Risks of Placental Abruption and Preterm Delivery in Patients Undergoing Assisted Reproduction

This cross-sectional study examines the risks of preterm delivery associated with assisted reproduction and placental abruption alone or in combination.


Introduction
Worldwide, assisted reproductive technology (ART) is being increasingly used for infertility treatment.The Centers for Disease Control and Prevention reported that 238 126 patients underwent 413 776 ART cycles in the US in 2021, resulting in 91 906 live births. 1 As delayed childbearing and improved access to ART become more common, these numbers will likely increase.
Assisted reproductive technology includes in vitro fertilization, cryopreservation of gametes or zygotes, intracytoplasmic sperm injection (ICSI), frozen embryo transfer (FET) cycles, and preimplantation genetic testing.7][8][9][10] Placental abruption is the premature detachment of the placenta from the uterine wall before delivery. 11,12In one study, singleton pregnancies conceived with ART experienced an 83% increase in abruption compared with spontaneously conceived pregnancies. 13In another study, patients who underwent ovulation induction or used ART were 2.4 times more likely to have abruption. 14Women with abruption have an increased risk of preterm labor compared with those with premature rupture of membranes or a medical indication for preterm delivery. 15Perinatal mortality risks among births with or without placental abruption were 102.7 and 6.2 per 1000 births, respectively, which further increased at earlier gestational ages. 16Given how ART, placental abruption, and preterm delivery affect neonatal morbidity and mortality, there is limited literature on the association among these 3 entities.
The use of ART can introduce processes that are not characteristic of in vivo reproduction.For example, ART may increase the risk of placental abnormalities due to oxidative, thermal, and mechanical stresses on molecules already undergoing epigenetic remodeling. 17While the exact pathophysiologic mechanisms are unknown, these conditions may arise from a similar etiology of inadequate remodeling of spiral arteries in early pregnancy causing uteroplacental ischemia. 18,19udies have shown the rate of preterm delivery increases as the pregnancy progresses, often due to medical indications-most commonly placental abruption, preeclampsia, and SGA birth, collectively referred to as ischemic placental disease (IPD). 20In one study, 21 ART pregnancies had 4 times the risk of preterm IPD compared with pregnancies not involving ART.
3][24][25] In the US, chronic hypertension affects a greater percentage of reproductive-age Black women (15%-19%) than White women (6%-10%) or Hispanic women (3%-10%), 25 which contributes to higher rates of placental abruption, preterm delivery, superimposed preeclampsia, cerebrovascular accidents, pulmonary edema, acute kidney failure, and pregnancy-related death.Hispanic women also have higher rates of placental abruption, eclampsia, preterm birth, and neonatal death compared with non-Hispanic White women despite the former having a lower risk of chronic hypertension. 22These disparities may be explained by similar differences in social determinants of health and access to care and insurance coverage. 22Fewer Black women are undergoing ART treatment despite an increasing prevalence of infertility among Black women and a decreasing prevalence of infertility among White women. 26We designed this large population-based study to determine the risk of placental abruption in relation to infertility treatment and to evaluate if exposures to both placental abruption and ART are associated with increased risk of preterm delivery (<37 weeks' gestation) beyond the risks conferred by each factor independently and stratified by race and ethnicity.

Methods
Data were obtained from the National Inpatient Sample (NIS) as part of the Healthcare Cost and The HCUP provided a uniform code for race and ethnicity based on original data from HCUP partner organizations.In the NIS, race and ethnicity were coded as Asian or Pacific Islander, Black, Hispanic, Native American, White, or other (not defined in the NIS dataset), and ethnicity took precedence over race.For example, a patient who was Hispanic and Black was assigned to the category of Hispanic. 28We evaluated 4 categories of race and ethnicity: Black, Hispanic, White, and other (including Asian or Pacific Islander, Native American, and other race and ethnicity).

Covariates
We

Statistical Analysis
We estimated the overall risks of placental abruption and preterm delivery by race, plurality, age group, and income quartile among patients who conceived with ART compared with those who conceived spontaneously, as well as the risks of preterm delivery among patients with spontaneous conceptions, ART conception, and placental abruption.We fit a logistic regression model to estimate the unadjusted and confounder-adjusted odds ratios (ORs and AORs, respectively) and 95% CIs stratified by the mother's race and ethnicity.
To estimate the biological interactions of ART conception and placental abruption with preterm delivery, we calculated the relative risk excess due to interaction (RERI) among patients with ART conception and placental abruption.The RERI describes departure from outcomes on a relative risk (RR) scale.A positive RERI implies that the number of cases attributable to 2 hazards in combination is larger than the sum of the numbers of cases that would be caused by each hazard separately. 31The formula to calculate the RERI is: RERI = RR 11 -RR 10 -RR 01 + 1, where RR ab is the RR in the group with X 1 exposure status a (1 = exposed; 0 = unexposed), and X 2 exposure status b (1 = exposed; 0 = unexposed) compared with the double-exposed group. 32 Downloaded from jamanetwork.comby guest on 07/14/2024 interaction, RERI greater than 0 implies a superadditive interaction, and RERI less than 0 implies a subadditive interaction. 33e statistical analysis was performed in SAS, version 9.4 (SAS Institute Inc).Data were analyzed from January 17 to April 18, 2024.

Results
Of the 78 901 058 hospital deliveries between 2000 and 2019, 391 780 (0.5%) of pregnancies were conceived with ART (Table 1).The mean (SD) maternal age was 27.9 (6.0) years, and The risks of placental abruption among spontaneous and ART conceptions are listed in Table 2.
The overall risk of placental abruption in pregnancies conceived by ART was higher than that in spontaneously conceived pregnancies (17 vs 11 per 1000 births, respectively).After adjusting for confounders, the risk of placental abruption was higher in all ART pregnancies compared with spontaneous pregnancies (AOR, 1.42; 95% CI, 1.34-1.51).
The risks of preterm delivery among spontaneous and ART conceptions, stratified by race, are listed in Table 3. Overall, the risk of preterm delivery was greater for pregnancies conceived by ART compared with spontaneous conception (AOR, 1.46; 95% CI, 1.42-1.51).
The risk of preterm delivery was greater in patients with both ART conception and placental abruption than in patients who experienced only 1 factor (555 per 1000 births vs 156 per 1000 births for ART conception alone and 421 per 1000 births for placental abruption alone) (Table 4).The RERI analysis showed an increase in preterm deliveries when patients had both ART conception and placental abruption (RERI, 2.0; 95% CI, 0.5-3.5)(Table 5).This association was attenuated in Black and Hispanic women.

Discussion
In this large epidemiological cross-sectional study, women who conceived by ART had an increased risk of placental abruption, and women who had both ART conception and placental abruption had an increased risk of preterm delivery compared with women with only 1 of these 2 risk factors.Our study is unique, to our knowledge, in that we explored the joint contributions of ART conception and placental abruption on the risk of preterm delivery.

JAMA Network Open | Obstetrics and Gynecology
Placental Abruption and Preterm Delivery in Patients Using Assisted Reproduction

ART and Placental Abruption
Among preterm births, ART conception has been associated with higher rates of placental infarction after adjusting for confounding factors, such as hypertensive disorders and fetal growth restriction. 34is brings into question: are placental abnormalities due to the processes of ART itself or due to the underlying infertility?A meta-analysis 35 evaluating abnormal placentation between ART conception vs spontaneous conception in patients with reduced fertility compared with a randomized population showed a smaller effect size, suggesting that decreased fertility likely contributes to placental disease.Patients with infertility may have hormonal dysfunction including hyperandrogenism, progesterone resistance, and hyperinsulinism, which appear to impair uterine placental mechanisms. 36Other studies have found that placentas in ART conceptions had increased placental thickness, suggesting placental pathology is due to ART rather than patient factors. 37Women who become pregnant via ART are exposed to supraphysiologic estradiol levels during ART cycles.Estradiol downregulates GATA3, a transcription factor expressed in trophectoderm during embryonic development, and decreased GATA3 leads to impaired trophoblast cell migration through downstream genes integral in placentation. 38Additionally, abnormal H19 methylation imprints were found in ART-derived human preimplantation embryos. 39In a mouse model, loss of H19 imprinting affected placental tissues more than embryonic tissue. 40In humans, levels of pregnancy-associated   Downloaded from jamanetwork.comby guest on 07/14/2024 plasma protein A, a protein involved in trophoblast invasion by contributing to maternal tolerance toward the fetus, were decreased in first-trimester pregnancies conceived by ART. 41Altogether, evidence suggests an association between abnormal hormonal levels influencing epigenetics and impaired trophoblastic invasion via ART processes.

ART and Preterm Delivery
While it has been widely reported that ART is associated with an increase in the risk of preterm delivery, 42 there is no consensus on what specific ART processes confer the greatest risk.In one study, the risk of preterm delivery was 1.5 times greater after ART conception compared with spontaneous conception (11.5% vs 7.7%), but there was no difference in preterm delivery rates with or without ICSI. 43In contrast, another study found an increased risk of medically indicated preterm delivery at 32 to 36 weeks' gestation in ART pregnancies without ICSI compared with ART pregnancies in which ICSI was used (OR, 1.6; 95% CI, 1.1-2.2). 44 The transfer of embryos at the blastocyst stage (day 5 or 6) is now preferred over transfer at the cleavage stage (day 3) due to increased pregnancy and decreased rates of multifetal gestation. 45Studies also suggest that blastocyst transfers and frozen-thawed transfers increase the risk of preterm delivery by 30% to 40%, 46 likely due to greater in vitro manipulation.
Superovulation has been associated with altered gene expression in endometrial remodeling during early implantation. 51 is uncertain if the increased risks of preterm delivery associated with ART conception and placental abruption are due to ART itself or to complications resulting from ART.These technologies may directly influence placental abnormalities, epigenetic and genetic changes due to environmental manipulation, and underlying subfertility. 52One study found that manipulation of mouse embryos in vitro affected blastocyst gene expression, potential for implantation, and placental development. 41On the other hand, ART pregnancies have been associated with an adverse maternal-fetal environment, 53 likely from compensatory mechanisms, which seem to have fewer implications for preterm delivery compared with fertility treatment or preterm delivery mediated by multiple gestations after ART.As single FET cycles are increasingly used that less frequently produce multiple gestations, the roles of direct factors from ART and compensatory factors from an adverse maternal-fetal environment may play a larger role.These unknown factors highlight the need for ongoing research on the mechanisms of preterm delivery, therapeutic targets, and modulating clinical practice.
We found a smaller increased risk of placental abruption in Black patients undergoing ART compared with the risk among White and Hispanic patients.The RERI analysis indicated that the risks of preterm delivery for the combination of ART conception and placental abruption were higher in preterm delivery among Black and Hispanic patients, but substantially higher among White patients.
[56] Preterm delivery rates may differ among races due to psychosocial stressors, epigenetic factors, and changes in placental, endothelial, and kidney biomarkers. 57,58The weathering hypothesis states that chronic exposure to social and economic disadvantage leads to an accelerated decline in physical health outcomes and can explain racial disparities in a variety of health conditions. 59 aggressive ovarian stimulation is required in Black women, and they have longer duration of infertility, higher body mass index, lower implantation rates, higher rates of fibroids, and higher rates of spontaneous abortion. 60Black race is a risk factor for placental abruption, and both Black race and placental abruption are risk factors for preterm delivery; yet, among preterm births, the risk of placental abruption is lower in Black women compared with White women after adjusting for socioeconomic status and medical risk factors. 54This seemingly contradictory finding suggests that other mechanisms of preterm birth may play a greater role in Black women, such as spontaneous preterm labor or premature rupture of membranes.Most patients undergoing infertility treatment are White or Asian, suggesting a racial disparity between the composition of the population receiving fertility treatment compared with the general population. 61There is a lack of data and consistent findings on ART outcomes in the Hispanic population 62 ; some studies suggest that Hispanic women with ART conception have similar rates of pregnancy, spontaneous abortion, and live births compared with White women with ART conception. 62Hispanic women are less likely to seek or receive infertility treatment and are more likely to have preterm delivery and low-birth-weight infants after ART compared with non-Hispanic White women undergoing ART. 63

Limitations
With information from an administrative database using ICD-9-CM and ICD-10-CM data, we were unable to determine the certainty of a diagnosis and in which hospitalization period it occurred.We included diagnoses of fertility testing, male factor infertility, procreative management, and female infertility but acknowledge that these ICD-9-CM and ICD-10-CM codes do not fully reflect the use of ART.Procreative management can be used as a diagnosis code for appointments that are not related to infertility.Patients with ART conception were diagnosed primarily with obstetric codes for pregnancy resulting from ART, supervision of pregnancy with a history of infertility, and complications associated with artificial fertilization.The Centers for Disease Control and Prevention estimates that 2.3% of infants born in the US are conceived using ART, 64 but our study identified an overall prevalence of only 0.5%.This suggests there could be data that are unaccounted for that could have some unknown impact on our results.Our dataset did not specify which specific ART processes were used or the subcategories of each type of ART.
The NIS does not track women across years, so we were unable to account for repeated pregnancy outcomes.Owing to data limitations, we were unable to distinguish between spontaneous and clinician-initiated preterm delivery.Although we found a greater risk of preterm delivery in patients who both conceived via ART and developed placental abruption, it is unclear whether preterm delivery and placental abruption are due to the processes of ART itself or due to diseases related to infertility.

Conclusions
To our knowledge, this cross-sectional study is the first to elucidate an additive association between the combination of placental abruption and ART conception on subsequent preterm delivery, but there are still unanswered questions about the mechanisms underlying this association.Our findings underscore the importance of counseling patients undergoing the demanding process of infertility treatment on risks to their pregnancy.The adverse effects of infertility treatments are continually being researched to direct potential targeted therapies.Finally, health care practitioners need to be aware of the complex interplay of race and ethnicity on preterm delivery and neonatal outcomes among pregnancies conceived by ART.

JAMA Network Open | Obstetrics and Gynecology Placental
Utilization Project (HCUP), sponsored by the Agency for Healthcare Research and Quality.The NIS is Abruption and Preterm Delivery in Patients Using Assisted Reproduction database of inpatient hospital stays from all HCUP-participating hospitals (28 states and 994 hospitals in 2000), with more states and hospitals participating in HCUP over time(48states plus the District of Columbia and 4568 hospitals in 2019).The NIS encompasses 98% of the US population and accounts for 7 to 8 million unweighted hospital discharges and 36 to 39 million weighted hospital discharges each year.This cross-sectional study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline 27 and, in accordance with 45 CFR §46, was deemed exempt from institutional review board review and the informed consent requirement because data used in the study were deidentified.

Table 1 .
Distribution of Women Who Conceived by ART and Those Who Conceived Spontaneously From the National Inpatient Sample, 2000 to 2019

Table 1 .
Distribution of Women Who Conceived by ART and Those Who Conceived Spontaneously From the National Inpatient Sample, 2000 to 2019 (continued) 30Other includes Asian or Pacific Islander, Native American, and other (not defined in National Inpatient Sample dataset).bDefinitions of hospital bed size vary by region, rural vs urban status, and technical status.29cDefinition of other varies by state.30

Table 2 .
Risks of Placental Abruption by Race and Ethnicity Among Patients Stratified by ART or Spontaneous Conception, National Inpatient Sample, 2000 to 2019 a Overall odds ratios were adjusted for year, age group, hospital bed size, hospital teaching status, location, region, maternal race and ethnicity, singleton vs multiple births, insurance, and income based on a survey logistic regression model.bOddsratiosbymaternalrace and ethnicity were adjusted for year, age group, hospital bed size, hospital teaching status, location, region, singleton vs multiple birth, insurance, and income based on a survey logistic regression model.cOtherincludesAsianor Pacific Islander, Native American, and other (not defined in National Inpatient Sample dataset).JAMA Network Open | Obstetrics and GynecologyPlacental Abruption and Preterm Delivery in Patients Using Assisted Reproduction JAMA Network Open.2024;7(7):e2420970.doi:10.1001/jamanetworkopen.2024.20970(Reprinted) July 10, 2024 6/13 Downloaded from jamanetwork.comby guest on 07/14/2024

Table 3 .
Risks of Preterm Delivery Associated With ART vs Spontaneous Conception and Stratified by Race and Ethnicity, National Inpatient Sample, 2000 to 2019 a a Preterm delivery occurred before 37 weeks' gestation.bOddsratios were adjusted for year, age group, hospital bed size, hospital teaching status, location, region, singleton vs multiple births, insurance, and income based on a survey logistic regression model (and additionally adjusted for mother's race and ethnicity for overall associations).cOther includes Asian or Pacific Islander, Native American, and other (not defined in National Inpatient Sample dataset).

Table 4 .
Preterm Delivery Risks Stratified by Race and Ethnicity Based on Conception by ART Alone, Placental Abruption Alone, or Both Factors, National Inpatient Sample, 2000 to 2019 a Other includes Asian or Pacific Islander, Native American, and other (not defined in National Inpatient Sample dataset).

Table 5 .
Association of Conception by ART and Placental Abruption, Alone and in Combination, With the Risk of Preterm Delivery by Race and Ethnicity, National Inpatient Sample, 2000 to 2019 a Abbreviations: ART, assisted reproductive technology; RERI, relative excess risk due to interaction.aOddsratioswereadjusted for year, age group, hospital bed size, hospital teaching status, location, region, singleton vs multiple births, insurance, and income based on a survey logistic regression model (and additionally adjusted for mother's race and ethnicity for overall associations).bOtherincludesAsian or Pacific Islander, Native American, and other (not defined in National Inpatient Sample dataset).JAMA Network Open | Obstetrics and GynecologyPlacental Abruption and Preterm Delivery in Patients Using Assisted Reproduction JAMA Network Open.2024;7(7):e2420970.doi:10.1001/jamanetworkopen.2024.20970(Reprinted) July 10, 2024 7/13

SUPPLEMENT 2. Data Sharing Statement
Placental Abruption and Preterm Delivery in Patients Using Assisted Reproduction SUPPLEMENT 1. eFigure.Prevalence of ART Deliveries and the Prevalence of Placental Abruption Among Women who Conceived by ART or Spontaneously: National Inpatient Sample, 2000-2019 eTable 1.International Classification of Diseases Codes Used in the Study eTable 2. Risks of Placental Abruption in Pregnancies Conceived by ART or Spontaneously, National Inpatient Sample, 2000 to 2019 eTable 3. Rates of Preterm Delivery (<37 Weeks' Gestation) in Pregnancies Conceived by ART or Spontaneously, National Inpatient Sample, 2000 to 2019 eTable 4. Rates of Preterm Delivery Among Pregnancies Conceived by ART or Spontaneously, National Inpatient Sample, 2000 to 2019 eTable 5. Relationship Between Conception by ART and Placental Abruption, Alone and in Combination, on the Risk of Preterm Delivery, National Inpatient Sample, 2000 to 2019