cies, and 82% of these pregnancies (25% of all cycles) resulted in live births [1]. Although efficiencies are not usually reported as the fraction of successful pregnancies per embryo transferred, 12% of embryos transferred in one study after preimplantation genetic diagnosis (PGD) implanted successfully (yielding a success rate of 19.9% when measured in the usual terms of pregnancy per cycle) [2].

Clinical characteristics of the male and female partners play a major role in determining the success rate of IVF treatment. For example, in 1994, the highest success was reported for couples in which the female partner was younger than 40 years old and the male had a normal semen analysis (24.5% live births per cycle). The lowest success was reported for women older than 40 years old with a male partner with a normal semen analysis (9% live births per cycle) or abnormal semen analysis (8.5% live births per cycle) [3].

The success rate of IVF may be constrained by the relatively high rate of pregnancy loss in humans. In unassisted reproduction, many pregnancies are lost before there is any clinical sign of their existence (“occult pregnancies”), and additional pregnancies are lost after they are detectable with hormone measurements but before they are detectable with ultrasonography (“chemical pregnancies”). According to one source [4], “more than 80% of [spontaneous] abortions occur in the first twelve weeks, and the rate decreases rapidly thereafter.” This contrasts with the frequent loss of cloned animal fetuses late in gestation.

IVF procedures involve the collection of eggs for fertilization. Any human reproductive cloning attempt would also involve this procedure, and the low efficiency of animal cloning suggests that a large number of eggs would have to be collected. The collection of these eggs would bring with it the risk of ovarian hyperstimulation syndrome in donors. The incidence of moderate and severe cases of this syndrome in studies in which more than 1000 IVF cycles were evaluated ranges from 0.8% [5] to 1.95% [6]. Maternal death resulting from the syndrome is rare enough that it is the subject of occasional case reports.

In the United States, multiple embryos are frequently implanted during an IVF cycle to increase the chances of a successful pregnancy [1]. That often results in multiple births, which are associated with risks of morbidity and mortality for the mother and, because of prematurity and low birth weight, for the children.

When IVF was first adopted in humans, no increase in the frequency of major malformations had been seen in IVF experiments in mice relative to normal animal reproduction [7]. That situation is in contrast with the data on animal cloning discussed in Chapter 3; cloned animals have markedly more problems, particularly severe abnormalities throughout gestation, than those animals produced by normal reproduction.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement