1

Descriptive analysis of the different phenotypes of cardiac

remodeling in fetal growth restriction

Short version of title: Cardiac Phenotypes

Mérida Rodríguez-López, MD MsC1, Mónica Cruz-Lemini, MD PhD1-2, Brenda

Valenzuela-Alcaraz, MD1, Laura Garcia-Otero MD1, Marta Sitges, MD PhD 3, Bart

Bijnens, PhD 4,5, Eduard Gratacos, MD PhD 1*, Fatima Crispi, MD PhD 1

Affiliations:

1BCNatal - Barcelona Center for Maternal-Fetal and Neonatal Medicine (Hospital

Clínic and Hospital Sant Joan de Déu), IDIBAPS, University of Barcelona, and Centre

for Biomedical Research on Rare Diseases (CIBER-ER), Barcelona, Spain

2Fetal Medicine and Surgery Research Unit, Children and Women's Hospital of

Querétaro, Mexico; Neurodevelopmental Research Unit "Dr. Augusto Fernández

Guardiola", Neurobiology Institute, Universidad Nacional Autónoma de México

(UNAM) Campus Juriquilla, Querétaro, Mexico.

3Cardiovascular Institute, Hospital Clinic, IDIBAPS, Barcelona, Spain;

4ICREA, Barcelona, Spain

5Universitat Pompeu Fabra, Barcelona, Spain.

*Corresponding author:

Eduard Gratacós, MD, PhD.

Department of Maternal-Fetal Medicine

BCNatal – Barcelona Center for Maternal-Fetal and Neonatal Medicine (Hospital Clínic

2

and Hospital Sant Joan de Déu). Sabino de Arana 1, 08028 Barcelona, Spain.

Phone: +34 93 227 9906; Fax: +34 93 227 5612. E-mail: gratacos@clinic.ub.es

Keywords: Fetal growth retardation; Ventricular Cardiac Remodeling;

Echocardiography; Phenotype.

Referees:

1. Jan Deprest, e-mail: Jan.Deprest@uzleuven.be, KU-Leuven University

2. Tim van Mieghen, e-mail: t.vanmieghem@gmail.com, KU-Leuven University

3. Ahmed Baschat, e-mail: ahmetbaschat@gmail.com, Johns Hopkins University

3

ABSTRACT

Objective: To identify the presence of different cardiac phenotypes among FGR.

Study Design: Fetal echocardiography was performed in 126 FGR (defined as birth

weight <10th centile) and 64 adequate for gestational age (AGA). Principal component

and cluster analyses were performed to identify different cardiac phenotypes among

FGR cases.

Results: Three different cardiac phenotypes were identified within FGR: globular,

elongated and hypertrophic. Most FGR cases (54%) were characterized by a ‘globular’

heart with the lowest left ventricular sphericity index (controls: median 1.78

(interquartile range 1.62-1.97), FGR-elongated: 1.92 (1.78-2.09), FGR-globular 1.44

(1.36-1.52) and FGR-hypertrophic 1.65 (1.42-1.77), P=0.001), while 29% of the cases

showed an ‘elongated’ left ventricle with nearly normal cardiac dimensions. Finally,

17% of the FGR showed a ‘hypertrophic’ phenotype with the highest values in left

ventricular wall thickness (controls: 1.22 mm/kg (1.1-1.67), FGR-elongated: 1.52 (1.28-

1.86), FGR-globular 1.65 (1.39-1.99) and FGR-hypertrophic 3.68 (3.45-4.71), P=0.001)

and cardiac dimensions. The globular and elongated phenotype showed fetoplacental

profile of late-onset FGR while the hypertrophic phenotype showed signs of early-onset

FGR The hypertrophic cluster also showed the worst perinatal results presenting the

lowest birthweight centile, gestational age at birth, Apgar score, and the highest

postnatal blood pressure and carotid intima media thickness.

Conclusions: FGR induces at least 3 different cardiac phenotypes, with early-onset

FGR cases associated with a hypertrophic response and worse perinatal outcomes. This

cardiac phenotypic classification may improve identification of those FGR cases with

the highest perinatal and long-term cardiovascular risk.

4

INTRODUCTION

Fetal growth restriction (FGR), defined as a failure to achieve its growth

potential, affects 7-10% of pregnancies1. It represents a leading cause of perinatal

morbidity and mortality1 and also a risk factor for long-term disabilities such as

cognitive impairment2 and cardiovascular diseases in adulthood3. FGR may arise at any

gestational age, either as the early forms, typically related to preeclampsia and severe

placental insufficiency, or as late-onset cases mostly associated with mild or subtle

forms of placental insufficiency. Both early and late-onset cases present cardiovascular

remodeling pre4, 5 and postnatally6, 7.

Echocardiographic signs of fetal systolic and diastolic dysfunction5, 8, 9 have been

demonstrated both in early and late FGR. Cardiac dysfunction usually reflects a cardiac

remodeling process, defined as a primary structural/shape change of the heart in order to

adapt to an insult. However, while there is consistency in reporting cardiovascular

remodeling, the specific structural changes of the fetal heart in FGR are still poorly

defined. Studies from the 90s suggested cardiomegaly as a typical sign of severe FGR10,

but increased4, 11, preserved12, 13 or decreased cardiac dimensions14 have been also

reported. The presence of myocardial hypertrophy is also controversial with results

suggesting increased4, 11, 15, lower14 or even normal16 wall thickness. Recently, a more

spherical cardiac shape has been reported as a typical feature in FGR5, 6, 8. We

hypothesized that FGR could present as different cardiac phenotypes, that might explain

this variability

The main objective of this study was to explore the existence of heterogeneity

expressed by different cardiac remodeling phenotypes within FGR, and to evaluate

5

whether these phenotypes have different perinatal and postnatal characteristics. To this

aim, fetal echocardiography was conducted in a large cohort of FGR cases and cluster

analysis was performed to identify cardiac morphometric phenotypes among FGR.

Then, perinatal and postnatal characteristics among the different cardiac clusters were

compared to normally grown fetuses. Finally, the discriminatory accuracy of

conventional feto-placental ultrasound was tested for the prediction of the cardiac

phenotypes.

6

MATERIALS AND METHODS Study design and participants

A prospective cohort study was conducted in FGR and adequate growth for

gestational age (AGA) fetuses, recruited in a tertiary university hospital from February

2010 to May 2012. FGR was defined by estimated fetal weight and confirmed birth

weight <10th centile, and AGA by birth weight between 10th-90th centile using local

growth curves17. Gestational age was calculated according to crown-rump length at first

trimester scan. Fetuses with structural or chromosomal abnormalities, multiple

pregnancies, pregestational diabetes, evidence of fetal infection, conceived by assisted

reproductive technologies or macrosomia (birth weight >95th centile) were excluded.

FGR cases who died in the perinatal period were excluded from the cluster analysis and

their results shown separately in the Supplementary material. Eighty percent of the

fetuses included in the present study were previously included in another publication

from our group4. The study protocol was approved by the local ethical committee and a

written consent form was obtained from parents.

Fetal ultrasound

Fetal ultrasound was performed using curved-array 2-6 MHz and phased-array

2-10 MHz transducers with a Siemens Sonoline Antares machine (Siemens Medical

Systems, Malvern, PA) using a standardized protocol4 as described below. It included

the assessment of estimated fetal weight, conventional feto-placental Doppler

parameters (Supplementary material), and fetal cardiac morphometry and function. All

measurements were performed simultaneously.

Fetal cardiac morphometry included the measurement of cardiac area, atrial

7

dimensions and ventricular cavity dimensions, sphericity indices and wall thickness.

Cardiac area was delineated in end-diastole from a 4-chamber view on 2D using the

area method4. Left and right atrial areas were delineated on 2D images from an apical or

basal 4-chamber view at ventricular end-systole18. Left and right ventricular sphericity

indices (LVSI and RVSI) were calculated by dividing the end-diastolic base-to-apex

ventricular length by the transverse ventricular diameters measured in 2D from an apical

or basal 4-chamber view. Left and right transverse ventricular diameters and wall

thickness were measured in M-mode from a transverse 4-chamber view at end-

diastole19. Relative wall thickness was calculated as (2*left ventricular free wall

thickness)/left ventricular transverse diameter. Right/ Left ventricular ratio was

calculated by dividing both transverse diameters. All cardiac morphometric

measurements (with the exception of sphericity indexes and relative wall thickness)

were normalized by dividing for estimated fetal weight.

Assessment of fetal cardiac function included the measurement of shortening

fractions, mitral and tricuspid annular plane systolic excursions (MAPSE and TAPSE)

and peak velocities, and left isovolumic relaxation time (IRT). Left and right shortening

fractions were obtained by M-mode from a transverse 4-chamber view and calculated

as: (ventricular diastolic diameter - ventricular systolic diameter)/ventricular diastolic

diameter20. MAPSE and TAPSE were calculated using real time M-mode from an apical

or basal 4-chamber view, measuring the maximum displacement of the valvular rings

between end-systole and end-diastole. Spectral tissue Doppler was applied in mitral and

tricuspid annulus to measure systolic (S’) and early diastolic (E’) peak velocities. Left

IRT was evaluated using conventional Doppler in a 4-chamber view measuring the time

interval from the closure of the aortic valve to the opening of the mitral valve using the

8

valve clicks as previously described21. All functional parameters (with the exception of

shortening fractions) were normalized into z-scores for gestational age22-24.

Perinatal and postnatal outcome

Perinatal characteristics including the presence of pregnancy complications,

delivery and neonatal data were collected from clinical records. Postnatal assessment

was planned around 6 months of corrected age. Infant’s anthropometric data was

collected by a trained nurse. Weight gain was calculated by subtracting birth weight

from infant weight. Infant’s systolic and diastolic blood pressures from the brachial

artery were measured by a trained physician using a validated ambulatory automated

device (Omron 5 Series; Omron Corporation, Kyoto, Japan). Finally, infant’s aortic

intima-media thickness (aIMT) was measured by ultrasound using a 12L-RS linear-

array 6.0-13.0 MHz transducer with a Vivid Q (General Electric Healthcare, Horten,

Norway) ultrasound equipment. Longitudinal clips of the far wall of the proximal

abdominal aorta in the upper abdomen were obtained, and aIMT measurements were

performed offline based on the trace method using commercially available software (GE

EchoPAC PC 108.1.x; General Electric Healthcare).

Statistical analysis

All feto-placental and echocardiographic parameters were normalized into z-

scores (when normality ranges were available) or by estimated fetal weight. Statistical

analysis was done using Stata IC version 14 (StataCorp. LP, College Station,TX). First,

principal component analysis was performed to select the most important fetal cardiac

morphometric parameters. One variable from each component was selected to create

9

the clusters based on the factor loading and its potential clinical relevance. Then, an

exploratory cluster analysis among FGR cases was performed. Cluster analysis is an

exploratory and data mining technique for identifying heterogeneity by grouping

patients25 in such a way that the individuals in the same group are very similar but

different from those in other groups. Hierarchical cluster analysis was performed to

visualize possible subgroups in a dendrogram graph. Then, non-hierarchical K-means

method was applied to confirm the number of clusters and to create subgroups. Finally,

bootstrapping resampling method was applied to validate the clusters results.

Once the clusters were determined, a descriptive analysis including the fetal

ultrasonographic, perinatal and postnatal characteristics was performed to assess the

differences between FGR clusters and AGA. In order to overcome the different

gestational age between groups, all cardiac parameters were normalized into Z-scores or

by estimated fetal weight. Data was expressed as mean ± standard deviation (SD),

median (interquartile range) or relative frequencies and compared to controls using

Student’s t-test, Wilcoxon-Mann-Whitney or χ2 test as appropriated. Differences among

FGR clusters were assessed by Kruskal-Wallis test, F-anova or χ2 test. To consider

multiple comparisons in all cardiac morphometric variables among FGR clusters,

multiple analysis of variance (MANOVA) was also performed. Baseline,

ultrasonographic and perinatal characteristics of the FGR group as a whole and AGA

are shown in the Supplementary Material, Table 1S, 2S.

Finally, logistic regression was used to estimate the discriminatory accuracy of

prenatal ultrasonographic data to identify the cluster with highest perinatal risk. Areas

under the receiver operating characteristic curves (AUC) were compared using DeLong

method.

10

RESULTS

Cardiac phenotypes within FGR

Left ventricular sphericity index and left ventricular free wall thickness were

selected by principal component analysis as the most representative echocardiographic

parameters in FGR (Supplementary Material, Table 3S). Cluster analysis using

hierarchical approach and Wards linkage identified 3 different subgroups among FGR

cases (Figure 1). K-means also suggested that 3 clusters could better represent the

differences among subgroups. Validation of the cluster by bootstrapping showed similar

results with 95.23% of agreement. Wilks' lambda test after MANOVA showed that

altogether morphometry parameters differed among clusters (p-value<0.001)

Figure 2 and Table 1 display the echocardiographic characteristics of the 3

phenotypes identified within FGR. Most FGR cases (54%) were characterized by a

‘globular’ heart with the lowest ventricular sphericity index, but also increased cardiac

dimensions and wall thicknesses. Secondly, 29% of the FGR cases showed an

‘elongated’ left ventricle (highest sphericity index) with nearly normal right ventricular

morphometry, decreased right/left ventricular ratio, and increased cardiac area and left

wall thicknesses. Finally, 17% of the FGR presented a ‘hypertrophic’ phenotype with

the highest wall thickness and cardiac dimensions values together with reduced

sphericity indices. All cardiac phenotypes presented reduced longitudinal motion

(measured by M-mode and tissue Doppler) and prolonged IRT as compared to AGA.

Comparison between AGA and each different cardiac phenotype group at a similar

gestational age showed similar results (Supplementary material).

Perinatal and postnatal characteristics of the different FGR cardiac phenotypes

11

Characteristics of the FGR subgroups as compared to AGA are shown in Table

2. Similar results were observed when each cluster was compared with those AGA with

comparable gestational age (Supplementary material, Tables 4S 5S). All FGR groups

showed significant changes in gestational age, fetal biometry, feto-placental Doppler,

prevalence of preeclampsia and infants’ height, weight, blood pressure and aIMT.

Comparison among FGR subgroups revealed that the hypertrophic phenotype presented

the worst feto-placental Doppler, the lowest gestational age at birth(95% of fetuses with

this phenotype were born preterm), birth weight(all cases<2200g), birthweight centile

(all cases <1st), 1 minute Apgar score, and highest prevalence of corticoid exposure,

preeclampsia and fetal distress during labor. The hypertrophic group also showed the

most significant postnatal changes with the lowest infants’ height, weight and body

mass index, and the highest weight gain, blood pressure and aIMT values.

Prediction of the hypertrophic phenotype

As the hypertrophic cluster showed the worst perinatal and postnatal outcomes,

we explored the potential predictive value of standard fetal ultrasonographic

characteristics for identifying this phenotype. Whereas all fetal ultrasonographic

parameters were significantly altered in the hypertrophic cluster, the highest

discriminatory accuracy was achieved by estimated fetal weight (AUC 0.98 (95%

confidence interval 0.95-0.99)), gestational age at scan (AUC 0.96 (0.93-0.99)) and

cerebroplacental ratio (CPR) (AUC 0.80 (0.67-0.93)). Figure 3 shows the distribution of

estimated fetal weight, gestational age at scan and CPR among the hypertrophic cluster

as compared to the other FGR subgroups. A cutoff value of 1500g in estimated fetal

weight showed 100% sensitivity and 87.62% specificity for predicting the hypertrophic

12

phenotype. A cutoff of 34 weeks at scan showed 100% sensitivity and 84.7% of

specificity. A CPR value below the 5th centile showed a sensitivity 80.9% and

specificity 73.3% for identifying the hypertrophic group. A composite score combining

estimated fetal weight (g) + gestational age (weeks) + 1.06 CPR (z-scores) + 2.70

ductus venosus PI (z-scores) + 1.08 aortic isthmus flow index (z-scores) could achieve

an AUC of 0.99 (95% confidence interval 0.97-1.00) (Supplementary material, Table

6S, Figure 1S).

13

DISCUSSION

In this study, we describe for the first time the presence of three different fetal

cardiac remodeling phenotypes within FGR. In particular, early-onset FGR was

associated to a ‘hypertrophic’ response and worse perinatal and postnatal outcomes,

while late-onset FGR displayed ‘globular’ or ‘elongated’ phenotypes with better

outcomes.

The most frequent cardiac adaptations observed amongst FGR were the ‘globular’

(56%) and ‘elongated’ (29%) phenotypes characterized by the most spherical geometry

affecting both ventricles and by a non-spherical left ventricle, respectively. These

phenotypes were associated with a mild but significant increase in fetal cardiac

dimensions and thickness together with subtle cardiac dysfunction and preserved ductus

venosus flow. Both phenotypes presented similar clinical and perinatal characteristics.

They were related to late-onset forms of FGR with mild signs of placental insufficiency

and slightly worse perinatal outcome as compared to controls. They also showed a small

increase in blood pressure and aIMT at 6 months after birth. These changes can be most

probably explained by mild placental insufficiency leading to subtle fetal hypoxia,

undernutrition and pressure overload during the last weeks of pregnancy in late FGR. A

normal ventricle is largely ellipsoid, with larger radius of curvature in the longitudinal

versus the circumferential direction. This implies that the wall stresses in the

longitudinal direction are much larger than the circumferential ones. If the shape

changes from ellipsoid to sphere, the radius of curvature (and thus the wall stress) in

the circumferential direction increases, but the radius of curvature in the longitudinal

direction decreases, thus decreasing the longitudinal (=maximal) wall stress. Therefore,

when a ventricle becomes more spherical, the wall stresses become more similar in all

14

directions and while the minimal wall stresses might increase, the maximal

(longitudinal) ones will decrease, thus keeping them within a range where the myocytes

can still appropriately function without a need for increased force development

(=hypertrophy)26. The fetal heart, given its flexibility may adapt by permanently

changing to a more spherical shape of one (the ‘elongated’ phenotype with the right

ventricle being more globular and thus pushing the septum towards the left ventricle and

therefore making it more elongated) or both ventricles (‘globular’ phenotype) thus

reducing wall stress to better tolerate pressure overload. These findings are consistent

with previous studies suggesting more globular hearts with subtle reduction of

longitudinal motion and impaired relaxation in FGR fetuses and children born near

term4, 5,. Future studies are warranted to better understand the differential cardiac

response (‘globular’ vs. ‘elongated’) among late-onset forms of FGR.

This study also described a less prevalent (17%) but more prominent cardiac adaptation

amongst FGR. The ‘hypertrophic’ cluster presented extensive myocardial hypertrophy,

cardiomegaly, an spherical shape and systolic and diastolic dysfunction. This cluster

was the most dissimilar to AGA and the most similar to those FGR who died in the

perinatal period (Supplementary Material). This cardiac adaptation was related to early-

onset FGR, abnormal feto-placental Doppler, the worst perinatal outcomes and the

highest values of postnatal blood pressure and aIMT, suggesting a higher long-term

cardiovascular risk for this subgroup. Most probably this cardiac adaptation reflects a

more severe and/or prolonged exposure to placental insufficiency requiring several

mechanisms for compensating. The initial reduction in sphericity index might not be

enough to compensate hypoxia and increased placental resistance. Then, hypertrophy

15

will be induced to increase contractility as well as decrease local wall stress; and

cardiomegaly might appear to cope with persistent volume overload26. This response of

early FGR is consistent with reports of the 90s (mainly focused on early-onset FGR)

suggesting cardiomegaly and ‘heart-sparing effect’ as a typical sign of FGR10, 11, and

also with more recent studies demonstrating increased myocardial wall thickness and

more spherical ventricles in FGR cases4, 11,15,and with data from experimental models

(usually mimicking severe forms)27, 28. Theoretical inconsistencies with previous studies

suggesting smaller and non-hypertrophic hearts could be explained by the lack of

normalization for body size, by the definition of FGR (mixing truly restricted and

constitutionally small fetuses) and by reporting means/medians between groups which

are more influenced by the values of the extreme and more prevalent cases. Our result

posits heterogeneity of cardiac remodeling as a plausible explanation for such

inconsistencies.

In the present study, a composite score including estimated fetal weight,

gestational age at scan and CPR permitted a high accuracy for identifying the cases with

a ‘hypertrophic’ response. This opens the opportunity for the detection of those FGR

cases with the worst fetal deterioration, cardiovascular adaptation, perinatal and

postnatal outcome.

This study has some strengths and limitations. The comprehensive fetal

echocardiographic assessment and the large sample size of cases enabled to identify

three different cardiac phenotypes within FGR. To our knowledge, this is the first

attempt to explore different cardiac phenotypes in fetuses by using cluster analysis. It

has advantages over the pure description of means/medians by allowing to identify

clinical heterogeneity and understand why some patients might be triggered to disease

16

while others (with a similar risk factor) did not. However, this is a non-inferential

technique that needs to be validated in other populations. Despite adjustment for

gestational age and fetal size, a residual confounding effect of gestational age might not

be completely rule out. Future studies are warranted to evaluate the postnatal

persistence of these cardiac phenotypes.

In conclusion, this study highlights the variability among the FGR population.

We demonstrated the presence of at least three phenotypically distinct subgroups of

cardiovascular remodeling in fetuses. While the ‘globular’ and ‘elongated’ phenotypes

correspond to late-onset FGR, the hypertrophic cluster corresponds to early and severe

FGR associated with the worst perinatal and postnatal outcomes. We additionally

showed that standard ultrasonography permit the detection of FGR cases with cardiac

hypertrophy and worse outcomes. An early identification of these high-risk subgroup

might be open opportunity for a more suitable follow-up and timely postnatal lifestyle

strategies29,30, that could improve the short-and-long-term outcomes of these fetuses.

17

REFERENCES

1. Figueras F, Gratacos E. Update on the diagnosis and classification of fetal

growth restriction and proposal of a stage-based management protocol. Fetal Diagn

Ther. 2014;36(2):86-98.

2. Aarnoudse-Moens CS, Weisglas-Kuperus N, van Goudoever JB, Oosterlaan J.

Meta-analysis of neurobehavioral outcomes in very preterm and/or very low birth

weight children. Pediatrics. 2009 Aug;124(2):717-28.

3. Risnes KR, Vatten LJ, Baker JL, Jameson K, Sovio U, Kajantie E, Osler M,

Morley R, Jokela M,Painter RC, Sundh V,Jacobsen GW,Eriksson JG, Sørensen TI,

Bracken MB. Birthweight and mortality in adulthood: a systematic review and meta-

analysis. Int J Epidemiol. 2011 Jun;40(3):647-61.

4. Cruz-Lemini M, Crispi F, Valenzuela-Alcaraz B, Figueras F, Gomez O, Sitges

M, Bijnens B,Gratacos E. A fetal cardiovascular score to predict infant hypertension

and arterial remodeling in intrauterine growth restriction. Am J Obstet Gynecol. 2014

Jun;210(6):552 e1- e22.

5. Perez-Cruz M, Cruz-Lemini M, Fernandez MT, Parra JA, Bartrons J, Gomez-

Roig MD, Crispi F,Gratacos E. Fetal cardiac function in late-onset intrauterine growth

restriction vs small-for-gestational age, as defined by estimated fetal weight,

cerebroplacental ratio and uterine artery Doppler. Ultrasound Obstet Gynecol. 2015

Oct;46(4):465-71.

6. Crispi F, Bijnens B, Figueras F, Bartrons J, Eixarch E, Le Noble F, Ahmed A,

Gratacos E. Fetal growth restriction results in remodeled and less efficient hearts in

children. Circulation. 2010 Jun 8;121(22):2427-36.

7. Crispi F, Figueras F, Cruz-Lemini M, Bartrons J, Bijnens B, Gratacos E.

18

Cardiovascular programming in children born small for gestational age and relationship

with prenatal signs of severity. Am J Obstet Gynecol. 2012 Aug;207(2):121 e1-9.

8. Cruz-Lemini M, Crispi F, Valenzuela-Alcaraz B, Figueras F, Sitges M, Bijnens

B, Gratacos E. Fetal cardiovascular remodelling persists at 6 months of life in infants

with intrauterine growth restriction. Ultrasound Obstet Gynecol. 2015 Sep 29.

9. Crispi F, Hernandez-Andrade E, Pelsers MM, Plasencia W, Benavides-Serralde

JA, Eixarch E, Le Noble F, Ahmed A, Glatz JF, Nicolaides KH, Gratacos E. Cardiac

dysfunction and cell damage across clinical stages of severity in growth-restricted

fetuses. Am J Obstet Gynecol. 2008 Sep;199(3):254.e1-8.

10. Baschat AA, Gembruch U, Reiss I, Gortner L, Diedrich K. Demonstration of

fetal coronary blood flow by Doppler ultrasound in relation to arterial and venous flow

velocity waveforms and perinatal outcome--the 'heart-sparing effect'. Ultrasound Obstet

Gynecol. 1997 Mar;9(3):162-72.

11. Veille JC, Hanson R, Sivakoff M, Hoen H, Ben-Ami M. Fetal cardiac size in

normal, intrauterine growth retarded, and diabetic pregnancies. Am J Perinatol. 1993

Jul;10(4):275-9.

12. Miyague NI, Ghidini A. Effect of fetal growth restriction on

ultrasonographically measured cardiac size. Early Hum Dev. 1997 Apr 25;48(1-2):93-8.

13. Mäkikallio K, Shah J, Slorach C, Qin H, Kingdom J, Keating S, Kelly E,

Manlhiot C, Redington A, Jaeggi E. Fetal growth restriction and cardiovascular

outcome in early human infancy: a prospective longitudinal study. Heart Vessels. 2015.

14. Cinar B, Sert A, Gokmen Z, Aypar E, Aslan E, Odabas D. Left ventricular

dimensions, systolic functions, and mass in term neonates with symmetric and

asymmetric intrauterine growth restriction. Cardiol Young. 2015 Feb;25(2):301-7.

19

15. Leipala JA, Boldt T, Turpeinen U, Vuolteenaho O, Fellman V. Cardiac

hypertrophy and altered hemodynamic adaptation in growth-restricted preterm infants.

Pediatr Res. 2003 Jun;53(6):989-93.

16. Crispi F, Bijnens B, Sepulveda-Swatson E, Cruz-Lemini M, Rojas-Benavente J,

Gonzalez-Tendero A, Garcia-Posada R, Rodriguez-Lopez M, Demicheva E, Sitges M,

Gratacos E. Postsystolic shortening by myocardial deformation imaging as a sign of

cardiac adaptation to pressure overload in fetal growth restriction. Circ Cardiovasc

Imaging. 2014 Sep;7(5):781-7.

17. Figueras F, Meler E, Iraola A, Eixarch E, Coll O, Figueras J, Francis A,

Gratacos E, Gardosi J. Customized birthweight standards for a Spanish population. Eur

J Obstet Gynecol Reprod Biol. 2008 Jan;136(1):20-4.

18. Lang RM, Badano LP, Mor-Avi V, Afilalo J, Armstrong A, Ernande L,

Flachskampf FA,Foster E, Goldstein SA, Kuznetsova T, Lancellotti P, Muraru D,

Picard MH, Rietzschel ER, Rudski L, Spencer KT, Tsang W,

Voigt JU. Recommendations for cardiac chamber quantification by echocardiography in

adults: an update from the American Society of Echocardiography and the European

Association of Cardiovascular Imaging. Eur Heart J Cardiovasc Imaging. 2015

Mar;16(3):233-70.

19. Crispi F, Gratacós E. Fetal cardiac function: technical considerations and

potential research and clinical applications. Fetal Diagn Ther. 2012;32(1-2):47-64.

20. DeVore GR, Siassi B, Platt LD. Fetal echocardiography. IV. M-mode

assessment of ventricular size and contractility during the second and third trimesters of

pregnancy in the normal fetus. Am J Obstet Gynecol. 1984 Dec 15;150(8):981-8.

21. Hernandez-Andrade E, Lopez-Tenorio J, Figueroa-Diesel H, Sanin-Blair J,

20

Carreras E, Cabero L, Gratacos E. A modified myocardial performance (Tei) index

based on the use of valve clicks improves reproducibility of fetal left cardiac function

assessment. Ultrasound Obstet Gynecol. 2005 Sep;26(3):227-32.

22. Cruz-Martinez R, Figueras F, Bennasar M, Garcia-Posadas R, Crispi F,

Hernandez-Andrade E. Normal reference ranges from 11 to 41 weeks' gestation of fetal

left modified myocardial performance index by conventional Doppler with the use of

stringent criteria for delimitation of the time periods. Fetal Diagn Ther. 2012;32(1-

2):79-86.

23. Comas M, Crispi F, Gomez O, Puerto B, Figueras F, Gratacos E. Gestational

age- and estimated fetal weight-adjusted reference ranges for myocardial tissue Doppler

indices at 24-41 weeks' gestation. Ultrasound Obstet Gynecol. 2011 Jan;37(1):57-64.

24. Gardiner HM, Pasquini L, Wolfenden J, Barlow A, Li W, Kulinskaya E, Henein

M. Myocardial tissue Doppler and long axis function in the fetal heart. Int J Cardiol.

2006 Oct 26;113(1):39-47.

25. Tzeng CR, Chang YC, Chang YC, Wang CW, Chen CH, Hsu MI. Cluster

analysis of cardiovascular and metabolic risk factors in women of reproductive age.

Fertil Steril. 2014;101(5):1404-10.

26. Bijnens B, Cikes M, Butakoff C, Sitges M, Crispi F. Myocardial Motion and

Deformation: What Does It Tell Us and How Does It Relate to Function. Fetal Diagn

Ther 2012; 32:5–16

27. Torre I, Gonzalez-Tendero A, Garcia-Canadilla P, Crispi F, Garcia-Garcia F,

Bijnens B, Iruretagoyena I, Dopazo J, Amat-Roldan I, Gratacos E. Permanent cardiac

21

sarcomere changes in a rabbit model of intrauterine growth restriction. PLoS One.

2014;9(11):e113067.

28. Gonzalez-Tendero A, Torre I, Garcia-Canadilla P, Crispi F, Garcia-Garcia F,

Dopazo J, Bijnens B, Gratacos E. Intrauterine growth restriction is associated with

cardiac ultrastructural and gene expression changes related to the energetic metabolism

in a rabbit model. Am J Physiol Heart Circ Physiol. 2013 Dec;305(12):H1752-60.

29. Rodriguez-Lopez M, Osorio L, Acosta-Rojas R, Figueras J, Cruz-Lemini M,

Figueras F, Bijnens B, Gratacos E, Crispi F. Influence of breastfeeding and postnatal

nutrition on cardiovascular remodeling induced by fetal growth restriction. Pediatr Res.

2016 Jan 15;79(1-1):100-6.

30. Skilton MR, Mikkilä V, Würtz P, Ala-Korpela M, Sim KA, Soininen P, Kangas,

AJ, Viikari JS, Juonala M, Laitinen T, Lehtimäki T, Taittonen L, Kähönen

M,Celermajer DS, Raitakari OT. Fetal growth, omega-3 (n-3) fatty acids, and

progression of subclinical atherosclerosis: preventing fetal origins of disease? The

Cardiovascular Risk in Young Finns Study. Am J Clin Nutr. 2013 Jan;97(1):58-65.

Conflicts of Interest: The authors report no conflict of interest.

Source of Funding: This work was supported by the Erasmus + Programme of the

European Union [grant number 2013-0040] [This publication reflects the views only of

the author, and the Commission cannot be held responsible for any use which may be

made of the information contained therein]; Instituto de Salud Carlos III and Ministerio

de Economia y Competitividad [grant numbers PI12/00801, PI14/00226, SAF2012-

37196 and TIN2014-52923-R], cofinanciado por el Fondo Europeo de Desarrollo

Regional de la Unión Europea “Una manera de hacer Europa”; The research leading to

22

these results has received funding from “la Caixa” Foundation; Cerebra Foundation for

the Brain Injured Child; Agència de Gestió d'Ajuts Universitaris i de Recerca [grant

number SGR_928]. B.V.A. was supported by Programa de Ayudas Postdoctorales from

Agència de Gestió d'Ajuts Universitaris i de Recerca [grant number: 2013FI_B

00667] and wishes to express her gratitude to the Mexican National Council of Science

and Technology (CONACyT, Mexico City, Mexico) for partially supporting her

predoctoral stay at Hospital Clínic, Barcelona, Spain.