To the Editor:
We read with interest the article by Fischer et al. which present a retrospective review of data on the effect of preimplantation genetic diagnosis (PGD) for carriers of a translocation with a history of three or more pregnancy losses (1). The authors conclude that PGD improves pregnancy outcome and dramatically reduces time to conceive. However, their data and comparative analysis did not convince us that their conclusions are justified, as outlined below.
Although the authors state that comparing pregnancy loss and live birth rates prior and subsequent to PGD is highly biased, they report the comparisons. Using the women as their own controls is not a fair comparison, especially without modelling to account for multiple pregnancies for each woman, since it always favours the intervention (2).
In addition, the comparison of PGD with non-PGD data (Table 1) is misleading. In the second column (which is not labelled), the numbers in the three PGD studies correspond to clinical pregnancies, not IVF-PGD cycles started, which would be more appropriate. Assuming all reported ongoing PGD pregnancies resulted in a live birth, we calculated from their data a live birth rate of 22% (60/272) per IVF-PGD cycle started, and 31% (60/192) cumulative live birth rate after 1.4 cycles of IVF-PGD per couple, not 87% (60/69) as reported in the article.
In the three non-PGD studies, numbers in the second column correspond to couples with a translocation, some of whom decided not to try again. The live birth rates included pregnancies that ended in preclinical, as well as clinical miscarriage.
In addition, the listing of non-PGD studies is far from complete. The large study by Franssen et al. (3) was omitted; it reported an 83% cumulative live birth rate, in a cohort of 247 carrier couples. With its inclusion, the overall live birth rate in the non-PGD studies would have been even higher than reported.
Finally, the authors inappropriately report the duration of the non-PGD studies (4, 5) as “timeframe to success”, but fail to report their own study duration. Instead, they state “timeframe to success” in their study is the average number of IVF-PGD cycles which resulted in a successful pregnancy. This resulted in the erroneous conclusion that IVF-PGD dramatically reduces the time to a successful pregnancy outcome.
In our opinion, their study does not show that IVF-PGD improves pregnancy outcome in translocation carriers with a history of recurrent pregnancy loss; in fact, it shows the opposite. Therefore, we cannot agree with the conclusions of Fischer et al. We hope that readers are aware of the difficulties of comparing PGD with non-PGD data, and that they will offer appropriate counselling to such patients, who need accurate and unbiased information about pregnancy outcomes with or without preimplantation genetic diagnosis.
Mary D. Stephenson, M.D., M.Sc.
University of Chicago,
Mariëtte Goddijn, M.D., Ph.D.
University of Amsterdam,
Amsterdam, The Netherlands
1. Fischer J, Colls P, Escudero T, Munne S. Preimplantation genetic diagnosis (PGD) improves pregnancy outcome for translocation carriers with a history of recurrent losses. Fertil Steril 2010;94:283-89.
2. Christiansen OB, Nybo Andersen AM, Bosch E, Daya S, Delves PJ, Hviid TV et al. Evidence-based investigations and treatments of recurrent pregnancy loss. Fertil Steril 2005;83:821-39.
3. Franssen MT, Korevaar JC, van der Veen F, Leschot NJ, Bossuyt PM, Goddijn M. Reproductive outcome after chromosome analysis in couples with two or more miscarriages: index-control study. BMJ 2006;332:759-63.
4. Goddijn M, Joosten JH, Knegt AC, van der Veen F, Franssen MT, Bonsel GJ, et al. Clinical relevance of diagnosing structural chromosome abnormalities in couples with repeated miscarriage. Hum Reprod 2004;19:1013-17.
5. Stephenson MD, Sierra S. Reproductive outcomes in recurrent pregnancy loss associated with a parental carrier of a structural chromosome rearrangement. Hum Reprod 2006;21:1076-82.
Published online in Fertility and Sterility doi:10.1016/j.fertnstert.2010.09.001
The Authors Respond:
We appreciate the comments of Doctors Stephenson and Goddijn, and, as we state in the paper, we concur that using the prior pregnancy history of a patient as a control for a subsequent intervention is inappropriate. This is the reason we have compared the treatment results to other studies (table 1) which clearly show a decrease in miscarriage rates from 26-64% to 13% after PGD.
This same table is questioned because we used pregnancies and not cycles started. However, as we are comparing IVF cycles to natural cycles, the only way to compare such in the literature is by comparing pregnancies to pregnancies as it is unknown how many natural cycles are undertaken and how many do not result in pregnancy.
As we have debated in the past with the authors, we may be looking at different populations. As stated in our paper, those seeking PGD may be traumatized by their history of losses. They may be “…emotionally or physically unable to continue conceiving naturally and comfortable with using ART.” These patients report that they cannot experience more losses and do not want to “waste” more time in trying to have a successful pregnancy naturally. In contrast, those patients conceiving naturally may not wish to undergo IVF, have the ability to do so, or even know this to be an option. They may not be as traumatized by losses, and in the face of little risk to deliver a child with an unbalanced form of the translocation, may not choose PGD.
Regardless, as stated in the paper, either choice has pros and cons as conceiving naturally and having a loss is not without physical, emotional and even financial risk. Obviously, a well randomized study would clarify if IVF with PGD can produce a successful pregnancy faster than no treatment. However, certain points are not arguable: PGD can detect translocations and transferring of an embryo free of the unbalanced form of such significantly reduces miscarriage rates. PGD is therefore a viable option for any individual with a translocation. Thus, PGD should be an option offered to all of these patients. Not providing patients with all options prevents them from making a truly informed decision as to their care. Therefore, if patients are counseled without bias as to all their options, they will be able to make the best decision for themselves.
Of note, since the paper was presented, the PGD field is moving towards full chromosome counts using DNA microarrays, which should further reduce the miscarriage rates, and towards using blastocyst biopsy, which may improve pregnancy rates.
Jill M. Fischer, M.S., C.G.C
Livingston, New Jersey
Santiago Munné, Ph.D.
Livingston, New Jersey
Published online in Fertility and Sterility doi:10.1016/j.fertnstert.2010.09.002