Ultrasonographic and Biochemical Markers of Human Fetal Cardiac Dysfunction in Placental Insufficiency
Background— Placental insufficiency may lead to fetal cardiovascular compromise. We sought to determine whether ultrasonographic parameters of fetal cardiovascular function correlate with umbilical arterial levels of biochemical markers of myocardial dysfunction and damage in placental insufficiency.
Methods and Results— In 48 fetuses with placental insufficiency, umbilical artery blood was obtained at delivery for assessment of N-terminal peptide of proatrial natriuretic peptide (NT-proANP) and cardiac troponin-T (cTnT). Group 1 fetuses (n=12) had normal NT-proANP and cTnT serum concentrations. Group 2 fetuses (n=25) showed increased NT-proANP (>1145 pmol/L) and normal cTnT values. Group 3 fetuses (n=11) had increased NT-proANP and cTnT (>0.10 ng/mL) levels. The ultrasonographic parameters of fetal cardiovascular function were compared between the groups. Pulsatility indices for veins of the ductus venosus, left hepatic vein, and inferior vena cava correlated significantly with NT-proANP levels. In group 3, ductus venosus, left hepatic vein, and inferior vena cava pulsatility indices for veins were higher (P<0.01) than in groups 1 and 2. The proportion of left ventricular cardiac output of combined cardiac output was greater (P<0.05) and that of right ventricle was smaller (P<0.05) in group 3 than in group 2. In group 3, tricuspid regurgitation was noted most often (P<0.05), and right ventricular fractional shortening was less (P<0.01) than in group 2.
Conclusions— Pulsatility in human fetal systemic veins correlated significantly with the cardiac secretion of ANP. Fetuses with myocardial damage demonstrate increased systemic venous pressure, a change in the distribution of cardiac output toward the left ventricle, and a rise in right ventricular afterload.
Received November 19, 2001; revision received February 27, 2002; accepted February 28, 2002.
Redistribution of arterial circulation has been documented in placental insufficiency.1 It is a protective mechanism maintaining oxygen supply to the most vital fetal organs, the heart and brain. In addition, redistribution of fetal cardiac output in favor of the left ventricle has been described in these complicated pregnancies.2 In placental insufficiency, abnormalities in the fetal venous circulation detected by Doppler ultrasonography have been proposed to indicate the need for delivery.3 We have shown that in placental insufficiency, fetuses with a normal umbilical vein blood flow profile have increased atrial natriuretic peptide (ANP) production,4 whereas an abnormal umbilical vein blood flow pattern is associated with fetal myocardial cell damage.5 In this study, we tested the hypothesis that there exists a relationship between ultrasonographic and biochemical markers of fetal cardiac dysfunction and myocardial cell damage in pregnancies complicated by placental insufficiency.
Forty-eight consecutive fetuses with signs of placental insufficiency (fetal growth <10th percentile growth curve and/or abnormal umbilical artery [UA] blood velocity waveform pattern5) referred to our perinatal unit between May 1998 and September 2000 were included in this cross-sectional study. The local ethics committee approved the research protocol, and all the women gave informed consent. The decision to break the study population into 3 subgroups based on UA concentrations of fetal N-terminal peptide of proatrial natriuretic peptide (NT-proANP) and cardiac troponin-T (cTnT) was made before data analysis. In addition, the investigators obtaining and analyzing ultrasonographic data were blinded to the biochemical data. Gestational age was confirmed by ultrasonographic examination before 20 weeks of gestation in all cases. All fetuses had a normal karyotype and no structural abnormalities.
Image-directed pulsed and color Doppler equipment (Acuson Sequoia 512, Mountain View) was used with a 4- to 8-MHz convex or a 5-MHz sector probe. The high-pass filter was set at minimum. The acoustic output of the system was controlled according to European Federation of Societies for Ultrasound in Medicine and Biology guidelines.6 An angle of <15 degrees between the vessel and the Doppler beam was accepted for analysis. Three consecutive cardiac cycles were assessed, and mean values were used for further analysis. The diameters of pulmonary (PV) and aortic (AoV) valves were measured in frozen real-time images during systole by using the leading-edge method.7 Three separate measurements of valve diameter were taken, and the mean value was used for the calculation of the cross-sectional area (CSA) of the valve.
Placental vascular impedance was assessed by calculating the UA pulsatility index value (PI=[peak systolic velocity−end-diastolic velocity]/time-averaged maximum velocity over the cardiac cycle). Umbilical artery blood velocity waveforms were obtained from free loops of the umbilical cord.
The distribution of arterial circulation was evaluated by calculating the descending aorta (DAo) and the middle cerebral artery (MCA) PI values and the DAo/MCA PI ratio. From aortic isthmus (AoI) blood velocity waveforms, time-velocity integrals (TVIs) of antegrade and retrograde components were obtained by planimetry of the area underneath the Doppler spectrum, and their ratio was calculated. Net blood flow was considered antegrade if the ratio was ≥1 and retrograde when the ratio was <1.8 Visualization of coronary artery circulation by color and pulsed Doppler ultrasonography was noted.9
Volumetric blood flows (Q) across PV and AoV were calculated (Q=CSA×TVI×FHR), in which FHR is the fetal heart rate. Right ventricular cardiac output (RVCO) equals the QPV and left ventricular cardiac output (LVCO) equals the QAoV, and their sum is combined cardiac output (CCO). The proportions (%) of RVCO and LVCO of CCO and weight-indexed RVCO, LVCO, and CCO were calculated. Because the time interval between the ultrasonographic examination and delivery was ≤4 days in all cases, actual birth weight was used for indexing purposes.
Systolic function of the fetal heart was assessed by calculating the right (RVeFo) and left (LVeFo) ventricular ejection forces by the following formula: (1.055×CSA×TVIac)×(PSV/TTP), in which TVIac is the TVI during the acceleration period of systole, PSV is the peak systolic velocity, and TTP is the time to peak velocity interval.10 Ventricular ejection forces were weight indexed. An index of myocardial performance (IMP), which describes the combined systolic and diastolic function of the heart, was calculated by the following formula: IMP=(ICT+IRT)/ET, in which ICT is the isovolumetric contraction time, IRT is the isovolumetric relaxation time, and ET is the ejection time.11 In the presence of tricuspid regurgitation (TR), dP/dT was calculated to assess the contractility of the ventricle.12
Diastolic function of the fetal heart was described by calculating the proportion (%) of left ventricular IRT of the total cardiac cycle.13 Mitral (MV) and tricuspid (TV) valve total TVIs were measured, and the ratio between TVIs of E-wave (early passive filling) and A-wave (atrial contraction) were calculated. In addition, MV and TV A-TVI/total TVI ratios, which reflect ventricular compliance, were obtained.13
Afterload of the fetal heart was described by assessing the ratio between the cardiac (CC) and thoracic circumferences (TC) at the level of a 4-chamber view.14 Right (RVFS) and left (LVFS) ventricular fractional shortenings were calculated from M-mode recordings using the following formula: ventricular fractional shortening [%]=[(inner diastolic diameter−inner systolic diameter)/inner diastolic diameter]×100).15 Tricuspid regurgitation was noted.16
From the fetal venous circulation, pulsatility index values for veins (PIV=[peak systolic velocity−velocity during atrial contraction]/time-averaged maximum velocity over the cardiac cycle) for the ductus venosus (DV), left hepatic vein (LHV), and inferior vena cava (IVC) were calculated.3 Atrial pulsations in the intra-abdominal umbilical vein were noted.
Intraobserver variability of Doppler parameters was analyzed in 23 women. Seventeen women had uncomplicated pregnancy and labor, and in 6 cases pregnancy was complicated by placental insufficiency and fetal growth restriction.
Biochemical Markers of Cardiac Dysfunction
Immediately after delivery, UA blood samples of ≥1 mL were drawn and centrifuged. Serum samples were stored at −80°C. Concentrations of fetal NT-proANP and cTnT (Enzymum-Test Troponin-T, Boehringer Mannheim Diagnostics) were determined.4,5⇓ We previously found that in neonates born after uncomplicated pregnancy and labor, the +2 SD level of NT-proANP was 1145 pmol/L4, which was chosen to represent the cut-off level. A clinically significant cTnT concentration was set at 0.10 ng/mL5.
Group 1 consisted of 12 fetuses with normal serum concentrations of UA NT-proANP (<1145 pmol/L) and cTnT (<0.10 ng/mL). Group 2 consisted of 25 fetuses with increased NT-proANP (>1145 pmol/L) and normal cTnT levels (<0.10 ng/mL). The time interval between the Doppler ultrasonographic examination and delivery ranged from 0 to 4 days, with a median value of 1 and 2 days in groups 1 and 2. Group 3 was comprised of 11 fetuses with elevated NT-proANP (>1145 pmol/L) and cTnT (>0.10 ng/mL) levels. The time interval between the Doppler ultrasonographic examination and delivery ranged from 0 to 2 days, with a median value of 0 days. Maternal hypertensive disorders were classified according to American College of Obstetricians and Gynecologists guidelines.17 Antenatal and perinatal data of the groups are given in Table 1.
Statistical analysis was performed by ANOVA when comparisons were made between the 3 groups and the data were normally distributed. If statistical significance was shown, the Scheffe F-test was used for additional analysis. If the data were not normally distributed, the nonparametric Kruskal-Wallis test was chosen. Categorical data were compared using the χ2 test. Linear regression analysis was used to show the relationship of NT-proANP concentration to PIV values in venous circulation. P≤0.05 was selected as the level of statistical significance.
Gestational age at delivery, birth weight, and 5-minute Apgar scores were significantly lower in group 3 than in groups 1 and 2. In group 3, NT-proANP concentrations were higher than in group 2 (Table 1). Cesarean delivery was performed because of signs of fetal distress in 6 of 12 cases in group 1 and in 9 of 25 and 8 of 11 cases in groups 2 and 3.
The mean intraobserver variability of RVCO and LVCO calculations was from 5.4% to 6.3% (95% CI, 4.7 to 7.9). The mean intraobserver variabilities of MV and TV total TVI calculations ranged from 5.0% to 5.2% (95% CI, 2.4 to 7.8), and the mean variability of ventricular ejection force calculations was 8.8% (95% CI, 4.7 to 12.8). The corresponding variability of UA, DAo, and MCA PI calculations was from 3.9% to 6.0% (95% CI, 2.5 to 9.5). The mean intra-observer variability of DV, LHV, and IVC PIV calculations ranged from 3.8% to 5.8% (95% CI, 1.9 to 7.5). In time-interval calculations, the corresponding variability was from 8.0% to 9.7% (95% CI, 5.4 to 13.2).
In group 3, UA PI values were higher than in groups 1 and 2 (P<0.001). No difference in the DAo and MCA PI values or in the DAo/MCA PI ratio was found between the groups. The detection rate of retrograde AoI net blood flow and the visualization rate of coronary artery circulation were similar among the groups (Table 2).
Weight-indexed RVCO, LVCO, and CCO did not differ between the groups. However, in group 3, the LVCO percent was greater (P<0.05) and the RVCO percent was smaller (P<0.05) than in group 2 (Table 3). Weight-indexed RVeFo and LVeFo did not differ between the groups. Right ventricular contractility, assessed by dP/dT of TR, was similar among the groups, with a mean value of 717 mm Hg/s in group 1 and 820 and 605 mm Hg/s in groups 2 and 3, respectively. No difference in the IRT percent was found. The total TVIs of TV and MV and their E/A TVI- and A-TVI/total TVI ratios did not differ between the groups. The IMP was similar among the groups (Table 4).
The CC/TC ratio was greater (P<0.05) in group 3 than in groups 1 and 2. In group 3, RVFS was less than in group 2 (P<0.01), whereas LVFS did not differ between the groups. The TR was more common in group 3 (6 of 11 patients) than in group 2 (3 of 25 patients) (P<0.03) (Table 5).
To examine the correlation between ultrasonographic parameters of cardiac function and gestational age, the study groups were combined (n=48). No statistically significant correlations were found between gestational age and ultrasonographic parameters of cardiac function.
The DV, LHV, and IVC PIV values were higher (P<0.01) in group 3 than in groups 1 and 2 (Table 6). In group 3, atrial pulsations in the umbilical vein were detected more often (P<0.01) than in groups 1 and 2. Significant correlations were found between NT-proANP concentrations (n=48) and the DV, LHV, and IVC PIV values (Figure).
Fetuses that demonstrated myocardial cell damage (cTnT >0.10 ng/mL) had signs of increased systemic venous pressure, a change in the distribution of cardiac output in favor of the left ventricle, and a rise in right ventricular afterload. In these cases, placental insufficiency was more severe and fetuses were delivered earlier than fetuses without biochemical evidence of myocardial cell damage.
Fetuses with myocardial cell damage had increased pulsatility in the DV, LHV, and IVC blood velocity waveforms. Studies on fetal lambs have shown that a rise in systemic venous pressure is associated with increased pulsatility in the systemic venous blood velocity waveforms.18 In the present study, significant correlations were found between NT-proANP concentrations and the DV, LHV, and IVC PIV values. The NT-proANP concentrations were highest in fetuses with myocardial cell damage. NT-proANP is released in equimolar amounts with ANP from atrial myocytes. The half-life of NT-proANP is longer than that of ANP. This is reflected in the proportionally larger increase in NT-proANP levels than in ANP concentrations. Therefore, NT-proANP is used to characterize endogenous secretion of ANP.19 Atrial stretch is the most important stimulus regulating the secretion of proANP-derived peptides. However, the ANP gene is very actively expressed in fetal and neonatal ventricle.20 Thus, increased afterload could be a major stimulus for ANP secretion in fetuses. Our findings suggest that in human fetuses, the DV, LHV, and IVC PIV values reflect atrial stretch and systemic venous pressure. It has been shown that changes in umbilical venous velocities originate in the fetal venous system and are transmitted to the placenta.21 Another explanation for increased pulsatility in the DV and LHV could be diminished placental volume blood flow in placental insufficiency. However, elevated NT-proANP levels suggest that increased pulsatility in the systemic venous blood velocity waveforms originated from the heart.
The right ventricular afterload was increased in fetuses with elevated cTnT levels. This is based on decreased RVFS, more frequent visualization of TR, and increased cardiac size. The right ventricle mainly reflects the circulation and resistance in the fetal lower body, placenta, and pulmonary bed. The left ventricle is responsible for the circulation in the coronary and cerebral arteries and fetal upper body. In fetal lambs, an acute increase in right ventricular afterload by means of ductal occlusion significantly decreased RVFS, and TR was immediately evident. All these changes resolved after ductal occlusion was released.22 It has also been shown that in human fetuses with increased vascular impedance in the descending aorta, RVFS is decreased with no change in the left ventricle.23 An abnormal fractional shortening could reflect myocardial compromise or an increase in the fetal ventricular workload. In this study, other parameters describing fetal cardiac systolic and diastolic functions did not differ between the groups, suggesting that decreased RVFS reflected increased afterload in these fetuses. The increased right ventricular afterload seemed to originate from the placenta. In normal pregnancies, the incidence of TR is ≈6% during the second half of gestation.16 In the present study, the incidence of TR was higher, especially in fetuses with myocardial cell damage (54.5%). However, similar TR dP/dT between the groups suggest that the right ventricle is able to maintain its contractility despite increased afterload. Fetal cardiomegaly can be caused by a variety of reasons, including volume overload and increased afterload. In placental insufficiency, placental volume blood flow and preload are usually decreased. In this study, cardiac size was increased in fetuses with more severe placental insufficiency, suggesting that increased afterload leads to a rise in relative cardiac size.
The weight-indexed CCO did not differ significantly between the groups, although a trend toward lower CCO was observed in group 3. In addition, a similar trend was found in weight-indexed RVCO, whereas weight-indexed LVCO did not show this trend. Therefore, in fetuses with myocardial cell damage, relatively greater proportion of CCO was directed through the left ventricle than in fetuses with only increased NT-proANP levels. This suggests that fetuses with increased right ventricular afterload shift their cardiac output from the right to the left ventricle. Previously, it has been shown that in growth-restricted fetuses with more severe placental insufficiency, RVCO percent was significantly less than in fetuses with relatively mild placental insufficiency.2
Similar weight-indexed ventricular ejection forces between the groups show that fetuses with elevated cTnT levels are able to maintain adequate systolic function of the heart. Ejection force, which is not affected by afterload, estimates the energy transferred from ventricular myocardial shortening to work done by accelerating blood into the circulation.24
Fetal cardiac diastolic function described by the IRT percent did not differ between the groups. Relaxation of the myocardium is an active process depending on the ability of the myocytes to accelerate calcium transport through Na+-Ca2+ channels,25 and IRT is the time interval needed for the ventricle to drop its pressure from a systemic to an atrial level. It seems that global myocardial function and ventricular compliance are not changed even in conditions with significant myocardial cell damage.
Redistribution of fetal arterial circulation, as determined by the DAo/MCA PI ratio, was similar among the groups. In addition, the incidences of retrograde AoI net blood flow and visualization of coronary artery blood flow did not differ between the groups. Retrograde AoI net blood flow has been associated with diminished oxygen delivery to the cerebral circulation.8 Visualization of coronary artery blood flow has been documented in growth-restricted fetuses, and it has been postulated that these fetuses are at a high risk of intrauterine death.26 Fetuses with myocardial cell damage demonstrated increased right ventricular afterload. A rise in pressure causes a greater demand for oxygen in the myocardium than an increase in volume load. The wall stress in the right ventricle is increased after a rise in afterload, thus increasing oxygen consumption by the myocardium. In the present study, this is supported by the fact that the right ventricular systolic diameter remained proportionally greater in fetuses with myocardial cell damage. However, it seems that fetal compensatory mechanisms, including coronary artery vasodilatation and increased blood flow, are able to meet the increased oxygen demand, because the incidences of demonstrable coronary artery circulation were similar among the groups. Studies on fetal lambs have shown large coronary vascular reserves in response to diminished oxygen content of the blood.27 It seems that fetal myocardial cell damage is more related to increased afterload and pressure than to diminished oxygen delivery to the myocardium. However, in the subendocardial region, the increase in wall tension is greater and the reserve of the coronary circulation is less than in the subepicardial area. Thus, the subendocardial area could be more vulnerable to hypoxemia in the human fetus.
The presence of abnormal systemic venous blood velocity waveforms has been suggested to indicate the need for delivery.3 Previously, we showed that in the presence of atrial pulsations in the intra-abdominal umbilical vein, fetal cTnT concentrations are elevated.5 The present results show that abnormalities in more proximal portion of the fetal venous circulation indicate fetal myocardial cell damage and increased systemic venous pressure.
In human fetuses, volumetric blood flow measurements by Doppler ultrasonography have been shown to be valid.7 The intraobserver variability of volumetric blood flow calculations in this study was comparable to that in previous reports. Intraobserver variability in fetal ventricular ejection force calculations has been <10%,10 which is similar to our study. The PI and PIV calculations showed intraobserver variability of <6%, which is comparable with previous reports. The cTnT assay has shown excellent reproducibility and a statistically significant ability to distinguish between values differing in the low range by 0.01 ng/mL.28 The sensitivity of the NT-proANP assay is 40 pmol/L. The intra-assay and interassay coefficients of variation are <10% and <15%, respectively.29 There is no significant transplacental passage of NT-proANP, and mode of delivery does not influence UA plasma ANP concentrations.4,30⇓ In addition, atrial natriuretic factor concentrations do not change with gestation.31 Gestational age at delivery and mode of delivery do not affect newborn cTnT levels, at least after 27 weeks’ gestation.5 One limitation of this study could be that some of the ultrasonographically obtained cardiovascular parameters change with advancing gestation in uncomplicated pregnancies.10,11,13,32⇓⇓⇓ However, in the present study in which all the pregnancies were complicated by placental insufficiency, cardiovascular parameters obtained by Doppler ultrasonography did not correlate with gestational age, which is in agreement with earlier studies.10,11,32⇓⇓
In conclusion, pulsatility in human fetal systemic veins correlated significantly with the cardiac production of ANP. Fetuses with myocardial cell damage demonstrate a rise in systemic venous pressure, a change in the distribution of cardiac output toward the left ventricle, and a rise in right ventricular afterload.
This work was supported by the Inkeri and Mauri Vänskä Research Foundation, the Paulo Research Foundation, and the Academy of Finland.
- ↵Wladimiroff JW, vd Wijngaard JA, Degani S, et al. Cerebral and umbilical arterial blood flow velocity waveforms in normal and growth-retarded pregnancies. Obstet Gynecol. 1987; 69: 705–709.
- ↵al-Ghazali W, Chita SK, Chapman MG, et al. Evidence of redistribution of cardiac output in asymmetrical growth retardation. Br J Obstet Gynaecol. 1989; 96: 697–704.
- ↵Hecher K, Campbell S, Doyle P, et al. Assessment of fetal compromise by Doppler ultrasound investigation of the fetal circulation: arterial, intracardiac, and venous blood flow velocity studies. Circulation. 1995; 91: 129–138.
- ↵Makikallio K, Vuolteenaho O, Jouppila P, et al. Umbilical artery N-terminal peptide of proatrial natriuretic peptide in hypertensive pregnancies and fetal acidemia during labor. Obstet Gynecol. 2001; 97: 23–28.
- ↵Makikallio K, Vuolteenaho O, Jouppila P, et al. Association of severe placental insufficiency and systemic venous pressure rise in the fetus with increased neonatal cardiac troponin T levels. Am J Obstet Gynecol. 2000; 183: 726–731.
- ↵European Committee for Medical Ultrasound Safety (ECMUS). Thermal teratology. Eur J Ultrasound. 1999; 9: 281–283.
- ↵Rasanen J, Wood DC, Weiner S, et al. Role of the pulmonary circulation in the distribution of human fetal cardiac output during the second half of pregnancy. Circulation. 1996; 94: 1068–1073.
- ↵Fouron JC, Skoll A, Sonesson SE, et al. Relationship between flow through the fetal aortic isthmus and cerebral oxygenation during acute placental circulatory insufficiency in ovine fetuses. Am J Obstet Gynecol. 1999; 181: 1102–1107.
- ↵Baschat AA, Gembruch U, Reiss I, et al. 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; 9: 162–172.
- ↵Rizzo G, Capponi A, Rinaldo D, et al. Ventricular ejection force in growth-retarded fetuses. Ultrasound Obstet Gynecol. 1995; 5: 247–255.
- ↵Tsutsumi T, Ishii M, Eto G, et al. Serial evaluation for myocardial performance in fetuses and neonates using a new Doppler index. Pediatr Int. 1999; 41: 722–727.
- ↵Tulzer G, Gudmundsson S, Rotondo KM, et al. Doppler in the evaluation and prognosis of fetuses with tricuspid regurgitation. J Matern Fetal Invest. 1991; 1: 15–18.
- ↵Tulzer G, Khowsathit P, Gudmundsson S, et al. Diastolic function of the fetal heart during second and third trimester: a prospective longitudinal Doppler-echocardiographic study. Eur J Pediatr. 1994; 153: 151–154.
- ↵Respondek M, Respondek A, Huhta JC, et al. 2D echocardiographic assessment of the fetal heart size in the 2nd and 3rd trimester of uncomplicated pregnancy. Eur J Obstet Gynecol Reprod Biol. 1992; 44: 185–188.
- ↵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; 150: 981–988.
- ↵Respondek ML, Kammermeier M, Ludomirsky A, et al. The prevalence and clinical significance of fetal tricuspid valve regurgitation with normal heart anatomy. Am J Obstet Gynecol. 1994; 171: 1265–1270.
- ↵American College of Obstetricians and Gynecologists. Hypertension in Pregnancy. ACOG technical bulletin No. 219. Washington, DC: American College of Obstetricians and Gynecologists, 1996.
- ↵Gudmundsson S, Gunnarsson GO, Hokegard KH, et al. Venous Doppler velocimetry in relationship to central venous pressure and heart rate during hypoxia in the ovine fetus. J Perinat Med. 1999; 27: 81–90.
- ↵Itoh H, Nakao K, Sugawara A, et al. γ-Atrial natriuretic polypeptide (γ ANP)-derived peptides in human plasma: cosecretion of N-terminal γ ANP fragment and α ANP. J Clin Endocrinol Metab. 1988; 67: 429–437.
- ↵Sagnella GA. Measurement and significance of circulating natriuretic peptides in cardiovascular disease. Clin Sci. 1998; 95: 519–529.
- ↵Reed KL, Anderson CF. Changes in umbilical venous velocities with physiologic perturbations. Am J Obstet Gynecol. 2000; 182: 835–838.
- ↵Tulzer G, Gudmundsson S, Rotondo KM, et al. Acute fetal ductal occlusion in lambs. Am J Obstet Gynecol. 1991; 165: 775–778.
- ↵Rasanen J, Kirkinen P, Jouppila P. Right ventricular dysfunction in human fetal compromise. Am J Obstet Gynecol. 1989; 161: 136–140.
- ↵Noble MIM, Trenchard D, Guz A. Left ventricular ejection in conscious dogs, II: determinants of stroke volume. Circ Res. 1966; 19: 148–152.
- ↵Mahony L. Calcium homeostasis and control of contractility in the developing heart. Semin Perinatol. 1996; 20: 510–519.
- ↵Baschat AA, Gembruch U, Harman CR. Coronary blood flow in fetuses with intrauterine growth restriction. J Perinat Med. 1998; 26: 143–156.
- ↵Reller MD, Burson MA, Lohr JL, et al. Nitric oxide is an important determinant of coronary flow at rest and during hypoxemic stress in fetal lambs. Am J Physiol. 1995; 269: H2074–H2081.
- ↵Lipshultz SE, Rifai N, Sallan SE, et al. Predictive value of cardiac troponin T in pediatric patients at risk for myocardial injury. Circulation. 1997; 96: 2641–2648.
- ↵Vuolteenaho O, Koistinen P, Martikkala V, et al. Effect of physical exercise in hypobaric conditions on atrial natriuretic peptide secretion. Am J Physiol. 1992; 263: R647–R652.
- ↵Kingdom JC, McQueen J, Connell JM, et al. Maternal and fetal atrial natriuretic peptide levels at delivery from normal and growth retarded pregnancies. Br J Obstet Gynaecol. 1992; 99: 845–849.
- ↵Ville Y, Proudler A, Abbas A, et al. Atrial natriuretic factor concentration in normal, growth-retarded, anemic, and hydropic fetuses. Am J Obstet Gynecol. 1994; 171: 777–783.
- ↵Rizzo G, Arduini D, Romanini C, et al. Doppler echocardiographic assessment of atrioventricular velocity waveforms in normal and small-for-gestational-age fetuses. Br J Obstet Gynaecol. 1988; 95: 65–69.