Assisted Conception Unit, Birmingham Women's Hospital, Edgbaston, Birmingham B15 T2G, UK
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key words: budget impact analysis/cost needed to treat/fertilization failure/intracytoplasmic sperm injection (ICSI)/number needed to treat
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() |
---|
Recent data from Fishel and colleagues have examined the concept that ICSI should be offered to all patients needing IVF because of the significantly higher fertilization rate (Fishel et al., 2000). ICSI has become more developed as a technique and popularized to a stage of routine laboratory service. For example, in the UK the Human Fertilisation and Embryology Authority (HFEA) reported a 14% rise in the use of ICSI in 1998/1999 compared with the previous year. Almost half of fresh embryo transfers (median 47% range 1674%) in this period were a result of ICSI treatment (Human Fertilisation and Embryology Authority, 2000
). This is consistent with data from the European register where 43% of the transfers were from ICSI (EIM/ESHRE, 2001
). Clearly, the use of ICSI is rising throughout the world and in some clinics it is the exclusive treatment of choice. Therefore, the issue of whether to use ICSI for all in-vitro inseminations needs to be critically discussed.
In this debate, we examine the arguments for and against the use of ICSI in cases where IVF would normally be used (non male-factor infertility). The issues discussed are fertilization rate, total failure of fertilization, embryo damage/blastocyst formation, cost effectiveness and safety.
Fertilization rate as a measure of effectiveness
One group has performed a randomized, prospective, multi-centred trial using sibling metaphase II oocytes in 221 patients to try to address the question of whether ICSI should be advocated for all couples (Fishel et al., 2000). The patients were divided into five groups. These included: Group 1 (37 patients): idiopathic previous failed IVF, where HIC was compared with ICSI using the partnersapos; spermatozoa; Group 2 (18 patients): idiopathic previous failed IVF with HIC, where conventional IVF was compared with ICSI using donor spermatozoa; Group 3 (36 patients): patients unsuitable for conventional IVF (male infertility), where IVF using donor spermatozoa was compared with ICSI using partnerapos;s spermatozoa; Groups 4 and 5 had metaphase II oocytes that had failed to fertilize by IVF and were re-inseminated by either HIC or ICSI. The clinical bottom line for groups 2 and 3 was that conventional IVF had a fertilization rate of 65.4% and ICSI 75.6%, with an absolute treatment effect of 0.102 [95% confidence interval (CI) 0.0250.179], generating a number needed to treat (NNT) of 10 (95% CI 640). The NNT is the number of sibling MII oocytes that need to be inseminated by ICSI to derive one additional zygote, compared with IVF. Although this figure is statistically significant, in clinical terms it means that, in this group of patients where normal spermatozoa were used in IVF, for every 10 sibling MII oocytes inseminated by ICSI, only one extra zygote is produced compared with insemination by conventional IVF.
In other studies, a lack of significant difference has been demonstrated in the fertilization rates obtained with ICSI and IVF in patients with non-male factor infertility (61 versus 67%) (Yang et al., 1996) and unexplained infertility (60.4 versus 54%) (Ruiz et al., 1997
). Nevertheless, we need to be cautious in the interpretation of the results presented by some of these studies, as for example, at the design stage, power and sample size statistics were often not sufficiently emphasized, thereby exposing the results to possible random errors (Fishel et al., 2000
). In addition to this, some studies (Ruiz et al., 1997
) are not randomized controlled trials. Closer scrutiny often shows that the only control possible due to ethical considerations was the use of sibling metaphase II (MII) oocytes. Often, in these studies, no explicit descriptions were made of what happened to oocytes allocated to ICSI, but found not to be MII after denudation (i.e. was intention-to-treat analysis performed?), or how investigators who randomized and performed in-vitro inseminations were blinded to embryo grading. These potential sources of error may serve to reduce the strength of evidence presented by the authors (NHS Centre for Reviews and Dissemination, 1999
). The implication is that often, what is presented as level Ib evidence in favour of ICSI (or in fact against) may, on critical appraisal, be found to be no better than level II or III. Therefore, larger carefully conducted studies are required on non male-factor patients to confidently address the question whether ICSI does result in significantly higher fertilization rates (and embryo development) in men with apparently normal semen.
Fertilization rate: an interim outcome measure
The use of fertilization rate instead of total failure of fertilization, or indeed clinical pregnancy rate as an outcome event has drawbacks. Fertilization rate is an interim outcome measure in an IVF programme, which may have little effect on the final outcome of a fresh cycle or that of a subsequent frozen embryo transfer. It is therefore difficult to judge whether or not to advocate ICSI over IVF based on fertilization rate alone. To illustrate the point, imagine a scenario with a mean recovery of 10 MII oocytes, fertilization rate of 65% from IVF, and 75% from ICSI. In the UK, a maximum of three embryos can be replaced in a treatment cycle. Frozen embryothaw success rates of 8190% for IVF and 8891% for ICSI have been described in prospective randomized studies (Damario et al., 1999; Hu et al., 1999
). Consequently, this would allow approximately the same number of frozen embryo transfer cycles for IVF or ICSI. It would therefore seem apparent that, if decision analysis was performed based on the above scenario, an improved fertilization rate alone might not be enough to advocate ICSI over IVF per cycle of treatment.
Total failure of fertilization
From the clinician and patientsapos; points of view, the rate of total failure of fertilization is a more useful outcome measure than fertilization rate. ICSI has an advantage, which in the UK is in the form of an HFEA regulation, requiring that only MII oocytes, assessed after cleaning the oocytecumuluscomplex, be injected. There is a prescribed oocyte quality and therefore a time limit to when insemination has to be accomplished during ICSI. For conventional IVF however, metaphase I (MI), MII, or luteinized post-maturity oocytes can be used.
Several studies have attempted to demonstrate a superiority of ICSI over IVF based on failed fertilization rates. For example, in a controlled study of 70 couples with either unexplained infertility or endometriosis who had failed to respond to intrauterine insemination, Ruiz and his colleagues, (Ruiz et al., 1997), found a clear benefit of ICSI over IVF (failed fertilization rates of 0 versus 11%) despite the lack of significant difference in the fertilization rates between the two methods (60.4 versus 54%). In this study, whereas metaphase II oocytes were used for ICSI, this was not the case for the IVF group, thus exposing the results to bias. In another example, an `auto-controlledapos; study of 662 sibling MII oocytes from patients with tubal disease and normozoospermic partners, found rates of total failure of fertilization of 3.6% (95% CI = 0.4 to 12.3) for ICSI and 12.5% (95% CI = 5.224.1) for IVF (Staessen et al., 1999
). This would appear to present a real difference, although the small sample size may have introduced type II error. These potential sources of error may have served to reduce the strength of evidence presented by the authors, and when considered may mean that the superiority described in favour of ICSI over conventional IVF may be a chance occurrence.
However, whilst scientific rigour indicates that the above studies have potential errors it does look as though ICSI may be of benefit in cases of fertilization failure with conventional IVF that can be predicted before treatment. Data from Liu and Baker illustrate this point (Liu and Baker, 2000). They have reported on 160 patients who have apparently normal semen but either fail to bind to the zona pellucida (ZP) or do not acrosome react (AR) in response to the ZP (disordered ZP-induced AR) and thus fail to have successful IVF conceptions. They estimate that, in their patient population, up to a third of normozoospermic men have disordered ZP-induced AR. Interestingly, ICSI was found to overcome these defects resulting in live births (Liu and Baker, 2000
).
In contrast, in our clinic, which is a tertiary referral centre, we do not see a high incidence of fertilization failure with IVF. The total failed fertilization rate for treatments between January 1999 and July 2000 was 1.5% (95% CI 0.043.8) for ICSI and 2.1% (95% CI 1.03.8) for IVF (Table I). Our data suggest that improved but flexible clinical and laboratory protocols can reduce the incidence of total failure of fertilization, although it is possible that in our centre we do not have a high incidence of normozoospermic men with dysfunctional spermatozoa (see Liu and Baker above) which may account for our low incidence of failed fertilization.
|
Economic analysis based on live birth rates
In the reporting period between 1998 and 1999, the HFEA showed an overall live birth rate per fresh treatment cycle for ICSI of 22.6% (4082/18042) significantly higher (P = 0.01) than the rate of 21.6% (5969/27617) for IVF.
However, ICSI is substantially more expensive than IVF. In our clinic, as in many parts of the UK, the cost difference is about £600 per fresh cycle completed (Philips et al., 2000). In the UK 25% of treatments are funded by the National Health Service (NHS) (Kerr et al., 1999
). We suggest that recommending the use of ICSI for all those needing IVF is unlikely to be considered a judicious use of scarce resources. To illustrate this, we used the HFEA 1998/99 data, assuming that the advantage of ICSI was sustained even when performed on the population of couples who would have had IVF for female or unexplained factors. We used the data comparing live birth rates per fresh cycle to derive the absolute treatment effect (ATE = 0.01), and NNT (NNT = 100) (see Table II
). Cost benefit analyses show that each live birth produced by ICSI costs £2000 extra. The main cost implication however is the incremental cost effectiveness or cost needed to treat (CNT). This shows that £60 000 will be needed to gain one additional live birth when ICSI is advocated for all patients requiring IVF. Budget impact analysis (BIA) shows that CNT (£60 000) can treat an extra 29 cycles of conventional IVF.
|
It is probable that the increases in congenital abnormalities observed in some ICSI children, such as sex chromosome abnormalities, are due to the use of sub-optimal male gametes. However, the technique itself may play a role in the formation of these abnormalities. In addition, there is experimental evidence that provides caution against the widespread use of ICSI, for example, the incorporation of exogenous DNA into spermatozoa and subsequent transmission to the offspring (Perry et al., 1999; Chan et al., 2000
). Experiments in mice have shown an enhancement of a genetic defect (sperm morphology) through ICSI (Akutsu et al., 2001
). The incorporation of foreign DNA combined with the possible enhancement of defects by bypassing the natural selection mechanisms now needs rigorous experimentation in the human.
The difficult question
The arguments above do not support the routine use of ICSI in all IVF treatments. However, there is a clear group of patients, e.g. those with normal but dysfunctional spermatozoa, that have zero or significantly reduced fertilization success at IVF (Barratt and Publicover, 2001). Such patients can be successfully treated by ICSI. Whilst it is possible, using sophisticated sperm function assays such as zona binding, to predict which men may have reduced success at IVF, such assays are impossible to use on a routine basis (Whitmarsh et al., 1996
). The question that all clinics therefore face is: at what IVF fertilization rate does ICSI become a more effective treatment than IVF? The answer is not clear-cut. We have attempted to address this by using the HFEA data (Templeton and Morris, 1998
; Human Fertilisation and Embryology Authority, 2000
). Table III
illustrates three scenarios where the IVF fertilization rate and number of embryos created, which have significant effects on live birth rate (LBR) varies. In these examples an IVF fertilization rate of <40% would only result in a maximum 20% LBR with no embryos available for freezing. Thus, the use of ICSI will result in a significantly higher LBR per fresh cycle and allow the possibility of 23 embryos for transfer in a subsequent cycle. Under such circumstances, it would be better to advise ICSI. However, the situation becomes less obvious when the fertilization rate is
50% as the LBR for fresh transfers are comparable with ICSI (28%) and, depending on the exact fertilization rate, a number of embryos are available for freezing. Of course, in these examples we have assumed that the spermatozoa look normal but are defective thus, embryos and pregnancies can be achieved by ICSI whereas conventional IVF would have been unsuccessful (Liu and Baker, 2000
). This is not always the case and our examples in Table III
only apply to such cases where ICSI can circumnavigate the use of defective male gametes.
|
![]() |
Conclusion |
---|
![]() ![]() ![]() ![]() ![]() ![]() |
---|
In summary, from both the safety and scientific viewpoints, ICSI should only be used in cases where success at IVF is regarded as unlikely.
![]() |
Acknowledgements |
---|
![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
Notes |
---|
After completion of this manuscript Bhattacharya and colleagues reported on a trial comparing ICSI and IVF in non-male factor and mild male factor infertility. Although live birth rates were not reported, implantation rates were similar between the two groups (Battacharya et al., 2001).
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() |
---|
Akutsu, H., Tres, L.L., Tateno, H. et al. (2001) Offspring from normal mouse oocytes injected with sperm heads from the azh/azh mouse display more severe sperm tail abnormalities than the original mutant. Biol. Reprod. 64, 249256.
Aytoz, A., Van den, A.E., Bonduelle, M. et al. (1999) Obstetric outcome of pregnancies after the transfer of cryopreserved and fresh embryos obtained by conventional in-vitro fertilization and intracytoplasmic sperm injection. Hum. Reprod., 14, 26192624.
Barratt, C.L.R. and Publicover, S.J. (2001) Interaction between sperm and zona pellucida in lae infertility. Lancet, in press.
Battacharya, S., Hamilton, M.P.R, Shaaban, M. et al. (2001) Conventional in-vitro fertilization versus intracytoplasmic sperm injection for the treatment of non-male factor infertility: a randomized controlled trial. Lancet, 357, 20752079.[ISI][Medline]
Bernardini, L., Martini, E., Geraedts, J.P. et al. (1997) Comparison of gonosomal aneuploidy in spermatozoa of normal fertile men and those with severe male factor detected by in-situ hybridisation. Mol. Hum. Reprod. 3, 431438.[Abstract]
Bonduelle, M., Camus, M., De Vos, A. et al. (1999) Seven years of intracytoplasmic sperm injection and follow-up of 1987 subsequent children. Hum. Reprod., 14, (Suppl. 1), 243264.[ISI][Medline]
Bourgain, C., Nagy, Z.P., De Zutter, H. et al. (1998) Ultrastructure of gametes after intracytoplasmic sperm injection. Hum Reprod., 13, (Suppl. 1) 107116.[Medline]
Chan, A.W., Luetjens, C.M., Dominko, T. et al. (2000) Foreign DNA transmission by ICSI: injection of spermatozoa bound with exogenous DNA results in embryonic GFP expression and live rhesus monkey births. Mol. Hum. Reprod., 6, 2633.
Damario, M.A., Hammitt, D.G., Galanits, T.M. et al. (1999) Pronuclear stage cryopreservation after intracytoplasmic sperm injection and conventional IVF: implications for timing of the freeze. Fertil. Steril., 72, 10491054. Erratum, Fertil. Steril., 73, 874.[ISI][Medline]
Dumoulin, J.C.M., Coonen, E, Bras, M. et al. (2000) Comparison of in-vitro development of embryos originating from either conventional in-vitro fertilization or intracytoplasmic sperm injection. Hum. Reprod., 15, 402409.
Dumoulin, J.C.M., Coonen, E., Bras, M. et al. (2001) Embryo development and chromosomal abnormalities after ICSI: effects of the injection procedure. Hum. Reprod., 16, 306312.
EIM/ESHRE (2001) The European IVF-monitoring programme for ESHRE. Assisted reproductive technology in Europe, 1997. Results generated from European registers by ESHRE. Hum. Reprod., 16, 384391.
Fishel, S., Aslam, I., Lisi, F. et al (2000) Should ICSI be the treatment of choice for all cases of in-vitro conception? Hum. Reprod., 15, 12781283.
Griffiths, T.A., Murdoch, A.P. and Herbert, M. (2000) Embryonic development in vitro is compromised by the ICSI procedure. Hum. Reprod., 15, 15921596.
Hawkins, M.M. and Barratt, C.L.R. (1999) Intracytoplasmic sperm injection. Science, 286, 5152.[ISI][Medline]
Hawkins, M.M., Barratt, C.L.R., Sutcliffe, A.G. and Cooke, I.D. (1999) Male infertility and increased risk of diseases in future generations. Lancet, 354, 19061907.[ISI][Medline]
Hewitson, L., Dominko, T., Takahashi, D. et al., (1999) Unique checkpoints during the first cell cycle of fertilisation after intracytoplasmic sperm injection in rhesus monkeys Nat. Med., 5, 133.[ISI][Medline]
Horne, G., Atkinson, A., Brison, D.R. et al. (2001) Achieving pregnancy against the odds: successful implantation of frozen-thawed embryos generated by ICSI using spermatozoa banked prior to chemo/radiotherapy for Hodgkinapos;s disease and acute leukaemia. Hum. Reprod., 16, 107109.
Hu, Y., Maxson, W.S., Hoffman, D.I. et al. (1999) A comparison of post-thaw results between cryopreserved embryos derived from intracytoplasmic sperm injection and those from conventional IVF. Fertil. Steril., 72, 10451048.[ISI][Medline]
Human Fertilisation and Embryology Authority (2000) Patientsapos; information and guide. URL http://www.hfea.gov.uk/frame2.htm
Kastrop, P.M., Weima, S.M., Van Kooij, R.J. et al. (1999) Comparison between intracytoplasmic sperm injection and in-vitro fertilization (IVF) with high insemination concentration after total fertilization failure in a previous IVF attempt. Hum. Reprod., 14, 6569.
Kerr, J., Brown, C., Balen, A.H. et al. (1999) The experiences of couples who have had infertility treatment in the United Kingdom: results of a survey performed in 1997. Hum. Reprod., 14, 934938.
Liu, D.Y. and Baker, H.W.G. (2000) Defective sperm-zona pellucida interaction: a major cause of failure of fertilization in clinical in-vitro fertilization. Hum. Reprod., 15, 702708.
Loft, A., Petersen, K., Erb, K. et al. (1999) A Danish national cohort of 730 infants born after intracytoplasmic sperm injection (ICSI) 19941997. Hum. Reprod ., 14, 21432148.
Luetjens, C.M., Payne, C. and Schatten, G.S. (1999) Non-random chromosome positioning in human sperm and sex chromosome anomalies following intracytoplasmic sperm injection. Lancet, 353, 1240.[ISI][Medline]
Manning, M., Lissens, W., Bonduelle, M. et al. (2000) Study of DNA-methylation patterns at chromosome 15q11-q13 in children born after ICSI reveals no imprinting defects. Mol. Hum. Reprod., 6, 10491053.
Naysmith, T.E., Blake, D.A., Harvey, V.J. et al (1998) Do men undergoing sterilizing cancer treatments have a fertile future? Hum. Reprod., 13, 32503255.[Abstract]
NHS Centre for Reviews and Dissemination (1999) Getting evidence into practice.Effective Health Care, 5, 116.
Perry, A.C., Wakayama, T., Kishikawa, H. et al. (1999) Mammalian transgenesis by intracytoplasmic sperm injection. Science, 284, 11801183.
Philips, Z., Barraza-Llorens, M. and Posnett, J. (2000) Evaluation of the relative cost effectiveness of treatments for infertility in the UK. Hum. Reprod., 15, 95100.
Ruiz, A., Remohi, J., Minguez, Y. et al. (1997) The role of in vitro fertilisation and intracytoplasmic sperm injection in couples with unexplained infertility after failed intrauterine insemination. Fertil. Steril., 68, 171173.[ISI][Medline]
Sakkas, D. Mariethoz, E., Manicardi, G. et al.. (1999) Origin of DNA damage in ejaculated human spermatozoa. Rev. Reprod. 4, 3137.
Shoukir, Y., Chardonnens, D., Campana, A. et al. (1998) Blastocyst development from supernumerary embryos after intracytoplasmic sperm injection: a paternal influence? Hum. Reprod., 13, 16321637.[Abstract]
Staessen, C., Camus, M., Clasen, K. et al. (1999) Conventional in-vitro fertilization versus intracytoplasmic sperm injection in sibling oocytes from couples with tubal infertility and normozoospermic semen. Hum. Reprod., 14, 24742479.
Templeton, A.A. and Morris, J.K. (1998) Reducing the risk of multiple births by transfer of two embryos after in vitro fertilisation. N.E.J.M.., 339, 573577.
Terada, Y. Luetjens, C.M., Sutovsky, P. et al. (2000) Atypical decondensation of the sperm nucleus, delayed replication of the male genome, and sex chromosome positioning following intracytoplasmic human sperm injection (ICSI) into golden hamster eggs: does ICSI itself introduce chromosome anomalies. Fertil. Steril., 74, 454460.[ISI][Medline]
Tesarik, J. (1998) Oocyte activation after intracytoplasmic injection of mature and immature sperm cells. Hum. Reprod., 13, (Suppl. 1), 117127.
Whitmarsh, A.J., Woolnough, M.J. Moore, H.D.M. et al. (1996) Biological activity of recombinant human ZP3 produced in vitro:potential for a sperm function test. Mol. Hum. Reprod., 2, 911919.[Abstract]
Wennerholm, U.B., Bergh, C., Hamberger, L. et al. (2000) Incidence of congenital malformations in children born after ICSI. Hum. Reprod., 15, 944948.
Yang, D., Shahata, M.A., al Bader, M. et al. (1996) Intracytoplasmic sperm injection improving embryo quality: comparison of the sibling oocytes of non-male-factor couples. J. Assist. Reprod. Genet., 13, 351355.[ISI][Medline]