Highlights
-
•
Forty percent of infants born to women with congenital heart disease experienced the composite adverse outcome of preterm delivery, small for gestational age and/or neonatal intensive care unit admission.
-
•
The risk of adverse neonatal outcomes was associated with maternal congenital heart disease severity.
-
•
In addition to congenital heart disease severity, certain maternal demographics, reproductive health history factors and pregnancy-related factors were also associated with an increased risk of neonatal adverse outcomes.
ABSTRACT
Background
Infants born to women with congenital heart disease (CHD) are at increased risk for adverse outcomes compared with the general population. There are few large, contemporary studies examining the relationship between CHD severity, maternal risk factors and neonatal outcomes.
Methods
Data on women with CHD who had a live birth from 2011 through 2014 in geographic areas in Georgia, North Carolina and Utah were analyzed. Maternal CHD was identified from clinical and administrative data collected for the CDC Lifespan CHD surveillance project. CHD was categorized as severe and non-severe (shunt, valve or shunt+valve lesions). Linked birth certificates provided information on neonatal outcomes and neonatal, maternal and pregnancy characteristics. Neonatal outcomes included preterm delivery (<37 weeks), low birth weight (<2,500 grams), small for gestational age (SGA), neonatal intensive care unit (NICU) admission, a composite of any adverse outcome (one or more of preterm delivery, SGA or NICU admission), and CHD. Neonatal outcomes were summarized using descriptive statistics. Differences by maternal CHD severity were assessed using Wilcoxon rank-sum and Chi-square tests for continuous and categorical variables, respectively. The association of maternal and pregnancy characteristics (race, pregestational diabetes (PGDM), hypertensive disorders of pregnancy (HDP), prior preterm birth, non-singleton pregnancy, maternal age, previous living live birth and CHD severity) with neonatal outcomes were assessed using logistic regression fit using generalized estimating equations clustered on mother and adjusted for mother’s age, birth year and geographical location of delivery.
Results
Of the 2,411 liveborn infants born to 1,982 women with CHD (34% with severe CHD), 19% were preterm, 18% were of low birth weight, 14% were SGA and 4.2% were diagnosed with CHD. Of the included infants, 25% required NICU and 40% experienced the composite adverse outcome. Infants born to mothers with severe CHD (vs non-severe) were more likely to experience the composite adverse outcome (aOR 2.1 [1.8, 2.5]) and CHD recurrence (aOR 17.1 [9.3, 31.5]). In a multivariable analysis, severe CHD (vs non-severe, aOR 2.2 [1.8, 2.7]), Black race (vs White, aOR 1.8 [1.4, 2.3]), PGDM (aOR 2.5 [1.3, 5.0]), hypertensive disorder of pregnancy (aOR 2.1 [1.5, 3.1]), prior preterm birth (aOR 2.0 [1.3, 3.1]), and non-singleton pregnancy (aOR 6.7 [3.8, 11.6]) were associated with increased risk of the composite adverse outcome.
Conclusions
Nearly half of infants born to women with CHD experienced the adverse composite neonatal outcome, with the highest risk among those born to women with severe CHD. In addition to CHD severity, risk factors for adverse neonatal outcomes among women with CHD included black race, hypertensive disorders of pregnancy, PGDM, prior preterm birth and non-singleton pregnancy. These findings can help provide individualized counseling about neonatal risks for women with CHD.
Background
Congenital heart disease (CHD) is the most common birth defect, affecting nearly 1% of births in the United States each year. Improvements in pediatric medical and surgical care has extended long-term survival of patients with CHD over the last several decades. As a result, the population of adults living with CHD now exceeds that of children, with an estimated 1.4 million adults living with CHD in the US. Accordingly, the number of women with CHD surviving to childbearing age is rapidly growing, and more women with CHD are able to experience pregnancy. ,,
During pregnancy, the cardiovascular system undergoes profound physiologic hemodynamic changes to support uteroplacental circulation. Although some women with CHD can adequately adapt to these changes and are able to tolerate pregnancy without complication, patients with more complex CHD are at increased for maternal and fetal adverse outcomes. , Likewise, infants born to women with CHD appear to be at increased risk for preterm birth, low birthweight, CHD recurrence and neonatal mortality, with a positive correlation between CHD severity and neonatal adverse outcomes. Despite this, there are few large, contemporary, US-based studies examining the relationship between CHD severity, maternal risk factors and neonatal outcomes. The aims of this analysis were to describe neonatal outcomes in infants born to mothers with CHD across in a large population in multiple geographic areas in the United States, compare neonatal outcomes by maternal CHD severity, and identify maternal risk factors for adverse neonatal outcomes.
Methods
This study was funded by the Centers for Disease Control and Prevention, Grant/Award Number CDC‐RFA‐DD15-1506. The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.
Study design and data source
The Centers for Disease Control (CDC) Lifespan project is a retrospective cohort study of individuals with CHD residing at any time from January 1, 2011 to December 31, 2013 in 1 of 5 geographic areas of the US: Colorado, North Carolina, Utah, 5 counties in metropolitan Atlanta, Georgia, and 11 counties in the state of New York. Their comorbidities were documented in the same three-year period (2011-2013). ,, This analysis evaluated neonatal outcomes of pregnancies in women with CHD and assessed risk factors for adverse neonatal outcomes. Individuals with CHD who received care in three geographic sites in Georgia, North Carolina, and Utah between January 1, 2011 and December 31, 2013 and gave birth between January 1, 2011 and December 31, 2014 were identified through linked clinical, administrative, and birth certificate data. Two additional funded sites, Colorado and New York, were excluded due to absence of linked infant birth certificates necessary for this analysis.
The Institutional Review Boards from Emory University in Georgia (GA), Duke University in North Carolina (NC), and Intermountain Health and University of Utah (UT) approved an analysis of deidentified data. The requirement for informed consent was waived by each site’s respective Institutional Review Board due to the de-identified data analysis.
Identification and classification of individuals with CHD
Individuals with CHD were defined by having at least one health encounter associated with a select CHD-related International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) code in the 745.xx to 747.xx range, as documented in the electronic health record between January 1, 2011 and December 31, 2013. Exclusion criteria included 1) ICD-9-CM code 745.5 (code for atrial septal defect and for patent foramen ovale, a normal variant) in isolation or in combination only with 746.9 (unspecified anomaly of the heart) and/or 746.89 (other specified anomalies of the heart), as these codes have low positive predictive value for true atrial septal defect, and 2) “other” CHD ICD-9-CM codes (ICD-9-CM 745, 745.7, 746, 746.8, 746.82, 746.84, 746.85, 746.87, 746.89, 746.9, 747, 747.2, 747.20, 747.21, 747.29, 747.4, 747.40, 747.49, 747.9), which have low positive predictive value for true CHD.
Based on ICD-9-CM codes, individuals were classified into one of five mutually exclusive CHD severity groups: severe, shunt, valve, shunt+valve (ie, both types of CHD present), and “other.” The severity groups were based on a previously described method that categorizes CHD diagnostic codes into mutually exclusive hierarchical groups based on cardiac anatomy, similar to Marelli et al. , Severe CHD included endocardial cushion defects, interrupted aortic arch, tetralogy of Fallot, total anomalous pulmonary venous return, tricuspid atresia, transposition complexes, truncus arteriosus, and univentricular hearts. Individuals with multiple CHD-related codes who had at least one severe code were classified as having severe CHD regardless of the number of non-severe codes. For this analysis, patients with shunt, valve or shunt+valve CHD were grouped together and defined as having “non-severe” disease.
Linkage to birth records
Linkage of birth certificates to women with CHD who were included in the Lifespan data was performed by sites using site-specific methodology. Linkage was performed within each study site by matching the name and date of birth of the identified individuals with CHD to the mother’s name (exact match or one-character difference) and (mother’s) date of birth on infant birth certificates filed between January 1, 2011 and December 31, 2014. Each linked birth certificate/infant was treated as an individual unit of analysis; thus, a single individual with CHD who gave birth to multiple infants (in one or multiple pregnancies) between January 1, 2011 and December 30, 2014 could be included multiple times in the analysis. Following linkage at each site, the identifying information was removed, and de-identified data were collected for analysis at Duke Clinical Research Institute. Non-viable infants (defined as <22 weeks of gestation) were excluded from the analysis.
Outcome variables
The primary outcome of interest was a composite adverse birth outcome defined as one or more of the following: preterm birth (less than 37 weeks of gestation), small for gestational age (SGA), and neonatal intensive care unit (NICU) admission. For infants with gestational age 22 weeks to 42 weeks of gestation at birth, SGA was defined as birthweight less than 10th percentile for gestational age on the Fenton growth charts. SGA for gestational age ≥43 weeks was determined by visual approximation of the Fenton growth charts. Additional neonatal outcomes of interest included gestational age at birth, birth weight (grams), low birth weight (<2,500 grams), need for assisted ventilation, sepsis, CHD recurrence, and non-cardiac congenital anomalies. All outcomes were derived from linked birth certificate data.
Independent variables
Independent variables collected included maternal age at delivery, race, education, present/past tobacco use, CHD severity, comorbidities (pre-gestational diabetes (PGDM), gestational diabetes (GDM), chronic hypertension (cHTN), hypertensive disorders of pregnancy (HDP; gestational hypertension, preeclampsia or eclampsia), and tobacco use), obstetric history (previous live births now living, previous preterm delivery, previous cesarean section, infertility treatment), and pregnancy/birth characteristics (non-singleton pregnancy, infant sex, birth year, and site). All independent variables were derived from linked birth certificate data.
Statistical analysis
Maternal/pregnancy characteristics and neonatal outcomes were assessed using descriptive statistics and reported as median (Q1, Q3) for continuous variables and % ( n ) for categorical variables. Differences by maternal CHD severity were assessed using Wilcoxon rank-sum and Chi-square tests for continuous and categorical variables, respectively. The association of maternal and pregnancy characteristics (race, PGDM, HDP, prior preterm birth, non-singleton pregnancy, maternal age, previous living live birth, and CHD severity) and neonatal outcomes was assessed using univariable and multivariable logistic regression fit using generalized estimating equations clustered on birthing person and adjusted for age at delivery, birth year, and geographical location of delivery. All analyses were performed using SAS software, Version 9.4 (SAS Institute Inc., Cary, NC).
Results
Baseline characteristics
Between January 1, 2011 and December 31, 2014, 2,411 liveborn infants born to women with CHD were identified ( Figure 1 ). Of the included infants, 34.1% ( n = 822) were born to women with severe CHD and 65.9% ( n = 1,589) with non-severe CHD: 39.98% ( n = 961) with a shunt defect, 20.9% ( n = 505) with a congenital valve defect and 5.1% ( n = 123) with both shunt and congenital valve disease. Demographic and clinical characteristics are shown in Table 1 .
Derivation of the study population. Flowchart diagram demonstrating the derivation of the analytic population of births from women with CHD. The final analytic population included 2,411 births from 1,982 women across 3 sites. CHD, congenital heart disease; EHR, electronic health record.
Table 1
Characteristics of mothers with CHD at the time of delivery, overall and by CHD severity.
| Characteristics | All births ( N = 2,411) | Severe ( N = 822) | Non-severe ( N = 1,589) | P -value |
|---|---|---|---|---|
| Demographics | ||||
| Maternal age at delivery (y) | .124 | |||
| N | 2,411 | 822 | 1,589 | |
| Median (Q1, Q3) | 28 (23, 33) | 29 (24, 33) | 28 (23, 33) | |
| Mother race | .069 | |||
| White | 1,589/2,321 (68.5%) | 538/817 (65.9%) | 1,051/1,504 (69.9%) | |
| Black or African American | 585/2,321 (25.2%) | 229/817 (28.0%) | 356/1,504 (23.7%) | |
| Asian/American Indian or Alaska Native/Other | 147/2,321 (6.3%) | 50/817 (6.1%) | 97/1,504 (6.4%) | |
| Mother education | .015 | |||
| 0-12 y of primary/secondary school | 837/2,334 (35.9%) | 293/787 (37.2%) | 544/1,547 (35.2%) | |
| At least some college (no graduate school) | 1,046/2,334 (44.8%) | 368/787 (46.8%) | 678/1,547 (43.8%) | |
| At least some graduate school | 451/2,334 (19.3%) | 126/787 (16.0%) | 325/1,547 (21.0%) | |
| Tobacco use prior to or during pregnancy | 135 (5.6%) | 40 (4.9%) | 95 (6.0%) | .402 |
| Selected maternal comorbidities | ||||
| Pre-gestational diabetes | 45/2,373 (1.9%) | 14/808 (1.7%) | 31/1,565 (2.0%) | .674 |
| Gestational diabetes | 100/2,373 (4.2%) | 35/808 (4.3%) | 65/1,565 (4.2%) | .838 |
| Chronic hypertension | 86/2,373 (3.6%) | 12/808 (1.5%) | 74/1,565 (4.7%) | <.001 |
| Hypertensive disorder of pregnancy | 135/2,373 (5.7%) | 36/808 (4.5%) | 99/1,565 (6.3%) | .062 |
| Pregnancy characteristics | ||||
| Previous live births | .002 | |||
| none | 898/2,347 (38.3%) | 272/804 (33.8%) | 626/1,543 (40.6%) | |
| one | 765/2,347 (32.6%) | 295/804 (36.7%) | 470/1,543 (30.5%) | |
| two or more | 684/2,347 (29.1%) | 237/804 (29.5%) | 447/1,543 (29.0%) | |
| Previous preterm | 103/2,373 (4.3%) | 25/808 (3.1%) | 78/1,565 (5.0%) | .032 |
| Previous cesarean section | 384/2,373 (16.2%) | 139/808 (17.2%) | 245/1,565 (15.7%) | .332 |
| Infertility treatment | 82/2,373 (3.5%) | 24/808 (3.0%) | 58/1,565 (3.7%) | .352 |
| Non-singleton pregnancy | 156 (6.5%) | 72 (8.8%) | 84 (5.3%) | .001 |
| Infant sex | .447 | |||
| Male | 1,247 (51.7%) | 434 (52.8%) | 813 (51.2%) | |
| Female | 1,164 (48.3%) | 388 (47.2%) | 776 (48.8%) | |
| Infant birth year | .010 | |||
| 2011 | 631 (26.2%) | 237 (28.8%) | 394 (24.8%) | |
| 2012 | 615 (25.5%) | 192 (23.4%) | 423 (26.6%) | |
| 2013 | 736 (30.5%) | 267 (32.5%) | 469 (29.5%) | |
| 2014 | 429 (17.8%) | 126 (15.3%) | 303 (19.1%) | |
| Site | <.001 | |||
| Georgia | 1,291 (53.5%) | 562 (68.4%) | 729 (45.9%) | |
| North Carolina | 603 (25.0%) | 174 (21.2%) | 429 (27.0%) | |
| Utah | 517 (21.4%) | 86 (10.5%) | 431 (27.1%) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
