ART in Europe, 2019: results generated from European registries by ESHRE

1This is the 23rd annual report of the European IVF-Monitoring (EIM) Consortium under the umbrella of ESHRE, assembling the data on ART, IUI, and fertility preservation (FP) reported by 40 participating European countries in 2019 (Supplementary Table S1 and Supplementary Data File S1).

2Eighteen previous annual reports published in Human Reproduction (https://www.eshre.eu/Data-collection-and-research/Consortia/EIM/Publications.aspx) and four in Human Reproduction Open, covered data on treatment cycles collected yearly from 1997 to 2018. As in previous reports, the manuscript contains the five most relevant tables. Twenty additional supplementary tables (Supplementary Tables S1, S2, S3, S4, S5, S6, S7, S8, S9, S10, S11, S12, S13, S14, S15, S16, S17, S18, S19, and S20) are available online on the publisher’s homepage. To allow easy comparison and assessment of trends, the presentation of the data is consistent with previous reports. For the fourth consecutive year, data on FP were collected and added to this report.

3Data were collected on an aggregate basis and were provided by 40 European countries, covering treatments with IVF, ICSI, frozen embryo transfer (FET), egg donation (ED), IVM, preimplantation genetic testing (PGT; pooled data), frozen oocyte replacement (FOR), IUI with husband’s/partner’s semen (IUI-H), and with donor semen (IUI-D). The report includes treatments that started between 1 January and 31 December in 2019. Data on pregnancies and deliveries represent the outcomes of treatments performed in 2019. Aggregated data on FP include numbers and types of cryopreserved material and interventions for the use of cryo-stored material between 1 January and 31 December in 2019.

4The national representatives of the 44 countries being members of the EIM consortium were asked to fill out the survey with the same data requirements as in 2018. A total of 10 modules on specific topics/questions were sent using software designed for the requirements of this data collection (Evidenze, former: Dynamic Solutions, Barcelona, Spain). Any identified inconsistency was clarified through direct contact between the administrator of the ESHRE central office (VG) and the national representative.

5The data were analyzed and presented similarly to previous reports. Footnotes to the tables were added to clarify some results reported by individual countries, when applicable.

6The terminology used was based on the glossary of The International Committee for Monitoring Assisted Reproductive Technology.

7Table 1 shows the number of clinics providing ART services with the different treatment modalities they offer and institutions performing IUI (IUI-H and IUI-D). Compared to 2018, the total number of reporting clinics (1488 versus 1422 in 2018) and the number of reported treatments (1 077 813 versus 1 007 598 in 2018, +7.0%) increased. Among the 51 European countries, 44 are EIM members including 28 that were members of the European Union (EU) at that time and 40 (39 in 2018) provided data (Supplementary Table S1). Non-EIM members are mainly small countries not offering ART services. Croatia, Cyprus, Georgia, and Romania did not deliver data in 2019 (9.1% of EIM members). In 21 countries (52.5% of reporting countries), all ART clinics participated in the reporting. Among 1774 (1552 in 2018) known IVF clinics in Europe, 1488 clinics reported data sets (83.9% versus 91.6% in 2018). The main differences with 2018 can be explained by the renewed, but still limited, participation of Turkey in 2019. As in 2018, the four European countries with the largest treatment numbers in 2019 were Russia (161 166; 155 949 in 2018), Spain (137 276; 140 498 in 2018), France (118 394; 106 884 in 2018), and Germany (107 136; 105 328 in 2018).

8The size of reporting clinics, as calculated based on the number of fresh and frozen cycles per year, was highly variable among and within countries, as seen in previous years (Supplementary Table S2). In 2019, as in 2018, clinics with cycle numbers between 200 and 499, and 500 and 999, were the most common (25.7% and 27.8%, respectively, versus 27.3% and 26.3%). The proportion of clinics performing more than 1000 treatment cycles per year was slightly higher than in 2018 (22.3% versus 21.0% in 2018). Small clinics with fewer than 100 treatment cycles per year were present in 24 countries (21 countries in 2018).

9Requirements of registries and reporting methods of the countries are shown in Supplementary Table S3. Data collection was either voluntary (15 out of 40 countries) or compulsory. Twenty-six countries reported all or a part of the treatment cycles to the national health authority. Among 19 countries with only partial reporting, data were mainly provided voluntarily (14 countries) to medical organizations (10 countries), to the national health authority (8 countries), or to a single individual who took the initiative to organize the data collection (2 countries).

10In contrast, complete reporting was most often achieved when data collection was compulsory (20/21 countries) and with data communication to the national health authority (all but three countries). Transfer of the data was mainly done on an aggregate basis (24 out of 40 countries).

11In 2019, 1 077 813 treatment cycles were reported to EIM (70 215 more than in 2018, +7.0%). Since 1997, the increasing numbers of clinics reporting to EIM resulted in a total of 12 804 411 treatment cycles and the birth of more than 2 479 254 infants (Table 2). As seen in Table 1, most countries reported similar numbers of treatment cycles as in 2018. Furthermore, the largest increments in reported treatment numbers were observed in France (+11 510, +10.8%) and Ukraine (+10 081, +35.6%). The largest reduction in reported treatment cycle numbers was seen in Denmark (−4757, −23.1%).

12Table 1 shows the number of treatment cycles per technique in 2019: ICSI remains the most used technology (427 980, 39.7%), versus 400 375 (39.7%) in 2018. Cycles with IVF, FET, ED, FOR, PGT, and IVM represented 14.9%, 31.2%, 7.6%, 0.6%, 5.9%, and 0.0005% of all cycles, respectively, in 2019. The distribution of the available techniques remained similar to 2018 (respectively, 16.2%, 30.7%, 8%, 0.5%, 4.8%, and 0.0005%). Reported cycle numbers with ICSI, FET, ED, PGT, IVM, and FOR increased, and only those with IVF decreased (−1.3%).

13The steepest rise in treatment numbers was observed for PGT (+32.7%; +29.5% in 2018), FOR (+15.7%; +4.5% in 2018), and FET (+8.5%; +14.0% in 2018).

14The highest proportions of FET treatments (calculated as FET/(FET + ICSI + IVF)) were reached in Armenia (60.8%), Czech Republic (52.2%), The Netherlands (51.7%), Ukraine (48.0%), Finland (47.2%), Belgium (47.1%), and Switzerland (45.9%) with an overall proportion of 36.3% and comparable to 35.5% in 2018 (Fig. 1A and B).

15Figure 1A shows the evolution and continuing preponderance of ICSI over conventional IVF. Among a total of 588 762 fresh treatments (ICSI + IVF), 72.7% (71.1% in 2018) were done with ICSI.

16The number of treatment cycles per million women of reproductive age and per million inhabitants is shown in Table 1 and Supplementary Table S4. Availability of ART treatments was calculated for the 21 countries with full coverage (Supplementary Table S4) showing a huge variability in availability when all techniques are considered (range per million women aged 15–45 years: 3943 in Lithuania to 19 393 in Czech Republic). Corresponding proportions of newborns resulting from ART ranged from 1.2% to 6.3% of all newborns in these countries. Among the countries with complete reporting to the national registry, proportions of ART infants above 5% were reached in Belgium, the Czech Republic, Denmark, Estonia, and Iceland.

17Table 3 shows pregnancy and delivery rates after IVF or ICSI and after FET (after both IVF and ICSI). Outcome data were calculated per aspiration, rather than per initiated cycle, as the numbers of initiated cycles have often been reported incompletely.

18Among the 40 reporting countries, 31 were able to provide both pregnancy and delivery rates per aspiration after IVF. After ICSI, 34 countries were able to provide both pregnancy and delivery rates per aspiration. For FET when considering thawing cycles, 33 countries were able to report pregnancy and delivery rates (32 in 2018). Supplementary Table S4 shows the number of deliveries for the 21 countries with full coverage.

19Significant variation in pregnancy and delivery rates (for all types of treatment cycles) was observed among different countries, as in previous years.

20Per aspiration, pregnancy rates (PRs) are shown in Fig. 2A and ranged from 16.0% to 53.1%. The delivery rates are shown in Fig. 2B and ranged from 10.6% to 29.4% in fresh cycles after IVF or ICSI (including the freeze-all cycles whether performed or not by the countries) (Table 3). For FET, pregnancy and delivery rates per thawing varied between 22.5% and 50.1% and between 7.2% and 41.4%, respectively. Overall, while higher pregnancy and delivery rates were recorded for FET cycles (per thawing) than for both fresh IVF and ICSI cycles (per aspiration) (Table 3; Supplementary Table S7), PRs per transfer in fresh cycles remained at the same level (34.6% for IVF and 33.5% for ICSI; Fig. 3A), but were slightly higher in FET cycles (35.8%), as were delivery rates per transfer (25.3% for IVF, 24.1% for ICSI, and 25.6% for FET), as in 2018 (Supplementary Tables S5, S6, and S7 and Fig. 3B).

21When considering the developmental stage of replaced embryos, the data showed PRs for blastocyst transfers to be higher (39.4%) than for cleavage-stage embryos (26.5%) in fresh IVF and ICSI cycles together (it was not possible to distinguish between IVF and ICSI). A similar picture was seen in FET cycles: 40.5% PRs for blastocyst transfers versus 26.9% for cleavage-stage embryos.

22Cycle numbers, aspirations, transfers, pregnancies, and deliveries in IVF, ICSI, and FET (after both IVF and ICSI) by country are presented in Supplementary Tables S5, S6, and S7.

23For the sixth year, ‘freeze-all’ cycles were collected (Supplementary Tables S5 and S6) including either freezing of all oocytes reported by 11 countries for IVF (11 in 2018 and 10 in 2017) and 19 countries for ICSI (17 in 2018 and 17 in 2017), or of all embryos by 26 countries for IVF (23 in 2018 and 22 in 2017) and 27 countries for ICSI (25 in 2018 and 27 in 2017). The highest proportions of freeze-all cycles per aspiration (oocytes and embryos together) were 61.0% (29.8% in 2018) and 61.9% (41.7% in 2018), respectively, for IVF and ICSI.

24ED cycle numbers were available for 20 countries (23 in 2018) although 27 (27 in 2018) provided outcome data (Supplementary Table S8). The highest numbers of ED cycles were reported from Spain, the Czech Republic, and Russia, as in 2018. The number of aspirations of donated oocytes was 34 406 (36 938 in 2018) resulting in 22 932 fresh transfers (24 148 in 2018), while the number of replacements of frozen oocytes (FOR) was 16 122 (16 130 in 2018). The PR per fresh ET was 50.5% (49.6% in 2018) for freshly donated oocytes and 44.8% (44.9% in 2018) for thawed oocytes. High variability was seen between countries, ranging from 31.5% to 100% for fresh oocytes and from 23.2% to 80.0% for thawed oocytes, as in previous years, although sometimes small numbers were observed. Overall (including also the transfers of frozen embryos), 25 156 deliveries were reported with donated eggs (25 760 in 2018 and 21 312 in 2017). Compared to cycles with own oocytes, pregnancy and delivery rates per transfer were higher for fresh (IVF and ICSI) and FET cycles together.

25Supplementary Tables S9 and S10 showed that the age distributions of women treated with IVF and ICSI, respectively, varied between countries. Some countries were not able to provide age categories (nine for IVF and six for ICSI). The highest percentage of women aged 40 years and older undergoing aspiration for IVF was reported in Greece (as in 2018), whereas the highest percentage of women aged <34 years was reported in Ukraine (as in 2018). For ICSI, the highest percentage of women aged 40 years and older undergoing aspiration was also reported in Greece (as in 2018), whereas the highest percentage of women undergoing aspiration aged <34 years was recorded in Sweden (as in 2018). An age-dependent decrease in pregnancy and delivery rates for IVF and ICSI cycles was reported, as expected. Pregnancy and delivery rates in women aged 40 years and older ranged between 6.5% and 56.7%, and 1.5% and 23.8%, respectively. These age-related declines were also visible in FET cycles (Supplementary Table S11) with recorded pregnancy and delivery rates among women aged 40 years and older ranging from 7.7% to 42.5% and 0% to 35.5%, respectively.

26As seen in Supplementary Table S12, the age of the recipient women had little influence on the outcomes of ED cycles.

27Differences in the number of embryos replaced per transfer after IVF and ICSI together, with multiple birth rates per subgroups defined by the number of embryos replaced, are presented in Table 4.

28Six countries reported neither the number of replaced embryos nor the multiplicity. Most transfers involved the replacement of one embryo (elective or not) (55.4% of cycles, as compared to 50.7% of single embryo replacements in 2018). The evolution of the proportions of replacements of one, two, and three or more embryos is shown in Fig. 4A.

29Twenty-three countries reported more than 50% single embryo transfers (17 in 2018) with seven reporting more than 75% single embryo transfers. None of the reporting countries carried out more than 50% of their transfers with three embryos. Among the seven countries recording transfers of four or more embryos, the highest proportion was found in Serbia (3.0%; 5.5% in Greece in 2018). For the third consecutive year, the embryonic developmental stage at transfer was recorded. Taking into account that the embryo stage at transfer was unknown in 18.4% of the fresh (IVF + ICSI) cycles, 52.8% (50.1% in 2018) of the transfers were performed at the blastocyst stage. The corresponding percentage for FET was 79.1% (73.9% in 2018). Information about the embryonic developmental stage was not available with respect to the number of embryos replaced.

30As a result of the decreasing number of embryos replaced per transfer, the global proportion of twin and triplet deliveries continued to decrease (Fig. 4B). Twin and triplet rates for fresh IVF and ICSI cycles together were 12.0% (range 0–26.9) and 0.3% (range: 0–3.8), respectively. Corresponding results for FET were 9.3% and 0.1%. Two countries reported rates of single embryo replacement above 95% in fresh cycles (100% in Iceland, 95.7% in Finland) and twin rates were as low as 0% (in Iceland, data from Finland were not available).

31Supplementary Tables S13 and S14 provide additional information on pregnancies and deliveries. The reported incidence of pregnancy loss was 19.9% after removing the data of those countries in which pregnancy loss was not documented (19.3% in 2018) after IVF + ICSI and 21.5% (21.4% in 2018) after FET. The proportion of recorded lost to follow-up pregnancies was 8.7% (7.2% in 2018) after IVF + ICSI and 8.6% (7.2% in 2018) after FET.

32Data on premature deliveries were available from 21 countries (21 countries in 2018). Premature delivery rates (for fresh IVF and ICSI, FET, and ED together) according to multiplicity are presented in Supplementary Table S15. The incidence of extremely preterm birth (20–27 gestational weeks at delivery) was 1.5% in singleton pregnancies (1.0% in 2018), 3.3% in twins (3.1% in 2018), and 12.2% in triplets (6.0% in 2018). Very premature birth rates (28–32 gestational weeks at delivery) were recorded in 3.4% of singletons (2.2% in 2018), 10.9% of twin pregnancies (9.7% in 2018), and 40.8% in triplet pregnancies (37.9% in 2018). The evolution of the proportions of premature deliveries (before 37 weeks) over the years according to multiplicity is shown in Fig. 5. Term deliveries (≥37 weeks) were achieved in 84.2% (83.1% in 2018) of singleton pregnancies, 44.0% (43.6% in 2018) of twin pregnancies, and 8.2% (8.1% in 2018) of triplet pregnancies.

33Complications and fetal reductions related to ART procedures were reported by 35 countries (34 in 2018) (Supplementary Table S16). The main reported complication was ovarian hyperstimulation syndrome (OHSS) (grades 3–5) with a total reported number of 1654 (1719 in 2018) corresponding to an incidence rate of 0.16% (0.17% in 2018). Other complications (1575; 1379 cases in 2018) were reported with a total incidence of 0.15% (0.14% in 2018) and with bleedings being the most frequent (0.1%, identical to 2018).

34Two maternal deaths were reported (3 in 2018). France reported a 32-year-old patient who died 6 weeks after oocyte retrieval because of a massive pulmonary embolism. In Russia, a patient died after IVF because of a pulmonary embolism after severe OHSS in early pregnancy.

35Fetal reductions were reported in 487 cases (509 in 2018), the majority from the UK, Belgium, and Ukraine.

36Table 1 includes PGT activities, which were reported from 27 countries (24 in 2018). The main contributors were Spain, Russia, Ukraine, Turkey, and Italy.

37The total number of treatment cycles was 64 089 representing 6.9% of initiated IVF + ICSI and FET cycles together (48 294; 7.1% in 2018).

38More details on PGT activities can be found in the annual reports of the ESHRE PGT Consortium.

39A total of 546 treatments with IVM were reported from nine countries (532 treatments from eight countries (Serbia) in 2018) (Table 1).

40Most IVM cycles were recorded in Belgium, as in 2018. A total of 188 transfers resulted in 37 pregnancies (19.7% per transfer) and 25 deliveries (13.3% per transfer).

41A total number of 6299 thawing cycles were reported by 22 countries (5444 from 22 countries in 2018) (Table 1) with Italy and the UK being the largest contributors (1361 and 998 cycles, respectively).

42Among 4402 transfers, 1075 resulted in pregnancies (29.5%; 29.5% in 2018) and 867 in deliveries (24.4%; 21% in 2018).

43Data on IUI-H or IUI-D were collected by a total of 1169 institutions (1271 in 2018) in 30 and 25 countries, respectively, as in 2018. Among 147 711 IUI-H (148 143 in 2018) and 51 651 IUI-D (50 609 in 2018) reported cycles, the numbers of IUI-H were the highest in France, Spain, Italy, Belgium, and Poland, and those of IUI-D were the highest in Spain, Belgium, Denmark, and the UK (Table 5 and Supplementary Tables S17 and S18).

44Delivery rates could be calculated for 139 870 IUI-H cycles (9.2%; 8.9% in 2018) and 45 436 for IUI-D cycles (12.1% versus 12.6% in 2018). Singleton deliveries were the most frequent regardless of the age group with an overall rate of 90.9% for IUI-H and 93.5% for IUI-D (91.2% in IUI-H, 93.4% in IUI-D in 2018). Twin and triplet rates were 8.7% and 0.4%, respectively, after IUI-H, and 6.2% and 0.2% after IUI-D, respectively (in 2018: 8.4% and 0.3%, respectively, after IUI-H and 6.4% and 0.2%, respectively, after IUI-D).

45Supplementary Table S19 provides an estimate of a cumulative delivery rate. The cumulative delivery rate was calculated as the ratio between the total number of deliveries from fresh embryo transfers and FET performed during 1 year (numerator) and the number of aspirations during the same year (denominator). The cumulative delivery rate thus differs from a true cumulative delivery rate, which is based on all transfers resulting from one aspiration. The calculation was based on data provided by 37 countries (36 countries in 2018) with an overall delivery rate of 31.4% (32.3% in 2018). The cumulative increase resulting from additional FET (overall delivery rates from fresh embryo transfers) reached 15.3% (14.4% in 2018). In some countries, more deliveries were reported after FET than after a fresh IVF/ICSI cycle. Consequently, as a result of the relative contribution of FET the cumulative PR was high in Armenia, Belgium, Czech Republic, Denmark, Finland, Iceland, Ireland, Kazakhstan, Latvia, Malta, Moldova, Poland, Spain, Sweden, Switzerland, Turkey, and Ukraine. The relative lowest contribution of FET to the cumulative PRs came from Germany, Hungary, Lithuania, Malta, Montenegro, North Macedonia, and Serbia.

46Thirteen countries reported data on cross-border reproductive care: Albania, Belarus, Czech Republic, Greece, Iceland, Ireland, Malta, Poland, Portugal, Slovenia, Spain, Switzerland, and Turkey. A total of 33 003 cycles (21 792 in 2018) were reported, 29.4% (21.5% in 2018) of which involved IVF/ICSI with the couple’s own gametes, 52.9% (52.6% in 2018) were oocyte reception cycles, and 13.7% (20.6% in 2018) were IVF or ICSI cycles with semen donation. Additionally, 2456 IUI with sperm donation (6791 in 2018) were registered. Information about the countries of origin was very incomplete and not reliable enough to draw any conclusion. The main reason reported by patients for crossing the borders was to seek less expensive treatment (43.0%; 25.0% in 2018). However, cross-border reproductive care was also reported to be performed because the treatment was of too low quality (29.7%; 42.3% in 2018) or not legal in the home country (13.8%; 21.1% in 2018).

47For the fourth year, data on FP were reported. Eighteen countries (16 in 2018 and 14 in 2017) provided data on a total number of 24 139 interventions (20 994 in 2018; 18 888 in 2017) (Supplementary Table S20) both for medical and non-medical reasons in pre- and post-pubertal patients. The majority of interventions consisted of the cryopreservation of ejaculated sperm (n = 11 592 with data from 16 countries, n = 10 503 with data from 14 countries in 2018) and the cryopreservation of oocytes (n = 10 784 with data from 15 countries, n = 9123 with data from 16 countries in 2018). Ovarian tissue cryopreservation was reported by 2 (2 in 2018) and 11 (11 in 2018) countries, respectively, for pre- and post-pubertal patients. The use of post-pubertal tissue through transplantation was reported by four countries (Greece, Italy, Slovenia, and Spain). Testicular tissue cryopreservation in post-pubertal patients and pre-pubertal boys was reported by 14 (8 in 2018) countries and by one country (1 in 2018), respectively.

48Between 1997 and 2019, the EIM Consortium of ESHRE reported on a total of over 12 million treatment cycles (12 804 411) resulting in the birth of over 2 million infants.

49The current 23rd annual report presents a comprehensive analysis of data on ART, IUI, and FP activities. The data are derived from compulsory or voluntary registries of 40 European countries (one more than in 2018). For the second time, the number of reported treatment cycles per year exceeded 1 million. Only a few countries opted out of participation (5 out of 44 EIM members as well as 7 non-EIM members including Azerbaijan, Kosovo, and 5 smaller countries not offering ART services). Furthermore, data could not be obtained from three member states of the EU (Croatia, Cyprus, and Romania), most probably because of economic, regulatory, or political factors, as suggested by a survey on medically assisted reproduction (MAR) activities.

50Overall, the number of European countries actively participating has remained relatively stable in recent years, with only slight fluctuations. There has been a continued increase in the reported number of treatment cycles (+7.0% compared to 2018). In contrast, the number of infants born from ART per year declined (−5.4% compared to 2018), as did the percentage of ART infants per national births (3.0%; 3.5% in 2018).

51Despite the well-known challenges associated with heterogeneous data collection systems in Europe and the lack of standardized indicators, the participation rate at the country level remains very high with 90.9% of EIM members contributing data after excluding those countries where ART services are not available. However, 21 countries (52.5% of EIM members) managed to submit data from all IVF institutions, resulting in 83.9% of all IVF institutions sending in their data (versus 91.6% in 2018). The main reason for this decrease is the newly installed participation of Turkey, where 26 out of 167 clinics (15.5%) submitted their data. Therefore, future efforts should prioritize the collection of complete data sets within each country.

52To enhance the quality of the data from participating countries, progress is expected through implementing a prospective cycle-by-cycle data collection (already established in 16 countries in 2019) with harmonized indicators. As an initial step, a minimum core data set with defined outcome parameters and collected items was established (https://www.eshre.eu/Data-collection-and-research/Consortia/EIM).

53Until better quality data are available, interpretation of the data should be carried out with caution. Better quality includes the harmonization of data collection systems across countries and registration of indicators, taking into account center/country-specific practices (e.g. freeze-all cycles, embryo transfer policy, PGT-A, etc.).

54Besides the current EU objective to enhance vigilance in the field of MAR, increased transparency regarding access to reproductive care and cross-border treatments for all stakeholders is equally important. Over the years, the EIM Consortium has constantly recorded significant variation in access to treatment between countries, with the number of ART cycles per million women aged 15–45 years ranging from 3943 in Lithuania to 19 393 in the Czech Republic, and per million inhabitants from 694 in Lithuania to 3621 in the Czech Republic.

55While such data are unique in Europe, the interpretation becomes more and more difficult owing to the historically estimated threshold of 1500 fresh ART cycles per million inhabitants. This threshold was previously considered necessary for adequate infertility care but technological advancements in the field have proved this outdated. Furthermore, cross-border patients also need to be considered when best estimates for sufficiency thresholds are established. Data on cross-border care were only available for 13 countries in 2019 (12 in 2018), indicating once more the need for a better pan-European registry.

56Concerning treatment modalities, ICSI remains the most commonly used technique with a trend to stabilization of its use over the last few years (Table 1 and Fig. 1). FET is the second most employed technique. Over the years, higher proportions of FET treatment cycles [FET/(FET + ICSI + IVF)] were recorded and have now reached a relatively stable level (36.3% in 2019 and 35.5% in 2018). Higher proportions of FET treatment cycles were observed in other large registries. However, the proportion of FET cycles varies considerably among countries with complete data sets (ranging from 10.9% to 60.8%) highlighting the considerable variability in practices. This is also observed for freeze-all cycles (Supplementary Tables S5 and S6) reported since 2014, with an overall proportion of 12.7% (11.4% in 2018) per aspiration (oocytes and embryos together) for IVF and 17.3% (15.5% in 2018) for ICSI. However, these numbers also vary among countries with proportions reaching as high as 61.0% (29.8% in 2018) and 61.9% (41.7% in 2018) for IVF and ICSI, respectively. Initial studies showed that the freeze-all strategy could be beneficial for subgroups of patients; however, the policy is being more and more frequently used for all patient categories.

57This variability among countries should be considered when interpreting the data, especially regarding the evolution of pregnancy and delivery rates after fresh IVF and ICSI cycles (per aspiration) as well as after FET cycles (per thawing) over time (Table 3 and Figs 2A and B and 3A and B). Indeed, the higher success rates recorded with FET (per thawing) compared to fresh IVF and ICSI (per aspiration), presented here for comparison with previous reports, can be misleading. Several factors that may influence outcomes should be taken into account.

58The PR per aspiration in IVF and ICSI seems to decrease over time. This can be seen in Fig. 2. Obviously, fresh cycles (per aspiration) can include cycles with no oocytes, cycles with failed fertilization, failed embryo development, and freeze-all cycles. Patients who benefit from embryo cryopreservation may even have a better prognosis.

59It should also be noted that recorded delivery rates per transfer were comparable, but even higher after FET than after both IVF and ICSI cycles. One of the explanations could be that more blastocyst transfers were recorded in FET cycles (79%) as compared to blastocyst transfers in fresh IVF and ICSI combined (53%). Such observations, made in large data collection sets, play a crucial role in identifying research questions and potential causes, such as the influence of hormonal support on miscarriage rates or neonatal outcomes in FET cycles.

60Cumulative delivery rates per cycle or per aspiration are better outcome indicators to assess treatment effectiveness. However, so far, the EIM consortium is unable to calculate true cumulative delivery and live birth rates since aggregated data are collected. As a result, the addition of outcomes from fresh and FET cycles within the same calendar year is used as a proxy indicator until a European cycle-by-cycle registry can be established. When data from 37 countries (36 countries in 2018) were included, an approximated ‘cumulative’ delivery rate of 31.4% (32.3% in 2018) was recorded during the 1-year period. The additional benefit derived from FET cycles (compared to delivery rates from fresh embryo transfers) varied widely, ranging from 1.9% to 53.4%, reflecting most likely differences in freezing policies and indications.

61Analyzing trends is important to inform the field about the adoption of data-driven approaches from registries and to assess subsequent modification of practices. For instance, the dissemination of EIM data sets increased awareness among professionals on the benefit of reducing the number of embryos replaced per transfer (Fig. 4A) to diminish multiple births (Fig. 4B). Consequently, most transfers now involve the replacement of a single embryo (elective or not) (55.4% of cycles versus 50.7% with single embryo replacement in 2018). Simultaneously, the proportion of both twin and triplet deliveries showed a small increase in 2019, although the overall trend shows a decrease (Fig. 4B). Twin and triplet rates for fresh IVF and ICSI cycles combined were 11.9% (range 0–26.9) and 0.3% (range: 0–3.8), respectively, while the corresponding results for FET were 8.9% and 0.1%.

62In the future, it is expected that efforts will lead to the ultimate objective of achieving the birth of a single healthy child per embryo transfer and thereby reducing the risks associated with multiple births, such as prematurity (Fig. 5).

63To promote singleton pregnancies through elective single embryo transfer and to reduce the time to pregnancy, embryo culture is often prolonged to the blastocyst stage. However, the benefit of blastocyst stage transfers on ART outcomes is still a matter of debate. In a multicenter randomized controlled trial in good prognosis IVF patients (≥4 available embryos), the live birth rate after fresh embryo transfer was higher in the blastocyst-stage group than in the cleavage-stage group (P = 0.035); however, a blastocyst-stage transfer policy did not, in this study, result in a significantly higher cumulative live birth rate compared to a cleavage-stage transfer policy.

64When analyzing the developmental stage of the replaced embryos, the data showed that PRs for blastocyst transfers were higher compared to cleavage-stage embryos (39.4% versus 26.5%, respectively) in fresh IVF and ICSI cycles combined. In FET, the rates were 40.5% for blastocyst transfers versus 26.9% for cleavage-stage embryos. However, it is important to note that while blastocyst transfers result in higher pregnancy and live birth rates per transfer, they also result in lower numbers of embryos available for transfer. This highlights the importance of true cumulative outcome parameters. Unfortunately, the available data did not include the assessment of the time to pregnancy.

65In addition to multiplicity and prematurity, other safety aspects of ART also remain underreported, among these the rate of complications for OHSS (0.16%, similar to 0.17% in 2018) and an incidence of all other complications being registered at 0.15% (0.14% in 2018). Reports on maternal deaths related to ART are even scarcer, with a best estimate of six maternal deaths per 100 000 IVF treatments directly related to IVF in a national cohort from The Netherlands, where OHSS and sepsis were the major causes. It is noticeable that two maternal deaths after ART were registered in 2019 (Supplementary Table S16), both caused by pulmonary embolisms. At least one case was associated with OHSS.

66Furthermore, while the age of recipients in ED cycles did not significantly affect the outcome of the cycle, risks associated with pregnancies in older women should not be overlooked as a potential safety aspect of the treatment. Indeed, a survey on the legislation and reimbursement aspects has shown that some countries do not have age limitations for recipients in ED cycles.

67The crucial role of registries in MAR activities regarding outcome parameters and safety is well established. The EIM data are also incorporated in the annual report of the worldwide IVF register from the International Committee for Monitoring Assisted Reproductive Technologies.

68To enable reliable comparisons of practices and to identify the safest and most efficient care, it is essential to enhance the quality of collected data and strive to complete and harmonized data throughout Europe. Besides the establishment of clear definitions of registered items, providing the countries and competent authorities with an adapted IT solution should be the next priority. The European monitoring of Medically Assisted Reproduction (EuMAR) aims to develop a pan-European registry of prospective cycle-by-cycle data on the use and outcomes of MAR treatments. EuMAR addresses the need for more transparency, surveillance, and biovigilance in MAR across country borders, including better data on the safety of MAR for offspring, donors, and recipients. These efforts align with the revision of the EU Directives on blood, tissues, and cells.