Objective
The objective of the study was to evaluate cardiovascular function in children who were small-for-gestational-age (SGA) fetuses.
Study Design
This was a prospective study including 100 controls and 50 children diagnosed in utero as SGA after 34 weeks subdivided into the following categories: SGA and intrauterine growth restriction (IUGR) according to the absence or presence, respectively, of weight centile less than 3 or abnormal cerebroplacental Doppler. Postnatal cardiovascular outcome was evaluated at 3-6 years of age by echocardiography, blood pressure, and carotid ultrasound.
Results
Both SGA and IUGR presented in childhood more globular hearts, reduced longitudinal motion, and impaired relaxation with an increase in radial function. Both groups showed increased blood pressure and carotid intima-media thickness. There was a linear tendency to worse cardiovascular results in IUGR as compared with SGA.
Conclusion
Fetal cardiovascular programming occurs in SGA, regardless of Doppler and weight centile. These findings challenge the concept of constitutionally small and warrant further investigation to identify predictors of cardiovascular outcome in SGA.
The heart is a central organ in the fetal adaptive mechanisms to placental insufficiency and cardiac dysfunction in utero is recognized among the essential pathophysiologic features of intrauterine growth restriction (IUGR). Cardiovascular adaptation in utero leads to fetal programming and persists in the form of subclinical cardiac and vascular dysfunction and remodeling in newborns and children. Primary cardiovascular remodeling is thought to be an important link in the long-known epidemiological relationship between IUGR and increased cardiovascular mortality in adulthood.
Fetal programming is not exclusive of severe forms of IUGR, but it also occurs in late-onset IUGR, often referred to as small-for-gestational-age (SGA). Considering that up to 6-10% of fetuses are born SGA, the importance of the problem and opportunities for public health cannot be overemphasized. However, it is unclear whether fetal cardiovascular programming occurs with the same magnitude in all forms of late-onset IUGR.
Although as a group they are associated with poorer perinatal and long-term postnatal outcomes, an important proportion of SGA fetuses have apparently normal neonatal outcomes. This has led to postulate that this diagnostic label is constituted by a mix of constitutionally small fetuses and true forms of IUGR because of placental late-onset insufficiency.
Over recent years several prognostic factors, including abnormal cerebroplacental and uterine artery Doppler, and birthweight below the third centile, have been identified as predictors of poor neonatal outcome, and consequently as surrogate markers of true late-onset IUGR among near term SGA fetuses. These factors are associated with poorer neonatal outcome, but it is unknown whether they also classify well the risk of long term abnormal outcome. Thus, their value to predict the presence of cardiovascular programming is unknown.
We evaluated the hypothesis that cardiovascular programming would occur exclusively or with increased magnitude in SGA fetuses with signs suggesting placental insufficiency. We conducted a prospective cohort study including 100 normal singleton pregnancies and 50 SGA cases, classified for the purposes of this study as SGA (n = 25) or IUGR (n = 25) according to estimated fetal weight centile and cerebroplacental Doppler. We evaluated cardiovascular outcome of children at 3-6 years of age by echocardiography, blood pressure, and carotid intima-media thickness.
Materials and Methods
Study population
This was a prospective cohort study including 200 3-6 years old children delivered after 34 weeks, subdivided into the following categories: (1) IUGR defined by birthweight below the third percentile or abnormal cerebroplacental ratio (below the fifth percentile); (2) SGA defined by birthweight between the third and 10th percentile together with normal cerebroplacental ratio; and (3) controls defined by birth weight above the 10th percentile with no pregnancy complications. Controls were matched 2 to 1 with cases by gestational age at delivery (±1 week). Eighty percent of the children included were already included in a previous study from our group.
The study protocol was approved by the local ethics committee and patients provided their written informed consent. Cases and controls were identified in fetal life among pregnancies attended in the Department of Maternal-Fetal Medicine in Hospital Clinic (Barcelona, Spain) and follow-up into childhood.
Exclusion criteria were structural or chromosomal anomalies, intrauterine infection, and monochorionic twin pregnancies. In all pregnancies gestational age was calculated based on the crown-rump length at first-trimester ultrasound. Fetal and neonatal weight centile were calculated according to local reference curves. Perinatal, demographic, and postnatal data were obtained by review of medical records and clinical databases and by parental interview at the time of study evaluation.
Low socioeconomic class was defined as routine occupation, long-term unemployment, or never worked for both the pregnant woman and her partner, according to the UK National Statistics Socio-Economic Classification. Familiar early cardiovascular history was defined as the presence of early cardiovascular disease (including congenital heart disease, coronary disease, hypertension, diabetes, or hypercholesterolemia) in expanded first-degree pedigree (males younger than 55 years; females younger than 65 years). Major neonatal morbidity was defined by the presence of bronchopulmonary dysplasia, necrotizing enterocolitis, intraventricular hemorrhage, periventricular leukomalacia, retinopathy, persistent ductus arteriosus, or sepsis.
Perinatal data
Cases underwent prenatal ultrasonographic examination using a Siemens Sonoline Antares (Siemens Medical Systems, Malvern, PA). Basic Doppler examination included umbilical artery and middle cerebral artery. Cerebroplacental ratio was calculated as a simple ratio of the middle cerebral artery pulsatility index divided by the umbilical artery pulsatility index. At delivery, gestational age, mode of delivery, birthweight, birthweight centile, Apgar score, and umbilical pH were recorded.
Postnatal follow-up
Cases and controls were followed up until 3-6 years of age. Data on the child′s medication and postnatal complications were obtained by parental interview. Anthropometric data, including height, weight, and body mass index were obtained at the time of examination, and percentiles were calculated according to local reference values.
Postnatal cardiac structure and function was assessed by echocardiography with a Siemens Sonoline Antares ultrasound system (Siemens Medical System) with a 2-10 MHz phased-array transducer following a standardized protocol. Children were studied when resting quietly or asleep. Cardiac scans were performed by 2 experienced physicians (F.C., J.B.) who were blinded to the allocation group. Structural heart integrity was confirmed by two-dimensional (2D) and Doppler echocardiography.
Cardiac morphometry
Cardiac morphometry included left and right sphericity index (defined as base-to-apex length divided by basal diameter) measured on an end-diastolic 2D apical 4-chamber view. Left ventricular end-diastolic diameter and septal and posterior wall thickness were measured by M-mode on a parasternal long-axis parasternal view and relative wall thickness calculated as posterior wall plus interventricular septum thickness divided by left ventricular diameter.
Systolic function
Systolic function evaluation included cardiac output, left ejection fraction, left ventricular thickening, mitral/tricuspid annular displacement (MAPSE/TAPSE), and systolic annular peak velocity (S′). Left stroke volume was calculated as cross-sectional aortic annulus area × aortic flow velocity time integral. Left cardiac output was calculated as left stroke volume × heart rate. Left stroke volume and cardiac output were normalized by child’s body surface area. Left ejection fraction was measured on 2D from a 4- and 2-chamber apical view following Simpson’s rule. Left ventricular thickening was calculated as posterior left free wall thickness in systole: in diastole divided by posterior left free wall thickness in diastole measured in a parasternal long-axis view by M-mode. Systolic mitral and tricuspid annular motion was assessed by M-mode (MAPSE/TAPSE) and spectral tissue Doppler (S′) on an apical 4-chamber view.
Diastolic function
Diastolic function was evaluated by peak early and late transvalvular filling (E/A) ratio, deceleration time of E velocity, isovolumetric relaxation time (IRT), early (E′) and late (A′) diastolic annular peak velocities, and E/E′ ratio. E deceleration time and E/A ratio were measured from mitral and tricuspid inflow velocities by conventional Doppler. IRT was measured from the end of the aortic wave to the beginning of the mitral early-filling wave. Mitral and tricuspid diastolic annular peak velocities were obtained by spectral tissue Doppler in an apical 4-chamber view. The E/E′ ratio was measured as previously described.
Vascular assessment included blood pressure and intima-media thickness (IMT) measurement. Systolic and diastolic blood pressures were measured with the child seated after 5-10 minutes resting and centiles were calculated according to standard normograms. Maximum and mean IMT was measured by ultrasound at both carotid arteries using a 13 MHz linear transducer. Measurements were performed offline based on a trace method with the assistance of a computerized program (Syngo Arterial Health Package; Siemens Medical Systems) as previously described and normalized by the child′s weight.
Statistical analysis
The primary outcome S′ was used to calculate sample size because of the high reported sensitivity to detect preclinical cardiac dysfunction in children. The sample size was calculated to enable us to observe a difference of 20% in S′ between the group of case subjects with severe IUGR and control subjects, with 85% power and a 5% type I risk. Basal mean and within-group SDs were estimated according to published normative data in children, which resulted in a required sample of 25 individuals in each group for S′. Therefore, a sample size of at least 25 individuals in each group was designed. Paired comparisons among study groups were adjusted for age, sex, and gestational age at delivery by linear (general linear model) or logistic regression analysis after confirming parametric assumptions by Kolmogorov-Smirnov test (data not shown). Cardiovascular outcomes were also adjusted for heart rate, maternal height, gestational diabetes, preeclampsia, and days in neonatal intensive care unit. Additionally, a linear polynomial orthogonal contrast was constructed for each model to test the hypothesis of linear/quadratic association across SGA/IUGR severity groups. Differences were considered statistically significant by P < .05. The software package SPSS 17.0 (SPSS, Chicago, IL) was used for the statistical analysis.
Results
Study populations and perinatal outcome
Study population characteristics and perinatal outcome are shown in Table 1 . Baseline characteristics were similar among the study groups with the exception of shorter parental height in children with SGA or IUGR. As expected, SGA cases showed a lower birthweight and higher prevalence of pregnancy complications and days in neonatal intensive care unit. Likewise, IUGR cases showed abnormal fetoplacental Doppler (with the exception of similar ductus venosus pulsatility index), lower Apgar scores and umbilical pH, and higher prevalence of pregnancy complications, cesarean section, and days in intensive care unit. Fetoplacental Doppler values, gestational age at delivery, and perinatal outcome were similar in SGA and controls.
| Characteristics | Controls | SGA | IUGR | Linear tendency P value |
|---|---|---|---|---|
| n | 100 | 25 | 25 | |
| Baseline characteristics | ||||
| Paternal height, cm | 176 (172–181) | 174 (170–177) | 172 (168–180) a | .006 |
| Maternal height, cm | 164 (160–168) | 162 (156–164) a | 159 (156–163) a | .056 |
| Maternal body mass index, kg/m 2 | 22 (21–25) | 22 (20–24) | 22 (20–25) | .559 |
| Maternal age, y | 33 (31–36) | 33 (32–37) | 34 (31–35) | .692 |
| Maternal nuliparity, % | 66 | 67 | 76 | .497 |
| Maternal smoking, % | 29 | 38 | 32 | .900 |
| Low socioeconomic level, % | 1 | 0 | 0 | .729 |
| Familiar early cardiovascular history, % | 16 | 9 | 9 | .332 |
| Prenatal corticoids, % | 7 | 0 | 12 | .400 |
| Fetoplacental Doppler | ||||
| Abnormal cerebro-placental ratio, % | 0 | 0 | 32 a | .001 |
| Cerebroplacental ratio | 0.2 (0–0.8) | –0.3 (–0.6 to –0.1) | –1 (–1.8 to –0.3) a | .001 |
| Umbilical artery PI | 0 (–0.3 to 0.5) | –0.6 (–0.8 to 0.2) | 0.8 (0–1.6) | .006 |
| Middle cerebral artery PI | 0.3 (0–0.9) | –1 (–1.2 to –0.9) | –0.8 (–1.8 to 0.7) a | .085 |
| Ductus venosus PI | 0 (–0.3 to 0.5) | –0.1 (–0.2 to 0.1) | –0.3 (–0.7 to 0.2) | .323 |
| Pregnancy outcome | ||||
| In vitro fecundation, % | 1 | 0 | 8 a | .018 |
| Gestational diabetes, % | 4 | 14 | 20 a | .004 |
| Preeclampsia, % | 0 | 5 | 4 | .137 |
| Delivery data | ||||
| Gestational age at delivery, wks | 39 (38–40) | 39 (39–40) | 39 (38–40) | .642 |
| Cesarean section, % | 24 | 19 | 40 | .135 |
| Birthweight, g | 3365 (3045–3600) | 2700 (2620–2800) a | 2530 (2195–2630) a | < .001 |
| Birthweight percentile | 56 (34–81) | 5 (4–7) a | 1 (0–2) a | < .001 |
| 1 min Apgar score ≤7 | 1% | 8% | 20% a | < .001 |
| Umbilical artery pH | 7.27 (7.23–7.31) | 7.24 (7.17–7.33) | 7.23 (7.16–7.26) a | .056 |
| Umbilical vein pH | 7.34 (7.31–7.37) | 7.32 (7.24–7.37) | 7.28 (7.25–7.32) a | .043 |
| Neonatal data | ||||
| Days in intensive care unit | 0 (0) | 0 (0–3) a | 2 (0–10) a | < .001 |
| Major neonatal morbidity, % | 0 | 0 | 0 | 1.00 |
a P < .05 compared with controls, calculated by linear or logistic regression adjusted for sex, age, and gestational age at delivery.
Postnatal constitutional characteristics
Follow-up characteristics at the time the children were assessed are shown in Table 2 . Children who were classified as SGA showed a lower proportion of males with similar age, ethnicity, and anthropometric results as compared with controls. Conversely, IUGR children showed a lower weight and body mass index percentile with similar height as controls.
| Characteristic | Controls | SGA | IUGR | Linear tendency P value |
|---|---|---|---|---|
| n | 100 | 25 | 25 | |
| Male sex, % | 48 | 19 a | 56 | .550 |
| Caucasian, % | 96 | 86 | 92 | .790 |
| Age, y | 5 (4–6) | 5 (4–5) | 4 (3–4) a | .001 |
| Height percentile | 53 (49–56) | 52 (49–53) | 50 (46–52) | .306 |
| Weight percentile | 57 (33–75) | 49 (24–69) | 29 (12–43) a | .005 |
| Body mass index percentile | 47 (25–61) | 26 (23–67) | 25 (12–44) a | .007 |
a P < .05 compared with controls, calculated by linear or logistic regression adjusted for gender, age, and gestational age at delivery.
Cardiac outcome
Both SGA and IUGR show significant changes in cardiac shape with more globular hearts and normal wall thickness ( Figure 1 ). Stroke volume, cardiac output, and left ventricular thickening were significantly increased in cases as compared with controls ( Table 3 ). Cases also showed a tendency to higher heart rate and left ejection fraction values. Longitudinal motion in systole was significantly reduced in IUGR with a nonsignificant tendency of lower results in SGA. Both IUGR and SGA showed signs of diastolic dysfunction, mainly measured as an increase in E deceleration, decrease in diastolic annular velocities, and a tendency to higher isovolumetric relaxation times. There was a significant linear tendency to worse cardiac function results in IUGR as compared with SGA.
| Characteristic | Controls | SGA | IUGR | Linear tendency P value |
|---|---|---|---|---|
| n | 100 | 25 | 25 | |
| Cardiac morphometry | ||||
| Left sphericity index | 1.82 (1.67–1.97) | 1.49 (1.34–1.55) a | 1.41 (1.27–1.52) a | < .001 |
| Right sphericity index | 1.58 (1.43–1.76) | 1.41 (1.24–1.54) | 1.41 (1.31–1.53) a | .014 |
| Relative wall thickness | 0.40 (0.35–0.44) | 0.38 (0.34–0.44) | 0.40 (0.37–0.47) | .713 |
| Systolic function | ||||
| Left ventricle | ||||
| Heart rate (bpm) | 94 (87–108) | 105 (97–109) | 99 (87–114) | .254 |
| Normalized left stroke volume (mL/m 2 ) | 49.7 (42.6–57.2) | 57.2 (48.1–69.4) a | 64.2 (58.6–73.5) a | < .001 |
| Normalized left cardiac output (L/min per square meter) | 4.53 (3.85–5.59) | 5.87 (4.71–7.81) a | 6.41 (5.85–7.16) a | .001 |
| Left ventricular thickening (mm) | 0.23 (0.07–0.38) | 0.33 (0.27–0.48) a | 0.33 (0.14–0.58) a | .023 |
| Left ejection fraction (%) | 68 (62–73) | 70 (67–74) | 71 (65–75) | .527 |
| MAPSE (mm) | 12.8 (11.2–15.1) | 9.4 (8.4–10.4) a | 9.8 (8.5–10.6) a | < .001 |
| Mitral S′ (cm/s) | 10 (9–11) | 10 (9–11) | 9 (8–10) a | .003 |
| Right ventricle | ||||
| TAPSE (mm) | 17.6 (15.4–19.5) | 16.5 (13.8–18.5) | 14.6 (12.8–16.1) a | .051 |
| Tricuspid S′ (cm/s) | 15 (14–17) | 16 (15–17) | 14 (12.5–15.5) a | .004 |
| Diastolic function | ||||
| Left ventricle | ||||
| Mitral E deceleration time (ms) | 88 (80–96) | 96 (84–104) | 96 (76–110) a | .046 |
| Mitral E/A ratio | 1.7 (1.4–1.9) | 1.8 (1.5–2) | 1.6 (1.5–2) | .450 |
| Mitral E′ (cm/s) | 19 (17–21) | 17.5 (16.5–19) | 17.5 (16.5–20) | .953 |
| Mitral A′ (cm/s) | 8 (7–9) | 7.5 (7–9) | 7 (6–9) a | .032 |
| Mitral E′/A′ | 2.33 (1.88–2.83) | 2.40 (2–2.65) | 2.43 (2–2.86) | .039 |
| Left IRT (ms) | 56 (50–64) | 64 (55–66) | 56 (48–60) | .382 |
| Mitral IRT′ (ms) | 48 (48–56) | 54 (52–60) a | 54 (48–60) | .861 |
| E/E′ | 5.26 (4.67–6.33) | 5.69 (5.13–6.23) | 5.37 (5.21–6.54) | .494 |
| Right ventricle | ||||
| Tricuspid E deceleration time, (ms) | 108 (96–124) | 94 (86–106) | 112 (104–124) a | .311 |
| Tricuspid E/A ratio | 1.4 (1.2–1.6) | 1.4 (1.2–1.8) | 1.6 (1.3–1.7) | .493 |
| Tricuspid E′ (cm/s) | 19 (18–21) | 20 (18–22) | 19 (17–21) | .184 |
| Tricuspid A′ (cm/s) | 11 (10–13) | 12.5 (10.5–14) | 11 (9–13) | .903 |
| Tricuspid E′/A′ | 1.68 (1.48–2) | 1.57 (1.46–1.8) | 1.64 (1.36–1.9) | .826 |
| Tricuspid IRT′ (ms) | 40 (36–46) | 40 (38–48) | 42 (40–48) | .111 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
