Most studies involving women with preexisting heart disease have highlighted the negative effects of pregnancy on maternal cardiac status but have focused less on the effect on the fetus and neonatal outcomes. However, some studies suggest an increased rate of neonatal morbidity and mortality in the newborns of women with heart disease, with mortality rates of 4%1 and fetal growth restriction in up to 8% of pregnancies being reported.2,3 The risk of fetal and neonatal complications has been associated with poor maternal functional class and left heart obstruction.2,3 In addition, deteriorating cardiac function may require early delivery, further compromising fetal well-being.
The maternal factors that determine fetal growth include nutrition, oxygenation, and cardiovascular adaptation to pregnancy.4 The latter is important because it determines uteroplacental perfusion, and although cardiac disease can be associated with poor nutrition and hypoxia, it seems likely that cardiovascular performance is the major influence on fetal growth. Fetal growth restriction is a common cause of perinatal morbidity and mortality and is increasingly recognized as a risk factor for cardiovascular and metabolic disease in later life.5 Understanding its etiology in cardiac disease may suggest ways to improve fetal growth generically. Our hypothesis was that cardiac disease limits the increase in cardiac output that occurs in normal pregnancy, leading to insufficiency of the uteroplacental blood supply and ultimately fetal growth restriction. To test this hypothesis, we performed a retrospective cohort study comparing birth weight percentiles in women with heart disease dividing this population into those in whom a substantial reduction of cardiac output would and would not be expected and comparing both with a control population.
The study cohort consisted of all women with congenital and acquired heart disease who delivered at Chelsea and Westminster Hospital between January 1994 and October 2010. For each woman identified, only her first pregnancy that progressed beyond 24 weeks of gestation was included in the study to avoid the bias affecting the overall results that could occur if more than one pregnancy occurred in a woman with a particular tendency to a poor pregnancy outcome. To allow comparison with our overall population, controlling for changes in demographics and practice over time, the women who delivered immediately before and immediately after each index pregnancy were used as controls. Only women with singleton pregnancies were studied. Twenty-five women with heart disease delivered before 1998 and, because data for control women were not available before this date, the first two women delivered on the same date in 1999 were used.
Data were obtained from the medical and obstetric charts. Supplementary information was taken from the Ciconia Maternity Information System. The data recorded for each pregnancy included age at booking, height, weight, parity, ethnicity, cardiac medications, and use of tobacco smoking and illicit drugs. For the control group, when important data such as birth weight were missing, the next case was used. Cardiac lesions were classified according to the International Pediatric and Congenital Cardiac Code.6 Previous cardiac surgical repairs and reoperation were noted. Prepregnancy hemodynamic status was assessed by echocardiography, cardiac magnetic resonance imaging, and clinical reports. The New York Heart Association functional class before pregnancy was ascertained. The patient population was further divided into four subgroups: those with acyanotic heart disease and a cardiac lesion associated with a normal cardiac output (group 1) or a cardiac lesion associated with a reduced cardiac output (group 2), those with cyanotic heart disease (group 3), and those using a β-blocker (group 4; Table 1).
Gestational age at delivery, admission to the neonatal intensive care unit, and birth weight were noted and birth weight percentiles were calculated using a computer-generated chart correcting for maternal ethnicity, parity, body mass index, gestational age, and the gender of the child.7,8
Adverse events throughout pregnancy and the postnatal period were classified as perinatal or obstetric complications as follows. Perinatal complications were preterm birth (delivery after 24 and before 37 completed weeks of gestation), small for gestational age (birth weight below the 10th percentile), perinatal mortality (stillbirth after 24 completed weeks of pregnancy and neonatal death up to 1 month after birth), and recurrence of congenital heart disease. Obstetric complications were antepartum hemorrhage (bleeding from the genital tract after 24 weeks of gestation), gestational hypertension (blood pressure 140/90 mm Hg or higher measured on two separate occasions at least 6 hours apart developing after 20 weeks of gestation), preeclampsia (gestational hypertension criteria with proteinuria of 300 mg/L or more in a 24-hour urine collection), eclampsia (preeclampsia with grand mal seizures), prelabor preterm rupture of membranes (spontaneous rupture of membranes before 37 weeks of gestation in the absence of regular painful contractions), and postpartum hemorrhage (blood loss 500 mL or more for vaginal delivery and 1,000 mL or more for caesarean delivery, or any blood loss requiring blood transfusion).
Data analysis was performed using SPSS 18 for Windows. Descriptive statistics are reported as frequency and median and interquartile range or mean value and standard deviations as appropriate. Comparison of continuous variables between groups was made by unpaired Student t tests when the data were normally distributed and, if not, then the Mann-Whitney U test was used. When comparing frequencies, we applied the χ2 test or Fisher exact test, when applicable. All tests used were two-tailed and P<.05 was considered statistically significant. Binary logistic regression was used to study the association of adverse perinatal outcome with maternal variables. Univariate analysis selected variables to be brought forward to multivariate analysis (P<.1). The present study was reviewed and approved by the Riverside Research Ethics Committee.
A total of 334 women with heart disease were identified. Three multiple pregnancies were excluded, leaving 331 women to be included in the study and compared with 662 women in the control group. The cardiac lesions identified in each subgroup are shown in Table 1. Baseline characteristics for all pregnancies are shown in Table 2.
Table 3 outlines the gestational age, birth weight and birth weight percentiles, and perinatal complications in the heart disease group and its subgroups and in the control group. Both gestational age and birth weight were significantly lower in the heart disease group, especially in those with reduced cardiac output or cyanosis. Median birth weight percentile was correspondingly lower in the same groups; 135 perinatal complications occurred in 111 (34%) pregnancies in women with heart disease, compared with 106 in 99 (15%) women in the control group (odds ratio 2.9, 95% confidence interval 2.1–3.9, P<.001).
Univariate analyses identified the variables New York Heart Association class I or higher, maternal cyanosis, myocardial dysfunction, and use of anticoagulation to be brought forward to multivariate analysis. Multivariate analysis is shown in Table 4. In combination, only myocardial dysfunction was significantly associated with adverse perinatal outcomes.
In the heart disease group, 111 obstetric complications occurred in 104 (31%) pregnancies. In the control population, 150 obstetric complications occurred in 149 (23%) pregnancies (odds ratio 1.57, 95% confidence interval 1.17–2.1, P=.003). When each individual obstetric complication was compared, only the risk of postpartum hemorrhage was significantly higher in the heart disease group (odds ratio 2.56, 95% confidence interval 1.79–3.66, P<.001).
In the presence of maternal heart disease, perinatal outcomes are impaired. In our series, perinatal complications occurred in 34% of women with heart disease compared to 15% of the control population. This rate is higher than previously reported.1–3 The most frequent complication, accounting for 61%, was a neonate born small for gestational age; both average birth weight and birth weight percentile in the heart disease group were significantly lower than in the control population. The mean gestational age at birth was significantly earlier in the heart disease population, which to some extent explains the difference seen in birth weight. However, the use of customized birth weight percentiles accounts for the confounding effects of gestation, parity, maternal body mass index, ethnicity, and the gender of the child. Other important factors associated with fetal growth restriction such as smoking, illicit drug use, and pregnancy-related hypertension were not significantly different between the two groups. Although we are unable to account for every confounding factor, our data suggest that the presence of maternal heart disease predisposes to a reduction in fetal growth rate. By dividing the cardiac lesions into subgroups, we estimated the effect of maternal cyanosis and a reduced cardiac output on fetal growth. Women with central cyanosis had the smallest newborns, with a median birth weight percentile of 13. Maternal hypoxia in women with cyanotic heart disease is known to be associated with impaired fetal growth, with maternal oxygen saturation inversely related to birth weight and fetal mortality.9,10 Similarly, at high altitudes (more than 2,500 m), chronic hypoxia leads to reduced birth weight.11 The chronic hypoxia of altitude interferes with the maternal circulatory adjustments to pregnancy such that blood volume is lower than at sea level, leading to a reduced uteroplacental blood flow.12 The reduced birth weight percentiles seen in the women with reduced cardiac output supports the hypothesis that cardiac output is an important factor in fetal growth. A negative correlation between cardiac output and birth weight has been demonstrated in normal pregnancies,13 and lower-than-average cardiac output has been demonstrated in mothers with idiopathic fetal growth restriction.14,15 This effect is presumably mediated via a reduced uteroplacental blood flow; however, the cardiovascular physiological response to pregnancy in women with heart disease has not been reported and we are now studying this prospectively.
The rate of preterm delivery in women with heart disease was significantly higher than in the control population (13% compared with 5%). Of these births, 67% were iatrogenic and 52% occurred before 34 completed weeks of gestation. We do not have data on the cause of preterm delivery in the control population; however, the proportion of preterm delivery in the general population that is iatrogenic has been reported to be 25.9%.16 Some studies have reported an increased rate of prelabor preterm rupture of membranes in women with heart disease.3,17 Our study showed a 1% rate of prelabor preterm rupture of membranes in women with heart disease, which is not significantly different than that of the control population. We are unable to comment on the long-term outcomes in the newborns of women with heart disease because full data on subsequent development are unavailable. However, preterm delivery (particularly at early gestational ages) and a reduced birth weight are likely to have significant clinical sequelae.
Univariate analyses identified the variables New York Heart Association class I or higher, maternal cyanosis, myocardial dysfunction, and use of anticoagulation to be brought forward to multivariate analysis. Of these, only myocardial dysfunction was significantly associated with adverse perinatal outcomes. This suggests that the effect of myocardial dysfunction is the most important variable contributing to the occurrence of an adverse perinatal outcome.
Obstetric complications occurred significantly more frequently in the cardiac group because of an increased rate of postpartum hemorrhage (68% of complications). This may be associated with the reduced syntocinon dose we use with cardiac patients to minimize its effects on the cardiovascular system (bolus doses can cause serious hypotension) or simply increased attention to recording maternal blood loss in this high-risk group.
Several potential limitations of our study must be noted. First, the retrospective design of the study required review of the medical charts to ascertain some of the data not included in the electronic archives, and some could not be collected because of deficient records. Second, the patient sample is subject to selection bias. Patients with more severe cardiac complications who receive counseling explaining pregnancy-related risk might choose not to proceed with pregnancy, and our study population therefore may represent a subset of women with heart disease at lower cardiac risk. The use of a nonmatched control population can introduce significant confounding factors. However, we chose not to match for characteristics such as height, weight, body mass index, and maternal age, because these can be part of the syndrome of cardiac disorder. Instead, we chose to examine their influence using logistic regression. In fact, only age was significantly different (Table 2), with the women in the cardiac group being significantly younger. It may be that as age advances and cardiac condition worsens in many cases, fertility declines faster in women with cardiac disease than in the general population, resulting in a deficit of older women in the cardiac study group. Maternal height, weight, and body mass index showed no significant differences (data not shown). Finally, only those women whose pregnancies progressed to a potentially viable gestation were included in our study, and thus we were unable to ascertain the risk of infertility or miscarriage before 24 weeks of gestation.
This cohort study demonstrates that women with cardiac disease are more likely than a those in a control population to give birth to newborns with low percentile birth weights at an earlier gestational age. This results in a higher rate of perinatal complications in these women. The strongest predictors of low birth weight and preterm delivery are maternal cyanosis or reduced cardiac output or both.
1. Drenthen W, Pieper PG, Roos-Hesselink JW, van Lottum WA, Voors AA, Mulder BJ, et al. Outcomes of pregnancy in women with congenital heart disease: a literature review. J Am Coll Cardiol 2007;49:2303–11.
2. Siu SC, Colman JM, Sorensen S, Smallhorn JF, Farine D, Amankwah KS, et al. Adverse neonatal and cardiac outcomes are more common in pregnant women with cardiac disease. Circulation 2002;105:2179–84.
3. Khairy P, Ouyang DW, Fernandes SM, Lee-Parritz A, Economy KE, Landzberg MJ. Pregnancy outcomes in women with congenital heart disease. Circulation 2006;113:517–24.
4. Neerhof MG. Causes of intrauterine growth restriction. Clin Perinatol 1995;22:375–85.
5. Barker DJ, Bull AR, Osmond C, Simmonds SJ. Fetal and placental size and risk of hypertension in adult life. BMJ 1990;301:259–62.
6. Franklin RC, Jacobs JP, Krogmann ON, Beland MJ, Aiello VD, Colan SD, et al. Nomenclature for congenital and paediatric, cardiac disease: historical perspective and the International Pediatric and Congenital Cardiac Code. Cardiol Young 2008;18(Suppl 2):70–80.
7. Gardosi J, Chang A, Kalyan B, Sahota D, Symonds EM. Customised antenatal growth charts. Lancet 1992;339:283–7.
8. Gardosi J, Mongelli M, Wilcox M, Chang A. An adjustable fetal weight standard. Ultrasound Obstet Gynecol 1995;6:168–74.
9. Presbitero P, Somerville J, Stone S, Aruta E, Spiegelhalter D, Rabajoli F. Pregnancy in cyanotic congenital heart disease. Outcome of mother and fetus. Circulation 1994;89:2673–6.
10. Spence MS, Balaratnam MS, Gatzoulis MA. Clinical update: cyanotic adult congenital heart disease. Lancet 2007;370:1530–2.
11. Mortola JP, Frappell PB, Aguero L, Armstrong K. Birth weight and altitude: a study in Peruvian communities. J Pediatr 2000;136:324–9.
12. Wilson MJ, Lopez M, Vargas M, Julian C, Tellez W, Rodriguez A, et al. Greater uterine artery blood flow during pregnancy in multigenerational (Andean) than shorter-term (European) high-altitude residents. Am J Physiol Regul Integr Comp Physiol 2007;293:R1313–24.
13. Desai DK, Moodley J, Naidoo DP. Echocardiographic assessment of cardiovascular hemodynamics in normal pregnancy. Am J Obstet Gynecol 2004;104:20–8.
14. Bamfo JEAK, Kametas NA, Turan O, Khaw A, Nicolaides, KH. Maternal cardiac function in fetal growth restriction. BJOG 2006;113:784–91.
15. Rosso P, Donoso E, Braun S, Espinoza R, Fernandez C, Salas SP. Maternal hemodynamic adjustments in idiopathic fetal growth retardation. Gynecol Obset Invest 1993;35:162–5.
16. Tucker JM, Goldenberg RL, Davis RO, Copper RL, Winkler CL, Hauth JC. Etiologies of preterm birth in an indigent population: is prevention a logical expectation? Obstet Gynecol 1991;77:343–7.
17. Ouyang DW, Khairy P, Fernandes SM, Landzberg MJ, Economy KE. Obstetric outcomes in pregnant women with congenital heart disease. Int J Cardiol 2009; May 1. [Epub ahead of print].