Assisted reproductive technologies in Canada: 2005 results from the Canadian Assisted Reproductive Technologies Register

Transcription

1 Assisted reproductive technologies in Canada: 2005 results from the Canadian Assisted Reproductive Technologies Register Joanne Gunby, M.Sc., a Francxois Bissonnette, M.D., b Clifford Librach, M.D., c and Lisa Cowan, M.Sc., d on behalf of the IVF Directors Group of the Canadian Fertility and Andrology Society a Pegasus Technologies, Burlington, Ontario; b OVO Fertility Clinic, Saint-Luc Hospital, and University of Montreal, Montreal, Quebec; c CReATe IVF Program, Sunnybrook Health Sciences Center, Women s College Hospital, and University of Toronto, Toronto, Ontario; and d Victoria Fertility Center, Victoria, British Columbia, Canada Objective: To present a report on assisted reproductive technologies (ART) cycles performed in 2005 in Canada. This is the fifth annual report from the Canadian ART Register (CARTR). Design: Prospective cohort study. Setting: Twenty-five of 25 ART centers in Canada. Participant(s): Couples undergoing ART treatment in Canada during Intervention(s): ART treatments, including IVF, intracytoplasmic sperm injection (ICSI), and frozen ET (FET). Main Outcome Measure(s): Clinical pregnancy, live birth, and multiple birth rates. Result(s): A total of 11,414 ART cycles was reported to CARTR. In 8195 IVF/ICSI cycles using the women s own oocytes, the clinical pregnancy rate per cycle started was 32.1% (37.5% per ET procedure), and the live birth rate was 25.6%; the multiple birth rate per delivery was 30.8%, with a triplet birth rate of 1.4%. IVF was performed in 40% of cycles and ICSI in 60% with similar pregnancy rates. One or two embryos were transferred in 68% of cycles; transferring more embryos did not increase the pregnancy rate. In 301 IVF/ICSI cycles using donor oocytes, the clinical pregnancy rate was 46.5%, and the live birth rate was 35.2%; the multiple birth rate was 33.3%, with no triplet birth. In 2498 FET cycles using the woman s own oocytes, the clinical pregnancy rate was 22.8%, and the live birth rate was 17.4%; the multiple birth rate was 24.5%, with a triplet birth rate of 1.6%. Compared with singletons, babies from multiple births had higher risks for preterm birth, low birth weight, and perinatal death. Conclusion(s): For 2005, CARTR achieved 100% voluntary participation from Canadian ART centers for the third consecutive year. Clinical pregnancy and live birth rates continued to increase in 2005 compared with previous years. (Fertil Steril Ò 2009;91: Ó2009 by American Society for Reproductive Medicine.) Key Words: Assisted reproductive technologies, pregnancy rates, in vitro fertilization, intracytoplasmic sperm injection, frozen ET, oocyte donation, multiple births The Canadian Assisted Reproductive Technologies Register (CARTR) was first established in 1999 for the collection of treatment cycle data from Canadian fertility centers that were using assisted reproductive technologies (ART), including IVF, intracytoplasmic sperm injection (ICSI), and frozen ET (FET). The IVF Directors Group of the Canadian Fertility and Andrology Society (CFAS) directs the CARTR program. Participation of ART centers in CARTR is voluntary. The first report from the Canadian ART Register, describing results from ART cycles performed in 2001, was published in 2005 (1). Subsequent reports described CARTR results from 2002 (2), 2003 (3), and 2004 (4). This is the fifth published annual report of Canadian ART outcomes. Received December 28, 2007; revised and accepted February 13, 2008; published online April 18, J.G. has nothing to disclose. F.B. has nothing to disclose. C.L. has nothing to disclose. L.C. has nothing to disclose. Supported by the Canadian Fertility and Andrology Society, Montreal, Quebec, Canada. Reprint requests: Ms. Joanne Gunby, 2534 Cavendish Drive, Burlington ON, L7P 4E4, Canada (FAX: ; gunbyj@ mcmaster.ca). The purpose of this paper is to report on ART cycles performed in Canadian centers in the 2005 calendar year and submitted to CARTR. MATERIALS AND METHODS As in previous years, CARTR collected data for 2005 using the Society for Assisted Reproductive Technology Clinical Outcome Reporting System, version 2 (Redshift Technologies Inc., New York). Staff at each center entered information about patient demographics, diagnosis, and obstetrical history; details of treatment; and pregnancy and birth outcomes for each ART treatment cycle initiated. The completed anonymous case records were sent electronically from each ART center to the CARTR coordinating center, where they were manually checked for accuracy and completeness. Corrections or clarifications were requested from the centers as necessary, but no on-site data validation from source documents was performed. The records from each center were then aggregated for data analysis using the computer program Statistical Package for the Social Sciences (SPSS), version 15 (SPSS Inc., Chicago). ART cycles started between January 1 and December 31, 2005, were submitted to CARTR in /09/$36.00 Fertility and Sterility â Vol. 91, No. 5, May doi: /j.fertnstert Copyright ª2009 American Society for Reproductive Medicine, Published by Elsevier Inc.

2 batch mode twice: once in mid-2006, when the pregnancy outcomes were known, for an internal interim report, and again in mid-2007, when all the birth outcomes were known, for this published report. It was not necessary to obtain Institutional Review Board approval for this study because data collection is one of the requirements for accreditation of centers providing ART services as organized by the CFAS in conjunction with the Canadian Council on Health Services. Although participation in accreditation is voluntary, most of the ART centers in Canada have agreed to the process and are obliged to inform patients that such data will be collected in a manner that is anonymous. These data from CARTR for 2005 were presented to the Medical and Laboratory Directors at the annual IVF Directors Meetings in November 2006 (pregnancy outcomes) and September 2007 (birth outcomes). A brief summary of the national clinical pregnancy and live birth rates was provided to the media, with the Directors permission, immediately after each meeting. Definitions of Outcomes The definitions established by the International Committee Monitoring Assisted Reproductive Technologies (ICMART) are followed by CARTR (5). A treatment cycle is considered to have started when a woman undergoing ovarian stimulation receives the first dose of gonadotropins or, in a nonstimulated cycle (e.g., for FET), when a decision is made to attempt ART treatment in that cycle. A canceled cycle is one that is stopped before the oocyte retrieval procedure or thawing of embryos. Clinical pregnancy includes intrauterine gestation (presence of a gestational sac on ultrasonography), ectopic pregnancy, and miscarriage diagnosed by histology. Cycles with only a positive pregnancy test (biochemical pregnancy) are not considered to have a clinical pregnancy. The implantation rate is the number of gestational sacs observed on ultrasonography divided by the number of embryos transferred. Pregnancy loss includes miscarriage and therapeutic abortion of a clinical intrauterine pregnancy occurring at %20 weeks gestation. Any pregnancy termination, either spontaneous or therapeutic, occurring after 20 weeks gestation with no liveborn infant is considered a stillbirth. A delivery is the birth of one or more infants, either living or not, after 20 weeks gestation. A live birth is a delivery that results in at least one living infant (but, if a multiple birth, may include one or more stillborn infants). A neonatal death is the death of an infant in the first 28 days of life. A multiple birth is the delivery of more than one infant, either live born or stillborn, including deliveries with all infants stillborn. High-order multiple births (triplets or more) are reported separately. Statistical Analysis The statistics used in this report are mainly descriptive, that is, rates, proportions, means, and medians. The c 2 -test was used occasionally to compare proportions. The c 2 -test with trend was used to evaluate the change over time in pregnancy, live birth, and multiple birth rates. Unless otherwise noted, the clinical pregnancy rate is reported per cycle started. Cycle cancellation, ectopic pregnancy, and other complications are reported per cycle started. The miscarriage or pregnancy loss rate is reported per intrauterine pregnancy. The live birth rate is reported per cycle started, excluding from both the numerator and the denominator cycles in which the outcome of the clinical pregnancy has not been reported. Because of these missing data, the live birth rates reported may underestimate the true live birth rates. The multiple birth rate, which includes stillbirths, is reported per delivery. RESULTS Participating Centers All 25 Canadian ART centers operating in 2005 contributed to CARTR for that year (listed in the Appendix). One of the 25 centers performed more than 1000 ART cycles in 2005, seven centers performed between 500 and 1000 cycles, 11 centers performed between 200 and 500 cycles, and six centers performed fewer than 200 cycles. Overall Results In total, 11,414 treatment cycles involving ART were reported to CARTR for Overall, 3443 ART cycles (30.2% of cycles started) resulted in a clinical pregnancy, at least 2713 cycles resulted in a delivery (24.0%), and at least 2687 cycles resulted in a live birth (23.8%). There were 108 cycles with ongoing pregnancies (3.8% of ongoing pregnancies) for which the birth outcome was not reported. Overall, there were at least 804 multiple births (29.6% of known births): 767 twin births (28.3%) and 37 triplet births (1.4%). The various procedures and their outcome rates are described in the following sections. The cycle outcomes of the four most common procedures are summarized in Table 1. IVF/ICSI with Own Oocytes IVF, including ICSI, was the most common procedure performed, with 8195 cycles reported. This category (referred to hereafter as IVF/ICSI), to distinguish it from those of oocyte donation (OD) and gestational carrier cycles, includes only cycles in which the parenting woman s own oocytes are used and the same woman receives the resulting embryos. However, cycles using donated sperm are included. Because the decision to use ICSI might not be made until the sperm and oocytes are assessed in the embryology laboratory, cycles canceled before oocyte retrieval cannot be classified by type of insemination procedure; thus, outcomes per cycle started can only be calculated for IVF and ICSI cycles grouped together Gunby et al. Canadian ART register 2005 Vol. 91, No. 5, May 2009

3 TABLE 1 Cycle outcomes for the four most common types of ART procedures. Outcome IVF/ICSI IVF/ICSI with OD FET FET with OD Cycles started Cancelled cycles 677 (8.3) 5 (1.7) 124 (5.0) 2 (1.4) (% of cycles started) Oocyte retrievals 7518 (91.7) 296 (98.3) 2374 (95.0) a 141 (98.6) a (% of cycles started) ETs (% of cycles started) 7001 (85.4) 284 (94.4) 2243 (89.8) 134 (93.7) Clinical pregnancy 2631 (32.1) 140 (46.5) 569 (22.8) 36 (25.2) (% per cycle started) Ectopic pregnancy 58 (0.7) 3 (1.0) 12 (0.5) 0 (% per cycle started) Miscarriage 389 (15.1) 23 (16.8) 109 (19.6) 8 (22.2) (% per intrauterine pregnancy) Therapeutic abortion 14 (0.5) 1 (0.7) 1 (0.2) 0 (% per intrauterine pregnancy) Delivery b (% per cycle started) 2097 (25.8) 102 (35.2) 437 (17.6) 27 (19.0) Live birth b (% per cycle started) 2076 (25.6) 102 (35.2) 433 (17.4) 26 (18.3) Singleton live birth b 1435 (17.7) 68 (23.4) 326 (13.1) 21 (14.8) (% per cycle started) Singleton delivery b 1451 (69.2) 68 (66.7) 330 (75.5) 22 (81.5) (% of deliveries) Twin delivery b (% of deliveries) 616 (29.4) 34 (33.3) 100 (22.9) 5 (18.5) Triplet delivery b (% of deliveries) 30 (1.4) 0 7 (1.6) 0 a Cycles with embryos thawed. b Cycles with unknown delivery status omitted. Gunby. Canadian ART register Fertil Steril Per IVF/ICSI cycle started, the clinical pregnancy rate was 32.1%, the live birth rate was 25.6%, and the singleton live birth rate was 17.7%. Donated sperm was used in 4.2% of cycles with oocytes retrieved. There were 58 ectopic pregnancies (0.7%), including six heterotopic pregnancies: one of the intrauterine gestations was nonviable, and five resulted in a singleton live birth. The pregnancy loss rate was 15.6% (miscarriage 15.1%, therapeutic abortion 0.5%). Of the 2097 known births (96% of ongoing pregnancies), 30.8% were multiple births (29.4% twins, 1.4% triplets). Included in these figures are eight pregnancies, three miscarriages, and five singleton live births that resulted from IUI performed after the IVF/ICSI cycle was cancelled. Rates for IVF and ICSI separately can only be provided per successful retrieval (i.e., one or more oocytes retrieved). Of 7485 IVF/ICSI cycles with a successful retrieval, 40.2% had insemination by standard IVF, 53.2% by ICSI, and 6.6% by IVF/ICSI split (some oocytes inseminated by each method). The clinical pregnancy rates per successful retrieval were 34.1%, 36.1%, and 32.1%, respectively. Including the IVF/ ICSI split cycles in the ICSI group, the clinical pregnancy rates were 34.1% for IVF and 35.7% for ICSI, and the live birth rates were 26.8% and 28.8%, respectively. The ectopic pregnancy rate was 0.8% with both IVF and ICSI, and the pregnancy loss rates were 16.9% (miscarriage 16.1%, therapeutic abortion 0.8%) and 14.7% (miscarriage 14.3%, therapeutic abortion 0.4%), respectively. Of 809 known births after IVF, 30.4% were multiple births (29.5% twins, 0.9% triplets); of 1283 known births after ICSI, 31.2% were multiple births (29.4% twins, 1.8% triplets). IVF/ICSI with Oocyte Donation IVF/ICSI with OD was reported in 301 cycles in In IVF/ICSI with OD, one woman undergoes ovarian stimulation and then donates some or all of the retrieved oocytes to another woman, usually anonymously. These oocytes are inseminated with sperm from the recipient s partner (or a sperm donor), and the resulting embryos are transferred to the uterus of the recipient. Information about the donor s age was not collected in In OD cycles, the clinical pregnancy rate per cycle started was 46.5%, the live birth rate was 35.2%, and the singleton live birth rate was 23.4%. Donated sperm was used in 12.7% of cycles with oocytes donated. There were three ectopic pregnancies (1.0%). The pregnancy loss rate was 17.5% (miscarriage 16.8%, therapeutic abortion 0.7%). Of 102 known births (90% of ongoing pregnancies), 33.3% were multiple births (all twins). Fertility and Sterility â 1723

4 Of 281 cycles with a successful retrieval and known insemination method, 35.6% had insemination by standard IVF, 52.0% by ICSI, and 12.5% by IVF/ICSI split. The clinical pregnancy rates per successful retrieval were 43.0%, 49.3%, and 51.4%, respectively. FET with Own Oocytes FET involves thawing embryos created and cryopreserved in a previous IVF/ICSI cycle and transferring them to the uterus of the woman who provided the oocytes in the original cycle. In 2005, 2498 such cycles were reported. Per cycle started, the clinical pregnancy rate was 22.8%, the live birth rate was 17.4%, and the singleton live birth rate was 13.1%. There were 12 ectopic pregnancies (0.5%). The pregnancy loss rate was 19.7% (miscarriage 19.6%, therapeutic abortion 0.2%). Of 437 known births (98% of ongoing pregnancies), 24.5% were multiple births (22.9% twins, 1.6% triplets). FET with Oocyte or Embryo Donation The category FET with OD includes transfer of cryopreserved embryos created from donor oocytes in a previous IVF/ICSI with OD cycle (110 cycles) and cryopreserved donated embryos (33 cycles). In the latter case, both the male and female gametes were provided by persons other than the intended parenting couple. The thawed embryos are transferred to the woman who intends to raise the child. In this category, the clinical pregnancy rate per cycle started was 25.2%, the live birth rate was 18.3%, and the singleton live birth rate was 14.8%. There was no ectopic pregnancy. The miscarriage rate was 22.2%. Of 27 known births (96% of ongoing pregnancies), 18.5% were multiple births (all twins). Gestational Carrier Cycles There were 103 cycles in which embryos were transferred into the uterus of a woman other than the one who intended to raise the child. Gestational carriers were used in 49 IVF/ ICSI and 29 FET cycles with the parenting woman s own oocytes and 17 IVF/ICSI and eight FET cycles with donated oocytes. Donated sperm was used in one cycle. In fresh embryo cycles using a gestational carrier, the clinical pregnancy rate per cycle started was 47.0%, the live birth rate was 37.1%, and the singleton live birth rate was 29.0%; in frozen embryo cycles, the rates were 32.4%, 18.9%, and 10.8%, respectively. Of the 40 clinical intrauterine pregnancies in gestational carriers, 15.0% ended in miscarriage; there were three ectopic pregnancies. Of 30 known births (88% of ongoing pregnancies), 26.7% were multiple births (all twins). Other Cycle Types Several other types of ART procedures that did not fit into the categories previously described were reported to CARTR for Natural (unstimulated) IVF was performed in 106 cycles, with clinical pregnancy rates of 11.3% per cycle started and 29.3% per ET and live birth rates of 7.8% and 21.1%, respectively, all singletons. Four cycles were reported in which oocyte retrieval was performed for the sole purpose of freezing oocytes. In 44 cycles, previously frozen oocytes were thawed and inseminated, with clinical pregnancy and live birth rates of 20.5%; four of nine births were multiple (all twins). Five cycles involving fresh donor embryos resulted in two pregnancies and two singleton live births. Two cycles of gamete intrafallopian transfer and one cycle of in vitro oocyte maturation were reported (no pregnancy). Six cycles were performed for the purpose of embryo banking. Preimplantation genetic diagnosis (PGD) was performed in 27 IVF/ICSI cycles, resulting in six pregnancies and six live births (five singleton and one twin). In addition, preimplantation genetic screening (PGS) for aneuploidy was reported for 58 IVF/ICSI cycles and one FET cycle, resulting in 12 pregnancies and 11 live births (nine singleton and two twin). However, use of PGD and PGS may be underreported as information on these techniques was not specifically requested in Birth Outcomes for All ART Procedures At least 3554 infants were born from all types of ART cycles started in 2005 in Canada: 1909 infants from 1909 singleton births (53.7% of infants), 1534 infants from 767 twin births (43.2%), and 111 infants from 37 triplet births (3.1%). An additional 108 pregnancies had no delivery information reported. Of these pregnancies, 23 had two viable fetuses and three had three viable fetuses at last report; thus, as many as 136 additional babies may have been born. Of the 1909 infants born as singletons, there were 21 stillbirths and 11 neonatal deaths, a total perinatal mortality rate of 1.7%. The median gestational age at birth was 39 weeks (range, weeks) for liveborn infants and 31 weeks (range, weeks) for stillborn infants. Preterm delivery (<37 weeks) occurred in 17.3% of births and very preterm delivery (<34 weeks) in 4.7% of births. The birth weight was >2500 g for 89.8% of liveborn singletons, g for 6.5%, g for 2.5%, and <1000 g for 1.2%. Some type of birth defect was reported for 46 infants (2.4%). Of the 1534 infants born as twins, there were 18 stillbirths and 23 neonatal deaths, a total perinatal mortality rate of 2.7%. The median gestational age at birth was 36 weeks (range, weeks) for live births and 25 weeks (range, weeks) for stillbirths. Preterm delivery occurred in 71.7% of births and very preterm delivery in 22.7% of births. Birth weight was >2500 g for 43.0% of liveborn twins, g for 36.2%, g for 17.1%, and <1000 g for 3.6%. Some type of birth defect was reported for 31 infants (2.0%). Of the 111 infants born as triplets, there was one stillbirth and four neonatal deaths, a total perinatal mortality rate of 1724 Gunby et al. Canadian ART register 2005 Vol. 91, No. 5, May 2009

5 4.5%. The median gestational age at birth was 31 weeks (range, weeks) for live births. Preterm delivery occurred in 100% of births and very preterm delivery in 78.8% of births. Birth weight was >2500 g for 3.3% of liveborn triplets, g for 24.4%, g for 56.7%, and <1000 g for 15.6%. Some type of birth defect was reported for three infants (2.7%). The information provided on birth defects was limited. Overall, some type of birth defect was reported for 80 infants (2.3%): eight cases of genetic defect (two stillbirths), 18 cases of cardiac defect (three stillbirths and two neonatal deaths), seven cases of limb defect, one case of cleft palate, and 46 cases of other unspecified defects (four stillbirths and five neonatal deaths). The risk of a couple experiencing perinatal death was related to multiple births. Perinatal death of one or more infants occurred in 1.7% of singleton deliveries, 3.5% of twin deliveries, and 8.1% of triplet deliveries (risk ratio, 2.2; 95% confidence interval, ; P¼.001, multiple vs. singleton). The risk of perinatal death of all infants was 1.7%, 1.8%, and 0%, respectively (P¼.90, multiple vs. singleton). By type of ART procedure, perinatal death of one or more infants occurred in 2.4% of deliveries resulting from IVF/ICSI cycles (3.0% in IVF cycles and 2.0% in ICSI cycles), 1.0% from OD cycles, and 1.8% from FET cycles. Effect of Female Age The clinical pregnancy and birth outcomes for women categorized into three age groups are given in Table 2. The mean female age was 35 years in IVF/ICSI and FET cycles and 41 years in OD cycles. The proportion of cycles in women aged 40 years and older was 18% in IVF/ICSI cycles, 16% in FET cycles, and 64% in OD cycles. In IVF/ICSI and FET cycles, the clinical pregnancy and live birth rates declined with female age, especially after age 40 years. In OD cycles, clinical pregnancy and live birth rates were similar in all age groups. The multiple birth rates declined gradually with age in IVF/ICSI cycles. In FET and OD cycles, age had no clear effect on multiple births because in both cases the middle age group had the lowest rate. In IVF/ICSI cycles using the woman s own oocytes, the age-related decline in ART success can be attributed to suboptimal outcomes at several steps in the process. The proportion of started cycles with successful retrieval decreased with age (93.7% for women aged <35 years, 90.7% for those aged years, and 86.6% for those aged R40 years), as did the mean number of oocytes retrieved (12.9, 10.9, and 9.4 respectively). In women who had one or more embryos replaced, the mean implantation rate declined with increasing female age (29.6%, 20.7%, and 11.1%), as did the clinical pregnancy rate (43.8%, 35.8%, and 23.6%), even though older women had more embryos transferred (mean, 2.1, 2.5, and 2.9). The proportion of women who had surplus embryos available for cryopreservation gradually decreased from the younger to older women (49.1%, 33.6%, and 20.9%). In women who achieved clinical intrauterine pregnancy, the pregnancy loss rate became higher as women aged (11.5%, 17.1%, and 31.7%). However, adverse birth outcomes were not found TABLE 2 Clinical pregnancy and birth outcomes by female age for the three most common ART procedures. Outcome/female age group IVF/ICSI IVF/ICSI with OD FET Mean female age, years (range) 35 (19 53) 41 (26 51) 35 (23 52) Cycles started by age, n (% of cycles within procedure): < (44.8) 43 (15.0) 1159 (47.0) (37.2) 59 (20.6) 920 (37.3) R (17.9) 184 (64.3) 389 (15.8) Clinical pregnancy by age, n (% per cycle started): < (38.6) 20 (46.5) 288 (24.8) (30.8) 26 (44.1) 211 (22.9) R (18.7) 87 (47.3) 65 (16.7) Live birth by age, n (% per cycle started) a : < (32.4) 14 (32.6) 229 (19.8) (24.0) 18 (31.6) 160 (17.5) R (12.0) 63 (36.0) 43 (11.1) Multiple birth by age, n (% per delivery) a : < (34.6) 5 (35.7) 67 (28.9) (26.8) 5 (27.8) 28 (17.4) R40 38 (21.6) 22 (34.9) 11 (25.6) a Cycles with unknown delivery status omitted. Gunby. Canadian ART register Fertil Steril Fertility and Sterility â 1725

6 to be related to advanced female age: the risks of preterm birth (36.3%, 33.0%, and 28.9%), very preterm birth (12.1%, 11.8%, and 8.4%), and perinatal death of one or more infants (2.4%, 2.9%, and 0.6%) were similar across age groups. Effect of Infertility Diagnosis In IVF/ICSI cycles, the reason for ART treatment was most commonly a single male (35% of cycles) or a single female (26%) infertility factor. Idiopathic or unexplained infertility was the diagnosis in 19% of cycles. Both female and male infertility factors were diagnosed in 9% of cycles, and more than one female factor in 4%. A diagnosis of other was given in 8% of cycles, so these cycles could not be classified. The clinical pregnancy rate per cycle started was highest when male factor infertility was the only diagnosis (35%) or was present in combination with a female factor (34%). Couples with idiopathic infertility had a clinical pregnancy rate of 31%. The clinical pregnancy rate was lowest in the presence of single (29%) or multiple (21%) female infertility factors without male factor infertility. These differences across diagnostic groups were statistically significant (P<.001). The distribution of primary diagnostic categories was quite different in IVF and ICSI cycles (Table 3). The most common primary diagnosis for couples having IVF was tubal factor, whereas for couples having ICSI it was male factor infertility. The clinical pregnancy rates per successful retrieval were similar across diagnostic categories for IVF (P¼.08) but were more variable for ICSI (P<.001), with the highest rates being achieved in couples with a primary diagnosis of male factor infertility (38%) or endometriosis (37%; Table 3). Effect of Number of Embryos Transferred The number of embryos transferred in IVF/ICSI cycles ranged from one to 13, with a mean of 2.4. A single embryo was transferred in 11% of transfer cycles. More commonly, either two (57% of cycles) or three (23% of cycles) embryos were transferred. More embryos were transferred in older women: the mean age of women receiving four or more embryos (9% of cycles) was 39 years, compared with 37 years for those receiving three embryos and 34 years for those receiving two embryos. Overall, the clinical pregnancy rate was 37.3% per ET procedure. Clinical pregnancy and birth outcomes by number of embryos transferred are shown in Table 4. The clinical pregnancy rate per ET procedure was low when only one embryo was transferred (20.1%). Transferring three or more embryos did not increase the clinical pregnancy rate beyond the high level observed with two embryos (41.8%); indeed, the clinical pregnancy rate declined when more than two embryos were transferred (34.9%). The mean implantation rates followed a similar pattern: 19.8% with one embryo, 28.2% with two embryos, 17.2% with three embryos, 13.3% with four embryos, and 8.2% with five or more embryos. Twenty-five percent of two-embryo transfers were performed on day 5 after oocyte retrieval, with 69% performed on day 3. The clinical pregnancy rates were higher on day 5 (46.0%) than on day 3 (41.4%). In contrast, only 7% of three-embryo transfers and 3% of four-embryo transfers were performed on day 5. Of IVF/ICSI cycles with more than one embryo transferred, the multiple birth rate was similar whether two embryos (33.1%) or three or more embryos (30.8%) were transferred (Table 4). The triplet birth rate was 3.0% when three or more embryos were transferred. The number of thawed embryos transferred in FET cycles ranged from one to eight, with a mean of 2.2. A single embryo was transferred in 19% of cycles, two embryos in 48%, three embryos in 27%, four embryos in 4%, and five or more embryos in 2%. Overall, the clinical pregnancy rate was 25.5% per ET procedure, increasing with number of embryos transferred, up to three (13.4% with one embryo, 26.2% with two TABLE 3 Clinical pregnancy rates per successful retrieval for IVF and ICSI cycles by primary infertility diagnosis category. a IVF ICSI Primary Diagnosis No. of cycles (% of all IVF cycles) No. of pregnancies (% per retrieval) No. of cycles (% of all ICSI cycles) No. of pregnancies (% per retrieval) Male factor 327 (12.4) 98 (30.0) 2268 (57.2) 854 (37.7) Tubal factor 896 (34.0) 278 (31.0) 478 (12.1) 130 (27.2) Idiopathic 674 (25.6) 232 (34.4) 568 (14.3) 191 (33.6) Endometriosis 306 (11.6) 120 (39.2) 252 (6.4) 94 (37.3) Ovulatory disorder 205 (7.8) 71 (34.6) 223 (5.6) 78 (35.0) Other 227 (8.6) 72 (31.7) 174 (4.4) 42 (24.1) a A total of 887 cycles with unreported primary diagnosis was omitted. Gunby. Canadian ART register Fertil Steril Gunby et al. Canadian ART register 2005 Vol. 91, No. 5, May 2009

7 TABLE 4 Clinical pregnancy rate per ET procedure and multiple birth rate per known birth by number of embryos transferred in IVF/ICSI cycles. No. of embryos transferred No. of cycles (% of all ET cycles) a No. of pregnancies (% per ET) No. of births (% of all births) a,b No. of total multiple births (% per birth) No. of triplet births (% per birth) (10.7) 143 (20.1) 110 (5.6) 4 (3.6) (57.3) 1601 (41.8) 1312 (66.3) 434 (33.1) 9 (0.7) (23.1) 557 (36.0) 416 (21.0) 124 (29.8) 12 (2.9) (5.8) 139 (35.6) 101 (5.1) 36 (35.6) 4 (4.0) 5 or more 205 (3.1) 51 (24.9) 41 (2.1) 12 (29.3) 1 (2.4) a A total of 318 cycles with unreported number of embryos transferred was omitted. b A total of 79 cycles with unknown delivery status was omitted. Gunby. Canadian ART register Fertil Steril embryos, 32.0% with three embryos, 30.9% with four embryos, and 18.4% with five or more embryos). The mean implantation rates were more uniform: 13.2% with one embryo, 16.7% with two embryos, 14.4% with three embryos, and 11.3% with four embryos, but only 4.6% with five or more embryos. The multiple birth rate gradually increased with the number of thawed embryos transferred: 2.4% with one embryo, 24.7% with two embryos, 28.8% with three embryos, 30.0% with four embryos, and 33.3% with five or more embryos. The triplet birth rate was 2.8% when three or more embryos were transferred. Effect of Day of ET In IVF/ICSI cycles, ET was performed on day 2 (after oocyte retrieval) in 5% of transfers, day 3 in 74%, day 5 in 18%, and day 6 in 2%. More embryos were transferred to each woman on day 2 and day 3 (mean, 2.8 and 2.4) than on day 5 and day 6 (mean, 2.0 and 1.9). The proportion of cycles with two embryos transferred was 33% on day 2, 53% on day 3, 79% on day 5, and 78% on day 6. The clinical pregnancy rates per ET procedure were 27.4% on day 2, 36.8% on day 3, 43.9% on day 5, and 27.2% on day 6. The mean implantation rates were 14.8%, 22.1%, 31.7%, and 19.2%, respectively. The multiple birth rates were 22.4% on day 2, 30.1% on day 3, 35.6% on day 5, and 24.2% on day 6. Effect of Surplus Embryos The availability of surplus embryos may be an indicator of embryo quality as well as embryo number. Thus, it is interesting to compare clinical pregnancy rates for cycles with and without embryos available for cryopreservation. The clinical pregnancy rate was 29.5% when all available embryos were transferred (61% of transfers) and 49.7% when surplus embryos were frozen (39% of transfers). The mean implantation rates were 17.2% and 33.0%, respectively. In Canada in 2005, a single embryo was transferred by choice in 126 IVF/ICSI cycles (18% of single ETs and 1.9% of all transfer cycles). The clinical pregnancy rate per ET procedure was 46.8% in elective single ETs, compared with 14.3% when only one embryo was available. The clinical pregnancy rate was 37.9% when elective single ET was done on day 3 (52% of transfers) and 58.9% when it was done on day 5 (44% of transfers). Two embryos were transferred by choice in 1858 IVF/ICSI cycles (49% of double ETs and 28% of all transfer cycles). In these cycles, the clinical pregnancy rate was 51.8%, compared with 32.4% when only two embryos were available. The clinical pregnancy rate was 51.5% when elective double ET was done on day 3 (74% of transfers) and 54.5% when it was done on day 5 (25% of transfers). The multiple birth rate was 36.1% with elective double ETand 28.1% when only two embryos were available. Complications and Pregnancy Reduction Complications were reported in 135 IVF/ICSI cycles (2.0% per cycle started). There were 111 cases of ovarian hyperstimulation syndrome (1.6% per cycle started), 45 of which (41%) required hospitalization; 17 complications related to medications (no hospitalization); and seven complications related to procedures (six hospitalizations). No maternal death was reported. Of 993 multiple pregnancies from all types of ART cycles, outcomes with regard to pregnancy reduction (loss of one or more but not all fetuses) were known for 967. Of these, 162 (16.8%) had pregnancy reduction (80% spontaneous, 18% therapeutic, and 2% type unknown) after ultrasonographic confirmation of fetal viability at about 8 weeks gestation. Of 879 pregnancies that were originally twins, reduction to one fetus occurred spontaneously in 12.6% and therapeutically in 0.3%, and loss of the whole pregnancy occurred in 3.6% (including one loss after a therapeutic reduction). Of 81 pregnancies that were originally triplets, reduction to two fetuses occurred spontaneously in 11.1% and therapeutically in 22.2%; reduction to one fetus occurred spontaneously in 9.9%, therapeutically in 6.2%, and type unknown in 2.5%; and loss of the whole pregnancy occurred in 3.7% (including Fertility and Sterility â 1727

8 one loss after a therapeutic reduction). Thus, only 46% of viable triplet pregnancies resulted in a triplet birth. Of seven pregnancies that originally had four or five fetuses, two were reduced spontaneously to two fetuses, three were reduced therapeutically to two fetuses, one was reduced (type unknown) to one fetus, and one pregnancy miscarried. DISCUSSION In this fifth annual publication from the Canadian ART Register, we have presented a report on ART cycles performed in Canada in All 25 ART centers operating in Canada in 2005 contributed to CARTR, representing 100% participation for the third consecutive year. The voluntary involvement of all Canadian centers in CARTR is a reflection of the enthusiastic commitment to the project by members of the IVF Directors Group and the dedication of their staff. CARTR is financed solely by funds generated from a small per-cycle fee charged to the participating centers. These very limited resources do not permit validation of data through auditing, but we are confident of data accuracy. Currently, data are submitted to CARTR in yearly batches. CARTR is working with the newly established Assisted Human Reproduction Agency of Canada to explore ways to improve data collection in future, including Internet-based, real time data submission. For 2005, one center again submitted summary data by type of ART treatment and female age group rather than cycle-specific data because of concerns relating to patient confidentiality. Therefore, cycles from that center had to be omitted from analyses involving detailed characteristics such as infertility diagnosis, day of ET, and number of embryos transferred. The trend over time toward increasing numbers of cycles reported to CARTR and increasing pregnancy and birth rates that was noted for 2002 (2), 2003 (3), and 2004 (4) continued into 2005 (Table 5). Compared with 2004, there was a 3.1% increase in the total number of ART cycles reported to CARTR for 2005, from 11,068 to 11,414. There was a 3.3% increase in the number of pregnancies resulting from all types of ART cycles, from 3332 in 2004 to 3443 in 2005, mainly because of the higher number of cycles performed, since the relative increase in clinical pregnancy rate was only 0.3% (from 30.1% to 30.2% per cycle started). However, there was a 9.4% increase in the number of live TABLE 5 Comparison of cycle outcomes from CARTR for the years Outcome CARTR 2003 (3) CARTR 2004 (4) CARTR 2005 P No. of clinics participating (%) 24 (100) 26 (100) 25 (100) Total no. of ART cycles reported 10,656 11,068 11,414 IVF/ICSI cycles: No. of cycles reported Cycles with ICSI, % <.0001 Cycles in women aged R40 years, % <.0001 Cycles with two or fewer embryos transferred, % Clinical pregnancy rate per cycle, % Live birth rate per cycle, % Singleton live birth rate per cycle, % Multiple delivery rate per delivery, % Triplet or more rate per delivery, % FET cycles: No. of cycles reported Clinical pregnancy rate per cycle, % Live birth rate per cycle, % Singleton live birth rate per cycle, % Multiple delivery rate per delivery, % Triplet or more rate per delivery, % OD cycles: No. of cycles reported Clinical pregnancy rate per cycle, % Live birth rate per cycle, % Singleton live birth rate per cycle, % Multiple delivery rate per delivery, % Triplet or more rate per delivery, % Gunby. Canadian ART register Fertil Steril Gunby et al. Canadian ART register 2005 Vol. 91, No. 5, May 2009

9 births reported, from 2456 in 2004 to 2687 in 2005, attributable to both the higher number of cycles performed and the 5.3% relative increase in live birth rate (from 22.6% to 23.8% per cycle started). The reason for the larger increase in live birth rate relative to the small increase in clinical pregnancy rate is more complete reporting of birth outcomes. In 2005, birth outcomes were missing for only 3.8% of ongoing pregnancies, compared with 8% in 2003 and However, there is still room for improvement in this regard. Although it is encouraging to see improvements in pregnancy, birth, and multiple birth rates from one year to the next, realistically, such differences may be attributable mainly to random variation, especially with the relatively small number of cycles involved in CARTR. For this reason, statistical testing has been done on the trend over 3 years, rather than just the difference between 2004 and However, multiple comparisons may produce some statistically significant results by chance alone. Therefore, the small changes over time described in the following paragraphs should not be overinterpreted. In IVF/ICSI cycles, there was only a small increase in the clinical pregnancy rate in 2005 over the previous year (Table 5; P for trend 0.23 over 3 years). Rates decreased slightly in the <35 years age group and increased slightly in the two older age groups. At 25.6%, the live birth rate per cycle started showed a relative increase of 5.8% over 2004, attributable to lower pregnancy loss rates in the two older age groups and more complete reporting of birth outcomes (P for trend 0.01 over 3 years). The proportion of IVF/ICSI cycles with only one or two embryos transferred continued to show a gradual increase, to 68% of ETs in 2005 (P for trend over 3 years), resulting in a similar gradual increase in the singleton live birth rate (P for trend over 3 years; Table 5). Although in 2003 and 2004 there had been relative decreases of about 10% in the multiple birth rate from the previous year, initiating what was thought to be an encouraging trend, we were disappointed to discover that 2005 showed a 10.8% relative increase (3.0% absolute increase), returning the multiple birth rate to a level only slightly lower than that of However, the low triplet birth rate of the previous 2 years was maintained (Table 5). In FET cycles, there was no change in clinical pregnancy rate from 2004 to 2005 (Table 5). However, the live birth and singleton live birth rates showed small relative increases of 5.5% and 7.4%, respectively. There was a 5.8% relative decrease in the multiple birth rate in 2005 compared with The triplet birth rate increased from 1.0% in 2004 to 1.6% in 2005, but this variability can be attributed to small numbers, as these percentages represent only four and seven triplet births, respectively. None of the outcomes for FET cycles showed any statistically significant change over the period (Table 5). In OD cycles reported to CARTR, there were small relative increases of 3.6% in the clinical pregnancy rate, 4.5% in the live birth rate, and 1.7% in the singleton live birth rate in 2005 over the previous year (Table 5). The multiple birth rate in OD cycles showed a relative increase of 2.5% in 2005, but there was no triplet birth. None of the outcomes for OD cycles showed any statistically significant change over the period (Table 5). The regression of the multiple birth rate in IVF/ICSI cycles can be attributed to a couple of factors. Although the number of elective single ETs continued to increase in 2005, the clinical pregnancy rate achieved (47%) did not match the impressive results of 2004 (58%), and the nonelective single ET pregnancy rate was also slightly lower (14% vs. 16%). Thus, in 2005, a smaller proportion of births (6% vs. 7%) resulted from transfer of a single embryo with its inherent lower risk of multiple birth. Because about two-thirds of births in IVF/ICSI cycles result from transfer of two embryos, the decrease in multiple births seen in 2004 was strongly influenced by a reduced multiple birth rate in this group (28.6%), although no explanation could be given for this finding (4). In 2005, the multiple birth rate in the two-embryo group (33.1%) returned to levels only slightly lower than those seen in 2003 (33.5%), again with no obvious reason for this change other than random variation. One possible explanation is an increase from 2004 to 2005 in the proportion of elective double ETs from 44% to 49% of all two-embryo transfers, 69% of which were performed in women <35 years of age. Women in this age group who have elective double ET represent 19% of all transfers after IVF/ICSI but 28% of clinical pregnancies, 30% of births, and 38% of multiple births. With a multiple birth rate of 39%, this group is clearly the most obvious target for elective single ET. The Society of Obstetricians and Gynecologists of Canada and the CFAS recently published a joint guideline on how many embryos should be transferred after IVF/ICSI (6). Their recommendation was a maximum of two embryos in women <35 years of age, a maximum of three embryos in women years, and a maximum of four embryos in women R40 years, with transfer of fewer embryos being considered in women with high-quality embryos and good prognosis (e.g., first or second ART attempt or previous successful ART cycle). In 2005, 2 years before the publication of these guidelines, Canadian ART centers were already meeting these recommendations in 83% of cycles in women <35 years (10% had transfer of one embryo), 91% of cycles in women years (60% had transfer of one or two embryos), and 92% of cycles in women %40 years (74% had transfer of one to three embryos). In the fall of 2006, the Bertarelli Foundation sponsored a consensus meeting about the issue of multiple gestations in Canada and the contributing role of ART (7). The panel was made up of Canadian experts from several fields, including fertility, neonatalology, and economics. It was clear that ART has been a contributing factor to the increase in multiple births in Canada over the last two decades, although non- ART fertility treatment through ovulation induction has also contributed to an unknown extent. The panel members Fertility and Sterility â 1729

10 agreed that the most desirable outcome of ART treatment is the delivery of a single healthy infant and that the best way to ensure this outcome is through single ET. However, poor understanding of the risks of multiple pregnancy by couples undergoing fertility treatment, their desire to maximize their chance of pregnancy by transferring several embryos, and the lack of government funding for ART treatment work against the widespread acceptance of single ET. The consensus meeting concluded by setting the goal of reducing ART multiple pregnancy rates in Canada by 50%. In summary, for the 2005 reporting year, 100% of Canadian centers again participated voluntarily in a compilation of ART cycles in Canada. Clinical pregnancy rates per cycle started of 32% in IVF/ICSI cycles, 47% in OD cycles, and 23% in FET cycles and live birth rates of 26%, 35%, and 17%, respectively, compare favorably with rates around the world (8, 9). Clinical pregnancy and live birth rates continued to increase in 2005 compared with previous years. Acknowledgments: The Clinical Outcome Reporting System computer program was generously provided to the Canadian Fertility and Andrology Society by the Society for Assisted Reproductive Technology (Birmingham, Alabama). The authors thank the personnel from each center responsible for data entry for CARTR for their hard work and devotion to detail. REFERENCES 1. Gunby J, Daya S. IVF Directors Group of the Canadian Fertility and Andrology Society. Assisted reproductive technologies (ART) in Canada: 2001 results from the Canadian ART Register. Fertil Steril 2005;84: Gunby J, Daya S. IVF Directors Group of the Canadian Fertility and Andrology Society. Assisted reproductive technologies (ART) in Canada: 2002 results from the Canadian ART Register. Fertil Steril 2006;86: Gunby J, Daya S. IVF Directors Group of the Canadian Fertility and Andrology Society. Assisted reproductive technologies (ART) in Canada: 2003 results from the Canadian ART Register. Fertil Steril 2007;88: Gunby J, Bissonnette F, Librach C, Cowan L. IVF Directors Group of the Canadian Fertility and Andrology Society. Assisted reproductive technologies (ART) in Canada: 2004 results from the Canadian ART Register. Fertil Steril 2008;89: Zegers-Hochschild F, Nygren KG, Adamson GD, de Mouzon J, Lancaster P, Mansour R, et al. The International Committee Monitoring Assisted Reproductive Technologies (ICMART) glossary on ART terminology. Fertil Steril 2006;86: Min JK, Claman P, Hughes E. Society of Obstetricians and Gynecologists of Canada, Canadian Fertility and Andrology Society. Guidelines for the number of embryos to transfer following in vitro fertilization. J Obstet Gynaecol Can 2006;28: Bissonnette F, Cohen J, Collins J, Cowan L, Dale S, Dill S, et al. Incidence and complications of multiple gestation in Canada: proceedings of an expert meeting. Reprod Biomed Online 2007;14: Andersen AN, Goossens V, Ferraretti AP, Bhattacharya S, Felberbaum R, Mouzon J, et al. Assisted reproductive technology in Europe, Results generated from European registers by ESHRE. Hum Reprod 2008;23: Centers for Disease Control and Prevention Assisted Reproductive Technology (ART) Report. Available at CANADIAN ART CENTERS REPORTING DATA TO CARTR FOR 2005 Victoria Fertility Center, Victoria, British Columbia University of British Columbia Center for Reproductive Health, Vancouver, British Columbia Genesis Fertility Center, Vancouver, British Columbia Regional Fertility Program, Calgary, Alberta Assisted Reproductive Technology at University of Saskatchewan (ARTUS), Saskatoon, Saskatchewan Heartland Fertility Clinic, Winnipeg, Manitoba London Health Sciences Center, London, Ontario Hamilton Health Sciences Center for Reproductive Care, Hamilton, Ontario ISIS Regional Fertility Center, Mississauga, Ontario Astra Fertility Center, Mississauga, Ontario NewLife Fertility Center, Mississauga, Ontario CReATe IVF Program, Toronto, Ontario LifeQuest Center for Reproductive Medicine, Toronto, Ontario Mt. Sinai Reproductive Biology Unit, Toronto, Ontario Toronto Center for Advanced Reproductive Technology (TCART), Toronto, Ontario IVF Canada and LIFE Program, Scarborough, Ontario Markham Fertility Center, Markham, Ontario The Fertility Center at the Ottawa Hospital, Ottawa, Ontario McGill University Reproductive Center, Montreal, Quebec Montreal Fertility Clinic, Montreal, Quebec OVO Fertility Clinic, Montreal, Quebec Procrea, Montreal, Quebec Procrea, Quebec, Quebec Conceptia Clinic, Moncton, New Brunswick Atlantic Assisted Reproductive Therapies (AART), Halifax, Nova Scotia 1730 Gunby et al. Canadian ART register 2005 Vol. 91, No. 5, May 2009