Department of Obstetrics & Gynaecology, University of Aberdeen, Aberdeen Maternity Hospital, Foresterhill, Aberdeen AB25 2ZD, UK E-mail: m.hamilton{at}abdn.ac.uk
Dear Sir,
We were interested to read the paper by Fishel et al. and intrigued by the title`Should ICSI be the treatment choice in all cases of in-vitro conception?' (Fishel et al., 2000).
This is an important clinical question and the authors are to be commended for raising the issue. Unfortunately the title of the paper is a little misleading since the data presented do not answer the question posed in the title, at least from a clinical perspective.
Firstly, the patients in this study, by definition, formed a selected group, who would normally have been offered ICSI by most clinics. They included couples with previous failed conventional IVF, or males with abnormal semen parameters, and as such could not be described as a routine IVF population. It would be inappropriate to extrapolate results based on this sample population to a general pool of IVF couples.
Secondly, the description of the study, as a prospective randomized trial, is suspect. The method of random allocation is not described, although the authors allude to the randomization of oocytes at MII stage of development. Conventional IVF would not allow for the identification of MII oocytes, unless they were denuded of surrounding cumulus cells. True randomization would surely apply to all oocytes available after oocyte retrieval. No information is given about the total number of oocytes derived from these patients. Subsequent management was variable, with some sub-groups receiving treatment with donor spermatozoa and others with the poor quality spermatozoa from the partner. Furthermore Groups 4 and 5, where failed fertilized oocytes, after an attempt at IVF, were reinseminated or injected, represent a clinical scenario irrelevant to clinical practice in the UK.
Thirdly, although the authors describe embryo quality as an important outcome relevant to the analysis, no data are shown to substantiate the claim that morphology, embryo cleavage rates and, importantly, implantation rates after embryo transfer, were comparable in the groups.
The authors have shown in this work that fertilization rates in cases of male factor infertility may be improved with ICSI, but this does not represent new knowledge (Van Steirteghem et al., 1993; Aboulghar et al., 1995). The question raised by the title of this paper demands a pragmatic prospective randomized trial, using a population of couples in whom IVF would be the current treatment of choice. Patients, rather than oocytes, should be randomized to receive either IVF or ICSI, and prior to the initiation of the study an appropriate power calculation should be carried out to define the patient numbers required to draw meaningful conclusions.
We have recently completed such a study involving over 400 such couples. Preliminary data have recently been presented (Bhattacharya et al., 2000) and the completed data set will be published elsewhere shortly.
In the study by Fishel et al. an important laboratory question has been answered, but it would be unwise to draw any clinical inference from the work.
References
Aboulghar, A.A., Mansour, R.T., Serour, G.I. and Amin, Y.M. (1995) The role of intracytoplasmic sperm injection (ICSI) in the treatment of patients with borderline semen. Hum. Reprod., 10, 28292830[ISI][Medline]
Bhattacharya, S., Shabaan, M., Khalaf, Y. et al. (2000) A randomised controlled trial of IVF and ICSI in non-male factor infertility. Abstracts of the 16th ESHRE Annual Meeting, Bologna, Italy. Hum Reprod., 15, O-148.
Fishel, S., Aslam, I., Lisi, F. et al. (2000) Should ICSI be the treatment for all cases of in-vitro conception? Hum. Reprod., 15, 12781283
Van Steirtegheim, A.C., Nagy, Z., Joris, H. et al. (1993) High fertilization and implantation rates after intracytoplasmic sperm injection. Hum. Reprod., 8, 10611066[Abstract]
CARE (Centres for Assisted Reproduction), The Park Hospital, Sherwood Lodge Drive, Burntstump Country Park, Arnold, Nottingham NG5 8RX UK Biogenesi, Casa di Cura, Villa Europa, Via Eufrate 27, 00144 Rome, Italy Vitalab, Box 46337 Orange Grove, 2119 Republic of South Africa
Dear Sir,
Drs Hamilton and Bhattacharya comment on our paper, `Should ICSI be the treatment of choice for all cases of in-vitro conception?' and correctly point out that we examined a `laboratory question'. Indeed, this should be the first port of call in any IVF procedure. But if the observations are performed with sibling oocytes during `real time' clinical IVF then surely a clinical embryology study has a `clinical perspective'. Semantics? Perhaps?
We are slightly bemused by the letter, but it implies some response to the four general criticisms and the overwhelming one that `the data presented does not answer the question posed in the title'. On the contrary, it is imperative that we consider our embryology first (possibilities for enhancing conception in vitro) before subjecting patients to a clinical trial; indeed launching into the latter prior to the former raises ethical concernsas we shall see later. In answer to the four criticisms:
Firstly, patients were distributed into five groups for analysisthat does not mean `by definition' they were `selected'. We did not, in the first instance, choose patients for whom IVF was entirely appropriate, as there is a clear ethical dilemma with such an approach. Perhaps in the year 2000 `most clinics' would choose ICSI for each of our five groups of patients (and we do not dispute this), however, that was not the situation during 199498 when this study was conducted; and there are still those today who would consider groups 1, 4 and 5 as appropriate for repeat IVF or IVF-HIC before commencing ICSI. Some specialists still offer donor sperm insemination in preference to the `risks of ICSI'; but most of us would not now accept that approach.
Secondly, the criticism of the `prospective randomized trial' is based on the concern that `conventional IVF would not allow for identification of MII oocytes .....'. Precisely! The fact that the oocyte-cumulus complexes (OCC) were chosen, and that the embryologist would be blind to the condition of the oocyte in the OCC, would support randomized distribution of the oocytes into the various groups.
However, we do agree with Drs Hamilton and Bhattacharya that the statement in our paper that `MII oocytes were inseminated ....' is not helpful and we were grateful to Drs Hamilton and Bhattacharyo for pointing this out. It was worded as such because our starting point for analysis was the MII oocyte. But, of course, it was the OCC'scontaining MII, MI or germinal vesicle oocytes within the OCCthat were inseminated. The total number of OCC's, MI and GV oocytes recovered was not presented because it adds no relevant information to the studyonly MII can be compared when assessing ICSI [and this informationfertilization rate per total number of oocytes or even per MII oocytescannot even be extracted from the abstract report of Drs Hamilton and Bhattacharya's own study! (Bhattacharya et al., 2000)].
The concern about the variable management of oocytes is not understoodit was precisely this approach to sibling oocytes that was the essence of the analysis. Drs Hamilton and Bhattacharya criticise the use of ICSI in 1-day old failed IVF oocytes as a `clinical scenario irrelevant to clinical practice in the UK'. However, scientific enquiry is neither limited to the borders of legal jurisdiction, nor centred solely on the practice in the UK. Indeed, reinsemination of 1-day old failed fertilization oocytes is common practice in many units outside the UK, and a number of healthy children have been born by this approach. The UK may yet have its restrictions reviewed in time to come.
Thirdly, Drs Hamilton and Bhattacharya express concern about our interpretation of embryo morphology. This puzzles us. We state clearly that this study assessed fertilization, cleavage and embryo quality (by morphological assessment). Where the data varied, for example in fertilization outcome, the data was shown. In relation to morphology we state `no significant difference in embryo quality was noted between the embryos within groups ....' and we reference our previous studies by comparison (Hall et al., 1995) on this issue. Our paper was not an embryo transfer-outcome based study for the reasons clearly stated and further commented on below.
The fourth criticism is that our study did not represent new knowledge. However, it was the first sibling oocyte study of its kind; it was, in its approach, comprehensive; it did contribute to the body of knowledge and, in Drs Hamilton and Bhattacharya's own words `an important laboratory question has been answered'.
Finally, their suggestion that `patients rather than oocytes should be randomized' raises a number of issues. The two studies are not mutually exclusive, which we believe Drs Hamilton and Bhattacharya recognize. But the use of patients `in whom IVF would be the current treatment of choice' and to expose `over 400 such couples' to this approach is worrisome. How is ethical approval given to use a highly invasive procedure that is clearly not indicated? The literature is littered with risk-concerns about the willy-nilly approach to ICSI, unless indicated. Patients in our study knew they had an option to have conventional IVF-treated embryos transferred. The alternate approach suggested by Drs Hamilton and Bhattacharya would provide only ICSI-embryos to patients who did not require ICSI.
Finally, the recent paper by Griffiths et al. (2000) supports our initial rejection of the outcome-based study as performed by Drs Hamilton and Bhattacharya as being of significant concern to be deemed probably unethical. Griffiths et al. demonstrated that 80% of ICSI embryos compared to 50% of IVF embryos (P < 0.01) were unable to form blastocysts; giving rise to the concern that the ICSI procedure `contributes to a reduced capacity for blastocyst formation'. Given these concerns, and the predicted outcome in the Dr Hamilton and Bhattacharya study in which they have concluded, `ICSI offers no clinical advantage over IVF in cases on non-male factor infertility', we feel even more justified in that the clinical embryology, laboratory approach of our study was the appropriate first method to answer the title of the study.
Notes
1 To whom correspondence should be addressed
References
Bhattacharya, S., Shabaan, M., Khalaf, Y. et al. (2000) A randomised control trial of IVF and ICSI in non-male factor infertility. Abstracts of the 16th ESHRE Annual Meeting, Bologna, Italy. Hum. Reprod., 15, 0148.
Griffiths, T.A., Murdoch, A.P. and Herbert, M. (2000) Embryonic development in vitro is compromised by the ICSI procedure. Hum. Reprod., 15, 15921596.
Hall, J., Fishel, S., Green, S. et al. (1995) Introcytoplasmic sperm injection versus high insemination concentration in-vitro fertilisation in cases of very severe teratozoospermia. Hum. Reprod., 10, 493496.[ISI][Medline]