|Home | About | Journals | Submit | Contact Us | Français|
To compare the effects of 2 different media on embryo morphology and development at days 2/3.
Six hundred seventy-six attempts from 512 couples were included in this prospective auto-controlled study. Sibling oocytes of all couples undergoing an IVF (n=286) or ICSI (n=390) attempt were randomly assigned to either GIII series (Vitrolife) or ISM (Medicult) media. Primary end points were fertilization and embryo morphology rates.
Fertilization rates in GIII series and ISM (IVF: 59.9 vs 62.0% and ICSI: 65.7 vs 66.8%) respectively were not different. GIII series showed an increase, compared to ISM, of early cleavage rate, (IVF: 25.8 vs 16.2% (p=0.005); ICSI: 40.8 vs 25.5% (p<0.0001), and good embryo morphology rate at day 2 [IVF: 64.6 vs 57.3% (p=0.01); ICSI: 74.2 vs 69.4 (p=0.03)] and at day 3 [IVF: 57.5 vs 49.0% (p=0.02); ICSI: 67.2 vs 61.6% (p=0.01)].
Embryo morphology at days 2/3 was significantly enhanced when the embryos were cultured in GIII series.
In vitro development of human embryos for 48 to 72 h can be achieved using a wide variety of media, ranging from simple balanced salt solutions to more complex media [1–3]. In the last decade, there has been an advance of the understanding of the pre-implantation embryo. This has led to significant changes in the composition of the culture media [4–6]. Furthermore, success rates in assisted reproduction technologies (ART) are influenced by a multitude of factors, the most important of which are patients’ age, number of oocytes retrieved per cycle, and the morphology of embryos transferred. Since the latter may be influenced by the culture conditions, clinicians and researchers have continued to search for optimal embryo culture media [4, 7–12].
High rates of multiple pregnancies following IVF or ICSI are largely ascribed and associated with risks of fetus, newborn and maternal complications. We recently confirmed that transferring fewer good morphology embryos reduced multiple pregnancies observed in ART . Thus, the implementation of suitable quality control, focused on the effectiveness of culture media to improve embryo morphology, is fundamental to the success of an IVF laboratory .
The aim of this prospective, auto controlled study was to compare two different series of media, defined as (i) G III series (G-FertTM; G1TM, G2TM) (Vitrolife, Kungsbacka, Sweden) and (ii) ISM1/2TM (Medicult, Lyon, France) to culture embryos during IVF or ICSI.
Between April 2005 and December 2006, 512 couples referred in our Infertility clinic for standard IVF or ICSI were asked and accepted to participate in this prospective auto controlled study. The trial has been submitted for acceptance and accepted to the local ethical desk. No funding was provided by any companies in this study. Patients with less than 4 oocytes at the time of ovarian puncture were excluded since this number is required to expect at least 1 embryo in each group. Patients with fertilization failure in both 2 groups or without metaphase 2 stage oocytes in one group were excluded as well. Women involved in the study were 33.2±4.7 years old, with 3.8±1.8 years of infertility and at the 2.1±1.4 IVF/ICSI attempt rank (mean ± SD). The principal IVF indications were: tubal disease (54.9%, 157 attempts), idiopathic infertility (15.7%, 45 attempts), endometriosis (10.1%, 29 attempts), polycystic ovarian syndrome (2.1%, 6 attempts) and mild male factor infertility, according to World Health Organization (WHO) criteria (WHO, 1999), (17.1%, 49 attempts). Indications for ICSI were male factor presence (84.7%, 327 attempts) according to WHO criteria (WHO, 1999) and previous IVF fertilization failure (15.3%, 59 attempts). Each patient could contribute with more than 1 cycle.
Based on prior evaluation of the ovarian function, three protocols of controlled ovarian stimulation (COS) were used. (i) The long agonist protocol used a first step of pituitary desensitization by GnRH agonist (Decapeptyl®, Ipsen-Biotech, Paris, France) followed by COS. (ii) The short protocol used a GnRH agonist administration started 2 days before COS. (iii) The antagonist protocol including COS associated to a daily GnRH antagonist injection (Cétrotide®, Serono, France) when the leader follicle have reached about 14 mm in diameter. In all cases, COS was performed with s.c. administration of recombinant human FSH (Puregon®, Organon, Saint Denis, France; Gonal-F®, Serono, Boulogne, France). (75 IU per ampoule) started with 2 ampoules per day and adapted from day 6 on an individual basis according to the ovarian response as assessed by sequential transvaginal ultrasonography and serum estradiol measurements. Human chorionic gonadotrophin (hCG; 10,000 IU, Organon, France) was administered i.m. when a consistent rise in serum oestradiol concentration was associated with the presence of ≥4 follicles ≥16 mm in diameter. Transvaginal oocyte retrieval was scheduled 36 h after hCG administration under ultrasound guidance and usual IVF techniques were performed. The luteal support was initiated on day 1 after oocyte retrieval with 400 mg/day of progesterone. In the absence of ovarian hyperstimulation syndrome (OHSS) risk, 2,500 and 1,500 units of hCG were administered 1 and 3 to 6 days after oocytes retrieval, respectively. In case of OHSS, assessed under ultrasound guidance by an ovarian diameter >10 cm and the presence of large peritoneal ascites, embryo transfers (ET) were always cancelled with cryopreservation of all embryos and women treated by corticoids and low-molecular-weight heparins.
Semen samples were collected by masturbation after 3 days of sexual abstinence and sperm were selected on a two density (90 and 45%) gradient PureSperm® (Nidacon International, Gothenburg, Sweden) centrifugation. Sperm concentration, mobility and vitality were assessed before and after selection, according to WHO guidelines .
ICSI was routinely performed on each metaphase 2 stage oocytes, as previously described by Palermo et al. . Denudation of all cumulus oocyte complexes (COCs) was done before ICSI into a ready to use 80 IU/ml hyaluronidase medium (Medicult, Lyon, France), following the instructions of the manufacturer. For an IVF attempt, COCs were inseminated with a concentration of 150,000 progressive motile sperm cells/ml, according to the French Guidelines . Injected oocytes or COCs were randomly allocated into 20 µl equilibrated drops of embryo culture media overlaid with paraffin oil (MediCult®, Lyon, France) in a 37°C and 5.5% CO2 humidified air atmosphere. Each oocyte and resulted embryo were cultured individually in one drop and daily observed. Fertilization was assessed, 18 to 20 h later, by the presence of two pronuclei (2PN) and two polar bodies, defining the fertilization rate (number of 2PN / number of metaphase 2 stage oocytes)×100). Fertilized oocytes were re-observed 25 h post insemination at day 1 to determine the early cleavage status (2PN, syngamic or cleaved). Embryos were cultured using appropriate media for 2 more days, with development monitored daily. Embryo morphology was assessed every day of culture. Embryos were scored according to criteria previously described . Briefly, embryos were classified as good morphology if they had <20% cytoplasmic fragmentations and (i) 3–5 blastomeres at day 2 and (ii) 6–9 cells at day 3. Top morphology embryos contained <20% fragmentations and 4 / 8 cells at day 2 / 3, respectively.
Sequences of embryo culture media defined the groups of the study. COCs treated by IVF were cultured either in (i) G FertTM from day 0 to day 1, G1/2TM from day 1 to day 3 (Group IVF G III series) or (ii) ISM1/2TM from day 0 to day 3 (Group IVF ISM). Metaphase 2 stage oocytes treated by ICSI were cultured either in (i) G1/2TM from day 0 to day 3 (Group ICSI G III series) or (ii) ISM1/2TM from day 0 to day 3 (Group ICSI ISM). When the number of retrieved COCs for IVF or metaphase 2 stage oocytes for ICSI was an odd number, the last was systematically allocated to the G III series group. Percentages of good and top morphology embryos at day 2 and day 3 were compared among the cohorts when the patients had obtained at least one embryo in each group.
The best morphology embryos were transferred mostly on day 3 after oocyte retrieval, irrespective of the culture media. A clinical pregnancy was assessed by a positive fetal heartbeat on transvaginal ultrasound at 5–6 weeks following ET. Clinical pregnancy (PR) and implantation (IR) rates were compared after homogeneous transfers of embryos totally resulting either from G III series or ISM culture media. Women were also separated into 2 groups, irrespective of the IVF/ICSI procedure, according to the occurrence or not of a clinical pregnancy. The female age, the mean number of total embryos transferred and the mean number of transferred embryos resulting from the G III series or ISM media groups were evaluated, and compared.
Data are presented as mean ± SD or counts and percentages unless otherwise stated. Percentages of good morphology embryos between G III series and ISM media were analyzed using conditional logistic regression models with generalised estimating equations to account for both correlation between observations from the same patient and the size of the cohort of embryos from each patient. Comparisons were performed with paired Wilcoxon rank sum test and Fisher’s exact test, with the level of significance set at 5%. Analyses were performed using R 2.2.1 statistical software (The R Foundation for Statistical Computing, Vienna, Austria).
In all, 2702 (IVF=1243; ICSI=1459) and 2431 (IVF=1153; ICSI=1278) oocytes were allocated to G III series and ISM culture media, respectively (IVF; ICSI: p<0.0001). According to the exclusion criteria of fertilization failure within the 2 groups, 113 and 110 oocytes were not analyzed in the G III series and ISM groups, respectively (Fig. 1). Data concerning the outcome of oocytes in G III series or ISM media are summarized in Table 1 for IVF and Table 2 for ICSI. Fertilization rates in G III series and ISM1 (IVF: 59.9 vs 62.0% and ICSI: 65.7 vs 66.8%) respectively were not statistically different. G III series showed a statistically significant increase, compared to ISM1, of the early cleavage rate, (IVF: 25.8 vs 16.2% (p=0.005); ICSI: 40.8 vs 25.5% (p<0.0001), and the good embryo morphology rate (i) at day 2 (IVF: 64.6 vs 57.3% (p=0.01); ICSI: 74.2 vs 69.4 (p=0.03)) and (ii) at day 3 (IVF: 57.5 vs 49.0% (p=0.02); ICSI: 67.2 vs 61.6% (p=0.03)). Furthermore, G III series showed a significant increase of top morphology embryo rate compared to ISM1 following (i) IVF (19.7 vs 18.1% at day 3 (p=0.04)) and (ii) ICSI: 41.3 vs 33.5% at day 2 (p=0.0005); 27.9 vs 18.6% at day 3: (p=0.03)).
In all, 227 clinical pregnancies were obtained, resulting from 675 IVF/ICSI attempts (PR=33.6%) and 590 ETs (PR=38.5%). Respectively 190 and 102 ETs were performed with embryos which resulted from culture with G III series and ISM media, with a PR of 44.2 and 35.3% (p=0.1) (Table 3). The number of transferred embryo per ET was significantly increased in homogeneous ETs from G III series when compared with ISM media (1.7 vs 1.5; p=0.0046) (Table 3). We observed a significant decrease of the number of embryos transferred from ISM media in the group of women who became pregnant, compared with those who did not (0.76 vs. 0.88 p=0.027) (Table 4).
Transferring fewer embryos to women increased reliance on the IVF laboratory to maximize embryo viability. Advances in our understanding of the pre-implantation embryo have led to significant changes in the philosophy of how best to culture the human embryo, and as a result, there has been a significant evolution in the culture media used in clinical IVF [18–21]. In this way, we performed a randomized prospective trial to test the ability of 2 different culture media sequences to produce viable human embryos. We observed, using a design where every patient was the control of herself eliminating every variable of confusion, that embryo morphology at days 2 and 3 was significantly enhanced when the embryos were cultured in G III series, especially during ICSI therapy, when compared with ISM1/2 media.
Since the composition of embryo culture media is mostly unknown, it is difficult to interpret such a difference. However, two differences could be observed between G III series and ISM1/2 media composition, which could explain differences showed in our study.
First is the presence of hyaluronan in G III series embryo culture media. The fluid of the reproductive tract is rich with macromolecules. One macromolecule that is present at increasing levels in the tract when the embryo is present is hyaluronan. Subsequently, it has been demonstrated that hyaluronan has an important role in embryo culture media. Hyaluronan can substitute for albumin in culture media but what is perhaps more significant for human IVF applications is that the addition of hyaluronan to a culture medium containing albumin significantly increased development but also improved the cryosurvival of embryos [22, 23]. In addition, recent randomized controlled trials in clinical IVF confirmed that elevating the levels of hyaluronan increased pregnancy and implantation rates, which was particularly evident in selected patient groups [24–26].
Second is the different form of glutamine within G III series (alanyl-glutamine) compared with ISM culture media (L-glutamine). One of the consequences of having amino acids, particularly glutamine, in the culture medium is that they breakdown spontaneously at 37°C to produce ammonium [27, 28]. The levels of ammonium that are produced by medium containing glutamine incubated at 37°C in just 24 h are significantly higher than those that have been shown to be inhibitory to embryo development [27, 28]. Significantly in media containing amino acids, this level of ammonium does not always alter blastocyst development; however, the cellular health of the blastomeres is affected significantly [27, 29]. Furthermore, at moderate levels of ammonium, fetal growth rates were affected such that resultant fetuses were smaller than control embryos [28–30]. Recently, a study found that the levels of ammonium that are produced after incubation of medium containing glutamine inhibit development of the human blastocyst in vitro . Although glutamine is quite labile in culture, it can be substituted with a stable dipeptide of glutamine such as alanyl glutamine or glycyl glutamine, which has the beneficial effects of glutamine without the problematic build-up of the toxin ammonium.
The clinical relevance of the ability for embryo culture media to produce higher good/top morphology embryos is debatable. Indeed, even if the use of sibling oocytes is the best methodology to test the embryo morphology between 2 media (rigorously comparable groups, without confounding factors from women or COS protocols) this procedure generates lots of mixed ETs from the 2 tested media, making the analysis of its clinical impact more difficult. Considering all ETs included mixed ETs where transferred embryos resulted from G III series and ISM media groups, the occurrence of a clinical pregnancy was related to a lower number of transferred embryos from ISM, compared to women who not became pregnant (Table 4). Furthermore, the mean number of transferred embryos from G III series in homogeneous ETs was statistically higher than that of embryos from ISM media (Table 3). These results could be explained by more embryos of good morphology, with a better viability, available to ET in the G III series patients. Another hypothesis to explain such results could be the higher number of oocytes allocated in G III series compared with that in ISM groups, either in IVF (p<0.0001) (Table 1) or in ICSI treatments (p<0.0001) (Table 2). Because of a similar fertilization rate into the 2 compared media, the number of subsequent transferred embryos was also higher into G III series (Table 4).
We failed in this study to show a significant difference concerning IR and PR in case of homogeneous ETs. Nevertheless, we observed a PR increase of 9% when all transferred embryos resulted from culture into G III series. A consistent 9% increase can be regarded as clinically important despite lack of statistical significance, which could be solely dependent on the sample size of the study and event rates in the groups. Then, the clinical relevance of such differences observed in our study should be confirmed only using a per-patient randomized clinical trial on an external larger series.
In conclusion, and within the limits of our culture procedures, better morphology embryos were obtained in G III series media that include both hyaluronan and a stable form of glutamine, compared with ISM media. In our opinion, our results strongly confirm that it is essential that high levels of quality control and assurance exist in the laboratory to enable any culture medium to perform to its maximum.
Capsule GIII series used to culture embryos from IVF/ICSI attempts leads to a better embryo morphology at days 2/3 than those resulting from ISM1/2 media.