ABCDEFG

UNIVERS ITY OF OULU P.O.B . 7500 F I -90014 UNIVERS ITY OF OULU F INLAND

A C T A U N I V E R S I T A T I S O U L U E N S I S

S E R I E S E D I T O R S

SCIENTIAE RERUM NATURALIUM

HUMANIORA

TECHNICA

MEDICA

SCIENTIAE RERUM SOCIALIUM

SCRIPTA ACADEMICA

OECONOMICA

EDITOR IN CHIEF

PUBLICATIONS EDITOR

Senior Assistant Jorma Arhippainen

University Lecturer Elise Kärkkäinen

Professor Hannu Heusala

Professor Olli Vuolteenaho

Senior Researcher Eila Estola

Information officer Tiina Pistokoski

University Lecturer Seppo Eriksson

Professor Olli Vuolteenaho

Publications Editor Kirsti Nurkkala

ISBN 978-951-42-8882-1 (Paperback)ISBN 978-951-42-8883-8 (PDF)ISSN 0355-3221 (Print)ISSN 1796-2234 (Online)

U N I V E R S I TAT I S O U L U E N S I S

MEDICA

ACTAD

D 985

ACTA

Zdravka Veleva

OULU 2008

D 985

Zdravka Veleva

FACTORS AFFECTINGTHE OUTCOME OF IVF/ICSI

FACULTY OF MEDICINE INSTITUTE OF CLINICAL MEDICINE, DEPARTMENT OF OBSTETRICS AND GYNECOLOGY,UNIVERSITY OF OULU

A C T A U N I V E R S I T A T I S O U L U E N S I SD M e d i c a 9 8 5

ZDRAVKA VELEVA

FACTORS AFFECTINGTHE OUTCOME OF IVF/ICSI

Academic Dissertation to be presented, with the assentof the Faculty of Medicine of the University of Oulu, forpublic defence in Auditorium 4 of Oulu UniversityHospital, on September 26th, 2008, at 12 noon

OULUN YLIOPISTO, OULU 2008

Copyright © 2008Acta Univ. Oul. D 985, 2008

Supervised byProfessor Juha S. TapanainenDocent Hannu Martikainen

Reviewed byDocent Antti PerheentupaDocent Anne-Maria Suikkari

ISBN 978-951-42-8882-1 (Paperback)ISBN 978-951-42-8883-8 (PDF)http://herkules.oulu.fi/isbn9789514288838/ISSN 0355-3221 (Printed)ISSN 1796-2234 (Online)http://herkules.oulu.fi/issn03553221/

Cover designRaimo Ahonen

OULU UNIVERSITY PRESSOULU 2008

Veleva, Zdravka, Factors affecting the outcome of IVF/ICSIFaculty of Medicine, Institute of Clinical Medicine, Department of Obstetrics and Gynecology,University of Oulu, P.O.Box 5000, FI-90014 University of Oulu, Finland Acta Univ. Oul. D 985, 2008Oulu, Finland

AbstractFertility declines with advancing age and the number of couples seeking infertility treatment at anolder age is constantly increasing. A top quality embryo is believed to have the highest potentialfor implantation and development into a child. A better understanding of the relative importanceof patient and treatment characteristics and of embryo quality could help to optimise the existingtherapeutic schemes and the safety of in vitro fertilisation/intracytoplasmic sperm injection (IVF/ICSI).

In this work, databases of five Finnish infertility clinics were studied retrospectively. Data ontreatments performed in the years 1994–2005 were collected. A total of 19,000 treatment cycleswere analysed. Special attention was paid to the relative significance of the transfer of top qualityembryos with regards to pregnancy, miscarriage, live birth and cost of treatment in the generalIVF/ICSI patient population and in groups with expected poor outcome.

The results showed that the transfer of a top quality embryo is associated with a better chanceof pregnancy and live birth. However, it does not diminish the probability of miscarriage. Bothlow and high BMI increase the miscarriage rate. Advancing age and a positive history of previousmiscarriages are also associated with a higher probability of miscarriage. In addition, the need forhormonal substitution in cases of frozen-embryo transfer is a risk factor of miscarriage, probablybecause of suboptimal endometrial function.

Since the transfer of several embryos leads to multiple pregnancies, which are associated witha high risk of maternal and fetal complications, elective single embryo transfer (eSET) of a topquality embryo allows all additional good quality embryos to be frozen and transferred later infrozen-thawed embryo transfer cycles. The present work demonstrates that eSET is a safetreatment strategy at least until the age of 40. However, it might not be performed in women withfewer than four collected oocytes, since the prognosis might remain poor even if the response isimproved in a following cycle.

When eSET is applied routinely and on a large scale, it diminishes treatment costs whileincreasing the number of deliveries occurring at term, making IVF/ICSI at the same time safer andmore affordable even to patients without access to reimbursed IVF treatment.

Keywords: cost effectiveness, elective single embryo transfer, fertilisation in vitro,frozen-thawed embryo transfer, low response, obesity, spontaneous abortion, top qualityembryo

5

Acknowledgements

The idea for this work originates back in the spring of 2002 when Juha S. Tapanainen asked me whether I would be interested in doing research in Oulu. He was bold enough to say the PhD would take only a couple of years and I was enthousiastic enough to believe it. However, in the six years that the thesis took to form I have had the pleasure of taking part in an exciting project with not only one, but two supervisors and an amazing amount of unanalysed data.

Juha did everything possible to arrange my smooth acclimatisation to life (and work!) in Oulu. His insight into reproductive endocrinology and capability to solve practical problems were invaluable guidance. Juha was always ready to provide a share of his deep knowledge of everything – from hormone measurements to gold mining in Lapland. He also introduced me to the Finnish language by giving me the only computer in the clinic with the Finnish version of Windows in it. Juha always found time to discuss or think things over, even when he obviously didn’t have that time because of his many commitments. At the same time, he never cared to know how many hours I had been working a week and always agreed to as many days of holidays as I asked for. He also arranged for the group grants which fuelled our work throughout the years.

I had the privilege of having Hannu Martikainen as my other supervisor. Hannu has been an inspiring supervisor and reliable mentor during all of my studies in Oulu, always with at least a few fresh study ideas. I am humbled by his ability to effortlessly grasp any new information in a matter of seconds. After I managed to survive the “tuntematon virhe” messages on my screen and also the University language courses, Hannu also very patiently spoke to me in Finnish until I started communicating my work progress in that language as well. Hannu has given me mental support during data transformation procedures and has calmly corrected all sorts of errors in early- and later-stage tables and texts. Last but not least, I am also grateful for his advice on Spanish paintings.

The heart of this project was the databases of the Oulu IVF clinics, which started working at a time when I had hardly ever heard of in vitro fertilisation. To these, for some of the projects I joined data from three other sources. During this work I had the honour of cooperating with Aila Tiitinen from the University of Helsinki who consistently and “ruthlessly” asked questions about the current project and thus helped create the more sophisticated discussion parts. Aila is also responsible for some of the fine tuning of criteria to use simultaneously on data from four different laboratories. Special thanks go to Christel Hydén-Granskog, the other

6

member of the HUKS team. It has been wonderful to rely on somebody who would cheerfully fill in missing data fields all the while doing pressing daily work, and who would also entertainingly explain all kinds of embryological particularities.

Data from the Family Federation of Finland in Helsinki were arranged for by Sirpa Vilska. She not only kindly assisted me during my data collecting trips, but also showed me the nicer aspects of the southern capital. I also thank Sinikka Nuojua-Huttunen from the Family Federation of Finland in Oulu, for data assistance (at a certain point almost daily) and for frequent advice on things such as ovarian stimulation and cross-country skiing.

Candido Tomás not only arranged for data from the AVA-Clinic in Tampere but also made and upgraded the Babe database together with Hannu. I am grateful for his always sending data on time and in perfect order. I thank Petri Karinen from the Department of Otorhinolaryngology for help with cost-effectiveness issues. Ilkka Järvelä’s early instructions on miscellaneous issues, including Babe and SPSS, should also be acknowledged. By far, this was an invaluable learning experience.

I am grateful to Anne-Maria Suikkari and Antti Perheentupa for useful suggestions in improving the thesis as external reviewers. I would also like to thank Mika Gissler from STAKES who kindly sent me the latest statistical data on Finnish IVF to include in the thesis, Kaarin Mäkikallio-Anttila for the critical assessment of the cost-effectiveness manuscript when it was still at the embryonic stage, Nick Bolton and Simon Brown for help with the English language.

This work has been financially supported by grants from the Sigrid Jusélius Foundation, the Finnish Academy and the Oulu University Scholarship Foundation.

Working with databases without formal training in statistics is not particularly easy, but this process was greatly helped by the excellent mathematics and physics teachers I had had at school. Thank you, Mrs. Stoyanova and Messrs. Stefanov, Lorensen and Ross! In Oulu, I am indebted to Risto Bloigu for being a compassionate listener and for his eagerness to discuss regressions even in the later hours of the day. I would also like to thank Esa Läärä for his enlightening course on the language R, which was a most wonderful demonstration of statistics’ anatomy and basic physiology.

This work would not have started without Aydin Tekay’s appealing website and especially without Markku Ryynänen who was my friendly tutor during the

7

unforgettable one-month exchange clerkship at the Department of Obstetrics and Gynecology after the fifth year in medical school.

I would like to thank all other staff at the Infertility Unit of the Oulu University Hospital and also of the sister infertility clinic of the Family Federation of Finland in Oulu for making me feel welcome here and for help with miscellaneous information and data retrieval. In particular, Laure Morin-Papunen and her family offered their friendship, sauna and pulla even during the very dark first winter. I would also like to thank Mauri Orava, Jouni Lakkakorpi and Riitta Koivunen for their constructive help and constant encouragement. I would also like to thank all doctors and nurses of the gynaecological policlinic of the Department of Ob/Gyn and especially Eila Suvanto-Luukkonen for kindness during my amanuenssi experience.

I am indebted to Rais Latypov for his crucial moral support during the early years of this work. I want to thank (in alphabetic order) my fellow Ph.D. students and junior “seniors”, some of whom nowadays have even higher positions: Hilkka Ijäs, Johanna Kesti, Katriina Rautio, Marianne Hinkula, Marja Simojoki, Marko Niemimaa, Mervi Haapsamo, Minna Jääskeläinen, Outi Uimari, Reeta Törmälä, Terhi Piltonen and Tommi Vaskivuo, for their stimulating talks and good company.

I thank all people who peeked at the kakkoshuone with a friendly smile and most of all Juha Räsänen for his quiet encouragement and inspiring work. It was a pleasure to rely in all bureaucratic matters on our secretaries Maire Syväri and Raili Puhakka who created calm even in the afternoon of important deadlines.

I extend warm thanks to Sari Ylipieti for making the transition to clinical work as easy as possible and for believing I can do the job of a Nordic general practitioner even when I was not sure of it myself. I am also grateful to Eija Sipilä for arranging my schedules in such a way as not to interfere with the completion of the thesis.

Thanks to my friends and to my family, I have been reminded that life exists inside and outside virtual reality but always real-time. They have always been ready to assist me in any imaginable way – gastronomical, musical, humorous, logistic – at any time of the day. It has been wonderful to feel the presence of you all near me.

My loving thanks to Vitali, for his everlasting, unconditional support throughout all these years. Without you, this work would probably have never come into existence.

Oulu, 25 August 2008 Zdravka Veleva

8

9

Abbreviations

Ach acetylcholine AMH anti-Müllerian hormone BMI body mass index CC clomiphene citrate CEA cost-effectiveness analysis CI confidence interval cSET compulsory single embryo transfer DET double embryo transfer ET embryo transfer eSBT elective single blastocyst transfer eSET elective single top quality embryo transfer FET frozen-thawed embryo transfer FISH fluorescent in situ hybridisation FSH follicle-stimulating hormone GH growth hormone GnRH gonadotropin-releasing hormone hCG human chorionic gonadotropin HMG human menopausal gonadotropin HR high ovarian response ICER incremental cost-effectiveness ratio ICSI intracytoplasmic sperm injection IGFBP-1 insulin-like growth factor-binding protein-1 IR implantation rate IVF in vitro fertilisation LBR live birth rate LH luteinising hormone LR low response LR→NR low response, followed by a normal response in a consecutive cycle MBR multiple live birth rate MR miscarriage rate NR normal response NR→LR normal response, followed by a low response in a consecutive cycle nt-eSET elective single embryo transfer of a non-top quality embryo OD oocyte donation OHSS ovarian hyperstimulation syndrome

10

OPU ovum pick-up PCOS polycystic ovary syndrome PCR polymerase chain reaction PGS prenatal genetic aneuploidy screening PR pregnancy rate ROS reactive oxygen species SD standard deviation tLBR term live birth rate WHO World Health Organisation

11

List of original articles

This thesis is based on the following articles, which are referred to in the text by their Roman numerals:

I Veleva Z, Järvelä IY, Nuojua-Huttunen S, Martikainen H & Tapanainen JS (2005) An initial low response predicts poor outcome in in vitro fertilization/intracytoplasmic sperm injection despite improved ovarian response in consecutive cycles. Fertilility and Sterility 83: 1384–1390.

II Veleva Z, Tiitinen A, Vilska S, Hydén-Granskog C, Tomás C, Martikainen H & Tapanainen JS (2008) High and low BMI increase the risk of miscarriage after IVF/ICSI and FET. Human Reproduction 23: 878–884. Epub 2008 Feb 15.

III Veleva Z, Vilska S, Hyden-Granskog C, Tiitinen A, Tapanainen JS & Martikainen H (2006) Elective single embryo transfer in women aged 36-39 years. Human Reproduction 21: 2098–2102. Epub 2006 Jun 1.

IV Veleva Z, Tomàs C, Tapanainen JS and Martikainen H (2008) Elective single-embryo transfer with cryopreservation improves the outcome and diminishes the costs of IVF/ICSI. Manuscript.

12

13

Contents

Abstract Acknowledgements 5 Abbreviations 9 List of original articles 11 Contents 13 1 Introduction 15 2 Review of the literature 17

2.1 History of IVF/ICSI ................................................................................ 17 2.2 Measuring the success of IVF/ICSI ........................................................ 19 2.3 Embryo quality........................................................................................ 20

2.3.1 Morphological .............................................................................. 20 2.3.2 Developmental stage .................................................................... 21 2.3.3 Genetics ........................................................................................ 22 2.3.4 Metabolism................................................................................... 23 2.3.5 Quality of frozen-thawed embryos ............................................... 24 2.3.6 Male and female factors and embryo quality ............................... 24

2.4 Endometrial receptivity........................................................................... 25 2.5 Female age .............................................................................................. 28

2.5.1 Ageing and the ovarian pool......................................................... 28 2.5.2 Ageing and oocyte quality............................................................ 29 2.5.3 Female age and outcome of IVF/ICSI.......................................... 30 2.5.4 Individual ageing rates ................................................................. 31

2.6 Low ovarian response ............................................................................. 32 2.6.1 Prognostic factors of LR............................................................... 33 2.6.2 Management of LR....................................................................... 34

2.7 BMI as an outcome predictor of IVF/ICSI.............................................. 35 2.7.1 Overweight, obese and insulin resistant women........................... 35 2.7.2 Underweight women .................................................................... 39

2.8 Main infertility diagnosis ........................................................................ 40 2.8.1 PCOS............................................................................................ 40 2.8.2 Other diagnoses ............................................................................ 41

2.9 FET ......................................................................................................... 41 2.10 Number of transferred embryos and pregnancy outcome ....................... 42 2.11 eSET with FET........................................................................................ 42

2.11.1 Physicians’ considerations............................................................ 44

14

2.11.2 Patients’ preferences ..................................................................... 45 2.12 Cost-effectiveness analysis ..................................................................... 45

2.12.1 Cost of IVF/ICSI treatment .......................................................... 47 2.12.2 Cost-effectiveness of eSET........................................................... 48

2.13 Alternatives to eSET ............................................................................... 48 3 Aim 51 4 Materials and Methods 53

4.1 Study population ..................................................................................... 54 4.2 Treatment protocols................................................................................. 54 4.3 Cost of IVF/ICSI ..................................................................................... 56 4.4 Statistical analyses of clinical data.......................................................... 56 4.5 Cost-effectiveness analysis ..................................................................... 58 4.6 Data protection ........................................................................................ 59

5 Results and discussion 61 5.1 Value of a top quality embryo in LR ....................................................... 61 5.2 Value of a top quality embryo in fresh and FET cycles........................... 63 5.3 Top quality embryos and miscarriage ..................................................... 63

5.3.1 Factors affecting MR: age ............................................................ 64 5.3.2 Factors affecting MR: BMI .......................................................... 67 5.3.3 Factors affecting MR: type of cycle ............................................. 67 5.3.4 Other factors affecting MR........................................................... 69

5.4 eSET and age........................................................................................... 69 5.4.1 Value of a non-top quality embryo in eSET ................................. 71

5.5 Embryo transfer policy in cycles with top quality embryo(s) ................. 71 5.5.1 Overall outcome in the years 1995–2004 ..................................... 71 5.5.2 Cost-effectiveness of eSET policy................................................ 75 5.5.3 Limitations of the study................................................................ 79

6 Conclusions 81 References 83 Original papers 109

15

1 Introduction

About 15% of all couples are involuntarily infertile [1] and require fertility treatment. Thirty years after the birth of the first infant after IVF, the number of children born worldwide as a result of IVF/ICSI already exceeds three million [2]. Clinical and laboratory procedures have been constantly improving and nowadays pregnancy rates of about 30% per transfer are routinely reported [3]. IVF/ICSI is therefore an important method for correcting unfavourable demographic indices despite associated costs, being as high as one quarter of annual household expenditure in some countries [4].

The biggest challenge for present-day IVF/ICSI is the high proportion of multiple pregnancies, which significantly affect even nationwide demographics [5]. Multiple pregnancies are associated with a worsened obstetric outcome as a result of prematurity [6]. Consequently, the infants have a significantly elevated risk of life-long complications such as cerebral palsy [7], language development delay and behavioural problems [6]. Because of this, the parents of multiples often suffer from exhaustion and have an elevated risk of depression [6].

The cause of multiple pregnancies is the transfer of several embryos at a time. In Europe, more than one embryo is transferred in 84% of IVF/ICSI cycles [3], while in the United States this number is even higher (≈99%) [8]. The transfer of one embryo is the only efficient way to minimise the number of multiple pregnancies. There is accumulating evidence that if several embryos are created and one of them is selected for transfer on the basis of good morphology, PR and LBR are high [9-11]. All other good quality embryos can be frozen and transferred later. This strategy results in a cumulative PR of up to 60% in young women [12, 13].

However, the eSET strategy is subject to criticism, as it is widely believed that if only one embryo is transferred, success rates will diminish and the price of treatment will rise [14-16]. Part of this scepticism is based on the fact that embryo morphology does not absolutely correlate with its pregnancy potential [17]. The zygote receives most of its cytoplasm from the oocyte and during the first five or six days embryo development is under the control of factors synthesised in the ooplasm [18]. This means that before embryos are transferred into the uterus, their potential cannot be fully predicted even if they have top quality morphology.

Finland is one of the pioneer countries as regards eSET. Experience with eSET at Oulu University Hospital started as early as in 1996. In the present study, the relative importance of the transfer of a top quality embryo on the outcome of

16

IVF/ICSI treatment was studied by using the computerised databases of five Finnish infertility clinics. Pregnancy, miscarriage and live birth rates were studied in women approaching the end of their fertile period, in fresh IVF/ICSI and in FET cycles. Most importantly, the overall effect of eSET on the outcome of the whole IVF/ICSI programme and on the costs of treatment was evaluated.

17

2 Review of the literature

2.1 History of IVF/ICSI

The milestones of IVF/ICSI are presented in Fig. 1. Initially, IVF was performed in a natural cycle. Since only one oocyte was collected at a time, the success rate was very low, with a LBR of only 9.6% [19]. Treatment was also limited only to women with spontaneous ovulation. The introduction of ovarian stimulation in 1981, first using clomiphene citrate [20] and later gonadotropins [21] and GnRH agonists [22] increased the number of collected oocytes and embryos available for transfer. At the same time, indications for IVF treatment broadened to anovulation of different aetiologies. Later on, ICSI allowed severe male reproductive disorders to be treated as well [23].

Initially, the simultaneous transfer of several embryos was favoured because of the low chance of implantation. In 1991, three embryos were transferred in 40.5% and more than 3 embryos in 24.6% of cycles, but the overall LBR per started cycle was only 12% [24]. Despite the low success rate, 25% of the deliveries were multiple births. After the identification of embryo quality characteristics it became clear that the transfer of several good quality embryos dramatically increased the multiple pregnancy rate, to 31–32% if three embryos were transferred [25, 26]. The proportion of triplet and higher-order multiple pregnancies also increased and was as high as 18% if three good quality embryos were transferred [27].

The high multiple pregnancy rate has exposed a large number of children to health risks. Compared with singleton births, perinatal mortality rates are at least four-fold higher for twins and at least six-fold higher for triplets [6]. Twins born after IVF have almost a ten-fold higher risk of being born prematurely than IVF singletons, and because of this a 3.8-fold increased risk of admittance to a neonatal intensive care unit [28]. These risks are even higher in children from higher-order pregnancies [6] who are born before the 37th gestational week with few exceptions (1%) [29]. Multiples also suffer from long-term medical and developmental problems. The risks of prematurity-associated neurological sequelae such as cerebral palsy are 3–7 times higher in twins and over 10 times higher in triplets, compared with singletons [6].

18

Fig. 1. Milestones of in vitro fertilisation.

The unacceptably high number of triplet and higher-order pregnancies prompted a restriction on the number of embryos transferred. The transition from triple embryo transfer to DET was started in 1993 [26, 27, 30] and was adopted early on in Europe, but not in the United States [31, 32]. More than five embryos were transferred in a total of 4350 cycles (about 10% of all cycles) in the United States during one calendar year [33].

As already observed in 1993 [27], the transfer of no more than two embryos at a time almost eliminated the triplet and higher-order multiple pregnancies, but not twin gestations and as a result the overall MBR remained high (25–40%) [31, 33-38]. Following publication of the morphological criteria for a top quality embryo [39, 40], the transition from DET to eSET was started. In eSET, one top quality embryo is transferred after ovum pickup. All other good quality embryos are frozen and can be transferred in a FET cycle. eSET was first described in women with medical contraindications to multiple gestation [9]. Since then, a

1975

1978

1981

19831984

19921993

199920002001

2006

First IVF pregnancy (ectopic) in a spontaneous cycle

First child born IVF after

Ovarian stimulation for IVF with clomiphene citrate

First birth after FET; start of Finnish IVF programme; gonadotropin stimulationOvarian stimulation with GnRH agonist

First pregnancy after ICSIChange from triple ET to DET initiated

ESHRE recommends SET

ASRM recommends SET

First study of eSET; publishing of top embryo criteriaESHRE recogni es multiple pregnancies complications as major IVF/ICSI

19

multitude of trials have shown the effectiveness of eSET in different selected groups of subjects, typically in their first IVF/ICSI cycle and aged less than 36 years [10, 41-45].

Currently, eSET is the recommended strategy in both Europe [35] and the United States [46], but clinical practice shows little adherence to the guidelines outside the Nordic countries [3, 5]. In 2003, there were 37,000 multiple births after IVF in Europe [3], the United States [5] and Canada [47]. More IVF cycles are performed each year and the total number of multiple births, especially twin births, is constantly increasing [3, 5, 8, 36, 37, 48] so that between 1998 and 2003, the number of twin births in the whole population of the United States grew by 17% [5]. eSET is practiced on a large scale in Finland, Belgium and Sweden, where it is regulated by law (in Belgium and Sweden) or by consent of IVF units (Finland) [3]. Finnish data from the years 1996–2006 show that the number of multiple births after IVF/ICSI decreased from 24% to 11%, while the LBR has remained stable [49]. Similarly, Swedish statistics also show that LBR was maintained at around 26% while the MBR decreased dramatically, from about 35% to around 5% in the period 1991–2004 [38].

2.2 Measuring the success of IVF/ICSI

Traditionally, IVF outcome has been expressed in terms of biochemical or clinical pregnancies and total live births. However, these parameters also include multiple gestations and births, which are considered to be complications of IVF because of the dimensions of their prematurity-related problems [6, 35]. This is why a lengthy discussion on the definition of the main outcome of IVF treatment ended without reaching consensus [50]. Competition between IVF clinics has resulted in outcome being expressed for selected groups of patients (biochemical PR/ET, for example). This can be illustrated with data from a study of BMI, in which the biochemical pregnancy rate/ET among women with normal BMI was 34.2% (1033/3018) while for the same group the LBR/cycle was “only” 20.8% (718/3457) [51]. In addition, the outcome of a stimulation cycle can be fully evaluated only if the outcome of FET is also taken into account in cumulative PR or LBR [52]. In 2003, the LBR after all fresh cycles performed in Finland was 21.5% while the cumulative LBR/OPU after one or several FET cycles per woman was much higher (31.0%) [3]. This can be explained by the policy of extensive use of cryopreservation in Finland. A drawback of cumulative outcome is that sufficient time should be allowed for all FET cycles to take place. This is

20

why the design of randomized studies almost never reflects general practice. For example, in a study of eSET, half of the patients had between four and 16 good quality embryos, but the study involved evaluation of the outcomes of the transfer of two embryos only [42]. In addition to several FET cycles, IVF/ICSI patients frequently undergo more than one stimulation cycle as well. While cumulative outcome from several stimulations has been reported [51, 53, 54], the full effect of treatment per subject has not been evaluated until now.

It is difficult to decide whether singleton pregnancies are the desired outcome of treatment (singleton LBR), as even singletons after IVF are often born prematurely [55]. However, with more than three million children born as a result of IVF/ICSI, the technology is not a novelty treatment anymore [2]. Criteria already in force for evaluation of spontaneous pregnancies should be used for the outcome of IVF/ICSI gestations as well. According to the WHO, the pregnancies with the best outcome are usually delivered between 38 and 42 weeks of gestation (term live birth). Therefore, the tLBR is the most logical measure of the clinical outcome of IVF/ICSI, with both singletons and twins included, since twins born at term generally have a better prognosis than premature ones. Up to now, the tLBR has been used only in a few investigations [56, 57].

2.3 Embryo quality

2.3.1 Morphological

The assessment of embryo quality in IVF/ICSI is essential and it determines the number of embryos to be transferred and frozen. Cleavage speed and degree of fragmentation were the first parameters to be assessed [58]. Other morphological aspects found to be related to embryo quality were cytoplasm appearance, blastomere irregularity and degree of fragmentation [59]. Together with cleavage rate, these factors were evaluated in the first embryo quality scoring method (the cumulative embryo score) which was aimed at optimising PR while avoiding triplet and other higher-order pregnancies [59].

At present, morphological evaluation of the embryo routinely includes zona pellucida assessment, as a thick zona pellucida has been found to affect fertilisation negatively [60]. Multinucleation of blastomeres is another factor that diminishes the chance of pregnancy [61]. Embryo quality is expressed not only in terms of morphology and cleavage speed two or three days after fertilisation but

21

also as regards the time of the first mitotic division [62, 63]. Embryos which complete the first mitotic division within 25–27 h after insemination have been associated with higher PR (40.5% vs. 31.3%), compared with late-cleaving ones [63].

The first grading score of individual embryos was published in 1995 after analysis of single embryo transfers on day 2; the best embryos had 4 regular cells and lacked anucleated fragments [39]. The criteria were further extended so that the definition of a top quality embryo included no multinucleated blastomeres, ≤20% anucleated fragments and four or five blastomeres on day 2, or seven or more blastomeres on day 3 [40]. The results of an analysis by Van Royen and colleagues suggested that eSET might be considered if a top quality embryo is available [40]. In a study of eSET, transfer of a single top quality embryo resulted in a PR of 44%, while after the transfer of one non-top embryo the PR was only 19% [64]. This grading system, with slight modifications, is used in most centres that perform eSET.

2.3.2 Developmental stage

Most often embryos are transferred on day 2 or 3 after fertilisation in order to allow selection of the most suitable ones for transfer. Cultures supporting later embryo development have been used since the late 1990s [65]. Embryo implantation into the endometrium takes place on day 5 and therefore the transfer of blastocysts on that day is aimed at improving the synchronicity of endometrial and embryonic development. Good quality blastocysts are also supposed to have higher pregnancy potential compared with cleavage stage (day 2–3) embryos, since not all embryos survive until the blastocyst stage.

There have been conflicting reports on the advantages of blastocyst transfer. In a recent meta-analysis the outcomes of prospective randomized trials were reviewed and no difference in PR per couple in good prognosis patients was found (day 2–3 38.8% vs. blastocyst 40.3%) [66]. Furthermore, there was no difference in MR per couple between the two groups. The rate of embryo freezing per couple was over two times higher in cleavage-stage transfers, while in the blastocyst group there were more cases in which no embryos were available for transfer (OR 3.21). To date, there is insufficient data for a comparison of the outcome of FET cycles among cleavage-stage and blastocyst groups.

22

2.3.3 Genetics

In the early embryo, cell cycle control is performed by factors synthesised in the oocyte before fertilisation. Expression of the embryonic genome increases gradually from the 8-cell stage onwards [18] and genetically abnormal embryos are gradually eliminated by natural selection. Due to maternal control of early divisions, a significant proportion of morphologically normal IVF/ICSI embryos have genetic abnormalities [67]. Consequently, it has been speculated that genetic analysis of the embryo might help in selection of those with normal composition and might improve the LBR.

In prenatal genetic analysis, a single cell is removed from the embryo. Usually, this happens on day 3 although it is also possible in the blastocyst stage when several cells can be biopsied at the same time with more reliable results [68]. Aneuploidy screening by means of FISH (PGS) is then performed on the biopsied cell(s) [69]. Genetic analysis by means of PCR can also be performed if there is risk of transmission of monogenetic diseases [70, 71]. In Finland, genetic analysis has been performed on embryos of carriers of translocations and of infantile neuronal ceroid lipofuscinosis [72].

Since PGS is an expensive procedure, it is currently performed in cases with an expected high risk of chromosomal abnormalities, such as advanced female age, low ovarian response and recurrent implantation failure [68, 73]. The major drawback of this procedure is that results are difficult to interpret because of embryo mosaicism, i.e. different genetic composition of the cells of the same embryo (Fig. 2). Mosaicism occurs because of mitotic errors in the cleaving embryo, and has been found in up to 70% of biopsied embryos [68]. Because of mosaicism, the risk of misdiagnosis is as high as 60% of cases with one biopsied cell [74]. In line with this fact, two recent meta-analyses revealed insufficient evidence for improved PRs after PGS [68, 75].

Mosaicism further complicates the results of genetic analysis, since the prognosis after transfer of a mosaic embryo might depend on the number of aneuploid cells. If this number is low (<40%), the embryo may have a relatively high chance of developing to the blastocyst stage (78%), while a higher number of abnormal cells may be associated with a lower survival rate (33%) [76]. This finding can be explained by activation of the embryonic genome during the morula stage (8–64 cells) [18] and the following mitotic arrest of abnormal cells, leading to death of either individual cells or of the entire embryo. The loss of abnormal embryos continues after implantation into the uterus, so that the

23

majority of the remaining abnormal embryos are lost before the 12th gestational week [18].

Fig. 2. Proportions of cytogenetically normal, abnormal and mosaic embryos at different stages of embryonic development. Modified, with permission, from Los et al. [18].

2.3.4 Metabolism

Because neither morphological criteria nor PGS can fully predict the potential of an embryo to develop into a healthy child, new non-invasive methods are currently being investigated. Analysis of embryo culture medium reveals the components of the medium used or released by the embryo, such as glucose, pyruvate and ammonium [77].

Since embryo metabolism is thought to be a critical determinant of viability, investigation methods at the level of the metabolite network (metabolomics) are currently being developed. It is considered that a viable human embryo possesses a unique metabolic fingerprint, which is expressed in culture medium as a metabolic footprint. The entire spectrum of interactions between embryo and culture medium is consequently studied using novel spectroscopic methods which show distinctly different spectra of embryos in cases of no implantation, and biochemical and clinical pregnancy [78].

0

20

40

60

80

100

1 2 4 8 16 32 64 hatch preg fetus

normal homogeneous abnormal mosaic

%

hatched pregnancy fetus blastocyst cell no.

24

2.3.5 Quality of frozen-thawed embryos

To date, there is no generally accepted grading system for frozen-thawed embryos, even though FET cycles have been performed successfully for about 15 years [79]. Various factors have been reported to affect the outcome after freezing and thawing, including cleavage stage [12, 80], pre-freeze morphological appearance [81, 82], the ovarian stimulation procedure used before oocyte collection [83] and the outcome of the fresh embryo transfer cycle [84, 85].

Not all embryo cells may survive after thawing. An embryo is considered to have survived cryopreservation if at least half of the initial blastomeres remain intact [86]. Survival rates are 50–80% [87, 88]. They are higher in morphologically normal embryos with no fragments and equally-sized blastomeres [89, 90] and in embryos with a relatively low cell number at freezing [83], as shown in studies on the outcome of day 2 and day 3 embryos [86, 91].

Reports concerning embryo survival and the transfer of damaged embryos are conflicting [81-83]. Overnight culture of thawed embryos allows embryos to be chosen for transfer if cleaving has restarted after thawing [92]. The implantation rate is higher in cleaved embryos compared with uncleaved ones (20% vs. 3%) [93]. Accordingly, a high rate of aneuploidy (75–80%) has been observed in morphologically good embryos which do not cleave after overnight culture [94, 95].

2.3.6 Male and female factors and embryo quality

Embryo grading cannot give an accurate prediction of pregnancy. This is why the evaluation of clinical factors helps predict the outcome of IVF/ICSI. The majority of these factors are characteristics of the female rather than the male patient. This can be explained by the fact that although the zygote contains equal amounts of chromosomal DNA of maternal and paternal origin, most of the cytoplasm and all the mitochondria originate from the oocyte. The results of several studies have indicated a relationship between oocyte and embryo quality [96-98], but the importance for early embryo development of factors related to spermatozoa is less clear. An analysis of cycles with shared oocytes in an OD programme revealed that embryo morphology (blastomere uniformity and fragmentation) is determined by oocyte quality, while the cleavage rate is affected by both oocyte and spermatozoa quality [99].

25

Paternal age Maternal age

20-29 years 30-34 years 35-44 years

20-29 years

30-34 years

35-39 years

Standard risk

(reference group)

High risk

OR 2.9 (95% CI 1.9 – 4.5)

40-64 years High risk Highest risk

OR 5.7 (95% CI 3.2 – 10.0)

Fig. 3. Couples’ age and miscarriage risk. Modified, with permission, from de la Rochebrochard et al. [100].

The outcome of clinical pregnancy is also affected more by maternal rather than paternal factors. A large multicentre European study of spontaneous pregnancies revealed that although both paternal and maternal age implied an increased risk of miscarriage with increasing age, advanced maternal age plays a more significant role (Fig. 3) [100]. In line with observations in spontaneous pregnancies, a significant increase in pregnancy loss (41.5% vs. 24.4%), a decrease in LBR (41.3% vs. 56.0%), and a decrease in blastocyst formation rate in an OD programme have been noted in men aged more than 50 years, compared with younger men [101]. These results have also been confirmed in IVF/ICSI patients using their own gametes [102].

2.4 Endometrial receptivity

It has been estimated that uterine receptivity accounts for about 31–64% of implantation [103]. A blastocyst can implant into the endometrium only during a short period of time called the window of implantation. It is believed that it lasts about 48 hours, beginning 6–10 days after the LH surge in a spontaneous cycle [104]. The window is advanced in clomiphene citrate- or gonadotropin-stimulated cycles [105-107] and can be delayed in steroid hormone replacement cycles for donor recipients [108].

26

Fig. 4. Factors affecting embryo implantation.

Embryo implantation is regulated by a multitude of factors (Fig. 4). Glycodelin is the major component of endometrial secretion and its expression is regulated by progesterone. Glycodelin has immunosuppressive properties and contributes to the maintenance of pregnancy during the first trimester [109]. It is believed that the appearance of glycodelin secretion reflects endometrial maturation, which is essential for embryo implantation [110]. Accordingly, concentrations of serum glycodelin have been found to be decreased in women with early pregnancy loss [111]. It is believed that the outcome of the first few days after implantation is determined by embryo morphology but that the continuation of pregnancy beyond 6 weeks is more dependent on the combination of embryonic and uterine factors [112]. In humans, there is accumulating evidence of a molecular dialogue between the developing embryo and the maternal endometrial epithelium [113-118]. This crosstalk involves, among other things, nutrition markers such as leptin [115, 119] as well as factors regulated by insulin, such as IGFBP-1 and αvβ3 integrin [120-122].

It is assumed that in cases in which the ovarian response is good and several high quality embryos are created, the uterine milieu is supportive of a pregnancy. Hormonal manipulation of the endometrium has recently been reviewed [123]. It is assumed that the asynchronous development of the endometrium during stimulation with GnRH agonists and gonadotropins is normalised with progesterone or hCG supplementation in the late luteal phase, so that stimulation has no major impact on actual endometrial receptivity. However, stimulation with

αvβ3 integrin receptor

IGFBP-I

IGFBP-I

αvβ3 integrin

Glycodelin

Endometrium

Myometrium

Blood vessel

27

gonadotropins and GnRH antagonists may be associated with impaired endometrial receptivity, compared with stimulation using a GnRH agonist, as shown in a recent analysis of top quality embryo cycles [124].

In IVF/ICSI, endometrial thickness is the only routinely assessed “marker” of uterine receptivity. Oocyte donation cycles are a good model for studying endometrial receptivity. In a study of OD cycles, lower PRs and LBRs were found in subjects with an endometrial thickness less than 8 mm before the start of progesterone administration, compared with those with a thicker endometrium [125]. However, endometrial thickness is a weak predictor of pregnancy [126, 127], since successful pregnancy has been established even in cycles with a maximal endometrial thickness of 4 mm [128, 129]. Other non-invasive methods of evaluation include Doppler and 3-D ultrasonography. However, they are not routinely performed and the results are conflicting [104].

While the window of implantation is relatively narrow in a spontaneous cycle, in an artificial cycle it can be extended to up to five weeks using fixed doses of estradiol. This has been extensively studied in cycles of recipients of donated oocytes [108, 130–133]. In an analysis of 3089 OD cycles, decreasing IRs and PRs were observed only after the 7th week of estrogen replacement [134]. Progesterone administration seems to allow normal endometrial maturation regardless of the length of estradiol therapy. However, a short duration of estradiol replacement (<10–14 days) is associated with a high MR (up to 53%) [130, 132]. Recently, it has been suggested that it is appropriate to start progesterone administration as soon as the endometrium is developed sufficiently, i.e. when it is at least 8 mm in thickness with a trilaminar ultrasonographic appearance, and to perform embryo transfer not before 3–4 days of progesterone treatment [135].

Uterine receptivity remains constant until the end of the fourth decade of life, as shown in a study of 17,300 OD cycles. Beyond the age of 38 years, implantation, pregnancy and live birth rates decreased while the frequency of miscarriages increased [136]. This observation has been supported by the results of other studies involving transfers of embryos created by the patients’ own gametes [137, 138].

28

2.5 Female age

Female age is a major determinant of the success rate of infertility treatment and was the first recognised prognostic factor in IVF/ICSI. Women are tending to delay childbirth until the third decade of life in growing proportions. In Finland, the percentage of women >35 years old giving birth for the first time increased from 6% in 1987 to 18% in 2006 [49]. Statistics from the United States show that the birth rate (per 1000) among women aged 35–39 years almost doubled in the period 1976–1998 (from 19.0 to 37.4) [139, 140]. Since fertility decreases rapidly with age [1], the number of women seeking infertility treatment because of advancing age is increasing.

2.5.1 Ageing and the ovarian pool

It is believed that women are born with a fixed number of oocytes arrested in the first meiotic division, and the numbers subsequently decline during their lives [141]. The number of germ cells reaches a maximum at mid-gestation during fetal life, and declines continuously thereafter. This trend is accentuated at about the age of 37–38, when there is an acceleration of the loss of oogonia [142]. Exhaustion of the oocyte pool occurs around the age of 50 years, leading to menopause.

Recently, this theory has been challenged by the discovery of germ cells that can differentiate into oocytes and follicles in mouse ovaries after birth [143]. These germ cells originate in the bone marrow and reach the ovaries via peripheral blood [144]. In a recent study of adult human ovaries it has been shown that progenitor cells from the tunica albuginea and cells from ovarian surface epithelium can differentiate into granulosa cells and oocytes [145, 146]. These data indicate that the pool of primary follicles in adult human ovaries might not be static, as previously believed, and that a cessation of follicular renewal, rather than a depletion of the static ovarian pool, might take place with advancement of age [147]. These new findings and their interpretations are the subjects of considerable scrutiny and criticism at the present time [148].

29

2.5.2 Ageing and oocyte quality

Neither the classical nor the new ovarian pool theory challenges the fact that oocyte numbers and quality decline with age. It has long been known that in spontaneous pregnancies the risk of chromosomal abnormalities increases with maternal age [149, 150]. Likewise, studies of human oocytes have revealed a growing number of chromosomal abnormalities in the oocytes of older women [151, 152]. As explained above, oocyte abnormalities are the main cause of embryonic chromosomal defects [152–154].

Causes of chromosomal abnormalities have been reviewed in a recent article [155] and are summarised in Fig. 5. The occurrence of aneuploidy has been classically attributed to chromosomal non-disjunction during either meiosis I or II [156]. However, premature separation of chromatids during meiosis might be the main factor contributing to the formation of oocytes with abnormal chromosomal complements [157]. This is caused by the gradual but constant age-related degradation of cohesins and other factors holding the four chromatids together during metaphase I.

Fig. 5. Causes of chromosomal abnormalities of human oocytes, leading to non-disjunction or to premature disjunction of chromatids during meiosis I. Based on Djahanbakhch et al. [155].

Human oocytes enter the first stage of meiosis during fetal life but are selected for further development years later in adult life. Until then, the oocyte’s mitochondria are exposed to ROS, which damage the mitochondrial DNA. This leads to abnormal oxidative metabolism and production of adenosine triphosphate, which might be responsible for abnormalities in the meiotic spindle [158].

ROS

hormonal imbalances

impaired perifollicular microcirculation

ageing granulosa cells

telomere shortening

telomerase defficiency

non-disjunction

premature disjunction

cohesin degradation premature separation of chromatids

damage to mitochondria

damage to meiotic spindle

mis-alignment

30

The decline in fertility in older women can also be explained by telomere shortening due to the combined effects of prolonged exposure to ROS and telomerase deficiency [159]. Telomeres are repetitive DNA sequences at chromosomal ends. Adequate telomere length is essential for the proper alignment of chromosomes during metaphase, but this length might be diminished in oocytes from older women because of damage caused by ROS. The enzyme telomerase protects against telomere shortening. Its activity in female germ cells is significant during early fetal life but is limited in later-stage oocytes.

The increased incidence of oocyte abnormalities with maternal age can also be explained by hormonal imbalances, abnormalities in follicular development due to ageing of the somatic cells surrounding the oocyte and impaired perifollicular microcirculation [155].

2.5.3 Female age and outcome of IVF/ICSI

The deleterious effect of increasing age on the outcome of IVF/ICSI is indisputable. Lower numbers of oocytes are collected and fewer embryos are created, leading to lower PRs and LBRs and increased MRs. It is generally considered that this decline starts after the age of 35 years and the effect is more pronounced after the age of 40. Accordingly, female age is either studied in groups or in linear regression models. Linear regression analysis of 1101 IVF cycles with ovum pickup showed a significant negative linear correlation between age and ongoing pregnancy rate [160]. The ongoing PR/ET declined from 26% in patients younger than 30 years of age to 9% in those aged 37 years, whereas the MR increased from 29% in women under the age of 40 to 50% in those ≥40 years old.

If a large sample of cases is studied, more complex interactions appear between age and the outcome of IVF/ICSI. A polynomial (cubic) relationship between age and LBR was found in an analysis of almost 37,000 IVF cycles [161] (Fig. 6), with a peak at about 30 years of age, after a number of confounding factors were studied simultaneously (previous live births, diagnosis, duration of infertility, number of previous unsuccessful IVF treatments). To date, this relationship has not been explained but it is possible that the determining factor will be identified in the future.

31

Fig. 6. Effect of age of women on LBR per cycle started, per egg collection and per ET. Reproduced, with permission, from Templeton et al. [161].

Despite the increased number of genetically abnormal embryos and resulting lower PRs, women over the age of 35 years who conceive after IVF/ICSI are still at risk of multiple pregnancy. According to US national surveys, multiple births in women aged 35–39 constitute 12–21% of all live births in this age group if two embryos are transferred [8, 32, 33]. If the number of embryos in this age group is increased from three to ≥7, the percentage of triplet and higher-order gestations increases from 4% to 22.3%, respectively, and the MBR to over 40% [32]. Multiple pregnancies are especially risky in older women, since the frequencies of complications such as gestational diabetes and pre-eclampsia are higher in these subjects [162, 163].

2.5.4 Individual ageing rates

The mechanisms that determine individual speeds of reproductive ageing are not fully understood. The importance of environmental and lifestyle factors such as oral contraceptive use, parity and smoking [164], chemotherapy, radiotherapy, pelvic surgery [165, 166], pelvic infection or tubal disease [167, 168] is well known. However, these factors do not explain all variations in the length of the fertile period. An investigation into early ovarian ageing, based on epidemiological data on variation of age at menopause, revealed that about 10% of women might be affected by an accelerated decline of fertility as early as at the age of 32 years [169].

32

Current knowledge about the genetics of reproductive ageing has recently been reviewed [170]. Estimates of heritability of age at menopause range from 30 to 85% and reproductive ageing is most likely determined by the interplay of many genes. Genes that have been identified so far include those that determine factors regulating the transition from primordial follicles into early growing follicles, blood circulation factors (coagulation factors V and VII and apolipoprotein E), steroid hormone conversion and apoptosis factors.

2.6 Low ovarian response

LR is observed when the outcome of ovarian stimulation is suboptimal, leading to a low chance of pregnancy and birth after IVF/ICSI. LR is one of the significant problems of IVF/ICSI because it occurs in up to 24% of cases [171]. The diagnosis, prognostic factors and treatment options for women with LR have been the subject of several reviews [170-176]. Although much is known about the nature of LR, numerous definitions exist because of the different diagnostic tools used in evaluation of the infertile patient.

They include the following:

– Low peak estradiol level during stimulation (<300 to <500 pg/ml). – Estradiol level <200 pg/ml on day 5 of stimulation. – Elevated day 3 FSH (≥7 to ≥15 mIU/ml). – Low number of collected oocytes (<3 to <5). – Low number of growing follicles (from <3 to <5 on day of hCG). – Female age ≥40 years. – Lately, the definition of poor response also includes poor embryo quality

[177].

It is believed that women with LR are in the early stages of ovarian ageing, even if they have regular menstrual cycles. This is illustrated by a study in which a lower number of antral follicles and higher FSH and progesterone levels were found in women with LR and regular menstrual cycles [178]. Furthermore, a LR in first IVF treatment is associated with an increased risk of early menopause [179]. Because of individual variations, LR can occur even in women of young calendar age [169, 180].

PGS studies of embryos of low responders have shown that 54–64% of the embryos in patients with poor prognosis and/or LR have chromosomal abnormalities [17, 73, 181, 182], although they look morphologically normal.

33

This may explain low PRs (3.2% per started cycle) [183] and the high MR (up to 90%) observed in women with LR [184-188].

2.6.1 Prognostic factors of LR

As a reflection of the different characteristics of ovarian reserve, several prognostic factors of LR have been proposed. Although women of ≥40 years have a five times greater risk of LR, compared with younger women [189], age has a low

predictive power [186, 190-195]. The basal FSH concentration was the first widely used endocrine marker of ovarian reserve that had better potential than age for predicting decreased ovarian function and diminished success rates after IVF [190, 191]. Elevations of basal (day 3) FSH may reflect declining ovarian reserve despite regular menstruation, as serum FSH levels in the early follicular phase begin to rise 5–6 years before menopause [196]. In a study of 758 stimulations, women with moderately elevated (15–24.9 mIU/ml) and elevated FSH levels (>25 mIU/ml) had lower ongoing PRs, compared with women with normal basal FSH of <15 mIU/ml (9.3% vs. 3.6% vs. 17%, respectively) [191]. Basal FSH is still widely used in the evaluation of infertility. However, high intercycle variability means that it does not accurately predict the response to gonadotropin stimulation [197].

Hormonal markers of low ovarian reserve used predominantly in the past include a high basal estradiol level [198], an elevated day 3 FSH/LH ratio [199] and a low day 3 LH level [200]. A low day 3 serum inhibin-B (<45 pg/ml) is associated with fewer collected oocytes after ovarian stimulation [201]. This can be explained by lower follicle numbers and respectively fewer granulosa cells in low responders. A previous LR or a low number of collected oocytes in previous treatment may also lead to a repeated low response in a subsequent stimulation [176].

The number of antral follicles (2–5 mm) visible in ultrasonography is a reflection of the actual resting oocyte pool [202–204]. The number of antral follicles decreases by about 60% between 22 and 42 years of age [204], at a rate of 5% per year before the age of 37 years and 17% per year thereafter [202]. The number of antral follicles before gonadotropin stimulation has been found to correlate with the number of collected oocytes [192] and no pregnancy was observed if fewer than four antral follicles were visible before stimulation in a study of 149 treatment cycles [205]. To date, the antral follicle count provides the best estimation of ovarian response, compared with age, BMI and hormonal

34

markers and consequently it is currently considered to be the best single tool in the prediction of LR [186, 192-195, 206].

The predictive value of AMH is also gaining popularity. AMH is a member of the transforming growth factor superfamily involved in the regulation of follicular growth and development [207]. In the ovary AMH is produced by the granulosa cells of early developing follicles. Compared with other hormonal tests, AMH seems to be the best marker reflecting the decline of reproductive age [207]. It can also be used to predict the outcome of IVF stimulation, since one study revealed antral follicle count and AMH to be equally important in predicting ovarian response [208]. Furthermore, IVF failure has been associated with AMH levels less than 1.1 ng/ml [209].

2.6.2 Management of LR

Long protocol with a GnRH agonist is considered the standard regimen in most IVF centres. However, only two studies have concerned investigation of the outcome of low responders in consecutive long protocol cycles with an increased gonadotropin dose. In the first study [210] the daily gonadotropin dose was increased to 450–600 IU, but the PR in previous low responders (6%) was much lower than that in women with a normal response (29%). In the second study [183], the PR among previous low responders remained as low as 3% despite an increase of the gonadotropin dose to 450 IU/day.

A multitude of other approaches have been attempted to overcome the low success rate associated with LR [171-175]. They are summarised in Table 1.

Table 1. Protocols used for treatment of low responders.

Changes of gonadotropin and/or GnRH analogue

administration:

Use of adjuvants or laboratory methods:

Increased gonadotropin dose with or without

change in protocol

Glucocorticoids (immunosuppression)

‘Milder’ pituitary suppression: gonadotropins alone,

CC+HMG/FSH, short and ultrashort protocols

Use of GH, GH-releasing factor,

pyridostigmine (increases GH secretion)

Replacing purified FSH with recombinant FSH L-arginine (donor of nitric oxide)

Luteal initiation of gonadotropins Low dose aspirin (improved microcirculation)

Luteal initiation of GnRH agonist (stop-Lupron) Routine use of ICSI

Pre-treatment with oral contraceptive pills or

progestogens

Assisted hatching

Natural cycles

35

Unfortunately, no stimulation protocol has been proved to be beneficial for low responders. A meta-analysis of the outcome with a GnRH agonist long protocol vs. short protocol vs. GnRH antagonist protocol revealed similar results with each protocol [211]. An ongoing randomised study of long vs. short vs. antagonist protocol may provide an answer to the problem of LR management [212].

2.7 BMI as an outcome predictor of IVF/ICSI

The BMI, calculated as the ratio between a subject’s weight and squared height, is a measure that can be used to evaluate energy storage levels in the body. It is modifiable and depends on genetic as well as lifestyle characteristics.

2.7.1 Overweight, obese and insulin resistant women

The prevalence of overweight (BMI 25–29.9 kg/m2) and obese adults (BMI ≥30 kg/m2) is rising around the world. In Western countries the prevalence of obesity is increasing at a rate of 1–6% per year [213]; this rate is now referred to as the obesity epidemic.

Adipose tissue is a complex metabolic and endocrine organ that responds to signals from circulating hormones, as adipocytes express receptors for pituitary hormones and hypothalamic peptides, including FSH and LH [214]. It also modifies steroid hormones and produces adipokines [215].

Leptin is the most studied adipokine. Its serum levels are in direct proportion to adipose tissue mass as well as to nutritional status in pregnant [216] and non-pregnant women [217] and during ovarian stimulation [218]. Leptin initiates pubertal development, accelerates GnRH pulsatility and stimulates pituitary release of LH and FSH [219]. In addition, leptin and its receptor are expressed in reproductive tissues, including the secretory endometrium [115, 220-222] in which they may regulate uterine angiogenesis [223]. They are also expressed in the blastocyst and participate in embryo implantation [115, 222].

Obesity is characterised by high leptin levels that are related to insulin resistance through altered fatty acid metabolism in skeletal muscle [224, 225] and a pro-inflammatory shift of the immune system [226, 227]. The inflammatory molecules TNF-α and IL-6 have been shown to alter insulin sensitivity by affecting different key steps in the insulin signalling pathway. Their serum concentrations are positively correlated with obesity, impaired glucose tolerance

36

and insulin resistance [213]. In addition, abnormal levels of TNF-α have been shown to be involved in the pathogenesis of miscarriage and recurrent miscarriage through resorption of the fetus [228].

Insulin resistance may lead to miscarriage through diminished endometrial production of IGFBP-1 and uterine αvβ3 integrin [229–231]. In addition, lower serum glycodelin levels have been described in both obese subjects and in women with miscarriage [232]. Changes in endometrial receptivity caused by insulin resistance are shown in Fig. 7.

Other adipokines with effects on reproductive functions include ghrelin, adiponectin and resistin. Ghrelin is another adipokine, the secretion of which is inversely proportional to BMI [219]. Ghrelin possibly acts via hypothalamic and pituitary receptors, and is known to inhibit embryo development and implantation in vitro [233]. Adiponectin and resistin modulate glucose and fat homeostasis, affect insulin action and thus may be another link between obesity, insulin resistance and fertility [234].

Data from large epidemiological studies indicate that chronic sleep deprivation is also related to obesity and insulin resistance. The mechanism has been recently reviewed [235]. Sleep deprivation worsens glucose metabolism, increases appetite, provides more time for eating and possibly decreases energy expenditure though tiredness. These in turn lead to insulin resistance and obesity. There are also indications that too much sleep (over 9 hours/day) can also lead to insulin resistance, although the exact mechanism remains unknown.

Fig. 7. Changes in the uterine milieu which contribute to increased MR in women with insulin resistance. Lower expression of IGFBP-I and αvβ3 integrin lead to diminished adhesion, and lower secretion of glycodelin together with a higher concentration of TNF-α contribute to the resorption of the conceptus.

Blood vessel

Endometrium

Myometrium IGFBP-I

IGFBP-I receptor

αvβ3 integrin

αvβ3 integrin receptor

Glycodelin

TNF-α

37

Fig. 8. Lifestyle characteristics, obesity and reproductive events. Summarised from Knutson et al. [235] and Pasquali et al. [246].

Fertility in obese women

Stein and Leventhal (1934) were the first to recognize the relationship between obesity and reproductive disturbances. The association between lifestyle characteristics, obesity and reproductive events is summarised in Fig. 8. Leptin resistance, low-grade inflammation and impaired glucose tolerance lead to early menarche, menstrual disorders, low fertility and various complications in pregnancy. Abdominal obesity particularly worsens the clinical features of menstrual irregularity and infertility because of increased hormonal disturbances [213]. The strength of the association between high BMI and anovulatory infertility found in several reports [236-238] has been evaluated in the Nurses’ Health Study [239]: the risk of anovulatory infertility increased from 1.3 in the group with a normal BMI to 2.7 in women with a BMI >32 kg/m2. Obesity has

early menarche

anovulation

oligo/amenorrhoea, menorrhagia

poor response to fertility drugs

disrupted embryonic-endometrial dialogue and miscarriage

pregnancy-induced hypertension, gestational diabetes,

Caesarean section

Down’s syndrome

glucose tolerance

obesity

leptin

leptin resistance

Low-grade inflammation:

inflammatory factors (TNFα, IL-6, others)

immunosuppressive factors (glycodelin)

food

exercise

sleep

38

also been found to be an independent cause of miscarriage [240], with MR between 17% [241] and 27% [242] in spontaneous pregnancies.

Obesity is an infertility factor that can be modified through changes in lifestyle. Even moderate weight reduction before infertility treatment might be beneficial [243, 244], since physical activity has been shown to improve insulin sensitivity [245].

Obesity and IVF/ICSI

Even though the consequences of obesity in women are well known, in IVF/ICSI there is controversy regarding the effect of obesity on the outcome of treatment.

Obese women require higher gonadotropin doses for ovarian stimulation [51, 247, 248], despite which, fewer oocytes are collected [51, 248]. However, similar numbers have also been reported [241]. In a study of 3600 IVF/ICSI cycles, an elevated BMI was found to be associated with decreased cumulative PR after controlling for multiple confounding factors [54]. Overweight women had only an 80% chance, obese only a 70% chance and very obese women only a 50% chance of pregnancy compared with women of normal weight, even after several IVF/ICSI cycles. Obesity also diminishes cumulative LBR. It was estimated that because of a lower success rate at every step of IVF/ICSI, only 17% of obese patients have a live birth, compared with 21% of normal-weight women [51].

Despite this evidence, it is difficult to draw final conclusions about the effect of obesity on the outcome of IVF/ICSI because of the contradictory results of miscarriage studies (Table 2). Results from different reports cannot be compared easily, as the definitions of obesity, scope of analysis, stimulation protocols and confounding factors studied differ significantly. However, the statistical methods used have been similar. In most analyses, outcomes among overweight and obese patients have been compared with the outcome of a reference group. Linear regression has also been used in one study [51]. Several studies have revealed a U-shaped relationship between BMI and PR or MR but a mathematical model explaining the relationship has not been developed [54, 242, 249].

In obesity studies, controversy regarding embryo quality and endometrial function exists as well. Obesity may affect the quality of embryos, as suggested in an article in which the mean grade of embryos of women with a BMI of ≥30 kg/m2 was lower than that in normal-weight and overweight women [250]. However, the results of other studies have shown no significant effect of obesity on embryo quality [51, 54, 247]. Reports that obesity increases the MR in OD

39

cycles indicate that obesity might affect the outcome of IVF/ICSI through diminished endometrial receptivity [251, 252]. This effect has not been observed consistently in all studies [253, 254].

Table 2. Summary of findings of previous studies on BMI and pregnancy loss before 12 weeks of gestation.

Study Groups Number of cycles

No effect of BMI on MR

Lashen et al. 1999 [241] <19; ≥27.9 kg/m2 333 started cycles

Winter et al. 2002 [249] WHO classification a 1196 pregnancies

Roth et al. 2003 [255] <20; 20–26.9; ≥27 kg/m2 494 pregnancies

Dechaud et al, 2006 [247] <20; 20–24.9; 25–29.9;≥30 789 started cycles

MR increases along with BMI

Fedorcsak et al. 2000 [248] <25; ≥25 kg/m2 383 pregnancies

Wang et al. 2001 [256] <20; 20–24.9; 25–29.9; 30–34.9; ≥35 kg/m2 1018 pregnancies

Wang et al. 2002 [242] WHO classification 2349 pregnancies

Fedorcsak et al. 2004 [51] WHO classification 5019 started cycles a BMI obesity groups according to WHO: underweight, <18 kg/m2; normal weight, 18–24.9 kg/m2;

overweight, 25–29.9 kg/m2; obese, 30–34.9 kg/m2; very obese, ≥35 kg/m2.

2.7.2 Underweight women

Similarly to obesity, a low body mass is also associated with menstrual disturbances and ovulatory infertility [257-262]. Furthermore, underweight women also have an increased risk of miscarriage [263, 264], intrauterine growth retardation [265] and preterm birth [266].

Not much is known about uterine function in women with low BMI. Low plasma leptin levels have been associated with early [216] and recurrent miscarriage [216, 267], possibly through suboptimal uterine angiogenesis [223], which may explain the reproductive failures observed in underweight women. However, it is not clear whether low BMI predicts a low chance of conception after IVF/ICSI. The IVF study with the highest number of underweight women had 441 subjects (12.3% of studied cycles) [54]. These women tended to have a reduced cumulative chance of pregnancy (OR 0.8, 95% CI 0.65–1.01), compared with normal-weight women. In a later smaller study from the same group, the MR also tended to be lower (OR 0.5, 95% CI 0.26–1.00) [256]. However, in the analysis with the highest proportion of underweight subjects (22%, 87 women) a similar LBR was observed in comparison with normal-weight subjects (20.8% vs.

40

15.2%). In other studies, women with a low BMI have constituted a much smaller proportion of the study population (2.7–3.0%) [51, 242, 249].

2.8 Main infertility diagnosis

2.8.1 PCOS

PCOS is the most common endocrine disorder in women of reproductive age. According to the Rotterdam consensus [268], it is diagnosed when two of the three following criteria are present: oligo/anovulation, clinical and/or biochemical signs of hyperandrogenemia, and polycystic ovaries (≥12 follicles measuring 2–9 mm in each ovary). The aetiology of PCOS is unknown. Insulin resistance is a main feature of the pathogenesis of the syndrome [269]. Obesity, especially the abdominal type, is another characteristic of PCOS. Obesity exacerbates endocrine disturbances of PCOS but even lean women with PCOS have fertility problems similar to those occurring in obese women without the syndrome [238, 270, 271] (Figs. 7 and 8).

As in obese women, weight reduction of even as little as 5% of body weight improves menstrual function and hormonal levels of obese patients with PCOS [272]. Treatment with metformin has been shown to improve insulin sensitivity, to restore ovulation [273-275] and to improve pregnancy outcome [276]. However, metformin therapy was recently found to be inferior to ovulation induction compared with clomiphene citrate [277].

In IVF/ICSI, PCOS is a known risk factor of OHSS [278], a rare [279] but serious complication of ovarian stimulation. More oocytes are usually collected from women with PCOS compared with subjects without the syndrome [270]. However, PCOS has been associated with lower cumulative PR after IVF/ICSI after controlling for obesity, age and other factors [54]. Earlier studies revealed an increased MR in these patients [280]. However, more recent research has showed that PCOS loses its independent effect on miscarriage after controlling for obesity and a number of other factors [242, 248, 256]. Furthermore, it has been demonstrated that insulin resistance increases the risk of miscarriage independently of BMI or PCOS status [281].

41

2.8.2 Other diagnoses

The main subgroups of female infertility can be defined as tubal factor (24% of all female causes), ovulatory dysfunction (34%), endometriosis (11%) and other (31%), including uterine factor or multiple, female-related factors [360]. Analyses of the effects of the main infertility diagnoses on outcome of IVF/ICSI have given inconsistent conclusions. Women with tubal factor infertility have been found to have lower LBR [30, 161, 282] and lower cumulative LBR [53]. Male factor infertility [282] and endometriosis [283] have also been associated with a low chance of live birth. Other factors with an effect on the outcome of IVF/ICSI are smoking, primary infertility, a long duration of infertility [282] and a high number of previous IVF/ICSI treatments [284].

2.9 FET

FET allows all good quality embryos to be stored and to cancel ET if the patient is at risk of OHSS [285]. According to data from 2003, Finland is the country with the highest number of FET cycles performed relative to the number of OPUs (56%) [3]. In other European countries, FET is performed to a much smaller extent (mean 20.5%, 1.0–39.9%).

The PR after FET is lower than that in fresh cycles (18% vs. 26%) [3]. This can be explained by embryo damage during the freezing and thawing procedure. An additional factor may be the fact that the best embryo(s) are always used in fresh cycles. Factors affecting the outcome of FET include infertility aetiology [286], age at the time of the fresh cycle [91, 286], type of ovarian stimulation protocol used in the fresh cycle [287], outcome of the fresh cycle [84, 85, 286], freezing method [288], as well as the embryo factors described above. In a study where embryos of similar grade were transferred, no difference was observed between the outcomes of spontaneous vs. hormonally substituted FET cycles (LBR 11% vs. 9.8%) [289].

Despite the lower PR, the risk of multiple pregnancy is elevated even in FET. In 2003, there were 993 multiple births after FET in Europe (MBR 15.1%) [3]. This can be explained by the number of embryos transferred in FET. A retrospective study revealed significantly higher LBR after the transfer of two embryos rather than one in FET SET (25.7% vs. 19.2%) but the MBR was ten times higher in the cycles with two (21.9%) compared with those with one embryo transferred (2.0%) [290].

42

2.10 Number of transferred embryos and pregnancy outcome

The chance of pregnancy increases along with the number of embryos transferred [32, 81, 290]. In a mathematical analysis it was found that the number of embryos implanted depends on the number of embryos transferred, so that the implantation probability of a given embryo is increased by 22% for each additionally implanted embryo [291]. This can be explained by the dialogue between the embryo and the endometrium before implantation [291] or by the increased placental mass in early multiple implantations, which produces more hCG and progesterone than singleton placentas [292, 293] and might affect important factors in implantation [119].

Embryo synergism continues after implantation so that the loss of a single embryo is lower in pregnancies with two or three gestational sacs, compared with singleton pregnancies [112, 292-294]. However, the benefits of multiple implantation are lost in later pregnancy. As mentioned earlier, multiple pregnancies often result in the premature births of infants with low birth weight. Even if up to 36% of twin gestations undergo a spontaneous reduction of one gestational sac [295–298], the outcomes of the resulting singleton pregnancies are not as good as in gestations with a single implanting embryo. The probability of low and very low birth weight are twofold higher in vanishing twin pregnancies, compared with single implantations [296]. A negative effect on the outcome may also be exerted by the number of embryos transferred but not developing into pregnancies visible in ultrasonography. Singletons born after DET have been shown to have higher ORs for preterm birth (1.8) and for low birth weight (3.4), compared with singletons after SET [295].

2.11 eSET with FET

During the nine years after the first study of eSET [9], considerable evidence has accumulated about the safety of this treatment option. Results from randomised and retrospective trials are shown in Table 3. In general, eSET is associated with a considerably lower multiple pregnancy rate compared with DET. However, PRs in randomised studies are lower in cases of eSET than in DET. In retrospective studies, similar PRs are observed with either transfer strategy.

Tabl

e 3.

Res

ults

from

eS

ET

stud

ies.

Stu

dy

S

ubje

cts

D

esig

n

PR

eSE

T vs

. DE

T

C

umul

ativ

e M

PR

eSE

T vs

. DE

T

Ran

dom

ised

Ger

ris e

t al.

1999

[41]

53 w

omen

<34

yea

rs, f

irst c

ycle

,

≥ 2

top

qual

ity e

mbr

yos

1

eSE

T vs

. 1 D

ET

38

.5 v

s. 7

4%

3.

9 vs

. 30%

Mar

tikai

nen

et a

l. 20

01 [1

0]

14

4 w

omen

, ≥ 4

goo

d qu

ality

embr

yos

C

umul

ativ

e ou

tcom

e af

ter

1 eS

ET

+ FE

T vs

. 1 D

ET

+ FE

T

47.3

vs.

58.

6%

5.

4 vs

. 26.

2%

Thur

in e

t al.

2004

[42]

631

wom

en <

36

year

s, ≥

2 g

ood

embr

yos

2

cycl

es e

SE

T +

FET

vs.

1 cy

cle

DE

T +

FET

47

.9 v

s. 5

2.6%

0.8

vs. 3

3.1%

a

Luka

ssen

et a

l. 20

05 [4

5]

10

7 w

omen

< 3

5 ye

ars,

firs

t cyc

le,

≥ 1

good

qua

lity

embr

yo

2

cycl

es e

SE

T vs

. 1 c

ycle

DE

T

41 v

s. 3

6%

0

vs. 3

7%

van

Mon

tfoor

t et a

l. 20

06 [2

99]

30

8 un

sele

cted

wom

en

1

eSE

T vs

. 1 D

ET

21

.4 v

s. 4

0.3%

0 vs

. 21.

0%

Ret

rosp

ectiv

e

Vils

ka e

t al.

1999

[9]

91

0 tra

nsfe

rs

eS

ET

if ris

k of

OH

SS

, pat

ient

pref

eren

ce o

r med

ical

indi

catio

n

29.7

vs.

29.

4%

0

vs. 2

3.9%

Tiiti

nen

et a

l. 20

01 [1

2]

A

ll tra

nsfe

rs in

199

8–19

99, 7

08

cycl

es

eS

ET

perfo

rmed

if p

atie

nt

pref

eren

ce o

r med

ical

indi

catio

n

38.6

vs.

40.

0%

2.

9 vs

. 26.

2% a

Ger

ris e

t al.

2004

[43]

367

wom

en <

38

year

s, fi

rst c

ycle

or p

revi

ous

deliv

ery,

≥ 1

top

qual

ity e

mbr

yo b

1

eSE

T vs

. 1 D

ET

49

.0 v

s. 5

4.0%

2.5

vs. 2

5.4%

van

Mon

tfoor

t et a

l. 20

05 [3

00]

32

6 w

omen

< 3

8 ye

ars,

≥ 1

top

qual

ity e

mbr

yo

C

umul

ativ

e ou

tcom

e af

ter t

hree

fresh

trea

tmen

t cyc

les

+ FE

T

41 v

s. 3

0%

0

vs. 3

3%

a mul

tiple

live

birt

h ra

te, b p

atie

nts

chos

e be

twee

n eS

ET

or D

ET

prio

r to

treat

men

t. In

the

eSE

T gr

oup,

eS

ET

was

car

ried

out i

f ≥1

top

qual

ity e

mbr

yo w

as

avai

labl

e, o

ther

wis

e D

ET

was

per

form

ed. R

esul

ts s

how

n ar

e fo

r gro

ups

base

d on

pat

ient

wis

hes.

43

44

Currently, eSET is practised on a large scale only in Finland, Sweden and Belgium [3]. The Finnish national birth registry has shown a steady decrease in the proportion of multiple births after IVF/ICSI from 24% in 1996, before the implementation of eSET, to 14% in 2002, while the LBR has remained unchanged [52]. Similar results have been reported from Belgium [301, 302] and also from Sweden [38, 64] where the proportion of cases of eSET reached 67.4% in 2004. Consequently, in the same year the Swedish national MBR was as low as 5.7% [38].

The eSET and multiple embryo transfer strategies have been compared using Swedish national data as an example of an eSET strategy and national data of the United States as a model of multiple ET [38]. Throughout the years since 1991, a continuous increase in LBR/ET was observed in the United States, whereas in Sweden it remained stable after 1993. A reduction by half in the number of embryos transferred took place in Sweden after the introduction of the law on eSET, followed by a considerable reduction in MBR. In the United States, the reduction in the number of embryos transferred was less pronounced (30%) and consequently the MBR decreased only slightly. Finally, the analysis showed that the LBR/embryo transferred was higher with eSET than with the transfer of several embryos (19% in Sweden and about 13% in the United States).

2.11.1 Physicians’ considerations

Countries in which eSET is practised on a large scale have state funding for IVF/ICSI [3]. State insurance has been found to affect embryo transfer practices. In a study of state-mandated insurance coverage for IVF/ICSI in the United States, a smaller proportion of transfers of at least 3 embryos was observed in states having insurance for IVF, compared with the states without insurance [303]. Utilisation of IVF services and treatment outcome were also studied in 360 fertility clinics in the United States, of which 31 clinics worked under complete insurance coverage [304]. State-mandated insurance for IVF was associated with increased use of treatment, fewer embryos transferred per cycle and a lower percentage of high order gestations. However, a lower PR/cycle was also observed in the clinics with insurance coverage. This may be a reason why infertility practitioners in places without infertility insurance are reluctant to decrease the number of embryos transferred and to consider eSET.

Before the latest nationwide data became available [38, 301, 302], several investigators had voiced the fear that if eSET was applied to a larger extent in

45

everyday practice, PRs per transfer would diminish considerably [14-16, 305]. Other concerns expressed were overall effectiveness and cost of treatment for the patient, since eSET might result in an increased cost of successful treatment and a longer time to achieve pregnancy.

2.11.2 Patients’ preferences

Patients’ preferences regarding the plurality of desired pregnancy have been subject to extensive investigation. Between 14% and 90% of IVF patients prefer twins rather than singletons [306-312]. In a Danish survey, only 23% of patients preferring twins expressed a positive attitude towards twins as the primary reason for their desire [313]. Many more patients wanted to maximise the chance of their child having siblings or to have as few IVF treatments as possible. Patients preferring the transfer of more than one embryo have also been concerned about the price of treatment [306, 308]. Many couples view the possibility of no pregnancy after treatment and the possibility of a multiple pregnancy as the only possible outcomes after IVF.

It is known that patients’ attitudes depend on the way health information is presented to them [359]. In addition, infertility patients seem to be unaffected by perceptions of the high risk that is associated with twins [306]. However, if success rates and prices are similar, 82% of patients would prefer eSET to DET [311]. Physician motivation also plays a role, as illustrated by a study in which 64% of couples chose to have eSET after careful explanation of the risks of multiple pregnancies [44].

2.12 Cost-effectiveness analysis

CEA is one of the techniques of economic evaluation designed to compare the costs and benefits of healthcare intervention in order to assess whether it is worth executing. The aim of CEA is to maximise the level of health effects relative to the resources available. It can be based on different types of data. With data from randomized studies, CEA tests the extent to which a form of intervention does more good than harm under ideal circumstances (efficacy, “Can it work?”). For a better estimation, data from healthcare practice under usual circumstances are studied instead (effectiveness, “Does it work in practice?”) [314].

In CEA, the total costs and total effects of a new and an old therapy are compared. Total costs can be grouped in four categories [315]:

46

– health care resources: costs of organising and operating a health care programme, including resources necessary to dealing with adverse effects caused by the programme;

– patient and family resources: out-of-pocket expenses of the patient or family members as well as the value of any resources that they contribute to the treatment process, such as formal care;

– productivity losses: patients or family members lose time from work while seeking treatment or participating in a health programme;

– other losses: social worker visits etc.

Depending on the desired endpoint, any of the cost categories or all of them can be evaluated.

The effects of health care interventions are quoted as numerical variables, such as number of births, years of life gained, millimetres of blood pressure lowered. In IVF, the outcomes most often used in economic analyses are live births or live-born children [316], although, as already mentioned, term live births have also recently been used [56, 57].

Total costs and total health effects are separately calculated for the new and old therapies. Possible results are presented in Fig. 9. Dominance is not common in economic evaluations, because in most cases more effective treatments are also more costly [317, 318]. The most common outcome of CEA is a new therapy which is more effective but also more costly, compared with the reference therapy. Whether it is an acceptable (cost-effective) option depends on the maximum amount of money that society is prepared to pay for the gain in effectiveness [315].

New therapy… … less effective … as effective … more effective

… less costly Non-dominance

Necessary to save money?

Weak dominance

Accept new therapy?

New therapy dominates old

Implement new therapy

… as costly Weakly dominated

Reject new therapy?

Non-dominance

Other considerarions?

Weak dominance

Accept new therapy?

… more costly New therapy is dominated

Reject new therapy

Weakly dominated

Reject new therapy?

Non-dominance

Additional funds available?

Fig. 9. Interpretation of the outcome of cost-effectiveness analyses.

47

Incremental cost-effectiveness ratios are useful tools when comparing two or more treatment strategies. They show the change in price for one unit of effect. The following formula is used in the calculation of an ICER:

costs total costs of old therapy total costs of new therapyICER

effects number of health effects with old therapy – number of health effects with new therapy

Δ −= =

Δ

It is never possible to estimate completely accurately the cost of healthcare. Likewise, some uncertainty always remains about the effects of different treatment strategies. A sensitivity analysis, which is an essential part of CEA, tests all the assumptions used in the model.

2.12.1 Cost of IVF/ICSI treatment

IVF/ICSI treatment is very costly, with prices differing considerably between countries. There is only one comparative study including data from European countries and the United States [4]. In 2002, the average price of a single IVF cycle was about €2700 in Finland, but about €3450 in other European countries (ranging from €1942 in the Netherlands to €4115 in Italy) and about €10,000 in the United States.

Depending on the country, treatment costs can be as high as 25% of annual household expenditure [4]. Use of treatment is therefore low throughout the world, with only 22% of infertile patients receiving IVF/ICSI, even in developed countries [319].

The actual cost of IVF/ICSI is even higher because of the cost of multiple pregnancies. Multiple pregnancies are associated with an increased risk of maternal complications necessitating absence from work and hospital stay, while twins and children born after higher-order multiple pregnancies frequently require hospitalisation [6, 28, 34, 320]. The family costs per multiple live birth are approximately 4-, 11- and 18-fold greater for twins, triplets and higher-order deliveries, respectively, compared with singleton deliveries [6]. Similarly, a study carried out in Oulu, Finland, has shown that the costs of maternal and pediatric care up to the end of the neonatal period are €5780 for an IVF singleton and €15,582 for an IVF twin pair [321].

In the United Kingdom, although multiple pregnancies after IVF/ICSI account for 27% of live births after treatment, they incur more than half of the direct cost of IVF/ICSI pregnancies up to the end of the first year after delivery

48

[322]. In the same study it was estimated that if eSET is implemented, savings in comparison with multiple pregnancies and births could allow the state funding of an additional 10,124 IVF cycles per year.

2.12.2 Cost-effectiveness of eSET

Up to now, the financial impact of eSET has been studied insufficiently [323]. Most cost-effectiveness analyses have been based on clinical data from randomised studies [43, 45, 324-327]. As explained above, this means that their conclusions are valid only in the setting of a randomised study, i.e. highly selected patients and a limited number of treatment cycles [315]. The results are shown in Table 4. The effect of FET cycles after eSET was evaluated only in two studies, and not to its full extent because of time limitations or assumptions overestimating the success rate [325, 326].

Because of different endpoints and costs evaluated, the outcomes of these studies have been re-analysed in a review [328]. eSET might be associated with smaller costs, compared with DET, if IVF treatment, pregnancy and paediatric care are evaluated together, but not if only the costs of IVF are analysed. The difference in effectiveness between eSET and DET was found to decrease substantially when eSET with FET was performed to patients with a good prognosis.

2.13 Alternatives to eSET

There are two current alternatives to eSET: eSBT and mild stimulation. The former is favoured by some clinics because of the high PR in fresh cycles following blastocyst transfer. Like eSET, eSBT has been shown to lower the MBR while preserving PR and LBR within an acceptable range with or without FET [329-332]. Two years of experience with eSBT has shown a decrease in MBR with no change in LBR [333], similarly to eSET [38, 52, 302]. However, with eSBT the MBR remains higher than with eSET (17% vs. 5%) [38, 333].

49

Table 4. Summary of findings of previous cost-effectiveness studies of eSET.

Results

eSET compared with DET

Study Subject data Design

Costs of

IVF

Outcome of

IVF

Wølner-Hanssen and

Rydhstroem 1998 [324]

Modelled data Modelled data Lower Similar

De Sutter et al. 2002

[325]

144 women,

>4 good embryos

<3 eSET + FET vs.

<3 DET + FET

Lower Worse

Gerris et al. 2004 [43] 367 women

< 38 yrs, first IVF or

prev. delivery

1 eSET vs. 1 DET Lower Better

Lukassen et al. 2005 [45] 107 women

< 35 yrs,

>2 good embryos

2 eSET vs. 1 DET Similar Similar

Thurin Kjellberg et al.

2006 [326]

661 women

< 36 yrs,

>2 good embryos

eSET + eSET/FET

vs. 1 DET

Lower Better

Fiddelers et al. 2006

[327]

308 women, first

IVF, >2 embryos

1 eSET vs. 1 DET Lower Worse

To date, the outcomes of eSET and eSBT have not been compared in a single study. The cumulative outcome with cleavage-stage embryos and blastocysts in other trials has been reviewed recently [66]. In two of the three identified studies, there was a higher cryopreservation rate as regards the cleavage-stage embryos and a higher implantation rate of the fresh blastocyst transfers, but the cumulative outcome was similar. In the third study there was a cumulative PR of 46% in the cleavage-stage group vs. only 27% in the blastocyst group [334].

The mild stimulation protocol is aimed at minimising the adverse effects of ovarian stimulation (down-regulation symptoms and multiple births) [335]. Gonadotropin administration is initiated on day 5 of the cycle and a GnRH antagonist is started on the day when at least one follicle has a diameter of ≥ 14 mm [336]. A larger proportion of euploid embryos has been observed after mild stimulation than after long protocol (51% vs. 35%), but the average number of embryos was the same (1.8), regardless of the protocol [337]. In addition, mild stimulation is associated with a high cancellation rate of 17–22.4% [336, 338], compared with stimulation using the long protocol (cancellation of about 1% of cycles performed at Oulu University Hospital).

50

A cost-effectiveness study has been carried out in which the outcomes of mild stimulation and of a long protocol with DET were compared [56]. Patients were randomised to one year of treatment, and FET was performed if cryopreserved embryos were available. The cumulative tLBRs/woman were similar: 43.4% with mild stimulation and 44.7% with the long protocol and DET. Mild stimulation was associated with lower total costs up to six weeks after expected delivery, but overall the costs of IVF treatment were similar in the two treatment groups.

51

3 Aim

At present, only 50% of women have a child following IVF or ICSI. There is insufficient knowledge about the interplay between characteristics of the patient, of the treatment and especially of embryo transfer. The overall aim of this work was to study the relative importance of the transfer of a top quality embryo with respect to features of the female patient and of laboratory data. The groups studied included the general IVF/ICSI patient population as well as treatment groups with expected worse outcome such as women with LR to ovarian stimulation, obese and underweight subjects as well as those of advanced reproductive age. In addition, the work involved assessment of the economic viability of the elective transfer of a single top quality embryo.

The specific aims were as follows:

1. To investigate the incidence of LR in subjects stimulated more than once and the outcomes of cycles with (a) top quality embryo(s).

2. To study factors that might affect first trimester miscarriage after fresh IVF/ICSI, spontaneous and hormonally substituted FET. To this end, we performed a multivariate analysis including embryo morphology and other baseline and treatment characteristics that have previously been shown to have an effect on miscarriage. Specific attention was paid to states of insulin resistance such as obesity and anovulation.

3. To verify the applicability of eSET of a top quality embryo in older women aged 36–39 years, by comparing the outcomes of eSET with those of DET in subjects of the same age.

4. To determine whether the overall outcome, costs of IVF/ICSI treatment and time to live birth are affected by the implementation of eSET in everyday clinical practice.

52

53

4 Materials and Methods

Fig. 10. Design of the studies included in this work.

Study III Older women

Women 36–39 years old 1224 fresh, 828 FET cycles

Always LR: three LR cycles 2 subjects, 6 cycles

LR→NR: First cycle LR, ≥1 NR cycle 43 subjects, 129 cycles

NR→LR: first cycle NR, ≥1 LR cycle 35 subjects, 105 cycles

666 spontaneous FET cycles

466 hormonally substituted FET cycles

Study II Miscarriage

8981 fresh, 7015 FET cycles

Study I Low responders

2237 women, 3846 fresh cycles

First three cycles analysed

≥ 1LR cycle and ≥3 fresh cycles 80 women/240 cycles, long protocol

Women with a clinical pregnancy: first pregnancy from IVF/ICSI

2198 fresh cycles

DET period: 1995–1999 826 women

1359 fresh cycles 589 FET cycles

eSET period: 2000–2004 684 women

1027 fresh cycles 683 FET cycles

Study IV Overall outcome

1510 women 2469 fresh, 1851 FET cycles

nt-eSET: (≥1 frozen embryo) 110 fresh cycles 128 FET cycles

cSET: (no frozen embryos) 194 fresh cycles

DET: 585 fresh cycles 280 FET cycles

eSET: 335 fresh cycles 420 FET cycles

54

4.1 Study population

The IVF Units of Oulu University Hospital and Helsinki University Central Hospital, the Oulu and Helsinki infertility clinics of the Family Federation of Finland, and the AVA-Clinic, Fertility Centre, Tampere, provided detailed data about IVF/ICSI and FET procedures carried out during the years 1994–2005. A total of 18,815 cycles have been analysed.

The design of the studies is illustrated in Fig. 10. Study I was based on data from the IVF Unit of Oulu University Hospital and the Family Federation of Finland, Oulu, from 1994–2002. Cycles of women stimulated <3 times or without LR (3473 cycles), as well as data on women who had undergone ovarian surgery (133 cycles) were excluded from the analysis.

Study II included the first clinical pregnancy cycles of women who were treated at the Infertility Clinics of Oulu and Helsinki University Hospitals and the Family Federation of Finland in Oulu and Helsinki, and the AVA-Clinic in Tampere during 1999–2004. Cycles with no pregnancy (11,946 cycles) and data on 701 non-first pregnancies were excluded from the analysis.

Study III involved analysis of the treatment outcomes of women aged 36–39 years who had undergone IVF/ICSI during the period 2000–2003 and FET during 2000–2004. Data for this study were collected from the Infertility Clinics of Oulu and Helsinki University Hospitals, and the Family Federation of Finland in Oulu and Helsinki.

Study IV was based on treatment cycles from Oulu University Hospital only: IVF/ICSI cycles performed in 1995–2004 and FET cycles up to 2005. All consecutive cycles of each woman were analysed. In cases with a live birth, subsequent cycles were excluded (738 fresh and 537 FET cycles).

4.2 Treatment protocols

IVF/ICSI protocols

Ovarian stimulation was performed using the long GnRH agonist protocol or the GnRH antagonist protocol [9, 12, 192, 339]. In a small proportion of cycles, other stimulation protocols were used (short and ultra-short protocols) (Study II). The starting gonadotropin dose was determined according to the patient’s age, BMI, antral follicular count in the baseline ultrasonographic scan and the outcome of previous infertility treatments. The cycle was cancelled if less than three growing

55

follicles are observed on day 10 of stimulation, resulting in a yearly cancellation rate of about 1%. Oocytes/embryos were cultured as previously described [290, 339]. A LR cycle was defined as yielding ≤3 oocytes, a NR cycle 4–14 oocytes, and a HR cycle ≥15 oocytes. Embryo transfer was carried out on day 2 or 3 after oocyte retrieval, depending on the day of the week – on Monday if oocytes were collected on Friday and 2 days after ovum pickup in all other cases. Such a strategy has been adopted to minimize work during the weekends. A top quality embryo had 4–5 evenly sized cells and <20% fragmentation when cultured for 2 days, or ≥8 cells and <20% fragmentation when cultured for 3 days [40]. Extra embryos were frozen on the day of embryo transfer, using a slow freezing protocol [290].

FET cycles

In spontaneous FET, ET was performed after spontaneous ovulation. The urinary LH surge was measured by means of a home test kit. Depending on the day of freezing of the embryo, ET was carried out four (2–5) days after a positive result. In some spontaneous FET cycles, luteal support with 200 mg Lugesteron/day (Leiras, Helsinki, Finland) was started on the day of ET and continued for two weeks.

In hormonally substituted FET cycles, estradiol valerate or 17β-estradiol (Estrofem, Novo Nordisk, Bagsværd, Denmark; Merimono, Novartis, Basel, Switzerland; Progynova, Schering, Berlin, Germany; Zumenon, Solvay Pharmaceuticals, Hannover, Germany) was administered at a daily dose of 4–6 mg. Vaginal micronized progesterone (400–600 mg Lugesteron/day) or Crinone vaginal gel (8%), 1.125 g twice a day (Serono, Geneva, Switzerland) was started when the endometrial thickness was ≥6 mm on cycle day 11–13 and treatment was continued until 9–12 gestational weeks after a positive pregnancy test result. In cases of a thinner endometrium, the estradiol dose was increased to 6–8 mg/day and progesterone administration was delayed until a thickness of ≥6 mm was reached. A GnRH agonist (Suprecur, 450 mg/day, Sanofi-Aventis, Paris, France, or Synarela, 400 mg/day, Pfizer, New York City, U.S.A.) was used in some cycles for pituitary down-regulation before starting endometrial preparation in order to prevent spontaneous ovulation and to control the timing of ET. Thawed embryos were graded using the same criteria as in the fresh cycles, although at present there is no consensus of opinion on the definition of a top quality embryo in FET cycles.

56

Pregnancy was detected by means of urinary hCG test (in Oulu) or by serum hCG measurement (in Helsinki and Tampere). Clinical pregnancies were diagnosed by transvaginal ultrasonography at gestational week 6–7 by visualisation of a gestational sac and cardiac activity.

4.3 Cost of IVF/ICSI

A “payer perspective” analysis of total treatment charges and medication costs up to the time of the pregnancy test was carried out on the two study periods in Study IV. Financial calculations were based on the pricelist of the IVF Unit of Oulu University Hospital for the year 2008 (Table 5). A 3% discounting rate was used. At Oulu University Hospital, the charge for freezing and cryostorage of supernumerary embryos is included in the charge of the fresh cycle. Costs that resulted from complications of ovarian stimulation (OHSS, bleeding, infections) were not taken into consideration.

Table 5. IVF/ICSI costs at the IVF Unit of Oulu University Hospital for the year 2008.

Unit Unit price, €

Fresh cycles

Stimulationa 1495

IVF 1542

ICSI 2158

Progesterone 14 days 34

FET cycles 600

Hormonal supportb 66 a includes GnRH agonist for long protocol, 2400 IU of recombinant FSH, 5000 IU of hCG. b includes LH

test and 14 days of vaginal progesterone (spontaneous FET) or 28 days of estradiol valerate and 14 days

of vaginal progesterone (hormonally substituted FET).

4.4 Statistical analyses of clinical data

For linear variables, log-transformation was used, if necessary, to normalise distributions before statistical testing by means of Student’s two-tailed t-test or one-way ANOVA with the Tukey multiple comparisons test as a post hoc analysis. Where distributions continued to be skewed after log-transformation, the non-parametric Mann–Whitney U or Kruskall–Wallis tests were used. For categorical variables, statistical significances in frequency differences between groups were evaluated by using the chi-square test. Statistical analysis was

57

performed with SPSS software, versions 10.1–16.0 (SPSS Inc., Chicago, IL, USA), with P<0.05 as the limit of significance. Unless otherwise indicated, results are expressed as mean ± SD.

In Study I, the primary outcome measure was the PR/cycle. Baseline and treatment characteristics, as well as the treatment outcome in the study groups were compared, using the overall mean PR/cycle during the study period in all cycles and in cycles with at least one top quality embryo transferred.

The primary outcome measure in Study II was the MR/cycle. Miscarriage was defined as pregnancy loss before 12 full weeks of gestation. Logistic regression was used to study the effect on MR of characteristics of the study subjects and the treatment cycles (see Table 6). Univariate logistic regression was first performed as regards all factors, after which the variables showing independent effects in these analyses were studied by means of multivariate logistic regression. Variables showing a non-significant effect were removed from the model in a stepwise manner. The fit of the final logistic model was assessed by means of the Hosmer–Lemeshow test.

In Study III, the PR/ET and the cumulative PR/OPU were the primary outcome measures. Baseline and treatment characteristics of the studied groups were also compared. In Study IV, the cumulative tLBR/woman was the clinical primary outcome measure. The analysis was not limited only to comparison of characteristics of the treatment groups. In order to assess the independent effect of the treatment period on clinical outcome, logistic regression with backward elimination after multicollinearity examination was performed on the cumulative LBR data. The independent clinical effect of the change in embryo transfer policy on the cumulative tLBR was evaluated by using the predicted probability for term live birth for each woman in the DET period. For the same subjects, the probabilities of term live birth with the regression coefficient of the eSET period were calculated. The difference between the two probabilities showed the change in clinical outcome from DET to eSET policy.

58

Table 6. Characteristics of treatment cycles examined by multivariate regression analysis in Study IV.

Type of factor Factors studied

Patient characteristics

(all cycles)

Age (categorical, linear, cubic)

BMI (four different models)

Categorical: obese vs. non-obesea

Categorical: WHO groupsb

Linear: exact BMI values

Quadratic: BMI and BMI2 c

Main infertility diagnosis

Primary vs. secondary infertility

Previous miscarriages: yes vs. no

Treatment type Fresh, spontaneous or hormonally substituted FET

Embryo characteristics

(all cycles)

Origin of the embryo: IVF vs. ICSI

Number of embryos transferred

Transfer of ≥1 top quality embryo

Fresh cycles only Stimulation protocol: GnRH agonist vs. GnRH antagonist

Gonadotropin dose

Endometrial thickness at end of stimulation

Number of oocytes:

Collected

Fertilised

Cleaved

Ovarian response: LR vs. NR vs. HR

Day 2 vs. day 3 fresh ET

eSET

Number of embryos frozen

Spontaneous FET only Use of luteal phase progesterone: yes vs. no

Hormonally substituted FET only Use of GnRH agonist: yes vs. no a Obesity was defined as BMI ≥27 kg/m2, as this is the threshold value above which insulin sensitivity

decreases significantly [340]. b BMI groups according to WHO: underweight, <18 kg/m2; normal weight, 18–24.9 kg/m2; overweight, 25–

29.9 kg/m2; obese, 30–34.9 kg/m2; very obese, ≥35 kg/m2. c BMI was centred at 23 kg/m2 (which approximates the mean BMI) in order to prevent possible co-

linearity between the BMI variables.

4.5 Cost-effectiveness analysis

Because of the multimodal distribution of the total treatment cost per woman, the treatment costs of the fresh and FET cycles per woman were analysed in a multivariate general linear model after log-transformation. The independent

59

effects of the treatment period, and clinical and laboratory characteristics were assessed after multicollinearity examination. Variables without an independent effect were removed from the model. The contrast estimate of the eSET period was used to calculate the change of the total treatment cost for the women from the DET period from a DET to an eSET policy.

Total costs and total term live births in the study periods were compared. The ICER was calculated using the following formula:

costsICER

effectstotal costs of DET period – total costs of eSET period

number of term live births during DET period – number of term live births during eSET period

Δ=

Δ

=

Sensitivity analysis was performed using the 95% CI of the number of fresh and FET cycles and of the tLBRs in the fresh and in the FET cycles. As the FSH dose required for stimulation varies substantially between subjects, the SD of the FSH dose was used in price calculations for the sensitivity analysis instead of the 95% CI. Finally, the range of the change in the predicted probabilities of term live birth and the range of the change in the total cost from the general linear model were used in estimating the independent effect of the change from DET to eSET policy.

In the year 2002, the cost of an IVF cycle in Finland was about 1.5-fold lower than in other European countries and 3.7-fold lower than in the United States [4]. Assuming similar inflation of health care costs in Finland, in the other European countries and in the United States, the ICER and its upper and lower confidence limits were multiplied by 1.5 and by 3.7 in order to facilitate interpretation of the results.

4.6 Data protection

Since all data were collected retrospectively, Ethics Committee permission was not sought. The personal identification code (henkilötunnus) of each female patient was linked to a working identification code at the time of data extraction. The files containing the original and the working codes were kept in each clinic according to its own data protection rules. The working databases contain only the working codes. Partners’ personal identification codes, as well as any names, addresses and phone numbers were not collected.

60

61

5 Results and discussion

5.1 Value of a top quality embryo in LR

Study I showed that LR cycles constituted 11.5% (441/3846) of all IVF/ICSI cycles performed from 1994 to 2002, which is consistent with previous reports [179, 341]. Of the 80 subjects with LR cycles who were stimulated three times, only two (2.5%) had a low ovarian response in all cycles. Therefore, results for recurrent low responders are not shown. The finding that LR is not a recurrent problem in a patient population consisting mostly of women of <40 years of age confirms a previous report showing that the majority of young poor responders with normal FSH values have normal responses in their second and third cycles [342].

The proportion of LR cycles in the groups with an initial LR and with an initial NR was similar (Table 7). The proportion of women with inactive ovaries at the beginning of ovarian stimulation tended to be higher in the LR→NR group (42.4%, 53/125) than in the NR→LR group (27.5%, 28/102, P=0.06). However, it was higher in both groups than in the general IVF/ICSI population (10.2%) [192]. These results confirm previous observations that a low antral follicle count is associated with lower numbers of retrieved oocytes [192, 205].

Stimulation characteristics were similar in the groups with an initial LR and with an initial NR. In LR→NR subjects, after LR in the first cycle the mean gonadotropin dose was increased from 2349.1 ± 685.3 to 2977.2 ± 977.8 IU (P<0.0001), which resulted in a higher number of collected oocytes (2.1 ± 0.9 vs. 6.7 ± 2.7, P<0.0001). PRs per cycle were similar between the study groups but were significantly lower (initial LR: P<0.0001; initial NR: P=0.01) when compared with the mean overall PR/cycle during the study period (27.2%). The fact that increasing the gonadotropin dose after a LR cycle results in an improved response with a low PR in the subsequent cycle is supported by the results of other studies [183, 210, 343-345].

As expected, the proportion of cycles with ≥ 1 top quality embryo transferred was lower than that observed in the general IVF/ICSI population (64.6% at the Oulu University Hospital in 2004). In these cycles, the total numbers of transferred embryos, especially the numbers of top quality embryos, did not differ between the two groups with LR cycles. However, following the transfer of top quality embryos, women with an initial LR had a significantly lower PR (17.8%

62

vs. 41.2%, P=0.03) and LBR (11.1% vs. 32.4%, P=0.03) than subjects with a NR in the first cycle. The PR and the LBR in the group with an initial LR were also lower than the overall mean PR (37.2%, P=0.007) and LBR (28.4%, P=0.01) associated with cycles with top quality embryos, but the outcomes among women with an initial NR were similar to the overall mean outcomes (P>0.6).

Table 7. General outcome of women with LR and outcome in cycles with top quality embryos.

NR→LR LR→NR Factor

n or mean SD or % n or mean SD or %

P value

All cycles

Number of LR cycles 42/105 40.0 61/129 47.3 0.3

Age, years 34.2 5.5 34.6 3.9 0.5

BMI, kg/m2 24.4 5.0 24.9 3.6 0.8

Total gonadotropin dose, IU 2522.4 861.5 2690.8 909.5 0.2

Oocytes 5.1 3.1 4.5 3.1 0.1

Fertilisation rate 61.4 31.0 59.2 32.4 0.6

Number of embryos transferred 1.7 0.9 1.5 0.9 0.07

Frozen 0.6 1.5 0.5 1.2 0.3

PR/cycle 17/105 16.2 13/129 10.1 0.2

LBR/cycle 11/105 10.5 8/129 6.2 0.3

≥1 top quality embryo transferred

Number of cycles 34/105 32.4 45/129 34.9 0.8

Number of top quality embryos transferred 1.38 0.49 1.33 0.48 0.7

Total number of transferred embryos 2.0 0.6 1.9 0.5 0.4

PR/cycle 14/34 41.2 8/45 17.8 0.03

LBR/cycle 11/34 32.4 5/45 11.1 0.03

Reference values:

Mean overall PR/cycle 1046/3846 27.2

Mean overall PR/cycle, ≥1 top quality embryo transferred 724/1946 37.2

Mean overall LBR/cycle, ≥1 top quality embryo transferred 553/1946 28.4

The poor outcome after an initial LR cycle, even in top quality embryo cycles, suggests that embryos of good morphological quality may still have a low pregnancy potential in this particular group of patients. Up to 64% of the embryos in patients with poor prognosis and/or low response have been found to have chromosomal abnormalities despite normal morphology [17, 73, 181, 182]. This can explain the low PR in the group with an initial LR. In a previous study in which LR and NR cycles were compared, a similar PR was found when the same number of top quality embryos was transferred in women aged <37 years, but the definition of

63

LR was not the one used in Study I (<5 oocytes compared with <4 in Study I) [346].

5.2 Value of a top quality embryo in fresh and FET cycles

At present, there is no consensus of opinion on the characteristics of a frozen-thawed top quality embryo. In the FET cycles in Study II, embryos were graded with the same criteria used for fresh embryos. The outcomes of the fresh cycles and of the spontaneous FET cycles analysed in Study II were similar in cycles with one top quality embryo transferred (Table 8).

Table 8. Outcome of fresh and spontaneous FET cycles after the transfer of one top quality embryo. Data from cycles collected for Study II.

Fresh cycles Spont. FET cycles Factor

n %

n %

P value

PR/ET, all cycles 745/2128 35.0 94/255 36.9 0.6

MR of cycles analysed in

Study IV

39/167 14.6 24/169 14.2 1.0

LBR, all cycles 610/2128 28.7 74/255 29.0 0.9

Top quality grading [40] takes into account several characteristics of the cleaving embryo: cleavage speed, regularity of blastomeres, absence of multinuclearity and degree of fragmentation. These factors have been found to be of importance for the outcome of frozen-thawed embryos [12, 80, 83, 93] and therefore it is not surprising that results in FET are similar to those of the fresh cycles.

5.3 Top quality embryos and miscarriage

In Study I, the MR in women with LR (36.7%, 11/30, P=0.005) was more than twofold higher than the overall mean MR during the study period (15.5%, 162/1046). An increase in miscarriage was also observed in the cycles with top quality embryos among subjects with LR (27.3%, 6/22, P=0.07), although as a result of low numbers the difference did not reach statistical significance. Analysis of the effect of embryo quality on miscarriage was extended by Study II. In the fresh cycles in Study II, the MR was 22.3% in LR (25/112), 13.9% (188/1353) in NR and 12.2% (87/713) in HR cycles (P=0.02, also for linear association). However, the proportions of cycles with top quality embryos

64

transferred were similar in the fresh cycles with and without miscarriage (58.2% vs. 61.4%, P=0.3) (Table 9).

In fact, the multivariate analysis carried out in Study II revealed that embryo morphology did not determine the MR after IVF/ICSI or FET. Factors with an independent effect on MR are shown in Table 10. The finding that embryo morphology does not affect the MR supports the results of a previous analysis in which it was concluded that embryo morphology determines implantation but that continuation of pregnancy also depends on uterine factors [112]. Likewise, embryo morphology at the time of freezing has not been found to affect the MR after FET [81]. These findings can be explained by research showing that during the 2–3 days before transfer the embryonic genome does not have full control over cell function [18]. Consequently, the genetic composition of an embryo cannot be predicted by its morphology.

5.3.1 Factors affecting MR: age

The higher MR in women with LR (Studies I and II) can be explained by the effect of ovarian ageing. In Study I, 35.7% (81/227) of the subjects with at least one LR cycle had inactive ovaries with few antral follicles, while a much lower incidence of inactive ovaries has been observed previously in unselected IVF/ICSI cycles (10.2%) [192]. The multivariate analysis carried out in Study II revealed that age and not ovarian response is a stronger factor in predicting the probability of miscarriage in fresh cycles. The detrimental effect of female calendar age on pregnancy loss is well documented [17, 73, 182, 284]. The final model in Study II included different age categories, since from a clinical point of view age as a categorical variable is more informative than age as a continuous one. Age <35 years was used as a reference. If the age of the woman was 35–39 years, the MR increased 1.5-fold, and if the age was ≥ 40 years, the MR increased 2.6-fold. Fig. 11 shows the modelled probability of miscarriage in different age groups. Although a cubic effect of age on MR has been previously described [161], such was not found in the multivariate analysis in Study II. A relatively low number of cases in the present analysis might explain this difference, even though to date this is the largest study of miscarriage

Tabl

e 9.

Cha

ract

eris

tics

of th

e st

udy

grou

ps in

Stu

dy II

.

Fres

h cy

cles

S

pont

aneo

us F

ET

Hor

mon

ally

sub

stitu

ted

FET

Mis

carr

iage

N

o m

isca

rria

ge

Mis

carr

iage

N

o m

isca

rria

ge

Mis

carr

iage

N

o m

isca

rria

ge

Cha

ract

eris

tics

n or

mea

n

% o

r

SD

n or

mea

n

% o

r

SD

n or

mea

n

% o

r

SD

n or

mea

n

% o

r

SD

n or

mea

n

% o

r

SD

n or

mea

n

% o

r

SD

n 30

3 13

.8

18

95

86.2

76

11.4

590

88.6

107

23.0

359

77.0

a

Age

, yea

rs

34.0

4.

9

32.6

4.

2b

34.5

3.

8

33.0

4.

1d

34.8

5.

1

33.1

4.

6e

BM

I, kg

/m2

23.6

4.

3

23.2

3.

6

22.9

3.

6

22.7

3.

2

25.3

5.

4

23.2

3.

7b

Obe

sity

(BM

I ≥27

kg/

m2 )

53

17.5

245

12.9

c

7 9.

2

55

9.3

33

30

.8

47

13

.1b

Dys

ovul

atio

n 39

13

.1

20

5 11

.1

5

6.6

33

5.

7

25

24.0

85

24.2

Sec

onda

ry in

ferti

lity

152/

298

51.0

945/

1845

51

.2

56

/76

73.7

355/

573

62.0

70/1

07

65.4

197/

355

55.5

Pre

viou

s m

isca

rria

ge

59/2

17

27.2

316/

1468

21

.5

21

/63

33

.3

95

/514

18

.5f

27

/68

39.7

65/2

45

26.5

Num

ber o

f em

bryo

s

trans

ferr

ed

1.5

0.5

1.

5 0.

5

1.8

0.5

1.

6 0.

5

1.7

0.5

1.

7 0.

5

≥1 to

p qu

ality

em

bryo

trans

ferr

ed

167/

287

58.2

1051

/171

1 61

.4

16

/73

21.9

186/

526

35.4

g

27/9

9 27

.3

86

/330

26

.0

eSE

T 11

3/30

3 37

.3

74

5/18

95

39.3

—

—

—

—

a P

<0.0

001,

com

pare

d w

ith th

e ot

her t

wo

treat

men

t typ

es. b

P<0

.000

1, c

P=0

.04,

d P

=0.

003,

e P

=0.0

02, f P

=0.0

02, g P

=0.0

3, c

ompa

red

with

the

mis

carr

iage

grou

p.

65

66

Table 10. Multivariate analysis of miscarriage <12 weeks.

Factor P value Odds ratio 95% CI a

BMI, kg/m2 0.3 0.98 0.94–1.02

BMI squared 0.013 1.01 1.002–1.011

Age

<35 years — 1

35–39 years 0.004 1.45 1.13–1.86

≥40 years <0.0001 2.64 1.83–3.82

Previous miscarriage

No — 1

Yes 0.002 1.51 1.17–1.95

Type of cycle

Fresh cycle — 1

Spontaneous FET 0.3 0.85 0.63–1.15

Hormonally substituted FET 0.002 1.65 1.21–2.27 a Predicted probability of miscarriage is b/(1+b), where b = Exp (– 2.308 – 0.024*(BMI-23) + 0.006*(BMI-

23)2 + 0.371 [age is 35–39 years (yes=1, no=0)] + 0.972 [age is ≥40 years (yes=1, no=0)] + 0.411

[previous miscarriage (yes=1, no=0)] – 0.162 [spontaneous FET (yes=1, no=0)] + 0.502 [hormonally

substituted FET (yes=1, no=0)]).

(3330 first pregnancy cycles). In the multivariate analysis, ovarian response had no independent effect on miscarriage because of collinearity with age when only one cycle per subject was analysed (Pearson’s correlation coefficient −0.12, P<0.0001). This underlines the association between LR and ovarian ageing [184–188]. Collinearity with age can also explain why in the fresh cycles the number of collected oocytes and the performance of eSET did not affect independently the probability of miscarriage.

The interaction between advancing age, eSET and miscarriage was studied more extensively in Study III. The MR of eSET patients aged 36–39 years was 18.9%. This value was slightly higher than previously reported MRs after eSET in younger women (15.3%–16.8%) [13, 42]. Another reflection of reproductive ageing was the fact that eSET was performed in 27% of women aged 36–39 years, while in the same time period 53% of women aged ≤35 years underwent eSET. However, the MR of the older eSET patients was considerably lower than the MR in the general population aged 35–39 years (25%) [347], indicating that women eligible for eSET in this age group may have a greater ovarian reserve compared with that of the general population of the same age.

67

5.3.2 Factors affecting MR: BMI

In Study II no effect of BMI on MR in the categorical (two and five groups) and linear multivariate regression models was found. A cubic component also did not reach significance. However, in the quadratic regression model, BMI independently affected the MR. Both underweight and obese women have an increased risk of miscarriage, since the relationship between BMI and the MR after IVF/ICSI or FET is not linear but U-shaped (Fig. 11).

This analysis clarifies the long-lived controversy concerning the effect of BMI on MR after IVF/ICSI [261]. In previous studies, underweight and obese women have been found to have lower PRs [54] and higher rates of early pregnancy loss [249] and miscarriage [51] but the results of these studies did not reach statistical significance as a result of to the linear or categorical regression models used. In addition, Study II included a relatively high number of underweight subjects (517 treatment cycles of women with BMI <20 kg/m2 and 113 cycles of women with BMI <18.5 kg/m2). An insufficient number of underweight subjects and/or the statistical methods used might be the reason(s) for the controversial results in previous studies [51, 241, 242, 247, 248, 256].

The higher MR in subjects with low and high BMI may be explained by the action of leptin. Leptin participates in the regulation of uterine angiogenesis [223] and embryo implantation [115, 222]. Low plasma leptin levels leading to insufficient trophoblast support may explain why underweight women have an increased MR. On the other hand, obesity is characterised by high leptin levels [225, 348] and insulin resistance [246]. These factors may be involved in miscarriage through deficient cross-talk between the embryo and the endometrium because of lower expression of the adhesion factors IGFBP-1 and uterine αvβ3 integrin [229-231]. A more hostile endometrial milieu reflected in higher levels of inflammatory markers such as TNF-α [228] and by lower levels of the immunosuppressor glycodelin [230, 232, 349] may also contribute to the higher MR observed in overweight and obese subjects.

5.3.3 Factors affecting MR: type of cycle

In Study II, a hormonally substituted FET cycle was found to increase the probability of miscarriage 1.65-fold (95% CI 1.21–2.27), while spontaneous FET did not affect MR significantly, compared with the fresh cycles. This finding cannot be explained by insufficient hormonal preparation of the endometrium. In

68

FET, endometrial thickness was greater in hormonally substituted (n=157) than in spontaneous FET cycles (n=106) (10.8 ± 2.3 vs. 9.6 ± 1.7 mm, P<0.0001) on the day of embryo transfer (mean: day 17.4 of the cycle for both groups). It is difficult to compare this observation with earlier ones owing to the small study groups in previous analyses of spontaneous and hormonally substituted FET cycles [289, 350–352]. In addition, no international guidelines on ultrasonographic monitoring of the endometrium in FET exist at the moment.

Fig. 11. Modelled probability of miscarriage <12 weeks.

The higher MR in the hormonally substituted cycles can be explained by the fact that hormonal substitution was used in women with ovulatory problems, the most likely aetiology of which was hyperinsulinemia and insulin resistance. Insulin resistance increases the risk of miscarriage independently of BMI or PCOS status [281] and can lead to a less receptive endometrial environment in hormonally

Age < 35 years

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

16 20 24 28 32 36 40

BMI, kg/m2

Prob

abili

ty o

f mis

carr

iage

Age 35-39 years

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

16 20 24 28 32 36 40

BMI, kg/m2

Prob

abili

ty o

f mis

carr

iage

Age ≥ 40 years

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

16 20 24 28 32 36 40

BMI, kg/m2

Prob

abili

ty o

f mis

carr

iage

Lines bottom to top: spontaneous cycle; spontaneous cycle, previous miscarriage(s); hormonally substituted cycle; hormonally substituted cycle with previous miscarriage(s).

69

substituted cycles, characterised by a relative deficiency of adhesion factors and glycodelin. The serum levels of glycodelin and estradiol correlate during ovarian stimulation [353] and it is possible that in hormonally substituted FET cycles endometrial estradiol levels, as well as those of glycodelin, remain lower than those in fresh cycles after ovarian stimulation. This is supported by the finding that in the fresh cycles, dysovulation was not associated with a higher MR, compared with other diagnoses.

5.3.4 Other factors affecting MR

Study II revealed that previous miscarriages also affected the MR. Women who had previously had ≥1 miscarriage were at a 1.6- to twofold greater risk of miscarriage after FET. A similar impact of previous miscarriages on the MR after IVF/ICSI has been observed previously [256], with the risk of miscarriage increasing slightly with the number of miscarriages [284]. This observation can also be explained by changes in glycodelin levels. Lower glycodelin levels have recently been observed in women with recurrent miscarriages [349].

The main infertility diagnosis had no effect on the MR. In this respect, the results of Study II are in accord with those of previous analyses, in which no independent effect of PCOS on MR after controlling for obesity [242, 248, 256] has been observed. The differences among studies in which lower success rates in cases of tubal factor infertility [30, 161, 282], male factor infertility [282] or endometriosis [283] have been found might be explained by different treatment strategies.

In summary, the transfer of a top quality embryo is not associated with a lower MR in either fresh or FET cycles. Both underweight and obese women have an elevated risk of miscarriage. PCOS does not have an independent effect, but women who need hormonally substituted FET because of anovulation have a higher MR. Age, ovarian reserve and previous miscarriages are the other factors that affect the probability of miscarriage during the first trimester of pregnancy.

5.4 eSET and age

In Study III the outcome of IVF/ICSI in women aged 36–39 years was analysed. The main outcome according to the type of embryo transfer is shown in Table 11. In the fresh cycles, the PR/ET (33.1 vs. 29.9%, P=0.3) and the LBR/ET (26.0 vs. 21.9%, P=0.2) did not differ significantly between the eSET and DET groups.

70

However, women in the eSET group had a higher cumulative PR (54.0% vs. 35.0%) and a higher cumulative LBR (41.8% vs. 26.7%, P<0.0001) compared with those in the DET group. The cumulative multiple birth rate in the eSET group was 1.7%, while in the DET group it was 16.6% (P<0.0001).

Table 11. Clinical outcome after eSET and DET in women aged 35–39 years.

eSET DET nt-eSET cSET Factor

n %

n %

n %

n %

Fresh cycles

PR/ET 111/335 33.1 175/585 29.9 21/110 19.1 b 38/194 19.6 b

LBR/ET 87/335 26.0 128/585 21.9 17/110 15.5 c 28/194 14.4 d

MR 0/111 0 31/175 17.7 a 1/21 4.8 0/38 0

Cumulative

PR/OPU 181/335 54.0 205/585 35.0 a 36/110 32.7 b —

LBR/OPU 140/335 41.8 156/585 26.7 a 32/110 29.1 c —

MR 3/181 1.7 34/205 16.6 a 1/36 2.8 — a P<0.0001, b P<0.001, c P<0.05, d P<0.01, compared with the eSET group.

Similar PRs/ET (30.8–34.5%) [10, 33, 354] and LBRs/ET (27.2–29.7%) [10, 13, 42, 354] in the fresh cycle after eSET have been found in previous studies on eSET in younger women. The finding is in contrast to that in a previous analysis of unselected IVF/ICSI cases, in which the chance of live birth among women aged ≥35 years was estimated to be 35% lower than in younger women [355]. This shows that embryo morphology is a stronger determinant of pregnancy than calendar age and suggests that patients should be selected for eSET on the basis of embryo criteria, rather than age.

The cumulative PR after one or two FET cycles in patients aged 36–39 years was 54%, with a live birth rate of over 40%. In younger women even higher values of up to 60% have been reported [12, 13]. The difference can be explained by the results of Study IV, in which higher numbers of collected oocytes (14.4 ± 6.7 vs. 11.9 ± 6.7, P=0.007) and viable embryos (6.9 ± 3.5 vs. 5.9 ± 3.2, P=0.045) were observed in eSET women ≤35 years old than in those aged 36–39 years. These results, however, show that eSET can be safely carried out in all women under the age of 40 years. The outcome after eSET in older women is very acceptable, especially when compared with the cumulative PR in unselected IVF/ICSI subjects with a mean age of 35 years (33.7%) [356].

71

5.4.1 Value of a non-top quality embryo in eSET

In Study III, eSET was also performed in a smaller group of subjects with only non-top quality embryos. As expected, the PR in fresh cycles was relatively low and was similar to that reported previously (18–19%) [12, 64]. However, through several FET cycles the cumulative PR almost doubled and was similar to the cumulative PR after DET. At the same time the MBR was only 2.8%, which is much lower than that observed in the DET group with two non-top quality embryos transferred (18.6%). These results show that eSET of a non-top quality embryo can be beneficial to women who have an increased risk of obstetric complications.

5.5 Embryo transfer policy in cycles with top quality embryo(s)

5.5.1 Overall outcome in the years 1995–2004

The dynamics of eSET and the cumulative outcome of IVF/ICSI at Oulu University Hospital during the years 1995–2004 were analysed in Study IV and are shown in Fig 12. eSET was started in 1996 and during the study period up to 2004 the proportion of eSET cycles increased from 4.2% to 46.6% (P<0.0001). At the same time, the cumulative MBR decreased from 17.1% in 1996 to 4.8% in the 2004 (P<0.0001).

Fig. 12. Dynamics of eSET and cumulative outcome of IVF/ICSI at Oulu University Hospital during the period 1995–2004.

%

0

1 0

2 0

3 0

4 0

5 0

1 9 9 5 1 9 9 6 1 9 9 7 1 9 9 8 1 9 9 9 2 0 0 0 2 0 0 1 2 0 0 2 2 0 0 3 2 0 0 4

e S E Tcu m u la tive P R /OP Ucu m u la tive L B R /OP Ucu m u la tive tL B R /OP Ucu m u la tive MB R

72

Study IV showed that in their starting cycles, women in the eSET and DET periods were comparable in respect of age, BMI, percentage of cases of ICSI and number of viable embryos (numbers not shown). Subjects in the eSET period had fewer fresh cycles (1.5 ± 0.8 vs. 1.6 ± 0.9, P=0.002) and more FET cycles (1.0 ± 1.3 vs. 0.8 ± 1.1, P=0.0001) than those in the DET period. The total number of cycles/woman (2.5 ± 1.7 vs. 2.4 ± 1.6, P=0.06) was similar in the two periods, as was the duration of treatment (eSET 8.8 ± 10.1 vs. DET 9.5 ± 11.8 months, P=0.77).

In contrast to fear of decreased effectiveness as a result of the widespread use of eSET [14, 15], the cumulative pregnancy, live birth and term live birth rates per OPU were higher in the eSET period than in the DET period (Table 12). However, the cumulative MBR/OPU was significantly lower in the eSET period (9.5% vs. 19.6%, P<0.0001). At the same time, the PR in FET cycles increased from 17% in the DET period to 24% in the eSET period, reflecting better embryo implantation potential. This can be explained by the fact that if two or more embryos are routinely transferred in a fresh cycle, only a few patients would have good quality embryo(s) to be frozen. Consequently, the chance of pregnancy in subsequent FET cycles would remain low.

Table 12. Cumulative outcome of IVF/ICSI during the period 1995–2004.

eSET period DET period Outcome factor

n %

n %

P value

Per OPU

Cumulative PR (%) 392/1026 38.2 450/1359 33.1 0.01

Cumulative LBR (%) 333/1026 32.5 371/1359 27.3 0.007

Cumulative MBR (%) 34/392 8.7 84/450 18.7 <0.0001

Cumulative tLBR (%) 287/1026 28.0 302/1359 22.5 0.001

Per woman

Cumulative PR (%) 381/684 55.7 428/826 51.8 0.1

Cumulative LBR (%) 333/684 48.7 373/826 45.0 0.2

Cumulative MBR (%) 34/381 8.9 84/428 19.6 <0.0001

Cumulative tLBR (%) 285/684 41.7 302/826 36.6 0.04

Taking into account all the fresh and FET treatments of each patient, the cumulative pregnancy and live birth rates per woman were similar in the two periods. The main clinical outcome measure of Study IV was the tLBR, which increased (41.7% vs. 36.6%, P=0.04) as a result of the decreased number of multiple births in the eSET period compared with the DET period (8.4% vs.

73

19.6%, P<0.0001). This finding is in agreement with Finnish, Belgian and Swedish national registry data showing a steady decrease in the proportion of multiple births after IVF/ICSI but no change in the LBR [38, 301, 357]. Similar results have also been shown in a clinic in the UK in which eSBT was used instead of eSET [333].

Logistic regression revealed that the single-embryo period independently increased the odds of term live birth (odds ratios 1.37 and 1.43 per woman and per OPU, respectively) (Tables 13 and 14).

Table 13. Results of logistic regression analysis of the odds of term live birth after all fresh and thawed-embryo cycles (taking into account the cumulative tLBR per woman)a.

Factor P value OR (95%CI)

Time period

eSET 0.01 1.37 (1.07–1.75)

DET Reference

Years of follow-up 0.052 0.92 (0.85–1.001)

Age 0.08

<30 years Reference

30–34 years 0.22 1.18 (0.81–1.52)

≥35 years 0.23 0.82 (0.59–1.13)

BMI2 0.001 0.993 (0.989–0.997)

Diagnosis 0.001

Endometriosis Reference

Dysovulation 0.71 0.91 (0.56–1.48)

Male factor 0.53 1.11 (0.80–1.55)

Tubal factor 0.01 0.62 (0.42–0.91)

Unexplained 0.03 0.65 (0.44–0.96)

Number of top quality embryos created <0.0001 1.15 (1.08–1.22) a Values in starting model: period (eSET vs. DET), years of follow-up, age (<30 years, 30–34 years, ≥35

years), BMI2 (centred at 24 kg/m2), diagnosis (excluding cases with multiple diagnoses), method of

fertilisation (IVF vs. ICSI), number of top quality embryos created.

74

Table 14. Results of logistic regression analysis of the odds of term live birth after one fresh and all consequent thawed-embryo cycles (taking into account the cumulative tLBR per OPU)a.

Factor P value OR (95%CI)

Time period

eSET 0.01 1.43 (1.09–1.88)

DET Reference

Years of follow-up 0.08 1.09 (0.99–1.19)

Age 0.01 0.993 (0.988–0.998)

<30 years 0.09

30–34 years Reference

≥35 years 0.28 0.74 (0.42–1.28)

BMI2 0.10 0.70 (0.46–1.07)

Diagnosis 0.006 0.54 (0.35–0.84)

Endometriosis 0.08 0.68 (0.44–1.04)

Dysovulation 0.09 1.35 (0.96–1.90)

Male factor Reference

Tubal factor <0.0001 1.26 (1.18–1.34)

Unexplained 0.08 0.68 (0.44–1.04)

ICSI 0.09 1.35 (0.96–1.90)

IVF Reference

Number of top quality embryos created <0.0001 1.26 (1.18–1.34) a Values in starting model: see Table 13.

The probabilities of term live birth among subjects in the double-embryo period were higher in the “eSET” than in the “DET” estimation (mean ± SD of the difference 0.060 ± 0.007, 95% CI 0.059–0.060, range 0.011–0.070).

More than 90% of the live births in Study IV occurred within four consecutive cycles (fresh or FET): 90.7% (302/333) of the live births in the eSET period and 93.3% (348/373) of the live births in the DET period (P=0.2). An evaluation of time until delivery for the patients in the study periods was also made. Even though the LBR was lower after FET than after fresh transfer, the time until delivery was similar in the DET and the eSET periods (6.7 ± 9.1 vs. 7.4 ± 9.3 months, P=0.2). This suggests that the time to delivery with eSET is not longer than with DET, as has been thought [307].

75

5.5.2 Cost-effectiveness of eSET policy

Cost

The multivariate general linear model showed that the eSET period was associated with a diminished cost of fresh cycles (OR 0.95, P<0.0001) but not the cost of FET cycles (P=0.07) (Table 15). The total treatment costs per woman decreased by €275 (range −€1184 to −€164). This was 4.7% (€275/5890) of the total treatment price per woman in the DET period.

Cost-effectiveness results

Before discounting, the total costs during the eSET period were €3,837,964 while in the DET period they were €4,865,304 (Table 16). After 3% discounting, the total costs per woman were €4584 in the DET period, lower than those in the DET period (€4942). The discounted number of term live births per woman was higher in the eSET period (0.261) than in the DET period (0.243), i.e. the eSET period dominated the DET period. The ICER was equal to −€19,889, meaning that €19,889 was saved per term live birth in the eSET period compared with the DET period.

Sensitivity analysis

Even when uncertainty in clinical parameters was added to the cost-effectiveness model, the eSET period was associated with more adjusted term live births and with lower adjusted total costs, compared with the DET period (Table 17). The confidence limits of the ICER were from −€38,602 to −€852.

By multiplying by 1.5, the “European” ICER became −€29,834 (confidence limits −€75,023 to −€1278). When a multiplication factor of 3.7 was used, the “American” ICER was −€73,589 per term live birth (confidence limits −€185,056 to −€3152).

76

Table 15. Results of the multivariate general linear model of treatment costs per woman in fresh and in thawed-embryo cycles a.

Factor P value OR (95% CI)

Period of treatment

eSET

Cost of fresh cycles <0.0001 0.95 (0.91–0.97)

Cost of FET cycles 0.07 1.03 (0.997–1.07)

DET Reference

Years of follow-up

Cost of fresh cycles <0.0001 0.96 (0.94–0.97)

Cost of FET cycles 0.4 0.995 (0.98–1.01)

Age

Cost of fresh cycles 0.16

<30 years Reference

30–34 years 0.73 1.01 (0.97–1.04)

≥35 years 0.37 0.97 (0.94–1.01)

Cost of FET cycles 0.03

<30 years Reference

30–34 years 0.13 0.98 (0.94–1.02)

≥35 years 0.008 0.94 (0.90–0.98)

Method of fertilisation

ICSI

Cost of fresh cycles <0.0001 1.07 (1.04–1.11)

Cost of FET cycles 0.02 0.96 (0.92–0.99)

IVF Reference

Number of top quality embryos

created

Cost of fresh cycles <0.0001 0.978 (0.971–0.984)

Cost of FET cycles 0.03 1.01 (1.001–1.02) a Values in starting model: see Table 13.

Table 16. Raw and adjusted costs (€) and term live births in the eSET and DET periods.

Cost factor eSET DET

Total costs 3,837,964 4,865,304

Costs of fresh cycles 3,383,250 4,473,172

Costs of FET cycles 454,714 392,133

Total costs per woman 5611 5890

After 3% discounting:

Total costs per woman 4584 4942

Term live births per woman 0.261 0.243

77

Table 17. Summary of variables used in sensitivity analysis and incremental cost-effectiveness ratio in selected sensitivity analyses (€ / term live birth).

Confidence interval

of ICER Variable

Baseline

estimate

Range used

Lower Upper

Baseline ICER −19,889

Number of fresh cycles in period

eSET period 1027 986.9–1071.0 a

DET period 1359 1307.1–1409.0 a −18,645 −10,365

Number of FET cycles in period

eSET period 683 610.1–747.9 a

DET period 589 325.3–650.0 a −50,015 −10,921

Term live birth rate, fresh cycle (%)

eSET period 16.4 14.1–18.6 a

DET period 17.7 15.6–19.7 a −23,723 −17,152

Term live birth rate, FET cycle (%)

eSET period 17.1 14.3–20.0 a

DET period 10.5 8.0–13.0 a −24,683 −16,562

Gonadotropin cost 1329 839–1735 b −22,865 −16,164

Discounting rate (%) 3 0–5 −19,561 −5471

Regression model

tLBR improvement in eSET

period

0.178 c 0.031–0.193 d

Improvement in total treatment

cost per woman in eSET period

−275.05 c −1183.53 to

−164.30 d

Mean ICER from regression

model

−1545 −38,602 −852

a 95% confidence intervals. b Standard deviation of the cost of the mean gonadotropin dose used. c Mean

from regression model. d Range of the difference.

Cost-effectiveness discussion

Study IV is the first evaluation of the overall cost-effectiveness of eSET policy in everyday clinical practice. By contrast, previous cost-effectiveness analyses have involved use of highly selected patients and predetermined treatment protocols, which have limited the validity of their results only to the settings of a randomised controlled trial [315]. When evaluating only costs of fresh cycles, these studies have revealed that eSET is inferior to DET either because of lower LBR [43, 45, 325, 327, 358] or because of higher costs [326].

78

The discrepancy with the present analysis can be explained by the high number of FET cycles during the eSET period, which are much cheaper than fresh cycles. In the eSET period, a higher proportion of all deliveries resulted from FET cycles, compared with the DET period (49% vs. 35%). This underlines the value of cryopreservation [10, 12], which until now has been underestimated.

Concerns about the economic impact of eSET policy [14, 15] are refuted by the present results. Study IV covered only the costs of IVF treatment, but it is expected that the long-term financial benefits of eSET will be significant. At Oulu University Hospital, if the costs of maternal and paediatric care up to the end of the neonatal period [321] were to have been included in the present analysis, the ICER would have increased about 1.9-fold to about €36,889 per term live birth in favour of eSET. Accordingly, earlier cost-effectiveness analyses have revealed lower overall costs with eSET if expenses for pregnancy, neonatal and paediatric care are also taken into account [43, 326, 327].

The fact that with eSET followed by cryopreservation a term live birth is more likely and less expensive can make IVF/ICSI treatment affordable for more patients. Even in developed countries only one fifth of infertile patients receive IVF/ICSI treatment [319], most likely because of the treatment costs, which can be as high as 25% of the annual household expenditure [4]. Therefore, eSET could also be beneficial in countries where patients pay for IVF/ICSI themselves. It can be estimated that if the price of IVF/ICSI decreases by 5% with eSET, as shown in the present study, a 15% increase in utilisation can be expected owing to the high price elasticity of IVF/ICSI treatment (3.0) [4].

In all countries in which eSET is practised on a large scale, a decreased number of multiple births has been reported, while the pregnancy rate has remained unchanged [3, 38, 52, 301], suggesting that the analysis made in Finland would also be valid in other countries. The results of Study IV support theconclusions made by Ledger et al. that many more IVF/ICSI cycles could be funded through savings made after the adoption of an eSET policy [322].

Mild ovarian stimulation together with eSET is another treatment strategy aimed at diminishing the costs of treatment and multiple births [56]. However, the outcome of mild stimulation seems to be inferior to the outcome of eSET after stimulation by means of the long protocol. eSET with the standard long protocol decreases the costs of IVF treatment and increases the tLBR, whereas with mild stimulation, the tLBR and costs of treatment remain similar to those associated with DET.

79

In summary, Study IV showed that the eSET policy dominates the policy of DET in women aged less than 40 years. More term live births are observed and substantial financial savings in IVF treatment itself are made without compromising the speed of treatment.

5.5.3 Limitations of the study

Data for the studies included in this thesis were collected retrospectively. As elsewhere in the Nordic countries, ovarian stimulation was monitored by ultrasonographic examination only. Hormonal analysis was not routinely performed, meaning that the analyses had to be limited only to anthropometric, ultrasonographic and laboratory (oocytes and embryos) variables. In Study I, analysis of ovarian activity was based on data of three groups (inactive, normal or polycystic-like ovaries, depending on the number of antral follicles) while the exact antral follicle counts were not recorded. Because of this, precise evaluation of the relationship between the number of antral follicles and ovarian response in three consecutive cycles was not possible. In Study II, despite consistency of data from the five participating clinics, numbers of gestational sacs were not available for all cycles analysed and could not be included in the multivariate analysis. However, this does not undermine the conclusions in Study II, since embryo morphology and age affect the MR independently of the number of gestational sacs [282]. Finally, an analysis of the importance of male factors as regards the outcome of top quality embryo transfer was found to be beyond the scope of the present thesis.

80

81

6 Conclusions

Transfer of a top quality embryo results in higher pregnancy and live birth rates, as shown in all studies included in the present thesis. However, if a previous cycle has resulted in less than four collected oocytes, the prognosis is poor even if the ovarian response is improved in the subsequent cycles and top quality embryos are created. On the other hand, when the ovarian response is diminished, the prognosis is good if a previous stimulation has resulted in at least four collected oocytes. These findings are important, since a low ovarian response seldom recurs (2.5%) and the majority of LR patients have both normal and low response cycles. In addition, the poor pregnancy and live birth rates after the transfer of a top quality embryo in subjects with an initial low response indicate that eSET might not be routinely performed among such women.

Transfer of a top quality embryo is also associated with better outcome in frozen-thawed cycles. The results show that an embryo with top quality characteristics before FET might have the same potential as a fresh top quality embryo. However, in contrast to these findings, the transfer of a top quality embryo does not diminish the probability of miscarriage after a fresh or a frozen-thawed embryo transfer. Multivariate analysis showed that the MR increases along with age and is elevated in cycles of women with low response as a result of a diminished ovarian reserve. Both extremes of BMI (underweight and obese) are also characterised by an increased probability of miscarriage, probably caused by altered endometrial function. Subjects with decreased glucose tolerance, such as obese women and those who need hormonal substitution for FET because of anovulation, have a suboptimal endometrium and a high MR. Miscarriages occur through defective embryo-endometrial crosstalk and through an endometrial milieu with increased inflammatory and decreased immunosuppressive factors. Finally, a previous miscarriage is another factor associated with higher probability of miscarriage in the treatment cycle.

Embryo morphology is a stronger determinant of pregnancy than calendar age. This is why elective transfer of a single top quality embryo is a safe treatment strategy regardless of the woman’s age, at least up to the age of 40 years. In this age group, eSET results in cumulative pregnancy rates of 50–60%, while multiple gestations occur in only about 2% of pregnancies. However, eSET is affected by age, since the proportion of suitable patients decreases with age. This is illustrated by the fact that during the time period analysed, eSET was

82

performed in 56% of the cycles of women of ≤35 years and in only 27% of the cycles of those aged 36–39 years.

The present thesis provides sufficient evidence that eSET is an effective treatment option. The proportion of eSET cycles increased from 4% in 1996 to 47% in 2004. At the same time, the cumulative multiple birth rate per ovum pickup decreased from 17% to 5%, while the cumulative pregnancy and live birth rates per woman remained stable. The increased use of eSET raised the proportion of pregnancies delivered at term as a result of the smaller number of multiple gestations. Contrary to predictions, the time required for treatment does not increase with eSET. More than 90% of births occur within four treatment cycles, but if eSET is performed, patients have relatively fewer fresh and more frozen-thawed cycles. Because frozen-thawed cycles cost less than fresh cycles, costs of IVF/ICSI treatment are smaller with eSET such that the implementation of eSET diminishes treatment costs at a rate of €19,825 per term live birth.

Future perspectives

This work resolved debates about the applicability of eSET and about the effect of obesity in IVF/ICSI, but many more questions remain unanswered. How much does the cumulative outcome vary with respect to the total number of top quality embryos? How can the outcome in hormonally substituted FET be improved? What is the effect of male factors on the outcome of top quality embryo transfer?

This work showed that use of a top quality embryo in FET has the same potential as a fresh top quality embryo. However, since ≈80% of embryos transferred in FET do not have top quality morphology, more detailed evaluation of frozen-thawed embryos is necessary in order to perform eSET in cryopreserved cycles and to decrease the multiple pregnancy rate further.

83

References 1. Templeton A, Fraser C & Thompson B (1991) Infertility--epidemiology and referral

practice. Hum Reprod 6(10): 1391–1394. 2. De Mouzon J (2006) IVF monitoring worldwide (ICMART). 22nd Annual Meeting of

the ESHRE, Prague, Czech Republic, 18–21 June 2006, Abstract O–193. 3. Nyboe Andersen A, Goossens V, Gianaroli L, Felberbaum R, de Mouzon J & Nygren

KG (2007) Assisted reproductive technology in Europe, 2003. Results generated from European registers by ESHRE. Hum Reprod 22(6): 1513-1525. Epub 2007 Apr 30.

4. Collins J (2002) An international survey of the health economics of IVF and ICSI. Hum Reprod Update 8(3): 265–277.

5. Dickey RP (2007) The relative contribution of assisted reproductive technologies and ovulation induction to multiple births in the United States 5 years after the Society for Assisted Reproductive Technology/American Society for Reproductive Medicine recommendation to limit the number of embryos transferred. Fertil Steril 88(6) 1554–1561.

6. ESHRE (2000) Multiple gestation pregnancy. The ESHRE Capri Workshop Group. Hum Reprod 15(8): 1856–1864.

7. Pinborg A, Loft A, Schmidt L, Greisen G, Rasmussen S & Andersen AN (2004) Neurological sequelae in twins born after assisted conception: controlled national cohort study. BMJ (7461): 311. Epub 2004 Jul 15.

8. Reynolds MA, Schieve LA (2006) Trends in embryo transfer practices and multiple gestation for IVF procedures in the USA, 1996–2002. Hum Reprod 21(3): 694–700. Epub 2005 Oct 27.

9. Vilska S, Tiitinen A, Hyden-Granskog C & Hovatta O (1999) Elective transfer of one embryo results in an acceptable pregnancy rate and eliminates the risk of multiple birth. Hum Reprod 14(9): 2392–2395.

10. Martikainen H, Tiitinen A, Tomas C, Tapanainen J, Orava M, Tuomivaara L et al. (2001) One versus two embryo transfer after IVF and ICSI: a randomized study. Hum Reprod 16(9): 1900–1903.

11. Bergh C (2005) Single embryo transfer: a mini-review. Hum Reprod 20(2): 323–327. Epub 2005 Jan 21.

12. Tiitinen A, Halttunen M, Harkki P, Vuoristo P & Hyden-Granskog C (2001) Elective single embryo transfer: the value of cryopreservation. Hum Reprod 16(6): 1140–1144.

13. Martikainen H, Orava M, Lakkakorpi J & Tuomivaara L (2004) Day 2 elective single embryo transfer in clinical practice: better outcome in ICSI cycles. Hum Reprod 19(6): 1364–1366. Epub 2004 Apr 22.

14. Bhattacharya S & Templeton A (2004) What is the most relevant standard of success in assisted reproduction? Redefining success in the context of elective single embryo transfer: evidence, intuition and financial reality. Hum Reprod 19(9): 1939–1942. Epub 2004 Jun 24.

15. Gleicher N & Barad D (2006) The relative myth of elective single embryo transfer. Hum Reprod 21(6): 1337–1344. Epub 2006 Feb 22.

84

16. Short R (2007) HFEA wants greater use of single embryo transfers in assisted reproduction. BMJ 334(7597): 766.

17. Magli MC, Gianaroli L, Munne S & Ferraretti AP (1998) Incidence of chromosomal abnormalities from a morphologically normal cohort of embryos in poor-prognosis patients. J Assist Reprod Genet 15(5): 297–301.

18. Los FJ, Van Opstal D & van den Berg C (2004) The development of cytogenetically normal, abnormal and mosaic embryos: a theoretical model. Hum Reprod Update 10(1): 79–94.

19. Edwards RG & Steptoe PC (1983) Current status of in-vitro fertilisation and implantation of human embryos. Lancet 2(8362): 1265–1269.

20. Hoult IJ, de Crespigny LC, O'Herlihy C, Speirs AL, Lopata A, Kellow G et al. (1981) Ultrasound control of clomiphene/human chorionic gonadotropin stimulated cycles for oocyte recovery and in vitro fertilization. Fertil Steril 36(3): 316–319.

21. Wortham JW Jr, Veeck LL, Witmyer J, Sandow BA & Jones HW Jr (1983) Vital initiation of pregnancy (VIP) using human menopausal gonadotropin and human chorionic gonadotropin ovulation induction: phase II–1981. Fertil Steril 40(2): 170–177.

22. Porter RN, Smith W, Craft IL, Abdulwahid NA & Jacobs HS (1984) Induction of ovulation for in-vitro fertilisation using buserelin and gonadotropins. Lancet 2(8414): 1284–1285.

23. Palermo G, Joris H, Devroey P & Van Steirteghem AC (1992) Pregnancies after intracytoplasmic injection of single spermatozoon into an oocyte. Lancet 340(8810): 17–18.

24. Cohen J (1993) World collaborative report 1991. VIII World congress on in vitro fertilization and alternate assisted reproduction. Kyoto, Japan.

25. Staessen C, Camus M, Bollen N, Devroey P & Van Steirteghem AC (1992) The relationship between embryo quality and the occurrence of multiple pregnancies. Fertil Steril 57(3): 626–630.

26. Tasdemir M, Tasdemir I, Kodama H, Fukuda J & Tanaka T (1995) Two instead of three embryo transfer in in-vitro fertilization. Hum Reprod 10(8): 2155–2158.

27. Staessen C, Janssenswillen C, Van den Abbeel E, Devroey P & Van Steirteghem AC (1993) Avoidance of triplet pregnancies by elective transfer of two good quality embryos. Hum Reprod 8(10): 1650–1653.

28. Pinborg A, Loft A & Nyboe Andersen A (2004) Neonatal outcome in a Danish national cohort of 8602 children born after in vitro fertilization or intracytoplasmic sperm injection: the role of twin pregnancy. Acta Obstet Gynecol Scand 83(11): 1071–1078.

29. Wang YA, Sullivan EA, Black D, Dean J, Bryant J & Chapman M (2005) Preterm birth and low birth weight after assisted reproductive technology-related pregnancy in Australia between 1996 and 2000. Fertil Steril 83(6): 1650–1658.

30. Templeton A & Morris JK (1998) Reducing the risk of multiple births by transfer of two embryos after in vitro fertilization. N Engl J Med 339(9): 573–577.

85

31. Nygren KG & Andersen AN (2001) Assisted reproductive technology in Europe, 1998. Results generated from European registers by ESHRE. European Society of Human Reproduction and Embryology. Hum Reprod 16(11): 2459–2471.

32. Schieve LA, Peterson HB, Meikle SF, Jeng G, Danel I, Burnett NM et al. (1999) Live-birth rates and multiple-birth risk using in vitro fertilization. JAMA 282(19): 1832–1838.

33. Kissin DM, Schieve LA & Reynolds MA (2005) Multiple-birth risk associated with IVF and extended embryo culture: USA, 2001. Hum Reprod 20 (8): 2215–2223. Epub 2005 Apr 14.

34. Pinborg A (2005) IVF/ICSI twin pregnancies: risks and prevention. Hum Reprod Update 11(6): 575–593. Epub 2005 Aug 25.

35. ESHRE Campus Course Report (2001) Prevention of twin pregnancies after IVF/ICSI by single embryo transfer. Hum Reprod 16(4): 790–800.

36. Jones HW (2003) Multiple births: how are we doing? Fertil Steril 79(1): 17–21. 37. Jones HW, Jr (2007) Iatrogenic multiple births: a 2003 checkup. Fertil Steril 87(3):

453–455. 38. Karlström PO & Bergh C (2007) Reducing the number of embryos transferred in

Sweden — impact on delivery and multiple birth rates. Hum Reprod 22(8): 2202–2207. Epub 2007 Jun 11.

39. Giorgetti C, Terriou P, Auquier P, Hans E, Spach JL, Salzmann J et al. (1995) Embryo score to predict implantation after in-vitro fertilization: based on 957 single embryo transfers. Hum Reprod 10(9): 2427–2431.

40. Van Royen E, Mangelschots K, De Neubourg D, Valkenburg M, Van de Meerssche M, Ryckaert G et al. (1999) Characterization of a top quality embryo, a step towards single-embryo transfer. Hum Reprod 14(9): 2345–2349.

41. Gerris J, De Neubourg D, Mangelschots K, Van Royen E, Van de Meerssche M & Valkenburg M (1999) Prevention of twin pregnancy after in-vitro fertilization or intracytoplasmic sperm injection based on strict embryo criteria: a prospective randomized clinical trial. Hum Reprod 14(10): 2581–2587.

42. Thurin A, Hausken J, Hillensjo T, Jablonowska B, Pinborg A, Strandell A et al. (2004) Elective single-embryo transfer versus double-embryo transfer in in vitro fertilization. N Engl J Med 351(23): 2392–2402.

43. Gerris J, De Sutter P, De Neubourg D, Van Royen E, Vander Elst J, Mangelschots K et al. (2004) A real-life prospective health economic study of elective single embryo transfer versus two-embryo transfer in first IVF/ICSI cycles. Hum Reprod 19(4): 917–923. Epub 2004 Feb 27.

44. De Neubourg D, Gerris J, Mangelschots K, Van Royen E, Vercruyssen M & Elseviers M (2004) Single top quality embryo transfer as a model for prediction of early pregnancy outcome. Hum Reprod 19(6): 1476–1479. Epub 2004 Apr 29.

45. Lukassen HG, Braat DD, Wetzels AM, Zielhuis GA, Adang EM, Scheenjes E et al. (2005) Two cycles with single embryo transfer versus one cycle with double embryo transfer: a randomized controlled trial. Hum Reprod 20(3): 702–708. Epub 2004 Dec 23.

86

46. The Practice Committee of the Society for Assisted Reproductive Tehcnology and the Practice Committee of the American Society for Reproductive Medicine (2006) Guidelines on number of embryos transferred. Fertil Steril 86(5 Suppl): S51–2.

47. Gunby J, Bissonnette F, Librach C & Cowan L (2008) Assisted reproductive technologies (ART) in Canada: 2004 results from the Canadian ART Register. Fertil Steril 89(5): 1123v132.

48. Jain T, Missmer SA & Hornstein MD (2004) Trends in embryo-transfer practice and in outcomes of the use of assisted reproductive technology in the United States. N Engl J Med 350(16): 1639–1645.

49. Gissler M & Heimo A (2008) Stakesin keräämät IVF-tilastot. STAKES. 50. Min JK, Breheny SA, MacLachlan V & 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(1):3–7.

51. Fedorcsák P, Dale PO, Storeng R, Ertzeid G, Bjercke S, Oldereid N et al. (2004) Impact of overweight and underweight on assisted reproduction treatment. Hum Reprod 19(11): 2523–2528. Epub 2004 Aug 19.

52. Tiitinen A, Hyden-Granskog C & Gissler M (2004) What is the most relevant standard of success in assisted reproduction?: The value of cryopreservation on cumulative pregnancy rates per single oocyte retrieval should not be forgotten. Hum Reprod 19(11): 2439–2441. Epub 2004 Aug 19.

53. Check JH, Lurie D, Callan C, Baker A & Benfer K (1994) Comparison of the cumulative probability of pregnancy after in vitro fertilization-embryo transfer by infertility factor and age. Fertil Steril 61(2): 257–261.

54. Wang JX, Davies M & Norman RJ (2000) Body mass and probability of pregnancy during assisted reproduction treatment: retrospective study. BMJ.;321(7272):1320–1321.

55. Koivurova S, Hartikainen AL, Gissler M, Hemminki E, Sovio U & Jarvelin MR (2002) Neonatal outcome and congenital malformations in children born after in-vitro fertilization. Hum Reprod 17(5): 1391–1398.

56. Heijnen EM, Eijkemans MJ, De Klerk C, Polinder S, Beckers NG, Klinkert ER, et al. (2007) A mild treatment strategy for in-vitro fertilisation: a randomised non-inferiority trial. Lancet 369(9563): 743–749.

57. Eijkemans MJ, Heijnen EM, de Klerk C, Habbema JD & Fauser BC (2006) Comparison of different treatment strategies in IVF with cumulative live birth over a given period of time as the primary end-point: methodological considerations on a randomized controlled non-inferiority trial. Hum Reprod 21(2): 344–351. Epub 2005 Oct 20.

58. Cummins JM, Breen TM, Harrison KL, Shaw JM, Wilson LM & Hennessey JF (1986) A formula for scoring human embryo growth rates in in vitro fertilization: its value in predicting pregnancy and in comparison with visual estimates of embryo quality. J In Vitro Fert Embryo Transf 3(5): 284–295.

87

59. Steer CV, Mills CL, Tan SL, Campbell S & Edwards RG (1992) The cumulative embryo score: a predictive embryo scoring technique to select the optimal number of embryos to transfer in an in-vitro fertilization and embryo transfer programme. Hum Reprod 7(1):117–119.

60. Bertrand E, Van den Bergh M & Englert Y (1995) Does zona pellucida thickness influence the fertilization rate? Hum Reprod 10(5): 1189–1193.

61. Jackson KV, Ginsburg ES, Hornstein MD, Rein MS & Clarke RN (1998) Multinucleation in normally fertilized embryos is associated with an accelerated ovulation induction response and lower implantation and pregnancy rates in in vitro fertilization-embryo transfer cycles. Fertil Steril 70(1): 60–66.

62. Shoukir Y, Campana A, Farley T & Sakkas D (1997) Early cleavage of in-vitro fertilized human embryos to the 2-cell stage: a novel indicator of embryo quality and viability. Hum Reprod 12(7): 1531–1536.

63. Lundin K, Bergh C & Hardarson T (2001) Early embryo cleavage is a strong indicator of embryo quality in human IVF. Hum Reprod 16(12): 2652–2657.

64. Saldeen P & Sundstrom P (2005) Would legislation imposing single embryo transfer be a feasible way to reduce the rate of multiple pregnancies after IVF treatment? Hum Reprod 20(1): 4–8. Epub 2004 Nov 26.

65. Gardner DK, Vella P, Lane M, Wagley L, Schlenker T & Schoolcraft WB (1998) Culture and transfer of human blastocysts increases implantation rates and reduces the need for multiple embryo transfers. Fertil Steril 69(1):84–88.

66. Blake DA, Farquhar CM, Johnson N & Proctor M (2007) Cleavage stage versus blastocyst stage embryo transfer in assisted conception. Cochrane Database Syst Rev (4): CD002118.

67. Angell RR, Aitken RJ, van Look PF, Lumsden MA & Templeton AA (1983) Chromosome abnormalities in human embryos after in vitro fertilization. Nature 303(5915): 336–338.

68. Shahine LK & Cedars MI (2006) Preimplantation genetic diagnosis does not increase pregnancy rates in patients at risk for aneuploidy. Fertil Steril 85(1): 51–6.

69. Munne S, Lee A, Rosenwaks Z, Grifo J & Cohen J (1993) Diagnosis of major chromosome aneuploidies in human preimplantation embryos. Hum Reprod 8(12): 2185–2191.

70. Moutou C, Gardes N & Viville S (2004) Duplex, triplex and quadruplex PCR for the preimplantation genetic diagnosis (PGD) of cystic fibrosis (CF), an exhaustive approach. Prenat Diagn 24(7): 62–69.

71. Sermon K, De Rijcke M, Lissens W, De Vos A, Platteau P, Bonduelle M et al. (2002) Preimplantation genetic diagnosis for Huntington's disease with exclusion testing. Eur J Hum Genet 10(10): 591–598.

72. Lahdetie J & Horelli-Kuitunen N (2001) [Preimplantation diagnostics]. Duodecim 117(22): 2257–2263.

88

73. Gianaroli L, Magli MC, Ferraretti AP & Munne S (1999) Preimplantation diagnosis for aneuploidies in patients undergoing in vitro fertilization with a poor prognosis: identification of the categories for which it should be proposed. Fertil Steril 72(5): 837–844.

74. Munne S (2002) Preimplantation genetic diagnosis of numerical and structural chromosome abnormalities. Reprod Biomed Online 4(2): 183–196.

75. Twisk M, Mastenbroek S, van Wely M, Heineman MJ, Van der Veen F & Repping S (2006) Preimplantation genetic screening for abnormal number of chromosomes (aneuploidies) in in vitro fertilisation or intracytoplasmic sperm injection. Cochrane Database Syst Rev (1): CD005291.

76. Sandalinas M, Sadowy S, Alikani M, Calderon G, Cohen J & Munne S (2001) Developmental ability of chromosomally abnormal human embryos to develop to the blastocyst stage. Hum Reprod 16(9): 1954–1958.

77. Leese HJ (1987) Analysis of embryos by non-invasive methods. Hum Reprod 2(1): 37–40.

78. Brison DR, Hollywood K, Arnesen R & Goodacre R (2007) Predicting human embryo viability: the road to non-invasive analysis of the secretome using metabolic footprinting. Reprod Biomed Online 15(3): 296–302.

79. Trounson A & Mohr L (1983) Human pregnancy following cryopreservation, thawing and transfer of an eight-cell embryo. Nature 305(5936): 707–709.

80. Testart J, Lassalle B, Belaisch-Allart J, Hazout A, Forman R, Rainhorn JD et al. (1986) High pregnancy rate after early human embryo freezing. Fertil Steril 46(2): 268–272.

81. Salumets A, Suikkari AM, Makinen S, Karro H, Roos A & Tuuri T (2006) Frozen embryo transfers: implications of clinical and embryological factors on the pregnancy outcome. Hum Reprod.;21(9):2368–2374. Epub 2006 May 9.

82. Van den Abbeel E, Van der Elst J, Van Waesberghe L, Camus M, Devroey P, Khan I et al. (1988) Hyperstimulation: the need for cryopreservation of embryos. Hum Reprod 3(Suppl 2): 53–57.

83. Hartshorne GM, Wick K, Elder K & Dyson H (1990) Effect of cell number at freezing upon survival and viability of cleaving embryos generated from stimulated IVF cycles. Hum Reprod 5(7): 857–861.

84. Toner JP, Veeck LL, Acosta AA & Muasher SJ (1991) Predictive value of pregnancy during original in vitro fertilization cycle on implantation and pregnancy in subsequent cryothaw cycles. Fertil Steril 56(3): 505–508.

85. El-Toukhy T, Khalaf Y, Al-Darazi K, O'Mahony F, Wharf E, Taylor A et al. (2003) Cryo-thawed embryos obtained from conception cycles have double the implantation and pregnancy potential of those from unsuccessful cycles. Hum Reprod 18(6): 1313–1318.

86. Lassalle B, Testart J & Renard JP (1985) Human embryo features that influence the success of cryopreservation with the use of 1,2 propanediol. Fertil Steril 44(5): 645–651.

89

87. Demoulin A, Jouan C, Gerday C & Dubois M (1991) Pregnancy rates after transfer of embryos obtained from different stimulation protocols and frozen at either pronucleate or multicellular stages. Hum Reprod 6(6): 799–804.

88. Senn A, Vozzi C, Chanson A, De Grandi P & Germond M (2000) Prospective randomized study of two cryopreservation policies avoiding embryo selection: the pronucleate stage leads to a higher cumulative delivery rate than the early cleavage stage. Fertil Steril 74(5):946–952.

89. Mandelbaum J, Junca AM, Plachot M, Alnot MO, Alvarez S, Debache C et al. (1987) Human embryo cryopreservation, extrinsic and intrinsic parameters of success. Hum Reprod 2(8): 709–15.

90. Testart J, Lassalle B, Forman R, Gazengel A, Belaisch-Allart J, Hazout A, et al. (1987) Factors influencing the success rate of human embryo freezing in an in vitro fertilization and embryo transfer program. Fertil Steril 48(1): 107–112.

91. Salumets A, Tuuri T, Makinen S, Vilska S, Husu L, Tainio R, et al. (2003) Effect of developmental stage of embryo at freezing on pregnancy outcome of frozen-thawed embryo transfer. Hum Reprod 18(9): 1890–1895.

92. Ziebe S, Bech B, Petersen K, Mikkelsen AL, Gabrielsen A & Andersen AN (1998) Resumption of mitosis during post-thaw culture: a key parameter in selecting the right embryos for transfer. Hum Reprod 13(1): 178–181.

93. Guerif F, Bidault R, Cadoret V, Couet ML, Lansac J & Royere D (2002) Parameters guiding selection of best embryos for transfer after cryopreservation: a reappraisal. Hum Reprod 17(5): 1321–1326.

94. Laverge H, Van der Elst J, De Sutter P, Verschraegen-Spae MR, De Paepe A & Dhont M (1998) Fluorescent in-situ hybridization on human embryos showing cleavage arrest after freezing and thawing. Hum Reprod 13(2): 425–429.

95. Iwarsson E, Lundqvist M, Inzunza J, Ahrlund-Richter L, Sjoblom P, Lundkvist O et al. (1999) A high degree of aneuploidy in frozen-thawed human preimplantation embryos. Hum Genet 104(5): 376–382.

96. Xia P (1997) Intracytoplasmic sperm injection: correlation of oocyte grade based on polar body, perivitelline space and cytoplasmic inclusions with fertilization rate and embryo quality. Hum Reprod 12(8): 1750–1755.

97. Loutradis D, Drakakis P, Kallianidis K, Milingos S, Dendrinos S & Michalas S (1999) Oocyte morphology correlates with embryo quality and pregnancy rate after intracytoplasmic sperm injection. Fertil Steril 72(2): 240–244.

98. Ebner T, Yaman C, Moser M, Sommergruber M, Feichtinger O & Tews G (2000) Prognostic value of first polar body morphology on fertilization rate and embryo quality in intracytoplasmic sperm injection. Hum Reprod 15(2): 427–30.

99. Salumets A, Suikkari AM, Mols T, Soderstrom-Anttila V & Tuuri T (2002) Influence of oocytes and spermatozoa on early embryonic development. Fertil Steril 78(5): 1082–1087.

100. de la Rochebrochard E & Thonneau P (2002) Paternal age and maternal age are risk factors for miscarriage; results of a multicentre European study. Hum Reprod 17(6): 1649–1656.

90

101. Frattarelli JL, Miller KA, Miller BT, Elkind-Hirsch K & Scott RT, Jr (2008) Male age negatively impacts embryo development and reproductive outcome in donor oocyte assisted reproductive technology cycles. Fertil Steril 90(1): 97–103.

102. Klonoff-Cohen HS & Natarajan L (2004) The effect of advancing paternal age on pregnancy and live birth rates in couples undergoing in vitro fertilization or gamete intrafallopian transfer. Am J Obstet Gynecol 191(2): 507–514.

103. Rogers PA, Milne BJ & Trounson AO (1986) A model to show human uterine receptivity and embryo viability following ovarian stimulation for in vitro fertilization. J In Vitro Fert Embryo Transf 3(2):93–98.

104. Aghajanova L, Hamilton AE & Giudice LC (2008) Uterine receptivity to human embryonic implantation: Histology, biomarkers, and transcriptomics. Semin Cell Dev Biol 19(2): 201–211.

105. Martel D, Frydman R, Glissant M, Maggioni C, Roche D & Psychoyos A (1987) Scanning electron microscopy of postovulatory human endometrium in spontaneous cycles and cycles stimulated by hormone treatment. J Endocrinol 114(2): 319–324.

106. Develioglu OH, Hsiu JG, Nikas G, Toner JP, Oehninger S & Jones HW, Jr (1999) Endometrial estrogen and progesterone receptor and pinopode expression in stimulated cycles of oocyte donors. Fertil Steril 71(6): 1040–1047.

107. Nikas G, Develioglu OH, Toner JP & Jones HW, Jr (1999) Endometrial pinopodes indicate a shift in the window of receptivity in IVF cycles. Hum Reprod 14(3): 787–792.

108. Navot D, Anderson TL, Droesch K, Scott RT, Kreiner D & Rosenwaks Z (1989) Hormonal manipulation of endometrial maturation. J Clin Endocrinol Metab 68(4): 801–807.

109. Seppälä M, Taylor RN, Koistinen H, Koistinen R & Milgrom E (2002) Glycodelin: a major lipocalin protein of the reproductive axis with diverse actions in cell recognition and differentiation. Endocr Rev 23(4):401–430.

110. Halttunen M, Kamarainen M & Koistinen H (2000) Glycodelin: a reproduction-related lipocalin. Biochim Biophys Acta 1482(1–2): 149–156.

111. Tulppala M, Julkunen M, Tiitinen A, Stenman UH & Seppala M (1995) Habitual abortion is accompanied by low serum levels of placental protein 14 in the luteal phase of the fertile cycle. Fertil Steril 63(4): 792–795.

112. Lambers MJ, Mager E, Goutbeek J, McDonnell J, Homburg R, Schats R et al. (2007) Factors determining early pregnancy loss in singleton and multiple implantations. Hum Reprod 22(1):275–279. Epub 2006 Sep 14.

113. Caballero-Campo P, Dominguez F, Coloma J, Meseguer M, Remohi J, Pellicer A et al. (2002) Hormonal and embryonic regulation of chemokines IL-8, MCP-1 and RANTES in the human endometrium during the window of implantation. Mol Hum Reprod 8(4): 375–384.

114. De los Santos MJ, Mercader A, Frances A, Portoles E, Remohi J, Pellicer A et al. (1996) Role of endometrial factors in regulating secretion of components of the immunoreactive human embryonic interleukin-1 system during embryonic development. Biol Reprod 54(3): 563–574.

91

115. Gonzalez RR, Caballero-Campo P, Jasper M, Mercader A, Devoto L, Pellicer A et al. (2000) Leptin and leptin receptor are expressed in the human endometrium and endometrial leptin secretion is regulated by the human blastocyst. J Clin Endocrinol Metab 85(12): 4883–4888.

116. Simon C, Gimeno MJ, Mercader A, O'Connor JE, Remohi J, Polan ML et al. (1997) Embryonic regulation of integrins beta 3, alpha 4, and alpha 1 in human endometrial epithelial cells in vitro. J Clin Endocrinol Metab 82(8):2607–2616.

117. Meseguer M, Aplin JD, Caballero-Campo P, O'Connor JE, Martin JC, Remohi J et al. (2001) Human endometrial mucin MUC1 is up-regulated by progesterone and down-regulated in vitro by the human blastocyst. Biol Reprod 64(2): 590–601.

118. Galan A, O'Connor JE, Valbuena D, Herrer R, Remohi J, Pampfer S, et al. (2000) The human blastocyst regulates endometrial epithelial apoptosis in embryonic adhesion. Biol Reprod 63(2):430–439.

119. Gonzalez RR, Simon C, Caballero-Campo P, Norman R, Chardonnens D, Devoto L et al. (2000) Leptin and reproduction. Hum Reprod Update 6(3): 290–300.

120. Irving JA & Lala PK (1995) Functional role of cell surface integrins on human trophoblast cell migration: regulation by TGF-beta, IGF-II, and IGFBP-1. Exp Cell Res 217(2): 419–427.

121. Giudice LC, Mark SP & Irwin JC (1998) Paracrine actions of insulin-like growth factors and IGF binding protein-1 in non-pregnant human endometrium and at the decidual-trophoblast interface. J Reprod Immunol 39(1–2):133–148.

122. Lessey BA, Damjanovich L, Coutifaris C, Castelbaum A, Albelda SM & Buck CA (1992) Integrin adhesion molecules in the human endometrium. Correlation with the normal and abnormal menstrual cycle. J Clin Invest 90(1):188–195.

123. Suikkari AM. (2008) In-vitro maturation: its role in fertility treatment. Curr Opin Obstet Gynecol 20(3): 242–248.

124. Orvieto R, Meltzer S, Rabinson J, Zohav E, Anteby EY & Nahum R (2008) GnRH agonist versus GnRH antagonist in ovarian stimulation: the role of endometrial receptivity. Fertil Steril 4:4.

125. Noyes N, Hampton BS, Berkeley A, Licciardi F, Grifo J & Krey L (2001) Factors useful in predicting the success of oocyte donation: a 3-year retrospective analysis. Fertil Steril 76(1): 92–97.

126. Puerto B, Creus M, Carmona F, Civico S, Vanrell JA & Balasch J (2003) Ultrasonography as a predictor of embryo implantation after in vitro fertilization: a controlled study. Fertil Steril 79(4): 1015–1022.

127. Dastidar KG & Dastidar SG (2003) Dynamics of endometrial thickness over time: a reappraisal to standardize ultrasonographic measurements in an infertility program. Fertil Steril 80(1): 213–215.

128. Check JH, Dietterich C, Check ML & Katz Y (2003) Successful delivery despite conception with a maximal endometrial thickness of 4 mm. Clin Exp Obstet Gynecol 30(2–3): 93–94.

92

129. Sundström P (1998) Establishment of a successful pregnancy following in-vitro fertilization with an endometrial thickness of no more than 4 mm. Hum Reprod 13(6): 1550–1552.

130. Navot D, Bergh PA, Williams M, Garrisi GJ, Guzman I, Sandler B et al. (1991) An insight into early reproductive processes through the in vivo model of ovum donation. J Clin Endocrinol Metab 72(2): 408–414.

131. Remohi J, Gallardo E, Guanes PP, Simon C & Pellicer A (1995) Donor-recipient synchronization and the use of gonadotrophin-releasing hormone agonists to avoid the premature luteinizing hormone surge in oocyte donation. Hum Reprod 10(Suppl 2): 84–90.

132. Borini A, Dal Prato L, Bianchi L, Violini F, Cattoli M & Flamigni C Effect of duration of estradiol replacement on the outcome of oocyte donation. (2001) J Assist Reprod Genet 18(4): 185–190.

133. Soares SR, Troncoso C, Bosch E, Serra V, Simon C, Remohi J et al. (2005) Age and uterine receptiveness: predicting the outcome of oocyte donation cycles. J Clin Endocrinol Metab 90(7): 4399–4404. Epub 2005 Mar 29.

134. Reis Soares S, Rubio C, Rodrigo L, Simon C, Remohi J & Pellicer A (2003) High frequency of chromosomal abnormalities in embryos obtained from oocyte donation cycles. Fertil Steril 80(3): 656–657.

135. Nawroth F & Ludwig M (2005) What is the 'ideal' duration of progesterone supplementation before the transfer of cryopreserved-thawed embryos in estrogen/progesterone replacement protocols? Hum Reprod 20(5): 1127–1134. Epub 2005 Feb 3.

136. Toner JP, Grainger DA & Frazier LM (2002) Clinical outcomes among recipients of donated eggs: an analysis of the U.S. national experience, 1996–1998. Fertil Steril 78(5): 1038–1045.

137. Navot D, Bergh PA, Williams MA, Garrisi GJ, Guzman I & Sandler B et al. (1991) Poor oocyte quality rather than implantation failure as a cause of age-related decline in female fertility. Lancet 337(8754): 1375–1377.

138. Navot D, Drews MR, Bergh PA, Guzman I, Karstaedt A, Scott RT, Jr et al. (1994) Age-related decline in female fertility is not due to diminished capacity of the uterus to sustain embryo implantation. Fertil Steril 61(1): 97–101.

139. Ventura SJ, Taffel SM & Mosher WD (1988) Estimates of pregnancies and pregnancy rates for the United States, 1976–85. Am J Public Health 78(5): 506–511.

140. Guyer B, Hoyert DL, Martin JA, Ventura SJ, MacDorman MF & Strobino DM (1999) Annual summary of vital statistics – 1998. Pediatrics 104(6): 1229–1246.

141. Zuckerman S (1951) The number of oocytes in the mature ovary. Recent Prog Horm Res 6: 63–108.

142. Faddy MJ, Gosden RG, Gougeon A, Richardson SJ & Nelson JF (1992) Accelerated disappearance of ovarian follicles in mid-life: implications for forecasting menopause. Hum Reprod 7(10): 1342–1346.

143. Johnson J, Canning J, Kaneko T, Pru JK & Tilly JL (2004) Germline stem cells and follicular renewal in the postnatal mammalian ovary. Nature 428(6979): 145–150.

93

144. Johnson J, Bagley J, Skaznik-Wikiel M, Lee HJ, Adams GB, Niikura Y, et al. (2005) Oocyte generation in adult mammalian ovaries by putative germ cells in bone marrow and peripheral blood. Cell 122(2): 303–315.

145. Bukovsky A, Caudle MR, Svetlikova M & Upadhyaya NB (2004) Origin of germ cells and formation of new primary follicles in adult human ovaries. Reprod Biol Endocrinol 2: 20.

146. Bukovsky A, Svetlikova M & Caudle MR (2005) Oogenesis in cultures derived from adult human ovaries. Reprod Biol Endocrinol 3: 17.

147. Bukovsky A, Caudle MR, Svetlikova M, Wimalasena J, Ayala ME & Dominguez R (2005) Oogenesis in adult mammals, including humans: a review. Endocrine 26(3): 301–316.

148. Skaznik-Wikiel M, Tilly JC, Lee HJ, Niikura Y, Kaneko-Tarui T, Johnson J et al. (2007) Serious doubts over "Eggs forever?" Differentiation 75(2): 93–99.

149. Hassold TJ & Jacobs PA (1984) Trisomy in man. Annu Rev Genet 18: 69–97. 150. Sherman SL, Freeman SB, Allen EG & Lamb NE (2005) Risk factors for

nondisjunction of trisomy 21. Cytogenet Genome Res 111(3–4): 273–280. 151. Lim AS & Tsakok MF (1997) Age-related decline in fertility: a link to degenerative

oocytes? Fertil Steril 68(2): 265–271. 152. Pellestor F, Andreo B, Arnal F, Humeau C & Demaille J (2003) Maternal aging and

chromosomal abnormalities: new data drawn from in vitro unfertilized human oocytes. Hum Genet 112(2): 195–203. Epub 2002 Oct 29.

153. Munne S, Alikani M, Tomkin G, Grifo J & Cohen J (1995) Embryo morphology, developmental rates, and maternal age are correlated with chromosome abnormalities. Fertil Steril 64(2): 382–391.

154. Benadiva CA, Kligman I & Munne S (1996) Aneuploidy 16 in human embryos increases significantly with maternal age. Fertil Steril 66(2): 248–255.

155. Djahanbakhch O, Ezzati M & Zosmer A (2007) Reproductive ageing in women. J Pathol 211(2): 219–231.

156. Nicolaidis P & Petersen MB (1998) Origin and mechanisms of non-disjunction in human autosomal trisomies. Hum Reprod 13(2): 313–319.

157. Wolstenholme J & Angell RR (2000) Maternal age and trisomy — a unifying mechanism of formation. Chromosoma 109(7): 435–438.

158. Zhang X, Wu XQ, Lu S, Guo YL & Ma X (2006) Deficit of mitochondria-derived ATP during oxidative stress impairs mouse MII oocyte spindles. Cell Res 16(10): 841–850. Epub 2006 Sep 19.

159. Keefe DL, Marquard K & Liu L (2006) The telomere theory of reproductive senescence in women. Curr Opin Obstet Gynecol 18(3): 280–285.

160. Padilla SL & Garcia JE (1989) Effect of maternal age and number of in vitro fertilization procedures on pregnancy outcome. Fertil Steril 52(2): 270–273.

161. Templeton A, Morris JK & Parslow W (1996) Factors that affect outcome of in-vitro fertilisation treatment. Lancet 348(9039): 1402–1406.

94

162. Saftlas AF, Olson DR, Franks AL, Atrash HK, Pokras R (1990) Epidemiology of preeclampsia and eclampsia in the United States, 1979–1986. Am J Obstet Gynecol 163(2): 460–465.

163. Xiong X, Saunders LD, Wang FL & Demianczuk NN (2001) Gestational diabetes mellitus: prevalence, risk factors, maternal and infant outcomes. Int J Gynaecol Obstet 75(3): 221–228.

164. van Noord PA, Dubas JS, Dorland M, Boersma H & te Velde E (1997) Age at natural menopause in a population-based screening cohort: the role of menarche, fecundity, and lifestyle factors. Fertil Steril 68(1): 95–102.

165. Lass A, Ellenbogen A, Croucher C, Trew G, Margara R, Becattini C et al. (1998) Effect of salpingectomy on ovarian response to superovulation in an in vitro fertilization-embryo transfer program. Fertil Steril 70(6): 1035–1038.

166. Sonmezer Mv Oktay K (2004) Fertility preservation in female patients. Hum Reprod Update 10(3): 251–266.

167. Keay SD, Liversedge NH & Jenkins JM (1998) Could ovarian infection impair ovarian response to gonadotrophin stimulation? Br J Obstet Gynaecol 105(3): 252–253.

168. Sharara FI (1999) "Poor responders" to gonadotropins and levels of antibodies to Chlamydia trachomatis? Fertil Steril 71(2): 388–389.

169. Nikolaou D & Templeton A (2003) Early ovarian ageing: a hypothesis: Detection and clinical relevance. Hum Reprod 18(6): 1137–1139.

170. Broekmans FJ, Knauff EA, te Velde ER, Macklon NS & Fauser BC (2007) Female reproductive ageing: current knowledge and future trends. Trends Endocrinol Metab 18(2): 58–65. Epub 2007 Feb 1.

171. Keay SD, Liversedge NH, Mathur RS, Jenkins JM (1997) Assisted conception following poor ovarian response to gonadotrophin stimulation. Br J Obstet Gynaecol 104(5): 521–527.

172. Ferraretti AP, Gianaroli L, Magli MC, Bafaro G & Colacurci N (2000) Female poor responders. Mol Cell Endocrinol 161(1–2): 59–66.

173. Macklon NS, Stouffer RL, Giudice LC & Fauser BC (2006) The science behind 25 years of ovarian stimulation for in vitro fertilization. Endocr Rev 27(2): 170–207. Epub 2006 Jan 24.

174. Tarlatzis BC, Zepiridis L, Grimbizis G & Bontis J (2003) Clinical management of low ovarian response to stimulation for IVF: a systematic review. Hum Reprod Update 9(1): 61–76.

175. Kligman I & Rosenwaks Z (2001) Differentiating clinical profiles: predicting good responders, poor responders, and hyperresponders. Fertil Steril 76(6): 1185–1190.

176. Broekmans FJ, Kwee J, Hendriks DJ, Mol BW & Lambalk CB (2006) A systematic review of tests predicting ovarian reserve and IVF outcome. Hum Reprod Update 12(6): 685–718. Epub 2006 Aug 4.

177. Fisch JD, Keskintepe L & Sher G (2008) Gonadotropin-releasing hormone agonist/antagonist conversion with estrogen priming in low responders with prior in vitro fertilization failure. Fertil Steril 89(2): 342–347. Epub 2007 Jun 11.

95

178. Beckers NG, Macklon NS, Eijkemans MJ & Fauser BC (2002) Women with regular menstrual cycles and a poor response to ovarian hyperstimulation for in vitro fertilization exhibit follicular phase characteristics suggestive of ovarian aging. Fertil Steril 78(2): 291–297.

179. de Boer EJ, den Tonkelaar I, te Velde ER, Burger CW & van Leeuwen FE (2003) Increased risk of early menopausal transition and natural menopause after poor response at first IVF treatment. Hum Reprod 18(7): 1544–1552.

180. Check JH, Peymer M & Lurie D (1998) Effect of age on pregnancy outcome without assisted reproductive technology in women with elevated early follicular phase serum follicle-stimulating hormone levels. Gynecol Obstet Invest 45(4): 217–220.

181. Pellicer A, Gaitan P, Neuspiller F, Ardiles G, Albert C, Remohi J et al. (1998) Ovarian follicular dynamics: from basic science to clinical practice. J Reprod Immunol 39(1–2): 29–61.

182. Gianaroli L, Magli MC, Ferraretti AP, Fortini D, Tabanelli C & Gergolet M (2000) Gonadal activity and chromosomal constitution of in vitro generated embryos. Mol Cell Endocrinol 161(1–2): 111–116.

183. Land JA, Yarmolinskaya MI, Dumoulin JC & Evers JL (1996) High-dose human menopausal gonadotropin stimulation in poor responders does not improve in vitro fertilization outcome. Fertil Steril 65(5): 961–965.

184. Nasseri A, Mukherjee T, Grifo JA, Noyes N, Krey L & Copperman AB (1999) Elevated day 3 serum follicle stimulating hormone and/or estradiol may predict fetal aneuploidy. Fertil Steril 71(4): 715–718.

185. Levi AJ, Raynault MF, Bergh PA, Drews MR, Miller BT & Scott RT, Jr (2001) Reproductive outcome in patients with diminished ovarian reserve. Fertil Steril 76(4): 666–669.

186. Bancsi LF, Broekmans FJ, Eijkemans MJ, de Jong FH, Habbema JD & te Velde ER (2002) Predictors of poor ovarian response in in vitro fertilization: a prospective study comparing basal markers of ovarian reserve. Fertil Steril 77(2): 328–336.

187. El-Toukhy T, Khalaf Y, Hart R, Taylor A & Braude P (2002) Young age does not protect against the adverse effects of reduced ovarian reserve — an eight year study. Hum Reprod 17(6): 1519–1524.

188. Lawson R, El-Toukhy T, Kassab A, Taylor A, Braude P, Parsons J et al. (2003) Poor response to ovulation induction is a stronger predictor of early menopause than elevated basal FSH: a life table analysis. Hum Reprod 18(3): 527–533.

189. Biljan MM, Buckett WM, Dean N, Phillips SJ & Tan SL (2000) The outcome of IVF-embryo transfer treatment in patients who develop three follicles or less. Hum Reprod 15(10): 2140–2144.

190. Toner JP, Philput CB, Jones GS & Muasher SJ (1991) Basal follicle-stimulating hormone level is a better predictor of in vitro fertilization performance than age. Fertil Steril 55(4): 784–791.

191. Scott RT, Toner JP, Muasher SJ, Oehninger S, Robinson S & Rosenwaks Z (1989) Follicle-stimulating hormone levels on cycle day 3 are predictive of in vitro fertilization outcome. Fertil Steril 51(4): 651–654.

96

192. Tomàs C, Nuojua-Huttunen S & Martikainen H (1997) Pretreatment transvaginal ultrasound examination predicts ovarian responsiveness to gonadotrophins in in-vitro fertilization. Hum Reprod 12(2):220–223.

193. Ng EH, Tang OS & Ho PC (2000) The significance of the number of antral follicles prior to stimulation in predicting ovarian responses in an IVF programme. Hum Reprod 15(9): 1937–1942.

194. Popovic-Todorovic B, Loft A, Lindhard A, Bangsboll S, Andersson AM & Andersen AN (2003) A prospective study of predictive factors of ovarian response in 'standard' IVF/ICSI patients treated with recombinant FSH. A suggestion for a recombinant FSH dosage normogram. Hum Reprod 18(4): 781–787.

195. Scheffer GJ, Broekmans FJ, Looman CW, Blankenstein M, Fauser BC, teJong FH et al. (2003) The number of antral follicles in normal women with proven fertility is the best reflection of reproductive age. Hum Reprod 18(4): 700–706.

196. Muasher SJ, Oehninger S, Simonetti S, Matta J, Ellis LM, Liu HC et al. (1988) The value of basal and/or stimulated serum gonadotropin levels in prediction of stimulation response and in vitro fertilization outcome. Fertil Steril 50(2): 298–307.

197. Scott RT, Jr., Hofmann GE, Oehninger S & Muasher SJ (1990) Intercycle variability of day 3 follicle-stimulating hormone levels and its effect on stimulation quality in in vitro fertilization. Fertil Steril 54(2): 297–302.

198. Licciardi FL, Liu HC & Rosenwaks Z (1995) Day 3 estradiol serum concentrations as prognosticators of ovarian stimulation response and pregnancy outcome in patients undergoing in vitro fertilization. Fertil Steril 64(5): 991–994.

199. Mukherjee T, Copperman AB, Lapinski R, Sandler B, Bustillo M & Grunfeld L (1996) An elevated day three follicle-stimulating hormone:luteinizing hormone ratio (FSH:LH) in the presence of a normal day 3 FSH predicts a poor response to controlled ovarian hyperstimulation. Fertil Steril 65(3): 588–593.

200. Noci I, Biagiotti R, Maggi M, Ricci F, Cinotti A & Scarselli G (1998) Low day 3 luteinizing hormone values are predictive of reduced response to ovarian stimulation. Hum Reprod 13(3): 531–534.

201. Seifer DB, Lambert-Messerlian G, Hogan JW, Gardiner AC, Blazar AS & Berk CA (1997) Day 3 serum inhibin-B is predictive of assisted reproductive technologies outcome. Fertil Steril 67(1): 110–4.

202. Scheffer GJ, Broekmans FJ, Dorland M, Habbema JD, Looman CW & te Velde ER (1999) Antral follicle counts by transvaginal ultrasonography are related to age in women with proven natural fertility. Fertil Steril 72(5): 845–851.

203. Gougeon A (1996) Regulation of ovarian follicular development in primates: facts and hypotheses. Endocr Rev 17(2): 121–155.

204. Reuss ML, Kline J, Santos R, Levin B & Timor-Tritsch I (1996) Age and the ovarian follicle pool assessed with transvaginal ultrasonography. Am J Obstet Gynecol 174(2): 624–627.

205. Chang MY, Chiang CH, Hsieh TT, Soong YK & Hsu KH (1998) Use of the antral follicle count to predict the outcome of assisted reproductive technologies. Fertil Steril 69(3): 505–510.

97

206. Haadsma ML, Bukman A, Groen H, Roeloffzen EM, Groenewoud ER, Heineman MJ, et al. (2007) The number of small antral follicles (2–6 mm) determines the outcome of endocrine ovarian reserve tests in a subfertile population. Hum Reprod 22(7): 1925–1931. Epub 2007 Apr 16.

207. La Marca A & Volpe A (2006) Anti-Mullerian hormone (AMH) in female reproduction: is measurement of circulating AMH a useful tool? Clin Endocrinol (Oxf) 64(6): 603–610.

208. van Rooij IA, Broekmans FJ, te Velde ER, Fauser BC, Bancsi LF, de Jong FH, et al. (2002) Serum anti-Mullerian hormone levels: a novel measure of ovarian reserve. Hum Reprod 17(12): 3065–3071.

209. Hazout A, Bouchard P, Seifer DB, Aussage P, Junca AM & Cohen-Bacrie P (2004) Serum antimullerian hormone/mullerian-inhibiting substance appears to be a more discriminatory marker of assisted reproductive technology outcome than follicle-stimulating hormone, inhibin B, or estradiol. Fertil Steril 82(5): 1323–1329.

210. Jenkins JM, Davies DW, Devonport H, Anthony FW, Gadd SC, Watson RH et al. (1991) Comparison of 'poor' responders with 'good' responders using a standard buserelin/human menopausal gonadotrophin regime for in-vitro fertilization. Hum Reprod.;6(7): 918–921.

211. Sunkara SK, Tuthill J, Khairy M, El-Toukhy T, Coomarasamy A, Khalaf Y et al. (2007) Pituitary suppression regimens in poor responders undergoing IVF treatment: a systematic review and meta-analysis. Reprod Biomed Online 15(5): 539–546.

212. Sunkara SK, Coomarasamy A, Khalaf Y & Braude P (2007) A three arm randomised controlled trial comparing Gonadotrophin Releasing Hormone (GnRH) agonist long regimen versus GnRH agonist short regimen versus GnRH antagonist regimen in women with a history of poor ovarian response undergoing in vitro fertilisation (IVF) treatment: Poor responders intervention trial (PRINT). Reprod Health 4(1): 12.

213. ESHRE (2006) Nutrition and reproduction in women. Hum Reprod Update 12(3): 193–207. Epub 2006 Jan 31.

214. Schaffler A, Binart N, Scholmerich J & Buchler C (2005) Hypothesis paper Brain talks with fat — evidence for a hypothalamic-pituitary-adipose axis? Neuropeptides 39(4): 363–367.

215. Ahima RS & Flier JS (2000) Adipose tissue as an endocrine organ. Trends Endocrinol Metab 11(8): 327–332.

216. Lage M, Garcia-Mayor RV, Tome MA, Cordido F, Valle-Inclan F, Considine RV et al. (1999) Serum leptin levels in women throughout pregnancy and the postpartum period and in women suffering spontaneous abortion. Clin Endocrinol (Oxf) 50(2): 211–216.

217. Andrico S, Gambera A, Specchia C, Pellegrini C, Falsetti L, Sartori E (2002) Leptin in functional hypothalamic amenorrhoea. Hum Reprod 17(8): 2043–2048.

218. Wunder DM, Kretschmer R & Bersinger NA (2005) Concentrations of leptin and C-reactive protein in serum and follicular fluid during assisted reproductive cycles. Hum Reprod 20(5): 1266–1271. Epub 2005 Mar 3.

98

219. Budak E, Fernandez Sanchez M, Bellver J, Cervero A, Simon C & Pellicer A (2006) Interactions of the hormones leptin, ghrelin, adiponectin, resistin, and PYY3-36 with the reproductive system. Fertil Steril 85(6): 1563–1581.

220. Alfer J, Muller-Schöttle F, Classen-Linke I, von Rango U, Happel L, Beier-Hellwig K, et al. (2000) The endometrium as a novel target for leptin: differences in fertility and subfertility. Mol Hum Reprod 6(7): 595–601.

221. Kitawaki J, Koshiba H, Ishihara H, Kusuki I, Tsukamoto K & Honjo H (2000) Expression of leptin receptor in human endometrium and fluctuation during the menstrual cycle. J Clin Endocrinol Metab 85(5): 1946–1950.

222. Cervero A, Horcajadas JA, MartIn J, Pellicer A & Simon C (2004) The leptin system during human endometrial receptivity and preimplantation development. J Clin Endocrinol Metab 89(5): 2442–2451.

223. Bouloumie A, Drexler HC, Lafontan M & Busse R (1998) Leptin, the product of Ob gene, promotes angiogenesis. Circ Res 83(10): 1059–1066.

224. Steinberg GR, Parolin ML, Heigenhauser GJ & Dyck DJ (2002) Leptin increases FA oxidation in lean but not obese human skeletal muscle: evidence of peripheral leptin resistance. Am J Physiol Endocrinol Metab 283(1): E187–192.

225. Dyck DJ, Heigenhauser GJ & Bruce CR (2006) The role of adipokines as regulators of skeletal muscle fatty acid metabolism and insulin sensitivity. Acta Physiol (Oxf) 186(1): 5–16.

226. Loffreda S, Yang SQ, Lin HZ, Karp CL, Brengman ML, Wang DJ, et al. (1998) Leptin regulates proinflammatory immune responses. Faseb J 12(1): 57–65.

227. Bastard JP, Maachi M, Lagathu C, Kim MJ, Caron M, Vidal H, et al. (2006) Recent advances in the relationship between obesity, inflammation, and insulin resistance. Eur Cytokine Netw 17(1): 4–12.

228. Christiansen OB, Nielsen HS & Kolte AM (2006) Inflammation and miscarriage. Semin Fetal Neonatal Med 11(5): 302–308. Epub 2006 May 6.

229. Apparao KB, Lovely LP, Gui Y, Lininger RA & Lessey BA (2002) Elevated endometrial androgen receptor expression in women with polycystic ovarian syndrome. Biol Reprod 66(2): 297–304.

230. Jakubowicz DJ, Essah PA, Seppala M, Jakubowicz S, Baillargeon JP, Koistinen R et al. (2004) Reduced serum glycodelin and insulin-like growth factor-binding protein-1 in women with polycystic ovary syndrome during first trimester of pregnancy. J Clin Endocrinol Metab 89(2): 833–839.

231. Giudice LC (2006) Endometrium in PCOS: Implantation and predisposition to endocrine CA. Best Pract Res Clin Endocrinol Metab 20(2): 235–244.

232. Jakubowicz DJ, Seppala M, Jakubowicz S, Rodriguez-Armas O, Rivas-Santiago A, Koistinen H et al. (2001) Insulin reduction with metformin increases luteal phase serum glycodelin and insulin-like growth factor-binding protein 1 concentrations and enhances uterine vascularity and blood flow in the polycystic ovary syndrome. J Clin Endocrinol Metab 86(3): 1126–1133.

99

233. Kawamura K, Sato N, Fukuda J, Kodama H, Kumagai J, Tanikawa H et al. (2003) Ghrelin inhibits the development of mouse preimplantation embryos in vitro. Endocrinology 144(6): 2623–2633.

234. Mitchell M, Armstrong DT, Robker RL & Norman RJ (2005) Adipokines: implications for female fertility and obesity. Reproduction 130(5): 583–597.

235. Knutson KL, Spiegel K, Penev P & Van Cauter E (2007) The metabolic consequences of sleep deprivation. Sleep Med Rev 11(3): 163–178. Epub 2007 Apr 17.

236. Zaadstra BM, Seidell JC, Van Noord PA, te Velde ER, Habbema JD, Vrieswijk B, et al. (1993) Fat and female fecundity: prospective study of effect of body fat distribution on conception rates. BMJ 306(6876): 484–487.

237. Hall LF & Neubert AG (2005) Obesity and pregnancy. Obstet Gynecol Surv 60(4): 253–260.

238. Pasquali R (2006) Obesity, fat distribution and infertility. Maturitas 54(4): 363–371. Epub 2006 May 24.

239. Rich-Edwards JW, Goldman MB, Willett WC, Hunter DJ, Stampfer MJ, Colditz GA, et al. (1994) Adolescent body mass index and infertility caused by ovulatory disorder. Am J Obstet Gynecol 171(1): 171–177.

240. Lashen H, Fear K & Sturdee DW (2004) Obesity is associated with increased risk of first trimester and recurrent miscarriage: matched case-control study. Hum Reprod 19(7): 1644–1646. Epub 2004 May 13.

241. Lashen H, Ledger W, Bernal AL & Barlow D (1999) Extremes of body mass do not adversely affect the outcome of superovulation and in-vitro fertilization. Hum Reprod 14(3): 712–715.

242. Wang JX, Davies MJ & Norman RJ (2002) Obesity increases the risk of spontaneous abortion during infertility treatment. Obes Res 10(6): 551–554.

243. Clark AM, Ledger W, Galletly C, Tomlinson L, Blaney F, Wang X, et al. (1995) Weight loss results in significant improvement in pregnancy and ovulation rates in anovulatory obese women. Hum Reprod 10(10): 2705–2712.

244. Clark AM, Thornley B, Tomlinson L, Galletley C & Norman RJ (1998) Weight loss in obese infertile women results in improvement in reproductive outcome for all forms of fertility treatment. Hum Reprod 13(6): 1502–1505.

245. Franks PW, Loos RJ, Brage S, O'Rahilly S, Wareham NJ & Ekelund U (2007) Physical activity energy expenditure may mediate the relationship between plasma leptin levels and worsening insulin resistance independently of adiposity. J Appl Physiol 102(5): 1921–1926. Epub 2007 Jan 18.

246. Pasquali R, Pelusi C, Genghini S, Cacciari M & Gambineri A (2003) Obesity and reproductive disorders in women. Hum Reprod Update 9(4): 359–372.

247. Dechaud H, Anahory T, Reyftmann L, Loup V, Hamamah S & Hedon B (2006) Obesity does not adversely affect results in patients who are undergoing in vitro fertilization and embryo transfer. Eur J Obstet Gynecol Reprod Biol 127(1): 88–93. Epub 2006 Jan 18.

100

248. Fedorcsák P, Storeng R, Dale PO, Tanbo T & Abyholm T (2000) Obesity is a risk factor for early pregnancy loss after IVF or ICSI. Acta Obstet Gynecol Scand 79(1): 43–48.

249. Winter E, Wang J, Davies MJ & Norman R (2002) Early pregnancy loss following assisted reproductive technology treatment. Hum Reprod 17(12): 3220–3223.

250. Metwally M, Cutting R, Tipton A, Skull J, Ledger WL & Li TC (2007) Effect of increased body mass index on oocyte and embryo quality in IVF patients. Reprod Biomed Online 15(5): 532–538.

251. Bellver J, Rossal LP, Bosch E, Zuniga A, Corona JT, Melendez F et al. (2003) Obesity and the risk of spontaneous abortion after oocyte donation. Fertil Steril 79(5): 1136–1140.

252. Bellver J, Melo MA, Bosch E, Serra V, Remohi J & Pellicer A (2007) Obesity and poor reproductive outcome: the potential role of the endometrium. Fertil Steril 88(2): 446–451.

253. Wattanakumtornkul S, Damario MA, Stevens Hall SA, Thornhill AR & Tummon IS (2003) Body mass index and uterine receptivity in the oocyte donation model. Fertil Steril 80(2): 336–340.

254. Styne-Gross A, Elkind-Hirsch K & Scott RT, Jr (2005) Obesity does not impact implantation rates or pregnancy outcome in women attempting conception through oocyte donation. Fertil Steril 83(6): 1629–1634.

255. Roth D, Grazi RV & Lobel SM (2003) Extremes of body mass index do not affect first-trimester pregnancy outcome in patients with infertility. Am J Obstet Gynecol 188(5): 1169–1170.

256. Wang JX, Davies MJ & Norman RJ (2001) Polycystic ovarian syndrome and the risk of spontaneous abortion following assisted reproductive technology treatment. Hum Reprod 16(12): 2606–2609.

257. Frisch RE (1987) Body fat, menarche, fitness and fertility. Hum Reprod 2(6): 521–533.

258. Frisch RE (1989) Body weight and reproduction. Science 246(4929): 432. 259. Frisch RE (1996) The right weight: body fat, menarche, and fertility. Nutrition 12(6):

452–453. 260. Scott EC & Johnston FE (1982) Critical fat, menarche, and the maintenance of

menstrual cycles: a critical review. J Adolesc Health Care 2(4): 249–260. 261. Frisch RE (1978) Population, food intake, and fertility. There is historical evidence for

a direct effect of nutrition on reproductive ability. Science 199(4324): 22–30. 262. Schweiger U, Laessle R, Pfister H, Hoehl C, Schwingenschloegel M, Schweiger M et

al. (1987) Diet-induced menstrual irregularities: effects of age and weight loss. Fertil Steril 48(5): 746–751.

263. Helgstrand S & Andersen AM (2005) Maternal underweight and the risk of spontaneous abortion. Acta Obstet Gynecol Scand 84(12): 1197–1201.

264. Maconochie N, Doyle P, Prior S & Simmons R (2007) Risk factors for first trimester miscarriage — results from a UK-population-based case-control study. BJOG 114(2): 170–186.

101

265. Doherty DA, Magann EF, Francis J, Morrison JC & Newnham JP (2006) Pre-pregnancy body mass index and pregnancy outcomes. Int J Gynaecol Obstet 95(3): 242–247. Epub 2006 Sep 27.

266. Morgan JF, Lacey JH & Chung E (2006) Risk of postnatal depression, miscarriage, and preterm birth in bulimia nervosa: retrospective controlled study. Psychosom Med 68(3): 487–492.

267. Laird SM, Quinton ND, Anstie B, Li TC & Blakemore AI (2001) Leptin and leptin-binding activity in women with recurrent miscarriage: correlation with pregnancy outcome. Hum Reprod 16(9): 2008–2013.

268. Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome (PCOS) (2004) Hum Reprod 19(1): 41–47.

269. Teede HJ, Hutchison SK & Zoungas S (2007) The management of insulin resistance in polycystic ovary syndrome. Trends Endocrinol Metab 18(7): 273–279. Epub 2007 Aug 16.

270. Heijnen EM, Eijkemans MJ, Hughes EG, Laven JS, Macklon NS & Fauser BC (2006) A meta-analysis of outcomes of conventional IVF in women with polycystic ovary syndrome. Hum Reprod Update 12(1): 13–21. Epub 2005 Aug 25.

271. Boomsma CM, Fauser BC & Macklon NS (2008) Pregnancy complications in women with polycystic ovary syndrome. Semin Reprod Med 26(1): 72–84.

272. Pasquali R, Antenucci D, Casimirri F, Venturoli S, Paradisi R, Fabbri R et al. (1989) Clinical and hormonal characteristics of obese amenorrheic hyperandrogenic women before and after weight loss. J Clin Endocrinol Metab 68(1): 173–179.

273. Morin-Papunen LC, Koivunen RM, Ruokonen A & Martikainen HK (1998) Metformin therapy improves the menstrual pattern with minimal endocrine and metabolic effects in women with polycystic ovary syndrome. Fertil Steril 69(4): 691–696.

274. Morin-Papunen L, Rautio K, Ruokonen A, Hedberg P, Puukka M & Tapanainen JS (2003) Metformin reduces serum C-reactive protein levels in women with polycystic ovary syndrome. J Clin Endocrinol Metab 88(10): 4649–4654.

275. Rautio K, Tapanainen JS, Ruokonen A & Morin-Papunen LC (2005) Effects of metformin and ethinyl estradiol-cyproterone acetate on lipid levels in obese and non-obese women with polycystic ovary syndrome. Eur J Endocrinol 152(2): 269–275.

276. Jakubowicz DJ, Iuorno MJ, Jakubowicz S, Roberts KA & Nestler JE (2002) Effects of metformin on early pregnancy loss in the polycystic ovary syndrome. J Clin Endocrinol Metab 87(2): 524–529.

277. Legro RS, Barnhart HX, Schlaff WD, Carr BR, Diamond MP, Carson SA et al. (2007) Clomiphene, metformin, or both for infertility in the polycystic ovary syndrome. N Engl J Med 356(6): 551–566.

278. Tummon I, Gavrilova-Jordan L, Allemand MC & Session D (2005) Polycystic ovaries and ovarian hyperstimulation syndrome: a systematic review*. Acta Obstet Gynecol Scand 84(7): 611–616.

102

279. Ragni G, Vegetti W, Riccaboni A, Engl B, Brigante C & Crosignani PG (2005) Comparison of GnRH agonists and antagonists in assisted reproduction cycles of patients at high risk of ovarian hyperstimulation syndrome. Hum Reprod 20(9): 2421–2425. Epub 005 May 12.

280. Homburg R (2006) Pregnancy complications in PCOS. Best Pract Res Clin Endocrinol Metab 20(2): 281–292.

281. Tian L, Shen H, Lu Q, Norman RJ & Wang J (2007) Insulin Resistance Increases the Risk of Spontaneous Abortion Following Assisted Reproduction Technology Treatment. J Clin Endocrinol Metab 92(4): 1430–1433.

282. Lintsen AM, Pasker-de Jong PC, de Boer EJ, Burger CW, Jansen CA, Braat DD et al. (2005) Effects of subfertility cause, smoking and body weight on the success rate of IVF. Hum Reprod 20(7): 1867–1875. Epub 2005 Apr 7.

283. Kuivasaari P, Hippelainen M, Anttila M & Heinonen S (2005) Effect of endometriosis on IVF/ICSI outcome: stage III/IV endometriosis worsens cumulative pregnancy and live-born rates. Hum Reprod 20(11): 3130–3135. Epub 2005 Jul 8.

284. Schieve LA, Tatham L, Peterson HB, Toner J & Jeng G (2003) Spontaneous abortion among pregnancies conceived using assisted reproductive technology in the United States. Obstet Gynecol 101(5 Pt 1): 959–967.

285. Tiitinen A, Husa LM, Tulppala M, Simberg N & Seppala M (1995) The effect of cryopreservation in prevention of ovarian hyperstimulation syndrome. Br J Obstet Gynaecol 102(4): 326–329.

286. Wang JX, Yap YY & Matthews CD (2001) Frozen-thawed embryo transfer: influence of clinical factors on implantation rate and risk of multiple conception. Hum Reprod 16(11): 2316–2319.

287. Van der Elst J, Van den Abbeel E, Camus M, Smitz J, Devroey P & Van Steirteghem A (1996) Long-term evaluation of implantation of fresh and cryopreserved human embryos following ovarian stimulation with buserelin acetate-human menopausal gonadotrophin (HMG) or clomiphene citrate-HMG. Hum Reprod 11(10): 2097–2106.

288. Van der Elst J, Camus M, Van den Abbeel E, Maes R, Devroey P & Van Steirteghem AC (1995) Prospective randomized study on the cryopreservation of human embryos with dimethylsulfoxide or 1,2-propanediol protocols. Fertil Steril 63(1): 92–100.

289. Gelbaya TA, Nardo LG, Hunter HR, Fitzgerald CT, Horne G, Pease EE et al. (2006) Cryopreserved-thawed embryo transfer in natural or down-regulated hormonally controlled cycles: a retrospective study. Fertil Steril 85(3): 603–609.

290. Hydén-Granskog C, Unkila-Kallio L, Halttunen M & Tiitinen A (2005) Single embryo transfer is an option in frozen embryo transfer. Hum Reprod 20(10): 2935–2938.

291. Matorras R, Matorras F, Mendoza R, Rodriguez M, Remohi J, Rodriguez-Escudero FJ et al. (2005) The implantation of every embryo facilitates the chances of the remaining embryos to implant in an IVF programme: a mathematical model to predict pregnancy and multiple pregnancy rates. Hum Reprod 20(10): 2923–2931. Epub 2005 Jul 21.

103

292. La Sala GB, Nicoli A, Villani MT, Gallinelli A, Nucera G & Blickstein I (2005) Spontaneous embryonic loss rates in twin and singleton pregnancies after transfer of top- versus intermediate-quality embryos. Fertil Steril 84(6): 1602–1605.

293. La Sala GB, Nucera G, Gallinelli A, Nicoli A, Villani MT & Blickstein I (2005) Lower embryonic loss rates among twin gestations following assisted reproduction. J Assist Reprod Genet 22(5): 181–184.

294. Glujovsky D, Shamonki MI & Bergh PA (2007) Embryonic synergism may reduce pregnancy loss: a multivariate regression analysis. Fertil Steril 87(3): 509–514. Epub 2006 Nov 29.

295. De Sutter P, Delbaere I, Gerris J, Verstraelen H, Goetgeluk S, Van der Elst J et al. (2006) Birthweight of singletons after assisted reproduction is higher after single- than after double-embryo transfer. Hum Reprod 21(10): 2633–2637. Epub 2006 Jun 19.

296. Pinborg A, Lidegaard O, la Cour Freiesleben N & Andersen AN (2005) Consequences of vanishing twins in IVF/ICSI pregnancies. Hum Reprod 20(10): 2821–2829. Epub 2005 Jun 24.

297. La Sala GB, Nucera G, Gallinelli A, Nicoli A, Villani MT & Blickstein I (2004) Spontaneous embryonic loss following in vitro fertilization: incidence and effect on outcomes. Am J Obstet Gynecol 191(3): 741–746.

298. Dickey RP, Taylor SN, Lu PY, Sartor BM, Storment JM, Rye PH et al. (2002) Spontaneous reduction of multiple pregnancy: incidence and effect on outcome. Am J Obstet Gynecol 186(1): 77–83.

299. van Montfoort AP, Fiddelers AA, Janssen JM, Derhaag JG, Dirksen CD, Dunselman GA et al. (2006) In unselected patients, elective single embryo transfer prevents all multiples, but results in significantly lower pregnancy rates compared with double embryo transfer: a randomized controlled trial. Hum Reprod 21(2): 338–343. Epub 2005 Oct 27.

300. van Montfoort AP, Dumoulin JC, Land JA, Coonen E, Derhaag JG & Evers JL (2005) Elective single embryo transfer (eSET) policy in the first three IVF/ICSI treatment cycles. Hum Reprod 20(2): 433–436. Epub 2004 Nov 18.

301. Gordts S, Campo R, Puttemans P, Brosens I, Valkenburg M, Norre J et al. (2005) Belgian legislation and the effect of elective single embryo transfer on IVF outcome. Reprod Biomed Online 10(4): 436–441.

302. Van Landuyt L, Verheyen G, Tournaye H, Camus M, Devroey P & Van Steirteghem A (2006) New Belgian embryo transfer policy leads to sharp decrease in multiple pregnancy rate. Reprod Biomed Online 13(6): 765–771.

303. Reynolds MA, Schieve LA, Jeng G & Peterson HB (2003) Does insurance coverage decrease the risk for multiple births associated with assisted reproductive technology? Fertil Steril 80(1): 16–23.

304. Jain T, Harlow BL & Hornstein MD (2002) Insurance coverage and outcomes of in vitro fertilization. N Engl J Med 347(9): 661–666.

305. van Wely M, Twisk M, Mol BW & van der Veen F (2006) Is twin pregnancy necessarily an adverse outcome of assisted reproductive technologies? Hum Reprod 21(11): 2736–2738. Epub 006 Jun 22.

104

306. Kalra SK, Milad MP, Klock SC & Grobman WA (2003) Infertility patients and their partners: differences in the desire for twin gestations. Obstet Gynecol 102(1): 152–155.

307. Gleicher N, Campbell DP, Chan CL, Karande V, Rao R, Balin M et al. (1995) The desire for multiple births in couples with infertility problems contradicts present practice patterns. Hum Reprod 10(5): 1079–10784.

308. Ryan GL, Zhang SH, Dokras A, Syrop CH & Van Voorhis BJ (2004) The desire of infertile patients for multiple births. Fertil Steril 81(3): 500–504.

309. Pinborg A, Loft A, Schmidt L & Andersen AN (2003) Attitudes of IVF/ICSI-twin mothers towards twins and single embryo transfer. Hum Reprod 18(3): 621–627.

310. Child TJ, Henderson AM & Tan SL (2004) The desire for multiple pregnancy in male and female infertility patients. Hum Reprod 19(3): 558–561. Epub 2004 Jan 29.

311. Murray S, Shetty A, Rattray A, Taylor V & Bhattacharya S (2004) A randomized comparison of alternative methods of information provision on the acceptability of elective single embryo transfer. Hum Reprod.;19(4): 911–916. Epub 2004 Feb 27.

312. Blennborn M, Nilsson S, Hillervik C & Hellberg D (2005) The couple's decision-making in IVF: one or two embryos at transfer? Hum Reprod 20(5): 1292–1297. Epub 2005 Feb 25.

313. Hojgaard A, Ottosen LD, Kesmodel U & Ingerslev HJ (2007) Patient attitudes towards twin pregnancies and single embryo transfer — a questionnaire study. Hum Reprod 22(10): 2673–2678. Epub 2007 Sep 1.

314. Haynes B (1999) Can it work? Does it work? Is it worth it? The testing of healthcareinterventions is evolving. BMJ 319(7211): 652–653.

315. Drummond MF, Sculpher MJ, Torrance GW, O'Brien BJ & Stoddart GL (2005) Methods for the economic evaluation of health care programmes. Oxford University Press.

316. Garceau L, Henderson J, Davis LJ, Petrou S, Henderson LR, McVeigh E et al. (2002) Economic implications of assisted reproductive techniques: a systematic review. Hum Reprod 17(12): 3090–3109.

317. Johannesson M (1995) Quality-adjusted life-years versus healthy-years equivalents. J Health Econ 14(1): 9–16.

318. Weinstein MC, Siegel JE, Gold MR, Kamlet MS & Russell LB (1996) Recommendations of the Panel on Cost-effectiveness in Health and Medicine. JAMA 276(15): 1253–1258.

319. Boivin J, Bunting L, Collins JA & Nygren KG (2007) International estimates of infertility prevalence and treatment-seeking: potential need and demand for infertility medical care. Hum Reprod 22(10): 2800.

320. Pinborg A, Loft A, Rasmussen S & Nyboe Andersen A (2004) Hospital care utilization of IVF/ICSI twins followed until 2–7 years of age: a controlled Danish national cohort study. Hum Reprod 19(11): 2529–2536. Epub 2004 Aug 19.

321. Koivurova S, Hartikainen AL, Gissler M, Hemminki E, Klemetti R & Jarvelin MR (2004) Health care costs resulting from IVF: prenatal and neonatal periods. Hum Reprod 19(12): 2798–2805. Epub 2004 Sep 23.

105

322. Ledger WL, Anumba D, Marlow N, Thomas CM & Wilson EC (2006) The costs to the NHS of multiple births after IVF treatment in the UK. BJOG 113(1): 21–25.

323. Scotland GS, McNamee P & Bhattacharya S (2007) Is elective single embryo transfer a cost-effective alternative to double embryo transfer? BJOG 114(1): 5–7. Epub 2006 Nov 2.

324. Wølner-Hanssen P & Rydhstroem H (1998) Cost-effectiveness analysis of in-vitro fertilization: estimated costs per successful pregnancy after transfer of one or two embryos. Hum Reprod 13(1): 88–94.

325. De Sutter P, Gerris J & Dhont M (2002) A health-economic decision-analytic model comparing double with single embryo transfer in IVF/ICSI. Hum Reprod 17(11): 2891–2896.

326. Thurin Kjellberg A, Carlsson P & Bergh C (2006) Randomized single versus double embryo transfer: obstetric and paediatric outcome and a cost-effectiveness analysis. Hum Reprod 21(1): 210–216. Epub 2005 Sep 19.

327. Fiddelers AA, van Montfoort AP, Dirksen CD, Dumoulin JC, Land JA, Dunselman GA et al. (2006) Single versus double embryo transfer: cost-effectiveness analysis alongside a randomized clinical trial. Hum Reprod 21(8): 2090–2097. Epub 2006 Apr 13.

328. Fiddelers AA, Severens JL, Dirksen CD, Dumoulin JC, Land JA & Evers JL (2007) Economic evaluations of single- versus double-embryo transfer in IVF. Hum Reprod Update 13(1): 5–13. Epub 2006 Nov 10.

329. Gardner DK, Surrey E, Minjarez D, Leitz A, Stevens J & Schoolcraft WB (2004) Single blastocyst transfer: a prospective randomized trial. Fertil Steril 81(3): 551–555.

330. Styer AK, Wright DL, Wolkovich AM, Veiga C, Toth TL (2008) Single-blastocyst transfer decreases twin gestation without affecting pregnancy outcome. Fertil Steril 89(6): 1702–1708.

331. Henman M, Catt JW, Wood T, Bowman MC, de Boer KA & Jansen RP (2005) Elective transfer of single fresh blastocysts and later transfer of cryostored blastocysts reduces the twin pregnancy rate and can improve the in vitro fertilization live birth rate in younger women. Fertil Steril 84(6): 1620–1627.

332. Ryan GL, Sparks AE, Sipe CS, Syrop CH, Dokras A & Van Voorhis BJ (2007) A mandatory single blastocyst transfer policy with educational campaign in a United States IVF program reduces multiple gestation rates without sacrificing pregnancy rates. Fertil Steril 88(2): 354–360. Epub 2007 May 9.

333. Khalaf Y, El-Toukhy T, Coomarasamy A, Kamal A, Bolton V & Braude P (2008) Selective single blastocyst transfer reduces the multiple pregnancy rate and increases pregnancy rates: a pre- and postintervention study. BJOG 115(3): 385–390.

334. Emiliani S, Delbaere A, Vannin AS, Biramane J, Verdoodt M, Englert Y et al. (2003) Similar delivery rates in a selected group of patients, for day 2 and day 5 embryos both cultured in sequential medium: a randomized study. Hum Reprod 18(10): 2145–2150.

335. Fauser BC, Devroey P & Macklon NS (2005) Multiple birth resulting from ovarian stimulation for subfertility treatment. Lancet 365(9473): 1807–1816.

106

336. Hohmann FP, Macklon NS & Fauser BC (2003) A randomized comparison of two ovarian stimulation protocols with gonadotropin-releasing hormone (GnRH) antagonist cotreatment for in vitro fertilization commencing recombinant follicle-stimulating hormone on cycle day 2 or 5 with the standard long GnRH agonist protocol. J Clin Endocrinol Metab 88(1): 166–173.

337. Baart EB, Martini E, Eijkemans MJ, Van Opstal D, Beckers NG, Verhoeff A et al. (2007) Milder ovarian stimulation for in-vitro fertilization reduces aneuploidy in the human preimplantation embryo: a randomized controlled trial. Hum Reprod 22(4): 980–988. Epub 2007 Jan 4.

338. Verberg MF, Eijkemans MJ, Macklon NS, Heijnen EM, Fauser BC & Broekmans FJ (2007) Predictors of low response to mild ovarian stimulation initiated on cycle day 5 for IVF. Hum Reprod 22(7): 1919–1924. Epub 2007 May 7.

339. Tomás C, Orava M, Tuomivaara L & Martikainen H (1998) Low pregnancy rate is achieved in patients treated with intracytoplasmic sperm injection due to previous low or failed fertilization in in-vitro fertilization. Hum Reprod 13(1): 65–70.

340. Campbell PJ & Gerich JE (1990) Impact of obesity on insulin action in volunteers with normal glucose tolerance: demonstration of a threshold for the adverse effect of obesity. J Clin Endocrinol Metab 70(4): 1114–1118.

341. Pellicer A, Ballester MJ, Serrano MD, Mir A, Serra-Serra V, Remohi J et al. (1994) Aetiological factors involved in the low response to gonadotrophins in infertile women with normal basal serum follicle stimulating hormone levels. Hum Reprod 9(5): 806–811.

342. Klinkert ER, Broekmans FJ, Looman CW & Te Velde ER (2004) A poor response in the first in vitro fertilization cycle is not necessarily related to a poor prognosis in subsequent cycles. Fertil Steril 81(5): 1247–1253.

343. van Hooff MH, Alberda AT, Huisman GJ, Zeilmaker GH & Leerentveld RA (1993) Doubling the human menopausal gonadotrophin dose in the course of an in-vitro fertilization treatment cycle in low responders: a randomized study. Hum Reprod 8(3): 369–373.

344. Kailasam C, Keay SD, Wilson P, Ford WC & Jenkins JM (2004) Defining poor ovarian response during IVF cycles, in women aged <40 years, and its relationship with treatment outcome. Hum Reprod 19(7): 1544–1547.

345. Karande VC, Jones GS, Veeck LL & Muasher SJ (1990) High-dose follicle-stimulating hormone stimulation at the onset of the menstrual cycle does not improve the in vitro fertilization outcome in low-responder patients. Fertil Steril 53(3): 486–489.

346. De Sutter P & Dhont M (2003) Poor response after hormonal stimulation for in vitro fertilization is not related to ovarian aging. Fertil Steril 79(6): 1294–1298.

347. Nybo Andersen AM, Wohlfahrt J, Christens P, Olsen J & Melbye M (2000) Maternal age and fetal loss: population based register linkage study. BMJ 320(7251): 1708–1712.

348. Enriori PJ, Evans AE, Sinnayah P & Cowley MA (2006) Leptin resistance and obesity. Obesity (Silver Spring) 14(Suppl 5): 254S–8S.

107

349. Salim R, Miel J, Savvas M, Lee C & Jurkovic D (2007) A comparative study of glycodelin concentrations in uterine flushings in women with subseptate uteri, history of unexplained recurrent miscarriage and healthy controls. Eur J Obstet Gynecol Reprod Biol 133(1): 76–80. Epub 2006 Oct 17.

350. Schmidt CL, de Ziegler D, Gagliardi CL, Mellon RW, Taney FH, Kuhar MJ et al. (1989) Transfer of cryopreserved-thawed embryos: the natural cycle versus controlled preparation of the endometrium with gonadotropin-releasing hormone agonist and exogenous estradiol and progesterone (GEEP). Fertil Steril 52(4): 609–616.

351. Sathanandan M, Macnamee MC, Rainsbury P, Wick K, Brinsden P & Edwards RG (1991) Replacement of frozen-thawed embryos in artificial and natural cycles: a prospective semi-randomized study. Hum Reprod 6(5): 685–687.

352. Al-Shawaf T, Yang D, al-Magid Y, Seaton A, Iketubosin F & Craft I (1993) Ultrasonic monitoring during replacement of frozen/thawed embryos in natural and hormone replacement cycles. Hum Reprod 8(12): 2068–2074.

353. Seppälä M, Martikainen H, Ronnberg L, Riittinen L & Kauppila A (1989) Suppression of prolactin secretion during ovarian hyperstimulation is followed by elevated serum levels of endometrial protein PP14 in the late luteal phase. Hum Reprod 4(4):389–391.

354. Tiitinen A, Unkila-Kallio L, Halttunen M & Hyden-Granskog C (2003) Impact of elective single embryo transfer on the twin pregnancy rate. Hum Reprod 18(7): 1449–1453.

355. Engmann L, Maconochie N, Tan SL & Bekir J (2001) Trends in the incidence of births and multiple births and the factors that determine the probability of multiple birth after IVF treatment. Hum Reprod 16(12): 2598–2605.

356. Ragni G, Allegra A, Anserini P, Causio F, Ferraretti AP, Greco E et al. (2005) The 2004 Italian legislation regulating assisted reproduction technology: a multicentre survey on the results of IVF cycles. Hum Reprod 20(8): 2224–2228. Epub 2005 Apr 7.

357. Tiitinen A (2004) Toward single embryo transfer policy. Acta Obstet Gynecol Scand 83(11): 1009–1011.

358. Wølner-Hanssen P & Rydhstroem H (1998) Cost-effectiveness analysis of in-vitro fertilization: estimated costs per successful pregnancy after transfer of one or two embryos. Hum Reprod 13(1): 88–94.

359. Dahl K, Kesmodel U, Hvidman L & Olesen F (2006) Informed consent: providing information about prenatal examinations. Acta Obstet Gynecol Scand 85: 1420–1425.

360. Wright VC, Chang J, Jeng G & Macaluso M (2006) Assisted reproductive technology surveillance - United States, 2003. MMWR Surveill Summ 55(4): 1–22.

108

109

Original papers

I Veleva Z, Järvelä IY, Nuojua-Huttunen S, Martikainen H & Tapanainen JS (2005) An initial low response predicts poor outcome in in vitro fertilization/intracytoplasmic sperm injection despite improved ovarian response in consecutive cycles. Fertililty and Sterilily 83: 1384–1390.

II Veleva Z, Tiitinen A, Vilska S, Hydén-Granskog C, Tomás C, Martikainen H & Tapanainen JS (2008) High and low BMI increase the risk of miscarriage after IVF/ICSI and FET. Human Reproduction 23: 878–884. Epub 2008 Feb 15.

III Veleva Z, Vilska S, Hyden-Granskog C, Tiitinen A, Tapanainen JS & Martikainen H (2006) Elective single embryo transfer in women aged 36-39 years. Human Reproduction 21: 2098–2102. Epub 2006 Jun 1.

IV Veleva Z, Tomàs C, Tapanainen JS and Martikainen H (2008) Elective single-embryo transfer with cryopreservation improves the outcome and diminishes the costs of IVF/ICSI. Manuscript.

The published papers reprinted with kind permissions from Oxford University Press (I and III) and Elsevier Limited (II).

Original publications are not included in the electronic version of the dissertation.

110

A C T A U N I V E R S I T A T I S O U L U E N S I S

Distributed byOULU UNIVERSITY LIBRARY

P.O. Box 7500, FI-90014University of Oulu, Finland

Book orders:OULU UNIVERSITY PRESSP.O. Box 8200, FI-90014University of Oulu, Finland

S E R I E S D M E D I C A

967. Huilaja, Laura (2008) Collagen XVII and pathomechanisms of junctionalepidermolysis bullosa and gestational pemphigoid

968. Heikka, Helena (2008) Sosiaali- ja terveysjohtajan työn sisältö ja kompetenssit

969. Juntti, Hanna (2008) Association of respiratory syncytial virus infection withasthma and atopic allergy

970. Winqvist, Satu (2008) Alcohol misuse in relation to traumatic brain injury. TheNorthern Finland 1966 birth cohort study

971. Makkonen, Saara (2008) Teknillisestä apulaisesta laboratoriohoitajaksi. Turunlaboratoriohoitajakoulutuksen kehitys vuosina 1955–1990

972. Liljeroos, Mari (2008) Toll-like receptor 2 (TLR2) and TLR4 signaling in the innateresponse against bacterial components

973. Vierimaa, Outi (2008) Multiple Endocrine Neoplasia Type 1 (MEN1) and PituitaryAdenoma Predisposition (PAP) in Northern Finland

974. Muhonen, Virpi (2008) Bone–Biomaterial Interface. The effects of surfacemodified NiTi shape memory alloy on bone cells and tissue

975. Oikarinen, Arja (2008) Kainuulaisten miesten terveyskäyttäytyminen —kulttuurinen näkökulma

976. Kantojärvi, Liisa (2008) Personality disorders in the Northern Finland 1966 BirthCohort Study

977. Valkealahti, Maarit (2008) The effects of bisphosphonates and COX-2 inhibitorson the bone remodelling unit

978. Hiiri, Anne (2008) Community-wide Oral Health Promotion in the PitkärantaDistrict of Russian Karelia – a case study

979. Koivunen, Kirsi (2008) Alaraajojen valtimonkovettumistautia sairastavienterveyteen liittyvä elämänlaatu sekä hoitomenetelmien kustannukset

980. Palosaari, Kari (2008) Quantitative and semiquantitative imaging techniques indetecting joint inflammation in patients with rheumatoid arthritis. Phase-shiftwater-fat MRI method for fat suppression at 0.23 T, contrast-enhanced dynamicand static MRI, and quantitative 99mTc-nanocolloid scintigraphy

981. Perkiömäki, Marja Riitta (2008) Craniofacial shape and dimensions as indicators oforofacial clefting and palatal form. A study on cleft lip and palate and Turnersyndrome families

ABCDEFG

UNIVERS ITY OF OULU P.O.B . 7500 F I -90014 UNIVERS ITY OF OULU F INLAND

A C T A U N I V E R S I T A T I S O U L U E N S I S

S E R I E S E D I T O R S

SCIENTIAE RERUM NATURALIUM

HUMANIORA

TECHNICA

MEDICA

SCIENTIAE RERUM SOCIALIUM

SCRIPTA ACADEMICA

OECONOMICA

EDITOR IN CHIEF

PUBLICATIONS EDITOR

Senior Assistant Jorma Arhippainen

University Lecturer Elise Kärkkäinen

Professor Hannu Heusala

Professor Olli Vuolteenaho

Senior Researcher Eila Estola

Information officer Tiina Pistokoski

University Lecturer Seppo Eriksson

Professor Olli Vuolteenaho

Publications Editor Kirsti Nurkkala

ISBN 978-951-42-8882-1 (Paperback)ISBN 978-951-42-8883-8 (PDF)ISSN 0355-3221 (Print)ISSN 1796-2234 (Online)

U N I V E R S I TAT I S O U L U E N S I S

MEDICA

ACTAD

D 985

ACTA

Zdravka Veleva

OULU 2008

D 985

Zdravka Veleva

FACTORS AFFECTINGTHE OUTCOME OF IVF/ICSI

FACULTY OF MEDICINE INSTITUTE OF CLINICAL MEDICINE, DEPARTMENT OF OBSTETRICS AND GYNECOLOGY,UNIVERSITY OF OULU