Study design

This study was a population-based retrospective registry study carried out in Sweden. We collected data from all IVF clinics, both public and private, in Sweden on IVF treatments performed from 2002 through 2013. The IVF clinics reported data on deliveries to the Swedish National Board of Health and Welfare for years 2002 through 2006, and for years 2007 through 2013, to the National Registry of Assisted Reproduction. All reported IVF singleton and twin deliveries with autologous oocytes were included in the study and divided into fresh blastocyst, frozen blastocyst, fresh cleavage-stage, and frozen cleavage-stage transfers. Using the unique personal identification number given to all citizens in Sweden we cross-linked data on deliveries after IVF with the Swedish Medical Birth Register (MBR), the Register of Birth Defects (formerly Register of Congenital Malformations), and the National Patient Register (formerly the Hospital Discharge Register) and collected information about maternal characteristics, pregnancy, delivery, and the neonatal period.

Data on deliveries included all live births and stillbirths. Before July 1, 2008, stillbirths ≥28 completed gestational weeks were included and from July 1, 2008, stillbirths ≥22 completed gestational weeks were included.

Data from the IVF registers included information about treatment, ie, blastocyst or cleavage-stage transfer, fresh or frozen-thawed embryo transfers, number of oocytes retrieved, number of embryos transferred, date of embryo transfer, singleton or twin gestation, and date of delivery. Data from MBR included maternal characteristics (ie, age, parity, body mass index [BMI], smoking habits, years of involuntary infertility) and data on delivery and neonatal outcomes. The MBR has been found to have high validity and covers almost all deliveries in Sweden since 1973.

Typically, blastocyst transfers were performed on days 5-6 and cleavage-stage transfers on days 2-3. Cryopreservation methods included both vitrification and slow freezing for blastocysts and slow freezing for cleavage-stage embryos.

Singleton deliveries after blastocyst transfers were compared with deliveries after cleavage-stage transfers and with deliveries after spontaneous conception (n = 1,196,394).

Outcomes

For singletons, the main neonatal outcome was birth defects. Other neonatal outcome measures were PTB (<37 weeks, <32 weeks [very PTB], <28 weeks), LBW (<2500 g), very LBW (<1500 g), small for gestational age (SGA) (<–2SD), large for gestational age (LGA) (>+2SD, >+3SD), macrosomia (>4500 g), low Apgar score (<7 at 5 minutes), perinatal mortality (defined as stillbirths and early neonatal deaths within 7 days after birth), child’s sex, and rate of cesarean delivery. The Swedish gender- and gestational age–specific growth standard was used as a reference for SGA and LGA. We also analyzed any composite neonatal adverse outcome (defined as any of PTB, LBW, SGA, or perinatal mortality) and severe composite adverse neonatal outcome (defined as any of very PTB, very LBW, or perinatal mortality).

Birth defects were defined according to 10th revision of the International Statistical Classification of Diseases and Related Health Problems ( ICD-10 ) codes beginning with Q. Minor birth defects such as preauricular appendix, tongue tie, patent ductus arteriosus in preterm infants, single umbilical artery, undescended testicle, hip luxation/subluxation, clicking hip, and nevus were excluded. Subanalyses were made within 6 different groups of birth defects: cardiac defects, neural tube defects, orofacial clefts, gastrointestinal defects, hypospadias, and limb reduction defects. Information on birth defects were collected from 3 sources: the MBR ( International Classification of Diseases codes given at the pediatric examination of the newborn), the Register of Birth Defects, and the National Patient Register. The Register of Birth Defects and the National Patient Register supplied information on birth defects during the infant’s first year of life.

Gestational age was determined from the second-trimester ultrasonography for the majority of spontaneous conception pregnancies, or from the first day of the last menstrual period if information from ultrasonography was not available. For IVF pregnancies gestational age was calculated from the day of embryo transfer, ie, day 5 or 6 for blastocyst and day 2 or 3 for cleavage-stage embryos. If the information was not available, gestational age was calculated from ultrasonography.

The maternal outcomes in singleton pregnancies included placenta previa ( ICD-10 code O44), placental abruption ( ICD-10 code O45), preeclampsia/eclampsia ( ICD-10 codes O14 or O15), preterm prelabor rupture of the membranes ( ICD-10 code O42), gestational diabetes ( ICD-10 code O244), and postpartum hemorrhage ( ICD-10 code O72; >1000 mL).

Outcomes for multiples included only birth defects.

Ethics

Permission for this study was given from the Regional Ethical Committee at the University of Gothenburg (Dnr 304/06, T876-14).

Statistical analysis

Descriptive data are given as numbers and percentages. Statistical analysis was performed on neonatal and maternal outcomes for singleton pregnancies after blastocyst transfer vs singleton pregnancies after cleavage-stage transfer and singleton pregnancies after spontaneous conception. Multivariable logistic regression analysis was used and adjustments were made for year of birth (entered as a continuous linear variable), maternal age (second-degree polynomial), parity (class variable: parity 1, 2-3, ≥4), maternal smoking (ordinal: 1 = no smoking, 2 = smoking 1-9 cigarettes per day, 3 = smoking ≥10 cigarettes per day), maternal BMI (continuous linear variable), years of involuntary childlessness (ordinal: 0 = no, 1 = 1-2 years, 2 = 3-4 years, 3 = ≥5 years), and child’s sex. Missing data on maternal smoking and BMI were replaced by the overall mean. Values not reported on years of involuntary childlessness were set at 0. When blastocyst and cleavage-stage transfer were compared, adjustment was also made for the number of oocytes retrieved, the number of embryos transferred, and for fresh and frozen transfers.

In a subanalysis of birth defects adjustment was also made for IVF/intracytoplasmic sperm injection (ICSI) for the time period 2008 through 2013. The reason for this limitation was that information on fertilization method was not available for the whole time period for the blastocyst group.

Crude odds ratios (OR) and adjusted ORs (AOR) with 95% confidence interval (CI) were calculated using software (Gauss; Aptech Systems Inc, Maple Valley, WA).

Sample size calculation

Power analyses (80% power) revealed that it was possible to detect a difference in the rate of birth defects of 0.9%, from 4.0-4.9% with 4800 singletons in the blastocyst group and 25,000 singletons in the cleavage-stage group.

Study design

This study was a population-based retrospective registry study carried out in Sweden. We collected data from all IVF clinics, both public and private, in Sweden on IVF treatments performed from 2002 through 2013. The IVF clinics reported data on deliveries to the Swedish National Board of Health and Welfare for years 2002 through 2006, and for years 2007 through 2013, to the National Registry of Assisted Reproduction. All reported IVF singleton and twin deliveries with autologous oocytes were included in the study and divided into fresh blastocyst, frozen blastocyst, fresh cleavage-stage, and frozen cleavage-stage transfers. Using the unique personal identification number given to all citizens in Sweden we cross-linked data on deliveries after IVF with the Swedish Medical Birth Register (MBR), the Register of Birth Defects (formerly Register of Congenital Malformations), and the National Patient Register (formerly the Hospital Discharge Register) and collected information about maternal characteristics, pregnancy, delivery, and the neonatal period.

Data on deliveries included all live births and stillbirths. Before July 1, 2008, stillbirths ≥28 completed gestational weeks were included and from July 1, 2008, stillbirths ≥22 completed gestational weeks were included.

Data from the IVF registers included information about treatment, ie, blastocyst or cleavage-stage transfer, fresh or frozen-thawed embryo transfers, number of oocytes retrieved, number of embryos transferred, date of embryo transfer, singleton or twin gestation, and date of delivery. Data from MBR included maternal characteristics (ie, age, parity, body mass index [BMI], smoking habits, years of involuntary infertility) and data on delivery and neonatal outcomes. The MBR has been found to have high validity and covers almost all deliveries in Sweden since 1973.

Typically, blastocyst transfers were performed on days 5-6 and cleavage-stage transfers on days 2-3. Cryopreservation methods included both vitrification and slow freezing for blastocysts and slow freezing for cleavage-stage embryos.

Singleton deliveries after blastocyst transfers were compared with deliveries after cleavage-stage transfers and with deliveries after spontaneous conception (n = 1,196,394).

Outcomes

For singletons, the main neonatal outcome was birth defects. Other neonatal outcome measures were PTB (<37 weeks, <32 weeks [very PTB], <28 weeks), LBW (<2500 g), very LBW (<1500 g), small for gestational age (SGA) (<–2SD), large for gestational age (LGA) (>+2SD, >+3SD), macrosomia (>4500 g), low Apgar score (<7 at 5 minutes), perinatal mortality (defined as stillbirths and early neonatal deaths within 7 days after birth), child’s sex, and rate of cesarean delivery. The Swedish gender- and gestational age–specific growth standard was used as a reference for SGA and LGA. We also analyzed any composite neonatal adverse outcome (defined as any of PTB, LBW, SGA, or perinatal mortality) and severe composite adverse neonatal outcome (defined as any of very PTB, very LBW, or perinatal mortality).

Birth defects were defined according to 10th revision of the International Statistical Classification of Diseases and Related Health Problems ( ICD-10 ) codes beginning with Q. Minor birth defects such as preauricular appendix, tongue tie, patent ductus arteriosus in preterm infants, single umbilical artery, undescended testicle, hip luxation/subluxation, clicking hip, and nevus were excluded. Subanalyses were made within 6 different groups of birth defects: cardiac defects, neural tube defects, orofacial clefts, gastrointestinal defects, hypospadias, and limb reduction defects. Information on birth defects were collected from 3 sources: the MBR ( International Classification of Diseases codes given at the pediatric examination of the newborn), the Register of Birth Defects, and the National Patient Register. The Register of Birth Defects and the National Patient Register supplied information on birth defects during the infant’s first year of life.

Gestational age was determined from the second-trimester ultrasonography for the majority of spontaneous conception pregnancies, or from the first day of the last menstrual period if information from ultrasonography was not available. For IVF pregnancies gestational age was calculated from the day of embryo transfer, ie, day 5 or 6 for blastocyst and day 2 or 3 for cleavage-stage embryos. If the information was not available, gestational age was calculated from ultrasonography.

The maternal outcomes in singleton pregnancies included placenta previa ( ICD-10 code O44), placental abruption ( ICD-10 code O45), preeclampsia/eclampsia ( ICD-10 codes O14 or O15), preterm prelabor rupture of the membranes ( ICD-10 code O42), gestational diabetes ( ICD-10 code O244), and postpartum hemorrhage ( ICD-10 code O72; >1000 mL).

Outcomes for multiples included only birth defects.

Ethics

Permission for this study was given from the Regional Ethical Committee at the University of Gothenburg (Dnr 304/06, T876-14).

Statistical analysis

Descriptive data are given as numbers and percentages. Statistical analysis was performed on neonatal and maternal outcomes for singleton pregnancies after blastocyst transfer vs singleton pregnancies after cleavage-stage transfer and singleton pregnancies after spontaneous conception. Multivariable logistic regression analysis was used and adjustments were made for year of birth (entered as a continuous linear variable), maternal age (second-degree polynomial), parity (class variable: parity 1, 2-3, ≥4), maternal smoking (ordinal: 1 = no smoking, 2 = smoking 1-9 cigarettes per day, 3 = smoking ≥10 cigarettes per day), maternal BMI (continuous linear variable), years of involuntary childlessness (ordinal: 0 = no, 1 = 1-2 years, 2 = 3-4 years, 3 = ≥5 years), and child’s sex. Missing data on maternal smoking and BMI were replaced by the overall mean. Values not reported on years of involuntary childlessness were set at 0. When blastocyst and cleavage-stage transfer were compared, adjustment was also made for the number of oocytes retrieved, the number of embryos transferred, and for fresh and frozen transfers.

In a subanalysis of birth defects adjustment was also made for IVF/intracytoplasmic sperm injection (ICSI) for the time period 2008 through 2013. The reason for this limitation was that information on fertilization method was not available for the whole time period for the blastocyst group.

Crude odds ratios (OR) and adjusted ORs (AOR) with 95% confidence interval (CI) were calculated using software (Gauss; Aptech Systems Inc, Maple Valley, WA).

Sample size calculation

Power analyses (80% power) revealed that it was possible to detect a difference in the rate of birth defects of 0.9%, from 4.0-4.9% with 4800 singletons in the blastocyst group and 25,000 singletons in the cleavage-stage group.