Since the Fontan procedure results in low pulsatile pulmonary blood flow similar to that seen in patients with a Glenn shunt, it may also be associated with abnormal distribution of flow to the lower lung lobes and with the development of pulmonary arteriovenous fistulae (PAVF). In 12 patients 0.8 to 4.5 years after Fontan procedure and in 20 patients 0.2 to 18 years after receipt of Glenn shunts we assessed ventilation (with '33Xe) and perfusion (after a peripheral injection of 99mTc-macroaggregated albumin) to compare upper to lower lobe distribution of blood flow with that in a control group. The presence of PAVF was assessed by radionuclide activity in kidneys and the brain and by a twodimensional echocardiographic contrast study. A decreased upper/lower lobe perfusion ratio was noted in 13 of 20 patients with Glenn shunts (65%) and correlated with the time after surgery (p < .05). Despite the shorter follow-up period, two of 12 (16%) patients who had undergone the Fontan procedure also had a decreased upper/lower lobe perfusion ratio, and one of these developed right heart failure. Brain and kidney radionuclide counts above control values were observed in all patients with Glenn shunts and in 11 of 12 patients who had the Fontan operation. However, in only five of 20 (25%) patients with Glenn shunts were PAVF confirmed by the two-dimensional echocardiographic contrast study. Three of the five patients with PAVF had Glenn shunts of long duration. While only two of five patients with PAVF had a decreased upper/lower lobe perfusion ratio at the time of the study, this abnormality may have been present in the other three at an earlier stage. Our data suggest that a decreased ratio of upper/lower lobe perfusion may be one of several factors associated with the development of PAVF after a Glenn shunt. Also, longer follow-up of patients who have had a Fontan procedure will be necessary to determine whether they are also a group at risk. Circulation 72, No. 3, 471-479, 1985 We therefore performed ventilation/perfusion lung scans to determine the distribution of pulmonary blood flow in patients after the Fontan procedure (RA-PA and RA-RV). We assessed the presence of PAVF by the scintigraphic appearance of 99mTc-macroaggregated albumin (MAA) in the brain and kidneys after a peripheral intravenous injection and by two-dimensional echocardiographic contrast studies. Our findings were compared with those obtained in patients with Glenn shunts and in normal subjects.

SUMMARYThe necropsy findings and characteristic angiocardiographic appearance of left atrial to coronary fenestration are presented. This fenestration is probably a fonne fruste of the developmental complex previously described of termination of the left superior vena cava in the left atrium, atrial septal defect, and absence of the coronary sinus. The partially unroofed coronary sinus provides an alternate or accessory route for blood flow when the atrium has no other functional exit. The angiocardiographic diagnosis is facilitated by selective injection of contrast material in the left atrium (or pulmonary vein) or left superior vena cava when the cava connects to the coronary sinus and left atrium via the fenestration.Congenital fenestration between the coronary sinus and left atrium is rare,'"' and when present may provide an alternative or accessory, though vital, route for the flow of blood in the patient with a severely obstructive, absent, or imperforate atrioventricular connection and inadequate interatrial septal communication.24 8-10 Sometimes, even less frequently, this unusual form of intracardiac fenestration may be an incidental necropsy finding or associated with an otherwise uncomplicated atrial septal defect.2 6 7 Though the morphological appearance of left atrial to coronary sinus window has been recorded in a few instances,'`0 this fenestration is rarely diagnosed during life and we wish, therefore, to present our pathological and angiocardiographic experience with six patients, and to comment on the circulatory and clinical implications of this anomaly. SubjectsA communication between the floor of the left atrium and coronary sinus (partially unroofed coronary sinus) was identified in six patients. The sex distribution was equal, and the age at diagnosis ranged from 2 months to 9 years and 7 months. The fenestration was diagnosed only at necropsy in two patients; at operation in one Received for publication 2 December 1980 patient, who is still alive; and by angiocardiography and confirmed at necropsy in three.All the patients had laevocardia with situs solitus. Two had obstructive anomalies of the tricuspid valve. A 2-month-old infant had tricuspid atresia, intact atrial septum, ventricular septal defect, and pulmonary stenosis and died after attempted systemic to pulmonary anastomosis. The fenestration was recognised at necropsy. The other patient, a boy aged 9 years and 7 months, had pulmonary atresia, an intact ventricular septum, a diminutive and hypertensive right ventricle, and persistence of the left superior vena cava. The diagnosis of coronary sinus to left atrial window was made by angiocardiography and confirmed at necropsy. The diagnosis was also made angiographically and confirmed at necropsy in two patients with obstructive anomalies of the mitral valve. One had mitral atresia with normal ventriculoarterial connections, and the other, mitral stenosis, Taussig-Bing type double outlet right ventricle, subpulmonary infundibular ventricular septal defect, a small ventricular sep...

SUMMARY The efficacy of treatment with spironolactone for congestive heart failure secondary to congenital heart disease was studied in 21 infants under 1 year of age. All received digoxin and chlorothiazide. In addition, group A (n = 10) was given supplements of potassium and group B (n = 11) received spironolactone. Daily clinical observations of vital signs, weight, hepatomegaly, and vomiting were recorded. Paired t test analysis showed significant reduction in liver size and weight (P< 01) and respiratory rate (P< 0 05) in group B, and less significant decreases in group A. The incidence of vomiting was slightly lower in group B. We conclude that the addition of spironolactone hastens and enhances the response to standard treatment with digoxin and chlorothiazide in infants with congestive heart failure.Spironolactone, a pharmacological antagonist of the adrenal mineralocorticoid,l has been used for some years in the treatment of congestive heart failure (CHF). By competitively binding to specific nuclear macromolecules in the distal convoluted renal tubular cells, this agent inhibits aldosterone's regulatory effects on electrolytes, increasing sodium excretion and decreasing potassium excretion.2 Spironolactone is particularly effective in the treatment of patients with CHF who are taking diuretics, because it neutralises the kaliuretic effect of the thiazides and enhances natriuresis.We have used spironolactone in the treatment of CHF for about 10 years at this hospital. However, its use in infants has been poorly documented. We therefore wished to compare the efficacy of spironolactone with that of a conventional potassium supplement, given in conjunction with digoxin and diuretic, in infants who had CHF secondary to congenital heart disease. Patients and methods

SUMMARY Regression equations were developed from left ventricular casts of known volumes to calculate left ventricular volumes from biplane cineangiography obtained in non-standard views. Volumes were calculated by Simpson's rule from casts of postmortem specimens from patients with congenital heart disease. The casts were divided into two groups: those that came from patients with abnormal right ventricular haemodynamic function (group 1, n = 11) and those that came from patients in which it was normal (group 2, n = 9). Biplane cinegrams were taken in conventional (anteroposterior/lateral, right anterior oblique/left anterior oblique) and nonconventional (long axis oblique, hepatoclavicular, and sitting up) projections. The true volume of each cast was determined from its weight and specific gravity. Correlations between measured and true volumes (r = 0-96 to 099) were excellent in all projections, although each projection overestimated the true volumes (slope = 0 72 to 0 94). The regression equations obtained from conventional views were significantly different from those from the non-conventional views; however, the regression slopes in group 1 were not different from those in group 2 in any view.Regression equations obtained by Simpson's rule do not seem to be affected by the haemodynamic state of the right ventricle. Different regression equations are required to measure left ventricular volumes from non-conventional angiograms.In angiographic measurement ofthe left ventricle the area-length method described by Dodge and colleagues has become popular.' This method was derived from the assumption that the left ventricular cavity could be represented as a three-dimensional ellipsoid of revolution. In complex congenital heart disease, however, the left ventricle is not usually elliptical and it is affected by right ventricular haemodynamic function. Data from our laboratory question whether the area-length method can be applied to the non-elliptical left ventricle.2Simpson's rule seemed to be independent of the Patients and methodsWe studied casts of the left ventricles of 20 hearts of children and infants who died of congenital heart disease. All clinical data and data on haemodynamic function were reviewed. There were four hearts with endocardial fibroelastosis; three with atrial septal defect; three with tetralogy of Fallot; two with complete transposition of great arteries with intact 182

No abstract