Comparison between day-2 embryos obtained either from ICSI or resulting from short insemination IVF: influence of maternal age*

Transcription

1 Human Reproduction vol.15 no.8 pp , 2000 Comparison between day-2 embryos obtained either from ICSI or resulting from short insemination IVF: influence of maternal age* Yves Ménézo 1,3 and Yona Barak 2 identified (Ménézo and Dale, 1994; Ménézo and Janny, 1997). 1 A recent study (Shoukir et al., 1998) noted that ICSI embryos Laboratoire Marcel Merieux, 1 Rue Laborde, BRON, France and 2 In Vitro Fertilization Unit, Herzliya Medical Center, have a lower developmental potential as measured by blastocyst 7 Ramot-Yam Street, Herzliya-on-Sea 46851, Israel formation. Jones et al. (1998) also confirmed the negative 3 impact of male factors on blastocyst formation. Gardner et al. To whom correspondence should be addressed at: Laboratoire Marcel Merieux, 1 Rue Laborde, Bron France, France. (1998) were the only ones who could not detect this observa- yves.menezo@insa-lyon.fr tion. One study (Gardner et al., 1998), however, did not identify a negative effect of ICSI on blastocyst formation. Short incubation time prevents deleterious effects of In order to understand properly the effect of the ICSI cumulus cell degeneration and excess spermatozoa in IVF technique on the developmental potential of ICSI compared embryos. We performed a short incubation (3 h) protocol with IVF embryos, we performed the short insemination time in 328 IVF cycles, in order to compare the developmental protocol (Gianaroli et al., 1996; Quinn et al., 1998) in potential of regular IVF embryos with those originating regular IVF. In fact, in terms of culture conditions, the short from 316 cycles entered our intracytoplasmic sperm injecinsemination protocol is similar to that of ICSI; the oocytes tion (ICSI) programme over the same period. Embryo are rapidly denuded, as after 3 h most of the cumulus cells transfers were performed in all patients on day 2. The are detached and removed by rinsing. They are no longer mean number of embryos transferred was 1.92 for the submitted to the possible deleterious effects of degenerating ICSI group and 1.73 for the IVF group (P < 0.007). This cumulus cells, or spermatozoa. We believe, therefore, that a was related only to the wishes of patients. However, the comparison such as this will lead to a better understanding of policy of the centre is to transfer a low number of embryos the paternal effect on the developmental potential of the in young patients in order to avoid multiple pregnancies. embryo. In addition, in order to discriminate a possible maternal All spare embryos were permitted to grow to the blastocyst effect, the patients were ranked by partner ages: 30, stage for freezing. Shortening incubation time did not and 35 years, as the age of 30 was shown as a shifting decrease fertilization rates. In our overall population, no point in fertility (Piette et al., 1990; Janny and Ménézo, 1996). difference was observed in the implantation rates per embryo for IVF (19%) or for ICSI (20%). An age-related decrease in embryo production was observed for both Materials and methods groups of patients (P < 0.01 for ICSI and P < for The present study, undertaken from April 1998 to the end of December IVF). The age-related decrease in embryo implantation 1998, was performed on patients who entered the IVF programme at was only significant for the IVF group (P < 0.03 for patients the Institut Rhonalpin, Laboratoire Marcel Merieux in Bron (France). <30 and >35 years of age). A significant overall decrease A total of 644 cycles was assessed, of which 316 were ICSI. In in blastocyst formation was observed for spare embryos 328 IVF cycles a short insemination procedure took place during after ICSI versus IVF (34.2 versus 43.8%; P < 0.05). The conventional IVF. IVF was performed on the first attempt when significance of this observation is discussed spermatozoa with at least 50% motility could be observed Key words: embryo implantation/icsi/ivf/short inseminamotile spermatozoa was in the total ejaculate; (ii) when in the specimen. ICSI was performed: (i) when the concentration of tion time the percentage of abnormal forms was over 90% according to the World Health Organization criteria (WHO, 1992); (iii) after one previous fertilization failure with no spermatozoa attached to the Introduction zona pellucida in one IVF cycle, in which at least three metaphase II (MII) oocytes had been retrieved. The paternal influence in embryogenesis has gained substantial This proportion of ICSI/IVF corresponds to our regular activity. attention since the appearance of techniques such as intracyto- The patient population in the ICSI group for ages 30, and plasmic sperm injection (ICSI) (Palermo et al., 1992). Treat- 35 years comprised 94, 140 and 82 cycles respectively totalling ment of men with poor quality spermatozoa by using ICSI has 316, and in the IVF group there were 61, 143, and 124 cycles improved our knowledge of sperm biology and several points respectively, totalling 328 in all. of negative impact, both genetic and epigenetic, have been The stimulation protocol was identical to that already described (Janny and Ménézo, 1994). Briefly, it included the use of gonadotrophin-releasing hormone analogues (GnRHa) in semi-long and/or short *Presented in part at the 1999 American Society for Reproductive protocols, followed by stimulation with urinary follicle stimulating Medicine, September 25 30, 1999, Toronto. hormone (ufsh), or recombinant FSH (rfsh). Ovulation was trig European Society of Human Reproduction and Embryology

2 Short insemination time in IVF: quality of IVF versus ICSI embryos gered between day 11 and day 13 of stimulation, by 9000 IU of human chorionic gonadotrophin (HCG). The oestradiol concentration at the time of ovulation was ~150 pg/ml/follicle. IVF procedure and culture media The follicle-rinsing medium was prepared in the laboratory, i.e. Earle s medium (Sigma, St Quentin Fallavier, France) supplemented with 0.4% of human serum albumin (LFB, 78000, Les ULIS, France) and 80 mg/l of gentamycin. The same medium was used for washing the spermatozoa which were processed by a two-layer pure-sperm gradient (IVF Science, Gothenburg, Sweden) for ICSI and regular IVF. Insemination of oocytes took place in 0.5 ml Universal-IVF (U-IVF, Medicult, Copenhagen, Denmark), in a 4-well dish (Nalge Nunc, Roskilde, Denmark). Three hours later, inseminated oocytes were placed in a fresh droplet of U-IVF medium. At this point, the spermatozoa and 90% of the cumulus were naturally removed. Only a few corona cells still surrounded the oocytes. Eighteen hours later, the oocytes were examined for the presence of pronuclei. As the oocytes were usually devoid of the majority of cumulus cells, a slight motion in a pipette totally removed the remaining corona cells around the zygotes. Observation of the fertilized oocytes was therefore easy and fast. Zygotes were then transferred into fresh droplets of Universal IVF medium (Medi-Cult). Embryo transfer took place h postinsemination. Supernumerary embryos were grown in a sequence of media similar to those previously described (Chouteau et al., 1998). Embryos that reached the blastocyst stage were frozen according to a protocol previously described (Ménézo et al., 1996; Ménézo and Veiga, 1997). All procedures were performed in droplets under oil (light oil; BDH, Poole, Dorset, UK), in a 5% CO 2 atmosphere in air. Figure 1. Mean SD number of cumulus oocyte complexes (COC) collected from ICSI and IVF patients in the various age groups. In the ICSI-treated group, the number of COC significantly decreased in the oldest group ( 35 years), in comparison with the two others (P 0.005). In the IVF group the decrease was significant between all three age groups; P ). Values shown above columns are mean number of COC. ICSI Within 1 h after ovum retrieval, oocytes were denuded in a solution of 80 IU hyaluronidase/ml BM1 medium (Ellios Biomedia, Paris, France). Sperm microinjection was performed in droplets under oil of BM1 medium containing 10% PVP, and the injected oocytes were then cultured in U-IVF (Medicult), for the subsequent 48 h. Fertilization was checked h later in the same droplets. The remaining supernumerary embryos were cultured for freezing at the blastocyst stage. As mentioned above, all the procedures were performed under oil, in a 5% CO 2 atmosphere in air. Embryo transfer All embryo transfers were performed in BM1 medium. Oocyte collection, maturation, fertilization, pregnancy rates per transfer and the implantation rates per embryo were recorded. Statistical analysis As it was not possible to determine precisely the number of MII oocytes in IVF, where they were cumulus-enclosed, in contrast to ICSI where MII oocytes had had cumulus cells removed, all calculations were made considering the number of collected cumulus oocyte complexes (COC). Statistical analysis was carried out using SPSS software. General descriptive statistics were performed for the entire population and Student s t-test was used to compare the two treatments. Quantitative data are presented as mean SD. Overall comparisons of means between the various groups were processed by analysis of variance (ANOVA), and two-by-two comparison by t-test was used to determine which groups were significantly different. For qualitative data, the overall comparisons of the distribution between the groups were processed by χ 2 analysis. When a difference Figure 2. Ratio of number of embryos formed and number of cumulus oocyte complexes (COC) in the ICSI and IVF populations of the various age groups. A higher ratio was noticed in the IVF group overall (P ). No difference was observed in the ratio when various age groups were compared within the entire population. However, a decrease in embryo formation was noticed in the older ICSI groups (age and 35 years), in comparison with the younger group (age 30 years); P 0.01; this was not observed in the IVF group. Values above columns are ratios. was found, a two-by-two comparison by χ 2 was used to determine which groups were significantly different. Results The entire population analysis was performed using ANOVA. Pairwise comparisons using the t-test were made in cases of significant F. The t-test that compared the whole population showed a higher maternal age for the IVF group: 1777

3 Y.Ménézo and Y.Barak Table I. A t-test comparison between ICSI versus IVF in the entire population ICSI IVF P No. cycles COC Embryos obtained a No. embryos transferred Pregnancy rate/transfer (%) NS Implantation rate per embryo (%) NS COC number of collected oocytes, obtained embryos and transferred embryos are presented as mean number per patient SD. Pregnancy and implantation rates are calculated per total numbers of oocytes or embryos. NS not significant. a Embryos with at least two cells on day 2. All polyploid, blocked at 2-pronuclear stage, or polyfragmented embryos were not included versus years for IVF and ICSI groups respectively (P 0001). The mean ages for the three IVF versus ICSI groups were as follows: (i) 30 years, versus ; (ii) years, versus ; (iii) 35 years, versus As shown in Figure 1, in the ICSI-treated group, the number of COC significantly decreased in the oldest group ( 35 years), in comparison with the two other age groups (P 0.005). In the IVF group the decrease was significant between all three age groups (P ). In both ICSI and IVF groups, an age-related decrease was also found in the number of embryos obtained. A higher embryo formation, per collected COC, was observed for IVF (P ; Figure 2). The mean number of embryos obtained was higher for the IVF group (P ; Table I). The mean number of embryos transferred was lower in the IVF group (1.7 versus 1.9; P 0.007), but the pregnancy rate per transfer and the implantation rates per embryos were similar (Table I). No difference was noticed in the mean number of embryos transferred, by age group. Differences were found between pregnancy rates (Figure 3) in the young group (42.6%), the year group (31.2%) and the group of 35 years (24.8%; χ ; P 0.001), in the entire population. Differences were also found in implantation rates between the youngest group and that of the 35 group (25 versus 16% respectively; P 0.02; Figure 4). No significant differences were found in pregnancy or implantation rates when the ICSI and IVF groups were compared (Figures 3 and 4). No differences in implantation rate were noticed between the various age groups within the ICSI population (Figure 4). However, a higher implantation rate was observed for the youngest group ( 30 years) compared to the oldest group ( 35 years) in the IVF patients (P 0.03). Regression and discriminate analyses gave no indication or model to predict success. Although it was not statistically significant, we observed an age-related increase in polyploidy for IVF embryos, that reached 10.7% for patients 35 years of age. A significant decrease in blastocyst formation was observed 1778 Figure 3. Pregnancy rates: a comparison of pregnancies per embryo transfer among the various age groups in the ICSI and IVF cycles. No difference was observed between the ICSI versus IVF cycles. Comparisons between the various age groups in the entire population showed differences between the groups (χ ; P 0.001), with the pregnancy rate of the youngest group being significantly higher than the others. A comparison among the various age groups within the ICSI population showed differences between the youngest group and the oldest ( 35 years) group; χ 2 6.6; P 0.01). Values above columns are pregnancy rates. Figure 4. Implantation rate: number of implanted embryos per number transferred in the various age groups of the ICSI and IVF populations. Implantation rates were similar when the ICSI and IVF groups were compared. A comparison between the various age groups in the entire population showed a higher implantation rate for the 30 years group (P 0.02). No difference was noticed between the various age groups within the ICSI population. A higher implantation rate was observed in the 30 years age group in the IVF population (P 0.03). Values shown above columns are implantation rates.

4 Short insemination time in IVF: quality of IVF versus ICSI embryos in the supernumerary ICSI versus IVF embryos [130/380 developmental potential than their corresponding IVF embryos. (34.2%) and 543/1241 (43.8%) respectively; P 0.05]. Moreover, as fast-cleaving embryos lead to the best pregnancy rates after transfer on day 2 or 3, as observed (Shoukir et al. Discussion 1997), this aspect was confirmed. Our lower embryo cleavage rate was observed for ICSI in The short insemination method did not decrease fertilization this study, which is either related to technical problems, or to and cleavage rates. In 1997, during a corresponding period these epigenetic or genetic problems. It has also been observed of time, when oocytes of 538 cycles were incubated with (Asch et al., 1995) that a considerable proportion of the sospermatozoa overnight as a routine procedure for insemination, called unfertilized eggs are fertilized, but are unable to perform our overall cleavage rate was 61%. In our current study, the even the first division. In ICSI, the quality of the injected cleavage rate was 57% in the 328 cycles, which were treated spermatozoon is randomly chosen; selection is performed on by the short insemination protocol. This confirms the results gross morphological aspect and motility (when possible). These from previous short co-incubation time studies of oocytes and sperm characteristics probably have nothing to do with the spermatozoa for fertilization (Gianaroli et al., 1996; Quinn real potential of the spermatozoon, in terms of affecting the et al., 1998). In general, for IVF embryos the polyploidy developing embryo. It is likely, therefore, that only some of observed by us, of up to 10.7% for patients 35 years of age, the injected spermatozoa have a good developmental potential is rather higher than that currently described in the literature (Hewitson et al., 1997). Moreover, Lundin et al. (1998) (8 10%) in this age range (Ho et al., 1994). Spermatozoa claimed that ICSI can increase genetic risk (Lundin et al., remained in contact with the oocytes for a shorter period, 1998). Although ICSI is performed to increase the chance of and most cumulus cells were removed while rinsing the oocytes fertilization, even with good spermatozoa, in most cases the 3 h post-insemination. Thus, the zygotes are cleaner and a real genetic capacity of the spermatozoon is unknown. It is simple motion with the denuding pipette retrieves the few admitted (Lundin et al., 1998) that, when possible, it is better remaining corona cells. This smooth removal may also have to perform regular IVF in order to try and avoid genetic risk, a positive impact on embryo quality. as again nothing is known about the injected spermatozoon, The age-related decline in ovarian production, determined suggesting that a study such as the current one to compare by the number of COC collected, has been previously observed ICSI and IVF would be useful. Obviously it would be necessary, (Piette et al., 1990; Janny and Ménézo, 1996; Figure 1). A ideally, to compare ICSI and IVF in the case of good higher implantation rate was observed as an outcome of IVF spermatozoa. This would facilitate determination of the contriwith short co-incubation time, when compared with the results bution related to the extra manipulation of the embryos due of routine IVF (e.g. overnight incubation with spermatozoa) to the injection only, in cases of ICSI, and the putative damage in the previous year during the same period of time (19% to the spindle, that follow (Hewitson et al., 1999). However, versus 13.5%) in our Unit, (Dossier FIVNAT 1999). this kind of investigation is ethically questionable. The possibility that reduced zona hardening may facilitate Thus, we assume that, in comparison to IVF, fewer embryos hatching and implantation has previously been mentioned may reach the blastocyst stage in ICSI. In addition, when (Waldenstrom et al., 1993). It is possible that the early removal embryo transfers are performed on day 2, as in the study of excess spermatozoa may avoid exposure of the embryos to presented here, a partial selection of embryos based on cleavage a high density of free radicals generated by supernumerary spermatozoa and degeneration of cumulus cells. Reactive speed and morphological appearance also occurs on that day. oxygen species induce zona hardening and tanning. Apart from Therefore, even fewer ICSI embryos will be able to reach the this aspect, they also have a deleterious effect on DNA. With blastocyst stage when compared to regular IVF. regard to these biochemical aspects, the short insemination Our data on spare embryos to some extent agree with these time is the only protocol which makes it possible to compare predictions. This simply indicates that not all the epigenetic ICSI and IVF embryos. Our data show no difference in and genetic problems mentioned earlier, and those related to pregnancy outcome between IVF and ICSI embryos in the poor sperm quality, can be overcome. The rescue is only overall population. partial for the overall ICSI embryo population. This is also in The implantation rates per embryo of ~19% were similar agreement with the observations of Shoukir et al. (1998) for ICSI and IVF embryos in our overall population. At first concerning decreased blastocyst formation for ICSI embryos, sight, this similarity is surprising: epigenetic-related problems, and it confirms the paternal effect on early embryogenesis such as centrosome function or dysfunction, (Sathananthan (Janny and Ménézo, 1994; Jones et al., 1998). Moreover, the et al., 1996; Van Blerkom, 1996) are expected to interfere apparently (but not significantly) lower implantation rate per with embryo quality. As mentioned earlier (Ménézo and Janny, embryo observed for ICSI versus regular IVF in younger 1997), ICSI may rescue embryos in some patients by avoiding patients (Figure 4) (aged 30 years) may also be linked to a delays in cell cycles through a direct sperm deposition in the paternal effect in these couples, where the maternal effect oocyte. The spare time helps to save maternal messenger is minimal. ribonucleic acid (mrna), which may partially overcome any According to the above data, we confirm that obtaining epigenetic defects. This agrees with previous observations early-stage embryos does not necessarily lead to a positive (Oehninger et al., 1996) that ICSI embryos, in cases of severe end-point, and furthermore, ICSI cannot universally cure major teratozoospermia, have higher morphological scores and higher sperm defects. 1779

5 Y.Ménézo and Y.Barak A study of blastocyst formation is currently taking place Shoukir, Y., Campana, A., Farley, T. et al. (1997) Early cleavage of in vitro fertilized human embryos to the 2-cell stage: a novel indicator of embryo for ICSI and IVF patients who enter our blastocyst transfer quality and viability. Hum. Reprod., 12, programme. This will facilitate ultimate determination of Shoukir, Y., Chardonnens, D., Campana, A. et al. (1998) Blastocyst whether blastocyst formation is impaired in ICSI patients, and development from supernumerary embryos after intracytoplasmic sperm injection: a paternal influence? Hum. Reprod., 13, whether blastocysts obtained for IVF and ICSI have the same Van Blerkom, J. (1996) Sperm centrosome dysfunction: a possible new class developmental (implantation) potential. of male factor infertility in the human. Mol. Hum. Reprod., 2, Waldenstrom, U., Hamberger, L, Nilsson, L. et al. (1993) Short time sperm egg exposure to prevent zona hardening. J. Assist. Reprod. Genet., 10, Acknowledgements World Health Organization (1992) Laboratory Manual for the Examination of Our thanks to Mrs Sara Intrater for statistical assistance. Human Semen and Semen Cervical Mucus Interaction, 3rd edn. Cambridge University Press, New York, pp References Asch, R., Simerly, C. and Ord, T. (1995) The stage at which human fertilization Received on September 30, 1999; accepted on April 26, 2000 arrests; microtubule and chromosome configuration in inseminated oocytes which failed to complete fertilization and development in humans. Hum. Reprod., 10, Chouteau, J., Girard, A., Sage, J.C. et al. (1998) the use of sequential media for blastocyst transfers. Hum. Reprod., 13 (Suppl. 4), Dossier FIVNAT (1999) Bilan de l année 98, French National Registry. Gardner, D.K., Schoolcraft, W.B., Wagley, L. et al. (1998) A prospective randomized trial of blastocyst culture and transfer in in vitro fertilization. Hum. Reprod., 13, Gianaroli L., Magli C., Ferraretti, A.P. et al. (1996) Reducing the time of sperm oocyte interaction in human in vitro fertilization improves the implantation rate Hum. Reprod., 11, Hewitson, L., Simerly, C. and Schatten, G. (1997) Inheritance defects of the sperm centrosome in humans and its possible role in male infertility. Int. J. Androl., 20 (Suppl.), Hewitson, L., Dominko, T., Takahashi, D. et al. (1999) Unique checkpoints during the first cell cycle of fertilisation after intracytoplasmic sperm injection in Rhesus monkey. Nature Med., 5, Ho, P.C., Yeung, W.S., Chan, Y.F. et al. (1994) Factors affecting the incidence of polyploidy in a human in-vitro fertilization program. Int. J. Fertil. Menopausal Stud., 39, Janny, L. and Ménézo, Y. (1994) Evidence for a strong paternal effect on human preimplantation embryo development and blastocyst formation. Mol. Reprod. Dev., 38, Janny, L. and Ménézo, Y. (1996) Maternal age effect on early human embryonic development and blastocyst Mol. Reprod. Dev., 45, Jones, G.M., Trounson, A.O., Lolatgis, N. et al. (1998) Factors affecting the success of human blastocyst develoment and pregnancy following in vitro fertilization and embryo transfer. Fertil. Steril., 70, Lundin, K., Hanson, Ch. and Hamberger, L. (1998) Are the new microfertilization techniques associated with an increased genetic risk to the offspring? Acta Obstet. Gynecol. Scand., 77, Ménézo, Y. and Dale B. (1994) Paternal contribution to successful embryogenesis. Hum. Reprod., 10, Ménézo, Y. and Janny, L. (1997) Influence of paternal factors in early embryogenesis. In Barrat, C., De Jonge, C., Mortimer, D. and Parinaud, J. (eds), Genetics of Human Male Infertility. EDK, pp Ménézo, Y. and Veiga, A. (1997) Cryopreservation of blastocyst. In vitro fertilization and assisted reproduction. Proc. Xth World Congress on In Vitro Fertilization and Assisted Reproduction, Vancouver, pp Ménézo, Y., Kauffman, R., Nicollet, B. et al. (1996) Efficiency of a simplified thawing protocol for human cocultured blastocysts. Proc. Am. Soc. For Reprod. Med., Fertil. Steril. Boston, P006. Oehninger, S., Kruger, T.F., Simon, T. et al. (1996) A comparative analysis of embryo implantation potential in patients with severe teratozoospermia undergoing in vitro fertilization with a high insemination concentration or intracytoplasmic sperm injection. Hum. Reprod., 11, Palermo, G., Joris, H., Devroy, P. et al. (1992) Pregnancies after intracytoplasmic injection of single spermatozoon into an oocyte. Lancet, 340, Piette, C., de Mouzon, J., Bachelot, A. et al. (1990) In vitro fertilization: influence of women s age on pregnacy rates. Hum. Reprod., 5, Quinn, P., Lydic, M.L., Ho, M. et al. (1998) Confirmation of the beneficial effects of brief coincubation of gametes in human in vitro fertilisation. Fertil. Steril., 69, Sathananthan, A.H., Ratnam, S.S., Ng, S.C. et al. (1996) The sperm centriole: its inheritance, replication and perpetuation in early human embryo. Hum. Reprod., 11,