Quanitification of left heart volume and systolic output in transposition of the great arteries

Evaluation of left ventricular volume using automatic border detection in children: a comparison with conventional off-line echocardiographic quantification

BACKGROUND

Evaluation of the clinical usefulness of the one-line automatic border detection system for determination of left ventricular volume in children in comparison to the conventional off-line method.

METHODS

Eighty consecutive patients in whom clear images were obtained by two-dimensional echocardiography were studied. Using the Hewlett-Packard Sonos 2500 with a 3.5 or 5.5 Mhz phased array transducer, all patients were studied in the apical four-chamber imaging plane for automatic border detection and apical four-chamber and two-chamber imaging planes for manual tracing. Left ventricular end-diastolic and end-systolic volumes were measured and compared using the bi-plane Simpson method.

RESULTS

Left ventricular end-diastolic volumes obtained by automatic border detection correlated well but were slightly underestimated compared to those obtained by manual tracing (r = 0.98). Left ventricular end-systolic volumes obtained by automatic border detection also correlated well with those obtained by manual tracing (r = 0.96). Left ventricular ejection fractions compared favorably. However, left ventricular volumes obtained using the classical Pombo M-mode echocardiography showed poorer correlation with those obtained by manual tracing methods.

CONCLUSIONS

Automatic border detection is a promising method for real-time estimation of left ventricular volume. In patients with good endocardial tracking, automatic border detection can be used for routine studies of cardiovascular disease, even in children.

The diagnostic evaluation of patients with complete transposition

SummaryThe vast majority of patients with complete transposition present during either the neonatal period or early infancy. Almost always the diagnosis and the categorization of associated lesions can be established by cross-sectional echocardiography. Doppler techniques usually provide enough information about flows to allow palliative or radical surgery to be undertaken without formal cardiac catheterization, because balloon atrial septostomy is carried out with echocardiographic imaging, often on the intensive care unit. Angiography is occasionally performed to visualize the pulmonary arteries or the aorta, and, if there is suspicion of a significant elevation of the pulmonary vascular resistance, it must be measured at cardiac catheterization. Neither cross-sectional echocardiography nor aortography is entirely reliable in identifying those patterns of the coronary arteries which are a contraindication to the arterial switch operation. There is, as yet, no evidence that either magnetic resonance imaging or transesophageal echocardiography are superior to more conventional methods of preoperative diagnosis in complete transposition and it seems unlikely that they will be required.

Ventricular development and function in complete transposition: angiocardiographic evaluation

We studied 50 left ventricular cineangiograms and 41 right ventricular cineangiograms of 40 patients with usual atrial arrangement (situs solitus) together with concordant atrioventricular and discordant ventriculo-arterial connexions (complete transposition), catheterized between 1 day and 12 months of age. Our purpose was to evaluate ventricular development and function. The patients were subdivided on the basis of associated lesions into groups with intact ventricular septum; with ventricular septal defect; with ventricular septal defect together with pulmonary stenosis and with pulmonary stenosis in isolation. Each group was further separated according to age into those patients below and above 60 days. Ventricular volumes, ejection fraction and the ratio between systolic pressure and end-systolic volume were evaluated for both ventricles. The left ventricular mass, stress, and the ratio of stress to end-systolic volume were also calculated. A volume overload leads to increased left ventricular volume even at birth. With an intact ventricular septum, the left ventricle in patients with complete transposition is normal at birth and also during the first weeks of life. Myocardial mass, however, does not increase proportionately with increase in volume as the patient grows and it remains inadequate by the age of one year. The left ventricular mass is also inadequate in patients with associated anomalies when the left ventricular pressure is less than 60 mm Hg. Moreover, the left ventricle in presence of an intact ventricular septum presents a decrease in contractility during the first year of life despite the finding of a satisfactory arterial oxygen saturation. The right ventricle has a normal volume at birth which increases during the first year due to a greater diastolic filling following atrial septostomy. We noted a progressive decrease in ejection fraction, however, which is related to various factors including a reduction in contractility.

Two-dimensional echocardiographic assessment of left atrial size in children

The ability of 2-dimensional echocardiography (2-D echo) to estimate end-systolic left atrial (LA) size and volume was assessed in 140 infants and children. These subjects were divided into 2 groups. Group A included 91 patients with normal LA volume and Group B included 49 patients with LA volume overload. Five echocardiographic views (left parasternal long-axis, left parasternal short-axis, apical 4-chamber, apical 2-chamber and subcostal 4-chamber) were used. From these views, the LA long-axis and minor-axis lengths were measured and the area was planimetered. These echocardiographically derived measurements were compared with angiographically calculated LA volume. Although all echocardiographic measurements correlated well with angiographic LA volume measurements, the echocardiographic area tracked better than length measurements. Echo LA volume was calculated using 5 single-plane and 3 biplane area-length methods. LA volume calculated from either single- or biplane methods correlated well with angiographically determined LA volume. The degree of correlation depended on the method used. Echocardiographic area and estimated LA volume measured from the parasternal long-axis and apical 2-chamber views best separated patients with LA volume overload from normal. Two-dimensional echo using these views accurately segregated all patients with a LA volume greater than 180% of normal and 15 of 21 patients (71%) with an LA volume between 138% and 179% of normal. Thus, 2-D echo is useful in the evaluation of LA size and volume in infants and children.

Rest and exercise right and left ventricular function late after the Mustard operation: assessment by radionuclide ventriculography

To evaluate ventricular function late after atrial repair of transposition of the great arteries (TGA), 26 asymptomatic patients had rest and exercise radionuclide ventriculography performed a mean of 9 years (range 5 to 15) after undergoing the Mustard operation. The mean resting right (systemic) ventricular (RV) ejection fraction (EF) was 0.50 +/- 0.10 (+/- 1 standard deviation); the RVEF was less than 0.45 in 8 patients. With exercise the RVEF increased in 9 patients and either failed to increase or decreased in 15 (including all 8 patients with resting values less than 0.45). The weight-adjusted work load performed was a first predictor of RV exercise response (sensitivity 87%, specificity 92%); patients whose RVEF increased did more work. The mean resting left (pulmonary) ventricular (LV) EF was 0.58 +/- 0.09; the LVEF was less than 0.50 in 3 patients. With exercise the LVEF increased in 14 patients and did not increase in 10 (including all 3 with resting values less than 0.50). The presence of complex ventricular arrhythmia documented on Holter monitoring was a first predictor of failure of the LVEF to increase with exercise (sensitivity 84%, specificity 71%). The patient's age, operative age, postoperative interval, residual arterial desaturation, preoperative large ventricular septal defect or pulmonary stenosis, postoperative pulmonary stenosis or superior vena caval obstruction, or performance of a second open-heart procedure was predictive of the rest or exercise EF of either ventricle.

Asymmetric distribution of the pulmonary blood flow between the right and left lungs in d-transposition of the great arteries

Pulmonary angiograms, radionuclide lung images and chest roentgenograms were evaluated regarding the incidence, magnitude and natural evolution of maldistribution of the pulmonary blood flow between the lungs in 63 patients with dextrotransposition of the great arteries. Approximately half of these patients had some degree of greater perfusion of the right relative to the left lung. A significant correlation was demonstrated between the incidence of this maldistribution of blood flow and the angulation between the main and the right pulmonary arteries. For any given angulation between these vessels, additional pulmonary stenosis increased the incidence of disparity in perfusion. Our observations suggest the following developmental mechanisms: The maldistribution in flow results from the abnormal rightward inclination of the main pulmonary artery in the transposition malformation which straightens the flow axis from the main to the right pulmonary artery. Under these circumstances the momentum of the blood in the main pulmonary artery carries the blood preferentially into the right pulmonary artery. This momentum is increased when there is stenosis of the left ventricular outflow tract. Consequent differences in the mechanical properties of the two pulmonary vascular beds can increase this maldistribution. The disparity in perfusion between the lungs is not present in newborns with d-transposition, appears to be progressive in severity and in time may result in almost complete cessation of effective perfusion of the left lung. The effect of the Mustard operation on this abnormality of flow is discussed.