Fetal cardiac remodeling and dysfunction is associated with both preeclampsia and fetal growth restriction


Preeclampsia and fetal growth restriction share some pathophysiologic features and are both associated with placental insufficiency. Fetal cardiac remodeling has been described extensively in fetal growth restriction, whereas little is known about preeclampsia with a normally grown fetus.


To describe fetal cardiac structure and function in pregnancies complicated by preeclampsia and/or fetal growth restriction as compared with uncomplicated pregnancies.

Study design

This was a prospective, observational study including pregnancies complicated by normotensive fetal growth restriction (n=36), preeclampsia with a normally grown fetus (n=35), preeclampsia with fetal growth restriction (preeclampsia with a normally grown fetus–fetal growth restriction, n=42), and 111 uncomplicated pregnancies matched by gestational age at ultrasound. Fetal echocardiography was performed at diagnosis for cases and recruitment for uncomplicated pregnancies. Cord blood concentrations of B-type natriuretic peptide and troponin I were measured at delivery. Univariate and multiple regression analysis were conducted.


Pregnancies complicated by preeclampsia and/or fetal growth restriction showed similar patterns of fetal cardiac remodeling with larger hearts (cardiothoracic ratio, median [interquartile range]: uncomplicated pregnancies 0.27 [0.23–0.29], fetal growth restriction 0.31 [0.26–0.34], preeclampsia with a normally grown fetus 0.31 [0.29–0.33), and preeclampsia with fetal growth restriction 0.28 [0.26–0.33]; P <.001) and more spherical right ventricles (right ventricular sphericity index: uncomplicated pregnancies 1.42 [1.25–1.72], fetal growth restriction 1.29 [1.22–1.72], preeclampsia with a normally grown fetus 1.30 [1.33–1.51], and preeclampsia with fetal growth restriction 1.35 [1.27–1.46]; P =.04) and hypertrophic ventricles (relative wall thickness: uncomplicated pregnancies 0.55 [0.48–0.61], fetal growth restriction 0.67 [0.58–0.8], preeclampsia with a normally grown fetus 0.68 [0.61–0.76], and preeclampsia with fetal growth restriction 0.66 [0.58–0.77]; P <.001). Signs of myocardial dysfunction also were observed, with increased myocardial performance index (uncomplicated pregnancies 0.78 z scores [0.32–1.41], fetal growth restriction 1.48 [0.97–2.08], preeclampsia with a normally grown fetus 1.15 [0.75–2.17], and preeclampsia with fetal growth restriction 0.45 [0.54–1.94]; P <.001) and greater cord blood B-type natriuretic peptide (uncomplicated pregnancies 14.2 [8.4–30.9] pg/mL, fetal growth restriction 20.8 [13.1–33.5] pg/mL, preeclampsia with a normally grown fetus 31.8 [16.4–45.8] pg/mL and preeclampsia with fetal growth restriction 37.9 [15.7–105.4] pg/mL; P <.001) and troponin I as compared with uncomplicated pregnancies.


Fetuses of preeclamptic mothers, independently of their growth patterns, presented cardiovascular remodeling and dysfunction in a similar fashion to what has been previously described for fetal growth restriction. Future research is warranted to better elucidate the mechanism(s) underlying fetal cardiac adaptation in these conditions.

Preeclampsia (PE) and fetal growth restriction (FGR) represent a major concern in public health, affecting 2–8% and 5–10% of all pregnancies, respectively, and being a leading cause of perinatal morbidity and mortality. Both syndromes share some pathophysiologic features, with a variable involvement of placental insufficiency and maternal cardiovascular maladaptation. Moreover, they occur concurrently in a nonignorable proportion, which could be different according to the population characteristics. In fact, every fifth case of PE also presents with FGR and about 50% of early-onset FGR cases will eventually coexist with PE ; however, in populations with a high burden of obesity among pregnant women, the prevalence of PE is extremely high, with few cases of FGR.

AJOG at a Glance

Why was this study conducted?

To answer the question: is there any fetal heart affectation in preeclampic pregnancies with normally grown fetuses? And, if there are any signs of fetal cardiac remodeling or dysfunction, are they similar to those described previously in fetal growth restriction?

Key findings

We recruited well-characterized pregnancies complicated by preeclampsia (with or without fetal growth restriction), normotensive fetal growth restriction, and uncomplicated pregnancies. Fetal hearts in complicated pregnancies were larger, more spherical, and thicker, with a similar pattern of fetal cardiac remodeling in preeclampsia and fetal growth restriction. Moreover, cardiac dysfunction was demonstrated by greater myocardial performance index, and cord blood B-type natriuretic peptide and troponin I, both in preeclampsia and fetal growth restriction.

What does this add to what is known?

This is the first study to demonstrate that pregnancies with preeclampsia and a normally grown fetus are associated with cardiac remodeling and dysfunction in a similar fashion to what has previously been described in fetal growth restriction.

Besides their association with perinatal morbidity and mortality in the short term, accumulating evidence suggests the existence of long-term cardiovascular consequences on both the mother and the offspring. Concerning the offspring, long-term effects are thought to be explained by fetal adaptations to adverse intrauterine conditions, including metabolic and cardiovascular programming. Recent studies consistently have demonstrated structural and functional cardiovascular changes in growth-restricted fetuses that persist into the postnatal life through infancy, childhood, and adolescence, supporting epidemiologic and experimental evidence linking low birthweight and cardiovascular morbidity and mortality later in life. In contrast, offspring from preeclamptic pregnancies showed cardiac structural and functional changes and greater blood pressure in childhood and adolescence. However, most of these studies did not include pure phenotypes of PE and FGR, ie, studies on FGR included pregnancies complicated by PE and vice versa, which prevents to differentiate the independent effect of each condition on the fetal heart.

Thus, the aim of this study was to evaluate the independent effect of PE and FGR on fetal cardiac structure and function. Well-characterized pregnancies with normotensive FGR, PE associated with FGR, and PE with a normally grown fetus were evaluated by echocardiography and cord blood myocardial biomarkers and compared with uncomplicated pregnancies.

Materials and Methods

Study population

This was a prospective, observational study including singleton pregnancies with a diagnosis of PE and/or FGR who attended the Departments of Maternal-Fetal Medicine at BCNatal (Barcelona, Spain) between July 2016 and December 2017. FGR was defined as estimated fetal weight (EFW) and birthweight below the 10th centile associated with either abnormal cerebroplacental ratio (<5th centile) or abnormal uterine arteries mean pulsatility index (PI) (>95th centile), or birthweight below the 3rd centile. EFW and birthweight centiles were assigned according to local standards. PE was defined as high blood pressure (systolic blood pressure ≥140 mm Hg and/or diastolic blood pressure ≥90 mm Hg on 2 occasions, at least 4 hours apart), developed after 20 weeks of gestation, with proteinuria (≥300 mg/24 hours or protein/creatinine ratio ≥0.3). Uncomplicated pregnancies with normotensive mothers and appropriate growth for gestational age fetuses—defined as EFW and birthweight above the 10th centile—were selected randomly from our general population to be included as controls and frequency paired with cases by gestational age at fetal echocardiography (±2 weeks). In all pregnancies, gestational age was calculated based on the crown–rump length at first trimester ultrasound. Pregnancies with chromosomal/structural anomalies or intrauterine infection were excluded. The study protocol was approved by the local ethics committee (HCB/2016/0253), and patients agreeing to participate provided their written informed consent.

Data collection and study protocol

The following data were recorded on enrollment: maternal age, ethnicity, body mass index, known chronic disease (ie, hypertension, diabetes mellitus), parity, obstetric history, mode of conception, and smoking status. Fetoplacental Doppler, EFW, and fetal echocardiographic parameters were obtained at diagnosis (or at enrollment for controls). Ultrasound studies were performed using a Siemens Sonoline Antares (Siemens Medical Systems, Malvern, PA) or a Voluson 730 Expert (GE Medical Systems, Milwaukee, WI) with 6–4-MHz linear curved-array probes. EFW was calculated using the Hadlock formula and centile based on local reference curves. All estimations were done in the absence of fetal movements and, when required, with the mother in voluntary suspended respiration. An angle of insonation of <30° between the vessel and the Doppler beam was accepted for analysis. The mechanical and thermal indices were maintained below 1, and the wall filter was set to 70 Hz.

Fetoplacental Doppler parameters were obtained from 3 or more successive waveforms in each vessel. Doppler examination included uterine arteries, the umbilical artery (UA), fetal middle cerebral artery (MCA), and aortic isthmus. Uterine artery PI was calculated as the average PI of the right and left arteries. UA-PI was measured from a free loop of the umbilical cord. MCA-PI was measured distal to the junction of the internal carotid artery in a transverse view of the fetal skull at the level of the circle of Willis. The cerebroplacental ratio was calculated as MCA-PI/UA-PI. The aortic isthmus PI was sampled downstream of the left subclavian artery and just upstream of the ductus arteriosus connection in a sagittal view simultaneously visualizing the aortic arch. Fetal echocardiography also was performed in all cases and controls, as described to follow.

At delivery, gestational age, birthweight, birthweight centile, Apgar scores, umbilical artery pH, admissions to the neonatal intensive care unit, and perinatal mortality were recorded. Those pregnancies with missing perinatal outcomes, unconfirmed diagnosis, or who delivered a small newborn not fulfilling the definition of FGR (ie, birthweight between the 3rd and the 10th centile associated with normal cerebroplacental ratio [≥5th centile] and normal uterine arteries PI [≤95th centile]) were excluded from further analysis, as shown in Figure 1 . In addition, cord blood from umbilical vein was collected at delivery to measure the concentration of B-type natriuretic peptide (BNP) and troponin I.

Figure 1

Recruitment flow chart

Fetal echocardiography data were collected from all the patients who accepted to participate in the study. Some of them were excluded from further analysis according to confirmed birthweight and perinatal outcomes that did not meet the inclusion criteria.

AGA , fetuses with appropriate growth for gestational age; FGR , fetal growth restriction.

Youssef et al. Fetal cardiac remodeling and dysfunction is associated with both preeclampsia and fetal growth restriction. Am J Obstet Gynecol 2020 .

Fetal echocardiography

Fetal echocardiography included a comprehensive examination to assess structural heart integrity and rule out cardiac defects following standard protocols. Then, fetal cardiac morphometry and function were evaluated. All the scans were performed by experienced sonographers in fetal echocardiography who were blinded to the study hypothesis. Researchers who performed the offline analysis of the echocardiographic images were blinded to the diagnosis.

Cardiac and ventricular dimensions were measured on 2-dimensional images from an apical or basal 4-chamber view at end-diastole. The cardiothoracic ratio was calculated as heart area divided by thoracic area. Left and right ventricular sphericity indices were calculated as base-to-apex length divided by basal diameter. Ventricular end-diastolic septal wall thicknesses were measured from a transverse 4-chamber view. Relative septal wall thickness was calculated as 2 times septal wall thickness divided by the left ventricular diastolic diameter. Ductus venosus PI was measured either in a mid-sagittal view of the fetal thorax or in a transverse plane through the upper abdomen prior to its entrance into the inferior vena cava, positioning the Doppler gate at the ductus venosus isthmic portion. Left myocardial performance index was obtained in a cross-sectional image of the fetal thorax, placing the Doppler sample volume on the medial wall of the ascending aorta and including the leaflets of the aortic and mitral valves. The clicks of the valves registered in the Doppler trace were used as landmarks to calculate the following time periods: isovolumetric contraction time from the closure of the mitral valve to the opening of the aortic valve, ejection time from the opening to the closure of the aortic valve, and isovolumetric relaxation time from the closure of the aortic valve to the opening of the mitral valve. Finally, the myocardial performance index was calculated as (isovolumetric contraction time + isovolumetric relaxation time) / ejection time, and normalized into z scores.

Placental evaluation

Placentas were fixed in 10% buffered formalin. Trimmed placentas were weighted, and samples were taken for routine processing adhering to a standard laboratory protocol. Diagnostics from pathology reports were classified according to Redline’s classification following 2014 Amsterdam Placental Workshop Group Consensus Statement.

Cord blood sampling and biomarkers assessment

Umbilical cord ethylenediaminetetraacetic acid–treated blood was obtained from the umbilical vein after cord clamp at delivery. Plasma was separated by centrifugation at 1500 g for 10 minutes at 4°C, and samples were immediately stored at −80°C until assayed. Concentrations of BNP and troponin I were measured using Siemens ADVIA Centaur BNP and Centaur CP troponin I assays, respectively. This analysis was performed in 50 cord samples from uncomplicated pregnancies and 30 samples from each group of the cases.

Statistical analysis

Data were analyzed with the statistical software STATA 14.2 (StataCorp LLC, College Station, TX). The study outcome was fetal cardiovascular assessment. The independent variable of interest was the presence of PE and/or FGR, and the covariates were the presence of chronic hypertension, diabetes, the mode of conception (via assisted reproductive technologies vs natural), and smoking during pregnancy. A sample size of 23 patients in each group of the cases and 69 controls was calculated by expecting 1 z-score differences in the myocardial performance index between cases and controls, for a given 5% α error and 80% power and 1:3 sampling ratio.

Results were expressed as median (interquartile range) or percentage as appropriate. Statistical analysis included the use of Student t or Mann–Whitney U tests and Pearson χ 2 test for continuous and categorical variables, respectively, to compare each group of the cases vs the controls. To evaluate the influence of covariates, comparisons of the cardiovascular parameters were adjusted for the presence of chronic hypertension, diabetes, assisted reproductive technologies, smoking, and fetal sex by multiple regression analyses. In addition to these covariates, BNP and troponin I levels also were adjusted for gestational age at sampling and the rate of cesarean deliveries for fetal distress. All reported P values are 2 sided. Differences were considered significant when P <.05.


Baseline and perinatal characteristics of the study population

A total of 224 pregnancies were included in data analysis. Baseline characteristics and perinatal outcomes are shown in Table 1 . The study groups were similar in terms of maternal baseline characteristics, except for significantly greater prevalence of chronic hypertension and pregestational diabetes among cases, greater proportion of conception by assisted reproductive technologies and maternal pregestational body mass index in preeclamptic mothers, and more smoking women in normotensive FGR as compared with controls. As expected, EFW, birthweight, and weight centiles were lower together with worse fetoplacental Doppler in FGR groups as compared with controls. In addition, gestational age at delivery was earlier in PE and/or FGR, with greater rates of cesarean deliveries and admissions to neonatal intensive care unit. FGR groups presented lower placental weight, with increased malperfusion lesions in the maternal side in PE with FGR group.

Table 1

Maternal, fetoplacental ultrasound, perinatal characteristics, and placental histopathologic findings of the study population

Normotensive AGA n=111 Normotensive FGR n=36 Preeclampsia AGA n=35 Preeclampsia FGR n=42
Maternal characteristics
Age, y 33.7 (30.7–36.7) 33.2 (30–37) 35.4 (32.4–38.1) 35.1 (32.3–37.6)
Caucasian ethnicity 55 (49.6) 24 (66.7) 20 (57.1) 21 (50)
Pregestational BMI, kg/m 2 22.5 (20.6–25.5) 22.3 (19.6–26.2) 23.1 (22.4–28) a 24.7 (21.3–27) a
Chronic hypertension 0 (0) 2 (5.6) a 6 (17.1) a 6 (14.3) a
Pregestational diabetes 0 (0) 2 (5.6) a 9 (25.7) a 3 (7.1) a
Nulliparity 64 (57.7) 16 (44.4) 21 (60) 26 (61.9)
Assisted reproductive technologies 3 (2.8) 0 (0) 5 (14.3) a 5 (11.9) a
Smoking during pregnancy 7 (6.3) 7 (20) a 3 (8.6) 5 (11.9)
Fetoplacental ultrasound
Gestational age at assessment, wk 33.4 (29.3–37.6) 33.1 (30.1–35.7) 34.1 (31.3–36.6) 32.5 (30.3–34.9)
Estimated fetal weight, g 2053 (1385–3086) 1152 (834–825) a 2077 (1674–2977) 1298 (699–1718) a
Estimated fetal weight centile 58 (36–78) 2 (0–5) a 30 (10–89) 2 (0–6) a
Uterine arteries mean PI (z score) –0.05 (–1.04 to 0.47) 1.8 (0.21 to 2.9) a 0.14 (–1.2 to 1.26) 2.98 (1.96 to 3.53) a
Umbilical artery PI (z score) –0.22 (–0.57 to 0.12) 0.44 (–0.92 to 1.23) a –0.05 (–0.59 to 0.49) 0.47 (–0.04 to 1.65) a
Middle cerebral artery PI (z score) –0.04 (–0.65 to 0.48) –0.77 (–1.67 to 0.15) a –0.14 (–0.57 to 0.33) –0.83 (–1.71 to –0.4) a
Cerebroplacental ratio (z score) –0.27 (–0.8 to 0.49) –1.17 (–2.24 to 0.33) a –0.46 (–1.15 to 0.13) –1.3 (–1.94 to –0.67) a
Aortic isthmus PI (z score) –0.31 (–1.04 to 0.46) 0.01 (–0.32 to 0.64) –0.75 (–1.41 to –0.07) 0.12 (–0.91 to 1.44)
Perinatal outcomes
Gestational age at delivery, wk 40.1 (39.1–40.9) 37.2 (34.9–37.6) a 37.1 (34.7–37.7) a 34.2 (32–35.9) a
Cesarean delivery 23 (18.7) 18 (50) a 26 (74.3) a 34 (72.3) a
Emergency cesarean delivery for fetal distress 2 (1.8) 5 (13.9) a 5 (14.3) a 6 (14.3) a
Male sex 45 (40.5) 19 (52.8) 13 (37.1) 24 (57.1)
Birthweight, g 3354 (3092–3650) 1994 (1665–2254) a 3010 (2470–3270) 1548 (1160–1830) a
Birth weight centile 49 (26–75) 1 (0–1.5) a 57 (21–85) 0 (0–1) a
Apgar score, 5 min, <7 0 (0) 0 (0) 0 (0) 6 (14.3) a
Umbilical artery pH 7.21 (7.15–7.26) 7.2 (7.13–7.25) 7.19 (7.14–7.23) 7.21 (7.11–7.23)
Admission to neonatal intensive care unit 4 (3.6) 17 (47.2) a 13 (38.2) a 35 (83.3) a
Perinatal mortality 0 (0) 0 (0) 0 (0) 4 (9.5) a
Placental weight and vascular histopathologic findings
Placental weight, g 465 (405–530) 293 (250–340) a 503 (365–590) 260 (205–325) a
Maternal side lesions
Maldevelopment, n (%) 0 (0) 3 (9.1) a 2 (6.5) 1 (2.4)
Malperfusion, n (%) 14 (15.6) 7 (21.2) 8 (25.8) 26 (61.9) a
Loss of integrity, n (%) 2 (2.2) 4 (12.1) 2 (6.5) 0 (0)
Fetal side lesions
Maldevelopment, n (%) 0 (0) 0 (0) 1 (3.2) 0 (0)
Malperfusion, n (%) 4 (4.4) 2 (6.1) 0 (0) 5 (11.9)
Loss of integrity, n (%) 10 (11.1) 2 (6.1) 6 (19.4) 4 (9.5)

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Aug 21, 2020 | Posted by in GYNECOLOGY | Comments Off on Fetal cardiac remodeling and dysfunction is associated with both preeclampsia and fetal growth restriction

Full access? Get Clinical Tree

Get Clinical Tree app for offline access