Objective-To assess the results of balloon dilatation of the aortic valve as first line treatment, for infints presenting in the first two months of life with cardiac failure caused by severe aortic stenosis. Design-A prospective study of all such infants presenting to a single centre over a three year period. Patients-Of 14 consecutive infants aged <2 months presenting with heart failure due to severe stenosis of the aortic valve, 12 underwent balloon dilatation of the valve. Two infants with hypoplastic left heart syndrome were excluded. Six patients had other significant cardiac malformations, nine had endocardial fibroelastosis. Four infants were moribund at presentation. Interventions-Balloon dilatation was performed through the femoral artery under general anaesthetic, with heparin and antibiotic prophylaxis. Maximum inflated balloon size was selected as equal to or slightly less than the diameter of the aortic valve measured echocardiographically. In two infants balloon dilatation of coexistent severe coarctation was performed at the same time. Results-Dilatation was achieved in all cases. Early survival rate (up to one month) was 75%; 50% survive up to three years. The two deaths occurring <24 hours after dilatation probably related to the infants' poor condition at presentation and the presence of significant left ventricular hypoplasia. The other early death was due to severe mitral regurgitation from papillary muscle dysfunction.All later deaths were related to associated malformations. The difference in left ventricular size (diameter and length) measured echocardiographically at presentation between survivors and those that died is significant (p < 0.002).

Objective-The origin of innocent systolic murmurs is uncertain. This study was designed to evaluate the relative contribution in the same subjects of the main factors implicated in the genesis of vibratory innocent systolic murmursthat is, the presence of left ventricular bands and increased left ventricular flow indices-and to examine evidence for turbulence in relation to these findings.Patients-40 children with innocent systolic murmurs and 40 paired controls without murmurs were matched for age and body surface area.Design-These subjects were examined prospectively by cross sectional echocardiography with pulsed and continuous wave Doppler and colour flow mapping.Setting-A tertiary referral centre for paediatric cardiology.Results-Left ventricular bands were significantly more prevalent in patients with innocent murmurs (70%) than in controls (48%) but only if apical bands were included (p = 0-039). Cases had significantly higher aortic peak flow velocities (p < 0-001), flow integrals (p = 0001), and stroke volumes (p = 0-014), and significantly lower heart rates (p = 0 033) than controls. There were no significant differences between cases and controls in mean aortic diameter, cardiac output, or ejection fraction. There was no evidence of turbulence in the left ventricle or outflow tract in any individual but curvilinear periodic vibrations and audible "whoops" were detected in the left ventricle by pulsed wave Doppler echocardiography. These were significantly more common in cases than controls (p < 0-01) but were not related to the presence of bands.Conclusions-Innocent systolic murmurs were unrelated to aortic size but they were strongly associated with higher aortic flow volume and velocity, with relative bradycardia, and with evidence of vibratory phenomena in the left ventricle. Left ventricular flow characteristics have also been investigated as a possible cause of the murmur. Schwarz et al reported that the peak velocities in the ascending and descending aorta were significantly greater in children with innocent murmurs and ascribed this to a narrower aortic diameter in this group compared with children without murmurs.7 These findings were confirmed by Sholler and colleagues8 but not by Van Oort et al who found no difference in the peak velocity, acceleration, or spectral width of the aortic flow signal between children with innocent murmurs and children without.9 There is currently no consensus on the origin of the musical murmur in structurally normal hearts.This study was designed to determine the prevalence of bands, the presence of turbulence about the bands and in the left ventricular outflow tract, and left ventricular flow characteristics and systolic function in children with innocent murmurs and without. Patients and methodsWe studied 40

The infant was commenced on a prostaglandin E2 infusion and transferred to the regional cardiothoracic unit for further investigation.An echocardiogram demonstrated normal intracardiac structures and connections with a structurally normal aortic valve and no evidence of coarctation. There was a small left to right shunt across a patent foramen ovale, right ventricular hypertrophy and moderate tricuspid reflux, indicating raised right ventricular pressure. The most impressive finding was that of a large mass in the aortic arch, extending from just proximal to the innominate artery to beyond the left subclavian artery (fig 1). The mass was irregular in shape with an echodense envelope. It projected into the innominate and left carotid arteries, partially obstructing flow up both carotids, left subclavian artery, and down the aorta (gradient 35 mm Hg). There was no evidence of an intracardiac mass.Magnetic resonance imaging (MRI) of chest and head was performed at 3 days of age. In addition to confirming the intra-aortic mass, it also showed areas of subacute infarction in the right frontal and left parietal cerebral cortex (fig 2A). Haemoglobin, platelet count, a clotting screen, and thrombophilia screen (including antithrombin III, protein C and S) were all within the normal range for age. Both parents had normal protein C activity. A diagnosis was made of an aortic arch thrombus with cerebral embolisation.Cardiopulmonary bypass or aortic cross clamping and surgical removal of the thrombus was not possible because it extended towards Figure 1 Two dimensional echocardiographic view in high right parasternal long axis plane showing the thrombus, partially lysed centrally, almost occluding the origin of the innominate artery. T= thrombus, A = innominate artery, AA =aortic arch, and PA =pulmonary artery.

Seven cases of single ventricle, six cases of hypoplasia of the right heart (tricuspid atresia), and one case of hypoplasia of the left heart were investigated with reflected ultrasound. In hypoplasia of the right heart there was abnormal anterior movement of the mitral valve; in hypoplasia of the left heart, abnormal anterior movement of the tricuspid valve, and in single ventricle, similiar movement of the anterior component of a single valve. In one case of single ventricle, two separate valves were identified at different depths. In none of the cases could a ventricular septum separating two ventricular cavities be identified. Ultrasound cardiography may be particularly useful as a non-invasive preliminary investigation of infants with congenital heart disease. Failure to demonstrate a ventricular septum combined with the presence of a single atrioventricular valve moving abnormally far anteriorly may be an expression of marked hypoplasia of the left or right ventricles or an anatomically single ventricle with one valve.