What is the most relevant standard of success in assisted reproduction?
Assessing the BESST index for reproduction treatment
Michael J. Davies1,
Jim X. Wang and
Robert J. Norman
Reproductive Medicine Unit, Department of Obstetrics and Gynaecology, University of Adelaide, Adelaide, Australia
1 To whom correspondence should be addressed at: Reproductive Medicine Unit, Department of Obstetrics and Gynaecology, University of Adelaide, Adelaide, Australia. e-mail: michael.davies{at}adelaide.edu.au
 |
Abstract
|
---|
It has been proposed that use of performance indicators for assisted reproduction treatment (ART) should be revised to better reflect the burden of treatment endured by a couple, and to place greater emphasis on the desired outcome of healthy babies. Recently, the BESST (birth emphasizing a successful singleton at term) score as a routine measurement for use in ART has been suggested. We applied the BESST index to a sample of ART patients and found that the BESST score was sensitive to the effects of patient age and extremes in the number of embryos transferred. However, the statistical properties of the index placed great weight on the effective implantation rate, which when applied to a time series in which implantation rates were improving, showed BESST scores to increase simultaneously with multiple pregnancy rates. This limits the completeness of the summary score as an expression of benefits compared to risks. A modified BESST index, not including cycles of initiated treatment as a denominator, indicated that the ART births contained a substantially lower proportion of BESST babies when compared to the entire population of births.
 |
Introduction
|
---|
There has been an increasing interest in using a new endpoint for success in assisted reproduction programmes. The basis for this is twofold. In the history of ART there have been numerous indicators for expressing the success of a treatment program, including implantation rate, pregnancy rate (usually clinical or ongoing pregnancy) per cycle (whether initiated, or oocyte recovery or embryo transfer cycle) and live birth rate per cycle. While each indicator has a valid application, the rationale for the use of a particular indicator should be explicit, as variation in numerator and denominator selection results in inconsistency of reporting and creates an opportunity for confusion in both the professional community and the recipient of care. Given the interests of the client and of the community, there is a clear need for an indicator reflecting progress toward the objective of infertility treatmentdelivery of a healthy baby following an unproblematic pregnancy. Consistent with an emphasis on outcomes that reflect the burden of treatment, Min et al. (2004
) advocate the use of an outcome measure in which success is defined as the delivery of live term birth of singleton baby. To this end, they have developed the BESST (Birth Emphasizing a Successful Singleton at Term) score, which is the chance of having a live term birth of singleton baby per ART cycle begun. The BESST outcomes are expressed as a proportion of all treatment cycles begunwhether or not the treatment has proceeded further. The rationale for this is that the inclusion of all cycles begun, regardless of treatment process, best represents the expectation by a couple, and better permits a patient commencing treatment to make an informed decision on their options at that time.
The motivations underlying these suggestions have undeniable merit. With the emergence of routine clinical services from experimental programs there is a need for a conceptual shift to endpoints in ART that reflect a number of interests, and which are explicitly patient oriented. A central problem is that pregnancy rate, however construed, is an inadequate endpoint for assessing ART success. Indeed, in the context where multiple embryos are transferred, the endpoint of pregnancy is highly correlated with multiple embryo transfer, which is known to be detrimental to the subjects clinical outcome and that of their baby(s). For instance, there is a growing imperative to address the prevalent and serious risks associated with iatrogenic multiple pregnancy resulting from multiple embryo transfer (ASRM, 1999
; ESHRE Capri Workshop, 2000). Resolving the dilemma of minimizing risks while maximizing the efficiency of care requires several endpoints to be considered simultaneously. In order to guide future clinical practice and inform patient choice, there is a need for clearly defined endpoints that express progress toward a range of outcomes, including the singleton live birth as suggested by the then Chairman of ESHRE (Evers, 2002
). We can envisage the need to develop endpoints related to efficacy, safety, patient satisfaction, community acceptability, and health economics of ART treatment. At present, pregnancy rate is being used as a principal measure of efficiency, and as an implied or partial surrogate endpoint for other outcomes. Systematic study on whether endpoints on each of these themes should be considered separately, or combined into a composite index for ease of reporting is likely to be highly instructive.
To assist in the development of the BESST score we have applied it to a time series of data in a large assisted reproduction program to assess whether it is sensitive to changes in clinical practice and outcomes.
 |
Subjects and methods
|
---|
This study used patient data recorded for infertility treatment in the Reproductive Medicine Unit in Adelaide, Australia, during the period 19822001 inclusive. A total of 17,054 initiated cycles, from 9045 women, were included in final analysis. The same inclusion and exclusion criteria as used by Min et al. (2004
) have been applied in this study. No other inclusion or exclusion criteria were applied.
The definitions of pregnancy and birth were also the same as reported by Min et al. (2004
). A pregnancy was first determined by the rise of serum
hCG above 10 IU/l on day 1618 after oocyte pick up (OPU). Ultrasound scan was used to detect fetal heart at about 6 weeks of gestation. After the first trimester, pregnant women were attended to by their own obstetricians. The outcomes of pregnancy were followed up by contacting the couple after the expected delivery date and further data on the outcomes of the pregnancy and delivery were obtained from database collected by the Perinatal Statistical Unit, Department of Human Service, South Australia.
We then calculated the BESST score per cycle begun, and also calculated the proportion of BESST births and compared it with that from the general population of South Australia.
The SPSS program was used to analyse the data.
2 test was used to detect the change of BESST rate over the study period and between the age and clinical treatment groups.
 |
Results
|
---|
The rate of BESST over a 20-year period is presented in Figure 1. There was an observed increase in the BESST score from
4% in 19831985 to over 10% for the period 19992000, rising to 15% for 2001. Analysis of the results over the three time periods, 19821985, 19861997 and 19982001, showed significant differences between the periods (P < 0.01). In order to assess which aspect of clinical performance has been responsible for the rise in BESST score, we have presented in Figure 2 a recent time series of the average number of embryos transferred per cycle of ART and the multiple pregnancy rate. It is apparent that over the period 19942002 there was a decline in the number of embryos transferred from 2.5 to 1.7 (P < 0.01). However, there was no consistent decline in the multiple pregnancy rate over the corresponding time period, and indeed, from 1998 to 2002 the multiple pregnancy rate rose from 17 to 20%. This may have been due to accumulated improvements in clinical and laboratory procedures that resulted in better implantation rates over the same time period. Hence, in the situation where implantation rates are increasing, the BESST score can increase simultaneously with the multiple pregnancy rate.

View larger version (18K):
[in this window]
[in a new window]
|
Figure 2. Time series showing the number of embryos transferred per embryo transfer cycle and multiple pregnancy rate over the period 19942002.
|
|
We also assessed the sensitivity of BESST for detecting differences between clinical practices and different age groups. To illustrate this, presented in Table I are the BESST scores for four maternal age groups and for the number of embryos, electively or non-electively, transferred. The BESST score was sensitive to the effect of maternal age, particularly with regard to the steep decline in positive outcomes after 35 years of age. The BESST score was also sensitive to the extremes of embryo transfer, with observed lower scores for non-elective transfer and for elective transfer of 3+ embryos compared to elective single embryo transfer. However, the difference was small and non significant between single and 2-embryo transfer groups. In our experience, single embryo transfer reduces the multiple pregnancy rate to that of the general population, compared to a multiple pregnancy rate of more than 20% in the 2-embryo transfer group.
Another potential use of BESST score is to show the proportion of optimal births in an ART program. Presented in Table I are the proportions of BESST in different age groups and with different number of embryos transferred. The results show that while younger women had the highest BESST score per cycle begun due to their higher implantation rate, they also had the lowest proportion of optimal births. In contrast, in the elective single embryo transfer group, there was consistency between the BESST score and the proportion of BESST births.
Figure 3 is a time series of the proportion of BESST in all ART births. The proportion averaged
60% (range 4880%) with no systematic change over the time period examined. In comparison, the proportion of BESST pregnancies in the general population of South Australia was in a narrow range around 90% over the same time period, which was substantially higher than overall ART births but comparable with that observed in the elective single embryo transfer group.
 |
Discussion
|
---|
Our calculated BESST score was found to co-vary with changes in ART practice over time. However, the index, by using initiated cycles as the denominator and due to the fact that over 70% of pregnancies were singleton pregnancies, was highly sensitive to changes in the overall pregnancy rate. This can result in the BESST score rising while the multiple pregnancy rate also increases, since both necessarily correlate with implantation rate. This limits the utility of the BESST score, as one impetus for the development of the BESST score was to provide a measure of the burden of adverse consequences experienced by patients, including for instance, multiple pregnancies that was unacknowledged in the use of crude pregnancy rates as the primary endpoint for ART programs. Hence, a non-significant variation in BESST score may contain a 20 percentage point or more variation between groups in the multiple pregnancy rates. Our results therefore highlight the difficulty of a single measurement serving all purposes, and particularly the difficulty of developing and using compound measures.
The BESST score indeed provides an accurate estimate at the commencement of treatment of the probability of a couple leaving with a term, live singleton baby. However, as the BESST score can be influenced dramatically by implantation rates, we can observe the public health paradox that the individual chance of successful treatment for a patient increases while the community burden of adverse consequences due to multiple pregnancies also rises unabated. The choice of denominator also underpins the reason why the BESST score was sensitive only between extremes in the number of embryos transferred.
BESST used as a proportion of all births did show the difference between elective single or double embryo transfer. It also showed that even non-elective multiple embryo transfer is associated with a lower proportion of BESST births. Using BESST as a parameter for comparing the proportion of optimal births between the ART group and the general population, we showed that there was a consistently lower proportion of unproblematic births following ART while the best outcome from elective single embryo transfer was probably comparable with that of the general population.
An additional issue related to the definition of BESST is the role of frozen embryo transfer (FET), pregnancies resulted from which should also be considered in the numerator for the BESST score.
 |
Conclusion
|
---|
While the motivation and the rationale for developing the BESST score is sound, its performance in assessing long term success with different numbers of embryos transferred indicates that supplemental information is necessary to accurately reflect the concurrent risk of multiple pregnancy associated with multiple embryo transfer (similar to the difference existing between the different ART programmes) and that additional information may be required to achieve the objectives of its design and to surpass the use of separate endpoints for evaluating ART practice. The proportion of BESST births in overall births may be used as the additional information for the overall normality of ART births in contrast to that in the general population.
 |
References
|
---|
Anonymous. (2000) Multiple gestation pregnancy. The ESHRE Capri Workshop Group. Hum Reprod 15,18561864.[Abstract/Free Full Text]
ASRM. (1999) Guidelines on number of embryos transferred. A practice committee report: A committee opinion. American Society for Reproductive Medicine, Birmingham, USA. This full bulletin can now be accessed only by members of ASRM on the following website: http://www.asrm.org/Professionals/movedtomembersonly.html (as of 7th March 2004).
Evers JL (2002) Female subfertility. Lancet 360,151159.[CrossRef][Medline]
Min JK, Breheny SA, MacLachlan V and Healy DL (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 19,37.[Abstract/Free Full Text]