Department of Obstetrics and Gynaecology, Solihull Hospital, Lode Lane, Solihull B91 2TL, UK and
University Department of Obstetrics and Gynaecology, Jessop Hospital for Women, Sheffield S3 7RE, UK
Dear Sir,
We read with interest the recent paper (Moreno et al., 1998), the idea of which is a valid contribution to the literature on the management of poor/low responders in in-vitro fertilization (IVF). Employing intracytoplasmic sperm injection (ICSI) to improve the fertilization rate in this group of patients echoes the intuition of many clinicians. However, we would like to raise several points of concern that weaken the conclusion drawn by the authors.
The authors failed to report on the method by which patients were allocated to IVF or ICSI and its timing in relation to the start of ovarian stimulation and oocyte recovery. Therefore, inadvertent bias or contamination of the study samples could not be excluded. Although the authors compared two groups of 52 cycles, they did not report on the number of patients in each group. Nevertheless, they went on to compare the `number of oocytes and embryos per patient' and `pregnancy rate per cycle' between the two groups in their study, which is rather confusing and misleading. Early follicular phase follicle stimulating hormone (FSH) concentrations should also be given to show that the groups would be expected to exhibit a similar response.
Pregnancy rate is the ultimate proof of the efficacy of any modality of assisted reproduction. The pregnancy rate per cycle in the ICSI group was 4% higher than the IVF group. Simple mathematical computation will show that ~2250 patients are required in each group to achieve a power of 90% in detecting a significant difference at 0.05 significance level. The sample size used by Moreno et al. will achieve a statistical power of <5%, i.e. (type 11 error) of >0.95. Therefore, there is a chance of 95% of falsely accepting the null hypothesis.
The authors did not report on the distribution of the cause of infertility, the number of previous IVF attempts and duration of infertility in the two study groups which have been shown to influence the pregnaw2ncy rate in IVF patients (Templeton, 1998). The authors administered eight ampoules of gonadotrophin (600 IU of FSH) to the patients who had previous cancelled cycles (34.6 and 38.5% in each group) despite the lack of supporting evidence. We have reported that exceeding a gonadotrophin dose of 300 IU of FSH is futile (Lashen et al., 1998
), besides such a high dose may be associated with lower pregnancy rates in IVF ((Stadtmour et al., 1994
),
In view of all these points it is difficult to accept the authors' conclusion as solid proof. Nevertheless, their effort in addressing the role of ICSI in the management of poor/low responders is very much appreciated and could be useful as a pilot study.
References
Lashen, H., Ledger, W., Lopez-Bernal, A. et al. (1998) Superovulation with high gonadotrophin dose for in vitro fertilisation: is it effective? J. Assist. Reprod. Genet., 15, 438443.[ISI][Medline]
Moreno, C., Ruiz, A., Simón, C. et al (1998) Intracytoplasmic sperm injection as a routine indication in low responder patients. Hum. Reprod., 13, 21262129.[Abstract]
Stadtmour, L., Ditkoff, E.C., Session, D. et al. (1994) High dosage of gonadotrophins are associated with poor pregnancy outcome after in vitro fertilisation-embryo transfer. Fertil. Steril., 61, 10581064.[ISI][Medline]
Templeton, A. (1998) Reducing the risk of multiple birth by transfer of two embryos after in vitro fertilisation. N. Engl. J. Med., 339, 573577.
Instituto Valenciano de Infertilidad, Guardia Civil 23, E-46020 Valencia, Spain
Dear Sir,
We appreciate the interest of Drs Lashen and Ledger in our manuscript published last year in Human Reproduction (Moreno et al., 1998). They are right that some information could be additionally added to the original paper. The method of randomization or the characteristics of the patients (number of previous attempts, etc) could be of interest and therefore we apologize for not including these data. They can be sure, however, that the method of randomization was adequate and there was no difference between groups in the characteristics mentioned. Sometimes, available data are not included in order to make the message clearer. However, they should not be confused concerning patients and cycles. From the careful reading of the manuscript, it is obvious that each cycle represents a single patient and none repeated treatment in the study.
Nevertheless, we have to stress that the point was basically fertilization and implantation rather than pregnancy rates as they suggest. We have always felt that implantation is a better parameter than pregnancy in showing differences between women using two different methods of fertilization. In this sense, we completely disagree with their statistical exercise because if one takes a look to the number of oocytes included in each group there is no reason for concern about the study.
In addition, we may agree with their findings of increasing the dose of gonadotrophins in low responders is not useful. In fact, we published it >10 years ago (Pellicer et al., 1987). However, this point is not relevant to the message of our manuscript, and we do not understand their comment.
References
Moreno, C., Ruiz, A., Simón, C. et al. (1998) Intracytoplasmic sperm injection as a routine indication in low responder patitients. Hum. Reprod., 13, 21262129[Abstract]
Pellicer, A., Lightman, A., Diamond, M.P. et al. (1987) Outcome of in vitro fertilization in women with low response to ovarian stimulation. Fertil. Steril., 47, 812815.[ISI][Medline]