Cardiovascular Remodeling and Dysfunction Across a Range of Growth Restriction Severity in Small for Gestational Age Infants　– Implications for Fetal Programming –

Abstract

Background: The purpose of this study was to clarify cardiovascular structure and function in small for gestational age (SGA) infants across a range of intrauterine growth restriction (IUGR) severity.

Methods and Results: This prospective study included 38 SGA infants and 30 appropriate for gestational age (AGA) infants. SGA infants were subclassified into severe and mild SGA according to the degree of IUGR. Cardiovascular structure and function were evaluated using echocardiography at 1 week of age. Compared with the AGA infants, both the severe and mild SGA infants showed increased left ventricular diastolic dimensions (severe SGA 10.2±2.4, mild SGA 8.2±1.3, and AGA 7.3±0.7 mm/kg, P<0.05 for all) and decreased global longitudinal strain (severe −21.1±1.6, mild −22.5±1.8, and AGA −23.8±1.8%, P<0.05 for all). Severe SGA infants showed a decreased mitral annular early diastolic velocity (severe 5.6±1.4 vs. AGA 7.0±1.3 cm/s, P<0.01) and increased isovolumic relaxation time (severe 51.3±9.2 vs. AGA 42.7±8.2 ms, P<0.01). Weight-adjusted aortic intima-media thickness and arterial wall stiffness were significantly greater in both SGA infant groups. These cardiovascular parameters tended to deteriorate with increasing IUGR severity.

Conclusions: SGA infants, including those with mild SGA, showed cardiovascular remodeling and dysfunction, which increased with IUGR severity. (Circ J 2016; 80: 2212–2220)

Fetal intrauterine growth restriction (IUGR) is a condition defined as a fetus with an estimated weight of less than the 10th percentile for gestational age,¹ and affects almost 15% of pregnancies;² small for gestational age (SGA) infants are usually defined as a newborn with birth weight less than the 10th percentile for gestational age. IUGR and SGA are associated not only with increased perinatal death and severe morbidity, but also with an increased risk of long-term complications.³

Editorial p 2096

Several epidemiological studies have suggested a clear association between IUGR and an increased risk of adult cardiovascular diseases (CVD), such as ischemic heart disease (IHD),⁴ and hypertension.⁵ These findings support the “fetal programming theory”, in which exposure to particular environmental conditions, such as chronic hypoxia and undernutrition because of placental insufficiency, can result in morphological and functional changes during the critical fetal developmental period. Furthermore, these changes may predispose the fetus to adult-onset chronic CVD.⁶ Previous experimental studies have revealed that altered heart structure and function (such as dilated cardiomyopathy and cardiac dysfunction) exist in the fetus and that these changes persist into adulthood.⁷ In humans, recent studies have demonstrated subclinical cardiac dysfunction and increased intima-media thickness (IMT) in fetuses with growth restriction.^8,9 Furthermore, Crispi et al demonstrated that children with a history of IUGR (mean age, 5 years) presented with cardiovascular morphological and functional changes that were related to fetal growth restriction (FGR).¹⁰ Endothelial function was also reduced in children who were born SGA.¹¹ However, the importance of cardiovascular findings in neonates with IUGR remains largely unclear.

Clinically, SGA is stratified into stages of severity according to infant birth weight as follows: neonates with birth weights <3rd percentile (–2 SD) are classified as having severe SGA, and neonates with birth weights between the 3rd and the 10th percentiles (−1.3 SD) are considered cases of mild SGA.¹² Recently, Sehgal et al demonstrated that severe SGA infants show subclinical cardiac dysfunction and increased IMT and aortic stiffness compared with appropriate for gestational age (AGA) infants;^13,14 however, those studies did not assess mild SGA infants, who represent most of the IUGR population and almost 10% of the general population. Importantly, Kaijser et al demonstrated in an epidemiological study that, similar to adults with a history of severe IUGR, adults with a history of mild IUGR are at increased risk of IHD.¹⁵ Thus, the aim of our prospective study was to compare cardiovascular structure and function in infants with various degrees of SGA severity with healthy AGA infants of a comparable gestational age.

Methods

Study Population

We performed a prospective case-control study comparing SGA infants with gestational age-matched AGA infants. Patients were recruited from June 2013 to September 2015 at the Shinshu University Hospital, Japan. SGA was defined as a birth weight <10th percentile for gestational age according to local percentile charts. Exclusion criteria for both groups were as follows: infants with structural or chromosomal anomalies, evidence of intrauterine or neonatal infection or neonatal asphyxia, infants that were heavy for gestational age, infants from monochorionic diamniotic (MD) twin pregnancies, and infants receiving inotropic support or with a gestational age <30 weeks. Infants whose mothers had complications such as hypertension, hypercholesterolemia or diabetes or who smoked or took drugs were also excluded from the study.

Transthoracic Doppler echocardiography and vascular ultrasonography were performed at 1 week of age in the patients and controls to avoid the period of acute hemodynamic changes that results from the transition from fetal to extrauterine circulation, the presence of a patent ductus arteriosus, and acute respiratory pathology during the early neonatal period.

The study protocol was approved by the Ethics Committee of Shinshu University Hospital, and written informed consent was given by the parents of the infants enrolled in the study.

Clinical Data Collection

Baseline data including infant height, weight, sex, Apgar score, mode of delivery and pH of the umbilical artery (UA) were collected at the time of the study examination, and their percentiles were calculated according to local reference charts. Maternal information, including the resistance index (RI, measured as peak systolic velocity–end-diastolic velocity/peak systolic velocity) of the UA Doppler was also collected. Noninvasive blood pressure (BP) measurements were obtained using an appropriate-sized cuff on the right leg while the infant was in a quiet state and positioned supine at the level of the sphygmomanometer (IntelliVue MMS X2, Philips Medical Systems, Andover, MA, USA); an average value of 3 readings was recorded.

Standard Echocardiography

Conventional, 2D, M-mode, pulsed and color Doppler echocardiography was performed by a single operator (Y.A.) using an IE33 (Philips Medical Systems) with pulsed-wave Doppler frequency 7- and 10-MHz probes. Left ventricular diastolic dimension (LVDd), LV posterior wall and fractional shortening were measured using M-mode echocardiography from a parasternal short-axis view. LV mass (LVM) was estimated using the area-length formula and the LVM index (LVMI) was obtained by correcting the value of LVM for body surface area. Transmitral and tricuspid peak early diastolic velocity (E), peak late diastolic velocity (A), the ratio of E to A (E/A), and the deceleration time of the E wave were measured using pulsed Doppler echocardiography. Pulsed-wave tissue Doppler imaging (TDI) recordings, peak systolic (s’), peak early diastolic (e’), and peak late diastolic (a’) velocities were measured at the lateral, septal mitral annulus and lateral tricuspid annulus, respectively. We measured the isovolumic relaxation time (IVRT) using TDI. These echocardiographic parameters were measured according to the guidelines promulgated by the American Society of Echocardiography.¹⁶ All parameters were measured during 3 consecutive heart cycles, and mean values were calculated.

Speckle-Tracking Echocardiography

Two-dimensional grayscale images were analyzed by speckle-tracking echocardiography using commercially available software (Q Lab version 9.1, Philips Medical Systems) for strain estimation. For myocardial deformation analysis, endocardial borders were traced on an end-diastolic frame in apical views and on an end-systolic frame in short-axis views. The images were optimized to visualize the myocardial walls. Peak longitudinal strain (LS) and peak circumferential strain (CS) were computed automatically to generate regional data from 6 segments and an average value for each view (Figure 1). Global LS (GLS) was measured by averaging all measured segments in the apical 4-chamber, 2-chamber, and long-axis views in 6 segments from each view (the 16-segments model) and the peak average CS was obtained from 6 segments measured in the mid-ventricular short-axis view. Tracking was automatically performed, and strain curves were accepted only after visual assessment confirmed good tracking. Any view in which 2 or more segments could not be tracked was excluded from the analysis. A minimum frame rate of 119 was used.

Figure 1.

Echocardiography images from appropriate for gestational age (AGA) and small for gestational age (SGA) subjects. (Top) Segmental peak systolic longitudinal strain curves using a 4-chamber view. The white curve represents mean global longitudinal strain (GLS). (Middle) Tissue Doppler waves illustrating early diastolic (e’) and systolic (s’) myocardial peak velocities, and left isovolumic relaxation time (IVRT). (Bottom) Abdominal aorta imaging illustrating intima-media thickness (IMT).

Vascular Assessment

A trained physician (Y.A.) performed all recordings using an ultrasound machine (Philips CX 50; Phillips, Bothell, WA, USA) equipped with a 15-MHz linear-array probe. A long-axis view of the abdominal aorta was recorded from the subxiphoid process. The images focused on the dorsal arterial wall, and gain settings and high-resolution boxes were used to optimize image quality. The aortic IMT (aIMT) was measured in a straight, unbranched 1-cm longitudinal segment of the abdominal aorta using a 15-MHz linear transducer as previously described.¹³ aIMT was measured at end-diastole using ultrasound calipers. Aortic diameter was measured in the systolic (Aos) and end-diastolic (Aod) phases. Aortic elasticity was assessed using the following indices:¹⁰ aortic strain (%)=100×(Aos−Aod)/Aod and aortic stiffness index (β)=ln[SBP/DBP]/[(Aos−Aod)/Aod]. SBP and DBP are systolic and diastolic blood pressures, respectively and “ln” represents the natural logarithm.

Reproducibility

Intraobserver and interobserver reproducibility for the analyses of strain, aIMT, Aod and Aos were determined in 10 randomly selected patients (5 SGA, 5 AGA infants) using the Bland-Altman method in a blinded manner. To assess interobserver reproducibility, 2 observers independently analyzed the data without knowing the other’s measurements. To assess intraobserver reproducibility, measurements were repeated after more than 4 weeks by the same observer. The bias (mean difference) and limits of agreement (1.96 SD of the difference) between the 1st and 2nd measurements were determined.

Statistical Analysis

Variables are expressed as the mean±SD or as the median range when appropriate. To compare clinical characteristics among the 3 groups, one-way analysis of variance and the post-hoc Bonferroni test were used for parametric variables, the Kruskal-Wallis test with post-hoc comparison using the Dunn multiple comparison test was used for nonparametric variables, and the χ² test was used for categorical variables. To evaluate the trend in cardiovascular parameters across growth restriction severity, the Jonckheere-Terpstra test was performed with post-hoc comparison. Values of P<0.05 were considered statistically significant. Statistical calculations were performed using SPSS (version 22; SPSS, Inc, Chicago, IL, USA).

Results

Clinical Characteristics

A total of 38 SGA infants, including severe SGA infants (birth weight <3rd percentile, n=18) and mild SGA infants (birth weight between the 3rd and 10th percentiles, n=20) were enrolled in this study; 30 AGA infants born during the same study period were used as a control group. SGA infants were ineligible if they met any of following exclusion criteria: GA <30 weeks (10 infants), MD twins (6 infants), major congenital anomalies (7 infants), inotropic support (2 infants), maternal drug use during pregnancy (3 infants), congenital infection (1 infant) and >2 segments with inadequate tracking in the speckle-tracking analysis (1 infant). The demographic data collected for the study infants and the mothers are shown in Table 1. Maternal characteristics were similar among the groups. The RI and pH of the UA of the severe SGA infants were worse than those of the mild and AGA infants. Although the mild SGA infants groups included significantly more female subjects than the other groups, no significant sex differences were found for the clinical characteristics and cardiovascular findings.

Table 1. Clinical Characteristics of Each Group

Variable	Controls (n=30)	Mild SGA (n=20)	Severe SGA (n=18)	P value
Maternal characteristics
Height, cm	157.9±4.7	154.6±5.5	157.7±5.8	NS
Weight, kg	50.1±5.3	53.1±6.7	53.9±9.8	NS
BMI	20.2±2.2	22.8±3.2*	21.8±3.1	0.009
Preeclampsia	3 (10)	9 (45)	11 (61)*	<0.01
Gestational diabetes	2 (7)	0 (0)	0 (0)	NS
Umbilical artery RI	0.61±0.09	0.58±0.08	0.72±0.21*,††	0.0066
Infant characteristics
Gestational age at delivery, weeks	35.5±0.86	36.1±1.72	35.2±0.59	NS
Birth weight, kg	2.261±0.215	1.978±0.276**	1.547±0.409**,†	<0.001
Birth weight percentile	35.9±17.5	6.4±2.3**	0.93±0.88**,††	<0.001
Birth weight SD	−0.40±0.50	−1.55±0.2**	−2.61±0.55**,††	<0.001
Birth height, cm	44.9±2.0	43.2±2.0**	40.0±4.6**	<0.001
Birth height percentile	39.2±22.4	11.2±8.7**	6.38±8.8**	<0.001
Birth height SD	−0.32±0.76	−1.35±0.49**	−2.09±0.99**,†	<0.001
Body surface area, m2	0.16±0.012	0.15±0.013**	0.13±0.025**	<0.001
M/F	22/8	6/14*	9/9	<0.01
Mode of delivery vaginal/cesarean	13/17	7/13	2/16	NS
5-min Apgar score	9 (7–10)	9 (7–10)	9 (8–9)	NS
Umbilical artery pH	7.35±0.08	7.33±0.07	7.26±0.05**,††	<0.001
Systolic BP, mmHg	62±5.5	63.6±8.9	64.2±5.8	NS
Diastolic BP, mmHg	32.6±5.5	33.4±7.3	39.2±5.6**,††	0.002

Data are presented as mean±SD, median (range), or number (%). *P<0.05 vs. controls, **P<0.01 vs. controls. ^†P<0.05 vs. mild SGA, ^††P<0.01 vs. mild SGA. BMI, body mass index; BP, blood pressure; NS, not significant; RI, resistance index; SGA, small for gestational age.

Cardiac Structure

The ductus arteriosus was spontaneously closed in all neonates in the 3 groups.

The LVDd index (mm/kg) was significantly higher in both the mild SGA (8.2±1.3) and severe SGA (10.2±2.4) infants than in the AGA infants (7.3±0.7). As growth restriction severity increased, LVDd tended to increase (trend P<0.001, Figure 2A). LVM, LVMI and relative wall thickness were similar among the groups.

Figure 2.

(A–F) Box-and-whisker plots and the results of the Jonckheere-Terpstra test of variables according to IUGR severity. The horizontal line in the middle of each box indicates the median, the top and bottom borders of the box show the 75th and 25th percentiles, respectively, and the whiskers mark the 90th and 10th percentiles. AGA, appropriate for gestational age; aIMT, aortic itima-media thickness; GLS, global longitudinal strain; IUGR, intrauterine growth restriction; IVRT, isovolumic relaxation time; LV, left ventricular; LVDd, LV end-diastolic dimension; SGA, small for gestational age.

Systolic Function

Although LV fractional shortening was similar among the groups, GLS was significantly decreased in both the mild and severe SGA infants compared with the AGA infants (Table 2, Figure 2B). Figure 2B shows a significant reduction in GLS from the controls (−23.8±1.8%) to the mild SGA infants (−22.5±1.8%) and to the severe SGA infants (−21.1±1.6%; trend P<0.001). Mitral septal and lateral s’ wave velocities were likewise significantly lower in the severe SGA infants. Mitral s’ velocity also tended to decrease significantly with growth restriction severity (trend P<0.001), whereas global CS was similar among the groups.

Table 2. Echocardiographic Characteristics of Each Group

Variable	Controls (n=30)	Mild SGA (n=20)	Severe SGA (n=18)	P value
Cardiac morphology
LVDd/kg	7.3±0.7	8.2±1.3**	10.2±2.4**,††	<0.001
LV posterior wall, mm	2.6±0.4	2.6±0.4	2.6±0.6	NS
Relative wall thickness	0.32±0.05	0.33±0.05	0.34±0.09	NS
LVMI	31.1±7.9	35.9±6.9	34.0±4.8	NS
Systolic function
Heart rate, beats/min	135±13	139±12	144±14*	0.012
Fractional shortening, %	29.2±3.8	32.0±4.7	29.8±4.8	NS
Stroke volume	3.1±0.58	2.9±0.6	2.1±0.68**,††	<0.001
Left cardiac output, ml/min	418±95	406±80	295±95**,††	<0.001
Left cardiac output index, ml/min/kg	2,598±533	2,769±463	2,309±385	NS
Septal s’ peak velocity, cm/s	4.6±0.6	4.7±0.6	4.0±0.7**,††	0.03
Mitral lateral s’, cm/s	5.6±1.1	5.4±0.8	4.8±1.0*,†	0.017
Tricuspid lateral s’, cm/s	6.9±1.0	7.1±0.8	6.5±1.2	NS
GLS, %	−23.8±1.8	−22.5±1.8*	−21.1±1.6**,†	<0.001
GCS, %	−29.5±4.3	−30.5±4.0	−29.8±4.2	NS
Diastolic function
Mitral E wave, cm/s	54.5±8.8	60.8±13.8	46.2±9.0	NS
Mitral A wave, cm/s	54.1±9.5	54.9±12.5	51.2±7.5	NS
Mitral E/A ratio	1.0±0.13	1.13±0.23	0.91±0.16	NS
Tricuspid E wave, cm/s	44.6±12.4	43.9±9.1	39.4±8.9	NS
Tricuspid A wave, cm/s	53.6±11.5	54.5±13.0	46.8±8.3	NS
Tricuspid E/A ratio	0.84±0.17	0.82±0.18	0.86±0.20	NS
Left IVRT, ms	42.7±8.2	45.7±7.4	51.3±9.2**	0.005
Mitral E deceleration time, ms	102±23.5	110±28.0	109±37.9	NS
Mitral lateral e’, cm/s	7.0±1.3	6.9±1.2	5.6±1.4**,††	0.004
Mitral septal e’, cm/s	5.4±0.9	5.7±1.1	4.5±0.8**,††	0.024
Tricuspid e’, cm/s	7.1±1.2	7.6±1.4	6.1±1.5	NS
E/e’ (LV lateral)	8.1±2.2	9.0±2.6	8.9±3.1	NS
E/e’ (septal)	10.2±1.7	10.9±2.5	10.4±2.2	NS
E/e’ (RV)	6.3±1.6	5.8±1.5	6.5±2.1	NS
MPI
LV MPI	0.27±0.08	0.25±0.08	0.33±0.08	NS
RV MPI	0.11±0.06	0.13±0.09	0.13±0.09	NS
Vascular assessment
aIMT, μm	540±60	568±69	564±66	0.035
Weight-adjusted aIMT, μm/kg	242±30	295±50**	392±121**,†	<0.001
Aortic strain, %	12.6±2.9	10.5±3.2	9.4±2.4**	0.001
Aortic stiffness	2.40±0.80	2.82±0.81	2.55±0.78	NS
Weight-adjusted aortic stiffness	1.05±0.33	1.48±0.45**	1.76±0.77**	<0.001

Data are presented as the mean±SD. *P<0.05 vs. controls, **P<0.01 vs. controls. ^†P<0.05 vs. mild SGA, ^††P<0.01 vs. mild SGA. aIMT, aortic intima-media thickness; GCS, global circumferential strain; GLS, global longitudinal strain; IVRT, isovolumic relaxation time; LV, left ventricular; LVDd, LV diastolic dimension; LVMI, LV mass index; MPI, myocardial performance index; NS, not significant; RV, right ventricular.

Diastolic Function

Mitral lateral e’ velocity was significantly decreased in the severe SGA infant group (5.6±1.4 cm/s) compared with the other groups (mild SGA, 6.9±1.2 and AGA, 7.0±1.3 cm/s, P<0.01, Table 2, Figure 2C). Mitral septal e’ velocity was also significantly lower in the severe SGA infants compared with the other groups. IVRT was prolonged in the severe SGA infant groups compared with the AGA group (severe SGA, 51.3±9.2 and AGA, 42.7±8.2 ms, P<0.01, Table 2). The reduction in mitral e’ velocity and the prolongation of IVRT were both significantly associated with growth restriction severity (trend P<0.05, Figures 2C,D).

Vascular Assessment

Weight-adjusted aIMT was significantly higher in the severe SGA group (392±121) and mild SGA group (295±50) than in the AGA group (242±30). Aortic strain was lower in the severe SGA infants (9.4±2.4%) than in the AGA infants (12.6±2.9%). Weight-adjusted aortic stiffness was also higher in both SGA groups than in the AGA group. There was a significant deteriorating trend in these vascular parameters as the growth restriction increased in severity (Figures 2E,F). These cardiovascular parameters were also linearly correlated with birth weight percentile (Figure 3).

Figure 3.

Regression plots between cardiovascular parameters and birth weight percentile. Red circles, severe SGA; blue circles, mild SGA; white circles, AGA. AGA, appropriate for gestational age; aIMT, aortic itima-media thickness; GLS, global longitudinal strain; IVRT, isovolumic relaxation time; LV, left ventricular; LVDd, LV end-diastolic dimension; SGA, small for gestational age.

Reproducibility

Intraobserver and interobserver reproducibilities for the analyses of strain, aIMT, Aos and Aod were determined in 10 randomly selected patients (5 SGA, 5 AGA infants). Figure 4 shows the Bland-Altman plots for interobserver variability (bias±and limits of agreement). Table 3 shows the interobserver and intraobserver reproducibility measurements.

Figure 4.

Bland-Altman plots of interobserver differences for (A) global longitudinal strain (GLS), (B) global circumferential strain (GCS), (C) Aod and (D) Aos. The dotted lines show mean differences and the solid lines show the 95% limits of agreement. Aortic diameter (Ao) was measured in the systolic (Aos) and end-diastolic (Aod) phases

Table 3. Inter- and Intraobserver Reproducibilities in an Assessment of Echocardiographic Parameters

Variable	Interobserver variability	Intraobserver variability
GLS	−0.7±1.9	−0.3±3.0
GCS	−0.7±5.9	0.9±6.8
aIMT	−0.9±57	−0.9±34
Aod	−1.5±21	1.6±18
Aos	5.0±24	5.2±18

Aortic diameter (Ao) was measured in the systolic (Aos) and end-diastolic (Aod) phases. Abbreviations as in Table 2.

Discussion

The main finding of the present study was that cardiac dysfunction early in life secondary to SGA might have important consequences for postnatal cardiac disease.

In this study, the SGA infants exhibited (1) relative LV dilatation, (2) longitudinal systolic dysfunction (decreased s’ wave velocity and GLS) and diastolic dysfunction (decreased e’ and prolonged IVRT), and (3) higher aIMT and aortic stiffness. These alterations in cardiovascular structure and function progressively worsened concomitantly with IUGR. These results indicated that cardiovascular dysfunction and remodeling are early events following SGA, even the milder forms, and that the magnitude of this event increases in proportion to the severity of the fetal condition. To the best of our knowledge, this report is the first to show that the degree of cardiovascular remodeling and dysfunction in SGA infants increases linearly with IUGR severity.

Although IUGR has various etiologies, most IUGR fetuses suffer from chronic hypoxia and/or undernutrition in utero.¹⁷ Because fetuses have several mechanisms to compensate for undernutrition, intrauterine hypoxia is strongly associated with fetal growth in utero.¹⁸ Previous experimental studies have demonstrated that models of chronic embryonic hypoxia can lead to FGR, cardiac chamber dilatation, reduced cardiac systolic and diastolic dysfunction and increased IMT; these changes persist into adulthood,^7,19 as does increased cardiac vulnerability to ischemia.²⁰ The molecular mechanism of fetal heart programming is unknown, but recent experimental studies have shown a possible mechanism in hypoxia-induced IUGR animal models. Tintu et al demonstrated that hypoxic embryonic hearts exhibit decreased cardiomyocyte numbers, increased collagen tissue and shifts in the expression of titin isoform N2BA (the larger and more flexible isoform) towards isoform N2B (the smaller and stiffer isoform) and that these alterations persist into adulthood.⁷ An experimental study using an IUGR rabbit model has shown that chronic hypoxia in utero results in sarcomere shortening.²¹ Those authors speculated that the shortened sarcomere might change the titin isoform, because titin is a large protein that determines sarcomere length. Interestingly, cardiac samples from human IUGR fetus autopsies have also demonstrated shorter sarcomeres.²² Shorter sarcomeres have been shown to lead to reduced contraction force²³ and diastolic cardiac function²⁴ in experimental studies. Short sarcomeres reduce energy consumption in IUGR hearts, which might reflect an adaptive mechanism that contributes to chronic oxygen and/or nutrient restriction in the uterus during fetal heart development or subclinical cardiac dysfunction, after cardiac remodeling, such as the increased cardiac fibrosis and dilated ventricles in SGA.

In the present vascular assessment, the increased weight-adjusted aIMT and aortic stiffness in SGA infants were in agreement with results from previous studies of neonates and children with IUGR.^13,14,25 Increased IMT and aortic stiffness are both risk factors for hypertension and CVD in adults.^26,27 The reasons for the hypertrophy of the intima-media layer and the increased aortic stiffness are likely to be multifactorial in fetuses with IUGR. Changes in local hemodynamic flow and pressure because of placental dysfunction in the fetal circulation may result in increased wall stress after hypertrophic changes in the arterial wall.²⁸ Alternatively, chronic hypoxia may lead to impaired elastin synthesis in the arterial wall during the critical fetal period, thus leading to increased arterial stiffness.²⁹ In this study, increased aIMT, aortic stiffness and DBP in SGA infants affected myocardial deformation; however, the SGA infants in our study showed no signs of concentric hypertrophy (increased RWT and increased LVM), which are characteristics of cardiac remodeling in hypertensive patients. Thus, it is unlikely that alterations in these vascular parameters contributed to impaired myocardial function in the SGA infants.

In this study, echo parameters of cardiac function (which were assessed using conventional echocardiography) were similar between the SGA and AGA groups. Both TDI and speckle-tracking echocardiography detected the presence of systolic and diastolic cardiac dysfunction in SGA infants, which revealed that these tools detect subclinical cardiac dysfunction more sensitively than conventional echocardiography.^30,31 In the speckle-tracking analysis performed in our study, GLS was decreased, but CS was maintained in both SGA groups. We speculate that chronic hypoxia in utero resulted in longitudinal dysfunction because decreased longitudinal movement is mediated by subendocardial and subepicardial fibers, which are highly sensitive to hypoxia.²¹ The observed significant increase in heart rate in the severe SGA group that was associated with decreased LS may represent a compensatory mechanism that preserves LV output. Crispi et al have presented similar findings in children (average age, 5 years) with a history of IUGR.¹⁰

Previous studies report contrasting results for diastolic function in growth-restricted neonates.^32,33 We speculate that these conflicting results may reflect the different stages of diastolic function. Although e’ was decreased in the present SGA group, other parameters related to diastolic function, such as E/A and E/e’, were similar between the severe SGA group and the control group in this study. This result may reflect mild diastolic dysfunction in the severe SGA group, which is associated with reduced LV relaxation but not yet elevated left atrial pressure.³⁴ The findings of subclinical cardiac dysfunction and increased aIMT in infants with IUGR in this study are in agreement with similar cardiovascular changes that have been observed in fetuses exhibiting growth restriction and in children born as growth-restricted infants.^10,13

In this study, we showed that SGA infants, including mild SGA infants, who represent the majority of SGA cases, exhibit primary remodeling and decreased function in the cardiovascular system immediately after birth. In a previous epidemiological study, Kaijser et al demonstrated an association between stratified FGR and adult IHD.¹⁵ Those authors revealed that mild fetal growth (birth weight ≤−2 to −1 SD below the mean) is significantly associated with adult IHD (hazard ratio, 1.54), similar to severe IUGR (birth weight ≤−2 SD; hazard ratio, 1.64). These findings are consistent with the results of our study, which showed that subjects with mild growth retardation also exhibit cardiovascular changes that may result in IHD in adults.

These results suggest that because of the high prevalence of IUGR it may be clinically relevant to recognize SGA in infants (including mild SGA) as a risk factor for CVD.

Single echo parameter that have sufficient predictive value for adverse outcomes have not been reported. Our study did not estimate whether echo parameters are predictive of short- or long-term adverse outcomes in SGA infants.

Crispi et al reported that a composite cardiovascular score integrating some echo parameters improved the predictive value for death.³⁵ Furthermore, they also reported that the cardiac dysfunction associated with IUGR fetuses deteriorated further as severity increased, together with increased biomarkers of heart failure and myocardial cell damage in cord blood.⁸ The combination of echo parameters with biomarkers of cardiac dysfunction and myocardial cell damage may aid risk stratification in SGA infants. Skilton et al provide evidence that omega-3 fatty acid administration in infants born with SGA improves arterial wall thickening and reduces SBP in adolescence.³⁶ Considering the advances that will be made in developing preventive treatments for adult CVD in these populations in the future, the findings from our study may be useful for predicting increased cardiovascular risk as early as possible and to encourage this high-risk population to improve their lifestyle and introduce treatments to prevent future adult-onset CVD.

Study Limitations

First, the sample size of the studied subgroups was small. Second, we performed the examination at 1 week of age to avoid the period during which acute hemodynamic changes occur in the transition from fetal to neonatal circulation; nevertheless, the effect of transitional circulation and changes in hormonal and sympathetic nervous systems during the early neonatal period may have affected cardiovascular function in our study population. Third, we examined only a single time point, and no follow-up was used to assess cardiovascular outcome. A long-term follow-up study in this population is warranted to clarify the pathophysiology of the relationship between fetal cardiovascular remodeling or dysfunction and CVD in adulthood. Fourth, although preeclampsia has been shown to affect the fetal myocardium,³⁷ we did not exclude mothers with preeclampsia because it is also a major cause of IUGR. This factor might have affected the results of this study. Finally, severity stratification based on birth weight might have changed during the fetal period.

Conclusions

In the present study, we found subclinical longitudinal cardiac dysfunction, LV remodeling, and increased IMT and stiffness in FGR neonates across a range of growth restriction severities. Importantly, even infants with mildly restricted SGA showed cardiovascular alterations during the neonatal period. Our findings support the hypothesis that the effect of physiology on fetal programming of cardiac structural and functional adaptations in response to IUGR may have consequences later in life. These findings may contribute to screening and preventive therapy for subjects at risk of developing CVD later in life. The long-term cardiovascular outcomes of the changes in SGA infants remain to be established in a prospective long-term follow-up study.

Financial Support Statement / Disclosure

The authors have no conflicts of interest or financial relationships relevant to this article to disclose.

Supplementary Files

Supplementary File 1

Methods

Please find supplementary file(s);

http://dx.doi.org/10.1253/circj.CJ-16-0352

References

• 1.   American College of Obstetricians and Gynecologists. ACOG Practice bulletin no. 134: Fetal growth restriction. Obstet Gynecol 2013; 121: 1122–1133.
• 2.    Ananth  CV,  Vintzileos  AM. Distinguishing pathological from constitutional small for gestational age births in population-based studies. Early Hum Dev 2009; 85: 653–658.
• 3.    Alberry  M,  Soothill  P. Management of fetal growth restriction. Arch Dis Child Fetal Neonatal Educ 2007; 92: 62–67.
• 4.    Barker  DJ,  Winter  PvD,  Osmond  C,  Margetts  B,  Simmonds  SJ. Weight in infancy and death from ischaemic heart disease. Lancet 1989; 2: 577–580.
• 5.    Ojeda  NB,  Grigore  D,  Alexander  BT. Intrauterine growth restriction: Fetal programming of hypertension and kidney disease. Adv Chronic Kidney Dis 2008; 15: 101–106.
• 6.    Palinski  W,  Napoli  C. Impaired fetal growth, cardiovascular disease, and the need to move on. Circulation 2008; 117: 341–343.
• 7.    Tintu  A,  Rouwet  E,  Verlohren  S,  Brinkmann  J,  Ahmad  S,  Crispi  F, et al. Hypoxia induces dilated cardiomyopathy in the chick embryo: Mechanism, intervention, and long-term consequences. PLoS One 2009; 4: e5155, doi:10.1371/journal.pone.0005155.
• 8.    Crispi  F,  Hernandez-Andrade  E,  Pelsers  MM,  Plasencia  W,  Benavides-Serralde  JA,  Eixarch  E, et al. Cardiac dysfunction and cell damage across clinical stages of severity in growth-restricted fetuses. Am J Obstet Gynecol 2008; 199: 254.e1–e8, doi:10.1016/j.ajog.2008.06.056.
• 9.    Zanardo  V,  Visentin  S,  Trevisanuto  D,  Bertin  M,  Cavallin  F,  Cosmi  E. Fetal aortic wall thickness: A marker of hypertension in IUGR children? Hypertens Res 2013; 36: 440–443.
• 10.    Crispi  F,  Bijnens  B,  Figueras  F,  Bartrons  J,  Eixarch  E,  Le Noble  F, et al. Fetal growth restriction results in remodeled and less efficient hearts in children. Circulation 2010; 121: 2427–2436.
• 11.    Faienza  MF,  Brunetti  G,  Delvecchio  M,  Zito  A,  De Palma  F,  Cortese  F, et al. Vascular function and myocardial performance indices in children born small for gestational age. Circ J 2016; 80: 958–963.
• 12.    Vayssière  C,  Sentilhes  L,  Ego  A,  Bernard  C,  Cambourieu  D,  Flamant  C, et al. Fetal growth restriction and intra-uterine growth restriction: Guidelines for clinical practice from the French College of Gynaecologists and Obstetricians. Eur J Obstet Gynecol Reprod Biol 2015; 193: 10–18.
• 13.    Sehgal  A,  Doctor  T,  Menahem  S. Cardiac function and arterial biophysical properties in small for gestational age infants: Postnatal manifestations of fetal programming. J Pediatr 2013; 163: 1296–1300.
• 14.    Sehgal  A,  Doctor  T,  Menahem  S. Cardiac function and arterial indices in infants born small for gestational age: Analysis by speckle tracking. Acta Paediatr 2014; 103: 49–54.
• 15.    Kaijser  M,  Bonamy  AK,  Akre  O,  Cnattingius  S,  Granath  F,  Norman  M, et al. Perinatal risk factors for ischemic heart disease: Disentangling the roles of birth weight and preterm birth. Circulation 2008; 117: 405–410.
• 16.    Lopez  L,  Colan  SD,  Frommelt  PC,  Ensing  GJ,  Kendall  K,  Younoszai  AK, et al. Recommendations for quantification methods during the performance of a pediatric echocardiogram: A report from the Pediatric Measurements Writing Group of the American Society of Echocardiography Pediatric and Congenital Heart Disease Council. J Am Soc Echocardiogr 2010; 23: 465–495.
• 17.    Marsál  K. Intrauterine growth restriction. Curr Opin Obstet Gynecol 2002; 4: 127–135.
• 18.    Botting  KJ,  Wang  KC,  Padhee  M,  McMillen  IC,  Summers-Pearce  B,  Rattanatray  L, et al. Early origins of heart disease: Low birth weight and determinants of cardiomyocyte endowment. Clin Exp Pharmacol Physiol 2012; 39: 814–823.
• 19.    Rouwet  EV,  Tintu  AN,  Schellings  MW,  van Bilsen  M,  Lutgens  E,  Hofstra  L, et al. Hypoxia induces aortic hypertrophic growth, left ventricular dysfunction, and sympathetic hyperinnervation of peripheral arteries in the chick embryo. Circulation 2002; 105: 2791–2796.
• 20.    Rueda-Clausen  CF,  Morton  JS,  Lopaschuk  GD,  Davidge  ST. Long-term effects of intrauterine growth restriction on cardiac metabolism and susceptibility to ischaemia/reperfusion. Cardiovasc Res 2011; 90: 285–294.
• 21.    Torre  I,  González-Tendero  A,  García-Cañadilla  P,  Crispi  F,  García-García  F,  Bijnens  B, et al. Permanent cardiac sarcomere changes in a rabbit model of intrauterine growth restriction. PLoS One 2014; 9: e113067, doi:10.1371/journal.pone.0113067.
• 22.    Iruretagoyena  JI,  Gonzalez-Tendero  A,  Garcia-Canadilla  P,  Amat-Roldan  I,  Torre  I,  Nadal  A, et al. Cardiac dysfunction is associated with altered sarcomere ultrastructure in intrauterine growth restriction. Am J Obstet Gynecol 2014; 210: 550.e1–e7, doi:10.1016/j.ajog.2014.01.023.
• 23.    Kobirumaki-Shimozawa  F,  Inoue  T,  Shintani  SA,  Oyama  K,  Terui  T,  Minamisawa  S, et al. Cardiac thin filament regulation and the Frank-Starling mechanism. J Physiol Sci 2014; 64: 221–232.
• 24.    Radke  MH,  Peng  J,  Wu  Y,  McNabb  M,  Nelson  OL,  Granzier  H, et al. Targeted deletion of titin N2B region leads to diastolic dysfunction and cardiac atrophy. Proc Natl Acad Sci USA 2007; 104: 3444–3449.
• 25.    Bradley  TJ,  Potts  JE,  Lee  SK,  Potts  MT,  De Souza  AM,  Sandor  GG. Early changes in the biophysical properties of the aorta in pre-adolescent children born small for gestational age. J Pediatr 2010; 156: 388–392.
• 26.    Furberg  CD,  Adams  HP Jr,  Applegate  WB,  Byington  RP,  Espeland  MA,  Hartwell  T, et al. Effect of lovastatin on early carotid atherosclerosis and cardiovascular events: Asymptomatic Carotid Artery Progression Study (ACAPS) Research Group. Circulation 1994; 90: 1679–1687.
• 27.    Mitchell  GF,  Wang  N,  Palmisano  JN,  Larson  MG,  Hamburg  NM,  Vita  JA, et al. Hemodynamic correlates of blood pressure across the adult age spectrum: Noninvasive evaluation in the Framingham Heart Study. Circulation 2010; 122: 1379–1386.
• 28.    Meyer  WW,  Lind  J. Iliac arteries in children with a single umbilical artery: Structure, calcification, and early artherosclerotic leisions. Arch Dis Child 1974; 49: 671–679.
• 29.    Thompson  JA,  Richardson  BS,  Gagnon  R,  Regnault  TR. Chronic intrauterine hypoxia interferes with aortic development in the late gestation ovine fetus. J Physiol 2011; 589: 3319–3332.
• 30.    Comas  M,  Crispi  F,  Cruz-Martinez  R,  Martinez  JM,  Figueras  F,  Gratacós  E. Usefulness of myocardial tissue Doppler vs conventional echocardiography in the evaluation of cardiac dysfunction in early-onset intrauterine growth restriction. Am J Obstet Gynecol 2010; 203: 1–7.
• 31.    Wang  J,  Khoury  DS,  Yue  Y,  Torre-Amione  G,  Nagueh  SF. Preserved left ventricular twist and circumferential deformation, but depressed longitudinal and radial deformation in patients with diastolic heart failure. Eur Heart J 2008; 29: 1283–1289.
• 32.    Fouzas  S,  Karatza  AA,  Davlouros  PA,  Chrysis  D,  Alexopoulos  D,  Mantagos  S, et al. Neonatal cardiac dysfunction in intrauterine growth restriction. Pediatr Res 2014; 75: 651–657.
• 33.    Gürses  D,  Seyhan  B. Evaluation of cardiac systolic and diastolic functions in small for gestational age babies during the first months of life: A prospective follow-up study. Cardiol Young 2013; 23: 597–605.
• 34.    Flachskampf  FA,  Biering-Sørensen  T,  Solomon  SD,  Duvernoy  O,  Bjerner  T,  Smiseth  OA. Cardiac imaging to evaluate left ventricular diastolic function. JACC Cardiovasc Imaging 2015; 8: 1071–1093.
• 35.    Crispi  F,  Hernandez-Andrade  E,  Benavides-Serralde  JA,  Padilla  N,  Acosta  R,  Gratacos  E. The use of a cardiac profile improves detection of heart dysfunction and prediction of poor perinatal outcome as compared with ductus venosus alone. Ultrasound Obstet Gynecol 2007; 30: 525.
• 36.    Skilton  MR,  Pahkala  K,  Viikari  JS,  Rönnemaa  T,  Simell  O,  Jula  A, et al. The association of dietary alpha-linolenic acid with blood pressure and subclinical atherosclerosis in people born small for gestational age: The Special Turku Coronary Risk Factor Intervention Project study. J Pediatr 2015; 166: 1252–1257.
• 37.    Balli  S,  Kibar  AE,  Ece  I,  Oflaz  MB,  Yilmaz  O. Assessment of fetal cardiac function in mild preeclampsia. Pediatr Cardiol 2013; 34: 1674–1679.