Clinical applications of echocardiography in infants and children. II. Estimation of aortic root diameter and left atrial size: a comparison between echocardiography and angiocardiography

Evolution of echocardiography in neonatal diagnosis

In the late 1960's, Edler and Lundström introduced ¿ultrasoundcardiography¿ for the evaluation of congenital heart disease. Initial evaluations using A- and M-mode echocardiography produced non-invasive diagnosis of many defects, including specific complex malformations such as hypoplastic left heart, Ebstein's malformation, endocardial cushion defect and transposition, all with single crystal techniques. Normal values for dimensions related to patient size and indices of function developed at that time remain as components of contemporary examinations. Two-dimensional imaging technology has evolved from 20 channels on Bom's linear array to 128-channel systems currently providing detailed imaging of structures as small as neonatal coronary arteries. The contribution of Doppler techniques for qualitative evaluation of blood flow characteristics has been greatly augmented by both the quantitative Doppler methods for accurate assessment of pressure gradients and pulmonary pressure, and by the development of color Doppler display of intracardiac and intravascular flow. These contributions have come from centers worldwide, with many initial and ongoing contributions from Lund. The evolution of instruments, and of application, now provides neonatal echocardiographic delineation of anatomic detail, function and hemodynamics of sufficient clarity and accuracy to replace the need for invasive study, or alternative technologies, in most cases.

Preoperative two-dimensional echocardiographic prediction of prosthetic aortic and mitral valve size in children

Preoperative two-dimensional echocardiograms (2D-ECHOs) were done on 18 patients to determine the accuracy of assessing the mitral and aortic anulus diameter in children undergoing valve replacement. Fourteen patients underwent primary valve replacement and four underwent repeat valve replacement. The mean age was 11.5 years (range 2 to 17 years). The 2D-ECHO measured mitral or aortic valve anulus was compared with the external diameter of the largest prosthetic valve that could be inserted. There was a strong correlation (r = 0.99, p less than 0.001, SEE = 1.0 mm) between the 2D-ECHO measurements and the prosthetic valve size in patients undergoing primary valve replacement, but the correlation (r = 0.16, p = NS) was poor for those undergoing repeat valve replacement. In conclusion, 2D-ECHO prediction of prosthetic mitral and aortic valve size is accurate in children undergoing primary valve replacement but is poor in those undergoing repeat valve replacement.

Aortic function during normal human pregnancy

Arterial rupture and dissection have been observed in both normal and complicated pregnancies. To understand how arterial mechanical properties might change during pregnancy, we measured aortic root diameter by M-mode echocardiography and obtained simultaneous cuff blood pressures during systole and diastole at rest and during isometric exercise. Measurements were obtained in 19 women at the end of normal gestation and again 3 to 6 months post partum. The aorta is larger and more compliant during normal human pregnancy, and these changes may not revert to prepregnant levels post partum. Increased aortic compliance, combined with decreased vascular resistance, may enhance left ventricular performance in normal pregnancy. Conversely, the presumed structural alterations leading to increased compliance may weaken diseased vessels leading to dissection or rupture during pregnancy.

Two-dimensional echocardiographic preoperative prediction of prosthetic aortic valve size

Aortic root diameter was measured by two-dimensional (2DE) and M-mode echocardiography in 20 adult patients (aged 29 to 69 years) undergoing aortic valve replacement in order to predict prosthetic aortic valve size. Eight patients had predominantly aortic stenosis, six had chronic aortic regurgitation, and six had acute severe aortic regurgitation secondary to infective endocarditis. 2DE measurements of aortic anulus diameter, as determined from the parasternal long-axis view, demonstrated a high correlation with actual prosthetic valve size implanted at surgery (r = 0.89, p less than 0.001, SEE 0.68 mm). 2DE exactly predicted actual prosthetic valve size in 12 of 20 patients (60%), was within 1 mm of prosthetic valve size in 6 of 20 patients (30%), and was within 2 mm of prosthetic valve size in two patients. In contrast, M-mode echocardiography failed to significantly predict aortic valve size (r = 0.14) because of its lack of two-dimensional anatomic orientation. Thus 2DE can safely and accurately predict preoperatively prosthetic aortic valve size and thereby be of great value in helping to avoid the problem of prosthesis-patient mismatch.

Assessment of left atrial dimensions by cross sectional echocardiography in patients with mitral valve disease

Left atrial dimensions were measured using cross sectional echocardiography in 37 patients with mitral valve disease and 30 normal subjects of similar ages. The anteroposterior (AP), superior-inferior (SI), and medial-lateral (ML) left atrial dimensions were determined at the end of ventricular systole using parasternal long and short axis and apical four chamber views (for SIa and MLa). To assess the reliability of these measurements cross sectional echocardiographic and angiographic left atrial volumes were compared in 19 patients with mitral valve disease, giving an excellent correlation. A moderate correlation was found between the anteroposterior dimension of the left atrium obtained using M mode echocardiography and that obtained using the parasternal short axis and long axis projections. In normal subjects a good correlation was found between SI and ML dimensions, while a lower correlation was found between SI and AP, and ML and AP dimensions. The SI dimension was the major axis of the left atrium and AP dimension the minor axis. In patients with mitral valve disease a good correlation was found between SI and ML dimensions, while SI and ML dimensions had a low correlation with AP dimensions. The AP dimension was the minor axis of the left atrium, while the SI and ML dimensions were not significantly different. All left atrial dimensions were significantly greater in patients with mitral valve disease than in normal subjects. Of 30 patients with at least one dimension increased, all three dimensions were abnormal in 16, two dimensions were increased in 10, and only one dimension was increased in four. AP, SI, and ML dimensions were abnormal in 25, 20, and 27 patients, respectively. Cross sectional echocardiography may provide a reliable estimate of left atrial dimensions. In patients with mitral valve disease a thorough examination of the left atrium using multiple cross sectional views is necessary to detect asymmetric left atrial enlargement and to measure the degree of left atrial dilatation.

Dimensions of the great arteries, semilunar valve roots, and right ventricular outflow tract during growth: normative angiocardiographic data

Systolic and diastolic diameters of the right and left pulmonary arteries (RPAD, LPAD), descending thoracic aorta (DTAD), right ventricular infundibulum (RVID), and pulmonary and aortic valve roots at the proximal, commissural and distal levels were estimated from angiocardiograms in 24 infants, children, and adolescents without heart disease, and correlated with body surface area (BSA), stroke volume (SV), cardiac output (CO), and ventricular volumes. The relationships between cardiovascular diameters and BSA were better expressed by a power function than by the other functions tried. We obtained different exponents for pulmonary and aortic valve annuli and the more distally measured great arteries (RPAD, LPAD, and DTAD), suggesting different growth patterns. The right ventricular infundibular shortening fraction (RVISF) was weakly correlated with BSA (r = -0.328), and the values obtained indicated constancy during normal growth. There was a direct proportional relationship between the pulmonary valve annulus diameter and the cube root of the right ventricular volume (r = 0.952), as well as between SV and cross-sections of the right pulmonary artery (RPAC; r = 0.916), left pulmonary artery (LPAC; r = 0.878) and descending thoracic aorta (r = 0.962). RPAC and LPAC were strongly correlated (r = 0.940), the RPAC being significantly larger than the LPAC.

Radionuclide angiographic and echocardiographic quantitation of left-to-right shunts in children with ventricular septal defect

Pulmonary to systemic blood flow ratios (Qp:Qs) were estimated in 16 children with ventricular septal defects using simultaneous echocardiography and radionuclide angiography, and compared to Qp:Qs measured at cardiac catheterization by the Fick principle method (Fick). When ratios of echographic left atrial dimensions (LAD) to body surface area (LAD/M2), body length (LAD/ht), and aortic root diameter (LAD/Ao) were compared to the Qp:Qs determined by Fick, the correlation coefficients were r = 0.70 for LAD/M2, r = 0.66 for LAD/ht, and r = 0.54 for LAD/Ao. The correlation coefficients between Qp:Qs by Fick, and left ventricular dimensions/M2 and fractional shortening of the left ventricle were not significant. The correlation coefficients between Qp:Qs and the ratios estimated by gamma-variate and area-ratio analysis of radioisotope pulmonary dilution curves were r = 0.92 and r = 0.84, respectively. Thus, radionuclide angiography provided more accurate quantitation of left to right shunting through a ventricular septal defect than echocardiography. However, difficulty in obtaining adequate bolus injections of the radioisotope may result in technical failures whereas echocardiographic measurement is possible in almost all pediatric patients. Finally, the gamma-variate method cannot accurately quantitate shunt ratios greater than 3.5 to 1.

Echocardiographic findings in patients with aortitis syndrome

Echocardiography was performed in 18 patients with the aortitis syndrome and in 20 age-matched normal volunteers. The aortic root dimension, the aortic dimension at the level of the sinotubular ridge, the aortic arch dimension, the left ventricular internal dimension, the left atrial dimension, the interventricular septal thickness, and the left ventricular posterior wall thickness were measured. All measurements, except for the left atrial dimension, were significantly greater in patients with aortitis syndrome than in the control subjects. We concluded (1) that the patients with the aortitis syndrome may have an enlarged or narrowed aorta, a dilated left ventricle and left atrium, and a thickened interventricular septum and left ventricular posterior wall; (2) that the incidence and the degree of these abnormalities depend on the presence of complications such as aortic regurgitation and arterial hypertension; and (3) that M-mode as well a cross-sectional echocardiography plays an important role in the assessment of the aorta and heart in the aortitis syndrome.

Cross-sectional echocardiographic characterization of aortic obstruction. 1. Supravalvular aortic stenosis and aortic hypoplasia

Cross-sectional echocardiographic and cineangiographic studies of the left ventricular outflow tract and ascending aorta were performed in five patients with supravalvular aortic stenosis (four hourglass and one hypoplastic). Visualization of the area of obstruction was possible in each patient using the cross-sectional system. In each case the echocardiographically determined diameter at the level of obstruction was within 3 mm of the similar angiographic value. Assessment of the extent of the lesion was possible in four of five cases. In three of these four cases the echocardiographic measurement was within 5 mm of the angiographic measurement while in the fourth the obstruction was felt to involve the total ascending aorta by both techniques. Determination of percent decrease in LVOT diameter from the aortic anulus to the level of obstruction was useful in defining obstruction and estimating severity. Cross-sectional echocardiography is a valuable noninvasive method for evaluating the ascending aorta in patients with supravalvular aortic stenosis.

Echocardiographic evaluation of left-to-right shunt in ventricular septal defect and persistent ductus arteriosus

Forty-five patients with either a ventricular septal defect or a persistent ductus arteriosus were assessed by echocardiography and cardiac catheterisation. Left atrial dimension was expressed either as a function of the body surface area (LAD cm per m2 BSA), or as a multiple of the aortic root dimension (LAD/AR), and was compared with the shunt size as determined by oximetry. A highly significant (P less than 0-001) regression relation was found for the group as a whole. A significant relation also existed for each individual group: ventricular septal defect, ventricular septal defect with pulmonary hypertension, and persistent ductus arteriosus. Regression equations were derived for the whole group. The value of echocardiography is in separating large from small shunts and in adding a dimension to the follow-up of the individual patient.

Applications of diagnostic ultrasound and radionuclides to cardiovascular diagnosis. Part II. Cardiovascular disease in the young

Echocardiography (ECHO) and radionuclide cardiography have had a significant impact on pediatric cardiology because they have proved to be sensitive enough to permit early diagnosis of many forms of heart disease and in some cases to estimate its severity and to provide information concerning ventricular performance. An overview of the anatomic and functional information that can be obtained from these two methods will be presented first, followed by details concerning the indications for their use and their relative clinical value in various acquired and congenital heart diseases. We have stressed particularly those facets of pediatric cardiac disease that differ most from those in the adult.