Birth Outcomes by Infertility Treatment: Analyses of the Population-Based Cohort: Massachusetts Outcomes Study of Assisted Reproductive Technologies (MOSART).

The outcomes of pregnancies conceived through assisted reproductive technology (ART) have been reported to be of lower birthweight and shorter gestation, even when limited to singleton births ( 1 – 5 ). It is unknown whether these decrements are due to parental characteristics or aspects of the ART treatment: this remains a primary challenge to infertility research ( 6 – 11 ). In addition, an acknowledged drawback of prior ART research in the US has been the self-reported nature of the outcomes data, which is typically reported by the patient herself or by her obstetrical provider. This study seeks to overcome these limitations by linking the Society for Assisted Reproductive Technology Clinic Outcomes Reporting System (SART CORS) data to the birth certificate and hospital utilization data. This is the second of a pair of analyses evaluating the effect of ART diagnoses and treatment parameters on the course and outcome of pregnancy ( 12 ). This within-ART set of analyses is part of a larger population-based study of ART in Massachusetts ( 12 – 17 ). The objective of this current analysis is to evaluate the contribution of specific ART treatment parameters to compromised maternal and child health outcomes. The health outcomes of interest include: plurality at birth, pregnancy hypertension, gestational diabetes, primary cesarean delivery, prematurity, low birthweight, small-for-gestation age and large-for-gestation birthweight, and birth defects.

Of the 9,092 linked pregnancies, there were 60 fetal deaths and 84 higher-order births (triplets and quadruplets); these pregnancies were excluded. The 8,948 ART cycles which resulted in live births (6,526 singleton pregnancies and 2,422 twin pregnancies of at least 20 weeks gestation and birthweights of at least 350 grams) were categorized by six ART treatment parameters: oocyte source (autologous versus donor), semen source (autologous versus donor), intracytoplasmic sperm injection (ICSI, no versus yes), assisted hatching (AZH, no versus yes), embryo state (fresh versus thawed), and number of embryos transferred (one versus two or ≥three). In addition, the ART factors of number of fetal heartbeats at the six-week ultrasound (one versus two or ≥three) and plurality at birth (singleton versus twins) were also evaluated. The descriptive statistics of the 8,948 ART pregnancies which resulted in live births by maternal and paternal characteristics are shown in Table I , by maternal health and reproductive history in Table II , and by pregnancy and birth outcomes in Table III (singletons) and Table IV (twins). The study population did not differ substantially by treatment parameters: more than 85% of mothers and fathers (with the exception of fathers using donor semen) were white non-Hispanic; 70% or more of both parents were college-educated ( Table I ). The use of donor gametes was more likely with older age: 66% of women over age 40 used donor oocytes and 33% of men over age 40 used donor semen. Assisted hatching was also more frequently used with older age: 22% of women and 33% of men over age 40. Number of embryos transferred also increased with older age. Among women ages 40 and older, 25% had three or more embryos transferred, compared to 10% having one embryo transferred and 11% having two embryos transferred. ART treatment parameters differed significantly by infertility diagnoses ( Table II ). Male factor was the diagnosis in about 33% of pregnancies, and was more likely to use donor semen (29.7%), ICSI (63.5%), and assisted hatching (44%). Women with the diagnosis of diminished ovarian reserve were more likely to use donor oocytes (50.4%) and donor semen (11.3%). Cleavage-stage embryos (day 2–3) were significantly more likely to have assisted hatching compared to blastocyst-stage embryos (34.3% vs 4.7%), and significantly more embryos were transferred. During the period of the study (2004–08), there were increases in the proportions of blastocyst-stage embryos transferred (from 6.2% to 18.6%, p<0.0001) and frozen cycles (11.3% to 12.8%, p =0.08). Although significantly more embryos were transferred in cycles using cleavage-stage compared to blastocyst-stage embryos ( Table II ), plurality did not differ substantially at either the six weeks ultrasound (66.1% and 68.2% had one fetal heartbeat, respectively, p = 0.14) or at birth (72.0% and 72.8% were singletons, p = 0.89). The pregnancy and birth outcomes for the 6,526 singleton pregnancies by ART treatment parameters are shown in Table III . More than 8% of pregnancies which resulted in singletons at birth were a higher plurality at six weeks gestation; this was significantly associated with the use of assisted hatching and number of embryos transferred. Pregnancy hypertension was more likely in pregnancies with donor oocytes (23.3%), donor semen (17.2%), and thawed embryos (16.0%). Vaginal delivery was less likely with donor oocytes (36.9%), assisted hatching (48.6%), thawed embryos (43.9%), and declined with increasing number of embryos transferred. Mean length of gestation was shorter and proportions of prematurity higher with donor oocytes and thawed embryos. LBW did not differ by ART treatment. Mean birthweight z-score and proportions of SGA and LGA did not differ significantly by ART treatment parameter, except embryo state: pregnancies from thawed embryos averaged significantly higher birthweight z-scores, lower proportion of SGA and higher proportion of LGA compared to pregnancies from fresh embryos. The rate of birth defects among singletons was 2.0%, and did not differ significantly by ART treatment parameter. The pregnancy and birth outcomes for the 2,422 twin pregnancies by ART treatment parameters were similar to those of singletons, and are shown in Table IV . Nearly 6% of pregnancies that resulted in twins at birth were a higher plurality at six weeks gestation (indicating fetal loss) and about 0.5% were singletons (indicating embryo splitting or monozygosity, or misreading of initial ultrasound); this was significantly associated with the use of assisted hatching and a greater number of embryos transferred. Pregnancy hypertension was more likely in pregnancies with donor oocytes (33.0%) and thawed embryos (32.3%). Vaginal delivery declined with increasing number of embryos transferred. Mean length of gestation and birthweight, and proportions of LBW, SGA, and LGA did not differ by treatment parameter, with the exception of embryo state: pregnancies from thawed embryos averaged significantly higher birthweight z-scores and lower proportions of SGA than pregnancies from fresh embryos. The rate of birth defects was 4.9%, and did not differ significantly by ART treatment parameter. The risks of adverse pregnancy and birth outcomes by plurality at birth and ART treatment parameters are shown in Table V . Twin versus singleton plurality at birth was associated with increased risks for all adverse outcomes (except LGA, for which there was a significantly decreased risk), ranging from 1.26 for gestational diabetes to 11.77 for prematurity. Adjustment for maternal and paternal demographic factors, ART diagnoses and treatment parameters, and maternal preexisting medical conditions increased the risks only slightly; additional adjustment for gestational diabetes and pregnancy hypertension attenuated the risks, but did not change them substantially. In contrast, the risks for adverse outcomes associated with ART treatment parameters were fewer and of less magnitude; no risks were associated with semen source, assisted hatching, or number of embryos transferred. For oocyte source, the use of donor oocytes was associated with increased risks for pregnancy hypertension, primary cesarean delivery, and prematurity; the use of ICSI with a greater risk for SGA; and fetal heartbeats >one at six weeks’ gestation with prematurity, LBW, and SGA. For embryo state, the use of thawed embryos was associated with an increased risk for pregnancy hypertension, but significantly lower risks for LBW and SGA. Gestational diabetes increased the risk for primary cesarean delivery, LGA, and birth defects; pregnancy hypertension was associated with greater risks for primary cesarean delivery, prematurity, and LBW.

This analysis indicates that plurality is the predominant ART treatment factor associated with adverse pregnancy and birth outcomes, with other treatment factors having much less or no effect. These findings are in accord with prior studies ( 22 ) and reviews ( 10 ), which have also concluded that the ART treatments have little or no effect on perinatal outcomes, except for the greater proportion of multiple births. This analysis found that three ART treatment factors did not have any significant effect on pregnancy and birth outcomes: semen source, assisted hatching, and number of embryos transferred; a fourth parameter, ICSI, was associated with a 20% increase in SGA which was of borderline significance. The use of frozen embryo transfer (FET) was associated with a significantly increased risk of pregnancy hypertension, but lower risks for LBW and SGA, a finding consistently reported in prior studies ( 23 – 27 ). We also found a higher proportion of LGA birthweight outcomes with the use of FETs in the univariate analyses (15.6% among singletons), a finding similar to the rates reported by Kato et al (19.0%) ( 27 ), but this relationship was not significant after adjusting for other factors. The use of FET has increased by more than 80% since 2006 due to better cryopreservation techniques, improved live birth rates, lower risk of ectopic pregnancies, and more physiologically normal hormonal and endometrial environments ( 27 – 33 ). Our findings on the effect of ICSI are similar to those reported by Katalinic et al ( 34 ), with comparable mean length of gestation and birthweights by plurality. Although the number of embryos transferred differed by maternal characteristics, it did not have a significant effect on the risk for adverse outcomes, in either the univariate or multivariate analyses. We have previously demonstrated, with a sample size of 23,645 singleton pregnancies and 14,083 twin pregnancies, that even when plurality at six weeks’ gestation is the same as plurality at birth, transferring a higher number of embryos has an increasing detrimental effect on fetal growth and birthweight-for-age ( 35 ). With the addition of more years of data to our current study, this relationship will be explored further. The effect of fetal loss reflects prior reports in both singletons ( 36 ) and twins ( 37 ) of a progressively increased adverse effect of higher plurality at six-weeks’ gestation than at birth. Other studies have demonstrated that when the fetal loss occurs later in gestation, the surviving child is at greater risk for severe neurodevelopmental consequences, including cerebral palsy ( 38 – 40 ). It has been suggested that the higher rates of cerebral palsy among children born from assisted conception might be an outcome of transferring more than a single embryo, occurring in a greater frequency with the early intrauterine death of an unrecognized twin ( 40 ). The MOSART study, which includes linking ART cycles to the vital records and hospital utilization data, represents the first time these datasets have been linked using direct identifiers from both datasets. ART national surveillance summaries are limited to birth outcomes reported by the patient herself or her obstetric provider ( 41 – 43 ). Prior studies ( 41 , 42 ) have relied on linkages between ART cycles and vital records using only maternal and infant dates of birth, or probabilistic algorithms ( 43 ). Although there is a high degree of comparability between the SART CORS and vital records ( 44 ), our study design assures more accurate linkage between ART treatment cycles, vital records, and the hospital discharge birth data, and a more complete picture of perinatal outcomes. Although this study has several unique advantages over prior ART research, it is also subject to a number of limitations. This study uses retrospective data from several centralized datasets and although this is advantageous to achieve large numbers, we had the disadvantage that data entered into the SART CORS system is not as rigorously controlled as data collected for a prospective research study. Likewise, the primary purpose of vital records is civil registration, with public health research and surveillance being a secondary use. One of the limitations of comparing our results to the published literature is that the latter is often based on data spanning decades, during which time both ART procedures and outcomes have improved. Although we utilized two sources for identifying pregnancy hypertension (hospital discharge records and the birth certificate), this outcome is subject to misclassification and inconsistent definitions, therefore the associations we report in this study should be interpreted with caution. Another limitation of this analysis is that it only includes women in Massachusetts. There may be significant demographic and outcome differences in patients in other regions of the country and with other healthcare systems, potentially limiting the generalizability of our findings. The addition of the remaining five years of this study (2009–13) will permit new analyses for higher-order multiple births (triplets and quadruplets), as well as by day of embryo transfer, stimulation protocols, and method of semen collection, factors for which we currently have small sample sizes. In summary, this analysis demonstrates that the major ART treatment risk factor for adverse pregnancy and birth outcomes is higher plurality. Other factors had smaller and less consistent effects on outcomes.

This retrospective cohort study included all live births in Massachusetts from July 1, 2004 through December 31, 2008 that linked to ART cycles in the Society for Assisted Reproductive Technology Clinic Online Reporting System (SART CORS) and the Pregnancy to Early Life (PELL) data system. Functioning within the Massachusetts Department of Public Health, the PELL system links records from birth and fetal death certificates, hospital discharges, and program data from child health and development programs. The PELL system, which operates within the Massachusetts Department of Public Health, links records from birth and fetal death certificates, hospital discharges, and program data from child health and development programs. The PELL data system has linked information on more than 99% of all births and fetal deaths in Massachusetts from 1998–2008 to corresponding hospital utilization data (hospital admissions, observational stays, and emergency room visits) for individual women and their children. There were a total of 860,654 linked deliveries through 2008. The Massachusetts Department of Public Health (MDPH) and the Massachusetts Center for Health Information and Analysis are the custodians of the PELL data. PELL is a relational data system composed of individual databases linked together by randomly-generated unique IDs for mother and infant. The PELL data system is housed at MDPH. The data source for ART data for this study was the SART CORS, which contains comprehensive data from more than 90% of all clinics performing ART in the US. Data were collected and verified by SART and reported to the Centers for Disease Control and Prevention in compliance with the Fertility Clinic Success Rate and Certification Act of 1992 (Public Law 102–493). SART maintains HIPAA-compliant business associates agreements with reporting clinics. In 2004, following a contract change with CDC, SART gained access to the SART CORS data system for the purposes of conducting research. The national SART CORS database for 2004–08 contains 642,927 ART treatment cycles. The database includes information on demographic factors (age, race/ethnicity); ART factors (infertility diagnoses, oocyte source and state, use of micromanipulation, number of embryos transferred); treatment outcomes (number of fetal heart beats on early ultrasound, early pregnancy loss); and pregnancy outcomes (live born, stillborn, length of gestation, plurality, and genders). The data in the SART CORS are validated annually ( 21 ) with some clinics having on-site visits for chart review based on an algorithm for clinic selection. During each visit, data reported by the clinic were compared with information recorded in patients’ charts. In 2012, records for 2,045 cycles at 35 clinics were randomly selected for full validation, along with 238 egg/embryo banking cycles ( 21 ). The full validation included review of 1,318 cycles for which a pregnancy was reported. Among the non-donor cycles, 331 were multiple-fetus pregnancies. Ten out of 11 data fields selected for validation were found to have discrepancy rates of ≤5%. The exception was the diagnosis field, which, depending on the diagnosis, had a discrepancy rate between 2.1% and 9.2%. The Massachusetts Outcome Study of Assisted Reproductive Technology (MOSART) project links data from the SART CORS with the PELL data system to evaluate pregnancy and child health outcomes on a population basis. A Memorandum of Understanding was executed between SART and the three entities that participate in the PELL project: Boston University, the Massachusetts Department of Public Health, and the Centers for Disease Control and Prevention. Human subjects approval was obtained from all entities and participating Universities. The study had the approval of the SART Research Committee. We constructed the MOSART database by linking the SART CORS and PELL data systems for all children born in Massachusetts hospitals to Massachusetts resident women between July 1, 2004 and December 31, 2008. The starting date was chosen based on the availability of SART CORS data (January 1, 2004) to allow us to capture any births associated with ART and the end date reflected the latest available data from both SART and PELL when we began the MOSART study. PELL data from July 1, 2004 and December 31, 2008 included 282,971 women with 334,152 deliveries resulting in 342,035 live births and fetal deaths which were linked to 42,649 ART cycles among 18,439 women from SART CORS. A deterministic five phase linkage algorithm methodology was implemented ( 13 ). ART deliveries for women with treatment cycles between January 1, 2004 and December 31, 2008 and that had either a Massachusetts patient zip code or in which the treatment clinic was located in Massachusetts were obtained from SART. The linkage was done by conveying to PELL a file containing patient identifiers but no cycle specific data. Data for 9,092 ART cycles resulting in deliveries were linked to PELL birth or fetal death certificates using mother’s first and last name, mother’s date of birth, father’s name, race of both parents, date of delivery, and number of babies born per delivery. Linked files were later de-identified by use of a linkage ID from which identifiers was removed. Methods for linkage have been described previously ( 13 ). The linkage rate was 89.7% overall and 95.0% for deliveries in which both zip code and clinic were located in MA. Independent variables included parental ages, race/ethnicity, and education; maternal pre-pregnancy medical conditions (chronic hypertension and diabetes mellitus); infertility diagnoses, plurality at the six-week ultrasound and plurality at birth; oocyte and semen sources; the use of ICSI and assisted hatching; day of transfer (for fresh, autologous cycles only), embryo state (fresh or thawed), and number of embryos transferred. Dependent variables included maternal morbidity (pregnancy hypertension, gestational diabetes, primary cesarean delivery) and infant morbidity (prematurity, low birth weight, and small-for-gestational age and large-for-gestational age birthweight, and birth defects). Parental ages at delivery were obtained from the birth certificates in PELL. Parental age was evaluated as both continuous and categorical variables (≤30, 31–34, 35–37, 38–40, 41–42, and ≥43). Parental race/ethnicity was also obtained from the birth certificate, and categorized as white, black, Asian, Hispanic, and other. Parental education was also obtained from the birth certificate and categorized as ≤high school or GED (General Education Development diploma), some college or Associate degree, or Bachelor degree or graduate school. Preexisting maternal medical diagnoses were identified in PELL from either the birth certificate or the hospital discharge delivery record (ICD-9 codes of 648.0 or 250 for diabetes mellitus; 401, 402, 403, 404, or 405 for chronic hypertension). ART treatment parameters were obtained from the SART CORS database, including diagnoses (male factor, endometriosis, ovulation disorders, tubal factors, uterine factors, other factors, and unexplained); oocyte and semen sources (autologous or donor); use of ICSI and assisted hatching (no or yes), number of embryos transferred (mean and standard deviation, and as 1, 2, ≥3), number of fetal heartbeats at the six-week ultrasound (1, 2, ≥3), and embryo state (fresh or thawed). Plurality at the six-week ultrasound was obtained from the SART CORS database and plurality at birth from the birth certificate. Length of gestation was calculated by using the SART CORS outcome date minus date of transfer and adding 17 days and the cycle day of transfer, effectively the outcome date minus the date of conception (or fertilization) plus 14 days. Deliveries prior to 37 completed weeks gestation were classified as premature and those which were 37 weeks or greater were classified as term. Birthweight on each liveborn infant was obtained from the birth certificate. Birthweights at each gestational age are normally distributed, and a z-score (or standard deviation score) is the deviation of the value for an individual from the mean value of the reference population divided by the standard deviation for the reference population ( 18 ). Birthweight z-scores were calculated to evaluate adequacy of weight-for-age using population-based standards, as recommended by Land ( 19 ) and modeled as continuous and categorical variables. We generated gender-, race/ethnicity-, and gestation-specific birthweight means and standard deviations using Massachusetts data for all live births from 1998–2008. Infants with z-scores of ≤1.28 (below the 10 th percentile for gestation) were classified as small-for-gestational age (SGA). Infants with z-scores of ≥1.28 (above the 10 th percentile for gestation) were classified as large-for-gestational age (LGA). Birthweights which were less than 2,500 grams were classified as low birthweight (LBW). In twin pregnancies, the first twin of each pair was classified as LBW or SGA as indicated above. Birth defects included major structural birth defects and chromosomal anomalies of medical, surgical, or cosmetic significance. Birth defects cases were ascertained through the Birth Defects Monitoring Program (BDMP) at the Massachusetts Department of Public Health (BD Reference) ( 20 ). Since 1999, the BDMP has conducted statewide, population-based active surveillance of birth defects among Massachusetts residents. The BDMP identified cases with structural birth defects diagnosed through one year of age from multiple sources, including delivery and specialty care hospitals, birthing centers, and vital records. Potential birth defect cases are assigned to trained abstractors who review maternal and infant medical records. All cases are coded according to the International Classification of Diseases, Ninth Revision, Clinical Modification, modified British Pediatric Association (ICD-9-CM/BPA) system (codes 740–759). Complex cases, cases with syndromes, and cases in which the infant died are reviewed by a clinical geneticist. Twin pregnancies were classified as having a birth defect when one or both infants were affected. We compared parental characteristics and pregnancy and birth outcomes by the six ART treatment parameters (oocyte source, semen source, use of ICSI and assisted hatching, embryo state, and number of embryos transferred) using analysis of variance for continuous variables and χ 2 for categorical variables, stratified by plurality at birth (singleton, twin). The association between the ART treatment parameter, as well as number of fetal heartbeats at six-weeks’ gestation and plurality at birth, and the risk of each of the eight adverse pregnancy and birth outcomes were computed as odds ratios and 95% confidence intervals from multivariate logistic regression models, adjusting for maternal and paternal demographic factors (age, race and ethnicity, education); ART diagnoses and treatment parameters; maternal preexisting medical conditions (chronic hypertension and other diabetes); and plurality at birth. The models of the six ART treatment parameters and fetal heartbeats at six weeks gestation were also initially adjusted for maternal pregnancy morbidities (pregnancy hypertension and gestational diabetes), but because the odds ratios and confidence intervals did not change substantially, these factors were not retained in the final models (results not shown). Because of the known greater risk of adverse outcomes with higher plurality at birth, three models were presented for the outcome of twin versus singleton birth: unadjusted; adjusted as described above; and additionally adjusted for gestational diabetes and pregnancy hypertension. In addition, because gestational diabetes and pregnancy hypertension may also be viewed as both an adverse outcome and as a factor in the causal pathway to the other adverse outcomes, they were modeled both ways. Results were considered significant with p values <0.05 for univariate analyses, and when the 95% confidence intervals did not include 1. All analyses were performed using the Statistical Package for the Social Sciences, version 19.0 (IBM SPSS, Inc., Chicago, IL, USA, 2010).