A real‐life prospective health economic study of elective single embryo transfer versus two‐embryo transfer in first IVF/ICSI cycles

Human Reproduction Vol.19, No.4 pp. 917±923, 2004 Advance Access publication February 27, 2004 DOI: 10.1093/humrep/deh188 A real-life prospective health economic study of elective single embryo transfer versus two-embryo transfer in ®rst IVF/ICSI cycles J.Gerris1,5, P.De Sutter2, D.De Neubourg1, E.Van Royen1, J.Vander Elst2, K.Mangelschots1, M.Vercruyssen1, P.Kok2, M.Elseviers3, L.Annemans4, P.Pauwels1 and M.Dhont2 1 Centre for Reproductive Medicine, Middelheim Hospital, Lindendreef 1, 2020 Antwerp, 2Centre for Reproductive Medicine, Ghent University Hospital, De Pintelaan 185, 9000 Ghent, 3Department of Biomedical Statistics, Antwerp University, Wilrijkstraat 10, 2650 Edegem and 4Department of Public Health, Ghent University, De Pintelaan 185, 9000 Ghent, Belgium 5 To whom correspondence should be addressed: E-mail: Key words: assisted reproductive technology/health economic analysis/single embryo transfer/single versus two-embryo transfer Introduction There has been a global increase in the incidence of multiple pregnancies, which is mainly due to the widespread application of assisted reproductive technologies. It is estimated that IVF and ICSI account for half of the increase while non-IVF treatments are responsible for the other half. According to national reports, the incidence of multiple pregnancy after IVF varies between 20 and 30%. Approximately 500 000 IVF/ICSI cycles are performed in the whole world each year (Nygren and Nyboe-Andersen, 2002). These result in ~100 000 ongoing pregnancies; a large number of the children resulting from these pregnancies are part of a set of twins or high order multiples (HOM). To give a rough estimate, if the global incidence for post-IVF/ICSI twinning is assumed to be 25% and for HOM to be 3%, these ~100 000 ongoing pregnancies will result in 72 000 singletons, 50 000 twin children and 9000 triplet children for a total of 131 000 children. Assuming an incidence of ~10% of severe complications and sequelae per child belonging to a set of twins or HOM (Wennerholm and Bergh, 2000), this means that each year, IVF/ICSI alone is responsible for ~6000 severely disabled children. In addition, non-IVF treatments are responsible for at least a similar number of multiples. Apart from severe complications, e.g. cerebral palsy, there are other physical and mental complications, as well as non-medical problems, such as educational, emotional, (neuro)linguistic, ®nancial, familial and sexual consequences that usually accompany the raising of twins (Dhont et al., 1999). A reasonable solution to the recent epidemic of twins would be to transfer only one embryo. However, the pressure on both doctors and patients to increase the success rate of an expensive and stressful treatment has made the transfer of two or more embryos the standard of care. Many clinicians are reluctant to introduce single embryo transfer (SET) because of fear of lowering their results. However, the judicious application of elective single embryo transfer is a feasible option which has been demonstrated to Human Reproduction vol. 19 no. 4 ã European Society of Human Reproduction and Embryology 2004; all rights reserved 917 BACKGROUND: We analysed the difference in maternal, neonatal and total costs after single (SET) versus double day 3 embryo transfer (DET). METHODS: We performed a two-centre prospective study of women in their ®rst IVF/ICSI cycle choosing between SET or DET. Infertility treatment data were gathered from a database; maternal and neonatal outcome data from a case report form (CRF); health economic data from medical acts registered in the CRF for the outpatient part and from hospital bills. SET was performed in 206/367 (56.1%) and DET in 161/367 (43.9%) women. RESULTS: In all, 367 transfers yielded 186 positive pregnancy tests, 148 ongoing pregnancies and 136 live deliveries (50.7, 40.3 and 37.1% per embryo transfer) of which 15 (11.0%) were twins. Live birth rate was 37.4% for SET, 36.6% for DET. Intention-to-treat analysis showed differences for: duration of pregnancy (SET: 39.0 6 1.4 versus DET: 38.3 6 2.2 weeks; P = 0.055), percentage prematurity (8.5 versus 23.8%; P = 0.033), percentage of neonates hospitalized (5.7 versus 17.9%; P = 0.121) and duration of neonatal hospitalization (6.3 6 2.2 versus C4700 6 3239 versus DET: = C8613 6 10 105; 10.3 6 10.1 days; P = 0.01). Total cost after DET was higher (SET: = C451 6 957 versus = C3453 6 8154; P < 0.001) and not to P = 0.105), due to signi®cantly higher neonatal costs (= C4250 6 2882 versus = C5160 6 4106; P = 0.152). CONCLUSIONS: This prospective differences in maternal costs (= health economic study shows that transfer of a single top quality embryo is equally effective as, but substantially cheaper than, double embryo transfer in women <38 years of age in their ®rst IVF/ICSI cycle. J.Gerris et al. Materials and methods Inclusion criteria Patients were recruited in two IVF/ICSI programmes including all accepted indications for this treatment. They had to ful®l the following inclusion criteria: <38 years of age at the time of embryo transfer; ®rst IVF/ICSI treatment ever or after a previous delivery, whether or not that pregnancy was the result of an infertility treatment. Sample size Prior to the initiation of the study, a calculation of the study group size was made using a Monte Carlo analysis for 10 000 patients entered into the model using ®gures from the only (theoretical) economic evaluation model comparing SET with DET available at the time the study was conceived (Clark, 1997; Wùlner-Hanssen and Rydstroem, 1998). In this model, the cost of one treatment cycle applying both SET or DET was estimated at = C2000; the average cost for a singleton C1625; the cost for a twin-complicated pregnancy at pregnancy at = = C4500; and neonatal costs were estimated at = C7000 per twin child and at = C2125 for a singleton. In our pre-study estimate, an ongoing pregnancy rate per cycle of 40% was used, based on prior published data (Gerris et al., 1999). The Monte Carlo evaluation provided an estimate of the SD of the expected cost. Based on this estimation, an alpha value of 0.05 and a power of 90%, the required sample size was calculated. It was calculated that a total number of 50 patients in each group would be suf®cient to demonstrate a signi®cant cost difference with P < 0.05. However, anticipating a lower than usual (25±30%) incidence of twins in the study of ~15%, it was calculated that a total number of patients of ~350 would be needed, even when unequally distributed (e.g. 25/75%) between both study arms. The present study sample of 367 patients eligible for analysis corresponds to these target numbers. Clinical and laboratory protocols All IVF/ICSI cycles were performed according to accepted standard protocols of pituitary suppression, ovarian stimulation, oocyte retrieval, gamete handling and embryo culture techniques, embryo 918 transfer and luteal supplementation, as described (...truncated)