Abstract
Is a modified double-lumen aspiration needle system with follicular flushing able to increase the mean oocyte yield by at least one in poor response IVF patients as compared to single-lumen needle aspiration without flushing?
Follicular flushing with the modified flushing system did not increase the number of oocytes, but increased the procedure duration.
Most studies on follicular flushing were performed with conventional double-lumen needles in patients who were normal responders. Overall, these studies indicated no benefit of follicular flushing.
Prospective, single-centre, randomized, controlled, open, superiority trial comparing the 17 G Steiner-Tan Needle® flushing system with a standard 17 G single-lumen aspiration needle (Gynetics®); time frame February 2015–March 2016.
Eighty IVF patients, 18–45 years, BMI >18 kg/m2 to <35 kg/m2, presenting with ≤ five follicles >10 mm in both ovaries at the end of the follicular phase were randomized to either aspirating and flushing each follicle 3× with the Steiner-Tan-Needle® automated flushing system (n = 40) or a conventional single-lumen needle aspiration (n = 40). Primary outcome was the number of cumulus-oocyte-complexes (COCs). Procedure duration, burden (Depression Anxiety and Stress Scale; DASS-21) and post-procedure pain were also assessed.
Flushing was not superior with a mean (SD) number of COCs of 2.4 (2.0) and 3.1 (2.3) in the Steiner-Tan Needle® and in the Gynectics® group, respectively (mean difference −0.7, 95% CI: 0.3 to −1.6; P = 0.27). Likewise no differences were observed in metaphase II oocytes, two pronuclear oocytes, number of patients having an embryo transfer and DASS 21 scores. The procedure duration was significantly 2-fold increased.
Testing for differences in the number of patients achieving an embryo transfer or differences in pregnancy rate would require a much larger sample size.
The use of follicular flushing is unlikely to benefit the prognosis of patients with poor ovarian response.
The Steiner-Tan Needles® and the flushing system were provided for free by the manufacturer. K.v.H. has received personal fees from Finox and non-financial support from Merck-Serono; M.D. has received personal fees from Finox and non-financial support from Merck-Serono. A.S.-M. has received personal fees and non-financial support from M.D., Ferring, Merck-Serono, Finox, TEVA. G.G. has received personal fees and non-financial support from M.D., Ferring, Merck-Serono, Finox, TEVA, IBSA, Glycotope, as well as personal fees from VitroLife, NMC Healthcare LLC, ReprodWissen LLC and ZIVA LLC.
NCT 02365350 (clinicaltrials.gov).
Sixth of February 2015.
Ninth of February 2015.
Introduction
The standard IVF procedure involves transvaginal retrieval of multiple oocytes from the ovary. Retrieval of an oocyte from a follicle relies on the release of the cumulus-oocyte-complex (COC) from the follicular wall into the follicular lumen, thereby allowing aspiration by a puncture needle. Flushing the follicle with multiple aspirations from the same follicle is used as a means to increase the ratio of COC per aspirated follicle. At least three recent meta-analyses have, however, concluded that flushing does not increase the oocyte yield (Wongtra-Ngan et al., 2010; Levy et al., 2012; Roque et al., 2012). Yet, all studies but one were performed in normo-responder patients and all previous studies suffer from the use of a potentially less than optimal flushing system, e.g. conventional double-lumen needles with hand held flushing by a syringe.
Due to the design of conventional double-lumen needles with vast dead space in the needle, the flow restriction of the flushing medium is considerable and, accordingly, only a low flow rate of flushing medium can be achieved at the tip of the needle despite the application of high pressure. The Steiner-Tan Needle® is a purpose-built new double-lumen needle that comes with a semi-automated flushing pump. The aim of the present study was to test the new needle system in a randomized clinical trial in poor responder patients. The primary interests were if the mean number of COCs can be increased and if a larger proportion of poor response patients indeed have at least one mature oocyte retrieved and subsequent embryo transfer. Secondary interests were procedure time and patient burden.
Materials and Methods
The present study was a prospective, single-centre, randomized, controlled, open, superiority trial comparing a standard 17 G single-lumen aspiration needle (Gynetics®) with the Steiner-Tan Needle® system for follicular flushing in patients with poor ovarian response. The study was conducted in the time frame February 2015–March 2016, approved by the Independent Ethics Committee (University of Luebeck, trial registration number 14–244), and was prospectively registered (NCT 02365350). The training period of the staff with the Steiner-Tan Needle® and flushing system went from November 2014 to February 2015. Patients with a BMI >18 kg/m2 and <35 kg/m2, between the age of 18 and 45, with an indication for IVF or ICSI, presenting with a total ≤ five follicles >10 mm in both ovaries combined at the end of the follicular phase of the treatment cycle were eligible for inclusion and were listed in the screening log. Exclusion criteria were one ovary absent (e.g. after ovarectomy) or one/both ovaries foreseeably difficult to be punctured (e.g. heterotopic site because of adhesions). Final oocyte maturation was induced by 5.000 IU urinary hCG as soon as the leading follicle reached a mean diameter of 18 mm or the day thereafter and oocyte-pick-up was scheduled 34–38 h thereafter. On the day of the decision to trigger final oocyte maturation, patients were randomized to either study group (Steiner-Tan Needle®) or control group (Gynetics®). Randomization was performed by one of the doctors doing the sonographic monitoring by opening a sealed, opaque and sequentially numbered envelope containing the allocation of the patient. The random sequence was software-generated and was produced by one of the authors (G.G.). Blocks of four were used. Randomized patients were sequentially logged in the randomization log.
The Steiner-Tan Needle® (enlarged view in lower panel) and flushing system consisting of the pump with warmed flushing medium and a vial warmer (to be used in conjunction with an aspirator pump and two foot pedals, one for activating the aspirator, one for activating the flushing device; not in picture).
In both groups, the number and sizes of follicles were assessed and recorded just before starting retrieval. After oocyte-pick, the total number of COCs was assessed and ICSI was performed as per standard operating procedure. The primary outcome measure was the mean number of COCs, with a separate analysis planned for the intention-to-treat population (e.g. population randomized; abbr.: ITT) and per protocol population (e.g. population indeed undergoing oocyte pick-up; abbr.: PP). Secondary outcomes were the mean oocyte retrieval rate (number of COCs divided by the number of aspirated follicles), the mean number of metaphase II (MII) oocytes, mean number of fertilized oocytes, the proportion of patients undergoing embryo transfer, the ongoing pregnancy rate, the mean duration of the procedure (insertion of the needle to retraction of the needle), the mean Depression Anxiety and Stress Scale Score (DASS-21) (Henry and Crawford, 2005) and a pain assessment by visual analogue scale (Carlsson, 1983) filled out within 2 h after the procedure.
The study hypothesis was that the new flushing system would increase the mean number of COCs retrieved by at least one from an expected control group mean of three. Group sample sizes of 37 and 37 patients achieve 81% power (alpha = 0.05) to detect a difference of 1.0 COC between the null hypothesis that both group means were 3.0 COCs and the alternative hypothesis that the mean number of COCs in the study group was 4.0 with estimated group standard deviations of 1.5 using a two-sided two-sample t-test. To account for an estimated 7.5% drop-out, it was planned to randomize 80 patients in total.
Chi-square, t-test and Mann–Whitney U-test were used, as appropriate, with a P-value of ≤0.05 considered significant. No correction for multiplicity of testing was applied. To account for the fact that three different operators performed the oocyte-pick-up procedure, a General Linear Model (GLM) was constructed with treatment group and physician entered as co-variates for the primary outcome number of COCs retrieved. The software used was SPSS v.20.
Results
Eighty patients were randomized (ITT), with 39 in each group being analysable (PP) (drop-out of 1 patient per group because of incorrect hCG administration) (Supplementary data, Fig. S1).
Table I shows demographics and outcomes. No increase in the mean COC number could be observed in the flushing group. Likewise, no statistical significant differences were found for MII oocytes (2.1 vs. 1.6, P = 0.14) and two pronuclear (2PN) oocytes (1.0 vs. 1.5, P = 0.16) in the study and control group, respectively. In the flushing and non-flushing group, 82% vs. 90%, 72% vs. 85%, 56% vs. 74% and 56% vs. 74% of patients had retrieval of at least one COC, at least one MII, at least one 2PN oocyte and transfer of at least one embryo (P = 0.67). The average duration of oocyte-pick-up was 3.9 (2.2) in the flushing group versus 1.9 (1.6) min in the non-flushing group (mean difference 1.9, 95% CI of the difference: 1.1 to 2.8; P < 0.001). The average score of the three DASS-21 dimensions were highly similar, as was the assessment of post-procedure pain. In the study and control group, respectively, 3/40 (7.5% per ITT) and 1/40 (2.5% per ITT) ongoing pregnancies or live births were recorded.
Demographics and treatment characteristics. Values are means (SDs) unless stated otherwise.
| Steiner-Tan Needle® system (flushing) | Single-lumen Gynetics needle (no flushing) | Difference, 95% confidence interval | P-value | |
|---|---|---|---|---|
| Age (years) | 37.5 (4.3) | 38.7 (5.0) | ||
| Weight (kg) | 68.7 (10.8) | 66.5 (11.2) | ||
| BMI (kg/m2) | 24.4 (3.9) | 23.0 (3.0) | ||
| Previous poor response, n (%) | 27 (76.5) | 24 (60.0) | ||
| Gonadotropin dose (IU) | 2143 (623) | 2154 (802) | ||
| Duration of stimulation (days) | 9.3 (1.8) | 9.5 (2.4) | ||
| Antagonist protocol, n (%) | 39 (97.5) | 35 (87.5) | ||
| Peak estradiol (pg/ml) | 869 (444) | 809 (401) | ||
| Peak progesterone (ng/ml) | 1.0 (0.5) | 1.3 (0.5) | ||
| General anaesthesia, n (%) | 31 (77) | 34 (85) | ||
| Follicles >10 mm on day of hCG | 3.5 (1.4) | 3.3 (1.4) | ||
| Follicles >10 mm on day of oocyte-pick-up | 4.2 (2.2) | 4.1 (1.8) | ||
| COCs | 2.4 (2.0) | 3.1 (2.3) | −0.7 0.3 to −1.6 | 0.27a |
| COC/punctured follicle | 0.6 (0.4) | 0.7 (0.5) | −0.2 0.03 to −0.4 | 0.09a |
| Zero COC retrieved, n (%) | 7 (17.5) | 5 (12.5) | 5% −11.2 to 21.1 | 0.75 |
| Steiner-Tan Needle® system (flushing) | Single-lumen Gynetics needle (no flushing) | Difference, 95% confidence interval | P-value | |
|---|---|---|---|---|
| Age (years) | 37.5 (4.3) | 38.7 (5.0) | ||
| Weight (kg) | 68.7 (10.8) | 66.5 (11.2) | ||
| BMI (kg/m2) | 24.4 (3.9) | 23.0 (3.0) | ||
| Previous poor response, n (%) | 27 (76.5) | 24 (60.0) | ||
| Gonadotropin dose (IU) | 2143 (623) | 2154 (802) | ||
| Duration of stimulation (days) | 9.3 (1.8) | 9.5 (2.4) | ||
| Antagonist protocol, n (%) | 39 (97.5) | 35 (87.5) | ||
| Peak estradiol (pg/ml) | 869 (444) | 809 (401) | ||
| Peak progesterone (ng/ml) | 1.0 (0.5) | 1.3 (0.5) | ||
| General anaesthesia, n (%) | 31 (77) | 34 (85) | ||
| Follicles >10 mm on day of hCG | 3.5 (1.4) | 3.3 (1.4) | ||
| Follicles >10 mm on day of oocyte-pick-up | 4.2 (2.2) | 4.1 (1.8) | ||
| COCs | 2.4 (2.0) | 3.1 (2.3) | −0.7 0.3 to −1.6 | 0.27a |
| COC/punctured follicle | 0.6 (0.4) | 0.7 (0.5) | −0.2 0.03 to −0.4 | 0.09a |
| Zero COC retrieved, n (%) | 7 (17.5) | 5 (12.5) | 5% −11.2 to 21.1 | 0.75 |
aGLM accounting for treatment group and physician performing the procedure. COC, cumulus-oocyte-complex.
Demographics and treatment characteristics. Values are means (SDs) unless stated otherwise.
| Steiner-Tan Needle® system (flushing) | Single-lumen Gynetics needle (no flushing) | Difference, 95% confidence interval | P-value | |
|---|---|---|---|---|
| Age (years) | 37.5 (4.3) | 38.7 (5.0) | ||
| Weight (kg) | 68.7 (10.8) | 66.5 (11.2) | ||
| BMI (kg/m2) | 24.4 (3.9) | 23.0 (3.0) | ||
| Previous poor response, n (%) | 27 (76.5) | 24 (60.0) | ||
| Gonadotropin dose (IU) | 2143 (623) | 2154 (802) | ||
| Duration of stimulation (days) | 9.3 (1.8) | 9.5 (2.4) | ||
| Antagonist protocol, n (%) | 39 (97.5) | 35 (87.5) | ||
| Peak estradiol (pg/ml) | 869 (444) | 809 (401) | ||
| Peak progesterone (ng/ml) | 1.0 (0.5) | 1.3 (0.5) | ||
| General anaesthesia, n (%) | 31 (77) | 34 (85) | ||
| Follicles >10 mm on day of hCG | 3.5 (1.4) | 3.3 (1.4) | ||
| Follicles >10 mm on day of oocyte-pick-up | 4.2 (2.2) | 4.1 (1.8) | ||
| COCs | 2.4 (2.0) | 3.1 (2.3) | −0.7 0.3 to −1.6 | 0.27a |
| COC/punctured follicle | 0.6 (0.4) | 0.7 (0.5) | −0.2 0.03 to −0.4 | 0.09a |
| Zero COC retrieved, n (%) | 7 (17.5) | 5 (12.5) | 5% −11.2 to 21.1 | 0.75 |
| Steiner-Tan Needle® system (flushing) | Single-lumen Gynetics needle (no flushing) | Difference, 95% confidence interval | P-value | |
|---|---|---|---|---|
| Age (years) | 37.5 (4.3) | 38.7 (5.0) | ||
| Weight (kg) | 68.7 (10.8) | 66.5 (11.2) | ||
| BMI (kg/m2) | 24.4 (3.9) | 23.0 (3.0) | ||
| Previous poor response, n (%) | 27 (76.5) | 24 (60.0) | ||
| Gonadotropin dose (IU) | 2143 (623) | 2154 (802) | ||
| Duration of stimulation (days) | 9.3 (1.8) | 9.5 (2.4) | ||
| Antagonist protocol, n (%) | 39 (97.5) | 35 (87.5) | ||
| Peak estradiol (pg/ml) | 869 (444) | 809 (401) | ||
| Peak progesterone (ng/ml) | 1.0 (0.5) | 1.3 (0.5) | ||
| General anaesthesia, n (%) | 31 (77) | 34 (85) | ||
| Follicles >10 mm on day of hCG | 3.5 (1.4) | 3.3 (1.4) | ||
| Follicles >10 mm on day of oocyte-pick-up | 4.2 (2.2) | 4.1 (1.8) | ||
| COCs | 2.4 (2.0) | 3.1 (2.3) | −0.7 0.3 to −1.6 | 0.27a |
| COC/punctured follicle | 0.6 (0.4) | 0.7 (0.5) | −0.2 0.03 to −0.4 | 0.09a |
| Zero COC retrieved, n (%) | 7 (17.5) | 5 (12.5) | 5% −11.2 to 21.1 | 0.75 |
aGLM accounting for treatment group and physician performing the procedure. COC, cumulus-oocyte-complex.
Discussion
Herein it is shown that flushing each follicle three-times by a modified double-lumen needle failed to increase the number of COCs recovered in poor ovarian response IVF patients. If flushing does not increase the number of oocytes, despite the use of an elaborate needle and flushing system, a larger efficiency trial on flushing is redundant in poor responders, since an increase in the number of oocytes is the only conceivable mechanism of action of increasing the pregnancy likelihood. Accordingly, the present study was designed on the primary outcome number of oocytes, not pregnancy rates.
Of note, it was not our intention to compare the Steiner-Tan Needle® with a conventional double-lumen needle, since we found no a priori convincing evidence that flushing is beneficial at all in the group of poor responders. Accordingly, in a first step, the concept of flushing with an optimized system was tested. Still we think that the results of the present study may be extrapolated to double-lumen needles and flushing in general. The failure to retrieve more oocytes with flushing in the present trial is most likely not a problem of a still suboptimal needle and flushing technique, but rather a consequence of the physiology of the ovarian follicle around the time of ovulation.
Our findings are in line with a recent randomized study in poor responders comparing conventional single and double-lumen needles for follicular aspiration in patients with ≤4 follicles ≥12 mm on day of hCG administration (Mok-Lin et al., 2013), which found that flushing did not increase the oocyte yield and observed a similar trend towards less COCs with flushing. A retrospective, two-armed comparison of the Steiner-Tan Needle® with a conventional non-flushing needle in IVM cycles is also concordant with the findings of the present study: no increase in the number of COCs was documented (Rose and Laky, 2013).
The procedure duration was significantly longer with flushing in the present trial and accordingly, anaesthesia time was also longer. Of note, setting-up the flushing system and preparing the flushing medium also increases effort on the side of the team, though this was not recorded herein.
In conclusion, flushing follicles 3× does not increase the number of oocytes in poor response IVF patients.
Supplementary data
Supplementary data are available at Human Reproduction online.
Authors’ roles
K.v.H. participated in designing the study, recruited subjects, collected and analyzed the data and wrote the final version of the manuscript. M.D. recruited and treated subjects, participated in the data analyses and data interpretation and revised the manuscript. A.S.-M. recruited and treated subjects, participated in the data analyses and data interpretation and revised the manuscript. G.G. designed the study, treated subjects, reviewed the analyses and revised the manuscript. The final manuscript and order of authorship has been approved by all authors.
Funding
The Steiner-Tan Needles® and flushing system were provided for free by the manufacturer.
Conflict of interest
K.v.H. has received personal fees from Finox and non-financial support from Merck-Serono; M.D. has received personal fees from Finox and non-financial support from Merck-Serono. A.S.-M. has received personal fees and non-financial support from MSD., Ferring, Merck-Serono, Finox, TEVA. G.G. has 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.
Acknowledgements
We acknowledge Dr Hans-Peter Steiner, Graz, Austria, inventor and patent-holder of the Steiner-Tan Needle® for training the team in the use of the needle and flushing system and support in all aspects.
