Accuracy of prenatal diagnosis of congenital heart defects.

OBJECTIVES: (1) To evaluate the detection and accuracy of fetal echocardiography for congenital heart defects (CHD). (2) To compare the detection and accuracy of a team of maternal fetal medicine specialists and radiologists (MFM/R) with those of perinatal cardiologists (PC). METHODS: All fetal echocardiograms performed from 1/91 to 6/95 were reviewed retrospectively. CHD diagnoses made prenatally were compared with postnatally-confirmed diagnoses. RESULTS: 106 heat defects were correctly detected prenatally. There were 10 incorrect diagnoses, 6 false-negatives and 4 false-positives. Overall detection rate (sensitivity) was 95%, while overall accuracy was 87%. Detection rate for MFM/R and PC were 91 and 97%, respectively (p = 0.4). Accuracy was 74% for MFM/R and 92% for PC (p = 0.02). CONCLUSIONS: CHD can be identified reliably by prenatal echocardiography. The involvement of a PC in the prenatal diagnosis of these defects may improve diagnostic accuracy.

Fetal heart rate and umbilical artery velocity variability in fetuses with congenital cardiac defects: a preliminary study.

OBJECTIVE: To examine whether variabilities in fetal heart rate and umbilical artery flow velocity are possible markers for hemodynamic dysfunction in fetuses with a congenital heart defect. METHODS: Doppler studies of the umbilical artery velocity waveform were performed at 20-35 weeks of gestation in 13 patients with a congenital heart defect. We determined absolute and variability values for heart rate and flow velocities from umbilical artery velocity waveforms of at least 18 s duration. We compared these findings with normal controls matched for gestational age. RESULTS: Fetuses with a congenital heart defect displayed decreased umbilical artery peak systolic and time-averaged velocities. However, variability in peak systolic and time-averaged velocities and fetal heart rate variability were increased compared with normal controls. Absolute fetal heart rates were similar between the two groups. CONCLUSIONS: Marked cardiovascular changes occur in the fetus with a congenital heart defect compared with the normal healthy fetus. We propose that variability in fetal heart rate and umbilical artery blood flow velocity could be additional markers for impaired homeostasis in the presence of fetal congenital heart disease.

Can Doppler systemic venous flow indices predict central venous pressure in children?

Tricuspid valve, superior vena cava (SVC), and hepatic vein Doppler patterns may be abnormal in right heart anomalies and have been used to predict high central venous pressure (CVP) in adults. The purpose of this study was to evaluate the relationship of these systemic venous flow indices to CVP in children. Children undergoing cardiac catheterization were studied prospectively using simultaneous recordings of mean CVP with pulsed-Doppler tracings of SVC, hepatic vein, and tricuspid valve flow. Systemic venous Doppler measurements included peak velocities and velocity time integrals for ventricular systole (S), ventricular diastole (D), and ventricular systole (B), and atrial systole (A). Tricuspid inflow Doppler E and A waves were recorded also. Patients with significant tricuspid stenosis or regurgitation, systemic venous obstruction, and nonsinus rhythm were excluded. The 42 patients ranged in age from 0.2 to 21.0 years and in weight from 3.0 to 68.0 kg. Mean CVPs ranged from 1 to 17 mmHg. Catheterization indications included hemodynamic evaluation (25 patients), transplant biopsy, (11 patients), and interventional procedures (6 patients). No SVC or tricuspid valve Doppler measurement correlated with CVP. Hepatic vein peak D, peak B, and peak A significantly correlated with CVP (r = 0.34 - 0.55; P < 0.05, linear regression). For all correlations, the r values were low with significant overlap among patients. Thus, in children, only hepatic vein peak velocities correlate with CVP. Because of the low r values and significant overlap among patients, the currently used Doppler indices have a low sensitivity for predicting CVP in this age group.

An unusual variant of total anomalous pulmonary venous connection with varices and multiple drainage sites.

Total anomalous pulmonary venous connection (TAPVC) is an uncommon cardiac anomaly that has also rarely been associated with esophageal atresia. We report an unusual case of esophageal atresia with TAPVC with several varices and multiple drainage sites into the superior vena cava and portal vein.

Spectrum and influence of hypoplasia of the left heart in neonatal aortic coarctation.

Obstruction of the left ventricular outflow tract may be associated with hypoplasia of the left heart, which importantly influences the options for treatment. Although the influence of the size of the left heart on the outcome for critical aortic stenosis has been described, less is known about the spectrum of such hypoplasia seen with neonatal aortic coarctation, and how this influences outcome. To determine, first, the spectrum and influence of hypoplasia of the left heart in neonatal coarctation, second, if the previously described critical values for adequacy of the left heart in neonates with critical aortic stenosis are applicable to neonates with coarctation, and, third, if any of the variables or associated abnormalities are risk factors for recoarctation, we studied 63 neonates who underwent repair of coarctation. From the initial echocardiogram, we measured multiple structures in the left heart, and calculated a score for adequacy as has been done for critical aortic stenosis. The sizes were compared to previously reported minimal values. We then analyzed the influence of the variables and the associated anomalies on outcome. There were no deaths. There was a broad spectrum of sizes that did not correlate with the need for re-intervention. The calculated score for adequacy would have predicted survival in only 56% of the patients, and 73% of the neonates had at least one parameter measured in the left heart below the previously reported minimal values. There is, therefore, a broad spectrum of sizes for the left heart in neonates with aortic coarctation that is not predictive of outcome. Minimal sizes, and the score for adequacy used for critical aortic stenosis, are not applicable to neonates with coarctation.

Influence of left ventricular cavity size on interventricular shunt timing and outcome in neonates with coarctation of the aorta and ventricular septal defect.

In this study, we reviewed the records and echocardiograms of 39 consecutive patients with coarctation of the aorta and ventricular septal defect who underwent neonatal coarctation repair to examine the influence of left heart size on ventricular shunting and outcome. We found smaller left heart structures (initial mitral and aortic annular diameters) to be associated with diastolic interventricular shunting and to be predictive of the development of mitral or aortic and/or subaortic stenosis.

Left heart hypoplasia and neonatal aortic arch obstruction: is the Rhodes left ventricular adequacy score applicable?

OBJECTIVE: Although the influence of small left heart structures on outcome of a biventricular repair in neonatal critical aortic stenosis is well documented, little is known about its effect in neonates with aortic arch obstruction and coarctation. The purpose of this study was to evaluate the influence of small left heart structures on early and late results of repair and the ability to achieve a biventricular repair in neonates with coarctation and aortic arch obstruction. PATIENTS: Neonates included in this study had a left ventricular adequacy score (as proposed by Rhodes and associates for critical aortic stenosis) that would have predicted a need for a univentricular (Norwood) repair. All were ductus dependent but had antegrade ascending aortic flow and a small but nonstenotic aortic valve (<30 mm Hg gradient). Twenty neonates aged 10 +/- 9 days were identified for the study with weights averaging 3. 1 +/- 0.6 kg. Selected left heart measurements obtained by preoperative echocardiography included the following: aortic anulus 5.3 +/- 0.3 mm, mitral anulus 8.4 +/- 1.0 mm, transverse aortic arch 3.4 +/- 0.6 mm, and left ventricular volume 25 +/- 4 mL/m2. All patients underwent coarctation repair by resection and extended end-to-end anastomosis to enlarge the transverse arch as needed. Three patients underwent simultaneous pulmonary artery banding because of a hemodynamically significant ventricular septal defect. These 3 patients have subsequently had their defects successfully closed without mortality. RESULTS: There were no early or late deaths at a follow-up of 38 +/- 16 months after the operation. Three patients (3/20, 15%) have had to undergo reintervention with balloon aortoplasty because of recurrent coarctation (gradient > 20 mm Hg) in 2 and resection of subaortic stenosis in 1. Late follow-up in the remaining patients reveals 1 with moderate subaortic stenosis (gradient = 43 mm Hg), 2 with mild aortic stenosis (gradient < 30 mm Hg), and 2 with mild to moderate mitral stenosis. At late follow-up, 16 patients (16/20, 80%) are completely free of symptoms and 4 (4/20, 20%) have mild residual symptoms. CONCLUSIONS: Biventricular physiology can be successfully achieved in neonates with small left heart structures and aortic arch obstruction with minimal mortality and excellent late functional results. Standard echocardiographic measurements used to predict the need for a univentricular repair in critical aortic stenosis are not valid for the neonate with aortic arch obstruction.

Echocardiographic diagnosis of thrombus originating from the ductus arteriosus.

Initial functional closure of the ductus arteriosus normally occurs within hours after birth, with permanent closure taking several weeks. The mechanism for ductal closure has been well studied and has not been shown to include thrombus formation. We describe a normal infant found to have a thrombus originating in the ductus arteriosus that occluded the ductus and subsequently extended into the left pulmonary artery, threatening to occlude it as well. This case illustrates the importance of echocardiography in making this rare diagnosis. It also emphasizes the role of echocardiography as an effective means of following the progression or regression of such a thrombus.

Usefulness of echocardiography for detection of coronary artery thrombi in patients with Kawasaki disease.

To evaluate the role of echocardiography for predicting and accurately detecting thrombi in patients with abnormal coronary arteries after Kawasaki disease, we reviewed the echocardiograms of 40 consecutive patients and compared echocardiographic findings with angiographic findings when available. Patients with Kawasaki disease who had coronary artery aneurysms > or =5 mm had significantly greater multivessel involvement, thrombi, and persistent coronary abnormalities than those with diameters <5 mm.

Size of ventricular structures influences surgical outcome in Down syndrome infants with atrioventricular septal defect.

To evaluate the relation between ventricular structure size and surgical outcome in Down versus non-Down syndrome infants with an atrioventricular septal defect, we reviewed the charts and echocardiograms of 44 consecutive infants (34 with Down syndrome) who underwent atrioventricular septal defect repair. Children with Down syndrome had significantly greater aortic valve diameters, left ventricular valve areas, and left/right atrioventricular valve area ratios as well as fewer adverse outcomes than non-Down syndrome children.

Doppler echocardiography distinguishes between physiologic and pathologic "silent" mitral regurgitation in patients with rheumatic fever.

BACKGROUND: The diagnosis of rheumatic fever is based on physical findings (major) and supporting laboratory evidence (minor) as defined by the Jones criteria. Rheumatic carditis is characterized by auscultation of a mitral regurgitant murmur. Doppler echocardiography, however, may detect mitral regurgitation when there is no murmur ("silent" mitral regurgitation), even in normal individuals. HYPOTHESIS: The hypothesis of this study was that physiologic mitral regurgitation can be differentiated from pathologic "silent" mitral regurgitation by Doppler echocardiography. METHODS: The study group consisted of 68 patients (2-27 years) with normal two-dimensional imaging and Doppler evidence of mitral regurgitation but no murmur. Patients with rheumatic fever (n = 37) met Jones criteria (chorea in 20, arthritis in 17). Patients without rheumatic fever (n = 31) were referred for innocent murmur (n = 7), abnormal electrocardiogram (n = 13), and chest pain (n = 11). Echoes were independently reviewed by two cardiologists blinded to the diagnosis. Pathologic mitral regurgitation was defined as meeting the following four criteria: (1) length of color jet > 1 cm, (2) color jet identified in at least two planes, (3) mosaic color jet, and (4) persistence of the jet throughout systole. Jet orientation was also noted. RESULTS: Using the above criteria, there was agreement in echo interpretation of pathologic versus physiologic mitral regurgitation in 67 of 68 patients (interobserver variability of 1.5%). Pathologic regurgitation was found in 25 (68%) patients with rheumatic fever but in only 2 (6.5%) patients without rheumatic fever (p < 0.001). The specificity of Doppler for detecting pathologic regurgitation was 94% with a positive predictive value of 93%. The color mitral regurgitant jet was posteriorly directed in all 25 patients with rheumatic fever. CONCLUSION: Pathologic "silent" mitral regurgitation of rheumatic fever can be distinguished from physiologic mitral regurgitation using strict Doppler criteria, particularly when the jet is directed posteriorly. These data support the use of Doppler echocardiography as a minor criterion for evaluating patients with suspected rheumatic fever.

Xq28-linked noncompaction of the left ventricular myocardium: prenatal diagnosis and pathologic analysis of affected individuals.

Isolated noncompaction of the left ventricular myocardium (INVM) is characterized by the presence of numerous prominent trabeculations and deep intertrabecular recesses within the left ventricle, sometimes also affecting the right ventricle and interventricular septum. Familial occurrence of this disorder was described previously. We present a family in which 6 affected individuals demonstrated X-linked recessive inheritance of this trait. Affected relatives presented postnatally with left ventricular failure and arrhythmias, associated with the pathognomonic echocardiographic findings of INVM. The usual findings of Barth syndrome (neutropenia, growth retardation, elevated urinary organic acids, low carnitine levels, and mitochondrial abnormalities) were either absent or found inconsistently. Fetal echocardiograms obtained between 24-30 weeks of gestation in 3 of the affected males showed a dilated left ventricle in one heart, but were not otherwise diagnostic of INVM in any of the cases. Four of the affected individuals died during infancy, one is in cardiac failure at age 8 months, and one is alive following cardiac transplant at age 9 months. The hearts from infants who died or underwent transplantation appeared, on gross examination, to be enlarged, with coarse, deep ventricular trabeculations and prominent endocardial fibroelastosis. Histologically, there were loosely organized fascicles of myocytes in subepicardial and midmyocardial zones of both ventricles, and the myocytes showed thin, often angulated fibers with prominent central clearing and reduced numbers of filaments. Markedly elongated mitochondria were present in some ventricular myocytes from one specimen, but this finding was not reproducible. Genetic linkage analysis has localized INVM to the Xq28 region, where other myopathies with cardiac involvement have been located.