffect model). Samenvoegen van de resultaten leverde een statistisch significant toegenomen kans op een klinische zwangerschap na embryo transfer met de zachte (643/2109) versus de stijve (488/2032) catheters op [P=0.01; OR = 1.39, 95% BI = 1.08 tot 1.79]. Bij analyse van de echte gerandomiseerde trials waren de resultaten nog steeds in het voordeel van de zachte embryo transfer catheters [zacht (432/1403) versus stijf (330/1402)], maar nu met een grotere significantie (P < ; OR = 1.49, 95% BI = 1.26 tot 1.77). De betere resultaten van zachte catheters zouden kunnen komen doordat zij minder schade aanrichten aan het endometrium. Hoe zachter het gebruikte materiaal, hoe minder kans op schade aan het endometrium en hoe minder kans op contracties van de uterus. De zachte transfer catheters volgen de natuurlijke curvatuur van het cavum uteri beter dan de steviger catheters, waardoor het risico om onder het endometrium aan de achterzijde van een uterus in anteflexie te komen, of om uterus contracties te veroorzaken mogelijk minder is. Wij concludeerden uit deze studie dat gebruik van zachte embryo transfer catheters leidt tot een significant hogere kans op een klinische zwangerschap vergeleken met stijve catheters. Hoofdstuk 8 behandelt de zevende vraag: Wat is het effect van het gebruik van luchtsloten in de transfer catheter? Om deze vraag te beantwoorden, werd een systematische review en meta-analyse van gerandomiseerde trials uitgevoerd om het gunstige of schadelijke effect van het gebruik van luchtsloten te bestuderen. Een electronische (PubMed, EMBASE, Cochrane Library) en handmatige zoekactie van de literatuur leverde drie studies op, waarvan er twee werden geincludeerd (298 vrouwen). De primaire uitkomstmaten waren levend geborenen, doorgaande en klinische zwangerschappen. De secundaire uitkomstmaten waren implantatie, miskraam, meerlingen en extra uteriene zwagerschappen en embryos achtergebleven in de catheter. Meta-analyse werd verricht met de Mantel Haenszel methode (fixed-effect model). De resultaten lieten zien dat er geen significante 252

263 verschillen waren tussen de twee methoden met betrekking tot levend geborenen (OR = 1.34; 95% BI = 0.59 tot 3.07), doorgaande zwangerschap (OR = 1.34; 95% BI = 0.59 tot 3.07) en klinische zwangerschap (OR = 1.13; 95% BI = 0.70 tot 1.83). Ook voor de secundaire uitkomsten waren er geen significante verschillen tussen de twee groepen. De conclusie was dan ook dat er onvoldoende bewijs is om te stellen dat het gebruik van de ene techniek superieur is aan de andere. Er zijn meer goede gerandomiseerde trials van voldoende grootte nodig om te bepalen of er meerwaarde is van een van beide technieken. Hoofdstuk 9 richt zich op de achtste vraag: Wat is de beste plaats in het cavum uteri om het embryo te plaatsen? Om deze vraag te beantwoorden werd een meta-analyse uitgevoerd van gerandomiseerde trials die verschillende plekken om een embryo te plaatsen met elkaar vergelijken. Electronische (PubMed, EMBASE, Cochrane Library, LILACS) en handmatige zoekacties werden uitgevoerd om trials te vinden. Uitkomstmaten waren levend geborenen, doorgaande zwangerschap en klinische zwangerschap. De lengte van het endometrium en de afstand van de fundus uteri tot de tip van de catheter werden gebruikt als variabelen. Van de zes geidentificeerde studies, werden drie trials (2170 embryotransfers) geincludeerd. Metaanalyse werd uitgevoerd met de Mantel-Haenszel methode, gebruikmakend van het fixed-effects model. De kans op een levend geborene en de kans op een doorgaande zwangerschap toonden een positieve trend bij plaatsing in het onderste halve deel van het cavum uteri. Voor afstand van de fundus uteri tot de tip van de catheter, waren alle kansen significant hoger voor de afstand van ~20 mm versus de afstand van ~10 mm. De resultaten van deze systematische review toonden aan dat er maar een beperkte mate van bewijs is voor de superioriteit van transfers laag in het cavum(~20 mm) in vergelijking tot de traditionele (~10 mm) transfers hoog in het cavum. Er zijn meer goede gerandomiseerde trials van voldoende grootte nodig om deze conclusie te bevestigen. 253

264 Hoofdstuk 10 heeft betrekking op de negende vraag: Verbetert het toepassen van echoscopie tijdens de embryo transfer de IVF uitkomsten als alle ingrepen door een en dezelfde persoon worden uitgevoerd? Om deze vraag te beantwoorden, verrichtten wij een gerandomiseerde klinische trial, waarin de embryo transfer plaats vond met of zonder echoscopische controle. Alle ingrepen werden door dezelfde persoon verricht. In de studie werden 378 vrouwen door het lot toegewezen aan transcervicale, intrauteriene embryo transfer met (183 vrouwen) of zonder echoscopische contole (190 vrouwen). Primaire uitkomsten waren levend geborenen, doorgaande en klinische zwangerschap per gerandomiseerde vrouw. Secundaire uitkomsten waren het aantal moeilijke transfers, bloed en/of slijm aan de catheter tip, spontane miskraam en extra uteriene zwangerschap. De resultaten van deze trial toonden aan dat de kans op een levend geborene/doorgaande zwangerschap significant hoger was in de embryo transfer groep met echoscopische controle [68/183 (40.98%)] dan in de groep zonder [50/190 (28.42%)] (O.R= 1.66, 95%BI= 1.07 tot 2.57). Bovendien was er een significant hoger aantal klinische zwangerschappen in de embryo transfer groep met echoscopische controle [75/183 (40.98%)] dan in de groep zonder [54/190 (28.42%)] (O.R= 1.75, 95%BI= 1.14 tot 2.69). Secundaire uitkomsten waren niet significant verschillend tussen de twee groepen. Het exacte mechanisme waarmee echoscopische controle de uitkomsten verbetert na embryo transfer is onduidelijk. Als mogelijke factoren zijn genoemd: bevestiging van de positie van de catheter tip binnen het cavum uteri, het correct bepalen van de plaats van transfer van het embryo en het terugbrengen van het aantal moeilijke embryo transfers of een beter inzicht in de anatomie van het kleine bekken. Wat het mechanisme ook moge zijn, het lijkt erop dat embryo transfer onder echoscopische controle in de handen van een en dezelfde persoon, significant de kans op doorgaande zwangerschap/klinische zwangerschap en levend geborenen vergroot in vergelijking tot embryo transfer zonder echoscopische controle. 254

265 Hoofdstuk 11 gaat in op de tiende vraag: Verbetert embryo transfer onder echoscopische controle de kans op zwangerschap en levend geborenen in vergelijking tot embryo transfer zonder echoscopische controle? Om deze vraag te beantwoorden voerden wij een systematische review van gerandomiseerde klinische trials uit, die embryo transfer met echoscopische controle en embryo transfer zonder echoscopische controle vergeleken. Electronische (PubMed, EMBASE, Cochrane Library, LILACS) en handmatige zoekacties werden uitgevoerd om trials te localiseren. Primaire uitkomstmaten waren levend geborenen, doorgaande zwangerschap en klinische zwangerschap. Secundaire uitkomstmaten waren implantatie, meerling zwangerschappen en miskramen. Bovendien werden de incidentie van extra uteriene zwangerschappen en moeilijke transfers geevalueerd. Er werden 25 studies gevonden, waarvan vijf werden geexcludeerd. Twintig studies (5,968 cycli met embryo transfer) konden dus geanalyseerd worden. Samenvoegen van de gegevens werd gedaan met de Mantel-Haenszel methode (fixed-effect model). De resultaten toonden aan dat er een significant hogere kans was op een levend geborene (odds ratio [OR] = 1.78, 95% [BI] = 1.19 tot 2.67), doorgaande zwangerschap (OR = 1.51, 95% BI = 1.31 tot 1.74), klinische zwangerschap (OR = 1.50, 95% BI = 1.34 tot 1.67), embryo implantatie (OR = 1.35, 95% BI = 1.22 tot 1.50), en makkelijke transfers (OR = 0.68, 95% BI 0.58 tot 0.81) na echoscopische controle. Er waren geen verschillen in kans op een meerling zwangerschap, extra uteriene zwangerschap of miskraam. Wij concludeerden daarom dat embryo transfer onder echoscopische controle de kans op een levend geborene, doorgaande zwangerschap en klinische zwangerschap signicant verhoogt vergeleken met embryo transfer zonder echoscopische controle en wij hopen dat deze feiten snel zullen worden overgenomen in de dagelijkse klinische praktijk. Klinici zullen wel hun eigen kosten effectiviteits analyses moeten doen om de beste combinatie van noodzakelijke apparatuur, personeel, en keuze van catheter te bepalen om de meest efficiente en kost-effectieve behandelingsmogelijkheden voor hun patienten te bieden. 255

266 Hoofdstuk 12 heeft betrekking op de elfde vraag: Wat is het effect van bloed aan de tip van de transfer catheter op de klinische uitkomst? Om deze vraag te beantwoorden, onderzochten wij in detail de gegevens van 943 vrouwen in de leeftijd van 18 tot 39 jaar, die 1,122 embryo tranfers ondergingen tussen 1 januari 2005 en 31 december Een semi-kwantitatief systeem werd gebruikt om de moeilijkheid van de transfer, de aanwezigheid van achtergebleven embryos, slijm en bloed aan de binnen en buitenkant van de transfer catheter na embryo transfer vast te leggen en te scoren om de correlatie van iedere factor met de kans op een klinische zwangerschap te bepalen. De resultaten toonden aan dat een gemakkelijke transfer (O.R = 4.52, 95% BI = 2.00 tot 10.20) en afwezigheid van bloed aan de buitenkant (OR = 2.20, 95% BI = 1.38 tot 3.52; P = 0.001) of aan de binnenkant (O.R = 1.63, 95% BI = 0.94 tot 2.82) van de transfer catheter na embryo transfer was geassocieerd met een verhoogde kans op klinische zwangerschap. Bovendien was er een correlatie tussen moeilijke transfers en de aanwezigheid van bloed. De aanwezigheid van achtergebleven embryos of slijm daarentegen beinvloedde de uitkomsten niet significant. De conclusie was dan ook dat een moelijke embryo transfer en de aanwezigheid van bloed aan de buiten of de binnenkant van de transfer catheter de kans op een klinische zwangerschap vermindert. Daarom moeten alle inspanningen erop gericht zijn de transfer zo gemakkelijk en atraumatisch mogelijk te laten verlopen. Hoofdstuk 13 behandelt de twaalfde vraag: Verbeteren de klinische uitkomsten van moeilijke embryo transfers met Propofol anaesthesie? Om deze vraag te beantwoorden, onderzochten wij in detail de gegevens van 198 vrouwen die een moeilijke embryo transfer hadden ondergaan tussen januari 2000 en december Wij includeerden 99 vrouwen die algehele anesthesie nodig hadden (Groep I), en eenvergelijkbare groep vrouwen zonder anesthesie (Groep II) als controle groep. Primaire uitkomstmaten waren implantatie en klinische zwangerschap. De resultaten lieten zien dat er geen significante 256

267 verschillen waren in implantatie graad (Groep I = 19.15%, Groep II = 20.86%) of kans op klinische zwangerschap (Groep I = 36.36%, Groep II = 33.33%) (O.R = 1.14, 95% BI = 0.64 tot 2.05; P = 0.77). De conclusie was dan ook dat het gebruik van Propofol anesthesie gedurende moeilijke embryo transfer de implantatie graad en kans op zwangerschap niet lijkt te verbeteren. Uiteraard zijn gerandomiseerde trials nodig om deze bevindingen te bevestigen. Conclusies 1. Passief strekken van de uterus en aspiratie van het cervix slijm dienen te worden verricht voorafgaand aan de terugplaatsing van de embryos als onderdeel van het standaard IVF protocol. 2. Zachte catheters onder echoscopische geleide dienen te worden gebruikt om accurate plaatsing van de embryos in de uterus mogelijk te maken op ongeveer 20 mm van de fundus uteri. 3. Het is van het grootste belang het aantal moeilijke terugplaatsingen te verminderen met iedere denkbare methode. 257

268 (الملخص العربى) Arabic summary تهدف هذه الدراسة الى الوصول الى الدليل العلمى لافضل طرق ارجاع الاجنة فى برنامج أطفال الانابيب / الحقن المجهرى. ذلك ان هناك عوامل آثيرة قد تو ثر على فاعلية ارجاع الاجنة وبالتالى على نسبة حدوث الحمل. ومن اجل تحقيق ذلك تم اجراء عدة دراسات اآلينيكية ودراسات احصاي ية مجمعة للاجابة على عدة تساؤلات. یتناول الفصل الثانى من هذه الرسالة الاجابة على السو ال الاول وهو : هل یفضل ارجاع الاجنة بعد مرور یومين ام ثلاثة من عملية سحب البویضات وللاجابة على هذا السو ال تم عمل دراسة مقارنة شبه عشواي ية بين ستماي ة وستة وعشرون حالة ارجاع للاجنة فى اليوم الثانى من عملية سحب البویضات وبين ثلاثمي ة وواحد حالة ارجاع للاجنة فى اليوم الثالث. وقد اظهرت هذه الدراسة انه ليس هناك اختلاف فى نسبة حدوث الحمل بين ارجاع الاجنة فى اى من اليومين وخلصت الدراسة الى ان یمكن نقل الاجنة فى اليوم الثانى او الثالث دون خوف على نتاي ج الحمل. اما الفصل الثالث فتناول الاجابة عن اهمية ازالة مخاط عنق الرحم قبل ارجاع الاجنة الى الرحم وقد قمنا بعمل دراسة مقارنة بين ازالة المخاط من عدمه حيث تبين ان ازالة المخاط یو دى الى زیادة فى نسبة حدوث الحمل زیادة ملحوظة برغم ان عدم ازالة المخاط یو دى الى سهولة اآبر فى ارجاع الاجنة وخلصت الدراسة الى ضرورة عمل داسة عشواي ية للتا آد من هذه النتاي ج. وتناول الفصل الرابع دراسة احصاي ية مجمعة لمعرفة مدى فاي دة امتلاء المثانة واستقامة الرحم فى تحسين نسبة حدوث الحمل حيث تم تجميع نتاي ج ثلاث دراسات اآلينيكية شملت اآثر من الف حالة ارجاع اجنة حيث تبين ان ملا المثانة یو دى الى سهولة اآبر فى ارجاع الاجنة مصحوبة بزیادة فى حالات الحمل وخلصت الدراسة الى التنبيه على ملا المثانة فى حالات ارجاع الاجنة لاطفال الانابيب / الحقن المجهرى. اما افصول من الخامس الى السابع فتناول الاجابة عن اهمية نوع القسطرة التى تستخدم فى ارجاع الاجنة حيث تم اجراء دراسة مقارنة عشواي ية بين نوعين من انواع القساطر اللينة ضمت اربعماي ة حالة حيث لم یتبين هناك اى فرق ملحوظ بين النوعين. ثم تم عمل دراسة لاآثر من ستماي ة وستين حالة اطفال انابيب للمقارنة بين القساطر اللينة والقساطر الغيرمرنة عندما یتم ارجاع الاجنة تحت اشراف الموجات فوق الصوتية حيث لم یتبين ایضا وجود فرق ملحوظ فى حدوث الحمل بين النوعين من القساطر اذا تمت عملية الارجاع تحت اشراف الموجات فوق الصوتية. ثم قمنا باجراء دراسة احصاي ية مجمعة للمقارنة بين القساطر اللينة واللقساطر غير المرنة بدون استخدام الموجات فوق الصوتية وضمت هذه الدراسة عشر دراسات مقارنة عشواي ية بها اآثر من اربعة ا لاف حالة حيث تبين ان هناك فرق ملحوظ جدا فى نسبة حدوث الحمل التى تزید بدرجة قویة عند استخدام القساطر اللينة عنها فى القساطر غير المرنة وذلك لتا ثيرها البسيط على بطانة الرحم وخلصت الدراسة المجمعة الى ضرورة استخدام القساطر اللينة من اجل تحسين فرص الحمل فى برنامج اطفال الانابيب / الحقن المجهرى. اما الفصل الثامن فتناول الاجابة عن اهمية وجود فقاعة هواي ية فى قسطرة ارجاع الاجنة حيث تم اجراء دراسة احصاي ية مجمعة شملت دراستين عشواي يتين (حوالى ثلاثمي ة حالة) وتبين من هذه الدراسة انه لایوجد فرق ملحوظ فى نسب حدوث الحمل سواء تم وضع فقاعة الهواء فى القسطرة ام لا وخلصت الدراسة الى ضرورة القيام بتجارب اآلينيكية جدیدة حتى یكون التقييم افضل من الناحية العلمية. 258

269 وتناول الفصل التاسع الاجابة عن سو ال : ماهو افضل موضع فى الرحم لارجاع الاجنة وللاجابة على هذا السو ال تم ایضا عمل دراسة مجمعة شملت ثلاث دراسات عشواي ية آبيرة (ضمت الفين و مي ة و سبعين حالة) حيث تبين انه یلزم المزید والمزید من الدراسات العشواي ية حتى نكتشف المكان المفضل لارجاع الاجنة والذى یو دى الى تحسن نسبة الحمل. ولمعرفة مدى اهمية اجراء ارجاع الاجنة تحت اشراف الموجات فوق الصوتية تم عمل دراسة مقارنة عشواي ية حيث تم الارجاع بواسطة طبيب واحد فى آل الحالات حيث تبين ان استخدام الموجات فوق الصوتية یو دى الى تحسن فى نسبة حدوث الحمل وللتا آد من هذه النتاي ج قمنا بعمل دراسة احصاي ية مجمعة شملت عشرین دراسة عشواي ية ضمت قرابة ستة ا لاف حالة حيث اآدت النتاي ج ان استخدام الموجات فوق الصوتية یزید من فرص الحمل والانجاب. اما الفصل الثانى عشر فقد تناول تا ثير وجود دم او مخاط على قمة القسطرة بعد ارجاع الاجنة على نسبة حدوث الحمل وتمت الاجابة بدراسة مقارنة تيبن من خلالها ان وجود دم على القسطرة بعد ارجاع الاجنة یكون مصاحب بانخفاض نسبة حدوث الحمل اما وجود المخاط فلایو دى الى فرق ملحوظ فى نتاي ج الحمل. وتناول الفصل الثالث عشر تا ثير استخدام المخدر العام فى حالات ارجاع الاجنة الصعبة وهل یو دى الى تحسين نسب الحمل ام لا وتمت الاجابة بعمل دراسة مقارنة بين الحالات الى تم اخذ المخدر العام فيها والحالات التى لم تا خذ مخدر عام وتبين عدم وجود فرق ملحوظ فى حدوث الحمل فى الحالتين وخلصت الدراسة الى ان المخدر العام ليس له تا ثير سلبى على نسبة حدوث الحمل. 259

270 Acknowledgements I would like to begin by thanking Allah (God) (SWT) for everything in my life, and for helping me to complete this work to the best of my abilities. To Prof. Dr. Fulco van der Veen, whose vision and meticulous supervision have allowed this work to come to light, for this I owe my endless appreciation in recognition of his great efforts, patience and support. To Prof. Dr. Hesham Al-Inany, my mentor and friend, I owe a lifetime of gratitude for all his efforts over the past few years. If it were not for Dr. Al-Inany s support, my career would never have taken off the ground. To Prof. Dr. Mona Aboulghar, who first introduced me to the fields of assisted reproduction and reproductive research. To Dr. Ragaa Mansour, Prof. Dr. Mohamed Aboulghar, Prof. Dr. Gamal Serour and all the staff members, nurses, administration and patients of the Egyptian IVF-ET center, thank you for making me who I am today. Lastly, but definitely not least, I would like to thank my family, especially my wife, for their great support and help in accomplishing this work. It was only through their belief, trust, love and support could I have ever performed this wonderful task. For that I am eternally grateful. 260

271 Curriculum Vitae Education 2001 Present: Postgraduate Healthcare & Hospital Administration Diploma Program (in progress), Institute of Management, American University in Cairo (AUC), Cairo, Egypt : Medical School Diploma (M.B.,B.Ch.), Faculty of Medicine, Cairo University, Cairo, Egypt. Professional Experience Sept Mar. 2006: Managing Biostatistician, the Egyptian & Middle East IVF Registries. Sept Mar. 2006: Postgraduate Researcher/ Clinical Associate/ Manager, the Biostatistics & Information Technology (IT) Department, the Egyptian IVF- ET Center. Mar Feb. 2001: Medical House Officer, Transitional Year Program, Cairo University Hospitals. Peer Reviewed Journal Publications: 1. Aboulghar MM, Aboulghar MA, Mansour RT, Serour GI, Amin YM, Abou-Setta AM. Pregnancy rate is not improved by delaying embryo transfer from days 2 to 3. European Journal of Obstetrics, Gynaecology and Reproductive Biology. 2003;107(2): Mansour RT, Aboulghar MA, Serour GI, Al-Inany HG, Amin YM, Abou- Setta AM. The use of gonadotropin-releasing hormone antagonist in a flexible protocol: a pilot study. American Journal of Obstetrics and Gynecology. 2003;189(2): Al-Inany HG, Wasseef M, Aboulghar MA, Mansour RT, Serour GI, Abou-Setta AM. Embryo Transfer under Propofol Anaesthesia: The Impact On Implantation and Pregnancy Rate. Middle East Fertility Society Journal. 2003;8(3): Kamal A, Fahmy I, Mansour RT, Abou-Setta AM, Serour GI, Aboulghar MA. Selection of individual testicular tubules from biopsied testicular tissue with a stereomicroscope improves sperm retrieval rate. Journal of Andrology. 2004;25(1):

272 5. Kamal A, Fahmy I, Mansour RT, Abou-Setta AM, Serour GI, Aboulghar MA. Outcome of repeated testicular sperm extraction and ICSI in patients with non-obstructive azoospermia. Middle East Fertility Society Journal. 2004;9(1): Abou-Setta AM. Transmission risk of Hepatitis C virus via semen during assisted reproduction: How real is it? Human Reproduction. 2004;19(12): Aboulghar MM, Al-Inany HG, Aboulghar MA, Serour GI, Mansour RT, Amin YM, Abou-Setta AM. Three dimensional endometrial volume measurement in prediction of IVF/ICSI outcome. Middle East Fertility Society Journal. 2005;10(1): Mansour RT and Abou-Setta AM. Assisted Reproductive Technology in Egypt, 2001: results generated from the Egyptian IVF registry. Middle East Fertility Society Journal. 2005;10(2): Abou-Setta AM, Al-Inany HG, Mansour RT, Serour GI, Aboulghar MA. Soft versus firm embryo transfer catheters for assisted reproduction: a systematic review and meta-analysis. Human Reproduction. 2005;20(11): Mansour R, Aboulghar A, Serour G, Amin Y, Abou-Setta AM. Criteria of a successful coasting protocol for the prevention of severe ovarian hyperstimulation syndrome. Human Reproduction. 2005; 20(11): Al-Inany HG, Abou-Setta AM, Aboulghar MA, Mansour RT, Serour GI. hmg versus recfsh for ovulation induction in developing countries: a cost-effectiveness analysis based on the results of a recent metaanalysis. Reproductive BioMedicine Online. 2006; 12(2): Abou-Setta AM. Firm Embryo Transfer Catheters for Assisted Reproduction: a systematic review and meta-analysis. Reproductive BioMedicine Online. 2006;12(2): Abou-Setta AM, Mansour RT, Al-Inany HG, Aboulghar M, Kamal A, Serour GI, Aboulghar MA. Intrauterine insemination catheters for assisted reproduction: a systematic review and meta-analysis. Human Reproduction. 2006;21(8): Youssef G, Ali AM, Alaa N, Makin B, Waly M, Abou-Setta AM. N- acetyl-cysteine in anovulatory women: The impact of postcoital test. Middle East Fertility Society Journal. 2006;11(2):

273 15. Mansour RT and Abou-Setta AM. Assisted Reproductive Technology in the Middle East, 2001: results generated from the Middle East IVF registry. Middle East Fertility Society Journal. 2006;11(3): Yehia M, Abou-Setta AM, Bedaiwy MA. DEBATE: Embryo transfer: does ultrasound guidance make a difference? Middle East Fertility Society Journal. 2006;11(3): Abou-Setta AM. Air-fluid versus fluid-only models of embryo catheter loading: a systematic review and meta-analysis. Reproductive BioMedicine Online. 2007;14(1): Eskandar M, Abou-Setta AM, El-Amin M, Almushait MA, Sobande AA. Removal of cervical mucus prior to embryo transfer improves pregnancy rates in women undergoing assisted reproduction. Reproductive BioMedicine Online. 2007;14(3): Abou-Setta AM. Effect of passive uterine straightening during embryo transfer: a systematic review and meta-analysis. Acta Obstetricia et Gynecologica Scandinavica. 2007;86(5): Aboulghar MA, Mansour RT, Al-Inany HG, Abou-Setta AM, Mourad L, Aboulghar MM, Serour GI. Paternal age and outcome of intracytoplasmic sperm injection. Reproductive BioMedicine Online. 2007;14(5): Abou-Setta AM. What is the best site for embryo deposition? A systemic review and meta-analysis. Reproductive BioMedicine Online. 2007;14(5): Mansour RT, Al-Inany HG, Abou-Setta AM, Murad L, Serour GI, Aboulghar MA. The use of sperm suspension as an ultrasonically visible material to explain the mechanism of a modified embryo transfer technique. Human Reproduction. 2007;22(7): Abou-Setta AM, Mansour RT, Al-Inany HG, Serour GI, Aboulghar MA. Difficult Embryo Transfer: The Impact of Propofol Anesthesia. Middle East Fertility Society Journal. 2007;12(2). 24. Abou-Setta AM, Mansour RT, Al-Inany HG, Aboulghar MM, Aboulghar MA, Serour GI. Among women undergoing embryo transfer, is the probability of pregnancy and live birth improved with ultrasound guidance over clinical touch alone? A systemic review and metaanalysis of prospective randomized trials. Fertility and Sterility. 2007;88(2):

274 25. Al-Inany HG, Abou-Setta AM, Aboulghar MA, Mansour RT, Serour GI. Efficacy and safety of human menopausal gonadotrophins versus recombinant follicle stimulating hormone: a systematic literature review and meta-analysis. Reproductive BioMedicine Online. 2007; 16(1): Aboulfotouh I, Abou-Setta AM, Khattab S, Mohsen IA, Askalani A. Firm versus Soft Embryo Transfer Catheters under Ultrasound Guidance: Does Catheter Choice Really Influence the Pregnancy Rates. Fertility and Sterility. 2007; July 17 [Epub ahead of print]. 27.Saldeen P, Abou-Setta AM, Bergh T, Sundström P, Holte J. A prospective randomized controlled trial comparing two embryo transfer catheters in an ART program. Fertility and Sterility. 2007; November 28 [Epub ahead of print]. 28. Bekkering GE, Abou-Setta AM, Bradley M, Kleijnen J. The application of quantitative methods for identifying and exploring the presence of bias in systematic reviews: PDE 5-inhibitors for erectile dysfunction. International Journal of Impotence Research. 2007; December 6 [Epub ahead of print]. 29. Eskandar MA, Abou-Setta AM, Almushait MA, El-Amin M, Mohmad SE. Ultrasound-guidance during embryo transfer: a prospective, single operator, randomized, controlled trial. Fertility and Sterility. (In-press). Cochrane Systematic Reviews: 30. Al-Inany HG, Abou-Setta AM, Aboulghar MA. Gonadotrophinreleasing hormone antagonists for assisted conception. Cochrane Database of Systematic Reviews. 2006, Issue 3. Art. No.: CD DOI: / CD pub Abou-Setta AM, Al-Inany HG, Farquhar CM. Levonorgestrel-releasing intrauterine device (LNG-IUD) for symptomatic endometriosis following surgery. Cochrane Database of Systematic Reviews. 2006, Issue 4. Art. No.: CD DOI: / CD pub Brown JA, Buckingham K, Abou-Setta A, Buckett W. Ultrasound versus 'clinical touch' for catheter guidance during embryo transfer in women. Cochrane Database of Systematic Reviews. 2007, Issue 1. Art. No.: CD DOI: / CD pub2. 264

275