- Email: [email protected]
ICSI
RBMOnline - Vol 15. No 3. 2007 310-315 Reproductive BioMedicine Online; www.rbmonline.com/Article/2925 on web 16 July 2007
!RTICLE 2EMOVAL OF CERVICAL MUCUS EFFECT ON PREGNANCY RATES IN )6&)#3) "IANCA 6ISSCHERS WAS BORN IN .OVEMBER 3HE COMPLETED HER MEDICAL STUDIES IN AT THE #ATHOLIC 5NIVERSITY OF .IJMEGEN IN THE .ETHERLANDS !FTER GRADUATING SHE SPECIALIZED IN 4ROPICAL -EDICINE INCLUDING TRAINING IN /BSTETRICS AND 'YNAECOLOGY IN 'ELRE (OSPITALS !PELDOORN )N n SHE WORKED AS A 4ROPICAL $OCTOR IN -ASINDI $ISTRICT (OSPITAL -ASINDI 5GANDA !FTER HER RETURN TO THE .ETHERLANDS SHE TOOK UP A POST AT THE $EPARTMENT OF 'YNAECOLOGY AND /BSTETRICS IN 4WEE 3TEDEN (OSPITAL 4ILBURG 2ECENTLY SHE STARTED HER SPECIALIST TRAINING IN 'YNAECOLOGY AND /BSTETRICS AT THE 5NIVERSITY (OSPITAL OF 'HENT "ELGIUM
$R "IANCA 6ISSCHERS "!*4 6ISSCHERS 23'- "OTS -& 0EETERS "7* -OL (*(- VAN $ESSEL $EPARTMENT OF /BSTETRICS AND 'YNAECOLOGY 5NIVERSITY (OSPITAL 'HENT $E 0INTELAAN 'HENT "ELGIUM $EPARTMENT OF 2EPRODUCTIVE -EDICINE $EPARTMENT OF %MBRYOLOGY 3T %LISABETH (OSPITAL 4ILBURG $EPARTMENT OF /BSTETRICS AND 'YNAECOLOGY !CADEMIC -EDICAL #ENTRE !MSTERDAM $EPARTMENT OF /BSTETRICS AND 'YNAECOLOGY -ÉXIMA -EDICAL #ENTRE 6ELDHOVEN 4HE .ETHERLANDS #ORRESPONDENCE 4EL E MAIL "6ISSCHERS YAHOOCOUK
!BSTRACT Cervical mucus may cover the embryo transfer catheter during passage of the cervical canal, interfering with the correct placement of the embryo(s) into the uterine cavity. The effect of removal of cervical mucus prior to embryo transfer in IVF/ intracytoplasmic sperm injection (ICSI) on live birth rate was studied. The study was set up as a single blind randomized controlled trial. Couples undergoing IVF/ICSI were randomly allocated to either removal of cervical mucus prior to embryo transfer, or a mock procedure. Randomization was done with stratification for age, cycle number and method of treatment. Primary outcome was live birth rate. A total of 317 couples were included and underwent 428 cycles, of which the outcome of 3 cycles was unknown. Baseline characteristics of both groups were comparable. Live birth occurred in 52 of 220 (24%) cycles in the treatment group and 42 of 205 (21%) cycles in the control group (risk difference 3%, 95% confidence interval −5−11%). It is unlikely that removal of cervical mucus prior to embryo transfer has a significant effect on live birth rate. A small effect, however, cannot be excluded. Keywords: cervical mucus, embryo transfer, randomized controlled trial
)NTRODUCTION
There is a growing interest in optimizing embryo transfer in IVF/ intracytoplasmic sperm injection (ICSI), and numerous attempts to improve the technique have been described, including the need for a gentle and atraumatic procedure (Nabi et al., 1997; Cevrioglu et al., 2006), performing a trial transfer before the actual procedure (Levi Setti et al., 2004; Neithardt et al., 2005), avoiding the use of a tenaculum (Lesny et al., 1999), using specific catheters (Ghazzawi et al., 1999; Meriano et al., 2000; Abou-Setta et al., 2006; Ata et al., 2007), location of deposition of the embryos in the uterine cavity (Egbase et al., 2000; Cavagna et al., 2006), embryo afterloading (Neithardt et al., 2005), flushing of the endometrial cavity prior to embryo transfer (Berkkanoglu et al., 2006), and performing the procedure under ultrasound guidance (Sallam et al., 2002; Sallam and Sadek, 2003). Scant data are available on the role of cervical mucus during embryo transfer.
Hypothetically, cervical mucus might act as a film covering the embryo transfer catheter during passage of the cervical canal. Thus, this ‘condom’ of cervical mucus might interfere with the correct placement of the embryo(s) into the uterine cavity. The embryo(s) might stick to the cervical mucus around the catheter and could be dragged along during withdrawal of the catheter. Consequently, it seems conceivable that meticulous removal of cervical mucus prior to embryo transfer can improve embryo placement in the uterine cavity and thus result in higher clinical pregnancy rates and eventually in higher live birth rates. Removal of cervical mucus may also minimize the inoculation of catheterbound micro-organisms and/or pro-inflammatory factors from the cervix into the uterine cavity, leading to clinical or sub clinical endometritis and a reduction in implantation and live birth rates.
© 2007 Published by Reproductive Healthcare Ltd, Duck End Farm, Dry Drayton, Cambridge CB3 8DB, UK
Article - Cervical mucus removal: effect on pregnancy rates - BAJT Visschers et al.
Three arguments support this preliminary hypothesis. During an in-vitro experiment, a transfer catheter filled with methylene blue was introduced into an artificial cervical canal filled with cervical mucus obtained from a patient on cycle day 13. An attempt to deliver a droplet of methylene blue failed, apparently due to the mucus acting as a condom that prevented the methylene blue from leaving the catheter. Secondly, on reviewing patient data, a higher clinical pregnancy rate was observed in a cohort of patients who routinely underwent cervical mucus removal, as compared with a cohort in which this procedure was not applied (clinical pregnancy rate 33 versus 25% per cycle). This finding might suggest that removal of cervical mucus prior to embryo transfer could indeed improve pregnancy rates. Thirdly, several authors recommend cervical mucus removal prior to embryo transfer (Nabi et al., 1997; Mansour and Aboulghar, 2002). An in-vitro study (Mansour et al., 1994) using methylene blue as dummy embryo transfer showed that the dye was extruded at the external os in 57% of the cases when the cervical mucus was not aspirated compared with 23% when the mucus was aspirated. Other studies showed that the appearance of mucus and/or blood at the catheter tip was associated with either an increased risk of retention of embryos or a compromised pregnancy rate (Nabi et al., 1997; Awoniyi et al., 1998), although the latter has been debated by others, who showed that the appearance of mucus and/or blood at the catheter tip can be used as an index of a difficult and traumatic procedure rather than the mucus being a compromising factor in embryo transfer (Levi Setti et al., 2004). Although all these data suggest a possible role of cervical mucus in embryo transfer, there is no clinical evidence on the effect of removal of cervical mucus on the outcome of IVF/ICSI. In view of this uncertainty, a study was undertaken to evaluate the role of cervical mucus removal prior to embryo transfer on pregnancy rates in IVF and ICSI.
-ATERIALS AND METHODS 0ATIENTS This study was designed as a single blind randomized controlled trial. Over a period of 1 year (April 2003 to March 2004) couples undergoing IVF or ICSI treatment were invited to participate. Patients originating from three hospitals (St Elisabeth Hospital in Tilburg, Jeroen Bosch Hospital in ’s-Hertogenbosch and Amphia Hospital in Breda, The Netherlands) were included. Embryo transfer was performed at the St Elisabeth Hospital in Tilburg in all patients. The study had institutional review board approval. Written and oral information on the study was given prior to the start of the IVF or ICSI treatment and informed consent was obtained before embryo transfer. Immediately prior to embryo transfer, women were randomly allocated to removal of cervical mucus (treatment group) or a mock procedure (control group). The randomization was done immediately before embryo transfer by the clinician performing the procedure. Randomization was performed with a computer program with stratification for age (≤ or >36 years), cycle number (first or higher cycle) and method of treatment (IVF or ICSI) using block randomization. In the treatment group cervical mucus was removed by the use of a cervical brush, as commonly used to collect cervical PAP smear tests (Cervex-Brush®; Rovers Medical Devices BV, Oss, The Netherlands). When cervical bleeding RBMOnline®
occurred, it was stopped by applying small cotton swabs. The patients in the control group underwent a mock procedure during which the ectocervix was touched with a cotton wool swab, but no endocervical removal of mucus occurred. Patients remained unaware of the procedure that was used. Clinical data (gravidity, parity, primary or secondary infertility, duration of infertility, cause of infertility, fresh or frozen embryos used, number of oocytes retrieved, number and quality of embryos transferred) were recorded in the electronic patient file, while the outcome of the randomization procedure was kept in a separate file. Apart from the doctors doing the actual embryo transfer, all physicians involved in treatment and follow-up of the patients and data analysis were blinded for the results of the allocation. After completion of the study the randomization log file contained the identification of the patient and the outcome of the randomization. These data were merged with the clinical data.
)6&)#3) PROCEDURE Ovarian stimulation, oocyte retrieval and IVF Ovarian stimulation was accomplished with FSH (Gonal F®; Serono Benelux BV, The Hague, The Netherlands) starting on day 2 of the cycle after pituitary down-regulation with leuproreline acetate (Lucrin®; Abbott BV, Hoofddorp, The Netherlands). When four to six leading follicles had reached a mean diameter of 18 mm, 10,000 IU of human chorionic gonadotrophin (HCG, Pregnyl®; Organon BV, Oss, The Netherlands) was administered i.m. Oocyte retrieval was scheduled to take place 34–36 h later by transvaginal ultrasound guidance after administering 0.5−1.0 mg alfentanil hydrochloride (Rapifen®; Janssen-Cilag BV, Tilburg, The Netherlands) and 0.5 mg atropine (Pharmachemie BV, Haarlem, The Netherlands) i.m. About 6 h later, the oocytes were inseminated and kept in HTF medium at 37°C. Fertilization was confirmed by the presence of two pronuclei 1 day after insemination.
Embryo development and morphology On the morning of embryo transfer, embryos were examined and the number of cells determined. Each embryo was scored according to its symmetry and the extent of fragmentation and granulation of blastomeres. Embryos with the best morphology and the most advanced stage of development were selected for transfer. Normally, up to two embryos were transferred in each group on day 2 or 3 after oocyte retrieval.
Technique of embryo transfer Patients were positioned in the lithotomy position and the cervix was exposed with a bivalve speculum. No disinfectant or medium was used. A Frydman catheter (CCD Laboratories, Paris, France) without the outer part was preloaded with embryos using the ‘three-drop procedure’ in which the embryos were separated by a bubble of air from a preceding and a following drop of medium. The catheter was then inserted through the external and internal cervical ostium using the markings on the catheter to ensure that embryos
Article - Cervical mucus removal: effect on pregnancy rates - BAJT Visschers et al.
were released at 2 cm distance from the fundus. Occasionally, the outer part of the Frydman catheter was used and/or grasping and dilatation of the cervix was performed when difficulty in introducing the catheter was encountered. Details of the transfer procedure including bleeding and excessive manipulation were carefully recorded. Thereafter, the patient was discharged and allowed to resume her usual activities. Intravaginal progesterone pessaries [Progestan® (Organon BV) 400 mg daily] supported the luteal phase.
Outcome The primary outcome parameter was live birth rate, defined as the birth of a living fetus (single or twins) beyond a gestational age of 26 weeks. Data on pregnancy and live birth rate were usually recorded in the database containing the electronic patient files. In cases where data were missing, the patient’s (obstetric) files were investigated, or information was requested from other hospitals. Secondary outcome parameters were clinical pregnancy rate, defined as a positive HCG test 2 weeks following embryo transfer) and number of ongoing pregnancies, defined as the presence of a viable pregnancy on ultrasound 10 weeks after embryo transfer.
3TATISTICAL ANALYSIS Before commencing this study, a power calculation was conducted. To show a 10% difference in live birth rate, 400 cycles were needed (β 0.20, α 0.05, two-sided test). Analysis was performed according to intention-to-treat. In both groups, live birth rates and pregnancy rates were calculated per cycle. Differences were assessed by calculating an absolute risk difference with 95% confidence interval.
2ESULTS A total of 428 cycles were included, collected in 317 patients. A flow chart is depicted in Figure 1. Three couples were lost to follow-up. Baseline characteristics of both groups were comparable (Table 1). Live birth occurred in 52 out of 220 cycles (24%) in the treatment group and 42 of 205 cycles (21%) in the control group (risk difference 3%, 95% CI −5−11%). Cycle outcomes are shown in Table 2. Clinical pregnancy occurred in 65 out of 220 cycles (30%) in the treatment group and in 63 out of 205 cycles (31%) in the control group (risk difference 1%, 95% CI −7−10%). Ongoing pregnancy occurred in 56 out of 220 cycles (25%) in the treatment group versus 46 out of 205 cycles (22%) in the control group (risk difference 3%, 95% CI −5−11%.). The twin rate was 14 out of 220 cycles (6%) in the treatment group versus 8 out of 205 cycles (4%) in the control group (risk difference 2%, 95% CI −2−7%). In addition, an analysis was performed for first cycles only (Table 3). Live birth occurred in 20 out of 62 cycles (32%) in the treatment group and 18 out of 66 cycles (27%) in the control group (risk difference 5%, 95% CI −11−21%). Clinical pregnancy occurred in 23 out of 62 cycles (37%) in the treatment group and in 23 out of 66 cycles (35%) in the control group (risk difference 3%, 95% CI −14−19%). Ongoing pregnancy occurred in 20 out of 62 cycles (32%) in the treatment group versus 18 out of 66 cycles (27%) in the control group (risk difference 5%, 95% CI −11−21%.). The twin rate was four out of 62 cycles (6%) in the treatment group versus three out of 66 cycles (5%) in the control group (risk difference 2%, 95% CI −6−10%).
%NROLMENT N CYCLES
#YCLES ALLOCATED TO INTERVENTION BRUSH N
#YCLES ALLOCATED TO INTERVENTION CONTROL N
#YCLES LOST TO FOLLOW UP N
#YCLES LOST TO FOLLOW UP N
#YCLES ANALYSED N
#YCLES ANALYSED N
#LINICAL PREGNANCY N
#LINICAL PREGNANCY N
/NGOING PREGNANCY N
/NGOING PREGNANCY N
,IVE BIRTH N
,IVE BIRTH N
Figure 1. Flow diagram of study design and outcome. RBMOnline®
Article - Cervical mucus removal: effect on pregnancy rates - BAJT Visschers et al.
Table 1. Clinical characteristics of patients in both groups with versus without removal of cervical mucus prior to embryo transfer.
Previous pregnancy Nulligravida Multigravida Age in years (mean, SD) Duration of subfertility in years (mean, SD) Subfertility Primary Secondary Cycle number 1 ≥2 Cause of infertility Male Tubal/uterine Hormonal Endometriosis Cervical factor Unknown Assisted reproduction procedure IVF ICSI Both Embryos Fresh Cryo No. of embryos transferred 1 2 Quality of embryos transferred Perfect Good Moderate Poor Unknown
Treatment group, n = 220 (%)
Control group, n = 205 (%)
131 (60.0) 89 (40.0) 33 (4.2) 3.2 (1.8)
101 (49.3) 104 (50.7) 33 (4.2) 3.3 (2.2)
142 (65.0) 78 (35.0)
117 (57.1) 88 (42.9)
63 (28.6) 157 (71.4)
66 (32.2) 139 (67.8)
125 (56.8) 46 (20.9) 9 (4.1) 9 (4.1) 7 (3.2) 24 (10.9)
114 (55.6) 35 (17.1) 8 (3.9) 9 (4.4) 11 (5.4) 28 (13.7)
121 (55.0) 96 (43.6) 3 (1.4)
117 (57.1) 86 (42.0) 2 (1.0)
181 (82.3) 39 (17.7)
178 (86.8) 27 (13.2)
50 (22.7) 170 (77.3)
37 (18.0) 168 (82.0)
24 (6.2) 184 (47.2) 107 (27.4) 17 (4.4) 58 (14.9)
39 (10.5) 172 (46.1) 110 (29.5) 14 (3.8) 38 (10.2)
Values in parentheses are percentages unless otherwise stated. ICSI = intracytoplasmic sperm injection.
RBMOnline®
Article - Cervical mucus removal: effect on pregnancy rates - BAJT Visschers et al.
Table 2. Outcomes for patients treated with cervical mucus removal versus the control group. Treatment group, n = 220 (%) HCG positive Gestational sacs 0 1 2 Pregnancy Ongoing 1 2 Abortion Live birth 1 2
Control group, n = 205 (%)
65 (30)
63 (31)
160 (73) 46 (21) 14 (6)
150 (73) 47 (23) 8 (4)
56 (25) 47 (21) 9 (4) 15 (7) 52 (24) 43 (20) 9 (4)
46 (22) 38 (19) 8 (4) 20 (10) 42 (21) 36 (18) 6 (3)
HCG positive Gestational sacs 0 1 2 Pregnancy Ongoing 1 2 Abortion Live birth 1 2
Treatment group, n = 62 (%)
Control group, n = 66 (%)
23 (37)
23 (35)
2 (3) 16 (26) 5 (8)
3 (5) 16 (24) 4 (6)
20 (32) 16 (26) 4 (6) 2 (3) 20 (32) 16 (26) 4 (6)
18 (27) 14 (21) 4 (6) 5 (8) 18 (27) 15 (23) 3 (5)
HCG = human chorionic gonadotrophin.
HCG = human chorionic gonadotrophin.
$ISCUSSION
the analysis as variables are not independent (Salim Daya, 2003). Furthermore, the intervention in one cycle may have a modifying effect in subsequent cycles, thereby influencing the outcome. When the analysis was limited to first cycles only, the results of the analysis did not change significantly (Table 3).
In this double-blind randomized study, no significant benefit of routine removal of cervical mucus prior to embryo transfer could be demonstrated. However, based on the findings, a small benefit cannot be excluded. Possibly, the preliminary hypothesis of a condom-like action of cervical mucus during embryo transfer was incorrect. Apparently, in-vitro experiments did not accurately reflect circumstances during in-vivo embryo transfer. Also, retrospective clinical observations made from earlier patient data, as described in the Introduction, are well known to be prone to considerable bias. On the other hand, a possible beneficial effect of removal of cervical mucus might be counteracted by the negative effect of the cervical brush manoeuvre. Manipulation of the cervix might enhance contractility of the uterus and then result in expulsion of embryos (Visser et al. 1993; Awoniyi et al., 1998; Mansour and Aboulghar, 2002; Alvero, 2003). The technique used to remove the mucus by means of a cervical brush might be more prone to this effect than others. Recently, Eskandar et al. concluded from their study that aspiration of cervical mucus prior to embryo transfer resulted in significantly higher pregnancy rates even when transfer was more difficult when the mucus had been aspirated (Eskandar et al., 2007).
Table 3. Outcomes of first cycles for patients treated with cervical mucus removal versus the control group.
Patients were allowed to participate in multiple cycles. Each cycle was considered as a new entry in the study and randomized accordingly. This technique, although commonly used in cohort studies and randomized trials in assisted reproduction techniques, is subject to debate. It has been pointed out that, when several cycles per patient are included, bias enters
To further improve ongoing pregnancy rates and live birth rate, use of a different catheter (Ghazzawi et al., 1999; Meriano et al., 2000; Abou-Setta et al., 2006; Ata et al., 2007), and ultrasound guidance during the embryo transfer (Sallam et al., 2003) will be considered in future policy, as recommended in the reviewed literature. In conclusion, removal of cervical mucus prior to embryo transfer in IVF/ICSI does not increase live birth rate in a significant way. To demonstrate a more subtle difference in live birth rate, a larger study is needed. Since cervical cleaning is such a simple and inexpensive procedure, even a small benefit could make it worthwhile.
!CKNOWLEDGEMENTS The authors thank the Department of Reproductive Medicine, St Elisabeth Hospital, Tilburg: CHJ Tiemessen, JM Bernards, MM Sprock, E Geenen. Department of Embryology St Elisabeth Hospital, Tilburg: Dr D Consten, Dr JH Marcelis, IVF Centre Amphia Hospital, Breda: Dr P van Bommel, IVF Centre Jeroen Bosch Hospital, ’s-Hertogenbosch: Dr C Hamilton.
2EFERENCES 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, 191–198.
RBMOnline®
Article - Cervical mucus removal: effect on pregnancy rates - BAJT Visschers et al.
Alvero R, Hearns-Stokes RM, Catherino WH et al. 2003 The presence of blood in the transfer catheter negatively influences outcome at embryo transfer. Human Reproduction 18, 1848–1852. Ata B, Isiklar A, Balaban B, Urman B 2007 Prospective randomized comparison of Wallace and Labotect embryo transfer catheters. Reproductive BioMedicine Online 14, 471–476. Awonuga A, Nabi A, Govindbhai J et al. 1998 Contamination of embryo transfer catheter and treatment outcome in in vitro fertilization. Journal of Assisted Reproduction and Genetics 15, 198–201. 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, 268–271. Cavagna M, Contart P, Petersen CG et al. 2006 Implantation sites after embryo transfer into the central area of the uterine cavity. Reproductive BioMedicine Online 13, 541–546. Cevrioglu AS, Esinler I, Bozdag G, H Yarali 2006 Assessment of endocervical and endometrial damage inflicted by embryo transfer trial: a hysteroscopic evaluation. Reproductive BioMedicine Online 13, 523–527. Egbase PE, Al-Sharhan M, Grudzinskas JG 2000 Influence of position and length of uterus on implantation and clinical pregnancy rates in IVF and embryo transfer treatment cycles. Human Reproduction 15, 1943–1946. Eskandar M, Abou-Setta AM, El-Amin M et al. 2007 Removal of cervical mucus prior to embryo transfer improves pregnancy rates in women undergoing assisted reproduction. Reproductive BioMedicine Online 14, 308–313. Ghazzawi IM, Al-Hasani S, Karaki R, Souso S 1999 Transfer technique and catheter choice influence the incidence of transcervical embryo expulsion and the outcome of IVF. Human Reproduction 14, 677–682.
Levi Setti PE, Colombo GV, Savasi V et al. 2004 Implantation failure in assisted reproduction technology and a critical approach to treatment. Annals of the New York Academy of Sciences 1034, 184–199. Mansour RT, Aboulghar MA 2002 Optimizing the embryo transfer technique. Human Reproduction 17, 1149–1153. Mansour RT, Aboulghar MA, Serour GI, Amin YM 1994 Dummy embryo transfer using methylene blue dye. Human Reproduction 9, 1257–1259. Meriano J, Weissman A, Greenblatt EM et al. 2000 The choice of embryo transfer catheter affects embryo implantation after IVF. Fertility and Sterility 74, 678–682. Nabi A, Awonuga A, Birch H et al. 1997 Multiple attempts at embryo transfer: does this affect in-vitro fertilization treatment outcome. Human Reproduction 12, 1188–1190. Neithardt AB, Segars JH, Hennessy S et al. 2005 Embryo afterloading: a refinement in embryo transfer technique that may increase clinical pregnancy. Fertility and Sterility 83, 710–714. Salim Daya 2003 Inappropriate use of cohort studies to formulate treatment guidelines. Fertility and Sterility 79, 27. Sallam HN, Sadek SS 2003 Ultrasound-guided embryo transfer: a meta-analysis of randomized controlled trials. Fertility and Sterility 80, 1042–1046. Sallam HN, Agameya AF, Rahman AF et al. 2002 Ultrasound measurement of the uterocervical angle before embryo transfer: a prospective controlled study. Human Reproduction 17, 1767–1772. Visser DS, Fourie FL, Kruger HF 1993 Multiple attempts at embryo transfer: effect on pregnancy outcome in an in vitro fertilization and embryo transfer program. Journal of Assisted Reproduction and Genetics 10, 37–43. Received 9 May 2007; refereed 18 May 2007; accepted 11 June 2007.
RBMOnline®
!RTICLE 2EMOVAL OF CERVICAL MUCUS EFFECT ON PREGNANCY RATES IN )6&)#3) "IANCA 6ISSCHERS WAS BORN IN .OVEMBER 3HE COMPLETED HER MEDICAL STUDIES IN AT THE #ATHOLIC 5NIVERSITY OF .IJMEGEN IN THE .ETHERLANDS !FTER GRADUATING SHE SPECIALIZED IN 4ROPICAL -EDICINE INCLUDING TRAINING IN /BSTETRICS AND 'YNAECOLOGY IN 'ELRE (OSPITALS !PELDOORN )N n SHE WORKED AS A 4ROPICAL $OCTOR IN -ASINDI $ISTRICT (OSPITAL -ASINDI 5GANDA !FTER HER RETURN TO THE .ETHERLANDS SHE TOOK UP A POST AT THE $EPARTMENT OF 'YNAECOLOGY AND /BSTETRICS IN 4WEE 3TEDEN (OSPITAL 4ILBURG 2ECENTLY SHE STARTED HER SPECIALIST TRAINING IN 'YNAECOLOGY AND /BSTETRICS AT THE 5NIVERSITY (OSPITAL OF 'HENT "ELGIUM
$R "IANCA 6ISSCHERS "!*4 6ISSCHERS 23'- "OTS -& 0EETERS "7* -OL (*(- VAN $ESSEL $EPARTMENT OF /BSTETRICS AND 'YNAECOLOGY 5NIVERSITY (OSPITAL 'HENT $E 0INTELAAN 'HENT "ELGIUM $EPARTMENT OF 2EPRODUCTIVE -EDICINE $EPARTMENT OF %MBRYOLOGY 3T %LISABETH (OSPITAL 4ILBURG $EPARTMENT OF /BSTETRICS AND 'YNAECOLOGY !CADEMIC -EDICAL #ENTRE !MSTERDAM $EPARTMENT OF /BSTETRICS AND 'YNAECOLOGY -ÉXIMA -EDICAL #ENTRE 6ELDHOVEN 4HE .ETHERLANDS #ORRESPONDENCE 4EL E MAIL "6ISSCHERS YAHOOCOUK
!BSTRACT Cervical mucus may cover the embryo transfer catheter during passage of the cervical canal, interfering with the correct placement of the embryo(s) into the uterine cavity. The effect of removal of cervical mucus prior to embryo transfer in IVF/ intracytoplasmic sperm injection (ICSI) on live birth rate was studied. The study was set up as a single blind randomized controlled trial. Couples undergoing IVF/ICSI were randomly allocated to either removal of cervical mucus prior to embryo transfer, or a mock procedure. Randomization was done with stratification for age, cycle number and method of treatment. Primary outcome was live birth rate. A total of 317 couples were included and underwent 428 cycles, of which the outcome of 3 cycles was unknown. Baseline characteristics of both groups were comparable. Live birth occurred in 52 of 220 (24%) cycles in the treatment group and 42 of 205 (21%) cycles in the control group (risk difference 3%, 95% confidence interval −5−11%). It is unlikely that removal of cervical mucus prior to embryo transfer has a significant effect on live birth rate. A small effect, however, cannot be excluded. Keywords: cervical mucus, embryo transfer, randomized controlled trial
)NTRODUCTION
There is a growing interest in optimizing embryo transfer in IVF/ intracytoplasmic sperm injection (ICSI), and numerous attempts to improve the technique have been described, including the need for a gentle and atraumatic procedure (Nabi et al., 1997; Cevrioglu et al., 2006), performing a trial transfer before the actual procedure (Levi Setti et al., 2004; Neithardt et al., 2005), avoiding the use of a tenaculum (Lesny et al., 1999), using specific catheters (Ghazzawi et al., 1999; Meriano et al., 2000; Abou-Setta et al., 2006; Ata et al., 2007), location of deposition of the embryos in the uterine cavity (Egbase et al., 2000; Cavagna et al., 2006), embryo afterloading (Neithardt et al., 2005), flushing of the endometrial cavity prior to embryo transfer (Berkkanoglu et al., 2006), and performing the procedure under ultrasound guidance (Sallam et al., 2002; Sallam and Sadek, 2003). Scant data are available on the role of cervical mucus during embryo transfer.
Hypothetically, cervical mucus might act as a film covering the embryo transfer catheter during passage of the cervical canal. Thus, this ‘condom’ of cervical mucus might interfere with the correct placement of the embryo(s) into the uterine cavity. The embryo(s) might stick to the cervical mucus around the catheter and could be dragged along during withdrawal of the catheter. Consequently, it seems conceivable that meticulous removal of cervical mucus prior to embryo transfer can improve embryo placement in the uterine cavity and thus result in higher clinical pregnancy rates and eventually in higher live birth rates. Removal of cervical mucus may also minimize the inoculation of catheterbound micro-organisms and/or pro-inflammatory factors from the cervix into the uterine cavity, leading to clinical or sub clinical endometritis and a reduction in implantation and live birth rates.
© 2007 Published by Reproductive Healthcare Ltd, Duck End Farm, Dry Drayton, Cambridge CB3 8DB, UK
Article - Cervical mucus removal: effect on pregnancy rates - BAJT Visschers et al.
Three arguments support this preliminary hypothesis. During an in-vitro experiment, a transfer catheter filled with methylene blue was introduced into an artificial cervical canal filled with cervical mucus obtained from a patient on cycle day 13. An attempt to deliver a droplet of methylene blue failed, apparently due to the mucus acting as a condom that prevented the methylene blue from leaving the catheter. Secondly, on reviewing patient data, a higher clinical pregnancy rate was observed in a cohort of patients who routinely underwent cervical mucus removal, as compared with a cohort in which this procedure was not applied (clinical pregnancy rate 33 versus 25% per cycle). This finding might suggest that removal of cervical mucus prior to embryo transfer could indeed improve pregnancy rates. Thirdly, several authors recommend cervical mucus removal prior to embryo transfer (Nabi et al., 1997; Mansour and Aboulghar, 2002). An in-vitro study (Mansour et al., 1994) using methylene blue as dummy embryo transfer showed that the dye was extruded at the external os in 57% of the cases when the cervical mucus was not aspirated compared with 23% when the mucus was aspirated. Other studies showed that the appearance of mucus and/or blood at the catheter tip was associated with either an increased risk of retention of embryos or a compromised pregnancy rate (Nabi et al., 1997; Awoniyi et al., 1998), although the latter has been debated by others, who showed that the appearance of mucus and/or blood at the catheter tip can be used as an index of a difficult and traumatic procedure rather than the mucus being a compromising factor in embryo transfer (Levi Setti et al., 2004). Although all these data suggest a possible role of cervical mucus in embryo transfer, there is no clinical evidence on the effect of removal of cervical mucus on the outcome of IVF/ICSI. In view of this uncertainty, a study was undertaken to evaluate the role of cervical mucus removal prior to embryo transfer on pregnancy rates in IVF and ICSI.
-ATERIALS AND METHODS 0ATIENTS This study was designed as a single blind randomized controlled trial. Over a period of 1 year (April 2003 to March 2004) couples undergoing IVF or ICSI treatment were invited to participate. Patients originating from three hospitals (St Elisabeth Hospital in Tilburg, Jeroen Bosch Hospital in ’s-Hertogenbosch and Amphia Hospital in Breda, The Netherlands) were included. Embryo transfer was performed at the St Elisabeth Hospital in Tilburg in all patients. The study had institutional review board approval. Written and oral information on the study was given prior to the start of the IVF or ICSI treatment and informed consent was obtained before embryo transfer. Immediately prior to embryo transfer, women were randomly allocated to removal of cervical mucus (treatment group) or a mock procedure (control group). The randomization was done immediately before embryo transfer by the clinician performing the procedure. Randomization was performed with a computer program with stratification for age (≤ or >36 years), cycle number (first or higher cycle) and method of treatment (IVF or ICSI) using block randomization. In the treatment group cervical mucus was removed by the use of a cervical brush, as commonly used to collect cervical PAP smear tests (Cervex-Brush®; Rovers Medical Devices BV, Oss, The Netherlands). When cervical bleeding RBMOnline®
occurred, it was stopped by applying small cotton swabs. The patients in the control group underwent a mock procedure during which the ectocervix was touched with a cotton wool swab, but no endocervical removal of mucus occurred. Patients remained unaware of the procedure that was used. Clinical data (gravidity, parity, primary or secondary infertility, duration of infertility, cause of infertility, fresh or frozen embryos used, number of oocytes retrieved, number and quality of embryos transferred) were recorded in the electronic patient file, while the outcome of the randomization procedure was kept in a separate file. Apart from the doctors doing the actual embryo transfer, all physicians involved in treatment and follow-up of the patients and data analysis were blinded for the results of the allocation. After completion of the study the randomization log file contained the identification of the patient and the outcome of the randomization. These data were merged with the clinical data.
)6&)#3) PROCEDURE Ovarian stimulation, oocyte retrieval and IVF Ovarian stimulation was accomplished with FSH (Gonal F®; Serono Benelux BV, The Hague, The Netherlands) starting on day 2 of the cycle after pituitary down-regulation with leuproreline acetate (Lucrin®; Abbott BV, Hoofddorp, The Netherlands). When four to six leading follicles had reached a mean diameter of 18 mm, 10,000 IU of human chorionic gonadotrophin (HCG, Pregnyl®; Organon BV, Oss, The Netherlands) was administered i.m. Oocyte retrieval was scheduled to take place 34–36 h later by transvaginal ultrasound guidance after administering 0.5−1.0 mg alfentanil hydrochloride (Rapifen®; Janssen-Cilag BV, Tilburg, The Netherlands) and 0.5 mg atropine (Pharmachemie BV, Haarlem, The Netherlands) i.m. About 6 h later, the oocytes were inseminated and kept in HTF medium at 37°C. Fertilization was confirmed by the presence of two pronuclei 1 day after insemination.
Embryo development and morphology On the morning of embryo transfer, embryos were examined and the number of cells determined. Each embryo was scored according to its symmetry and the extent of fragmentation and granulation of blastomeres. Embryos with the best morphology and the most advanced stage of development were selected for transfer. Normally, up to two embryos were transferred in each group on day 2 or 3 after oocyte retrieval.
Technique of embryo transfer Patients were positioned in the lithotomy position and the cervix was exposed with a bivalve speculum. No disinfectant or medium was used. A Frydman catheter (CCD Laboratories, Paris, France) without the outer part was preloaded with embryos using the ‘three-drop procedure’ in which the embryos were separated by a bubble of air from a preceding and a following drop of medium. The catheter was then inserted through the external and internal cervical ostium using the markings on the catheter to ensure that embryos
Article - Cervical mucus removal: effect on pregnancy rates - BAJT Visschers et al.
were released at 2 cm distance from the fundus. Occasionally, the outer part of the Frydman catheter was used and/or grasping and dilatation of the cervix was performed when difficulty in introducing the catheter was encountered. Details of the transfer procedure including bleeding and excessive manipulation were carefully recorded. Thereafter, the patient was discharged and allowed to resume her usual activities. Intravaginal progesterone pessaries [Progestan® (Organon BV) 400 mg daily] supported the luteal phase.
Outcome The primary outcome parameter was live birth rate, defined as the birth of a living fetus (single or twins) beyond a gestational age of 26 weeks. Data on pregnancy and live birth rate were usually recorded in the database containing the electronic patient files. In cases where data were missing, the patient’s (obstetric) files were investigated, or information was requested from other hospitals. Secondary outcome parameters were clinical pregnancy rate, defined as a positive HCG test 2 weeks following embryo transfer) and number of ongoing pregnancies, defined as the presence of a viable pregnancy on ultrasound 10 weeks after embryo transfer.
3TATISTICAL ANALYSIS Before commencing this study, a power calculation was conducted. To show a 10% difference in live birth rate, 400 cycles were needed (β 0.20, α 0.05, two-sided test). Analysis was performed according to intention-to-treat. In both groups, live birth rates and pregnancy rates were calculated per cycle. Differences were assessed by calculating an absolute risk difference with 95% confidence interval.
2ESULTS A total of 428 cycles were included, collected in 317 patients. A flow chart is depicted in Figure 1. Three couples were lost to follow-up. Baseline characteristics of both groups were comparable (Table 1). Live birth occurred in 52 out of 220 cycles (24%) in the treatment group and 42 of 205 cycles (21%) in the control group (risk difference 3%, 95% CI −5−11%). Cycle outcomes are shown in Table 2. Clinical pregnancy occurred in 65 out of 220 cycles (30%) in the treatment group and in 63 out of 205 cycles (31%) in the control group (risk difference 1%, 95% CI −7−10%). Ongoing pregnancy occurred in 56 out of 220 cycles (25%) in the treatment group versus 46 out of 205 cycles (22%) in the control group (risk difference 3%, 95% CI −5−11%.). The twin rate was 14 out of 220 cycles (6%) in the treatment group versus 8 out of 205 cycles (4%) in the control group (risk difference 2%, 95% CI −2−7%). In addition, an analysis was performed for first cycles only (Table 3). Live birth occurred in 20 out of 62 cycles (32%) in the treatment group and 18 out of 66 cycles (27%) in the control group (risk difference 5%, 95% CI −11−21%). Clinical pregnancy occurred in 23 out of 62 cycles (37%) in the treatment group and in 23 out of 66 cycles (35%) in the control group (risk difference 3%, 95% CI −14−19%). Ongoing pregnancy occurred in 20 out of 62 cycles (32%) in the treatment group versus 18 out of 66 cycles (27%) in the control group (risk difference 5%, 95% CI −11−21%.). The twin rate was four out of 62 cycles (6%) in the treatment group versus three out of 66 cycles (5%) in the control group (risk difference 2%, 95% CI −6−10%).
%NROLMENT N CYCLES
#YCLES ALLOCATED TO INTERVENTION BRUSH N
#YCLES ALLOCATED TO INTERVENTION CONTROL N
#YCLES LOST TO FOLLOW UP N
#YCLES LOST TO FOLLOW UP N
#YCLES ANALYSED N
#YCLES ANALYSED N
#LINICAL PREGNANCY N
#LINICAL PREGNANCY N
/NGOING PREGNANCY N
/NGOING PREGNANCY N
,IVE BIRTH N
,IVE BIRTH N
Figure 1. Flow diagram of study design and outcome. RBMOnline®
Article - Cervical mucus removal: effect on pregnancy rates - BAJT Visschers et al.
Table 1. Clinical characteristics of patients in both groups with versus without removal of cervical mucus prior to embryo transfer.
Previous pregnancy Nulligravida Multigravida Age in years (mean, SD) Duration of subfertility in years (mean, SD) Subfertility Primary Secondary Cycle number 1 ≥2 Cause of infertility Male Tubal/uterine Hormonal Endometriosis Cervical factor Unknown Assisted reproduction procedure IVF ICSI Both Embryos Fresh Cryo No. of embryos transferred 1 2 Quality of embryos transferred Perfect Good Moderate Poor Unknown
Treatment group, n = 220 (%)
Control group, n = 205 (%)
131 (60.0) 89 (40.0) 33 (4.2) 3.2 (1.8)
101 (49.3) 104 (50.7) 33 (4.2) 3.3 (2.2)
142 (65.0) 78 (35.0)
117 (57.1) 88 (42.9)
63 (28.6) 157 (71.4)
66 (32.2) 139 (67.8)
125 (56.8) 46 (20.9) 9 (4.1) 9 (4.1) 7 (3.2) 24 (10.9)
114 (55.6) 35 (17.1) 8 (3.9) 9 (4.4) 11 (5.4) 28 (13.7)
121 (55.0) 96 (43.6) 3 (1.4)
117 (57.1) 86 (42.0) 2 (1.0)
181 (82.3) 39 (17.7)
178 (86.8) 27 (13.2)
50 (22.7) 170 (77.3)
37 (18.0) 168 (82.0)
24 (6.2) 184 (47.2) 107 (27.4) 17 (4.4) 58 (14.9)
39 (10.5) 172 (46.1) 110 (29.5) 14 (3.8) 38 (10.2)
Values in parentheses are percentages unless otherwise stated. ICSI = intracytoplasmic sperm injection.
RBMOnline®
Article - Cervical mucus removal: effect on pregnancy rates - BAJT Visschers et al.
Table 2. Outcomes for patients treated with cervical mucus removal versus the control group. Treatment group, n = 220 (%) HCG positive Gestational sacs 0 1 2 Pregnancy Ongoing 1 2 Abortion Live birth 1 2
Control group, n = 205 (%)
65 (30)
63 (31)
160 (73) 46 (21) 14 (6)
150 (73) 47 (23) 8 (4)
56 (25) 47 (21) 9 (4) 15 (7) 52 (24) 43 (20) 9 (4)
46 (22) 38 (19) 8 (4) 20 (10) 42 (21) 36 (18) 6 (3)
HCG positive Gestational sacs 0 1 2 Pregnancy Ongoing 1 2 Abortion Live birth 1 2
Treatment group, n = 62 (%)
Control group, n = 66 (%)
23 (37)
23 (35)
2 (3) 16 (26) 5 (8)
3 (5) 16 (24) 4 (6)
20 (32) 16 (26) 4 (6) 2 (3) 20 (32) 16 (26) 4 (6)
18 (27) 14 (21) 4 (6) 5 (8) 18 (27) 15 (23) 3 (5)
HCG = human chorionic gonadotrophin.
HCG = human chorionic gonadotrophin.
$ISCUSSION
the analysis as variables are not independent (Salim Daya, 2003). Furthermore, the intervention in one cycle may have a modifying effect in subsequent cycles, thereby influencing the outcome. When the analysis was limited to first cycles only, the results of the analysis did not change significantly (Table 3).
In this double-blind randomized study, no significant benefit of routine removal of cervical mucus prior to embryo transfer could be demonstrated. However, based on the findings, a small benefit cannot be excluded. Possibly, the preliminary hypothesis of a condom-like action of cervical mucus during embryo transfer was incorrect. Apparently, in-vitro experiments did not accurately reflect circumstances during in-vivo embryo transfer. Also, retrospective clinical observations made from earlier patient data, as described in the Introduction, are well known to be prone to considerable bias. On the other hand, a possible beneficial effect of removal of cervical mucus might be counteracted by the negative effect of the cervical brush manoeuvre. Manipulation of the cervix might enhance contractility of the uterus and then result in expulsion of embryos (Visser et al. 1993; Awoniyi et al., 1998; Mansour and Aboulghar, 2002; Alvero, 2003). The technique used to remove the mucus by means of a cervical brush might be more prone to this effect than others. Recently, Eskandar et al. concluded from their study that aspiration of cervical mucus prior to embryo transfer resulted in significantly higher pregnancy rates even when transfer was more difficult when the mucus had been aspirated (Eskandar et al., 2007).
Table 3. Outcomes of first cycles for patients treated with cervical mucus removal versus the control group.
Patients were allowed to participate in multiple cycles. Each cycle was considered as a new entry in the study and randomized accordingly. This technique, although commonly used in cohort studies and randomized trials in assisted reproduction techniques, is subject to debate. It has been pointed out that, when several cycles per patient are included, bias enters
To further improve ongoing pregnancy rates and live birth rate, use of a different catheter (Ghazzawi et al., 1999; Meriano et al., 2000; Abou-Setta et al., 2006; Ata et al., 2007), and ultrasound guidance during the embryo transfer (Sallam et al., 2003) will be considered in future policy, as recommended in the reviewed literature. In conclusion, removal of cervical mucus prior to embryo transfer in IVF/ICSI does not increase live birth rate in a significant way. To demonstrate a more subtle difference in live birth rate, a larger study is needed. Since cervical cleaning is such a simple and inexpensive procedure, even a small benefit could make it worthwhile.
!CKNOWLEDGEMENTS The authors thank the Department of Reproductive Medicine, St Elisabeth Hospital, Tilburg: CHJ Tiemessen, JM Bernards, MM Sprock, E Geenen. Department of Embryology St Elisabeth Hospital, Tilburg: Dr D Consten, Dr JH Marcelis, IVF Centre Amphia Hospital, Breda: Dr P van Bommel, IVF Centre Jeroen Bosch Hospital, ’s-Hertogenbosch: Dr C Hamilton.
2EFERENCES 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, 191–198.
RBMOnline®
Article - Cervical mucus removal: effect on pregnancy rates - BAJT Visschers et al.
Alvero R, Hearns-Stokes RM, Catherino WH et al. 2003 The presence of blood in the transfer catheter negatively influences outcome at embryo transfer. Human Reproduction 18, 1848–1852. Ata B, Isiklar A, Balaban B, Urman B 2007 Prospective randomized comparison of Wallace and Labotect embryo transfer catheters. Reproductive BioMedicine Online 14, 471–476. Awonuga A, Nabi A, Govindbhai J et al. 1998 Contamination of embryo transfer catheter and treatment outcome in in vitro fertilization. Journal of Assisted Reproduction and Genetics 15, 198–201. 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, 268–271. Cavagna M, Contart P, Petersen CG et al. 2006 Implantation sites after embryo transfer into the central area of the uterine cavity. Reproductive BioMedicine Online 13, 541–546. Cevrioglu AS, Esinler I, Bozdag G, H Yarali 2006 Assessment of endocervical and endometrial damage inflicted by embryo transfer trial: a hysteroscopic evaluation. Reproductive BioMedicine Online 13, 523–527. Egbase PE, Al-Sharhan M, Grudzinskas JG 2000 Influence of position and length of uterus on implantation and clinical pregnancy rates in IVF and embryo transfer treatment cycles. Human Reproduction 15, 1943–1946. Eskandar M, Abou-Setta AM, El-Amin M et al. 2007 Removal of cervical mucus prior to embryo transfer improves pregnancy rates in women undergoing assisted reproduction. Reproductive BioMedicine Online 14, 308–313. Ghazzawi IM, Al-Hasani S, Karaki R, Souso S 1999 Transfer technique and catheter choice influence the incidence of transcervical embryo expulsion and the outcome of IVF. Human Reproduction 14, 677–682.
Levi Setti PE, Colombo GV, Savasi V et al. 2004 Implantation failure in assisted reproduction technology and a critical approach to treatment. Annals of the New York Academy of Sciences 1034, 184–199. Mansour RT, Aboulghar MA 2002 Optimizing the embryo transfer technique. Human Reproduction 17, 1149–1153. Mansour RT, Aboulghar MA, Serour GI, Amin YM 1994 Dummy embryo transfer using methylene blue dye. Human Reproduction 9, 1257–1259. Meriano J, Weissman A, Greenblatt EM et al. 2000 The choice of embryo transfer catheter affects embryo implantation after IVF. Fertility and Sterility 74, 678–682. Nabi A, Awonuga A, Birch H et al. 1997 Multiple attempts at embryo transfer: does this affect in-vitro fertilization treatment outcome. Human Reproduction 12, 1188–1190. Neithardt AB, Segars JH, Hennessy S et al. 2005 Embryo afterloading: a refinement in embryo transfer technique that may increase clinical pregnancy. Fertility and Sterility 83, 710–714. Salim Daya 2003 Inappropriate use of cohort studies to formulate treatment guidelines. Fertility and Sterility 79, 27. Sallam HN, Sadek SS 2003 Ultrasound-guided embryo transfer: a meta-analysis of randomized controlled trials. Fertility and Sterility 80, 1042–1046. Sallam HN, Agameya AF, Rahman AF et al. 2002 Ultrasound measurement of the uterocervical angle before embryo transfer: a prospective controlled study. Human Reproduction 17, 1767–1772. Visser DS, Fourie FL, Kruger HF 1993 Multiple attempts at embryo transfer: effect on pregnancy outcome in an in vitro fertilization and embryo transfer program. Journal of Assisted Reproduction and Genetics 10, 37–43. Received 9 May 2007; refereed 18 May 2007; accepted 11 June 2007.
RBMOnline®