What is the most relevant standard of success in assisted reproduction?

The importance of informed choice

William Buckett1,2 and Seang Lin Tan1

1 Department of Obstetrics and Gynecology, McGill University, Montréal, Canada

2 To whom correspondence should be addressed at: Department of Obstetrics and Gynecology, McGill University, Royal Victoria Hospital, 687, avenue des Pins Ouest, Montréal, Canada H3A 1A1. e-mail: william.buckett{at}muhc.mcgill.ca


    Abstract
 Top
 Abstract
 Introduction
 Measurement of success
 The risks of multiple...
 Conclusions
 References
 
Whilst there is a need to emphasize the birth rates of healthy infants when reporting success rates following infertility treatment, is ‘the singleton term gestation per cycle started’ the most appropriate measure? Although there are many ways to measure the efficacy of medical intervention, none of these exclude from the ‘successes’ those in whom complications have occurred. The safe delivery of healthy twins at term, while possibly not the ideal outcome, should still be regarded as a treatment success. Although multiple pregnancy, in particular high order multiple pregnancy, continues to be a major problem associated with assisted reproductive treatments and also with ovulation induction, appropriate counselling concerning the risks of twin and triplet pregnancies will allow couples to make informed choices and allow clinicians to continue to individualize treatment plans.

Key words: assisted reproduciton/success rates/multiple gestation/informed choice


    Introduction
 Top
 Abstract
 Introduction
 Measurement of success
 The risks of multiple...
 Conclusions
 References
 
There continues to be much controversy surrounding the reporting of success rates of infertility treatment in general and assisted reproductive treatment (ART) in particular. The different denominators used by the HFEA in the UK (clinical pregnancies per cycle started) and SART/ASRM in the USA (clinical pregnancies per oocyte retrieval) are a case in point. Added to this, the current World Health Organization recommendation that ART rates emphasize the birth rates of healthy infants, as well as malformations, neonatal morbidity and mortality, and pregnancy complications (World Health Organization 2002Go), confirm the need for improved ways of reporting treatment success in infertility. However, is the singleton, term gestation per cycle started the most appropriate measure? (Min et al., 2004Go)

Multiple gestation is now a major problem associated with ART and ovulation induction therapies. The multiple gestation rates, particularly high order multiple gestation (triplets and above), can be reduced by reducing the numbers of embryos transferred, including the introduction of elective single embryo transfer (SET) (Vilska et al., 1999Go; Martikainen et al., 2001Go; Gerris et al., 2002Go; De Sutter et al., 2003Go; Soderstrom-Anttila et al., 2003Go). However, most authorities would concede, rather than adopting an absolute limit of the number of embryos that can be transferred, that the number of embryos transferred should be individualized for each couple (Van Steirteghem et al., 2003Go). Improvements in ovarian stimulation, embryo selection, preimplantation aneuploidy screening and embryo cryopreservation techniques will also allow the number of embryos transferred to be reduced with less impact on the overall success of treatment.

However, the adoption of the singleton term gestation per cycle started as the measure of success would undoubtedly focus on, and increase the use of, elective SET. Is this in all our patients’ best interests? There are major national and international differences in the demographics of couples seeking treatment and also in the provision of treatment of infertility as well as the funding of health care as a whole. Although programmes which have introduced elective SET for a significant proportion of couples have not demonstrated any decline in clinical pregnancy rates (Gerris et al., 2002Go; Tiitinen et al., 2003Go), data from Canada (Canadian Fertility and Andrology Society, 2003Go) and the USA (Toner, 2002Go) continue to show a year-on-year increase in clinical pregnancy rates. Therefore, what may be most appropriate for couples in one country or region may not necessarily be ideal elsewhere.


    Measurement of success
 Top
 Abstract
 Introduction
 Measurement of success
 The risks of multiple...
 Conclusions
 References
 
Success of medical intervention is measured in a variety of ways. Usually this is by resolution of symptoms or disease; for example, the efficacy of antimicrobial chemotherapy is measured by the eradication of infection. Other ways of measuring efficacy concern survival or death, and this has been modified to the 2-, 5- and 10-year survival rates used when comparing interventions for terminal disease. However, many interventions improve symptoms rather than eradicate disease, and the success of these treatments is measured by resolution of the presenting symptoms (e.g. mobility following hip replacement surgery, or reduction in menstrual loss following medical treatment for menorrhagia). The efficacy of these non-curative interventions can be widened by the use of quality of life measures.

Despite these different and often complementary methods of measuring the success of particular medical interventions, none seeks to exclude from the ‘successes’ those in whom complications have occurred; for example, a urinary or wound infection does not reduce the efficacy of a hysterectomy in the treatment of persistent refractive menorrhagia. In other words, suffering a complication of treatment does not mean that the treatment has failed!

Many couples seeking ART also make decisions concerning other treatments for infertility, for example reversal of tubal ligation versus IVF for previous tubal ligation, or intrauterine insemination versus IVF for unexplained infertility in a woman over 38 years of age. To allow appropriate comparisons, consistent measures of success for all types of infertility treatment are therefore appropriate.

The inconsistency of excluding cases where complications have occurred (namely multiple pregnancy) from overall rates of success would also deny the existence of the concept of a partial or acceptable success; for example, a woman who had suffered repeat mid-trimester pregnancy losses would accept that delivery at 34 weeks would still represent a treatment success, although the delivery would still be pre-term and have consequent but reduced risks.


    The risks of multiple pregnancy
 Top
 Abstract
 Introduction
 Measurement of success
 The risks of multiple...
 Conclusions
 References
 
It is well established that multiple pregnancy is associated with increased maternal and neonatal risks and that these risks increase with the number of fetuses. Nevertheless, many couples with infertility undergoing ART see a twin delivery as the ideal outcome (Child et al., 2004Go). While informed decision making is paramount, a closer look at the relative and absolute risks of twin pregnancy versus singleton pregnancy is warranted.

Raw maternal mortality rates are 2–3 times higher in all multiple pregnancies (10.2–14.9 per 100 000) compared with all singleton pregnancies (4.4–5.2 per 100 000) in Europe (Senat et al., 1998Go). Although this is significant, it is less than the increased risk of thromboembolism whilst taking third-generation oral contraceptive pills (25 per 100 000) compared with non-users (5 per 100 000) (Drife, 2002Go).

Neonatal mortality is also significantly higher in multiple pregnancy and increases with the number of fetuses. Raw neonatal mortality rates are ~5 times higher for twin pregnancies (31 per 1000 live births) compared with singleton pregnancies (6 per 1000 live births) (Russell et al., 2003Go). Neonatal mortality rates for triplet pregnancy, however, are ~20 times higher, i.e. in the region of 100–120 per 1000 live births (Doyle., 1996Go). Therefore, while most couples would wish to avoid the 10–12% chance of neonatal death as well as significant morbidity associated with triplet pregnancies, many may chose to accept a 3% versus 0.6% chance of neonatal death when comparing a twin versus a singleton pregnancy. These risks do not account for monochorionicity, which is associated with poorer outcomes and is more common in spontaneous twin pregnancies compared with twin pregnancies resulting from ART (Derom et al., 2001Go).

Cerebral palsy is the most significant neurological impairment associated with multiple birth and is also increased by 4–6 times in twin delivery (9–12 per 1000 first year survivors) compared with singleton delivery (2.3 per 1000 first year survivors) (Petterson et al., 1993Go; Pharoah and Cooke, 1996Go). Again, the risk in triplet deliveries is exponentially higher at 18–20 times (45 per 1000 first year survivors). Similarly, some couples may accept the increased risks of twin pregnancy, but not those associated with a triplet pregnancy.

Although much more difficult to quantify, multiple births are also associated with a higher prevalence of maternal depression (Thorpe et al., 1991Go), parental stress and sibling problems (Bryan, 2003Go).


    Conclusions
 Top
 Abstract
 Introduction
 Measurement of success
 The risks of multiple...
 Conclusions
 References
 
While improved methods of measuring success in ART need to be evaluated, adopting the rate of singleton term delivery per cycle started has several disadvantages. First, it is not a consistent way of reporting treatment success when compared with other infertility treatments or indeed other interventions in medicine as a whole. Secondly, its adoption would lead to a global or blanket increase in the use of SET (rather than continued individualized treatment plans). This may not be of benefit to all patients, and increased education and dissemination of the risks of twin and higher order multiple pregnancy would be the most appropriate way to reduce these risks of ART.


    References
 Top
 Abstract
 Introduction
 Measurement of success
 The risks of multiple...
 Conclusions
 References
 
Bryan E (2003) The impact of multiple preterm births on the family. Br J Obstet Gynecol 110 (Suppl 20),24–28.

CanadianFertilityandAndrology Society (2003) Human assisted reproduction live birth rates for Canada. 6th Annual Meeting of the Directors of Canadian IVF Centres, Victoria, British Columbia, Canada

Child TJ, Henderson AM and Tan SL (2004) The desire for multiple pregnancy in male and female infertility patients. Hum Reprod 19,558–561.[Abstract/Free Full Text]

Derom C, Derom R and Vlientinck R (2001) The risk of monozygotic twinning. In Blickstein I and Keith LG (eds), Iatrogenic Multiple Pregnancy: Clinical Implications. Parthenon Publishing, New York, pp. 9–19.

DeSutter P, Van der Elst J, Coetsier T and Dhont M (2003) Single embryo transfer and multiple pregnancy reduction in IVF/ICSI: a five-year appraisal. Reprod Biomed Online 6,464–469.[Medline]

Doyle P (1996) The outcome of multiple pregnancy. Hum Reprod 11 (Suppl 4),110–117.[Medline]

Drife J (2002) Oral contraception and the risk of thromboembolism: what does it mean to clinicians and their patients? Drug Saf 25,893–902.[Medline]

Gerris J, De Neubourg D, Mangelschots K, Van Royen E, Vercruyssen M, Barudy-Casquez J, Valkenburg M and Ryckaert G (2002) Elective single day 3 embryo transfer halves the twinning rate without decreasing the ongoing pregnancy rate of an IVF/ICSI programme. Hum Reprod 17,2626–2631.[Abstract/Free Full Text]

Martikainen H, Tiitinen A, Tomàs C, Tapanainen J, Ovara M, Tuomivaara M, Vilska S, Hyden-Granskog C and Hovatta O for the Finnish ET study group (2001) One versus two embryo transfer after IVF and ICSI: a randomized study. Hum Reprod 16,1900–1903.[Abstract/Free Full Text]

Min JK, Breheny SA, MacLachlan V and Healy DH (2004) What is the most relevant standard of success in assisted reproduction? The singleton, term gestation, live birth rate per cycle initiated: the BESST endpoint for assisted reproduction. Hum Reprod 19,3–7[Abstract/Free Full Text]

Petterson B, Nelson K and Watson K, (1993) Twins, triplets and cerebral palsy births in Western Australia in the 1980s. Br Med J 307,1239–1243.[Medline]

Pharoah PO and Cooke T (1996) Cerebral palsy and multiple births. Arch Dis Child Fetal Neonatal Ed 75,F174–F177.[Medline]

Russell RB, Petrini JR and Damus K (2003) The changing epidemiology of multiple births in the United States. Obstet Gynecol 101,129–135.[Abstract/Free Full Text]

Senat M, Ancel P, Bouvier-Colle MH and Breart G (1998) How does multiple pregnancy affect maternal morbidity and mortality? Clin Obstet Gynecol 41,78–83.[Medline]

Soderstrom-Anttila V, Vilska S, Makinen S, Foudila T and Suikkari AM (2003) Elective single embryo transfer yields good delivery rates in oocyte donation. Hum Reprod 18,1858–1863.[Abstract/Free Full Text]

Thorpe K, Golding J, MacGillivray I and Greenwood R (1991) Comparison of prevalence of depression in mothers of twins and mothers of singletons. Br Med J 302,875–878.[Medline]

Tiitinen A, Unkila-Kallio L, Halttunen M and Hyden-Granskog C (2003) Impact of elective single embryo transfer on the twin pregnancy rate. Hum Reprod 18,1449–1453.[Abstract/Free Full Text]

Toner JP (2002) Progress we can be proud of: US trends in assisted reproduction over the first 20 years. Fertil Steril 78,943–950.[CrossRef][Medline]

VanStierteghem A, Gerris J and Frydman R (2003) Individualization of the number of embryos transferred. In: Infertility therapy associated multiple pregnancies (births); an ongoing epidemic. Reprod. Biomed. Online 7 (Suppl 2),18–19.

Vilska S, Tiitinen A, Hyden-Granskog C. and Hovatta O (1999) Elective transfer of one embryo results in an acceptable pregnancy rate and eliminates the risk of multiple birth. Hum Reprod 14,2392–2395.[Abstract/Free Full Text]

WorldHealth Organization (2002) Medical, Ethical and Social Aspects of Assisted Reproduction: Report of a WHO Meeting. WHO Publications, Geneva, Switzerland, pp. 363–376.