Review question
In women undergoing in vitro fertilisation (IVF) and intracytoplasmic sperm injection (ICSI) what are the best methods to prevent embryos from being expelled after the embryo transfer (ET) procedure?
Background
In recent years assisted reproduction has become more evidence based in an attempt to determine what techniques truly work. This has led to major changes in the way embryos are transferred and what therapies are given to women before and after the transfer. Even so, one aspect that is still not well established is whether or not there are ways to decrease the movement or expulsion rate of embryos from the uterus following transfer. Based on this lack of information, combined with the need to clearly improve outcomes by reducing the risk of losing embryos following transfer, Cochrane authors decided to systematically locate and review the best available evidence regarding interventions for the prevention of embryo expulsion in women undergoing IVF and ICSI.
Study characteristics
Following meticulous searches of major databases (all databases searched June 2013, MDSG register last searched June 2014) and conference proceedings we were able to locate four trials, with a total of 1392 women. These were all prospective, randomised controlled trials comparing two competing post-ET interventions or an intervention versus no treatment on clinical outcomes in women undergoing IVF and ICSI. Each study recruited from 164 to 639 infertile women of reproductive age and all were conducted in IVF centres; none of them had conflicts of interest regarding study funding.
Key results
Our primary measure of success, live birth rate, was not reported in any of the included trials.
There was insufficient evidence to support a specific length of time for women to remain recumbent, if at all, following ET, nor was there sufficient evidence to recommend the use of fibrin sealants. There was very limited evidence to support the use of mechanical pressure to close the cervical canal. Further large well-designed studies are required to determine the true effectiveness and safety of these interventions.
There was no evidence that any of the interventions had an effect on adverse event rates, but data were too few to reach any conclusions.
Quality of the evidence
The quality of the evidence was low or very low for all comparisons. The main limitations were failure to report live births, imprecision and risk of bias. Study risk of bias was variable, with the reporting of a proper method of randomisation and allocation concealment demonstrated in most trials while only one trial clearly blinded both the patients and the clinicians to the intervention received.
There is insufficient evidence to support any specific length of time for women to remain recumbent, if at all, following embryo transfer, nor is there sufficient evidence to recommend the use of fibrin sealants added to the embryo transfer fluid. There is very limited evidence to support the use of mechanical pressure to close the cervical canal following embryo transfer. Further well-designed and powered studies are required to determine the true effectiveness and safety of these interventions.
In women undergoing in vitro fertilisation (IVF) and intracytoplasmic sperm injection (ICSI), embryos transferred into the uterine cavity can be expelled due to many factors including uterine peristalsis and contractions, low site of deposition and negative pressure generated when removing the transfer catheter. Techniques to reduce the risk of embryo loss following embryo transfer (ET) have been described but are not standard in all centres conducting ET.
To evaluate the efficacy of interventions used to prevent post-transfer embryo expulsion in women undergoing IVF and ICSI.
We searched the Menstrual Disorders and Subfertility Group Specialised Register of controlled trials to June 2014 and PubMed, MEDLINE, EMBASE, CENTRAL, PsycINFO, CINAHL, World Health Organization ICTRP, and trial registers from inception to June 2014, with no language restrictions. Additionally, we handsearched reference lists of relevant articles, and ESHRE and ASRM conference abstracts.
We included randomised controlled trials (RCTs) of interventions used to prevent post-transfer embryo expulsion in women undergoing IVF and ICSI. Two review authors independently screened titles and abstracts and reviewed the full-texts of all potentially eligible citations to determine whether they met our inclusion criteria. Disagreements were resolved by consensus.
Two review authors independently extracted data and assessed the risk of bias of included trials using standardised, piloted data extraction forms. Data were extracted to allow intention-to-treat analyses. Disagreements were resolved by consensus. The overall quality of the evidence was rated using GRADE methods.
We included four RCTs (n = 1392 women) which administered the following interventions: bed rest (two trials), fibrin sealant (one trial), and mechanical closure of the cervix (one trial). Our primary outcome, live birth rate, was not reported in any of the included trials; nor were the data available from the corresponding authors. For the ongoing pregnancy rate, two trials comparing more bed rest with less bed rest showed no evidence of a difference between groups (odds ratio (OR) 0.88; 95% confidence interval (CI) 0.60 to 1.31, 542 women, I2 = 0%, low quality evidence). Secondary outcomes were sporadically reported with the exception of the clinical pregnancy rate, which was reported in all of the included trials. There was no evidence of a difference in clinical pregnancy rate between more bed rest and less bed rest (OR 0.88; 95% CI 0.60 to 1.31, 542 women, I2 = 0%, low quality evidence) or between fibrin sealant and usual care (OR 0.98; 95% CI 0.54 to 1.78, 211 women, very low quality evidence). However, mechanical closure of the cervix was associated with a higher clinical pregnancy rate than usual care (OR 1.92; 95% CI 1.40 to 2.63, very low quality evidence). The quality of the evidence was rated as low or very low for all outcomes. The main limitations were failure to report live births, imprecision and risk of bias. Overall, the risk of bias of the included trials was high. The use of a proper method of randomisation and allocation concealment was fairly well reported, while only one trial clearly reported blinding. There was no evidence that any of the interventions had an effect on adverse event rates but data were too few to reach any conclusions.