If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Gynaecological Endocrinology and Reproductive Medicine, University Hospital of Schleswig-Holstein, Campus Luebeck, Ratzeburger Allee 160, 23538, Luebeck, Germany
Department of Gynaecological Endocrinology and Reproductive Medicine, University Hospital of Schleswig-Holstein, Campus Luebeck, Ratzeburger Allee 160, 23538, Luebeck, Germany
A systematic literature review and meta-analysis was conducted to evaluate the effect of follicular flushing on clinical outcomes (primary outcome: mean number of cumulus–oocyte–complexes [COC]) in poor-response IVF patients). The bibliographic databases OvidMedline (includes Pubmed), Cochrane Library and Web of Science were searched electronically for randomized controlled trials (RCT) comparing follicular flushing with no flushing. Three RCT with a total of 210 patients could be included. The mean number of COC did not increase with flushing (weighted mean difference: −0.45 COC, 95% CI −1.14 to 0.25, I2 = 70%; P = 0.21; three RCT, n = 210). Mean number of metaphase II oocytes and the proportion of randomized patients having at least one COC retrieved were no different between groups. No difference was observed between groups for mean number of embryos, the proportion of randomized patients achieving embryo transfer, clinical pregnancy and live birth rates. Procedure duration was significantly increased with flushing (P = 0.0006). A positive effect of flushing on any of the investigated outcomes could not be observed in the existing literature in patients with poor ovarian response. Flushing is unlikely to significantly increase the number of oocytes, and the routine use of follicular flushing should, therefore, be scrutinized.
). Flushing follicles with multiple aspirations from the same follicle is used to increase the number of cumulus–oocyte–complexes (COC) retrieved from a given number of follicles. For that purpose, double lumen needles consisting of one lumen for aspiration and a second lumen for flushing of the ovarian follicle were developed (
A new follicle aspiration needle set is equally effective and as well tolerated as the standard needle when used in a prospective randomized trial in a large in vitro fertilization program.
), it is important to retrieve the maximum number of COCs from a given number of follicles that have developed in response to stimulation with follicle-stimulating hormone (FSH). A fraction of patients will, despite high FSH doses, produce only a small number of growing follicles (
). In case of extreme ovarian follicular depletion, mono-follicular growth may be the maximally achievable outcome of ovarian stimulation. In such patients, retrieval of at least one COC determines if embryo transfer (and thereby a chance of pregnancy achievement) may occur at all. Accordingly, in patients with poor ovarian response, the use of follicular flushing and double lumen needles has become popular for decreasing the risk of oocyte retention within the follicle. For the Austrian and German market, it is estimated from needle sales that up to 18% of oocyte retrieval procedures are carried out with flushing (Gynemed company, Lensahn, Germany; personal communication).
Apart from flushing, several approaches have been investigated to improve IVF outcomes of poor responders. For only a few of these interventions, a proof of potential benefit for patients exists, (e.g. use of androgens in poor responders (
Addition of growth hormone to gonadotrophins in ovarian stimulation of poor responders treated by in-vitro fertilization: a systematic review and meta-analysis.
)), which is why poor response is still rather challenging in IVF treatment.
In normal responding patients, randomized studies and meta-analyses comparing flushing with double lumen needles with conventional follicular aspiration with single lumen needles failed to prove superiority of the concept of follicular flushing (
). Neither meta-analysis found increased oocyte retrieval but agreed on prolonged procedure times when carrying out follicular flushing in normo-responders. In patients with poor ovarian response who might benefit the most from flushing, however, this intervention has remained popular despite negative evidence in normo-responders. In the latest Cochrane review (
), only one, single, small, study in low-response patients was available, thus limiting definite inferences on the utility of flushing in this patient population.
The aim of the present review and meta-analysis, therefore, was to systematically evaluate the primary outcome mean number of COC, retrieved from randomized controlled trials (RCT) on follicular flushing in IVF patients with poor ovarian response.
Materials and methods
Electronic literature search for identification of randomized controlled trials
The electronic literature search was built on the existing search strategy of the Cochrane library of
. The electronic literature search of the present meta-analysis covered the time point until 2017 and was conducted in the databases OvidMedline (includes PubMed), Cochrane Library and Web of Science. The literature search aimed to identify RCT from which comparative data on clinical outcome after follicular flushing versus no follicular flushing in poor ovarian response IVF patients were retrieved. The computerized literature search was conducted using various combinations of involved terminology and key words (SupplementaryAppendix).
Selection criteria
Randomized controlled trials in IVF patients termed as poor ovarian responders by the investigators were considered for inclusion in this review. No exclusion criteria were set for patient number, number of follicles or relating to specific flushing techniques. Two review authors (KN, GG) independently scanned titles and abstracts identified from the searches. Potentially relevant trials were selected and independently assessed for inclusion in this review. A flowchart of identified and analysed trials based on PRISMA guidelines is presented in Figure 1 (
High-dose gonadotrophin stimulation; ‘majority’ antagonists protocols; combinations with oestradiol patch in previous cycle; clomiphene or letrozol during stimulation in both groups; 11.2 (control group) and 10.8 days of stimulation.
Randomized, open trial comparing a modified double-lumen needle follicular flushing system with a single-lumen aspiration needle in IVF patients with poor ovarian response.
Computer-generated sequence; concealment of allocation by a sealed, opaque and sequentially numbered envelope.
Mean number of COCs.
80
One oocyte difference between groups (from 3 to 4) (α = 0.05) with 81% power (β = 0.2) based on a SD of 1.5.
Five or fewer follicles >10 mm on day of hCG administration.
Follicular flushing with a 17-gauge (Steiner-Tan®) needle with pseudo double-lumen (=one lumen which is used for flushing and aspiration via a three-way stop) (Graz, Austria)
High-dose gonadotrophin stimulation, GnRH antagonist protocol in 87.5% (control group) and 97.5% (flushing), 9.5 (control group) and 9.3 days of stimulation.
Random numbers table; concealment of allocation by consecutively numbered opaque, sealed envelopes.
Mean number of MII oocytes.
80
One oocyte difference between groups (from 3 to 4) (α = 0.05 with 80% power (β = 0.2) based on a SD of 1.48 .
Five or fewer follicles >13 mm and serum progesterone level <1.5 ng/ml on day of hCG and known poor functional ovarian reserve, AFC <6 in both ovaries together with an anti-Müllerian hormone level <0.8 ng/ml; exclusion of monofollicular growth.
Follicular flushing with a 17-gauge double-lumen needle (Cook Ireland, Limerick, Ireland).
High-dose gonadotrophin stimulation, GnRH agonist protocol in 47.5% (control group) and 57.5% (flushing), antagonist 5% in both groups, combination with letrozol and GnRH antagonist 47.5% (control group) and 37.5% (flushing).
Primary and secondary outcome, data extraction and analysis
As primary outcome criteria, mean number of COCs aspirated on follicular puncture was chosen. The secondary outcome criteria were mean number of metaphase II oocytes, proportion of patients having at least one COC, mean number of embryos, proportion of patients achieving embryo transfer, clinical pregnancy rate, live birth rate and procedure time of follicle aspiration. Study features and results were assembled in tabular form, and a formal meta-analysis was conducted. The dichotomous data results for each study were expressed as a risk difference (RD) or relative risk (RR) with 95% confidence intervals (CI). These results were combined for meta-analysis with the software RevMan 5.3 (The Nordic Cochrane Centre, The Cochrane Collaboration, 2014), using the Mantel–Haenszel method. All outcomes were extracted based on the denominator patients randomized (intention-to-treat (ITT)).
When the outcome of interest was of a continuous nature, the differences were pooled across the studies, which provided information on this outcome parameter, resulting in a weighted mean difference (WMD) with 95% CI. The heterogeneity between studies was tested using the Q statistic (
). Statistical heterogeneity was assessed by the measure of the I2. An I2 measurement greater than 50% was taken to indicate a substantial heterogeneity (The Cochrane Collaboration, 2011). If substantial heterogeneity was detected, a random effect model was used instead of a fixed effect model.
Results
Included studies
One study, identified in the clinical trial registry, had to be excluded, after contact with the principal investigator, as the study is still ongoing (
Randomized, open trial comparing a modified double-lumen needle follicular flushing system with a single-lumen aspiration needle in IVF patients with poor ovarian response.
) numbers between flushing and no flushing; in all three trials the necessary sample size was determined a-priori. In the control group of all trials, a conventional single lumen needle was used. All trials were conducted between 2013 and 2017.
Grading of studies and publication bias
All three included studies reported adequately on primary outcomes (mean number of COCs, mean number of MII oocytes), which were defined by all studies a priori. All three studies reported on live birth rate, rate of embryo transfer and patients having at least one oocyte, whereas only two studies analysed mean number of embryos. By necessity, physicians carrying out the follicular puncture were not blinded in any of the studies. A bias owing to allocation concealment was sufficiently avoided by all included studies by using opaque, sealed envelopes, which were opened before follicular puncture. A valid randomization was performed either by using computer-generated sequences (
Randomized, open trial comparing a modified double-lumen needle follicular flushing system with a single-lumen aspiration needle in IVF patients with poor ovarian response.
Randomized, open trial comparing a modified double-lumen needle follicular flushing system with a single-lumen aspiration needle in IVF patients with poor ovarian response.
Randomized, open trial comparing a modified double-lumen needle follicular flushing system with a single-lumen aspiration needle in IVF patients with poor ovarian response.
None of the trials reported a difference in baseline characteristics, such as age, body weight or total gonadotrophins used. No trial reported differences in the use of different protocols for ovarian stimulation between groups. In two studies, hCG was used for triggering final oocyte maturation (
Randomized, open trial comparing a modified double-lumen needle follicular flushing system with a single-lumen aspiration needle in IVF patients with poor ovarian response.
Comparison of the primary outcome mean number of COCs
Follicular flushing did not increase the mean number of COCs on follicular puncture (WMD −0.45 COC, 95% CI −1.14 to 0.25, I2 = 70%; P = 0.21; three RCT, n = 210 (Figure 2). Standard deviation (SD) was not reported for mean number of COCs in one of the RCT (
Figure 2(a) Mean number of cumulus-oocyte-complexes (COC); (b) mean number of metaphase II (MII) oocytes; (c) proportion of patients having at least one COC; (d) mean number of embryos.
Follicular flushing did not increase mean number of MII oocytes (WMD 0.09 MII oocytes, CI −0.40 to 0.59, I2 = 64%; P = 0.71; three RCT, n = 210). The proportion of patients having at least one oocyte retrieved at follicular aspiration was not different between groups (RD 0.01%, 95% CI −0.07 to 0.09, I2 = 0%; P = 1.00; three RCT, n = 210) (Figure 2).
), which did not differ between groups (WMD −0.41 embryos, 95% CI −1.29 −0.47, I2 = 90%; P = 0.36; two RCT, n = 130 patients) (Figure 2).
The proportion of randomized patients achieving embryo transfer was not different between groups (risk difference −0.07%, 95% CI −0.19 to 0.06, I2 = 0%; P = 0.29; three RCT, n = 210) (Figure 3).
Figure 3(a) Proportion of patients achieving embryo transfer; (b) Clinical pregnancy rate; (c) Live birth rate; (d) Mean procedure time of follicle aspiration.
The clinical pregnancy rate was reported by all three trials and was not different between groups (RR 0.63, 95% CI 0.14 to 2.79; I2 = 62%; P = 0.54; three RCT, n = 210), as was live birth rate (RR 0.81, 95% CI 0.42 to 1.58, I2 = 36%; P = 0.54; three RCT, n = 210) (Figure 3).
Procedure duration
All three studies reported prolonged procedure time for follicular aspiration with flushing (WMD 1.63 minutes, 95% CI 0.7 to 2.57, I2 = 81%; P = 0.0006; three RCT, n = 210) (Figure 3).
Discussion
The present meta-analysis collates data from three randomized controlled trials, which were all conducted to test for an increase in COCs or MII oocytes to be retrieved at oocyte retrieval. None of the individual studies found such an increase and, accordingly, the combined analysis shows no benefit of flushing for increasing the oocyte yield in poor responders. This conclusion is robust, as the combined sample size is large enough to test for clinically relevant differences in oocyte numbers.
Poor ovarian response was defined differently within each study. In 2011, the ESHRE Bologna criteria described poor ovarian response as the presence of two out of three of the following criteria: advanced maternal age, previous poor ovarian response (three oocytes or fewer after conventional stimulation), an abnormal ovarian reserve test (AFC 5–7, AMH <0.5 −1.1 ng/ml), (
ESHRE working group on Poor Ovarian Response Definition ESHRE consensus on the definition of ‘poor response’ to ovarian stimulation for in vitro fertilization: the Bologna criteria.
Randomized, open trial comparing a modified double-lumen needle follicular flushing system with a single-lumen aspiration needle in IVF patients with poor ovarian response.
) combined sonographic criteria with AMH levels. The present systematic review collated studies testing a rather mechanistic intervention in women likely to fall under the Bologna poor response definition (based on the number of COCs in Figure 2), and we therefore think that the study findings may well be extrapolated to a Bologna criteria poor response population. The latest meta-analysis on flushing by
already included a subgroup analysis for patients with low ovarian response and found no increase of oocyte retrieval for flushing. This analysis, however, contained data from only one trial with 30 patients (
). The present meta-analysis evaluates further data from three RCT, which became available between 2013 and 2017, allowing the analysis of a combined sample of 210 randomized patients.
On the basis of the available evidence, it is plausible that flushing will not benefit even natural cycle IVF patients with a single dominant follicle. The failure of the retrieval of a COC from a given follicle is most likely a problem of the maturity and healthiness of the oocyte, cumulus cells and mural granulosa cells, respectively, rather than a problem of a technically insufficient aspiration procedure.
For the outcome pregnancy rate and live birth rate, the combined sample size is still too small, and the studies show too much heterogeneity in outcome for confident conclusions to be drawn. The only conceivable mechanism of positive action of follicular flushing, however, is the increase in COC numbers, which should enable more patients to undergo embryo transfer with more embryos. As such an increase in COC number is not present, further studies are not warranted.
One RCT described a negative influence of flushing on embryos and a decreased chance of pregnancy and live birth (
), which was not supported by the other two studies. Of note, the control group pregnancy rate was extremely high in this study, and accordingly, the conclusion of a reduced chance of pregnancy with flushing may well represent a chance finding in a small study. Still, a negative effect of flushing should not be ruled out. An increase in intra-follicular pressure, an increased procedure time, a change in the paracrine milieu of the oocytes by dilution with the flushing medium after retrieval or the presence of spilled flushing medium in the pelvis may all be detrimental factors.
Adverse events were not systematically assessed in the studies included herein. Only
Randomized, open trial comparing a modified double-lumen needle follicular flushing system with a single-lumen aspiration needle in IVF patients with poor ovarian response.
reported no difference in anxiety, depression and pain using the DASS 21 score and a visual pain scale immediately after the procedure.
For the quantitative outcome procedure duration, the meta-analysis of the trials is homogenous in effect direction and shows a significantly prolonged procedure time for follicular aspiration when using flushing. This finding comes as no surprise, and confirms previous meta-analyses for normal response IVF patients (
). The mean increase is only 2 min, which is likely of no clinical relevance for the patient in anaesthesia time. Flushing, however, increases the effort for the team (preparation and equilibration of the flushing medium), as well as the financial burden, as double lumen needles and flushing media are costly.
A limitation to the external validity of the present systematic review is that all studies did not specifically assess poor response patients as defined by the ESHRE Bologna criteria. Accordingly, the conclusions to be drawn from this systematic review should be restricted to IVF patients undergoing ovarian stimulation and presenting with few follicles on the day of hCG administration. A further limitation is the small number of available studies and the presence of heterogeneity for some outcomes. For the primary, quantitative outcome, oocyte number, the studies are, however, sufficiently homogenous, and the sample size large enough for relevant differences to be excluded. This; however, is not the case for all other outcomes, especially dichotomous outcomes such as number of patients achieving embryo transfer or pregnancy rate.
In conclusion, existing evidence discourages the use of follicular flushing in poor responders.
Appendix. Supplementary material
The following is the supplementary data to this article:
Addition of growth hormone to gonadotrophins in ovarian stimulation of poor responders treated by in-vitro fertilization: a systematic review and meta-analysis.
A new follicle aspiration needle set is equally effective and as well tolerated as the standard needle when used in a prospective randomized trial in a large in vitro fertilization program.
Randomized, open trial comparing a modified double-lumen needle follicular flushing system with a single-lumen aspiration needle in IVF patients with poor ovarian response.
Kay Neumann is a physician. He is currently undergoing his subspeciality training at the Department of Gynaecological Endocrinology and Reproductive Medicine of the University of Luebeck, Germany. He has previously done research, during a fellowship at the Dept. of Molecular Biophysics and Physiology of the Rush University in Chicago, USA in 2010–2011. He received his medical degree and did his doctorate thesis at the University of Goettingen, Germany in 2013.
Key message
A positive effect of flushing in poor response IVF patients could not be observed in published research.
Article info
Publication history
Published online: December 29, 2017
Accepted:
December 15,
2017
Received in revised form:
October 11,
2017
Received:
April 28,
2017
Declaration: GG received personal fees and non-financial support from MSD, Ferring, Merck-Serono, Finox, TEVA, IBSA, Glycotope, as well as personal fees from VitroLife, NMC Healthcare LLC, ReprodWissen LLC, and ZIVA LLC.
30709-5&id=rbmo1873-fig-0001.jpg){kind=link}
30709-5&id=rbmo1873-fig-0002.jpg){kind=link}
30709-5&id=rbmo1873-fig-0003.jpg){kind=link}